Mechanical and dynamic mechanical properties of polystyrene composites reinforced with cellulose fibers

Abstract

The mechanical and dynamic mechanical properties of cellulose fibers-reinforced polystyrene composites were investigated as a function of cellulose fiber content and coupling agent effect. The composites were prepared using a corotating twin-screw extruder and after injection molding. Three levels of filler loading (10, 20, and 30 wt%) and a fixed amount of coupling agent (2 wt%) were used. The results showed that a cellulose fiber loading of more than 20 wt% caused decrease in the mechanical properties. The addition of coupling agent substantially improves the mechanical and dynamic mechanical properties. The use of coupling agent improved the storage modulus and reduced the damping peak values of the composites due to the improved interfacial adhesion. The height of the damping peak was found to be dependent on the content of cellulose fiber and the interfacial adhesion between fiber and matrix. The adhesion factor values confirm that the better adhesion occurs when coupling agent is used.

Introduction

In the last years, great attention has been dedicated to the exploitation of cellulosic fibers as reinforcement for thermoplastic composite materials, replacing glass fibers and other synthetic materials. ^{1 –3} This occurs not only due to environmental concerns but also for providing a unique combination of high performance, great versatility, and processing advantages at favorable cost. ¹

The use of cellulosic fibers in the production of thermoplastic composite materials is highly beneficial as these fibers improve the stiffness and strength of polymers. ^4,5 Moreover, these fibers are cheap, biodegradable, and come from renewable resources. ^6,7 Unlike glass fibers, they reduce the wear machinery during processing, present low density, and produce no health hazards. ^6,7 However, some drawbacks are associated with the usage of cellulosic fibers as reinforced agents in composites. The incompatibility between hydrophilic fibers and hydrophobic polymer matrices, the tendency to form aggregates during processing, and the poor resistance to moisture reduce the use of these fibers as reinforcements in thermoplastic polymers. ^1,6,7

The stress transfer from the matrix to the fiber depends on fiber–fiber and fiber–matrix interactions. ¹ Enhancement of the compatibility between thermoplastic polymer and cellulosic filler has attracted much attention from researchers as the interfacial adhesion plays an important role in determining the mechanical, thermal, and dynamic mechanical properties of the composites. To improve the adhesion between fiber and matrix, a coupling agent is normally used. The most used coupling agent is a polymer modified with maleic anhydride that promotes interactions between fiber and polymer matrix responsible for stress transfer from the matrix to the fibers. ^{8 –10} However, the polymer layer in contact with the fiber surface has different properties from the bulk matrix because of fiber/polymer interactions due to immobilization of the matrix chains, electrostatic forces, or chemical bonds. ¹¹ So, adhesion is a major contributor to the better performance of composite materials and controlling the load transfer ability of the matrix/reinforcement interface. ¹² Usually, stronger interfaces result in stronger, but more brittle, composite materials. ¹²

Dynamic mechanical analysis (DMA) is an effective method to study the relaxations in polymers and thereby the behavior of the materials under various conditions of stress, temperature, and composition of fiber composites and its role in determining the mechanical properties. ^13,14 The dynamic mechanical properties, such as, storage modulus, loss modulus, and mechanical damping, provide an insight into the level of interaction between polymer matrix and fiber reinforcement and can be used to evaluate the adhesion factor when a coupling agent is used. ^{11 –14}

On this context, the aim of this work is to investigate the performance of polystyrene (PS) composites reinforced with cellulose fibers by means of mechanical and dynamic mechanical properties, focusing on understanding the effect caused by the addition of the coupling agent on the interfacial adhesion between fiber and matrix.

Experimental

Materials

Crystal PS (N1921) was used as matrix; its melt flow index is 20 g/10 min (200°C/5 kg). It was supplied by Innova S/A (Triunfo, Rio Grande do Sul, Brazil). Bleached cellulose fibers from Pinus taeda were used as filler and supplied by Cambará S/A (Rio Grande do Sul). The poly(styrene-co-maleic anhydride) oligomer, styrene maleic anhydride (SMA), used as a coupling agent was supplied by Sartomer (Exton, Pennsylvania, USA). The coupling agent, SMA2000, contains 30 wt% of maleic anhydride groups and a weight average molecular weight of 7500 g mol⁻¹.

