Silane-Modified Graphene Oxide as a Compatibilizer and Reinforcing Nanoparticle for Immiscible PP/PA Blends.
Polymer blending is a cost-effective way to generate new materials with enhanced properties. Most of the polymer blends are immiscible due to unfavorable enthalpy of mixing, exhibiting phase separation. Interfaces in immiscible blends exhibit low interfacial adhesion resulting in significant decrease in the blends' mechanical properties. Hence, there is a need to improve the interfacial adhesion between the polymer blend constituents by physical or chemical means [1, 2].
Compatibilizers are usually used to remedy the interfacial shortcomings. Block or graft copolymers are traditionally used to improve the compatibility between immiscible polymer blends. Ideally, compatibilizing copolymers form covalent bonds with the incompatible polymer phases and thus enhance the interfacial adhesion. However, compatibilizers based on copolymers require complicated synthesis in order to meet the required blends' compositions [3, 4]. In recent years, nano inorganic compatibilizers like nanoclays, layered silicates [3, 5-7], and carbon nanotubes [8, 9] were proposed as a possible replacement for conventional organic compatibilizers. Lately, graphene oxide (GO) has been used as a compatibilizer in immiscible thermoplastic polymer blends and as a reinforcing additive [10, 11]. GO is a single layer of graphite oxide, consisting of sp2 and sp3 hybridization due to carboxyls, hydroxyls, epoxies, ethers, and carboxylate groups on the basal plane and the edges [12, 13]. These active groups can interact chemically as well as reinforce polymer blends due to its large lateral to width aspect ratio and their inherent mechanical properties. These attributes could make GO a perfect compatibilizer for immiscible polymer blends [14-16].
Some research studies have demonstrated the possibility of GOs to be potential compatibilizers for polar immiscible blend like Nylon 6/poly(vinylidene fluoride) using solution blending. In these composites, the GO acts as coupling agent for both Nylon 6 and poly(vinylidene fluoride) components due to its amphiphilic characteristics. With the addition of GO, the dispersion of poly(vinylidene fluoride) in the major Nylon 6 phase becomes more uniform . Moreover, GO was studied as a compatibilizer in immiscible PA/PPO, poly(phenylene oxide), blends. The addition of GO into PA/PPO blends minimized their interfacial tension and thus improved the compatibility. Only 1 wt% of GO was found to enhance significantly the mechanical properties of the blends' tensile strength by 87% .
When a rigid particle is incorporated with the aim to reinforce a polymer, the interfacial tension between the polymer and the particle is not always low enough, and thus, there is a need to modify the particle surface or add a proper compatibilizer to achieve good compatibility with the polymer. Furthermore, compatibilizers improve the dispersion of the filler particles and reduce the agglomerate's size resulting in property enhancement of the hybridized polymer [17, 18]. In this work, silane-functionalized GO was studied as a potential compatibilizer for immiscible polymer blends. Due to the commercial availability of various functional silanes, covalent bonding could be obtained between organic and inorganic constituents [19, 20] using the appropriate functional silane. Yang et al. , alkylated by silanes, reduced GO (rGO) via covalent functionalization using dodecyl (CI2), octyl (C8), and propyl (C3) trimethoxysilanes. As a result, the rGO nanoparticles (NPs) were fully dispersed in polypropylene (PP) dissolved in xylene and led to increase of the PP's mechanical properties. Fourier transform infrared (FTIR) demonstrated that the alkyltrimethoxysilanes have been grafted on the surface of the rGO. The improvement in tensile strength and Young's modulus of the nanocomposites increased by 19.7% and 16.5%, respectively, compared to the neat PP . The ability to control the distribution and dispersion of GO in a polymeric blend of PLA and ethylene/n-butyl acrylate/glycidyl methacrylate terpolymer elastomer (EBA-GMA) was explored , using silane-functionalized GO. Hence, a variety of silanes were used, such as 3-aminopropyl triethoxysilane (APTES), 3-isocyanatopropyl triethoxysilane (95%), 3-glycidyloxypropyl trimethoxysilane (98%), 3-(2-aminoethylamino) propyl trimethoxylsilane (2APS, 97%), and 3-mercaptopropy 1 trimethoxysilane (MSH, 95%). The functionalization was carried out by reacting the GO with the silanes in ionic liquid. The GO-g-silane was first mixed with PLA, and then the PLA/GO-g-silane was compounded with EBA-GMA using twinscrew micro-extruder. GO-APS, GO-2APS, and GO-MSH formed network-like microstructures in the terpolymer elastomer phase. The selective location and distribution of GO-g-silane in the rubbery phase indicate migration of the GO-g-silane through the interface .
In the present study, GO was modified by aminosilane (AS) in low pH aqueous media for possible condensation between the oxygen groups on GO and the resulting silanol groups. Consequently, the GO and AS-modified GO (GOAS) were used as compatibilizers for PP/polyamide (PA) immiscible blends and were incorporated in the melt state, thus avoiding the use of solvents as was the case in the previous studies [10, 11, 21, 22]. Various levels of GO oxidation and GOAS modifications were used to investigate its effect on the resulting blends' morphologies and derived mechanical properties.
