Preparation and characterization of composites using blends of divinylbenzene-based hyperbranched and linear functionalized polymers

In this study, hyperbranched polymers were explored as matrix modifiers to create E-glass fiber (GF) reinforced polymer composites with enhanced mechanical properties. Hyperbranched polymers have lower viscosities than their linear equivalents, potentially providing enhanced fiber wet out leading to improved stress transfer. Hyperbranched (HB), hydrogenated hyperbranched (H-HB), and linear functional (LF) divinyl benzene were blended with linear polystyrene (LP) to form a range of composite matrix formulations. Blends of the HB and LP polymers were used since the neat hyperbranched polymers alone proved to be highly brittle when formed into a film. A neat LP-GF composite was also prepared as control. Of the three matrix modifiers considered, only the H-HB provided an improvement in mechanical properties in comparison to LP-GF. With the addition of 10 and 20 wt% H-HB, respectively, the flexural modulus increased by 25% ( p < 0.05) and 36% ( p < 0.05) and flexural strength increased by 15% ( p < 0.05) and 31% ( p < 0.005). The enhanced mechanical properties were attributed to better fiber wetting along with crystallization observed with the addition of 20 wt% H-HB. The non-reactive ethyl ( CH 2 CH 3 ) chain end group of the macromolecular H-HB resulted in a plasticizing effect, which in turn improved its wettability. The LP:HB polymer blends, on the other hand, underwent crosslinking due to the presence of the vinyl ( CH CH 2 ) chain ends leading to poor wettability in comparison to the LP:H-HB and LP:LF blended films and hence lower mechanical properties.

HB polymer to modify an epoxy matrix based composite and reported a 2.5 fold increase in fracture toughness (K IC ) with addition of 5 phr (parts per hundred rubber) of HB and also reported that this increase was obtained without affecting the viscosity, processability, and glass transition temperature of the epoxy resin. DeCarli et al. 4 also reported a 224% increase in fracture toughness for carbon fiber reinforced epoxy anhydride composite on addition of 10 wt% of epoxy terminated HB polymer (Boltorn E1) as a toughening agent.
In addition, HB polymers functionalized with vinyl reactive groups have been used as fiber/filler processing aids for thermoset epoxy composites. 1,3,5,6 For example, Li et al. 2  In the field of thermoplastic composites, HB polymers were investigated to improve fiber/matrix adhesion and filler dispersion. 7 Sun et al. 8 used sisal fibers grafted with poly(amidoamine) dendrimer as reinforcement to produce sisal fiber-reinforced polypropylene composites. They studied the effects of dendrimer generations on the mechanical properties of composites. For generation 2.0 dendrimer, they found that the tensile, flexural, and impact strength of the composites (at 30 wt% fiber loading) increased by 29%, 13%, and 54%, respectively. Lu et al. 9 prepared sisal fiber/polypropylene/carboxyl terminated HB polymer composites and reported a 21.5% and 9.7% increase in impact and flexural strength, respectively, for the HB modified composites in comparison to unmodified polymer fiber composites. Wong et al. 10 reported improved brittleness of a polylactic acid matrix via addition of an HB polymer and also reported that with addition of 10% v/v HBP, the toughness of the composite doubled. They suggested that the improved wetting of the fibers by the matrix (when HB was present) had improved the toughness.
On the other hand, non-fiber, composites have also been prepared with HB polymers to exploit their unique properties. Zhou et al. 11 reported on the modification of multi-wall carbon nanotube (MWCNT) with an HB polymer containing UV reactive functional groups. The resin was cured under UV irradiation and revealed improved tensile strength and toughness values by 41% and 105%, respectively, with addition of only 0.1 wt% MWCNT. 11 Other studies also report on the use of HB containing composites due to their unique properties. For example, HB Polymers have facilitated a wide range of applications in inorganic-organic composite materials, 12 15 The temperatures utilized in this study for composite manufacture were determined from a previous rheological study of pure LP and its blends. In that study it was observed that LP started to show flow behavior at around 190 C,   whereas LP:HB 90-10 and 80-20 did not show any flow behavior   from room temperature up to 200 C. Whereas, the LP:H-HB showed   flow behavior at around 170 C while LP:LF 90-10 and 80-20 showed flow behavior around 135 C.
In this study, HB polymers were used in blends, with their linear analogs, in order to create the matrix materials used in the preparation of polymer E-Glass fiber composites. Furthermore, in the manufacture of these composites, the E-Glass fiber was used as received so that the study concentrated solely on the effects of matrices on the mechanical properties of the polymer fiber composites. The mechanical properties of the LP film and blends prepared at both room temperature (R.T) and heat treated (H.T) at elevated temperatures during composite manufacture were also evaluated. Furthermore, the flexural properties of the composites produced were evaluated and compared.   20) were prepared using the methods described in our previous paper. 15 The LP film and the blends were prepared by dissolving specific amounts of the polymers in chloroform (4% w/v). The solution was stirred for 3 h using a magnetic stirrer for homogenization at room temperature (~25 C) and then poured into a PTFE mold. The solvent was then allowed to evaporate at room temperature (25 C). The films were further dried in an oven at 50 C for 3 days to remove any residual solvent. were prepared using the methods described in a prior report. 15 The LP film and the blends were prepared by dissolving specific amounts of the polymers in chloroform (4% w/v). The solution was stirred for 3 h using a magnetic stirrer to create a homogeneous mixture at room temperature (~25 C) and then poured into a PTFE mold. The solvent was then allowed to evaporate at room temperature (25 C). The films were further dried in an oven at 50 C for 3 days to remove any residual solvent.