Composite preparation

The cellulose sheets were ground in a rotary knife mill, (Marconi, model MA 580, Brazil) and the fiber length used was between 100 μm and 270 μm. Then cellulose fiber was dried in an oven for 24 h at 70°C temperature. The composites were prepared in a corotating twin-screw extruder equipment (model MH-COR-20-32-LAB, MH Ltda, Brazil), with a screw speed of 200 r min⁻¹ in the composition of 10, 20, and 30 wt% of cellulose without and with 2 wt% of SMA as a coupling agent. The temperature profile used in the extruder heating zones from 1 to 9 was 100, 160, 175, 185, 190, 170, 180, 185, and 190°C, respectively.

The samples were dried in an oven for 24 h at 80°C temperature before injection molding. The specimens for mechanical tests, according to ASTM D790 and ASTM D256 standard, were injection molded, Himaco LH model 150-80, between 170°C and 180°C, using a preheated mold of 60°C ± 2°C. The composite formulation developments are presented in Table 1.

Table 1.

Composite formulations (percent by weight).

Sample code	PS content	Cellulose fiber content	Coupling agent content
PS	100	0	0
C100	90	10	0
C200	80	20	0
C300	70	30	0
S102	88	10	2
S202	78	20	2
S302	68	30	2

PS: polystyrene.

Characterization

The flexural tests were performed according to ASTM D790 standard at a flexural speed of 1.5 mm min⁻¹ using an EMIC DL 3000 testing machine (Brazil). Izod impact strength was measured with a CEAST Resil 25 (Akron, Ohio, USA) and a 1 J pendulum using unnotched specimens according to ASTM D256 standard. Each test value was calculated as the average of at least five independent measurements.

The fiber/matrix interface was investigated by scanning electron microscopic (SEM) analysis, using a Shimadzu Superscan S-550 scanning electron microscope, operating at an accelerating voltage of 15 kV. The samples were cryo-fractured and the surfaces were sputter coated with gold in order to eliminate electron charging.

Rectangular specimens of composites with and without coupling agents, with dimensions of 50 × 13 × 3.5 mm³, were subjected to dynamic mechanical testing using an Anton Paar Physica MCR 101 oscillatory rheometer (Germany) operating in DMA mode. The measurements were carried out in the torsion mode of the equipment and the corresponding viscoelastic properties were determined as a function of temperature. The temperature range used in the experiment was from 23°C to 120°C at a heating rate of 3°C min⁻¹ under nitrogen flow. The samples were scanned at a fixed frequency of 1 Hz with a dynamic strain of 0.1%.

Results and discussion

Mechanical properties

The mechanical properties of the composites without and with coupling agent are presented in Table 2. It can be clearly observed that for composites without SMA, increasing the cellulose fiber content increased the composite flexural modulus but decreased the values of the flexural strength and flexural strain. The changes in flexural properties of PS/cellulose composites due to the addition of cellulose fibers can be considered as follows. The flexural modulus progressively increases with filler content, probably due to the fact that the cellulose fiber is more rigid than the polymer matrix. ^3,15 The decrease in flexural strength and flexural strain with increasing filler loading can be attributed to poor dispersion of the fibers in the matrix. ¹⁶ The cellulose fibers tend to form agglomerates, due to strong interfiber hydrogen bonds, and thus dispersion of the individual fibers was inhibited as the filler content increased. ¹⁶ Furthermore, the poor interfacial bonding between hydrophilic filler and hydrophobic matrix also led to the decreased flexural strength. ^16,17

Table 2.

Flexural and impact properties of PS/cellulose fiber composites.

Sample	Flexural strength (MPa)	Flexural strain (%)	Flexural modulus (MPa)	Impact strength (J m−1)
PS	38.27 ± 1.21	1.48 ± 0.06	2904 ± 94	129.70 ± 7.24
C100	40.33 ± 2.10	1.39 ± 0.06	3438 ± 98	94.01 ± 6.34
C200	37.20 ± 1.80	1.22 ± 0.11	3629 ± 118	108.51 ± 6.57
C300	32.78 ± 1.40	0.90 ± 0.04	4180 ± 154	101.81 ± 4.47
S102	66.05 ± 1.21	1.96 ± 0.05	3871 ± 57	126.34 ± 3.24
S202	69.83 ± 1.39	1.86 ± 0.07	4611 ± 51	120.19 ± 8.07
S302	71.26 ± 0.68	1.50 ± 0.03	5261 ± 83	102.47 ± 3.81

PS: polystyrene.