Graphite (Grade 3772 and 3775 from Asbury) served as the raw material to prepare the GO. Chemicals used for the oxidation process included sodium nitrate (NaN[O.sub.3], ReagentPlus[R], [greater than or equal to]99.0%. from Sigma-Aldrich), concentrated sulfuric acid ([H.sub.2]S[O.sub.4] 95%-98% AR-p, from Bio-Lab), potassium permanganate (KMn[O.sub.4], ACS reagent, [greater than or equal to]99.0%, from Sigma-Aldrich), phosphoric acid ([H.sub.3]P[O.sub.4] from Sigma-Aldrich), hydrogen peroxide (30% v/v, meets EP/BP spec, from Bio-Lab), hydrochloric acid (HCl 37% AR, from Bio-Lab), and acetic acid (Baker analyzed[R] A.C.S reagent, from J. T. Baker). Medium viscosity PA12 (Grilamid L20 from EMS Grivory) and PP homopolymer (Capilene R 50 from Carmel Olefins) were used. The silane selected for GO modification was 3-aminopropyl triethoxysilane (APTES, from Fluka).
Synthesis of GO and Silane-Modified GO
GO was prepared from graphite powder by Hummer's method  (Fig. 1). The procedure used was as follows: Graphite (1 g) and NaN[O.sub.3] (0.5 g) were stirred in [H.sub.2]S[O.sub.4] (23 mL). The components were mixed in a 1-L beaker, and the solution was cooled to 0[degrees]C as a safety measure. Next, KMn[O.sub.4] (3 g) was added to the suspension. The beaker was removed from the ice bath, and the temperature was raised to 35[degrees]C [+ or -] 3[degrees]C and maintained for 30 min. Afterward, distilled water (46 mL) was added to the suspension, and the temperature was raised to 70[degrees]C-80[degrees]C and maintained for 15 min. The reaction was terminated by adding water (140 mL) and 30% hydrogen peroxide (10 mL). The resulting GO was separated by vacuum filtration and washed five times with dilute HCl and distilled water to remove the remaining Mn ions and acid, respectively. The purified GO was freeze-dried. To achieve various oxidation levels, different amounts of KMn[O.sub.4] (0.25, 0.5, 1, 2, 3 g) were used. Modifications of GO by AS was carried out in water with the aim of modifying the surface tension of the GO and hence control the localization of the GO in the PP/PA blends. The procedure used was as follows: 1 g of GO was dissolved and stirred in 500 mL deionized water to form a stable GO dispersion; 4 mL of AS and 4 mL of deionized water were dissolved in 400 mL isopropyl alcohol. Acetic acid was added to adjust the pH to 3-4. The solution was stirred for 30 min. The silane solution was added to the GO dispersion and mixed at 70[degrees]C for 30 min. Finally, the modified GO (GOAS) was separated by vacuum filtration.
Preparation of PP/PA/GO and PP/PA/GOAS
The PP/PA/GO (50:50, w/w PP/PA) composites were prepared by melt mixing. A predetermined amount of GO and modified GO, respectively, were mixed with PP and PA by using a Thermo-Haake Mixer at 210[degrees]C with a screw speed of 200 rpm for 15 min, yielding PP/PA/GO or PP/PA/GOAS blends with different GO loadings. At first, the kinetic effect was studied. GO (or GOAS) was premixed with PP for 15 min at 190[degrees]C to form PP-GO (or PP-GOAS), and then PP-GO (or PP-GOAS) was mixed with PA for another 15 min at 210[degrees]C to form PP-GO/PA (or PP-GOAS/PA) nanocomposite blends. Furthermore, different concentrations of GOAS (1%-4%) were incorporated to the PP/PA blend to investigate the rheological properties.
The oxidation level of GO was determined by X-ray photoelectron spectroscopy (XPS). Electrical resistivity was determined for GO having different degrees of oxidation in order to examine the dependency of the electrical properties on the oxidation level. The silane modification of GO was analyzed by FTIR spectroscopy and contact angle (CA) measurements. The morphologies of GO and the GOAS-containing blends were analyzed by high-resolution scanning electron microscopy (HRSEM). To study the thermal stability of GO and GOAS, thermogravimetric analysis (TGA, Model Q50 by TA Instruments) and differential scanning calorimetry (DSC, Model Q200 by TA Instruments) were used in an inert environment at a heating rate of 10[degrees]C/min. Dynamic mechanical analysis (DMA, Model Q800 by TA Instruments) was carried out to determine the mechanical properties of the blends and the changes in glass transition temperatures (Tgs). Samples of 1,5-mm thickness were loaded in tension mode with strain control and 1 Hz frequency at a temperature ramp of 3[degrees]C/min. Parallel plate viscometry was used to investigate the rheological properties of the nanocomposites. A (Discovery HR-1) hybrid rheometer with 25 mm diameter parallel plates and 2 mm gap was used. Dynamic frequency measurements were taken in the range 0.1-100 Hz, with a fixed temperature of 210[degrees]C. Capillary rheometry of the neat polymers was also performed at 210[degrees]C. Finally, the effect of GOAS on the mechanical properties of the PP/PA blend at room temperature was determined using a mechanical tester (LLOYD Instruments, model LR 10 K).
