• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Polymerization compounding composites on nylon-6,6/short glass fiber

Posted on: Thursday, 28 August 2003, 06:00 CDT

Nylon-6,6 was grafted onto the surface of short glass fibers through the sequential reaction of adipoyl chloride and hexamethylenediamine onto the fiber surface. Grafted and unsized short glass fibers (USGF) were used to prepare composites with nylon- 6,6 via melt blending. The glass fibers were found to act as nucleating agents for the nylon-6,6 matrix. Grafted glass fiber composites have higher crystallization temperatures than USGF composites, indicating that grafted nylon-6,6 molecules further increase crystallization rate of composites. Grafted glass fiber composites were also found to have higher tensile strength, tensile modulus, dynamic storage modulus, and melt viscosity than USGF composites. Property enhancement is attributed to improved wetting and interactions between the nylon-6,6 matrix and the modified surface of glass fibers, which is supported by scanning electron microscopy (SEM) analysis. The glass transition (tan [delta]) temperatures extracted from dynamic mechanical analysis (DMA) are found to be unchanged for USGF, while in the case of grafted glass fiber, tan [delta] increases with increasing glass fiber contents. Moreover, the peak values (i.e., intensity) of tan [delta] are slightly lower for grafted glass fiber composites than for USGF composites, further indicating improved interactions between the grafted glass fibers and nylon-6,6 matrix. The Halpin-Tsai and modified Kelly-Tyson models were used to predict the tensile modulus and tensile strength, respectively.

INTRODUCTION

Short fiber thermoplastic composites are attracting more and more attention because of their widespread applications. These composites were developed mainly to bridge the gap between continuous fiber laminates, used as structural components by the aircraft and aerospace industry, and unreinforced polymers used largely for non- load bearing applications (1). Compared to their matrices, short fiber thermoplastic composites exhibit improved mechanical, electrical and thermal properties. Moreover, they are reshapable, recyclable and their processing does not generate any volatile organic contaminant, which are important assets considering the growing environmental concerns in our era.

It is well recognized that dispersion, wetting and interaction between fiber and polymeric matrix are critical factors in designing fiber reinforced polymer composites. In the last decades, considerable effort has been made to modify the fiber-matrix interface. In this context, various fiber surface modification techniques have been developed. The most common method used is to treat the glass fibers with low-molecular weight coupling agents, dispersants, or surfactants (2). However, no satisfactory results have yet been obtained in forced deterioration tests with such coupling agents. A growing number of grafting techniques have been proposed for glass fibers, including radical (3-6), ionic (7), coordination polymerization (8) and polycondensation (9-14). The grafted glass fibers exhibit improved interface interaction which results in enhanced mechanical properties. For example, Hashimito and co-workers (3) reported that the copolymerization of styrene and glycidyl methacrylate onto glass fibers leads to an increase of 12.7% and 29.2% in flexural strength and interlaminar shear strength of glass fiber-epoxy composites, respectively.

Nylon-6,6 is one of the most widely used engineering thermoplastics and is extensively used in discontinuous glass fiber reinforced composites because of its excellent physical and processing properties. In order to obtain good fiber/matrix interface, Botelho et al. (9) and Idemura et al. (10) performed an in-situ interfacial polycondensation of nylon-6,6 and fibers by putting fibers, adipoyl chloride and hexamethylenediamine together in a reactor. This resulted in an enhanced interface as compared to that of composites prepared by melt blending of nylon-6,6 and unsized fibers. In fact, very few if any direct chemical bonds were created between the fibers and nylon-6,6, and therefore the interface was not improved enough. Recently, Salehi-Mobarakeh, Ait- Kadi and Brisson (15) carried out surface treatments of long DuPont Kevlar fibers in a lab set-up through sequential reactions of adipoyl chloride and hexamethylenediamine with the fiber surface. The same method was used to graft nylon-6,6 on the surface of long glass fibers, the interfacial shear stress of grafted glass fibers increased by 59% as compared to untreated glass libers. Moreover, the composites showed decreased water absorption (13). Because the same polymer was synthesized at the fiber surface and used as the matrix, this process was called "polymerization compounding" (16, 17). This approach has also been used by other researchers (18).