Table 2 presents that the Izod impact strength of the composites without coupling agent is lower than the matrix. The introduction of reinforcements will introduce weak interfacial regions and the concentration of stress at the fiber ends, thereby decreasing the impact strength. ¹⁸ The cellulose fibers are brittle and reduce the polymer chain mobility and, therefore, its ability to absorb energy during fracture propagation. The poor interfacial bonding between filler and matrix causes micro-cracks to occur at the point of the impact, which causes the cracks to easily propagate in the composite, thereby reducing the impact strength when no coupling agents were used. ^19,20

On the other hand, the mechanical properties increased when the coupling agent is used, as shown in Table 2. The improvement in flexural strength of the composites with coupling agent is approximately 64%, 88%, and 117% higher than the composites without SMA 2000, respectively. This behavior can be attributed to the esterification reaction that occurs between the OH groups present at the surface of cellulose fibers and the anhydride groups of the coupling agent. ^3,21 A hypothetical model of the interface of the PS/cellulose fibers is shown in Figure 1.

Figure 1.

Hypothetical model of the interface of PS/cellulose fiber composites. PS: polystyrene.

The flexural strain also increased after coupling agent addition. When coupling agent is used, the dispersion of the cellulose fibers in the matrix increases and consequently the fiber–fiber interactions are reduced, thereby allowing the polymer matrix to transfer the applied stress to the fibers leading to the increase in deformation under flexural stress. As a similar effect, the improvement in interfacial adhesion enhanced the stress transfer from the polymer matrix to the stiffer cellulose fibers and flexural modulus also increase. ^21,22 The impact strength was, in general, higher for composites with SMA than for composites without coupling agent. The presence of SMA improved the cellulose fiber dispersion and led to a more uniform distribution of the applied stress. Therefore, more energy for debonding and fiber pullout is required, and thus the impact strength increases. ¹⁸

Morphological characterization

The distribution of the fiber in the polymer matrix and compatibility between the two components can be observed by SEM analysis. The SEM micrographs of non-treated composites and those treated with 20 wt% of cellulose fibers are shown in Figure 2(a) and (b), respectively. In Figure 2(a), examination of the cryo-fracture surface of the composite without coupling agent indicated the presence of voids indicating fiber pullout and larger gaps between the fiber and matrix, which is the evidence of weak interfacial adhesion at the interface. ^{6 –8} The SEM micrograph of the treated composite in Figure 2(b) shows strong bonding with traces of matrix attached in the cellulose fiber surface and reduced evidence of fiber pullout. This result demonstrates that SMA addition to the composites provides strong interfacial adhesion and good wetting, as evidenced by the almost complete absence of voids in the polymer matrix and gaps between the fiber and the matrix, ^8,18 and corroborates with higher improvement observed in mechanical properties.

Figure 2.

SEM micrographs of composites with 20 wt% of cellulose fibers without (a) and with (b) coupling agent. SEM: scanning electron microscopic.

Dynamic mechanical properties

The variation in the storage modulus (E′) of the PS/cellulose fiber composites as a function of temperature can be observed in Figure 3. There is an increase in the modulus of the PS matrix with the incorporation of cellulose fibers over the entire region. This may be due to the increase in the stiffness of the matrix due to the reinforcement effect imparted by the fibers. ^23,24 As the temperature increases, E′ decreases and thus there is as sharp decline in the E′ value in the glass transition region. This behavior can be attributed to the increase in the molecular mobility of the polymer chains above glass transition temperature (T _g). ^23,24 The drop in the modulus on passing through the T _g is lower for the reinforced composites compared with the PS matrix. That is, the difference between the moduli in the glassy state and rubbery state is smaller in the composites than in the neat PS, which clearly indicates the reinforcing effect of the cellulose fibers.

Figure 3.