RESULTS AND DISCUSSION
Synthesis and Characterization of GO
The effectiveness of graphite oxidation according to Hummer's method can be followed by determining the carbon to oxygen ratio. Oxidation levels having an atomic ratio of 2.1-2.9 are considered adequate . Figure 2 exhibits the XPS results of the GO NPs. The main peaks are at 284.8, 286.7, and 288.2 eV representing C--C and C--H, C--O, and C=O, respectively [24, 25]. The GO contains approximately 70% of carbon and 30% of oxygen. The carbon to oxygen ratio is 2.3, indicating an appropriate level of oxidation. Figure 3A displays a typical HRSEM of GO. As evident, the thickness of the exfoliated graphite stacks is about 3 nm, which is equivalent to about 10 graphene layers. Figure 3B presents the texture of GO showing an exfoliated morphology.
Different degrees of oxidation were prepared by varying the amount of KMn[O.sub.4] used. It was found that the oxidation level influenced the exfoliation level of the graphene sheets and their electrical properties: higher oxygen levels (~40%) resulted in excellent exfoliation but low electrical conductivity, and conversely, lower oxygen levels (~10%) led to a low degree of exfoliation but demonstrated relatively high electrical conductivity. Figure 4 presents the relationship between the XPS oxygen level, the HRSEM morphology, and the electrical resistivity.
TGA and DSC characterizations were carried out to determine the thermal stability of GO at different levels of oxygen. TGA results for GO with 10% of oxygen (GO-10%), 20% of oxygen (G0-20%), and 30% of oxygen (G0-30%) are given in Fig. 5. As can be seen, GOs at different levels of oxygen display different weight losses. The weight loss up to 100[degrees]C is attributed to the removal of water and other volatile residues from the GO. The most significant weight loss is between 180[degrees]C and 200[degrees]C and is due to the decomposition of oxygen-containing moieties in the GO layer to form rGO [26, 27]. The G0-30% loses 40% of its weight, G0-20% loses 18% of its weight, and GO-10% loses 4% of its weight. Table 1 summarizes the exothermic enthalpies of the reduction process of GOs at different levels of oxygen. The major exothermic peaks around 200[degrees]C can be attributed to the self-proportionation reaction of oxygen in the GO and the formation of CO, C [O.sub.2], [H.sub.2]O, and carbon [28, 29]. As noticed, the amount of released enthalpy increases with the level of oxygen, from 219 J/g at 10% oxygen to 1,724 J/g at high level (30%) of oxygen. DSC results support the TGA findings. Accordingly, the GO is reduced to produce rGO. Moreover, in cases where GO is incorporated into polymers in the melt state using temperatures approaching 200[degrees]C, self-proportionation of the GO may occur and may lead to change of its surface attributes.
Localization of GO in PP/PA Blends
The localization of the GO NPs in the polymer blend is one of the most important parameters affecting the polymer blend properties. The NPs may be in the PP phase, the PA phase, or at the interface between the incompatible polymers. The driving force for localization of the NPs is determined by the thermodynamic interactions between the constituents, and the actual localization is affected by the kinetics during melt mixing. From the thermodynamics point of view, the localization of a filler in polymer blends can be predicted by the wetting coefficient ([[omega].sub.[alpha]]) according to Young's equation :
[[omega].sub.[alpha]] = [[gamma].sub.filler-polymer B] - [[gamma].sub.filler-polymer A]/[[gamma].sub.polymer A-polymer B] (1)
where [[omega].sub.[alpha]] is the wetting coefficient, [[gamma].sub.filler-polymer] is the interfacial tension between the filler and the respective polymer (A or B), and [[gamma].sub.polymer A - polymer B] is the interfacial tension between the two polymers. When [[omega].sub.[alpha]] > 1, the filler is located within phase A, when -1 < [[omega].sub.[alpha]] < 1, the filler is located at the interface, and when [[omega].sub.[alpha]] < -1, the filler is located within phase B. The interfacial tension between the two components 1 and 2 ([[gamma].sub.c1-[c.sub.2]]) can be calculated by the geometric mean equation :
[[gamma].sub.c1-[c.sub.2]] = [[gamma].sub.c1] + [[gamma].sub.c2]-2([square root of ([[gamma].sup.d.sub.c1][[gamma].sup.d.sub.c2])] + [square root of ([[gamma].sup.p.sub.c1][[gamma].sup.p.,sub.c2])] (2)
where [[gamma].sub.c1] and [[gamma].sub.c2] are the surface energies of components 1 and 2, [[gamma].sup.d.sub.c1] and [[gamma].sup.d.sub.c2] are the dispersive part of the surface energies of components 1 and 2, and [[gamma].sup.p.sub.c1] and [[gamma].sup.d.sub.c2] are the polar part of the surface energies of components 1 and 2. Table 2 summarizes the calculated surface energies of the PP/PA/GO system. GO and rGO surface energies were taken from Ref. , and the polar and dispersive components were deduced by solving Eq. 2 for several solvents (CA data from Ref. ) and averaging.