| Preparation of composites
The composite samples were prepared via a film stacking process. The polymer films were stacked alternately with woven E-glass fiber mats into a 1 mm thick mold cavity between two metallic plates. The width and length of the mold were 60 Â 60 mm 2 , respectively. For composites with matrices of LP, LP:HB, and LP:H-HB, the entire stack was then heated in the press for 10 min at 200 C and then pressed for another 10 min at 40 bar. However, for the composite with LP:LF matrices, the polymer fiber stack was heated at 140 C (rather than 200 C) for 10 min and then pressed for 10 min at 40 bar pressure.
After pressing, the composites were cooled, while maintaining the same pressure, at a rate~10 C min À1 to below the T g of LP (~101 C) for LP-GF, LP:HB-GF, and LP:H-HB-GF composites. Meanwhile, for LP:LF-GF composite, cooling was maintained at the same pressure at a rate~10 C min À1 to below the T g of LF (~80 C). LP-GF composite was prepared to use as control.
The resulting laminated composites were cut into 25 mm length Â 10 mm width coupons for flexural testing, using a Diamond cutter.
The composites prepared in this study with their respective sample codes, polymer blends, and volume fractions of polymer blends and fibers are presented in Table 1.

| Dynamic viscosity measurement
The dynamic viscosity was determined using an Anton-Paar 302 rheometer. Measurements were performed using a 25 mm parallel plate gap, which was adjusted between 0.5 and 0.6 mm. All tests were performed with a logarithmically increasing shear rate range between 0.01 s À1 and 100 s À1 . Viscosity of LP, LP:HB, and LP:H-HB blends were measured at 200 C and LP:LF was measured at 140 C to match the processing temperatures used during composite manufacture. All samples were tested in triplicate.

| Differential scanning calorimetry (DSC)
LP:H-HB 90-10 and 80-20 blends were investigated for thermal properties using a DSC (Q2000, TA instruments, UK). Samples (approximately 5 mg) were heated from 25 C to 200 C at a heating rate of 10 C min À1 under nitrogen gas flow (100 ml min À1 ). After heating, the samples were subsequently cooled to room temperature (i.e., at a rate of approximately 20 C min À1 ) before ramping again to 200 C at the same heating rate. The data were taken from the second cycle.

| Burn off tests
where, P is the percentage loss on ignition.
m 1 is the mass of the container. m 2 is the initial total mass of the container plus the specimen.

| Flexural testing
The flexural strength and modulus of the composite samples (10 Â 25 mm 2 ) were evaluated by flexural (three-point bending) tests using an Instron 5969 testing machine (Software-QMAT). These measurements were done according to the standards BS EN ISO 14125:1998. A crosshead speed of 0.5 mm min À1 and a 5 kN load cell was used. Flexural studies were conducted using three repeat specimens.

| Scanning electron microscopy
Scanning electron microscope images were taken to examine the cross section of the freeze-fractured composite plates. The specimens were carbon coated prior to examination and viewed with a JEOL 6400 SEM scanning electron microscope operated at 10 kV in secondary electron mode (SE).

| Statistical analysis
Statistical analysis on a sample group (more than two specimens) was performed using Tukey's Multiple Comparison Test (95% confidence intervals) through a one-way analysis of variance (ANOVA), employing Graph Pad Prism software (version 5.01).

| Viscosity
The temperature of composite making and as well as viscosity study of LP and its blends were based on a previously performed rheological study. In that study, LP showed flow behavior (loss modulus became higher than storage modulus; G 00 > G 0 ) at~190 C and LP:H- The viscosity of the LP and blends are presented as a function of shear rate in Figure 1. Both LP and the blends showed shear thinning behavior. It has been reported that HB polymers show Newtonian behavior due to the absence of chain entanglement. 16,17 The shear thinning behavior of LP and blends was, therefore, attributed to the contribution of polymer chain entanglements from LP. 18 The viscosity of LP: HB blends were found to be highest (5.9 Â 10 4 Pa.s) at 200 C at low shear rate (0.01 s À1 ). This behavior could be explained taking into account the crosslinking effect of the HB polymer at that temperature, which was confirmed from the previous rheological study performed as described in, 15

| Tensile test
The tensile modulus data of LP and blends both at room temperature (R. T) and heat treated (H.T) are presented in Figure 3(A, B)     have been previously reported for use as matrix modifiers for thermoplastic polymer composites to improve processability. 7

| CONCLUSIONS
In conclusion, after heat treatment the tensile modulus of LP:HB of 90-10 and 80-20 increased significantly in comparison to LP (p < 0.005), which was suggested to be due to crosslinking, and their