To extend the "polymerization compounding" approach to short glass fibers, we designed a lab-scale reactor to carry out nylon- 6,6 grafting onto short glass fibers, and short glass fibers with different nylon-6,6 molecular weight were prepared (19). In this paper, nylon-6,6 grafted short glass fibers and unsized short glass fibers were used to prepare nylon-6,6 composites via melt-blending. The effect of grafting nylon-6,6 onto the glass fiber surface on the thermal, mechanical, dynamic mechanical and rheological properties will be reported.

EXPERIMENTAL

Materials

Nylon-6,6 (Zytel(R) 103 FHS NCO10) was obtained from DuPont Co. Canada. E-glass short glass fibers (SGF, 123D-10P CRATEC^sup Plus^) with average length 4 mm and diameter 10 [mu]m were provided by Owens Corning Company. The unsized short glass fibers (abbreviated as USGF) was obtained by heating SGF for 4 hours at 500[degrees]C. The USGF was grafted with nylon-6,6 monomers, adipoyl chloride and hexamethylenediamine in cyclohexane as described previously (19). The nylon-6,6 grafted fibers are abbreviated as SGF-N (N stands for the number of polycondensation unit of nylon-6,6), SGF-10, SGF-20, SGF-35 and SGF-50 grafted glass fibers were used in this study. Nylon-6,6 and the fibers were vacuum dried 48 hours at 80[degrees]C before use.

Composite Preparation

A Haake-Buchler Rheocorder System 40 mixer equipped with 45 mL mixing chamber was used for the mixing of nylon-6,6 with both unsized and grafted short glass fibers at 265[degrees]C. The rotational speed was set at 45 rpm and mixing was maintained until a constant torque was reached. The weight fraction of glass fibers in the composites ranges from 5 to 30%. 0.3 wt% of Ciba-Geigy Irganox 1098 was added to prevent degradation of nylon-6,6. All samples, including tensile dogbone samples, dynamic mechanical analysis and rheological samples, were pressed using a Carver laboratory press at 280[degrees]C for 10 min under dried nitrogen atmosphere and were vacuum stored at 25[degrees]C one week before testing. All composites will be designated as follows: aa/bb-cc where aa is the type of glass fibers, either USGF or SGF-N, bb is the weight fraction of nylon-6,6 and cc the weight fraction of glass fibers. For example, SGF-50/80-20 contains 80% nylon-6,6 and 20% SGF-50. Composite compositions are listed in Table 1.

Differential Scanning Calorimetry (DSC)

The crystallization characterization was carried out under non- isothermal conditions using a Perkin-Elmer DSC-7 apparatus. The temperature and enthalpy readings were calibrated with indium at 10[degrees]C/ min; all experiments were conducted under nitrogen atmosphere. A small piece of the composites (ca. 10 mg) was used for each experiment. The samples were heated from 150[degrees]C to 300[degrees]C and maintained at this temperature for 10 min to eliminate the effects of preconditioning. The samples were cooled from 300[degrees]C to 150[degrees]C at constant rates of 10[degrees]C/min (cooling cycle), held for 5 min at this temperature, and then heated again to 300[degrees]C at a rate of 10[degrees]C/min (heating cycle). From the thermograms, transition temperatures and enthalpies were obtained and the degree of crystallinity determined.

Single Fiber Composite (SFC) Tests

The critical fiber length (l^sub c^) of USGF and SGF-20 was determined by an embedded-single-fiber technique. Thin sheets of nylon-6,6 were prepared by compression molding from nylon-6,6 pellets using a Carver Laboratory press at 285[degrees]C. A long glass fiber was put between two nylon-6,6 sheets and aligned under tension and carefully molded by compression using a Carver Laboratory press at 285[degrees]C for 10 min under nitrogen atmosphere. Resulting samples were then quenched with cold water. SFC samples were cut in a dogbone shape (ASTM type I) from thin molded films according to the ASTM D-638 method, the fiber being oriented in the longitudinal direction. The specimens were tested under tensile load at a crosshead speed of 0.5 mm/min up to 10% strain, which is well above the elongation at break of glass fibers (~ 2%). After testing, fiber fragment lengths were observed under an optical microscope (Olympus SZ-STU2) and measured using the Image- pro software (Media Cybernetics, L.P.).

Table 1. Compositions of Nylon-6,6 Composites and Their Thermal Properties.

Tensile Properties

Tensile strength and modulus tests were carried out by means of an Alliance RT/50 type mechanical tester according to ASTM D 638-96 and using V-type samples at a loading speed \of 10 mm/min.