Based on the surface energies in Table 2, the calculated interfacial tensions are [[gamma].sub.GO-PP] = 35.7 mJ/[m.sup.2], [[gamma.sub.GO-PA] = 15.7 mJ/[m.sup.2], and [[gamma.sub.PP-PA] = 4.0 mJ/[m.sup.2]. Introducing these values into Eq. 1 leads to a calculated wetting coefficient for GO of [[omega].sub.GO] = - 4.9. Hence, the GO is expected to locate within the PA phase. In order to locate the GO at the interface, its surface energy needs to be reduced, and this can be done by surface treatment of the GO. Thus, a surface treatment was carried out by using AS having a surface energy of 33 mJ/[m.sup.2]. The surface energy of the modified GO with AS (GOAS) was calculated by measuring CAs using pressed dry samples leading to the result of [[gamma].sup.total.sub.GOAS] = 38.26 mJ/[m.sup.2], which is lower than [[gamma].sup.total.sub.GO], due to the effect of the AS groups. The calculated wetting coefficient for GOAS is [[gamma].sub.GOAS] = - 2.6. Thus, GOAS is expected to locate within the PA phase with a tendency to locate at the interface (close to [omega] of -1). The latter morphology could be achieved by controlling the residence time during compounding of the blend. Interestingly, rGO with a wetting coefficient of [[omega].sub.rGO] = -4.3 is also expected to remain in the PA phase, despite the lower polar component and lower surface energy.
Characterization of GOAS
FTIR was used to confirm the formation of covalent bonding between GO and AS. Figure 6 exhibits the FTIR results of GO-30%, GOAS-30%, and AS itself. GOAS spectrum shows three new peaks compared with GO, at 696, 1,066 and 1,382 [cm.sup.-1] which represent Si--O--C, Si--O, and C--H bonds in the AS, respectively. In addition, the GO spectrum contains two peaks at 1,235 and 1,724 [cm.sup.-1] due to hydroxyl, carboxyl, and epoxy groups that disappear in the GOAS spectrum, indicating that these oxygen groups reacted with the silanol groups of the AS [33-35].
TGA and DSC measurements were carried out to examine the thermal stability of GOAS compared with GO. Figure 7A displays the thermal decomposition of GO and GOAS at different oxidation levels that have been obtained at ~200[degrees]C. Figure 7B describes the differential weight change. The results demonstrate the effect of AS treatment on the stability of GO. For the highest degree of oxidation, 30%, the AS treatment reduced the weight loss by 48% and for the 20% oxidation by 27%. Regarding 10% oxidation, the weight loss is the same for GO and GOAS. This may reflect the level of covalent bonding between GO and AS, being higher as the oxygen content increases. Table 3 summarizes the decomposition enthalpies based on DSC characterization. As can be noticed, the AS treatment moderated the exotherm. For 30% oxidation, the enthalpy was reduced by 55% and for 20% oxidation only by 16%. While GO is reduced to rGO, the GO-treated AS moderates the reduction of GO and forms rGOAS that is more gravimetrically stable.
The above results may lead to the assumption that the GO and GOAS were in situ reduced thermally to their respective rGO and rGOAS states, during the melt blending process that took place at 210[degrees]C.
Characterization of the PP/PA Blend
Morphology. The kinetics effect. In order to follow the migration kinetics of the rGO and rGOAS NPs toward their thermodynamic equilibrium state, a variety of blending sequences and times were devised. The localization of the NPs was determined by analyzing the resulting morphologies. Figure 9 displays the differences in the morphologies between the neat PP/PA blend, GO simultaneously blended with PP and PA, and GO premixed with PP followed by mixing with PA. The morphology of the neat PP/PA blend is co-continuous  with clear phase separation (Fig. 8A). With the addition of GO (Fig. 8B), the morphology is still co-continuous with clear phase separation. Distinctively, when the GO (Fig. 8C) or rGOAS (Fig. 8D) was premixed with PP and then mixed with PA, the morphology changed, demonstrating a lower degree of phase separation and good interphase wetting.
Localization of GOAS and rGO. To control the localization of GO in the blend by manipulating the thermodynamic affinities, the GO NPs were surface treated with AS. Since AS is hydrophilic, it tends to drive the GO to the PA phase. To investigate the kinetics effect on the NP localization, the GOAS was premixed with the PP (PP-rGOAS) and then with PA (PP-rGOAS/ PA). The resulting blend morphologies were analyzed following different durations of 4, 10, and 15 min. Figure 10 depicts the HRSEM of the nanocomposite at different times. After 4 min, a phase-separated morphology is evident and the rGOAS is located within the PP phase. After 10 min mixing, a phase-separated morphology still exists with initial migration of the rGOAS from the PP phase. Following 15 min of mixing, rGOAS is located at the interface and the morphology is more uniform with almost no clear phase separation. As can be concluded, the GO thermodynamic affinity was changed as a result of its treatment with AS, driving GOAS from the PP phase toward the interface. As shown, with the proper kinetics and duration of blending, GOAS can be driven to the interface. This is be supported by the calculated interfacial tensions of GOAS-PP (13.0 mJ/[m.sup.2]) versus GOAS-PA (2.6 mJ/[m.sup.2]) following Eq. 2 and the data in Table 2.