Effect of cellulose fiber loading (a) and coupling agent addition (b) with temperature on the storage modulus of PS/cellulose fiber composites. PS: polystyrene.

The dynamic mechanical properties of composites are influenced by an improvement in the fiber/matrix adhesion. The effect of the use of coupling agents on the storage modulus of PS/cellulose fiber composites is shown in Figure 3(b). The E′ values for the treated composites are higher than those for the untreated composites. An improvement in the storage modulus observed when SMA was used is due to the better interfacial adhesion between the wood flour and the PS matrix. The SMA increases the fiber–matrix adhesion, causing reduced molecular mobility in the interfacial region. The sufficiently lower molecular weight of SMA2000 allows better diffusion in the polymer matrix, which indicates easier entanglement with the PS matrix. ^21,25 Furthermore, the maleic anhydride groups attached to the coupling agent PS chains cause strong interfacial interaction, probably due to the formation of chemical bonds between the maleic anhydride groups and the hydroxyl groups of the cellulose, ³ which could result in an increase in E′ value because of improvement in compatibility between filler and polymer matrix, as shown in Figure 2(b).

The effectiveness of fillers on the storage modulus of the composites can be represented by a coefficient C given in equation (1) ^12,26:

$C=\frac{{\left({E^{\prime }}_{\text{g}}/{E^{\prime }}_{\text{r}}\right)}_{\text{comp}}}{{\left({E^{\prime }}_{\text{g}}/{E^{\prime }}_{\text{r}}\right)}_{\text{resin}}},$

1

where $E_g^{\prime }$ and $E_r^{\prime }$ are the storage modulus values in the glassy and rubbery regions, respectively. The E′ values measured at 35°C and 120°C were employed as the $E_g^{\prime }$ and $E_r^{\prime }$ , respectively. A high C value indicates that the filler is less effective. The values obtained for different composites are given in Table 3. It can be observed that for composites with the cellulose fiber loading, the C value decreases and the lowest values were obtained for the composites with 30 wt% of cellulose. The composites with coupling agent present the lowest C values when compared with the composites without coupling agent. This result confirms that after using the coupling agent, more stress was transferred from the matrix to the fiber, corroborating the results of mechanical properties.

Table 3.

Variation of C with cellulose fiber loading and coupling agent treatment.

Sample	C value	T g (°C)
PS	–	99.5
C100	0.440	101.0
C200	0.192	100.8
C300	0.092	102.0
S102	0.358	101.0
S202	0.156	101.0
S302	0.055	100.9

PS: polystyrene; T _g: glass transition temperature.

The loss modulus (E″ is a measure of the energy dissipated or lost as heat per cycle under deformation, and measures the viscous response of the material. ¹² Figure 4 shows the variation in E″ as a function of the temperature for the different composite studied. The incorporation of cellulose fibers caused broadening of the E″ peak when compared to the polymer matrix. The peak broadening is an indication of a change in relaxation mechanisms. ¹² This can be attributed to the increase in relaxation time caused by the larger concentration of cellulose and because of agglomeration of some cellulose fibers originated from the formation of intramolecular hydrogen bonds between the neighborhood cellulose fibers due to the presence of free hydroxyl groups in cellulose surface increasing material heterogeneity. ^27,28

Figure 4.

Variation in loss modulus values with temperature for composites with different cellulose fibers loadings without (a) and with (b) coupling agent.

The values for the E″ of the composites increased with the increasing cellulose fiber content. The higher value is due to the increase in internal friction, which enhances the dissipation of energy. ^23,24 Additionally, the presence of cellulose fibers with a high modulus value reduces the flexibility of the composite material by introducing constraints on the segmental mobility of the polymeric molecules ^11,24 at the temperatures studied. In general, the composites with coupling agent present higher value of E″ throughout the temperature studied when compared with composites without SMA 2000. As explained earlier, the loss modulus is a measure of the viscous response of the material, thus the sufficient number of maleic anhydride groups in SMA 2000 can lead to the formation of strong chemical bonds between decoupling agent and the cellulose fibers, thereby increasing the viscous response of the composite.