Thermomechanical Properties. The effects of the various oxygen levels of GO and GOAS and the derived thermodynamic and kinetic attributes on the mechanical properties of the resulting nanocomposite PP-PA blends were studied. In this part of the work, the premixing method and total time of mixing for 15 min were used. Figure 10 demonstrates the influence of different oxygen levels on the storage and loss modulus of PPrGOAS/PA blend compared with the PP-rGO/PA with 1 wt% of rGO and rGOAS, respectively. At low levels of oxidation (10%-20%), the storage modulus of the two blends was increased. It should be noticed that the location of the rGOAS10% and rGOAS-20% was at the PP/PA interface, whereas in the case of the unmodified rGO-10% and rGO-20%, the NPs were located at the PP phase (Fig. 11). At a high level of oxidation (30%), the location of the rGOAS-30% was not conclusive as the boundaries of the phases were unclear. For the blends containing rGOAS-30%, the resultant modulus was lower than the modulus obtained for the neat blend. This may be attributed to the large surface defects of the GO NPs resulting from the oxidation reaction. Although it is expected that the silane should provide partial protection to the GO, it is probably not enough, and large level of oxidation takes place generating surface defects.
Tgs of the polymers were characterized by the loss modulus (E") peaks of the DMA . Generally, in case of incompatible or partially compatible binary or ternary polymer blends, two or three peaks may appear, respectively, each peak representing the Tg of the respective blend component. Figure 12 displays the loss modulus results, and Table 4 summarizes the Tgs for PP-rGO/PA and PP-rGOAS/PA blends. Blends containing GO at all levels of oxidations display two peaks (Tg transitions) at around 2[degrees]C and 35[degrees]C representing those of PP and PA, respectively. Also, there is a slight increase in Tg as a result of adding GO. Addition of NPs to polymers results in a restriction of the segmental motion and thus increases the Tg . PP-rGOAS (20%)/PA is the only blend that demonstrates three Tg peaks, each peak corresponds to a different phase. The left most ([??]) and the right most ([??]) peaks in Fig. 12B correspond to the Tg of PP and PA, respectively, and the formation of an interphase in the case of GOAS-20% may explain the third peak ([??]). As shown in Fig. 11D, the rGOAS is located at the interface. Based on the latter result, it was concluded that GOAS with 20% oxygen possesses the optimal surface tension for the PP/PA blend.
Rheological Characterization. Rheological analysis was carried out using various concentrations of rGOAS-20% to study the effect of network formation of the NPs in the PP/PA blend on the rheological properties. The NP network formation could be studied by following the rheological percolation threshold, where a transition from liquid to solid-like behavior takes place when a 3D network is formed. The NPs' 3D network affects the viscoelastic behavior and the resulting solid-state properties of the nanocomposite. When using oscillatory rheology, the percolation threshold is identified by the appearance of a plateau of the storage modulus, G', at low frequencies . Accordingly, five blends with different concentrations (0%-4%) of rGOAS were prepared and characterized using a parallel plate rheometer. Figure 13A describes the storage modulus as a function of the angular frequency of the different levels of rGOAS-containing blends. As can be seen, the storage modulus is increased with the addition of GOAS-20%, with a maximum at 4%. As can be noticed, a plateau was obtained above a concentration of 2% rGOAS. Figure 13B shows (linear scale) the variation of G' with rGOAS concentration at 0.1 Hz indicating the formation of a percolated network between 2% and 3% of rGOAS.
The linear viscoelasticity of incompatible polymer blends has been studied by Palierne and others . Palierne's constitutive equation relates the linear viscoelastic behavior with the interfacial tension and predicts an increase in storage modulus, G', with interfacial tension, dispersed-phase volume fraction and inversely with drop diameter and viscosity ratio of the dispersed phase to matrix phase. Hence, G' values were extracted from the plateau region in Fig. 13, taking into account the viscosities from capillary rheometry at 210[degrees]C (1,300 and 930 Pa s for PA and PP, respectively). The domain sizes were taken from the SEM micrographs. Figure 14 shows that the radius of the domains, R, decreases from 15 to 5 pm. Thus, the value of [gamma]/R could be calculated, where [gamma] is the interfacial tension and R the size of the dispersed domains (Palierne's notation in Ref.  is [alpha]/R). As shown, [gamma] increased with percent GOAS, from 3.2 mJ/[m.sup.2] at 1% GOAS to 6.9 mJ/[m.sup.2] at 4% GOAS. These values are in line with the calculated blend interfacial tension, [[gamma].sub.PP - PA], and confirm as expected that G' increases with [gamma]. Moreover, as a result of GOAS migration from the PP phase into the interphase or the PA phase, [gamma] is shown to decrease, as expected for a compatibilized system.
Tensile Properties of rGOAS-Containing Blends. In addition to DMA storage modulus characterization, the tensile properties of the rGOAS-containing PP/PA blends, prepared by compression molding, were investigated. The mechanical tensile properties of PP/PA blends compatibilized with rGOAS-20% at various loadings are given in Table 5. The average tensile moduli and tensile strengths of the neat PP/PA blend are 1,530 and 12 MPa, respectively. The average tensile moduli of PP-rGOAS (20%)/PA-1%, PP-rGOAS (20%)/PA-2%, PP-rGOAS (20%)/PA-3%, and PP-rGOAS (20%)/PA-4% are 1,800, 2,060, 2,110, and 2,050 MPa, respectively, and their corresponding tensile strengths are 18, 25, 24, and 18 MPa, respectively. As can be observed, rGOAS reinforces the PP/PA blends in the range of 2%-3% NPs' weight concentration (at this range, the percolation threshold was observed), indicating a significant strength increase of 108% and modulus increase of 38% compared with the unreinforced PP/PA blend. At 4% rGOAS concentration, both the strength and modulus decrease possibly due to agglomeration of the NPs. The reinforcing effect may be attributed to the compatibilization effect of the rGOAS and its localization at the PP/PA interface as can be observed in Fig. 14. For all the blends, the localization of the GOAS is in both phases, PP and PA, and at the interface. At a concentration of 2% and 3%, GOAS coated the PP phase (Fig. 14B and C). This could be an indication of the formation of the percolated network. Figure 14D displays the agglomeration of GOAS in the PP phases in the case of 4% loading of the NPs.
Incorporation of GOAS to PP/PA blends generated a unique morphology leading to both strength and modulus increases. It is generally accepted that an increase in modulus leads to strength decrease. As can be seen, the elongation to break decreases by 12% with the addition of 2 wt% GOAS; however, the increases in modulus and strength are more significant. Figure 15 describes the stress-strain curves of PP-GOAS (20%)/PA. The area under the stress-strain curve is the amount of energy/volume that the material absorbs indicative of its toughness. Table 6 summarizes the calculated energy/volume for each blend. At the percolation threshold of the NPs (2% and 3%), the energy to break is 30% higher compared with the neat blend and then decreases at higher NP concentration. Hence, it can be concluded that at the range of 2%-3% NPs, the highest toughness was realized demonstrating the compatibility effect of GOAS.
GOAS could be used as a compatibilizer for hydrophobic/ hydrophilic polymers like PP/PA blend system by tailoring the thermodynamics and kinetics control of the system during melt blending. Various oxidation levels of GO affected the electrical properties of the GOs, the exfoliation levels of GO layers, and its localization in the blend's phases. GO was found to be unstable above 200[degrees]C, which is within the processing conditions of PP/PA. GO and GOAS undergo reduction at ~200[degrees]C; nevertheless, the AS treatment moderates the reduction, and hence, GOAS is more thermally stable. Furthermore, the reduction process of GO depends on the mixing time.
The rGO thermodynamic attributes changed as a result of treatment with AS (hydrophilic character), and thus, it may provide the means to control its localization in the hydrophobic phase (PP), hydrophilic phase (PA), or at the interface.
GO and GOAS containing 10%-20% oxygen were localized at the PP phase or the blend interface and led to an increase in storage modulus. GO and GOAS with 30% oxygen reduced the storage modulus due to defect formation on the rGO/rGOAS NP surface, due to the reduction reaction during blend mixing. PPrGOAS (20%)/PA was the only blend that exhibited three Tg peaks. It was concluded that the intermediate Tg might indicate the formation of an interphase.
Rheological percolation was achieved between 2% and 3% of rGOAS (20% oxy), resulting in improvements in the solid state of tensile strength, tensile modulus, and toughness of 108%, 38%, and 30%, respectively compared with neat PP/PA.
Finally, it was concluded that rGOAS based on 20% oxygen could serve as a model compatibilizer and reinforcing NP for hydrophobic/hydrophilic polymer blends.
[1.] M.S. Han, B.H. Lim, H.C. Jung, J.C. Hyun, S.R. Kim, and W. N. Kim, Korea-Australia Rheology Journal, 13(4), 169 (2001).
[2.] P.K.S. Mural, A. Baneijee, S. Rana, and A. Shukla, J. Mater. Chem. A, 2, 17635 (2014).
[3.] M. Si, T. Araki, H. Ade, A.L.D. Kilcoyne, R. Fisher, J. C. Sokolov, and M.H. Rafailovich, Macromolecules, 39(14), 4793 (2006).
[4.] J. Tong, H.X. Huang, and M. Wu, Compos. Sci. Technol., 149, 286 (2017).
[5.] Y. Wang, Q. Zhang, and Q. Fu, Macromol. Rapid Commun, 24 (3), 231 (2003).
[6.] Q. Zhang, H. Yang, and Q. Fu, Polymer, 45(6), 1913 (2004).
[7.] S. Sinha Ray, S. Pouliot, M. Bousmina, and L.A. Utracki, Polymer, 45(25), 8403 (2004).
[8.] S .M. Panamoottil, P. Potschke, R.J.T. Lin, D. Bhattacharyya, and S. Fakirov, Express Polym. Lett, 7(7), 607 (2013).
[9.] A. Cayla, C. Campagne, M. Rochery, and E. Devaux, Synth. Metals, 161(11-12), 1034 (2011).
[10.] J. Yang, C. Feng, J. Dai, N. Zhang, T. Huang, and Y. Wang, Polym. Int, 62(7), 1085 (2013).
[11.] Y. Cao, J. Zhang, J. Feng, and P. Wu, ACS Nano, 5(7), 5920 (2011).
[12.] J.Y. Jang, M.S. Kim, H.M. Jeong, and C.M. Shin, Compos. Sci. Technol., 69(2), 186 (2009).
[13.] J.A. Yan and M.Y. Chou, Phys. Rev. B, 82(12), 21 (2010).