Mechanical damping is an important parameter in relation to the dynamic behavior of fiber-reinforced composites. The damping properties of the material give the balance between the elastic phase and viscous phase in a polymeric structure. ²⁴ In composites, damping is influenced by the incorporation of fibers. ^27,28 The mechanical damping values for the composites are lower than those for the neat PS, as shown in Figure 5. It was found that as the amount of cellulose fibers in the composite increases, the tan δ value decreases. The incorporation of cellulose fibers reduces the height of the tan δ peak. As the cellulose content increases, the system becomes more rigid. As a result, the restriction of the polymer chains increases. The lowering of the damping curves on the addition of cellulose fibers, compared with the neat PS, is also due to the decrease in the volume fraction of the matrix. ^26,27 This affects the mechanical properties and, in turn, the damping.

Figure 5.

Effect of temperature on the tan δ value for composites without (a) and with (b) coupling agent.

The T _g values obtained from tan δ peak for the matrix and composites are presented in Table 3. When compared with the polymer matrix, the T _g values for the composites were found to be slightly shifted to higher temperatures with the incorporation of cellulose fibers, which may be attributed to the restriction imposed by the fibers on the molecular motion in the matrix. ²⁹

The energy dissipation will occur in the polymer matrix at the interface and a strong interface is characterized by less energy dissipation. ²⁴ Thus, the tan δ peak of the untreated composite was lower in comparison to the neat PS and higher when compared with the treated composites. This behavior demonstrates that a composite material with poor interfacial bonding between the fibers and matrix will tend to dissipate more energy, showing a higher damping peak in comparison to material with a strongly bonded interface. ¹¹ This fact is substantiated by the SEM micrographs in Figure 2.

Through DMA, the adhesion factor A, proposed by Kubát et al., ³⁰ can be calculated to investigate the effects of different surface treatments on the adhesion between filler and matrix. The adhesion factor can be expressed in terms of the relative damping of the composite and the polymer matrix and the volume fraction of the filler as expressed in equation (2) ^30,31 :

$A=\frac{1}{\left(1-v_{\text{f}}\right)}\frac{\tan {\delta }_{\text{c}}}{\tan {\delta }_{\text{m}}}-1,$

2

where the subscripts c and m denote composite and matrix, respectively, and v _f is the corresponding volume fraction of the filler. At high levels of interface adhesion, the molecular mobility surrounding the filler is reduced, and this reduces the tan δ _c values and consequently the A values. Thus, a low value of the adhesion factor A suggests improved interactions at the matrix–filler interface. ^30,31

The peak height of tan δ and adhesion factor values for the treated and untreated composites are shown in Table 4. The peak height was reduced with the fiber loading compared with the polymer matrix. This behavior may be attributed to the lowering of molecular mobility of polymer matrix chains upon the addition of rigid cellulose fibers. On the other hand, the adhesion factor for treated composites is lower than for untreated. This result confirms that when SMA 2000 is used, the interfacial adhesion between filler and matrix improves. This envisages that a composite material with poor interfacial bonding between the fibers and matrix will tend to dissipate more energy, showing high magnitude of damping peak in comparison to a composite material with strongly bonded interface. ¹¹ This fact is supported by the results obtained in mechanical properties and SEM morphology of the composites studied in this work.

Table 4.

Peak height and adhesion factor for samples studied.

Sample	Peak height	Adhesion factor
PS	3.59	–
C100	2.30	−0.274
C200	1.54	−0.438
C300	1.14	−0.513
S102	2.05	−0.343
S202	1.42	−0.477
S302	0.97	−0.584

PS: polystyrene.

Conclusions

The study on the mechanical and dynamic mechanical properties of cellulose fibers-reinforced PS composites revealed that the behavior of a composite is dependent on the wood cellulose fiber loading and the nature of the interface. The incorporation of cellulose fibers into the PS matrix without coupling agent results in poor interfacial adhesion and reduces the mechanical properties evaluated. However, when using the coupling agent, the mechanical properties were substantially improved. The use of coupling agent further improved the storage modulus of the composites as a result of the improved cellulose fiber/matrix adhesion. SEM micrographs indicated strong bonding and good wetting between the cellulose fibers and PS matrix. The lower adhesion factor values obtained for the treated composites demonstrate that in all formulations the polymer–fiber adhesion was improved when compared with the untreated composite and as a result the mechanical properties were increased.