[14.] Y. Hao, X. Zhao, J. Dong, and Q. Zhang, Int. J. Polym. Sei, 2017 (2017).
[15.] S. Ye, Y. Cao, J. Feng, and P. Wu, RSC Adv., 3(21), 7987 (2013).
[16.] X. Zhang, X. Xue, Q. Yin, H. Jia, J. Wang, Q. Ji, and Z. Xu, Compos. Part B: Eng., Ill, 243 (2017).
[17.] B.G. Eda and M. Chhowalla, Adv. Mater, 22, 2392 (2010).
[18.] W. Li, B. Zhou, M. Wang, Z. Li, and R. Ren, J. Mater. Sci., 50 (16), 5402 (2015).
[19.] K.L. Mittal, ed., Silanes and Other Coupling Agents, Vol. 5, CRC Press, Leiden, The Netherlands, (2009).
[20.] S.P. Lonkar, Y.S. Deshmukh, and A.A. Abdala, Nano Res., 8(4), 1039 (2015).
[21.] X. Yang, X. Wang, J. Yang, J. Li, and L. Wan, Chem. Phys. Lett., 570, 125 (2013).
[22.] Y. Fu, L. Liu, and J. Zhang, ACS Appl. Mater. Interfaces, 6(16), 14069 (2014).
[23.] W.S. Hummers and R.E. Offeman, J. Am. Chem. Soc., 80(6), 1339 (1958).
[24.] E. Pourazadi, E. Haque, W. Zhang, T. Harris, and A.I. Minett, Chem. Commun., 49, 11068 (2013).
[25.] J.F. Dai, G.J. Wang, and C.K. Wu, Chromatographia, 77(3-4), 299 (2014).
[26.] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. Binh, T. Nguyen, and R. S. Ruoff, Carbon, 45(7), 1558 (2007).
[27.] A. Lerf, H. He, M. Forster, and J. Klinowski, J. Phys. Chem. B, 102(23), 4477 (1998).
[28.] G. Wang, Z. Yang, X. Li, and C. Li, Carbon, 43(12), 2564 (2005).
[29.] I. Dekany, R. Kruger-Grasser, and A. Weiss, Colloid Polym. Sci, 276(7), 570 (1998).
[30.] F. Gubbels, R. Jerome, P. Teyssie, E. Vanlathem, R. Deltour, A. Calderone, V. Parents, and J.L. Bredas, Macromolecules, 27 (7), 1972 (1994).
[31.] A. Goldel, G. Kasaliwal, and P. Potschke, Macromol. Rapid Commun, 30(6), 423 (2009).
[32.] S. Wang, Y. Zhang, N. Abidi, and L. Cabrales, Langmuir, 25 (18), 11078 (2009).
[33.] C.Y. Lee, J.H. Bae, T.Y. Kim, S.H. Chang, and S.Y. Kim, Compos. Part A: Appl. Sci. Manuf., 75, 11 (2015).
[34.] A. Ahmadi, B. Ramezanzadeh, and M. Mahdavian, RSC Adv., 6 (59), 54102 (2016).
[35.] A. Munajad, C. Subroto, and S. Suwarno, Energies, 11(2), 1 (2018).
[36.] P. Potschke and D.R. Paul, J. Macromol. Sci. - Polym. Rev., 43 (1), 87 (2003).
[37.] J. Ren, H. Fu, T. Ren, and W. Yuan, Carbohydr. Polym, 77(3), 576 (2009).
[38.] L. Valentini, J. Biagiotti, J.M. Kenny, and M.A. Lo, J. Appl. Polym. Sci., 89(10), 2657 (2003).
[39.] L. Zonder, A. Ophir, S. Kenig, and S. McCarthy, Polymer, 52 (22), 5085 (2011).
[40.] D. Graebling, R. Muller, and J.E. Palierne, J. Phys. IV Colloq., 03(C7), 1525 (1993).
Adi Kol, Samuel Kenig , Naum Naveh
Polymers and Plastics Engineering Department, Shenkar College, Ramat Gan, Israel
Correspondence to: S. Kenig; e-mail: email@example.com
Published online in Wiley Online Library (wileyonlinelibrary.com).
Caption: FIG. 1. (A) Synthesis of GO via Hummer's method and (B) GOAS via hydrolysis of AS. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 2. XPS results of GO. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 3. HRSEM micrographs: (A) average thickness of the GO and (B) morphology of GO.
Caption: FIG. 5. DSC results of GO with 10%, 20%, and 30% oxygen. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 6. FTIR spectrographs of GO and GOAS at 30% oxygen. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 7. (A) TGA of GO and GOAS at different oxygen levels. (B) Differential TGA of GO and GOAS at different oxygen levels. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 8. Kinetic effects in (A) PP/PA, (B) PP/PA-rGO, (C) PP-rGO/PA, and (D) PP-rGOAS/PA. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 9. PP/PA/GOAS morphology after (A) 4 min mixing--phase separation, rGOAS in PP phase. (B) 10 min mixing--clear phase separation, rGOAS starts to migrate from PP phase. (C) 15 min mixing--rGOAS at the interface. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 10. DMA results: (A) storage modulus of PP-rGO/PA at various oxygen levels. (B) Storage modulus of PPrGOAS/PA at various oxygen levels. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 11. The location of rGO and rGOAS in the blend with 10%, 20%, and 30% of oxygen. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 12. DMA results: (A) loss modulus of PP-rGO/PA at various oxygen levels. (B) Loss modulus of PP-rGOAS/PA at various oxygen levels. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 13. Rheological properties (G' vs. [omega]) of PP-rGOAS (20%)/PA at different loadings. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 14. Morphology of PP-GOAS (20% oxygen)/PA at various GOAS concentrations: (A) 1%, (B) 2%, (C) 3%, (D) 4%. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 15. Stress-strain curves of PP-GOAS (20% oxygen)/PA at various GOAS concentrations. [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 1. Exothermic enthalpies of GOs at various oxygen levels. Oxygen level Enthalpy (J/g) Decomposition of GO (%) temperature ([degrees]C) 10 219 209 20 1,059 214 30 1,724 221 TABLE 2. Surface energies of the PP/PA/GO system (GO and rGO reference values from ). Polymer/ [[gamma].sup. [[gamma]. [[gamma]. filler total.sub.s] sup.d.sub.s] sup.p.sub.s] (mJ/[m.sup.2]) (mJ/[m.sup.2]) (mJ/[m.sup.2]) PA 32.66 18.18 14.48 PP 28.26 24.58 3.68 GO 62.1 7.0 55.1 rGO 46.7 4.0 42.7 GOAS 38.3 11.52 26.74 [[gamma].sup.d.sub.s], dispersive contribution; [[gamma].sup.p.sub.s], polar contribution; [[gamma].sup.total.sub.s], total surface energy. TABLE 3. Enthalpies and decomposition temperatures of GO and GOAS at different levels of oxygen. Oxygen Enthalpy Decompositiontemperature level (%) (J/g) ([degrees]C) GO 10 219 209 20 1,059 214 30 1,724 221 GOAS 10 217 211 20 889 209 30 779 201 TABLE 4. Glass transition temperatures of PP-rGO/PA and PP-rGOAS/PA at various oxygen levels. Sample Tg of PP phase [DELTA]r of Tg ([degrees]C) PP/PA -1.24 -- PP-rGO (10% oxy)/PA -3.49 -2.25 PP-rGO (20% oxy)/PA -0.63 +0.61 PP-rGO (30% oxy)/PA -2.06 -0.82 PP-rGO AS (10% oxy)/PA -4.75 -3.51 PP-rGOAS (20% oxy)/PA -2.65 -1.41 PP-rGOAS (30% oxy)/PA -0.74 +0.5 Sample Tg of PA phase [DELTA]r of Tg ([degrees]C) PP/PA 31.39 -- PP-rGO (10% oxy)/PA 40.39 +9 PP-rGO (20% oxy)/PA 38.23 +6.84 PP-rGO (30% oxy)/PA 33.42 +2.03 PP-rGO AS (10% oxy)/PA 38.43 +7.04 PP-rGOAS (20% oxy)/PA 29.98 43.69 -1.41 +12.3 PP-rGOAS (30% oxy)/PA 33.61 +2.22 TABLE 5. Mechanical properties of PP-GOAS (20% oxy)/PA and PP/PA blends at different NP levels. Blends Tensile Young strength (MPa) modulus (MPa) PP/PA 12 [+ or -] 0.7 1,530 [+ or -] 45 PP-GOAS (20%)/ 18 [+ or -] 0.2 1,800 [+ or -] 44 PA (1%) PP-GOAS (20%)/ 25 [+ or -] 0.4 2,060 [+ or -] 126 PA (2%) PP-GOAS (20%)/ 24 [+ or -] 0.7 2,110 [+ or -] 111 PA (3%) PP-GOAS (20%)/ 18 [+ or -] 0.3 2,050 [+ or -] 135 PA (4%) Blends Elongation at break (%) PP/PA 4.2 [+ or -] 0.4 PP-GOAS (20%)/ 2.1 [+ or -] 0.6 PA (1%) PP-GOAS (20%)/ 3.7 [+ or -] 0.7 PA (2%) PP-GOAS (20%)/ 3.5 [+ or -] 0.8 PA (3%) PP-GOAS (20%)/ 2.3 [+ or -] 0.1 PA (4%) TABLE 6. Energy to break of PP-GOAS (20% oxy)/PA and PP/PA blends at different NP levels. Concentration Strength at Energy to of GOAS (%) maximum load (MPa) break (MJ/[m.sup.3]) 0 12 [+ or -] 0.7 0.38 1 18 [+ or -] 0.1 0.30 2 25 [+ or -] 0.4 0.50 3 24 [+ or -] 0.7 0.51 4 18 [+ or -] 0.3 0.11 FIG. 4. Morphology and electrical surface resistivity of GO at different oxidation levels. [Color figure can be viewed at wileyonlinelibrary.com] OXIDATION DEGREE[%] 0.25 300 ~10% [OMEGA]/sq 0.5 500 ~15% [OMEGA]/sq 1 2500 ~20% [OMEGA]/sq 2 2x[10.sup.8] ~25% [OMEGA]/sq 3 5x[10.sup.10] ~30% [OMEGA]/sq 6 ~40%