Determination of Number Average Fiber Length and Fiber Length Distribution

Nylon-6,6 composites were burned at 500[degrees]C for 2 hours to remove the matrix. The collected glass fibers were vacuum sputtered with gold and observed under a JEOL JSM 840A scanning electron microscopy (SEM). Fiber length measurements were performed by following manually fiber image traces from SEM pictures with an electronic pen using the Image-pro software (Media Cybernetics, L.P.). Two parameters were determined: the fiber length distribution and the number average fiber length. l ^sub f^, which is equal to [Sigma]n^sub i^l^sub i^/[Sigma]n^sub i^, where n^sub i^ is the number of fibers within a certain length range (in this case 0.01 mm) centered at l^sub i^.

Observation of Cryofractured Surfaces of Composites

Surface morphology of the composites was examined by SEM (JEOL JSM 840A). The samples were cryofractured in liquid nitrogen and vacuum metallized with gold before analysis.

Dynamic Mechanical Analysis

Dynamic mechanical analysis was performed on a Rheometric Solid Analyzer (RSA-II, Rheometric Scientific) using a three-point bending geometry. Sample dimensions were 52 x 6.0 x 2.0 mm. Temperature sweep tests (from 25 to 250[degrees]C) were all carried out at a frequency of 1 Hz with a heating rate of 2[degrees]C/min and under a controlled strain of 2 x 10^sup -4^ which was in the range of linear deformation. The viscoelastic properties, such as the storage modulus (E'), loss modulus (E''), and mechanical loss factor, tan [delta] = E'/E'', were recorded as a function of temperature.

Rheological Analysis

Composites were analyzed on a rotational rheometer (ARES, Rheometric Scientific), using 25 mm diameter parallel plates. Strain sweep tests were performed in order to determine the linear viscoelasticity domain. Frequency sweep tests from 0.1 to 500 rad s^sup -1^ were then performed at 280[degrees]C and 0.5% strain under a nitrogen atmosphere and the complex viscosity ([eta]*) was recorded.

RESULTS AND DISCUSSION

Thermal Behavior

Typical DSC curves of a pure nylon-6,6 and a SGF-10/70-30 composite during cooling and heating cycles are shown in Fig. 1. Two melting peaks occur during heating cycle for all nylon-6,6 composites in this study. The same observation was reported by other authors (20), who attribute the formation of two melting peaks to the secondary crystallization of nylon-6,6. The thermal parameters, including crystallization temperature (T^sub c^), maximum melting temperature (T^sub m^), heat of fusion ([Delta]H^sub f^) and the percentage of crystallinity which were determined or calculated from the differential scanning thermograms are summarized in Table 1. The relative crystallinity (X^sub c^) of nylon-6,6 in composites was determined by using the following relationship:

Fig. 1. Typical DSC curves of nylon-6,6 and its composite.

where W^sub f^ was the weight fraction of glass fibers, [Delta]H^sub f^ the heat of fusion of composites during the heating cycle and [Delta]H^sup 0^^sub f^ the heat of fusion of perfect nylon- 6,6 crystals, for which a value of 196 J/g was used (21). The effect of glass fibers on T^sub m^ of nylon-6,6 is only marginal and indicates no clear tendency. The determined X^sub c^. values for pure nylon-6,6 and nylon-6,6 in composites show no significant difference, indicating that neither unsized glass fiber nor grafted glass fiber affected the degree of crystallinity. This conclusion is in agreement with the observations of Klein, Selivansky and Marom (20), which have shown that the presence of glass fibers affects crystallization kinetics but not its thermodynamics. Therefore, the percentage of crystallinity of the nylon-6,6 component remains unchanged with respect to contents.

In the present study, because the glass fibers were grafted with nylon-6,6, which was also used as the composite matrix, the effect of glass fibers and of sizing on the T^sub c^ of composites were investigated. From Table 1, it can be seen that the addition of glass fibers causes an increase in T^sub c^ of the composites, which further increases with increasing glass fiber contents. This phenomenon was also observed by Amash and Zugenmaier (22) on the polypropylene/glass fiber system and can be explained by the assumption that glass fibers act as nucleating agents, which increases the crystallization rate of nylon-6,6. Indeed, the presence of nucleating agent results in a reduction in the activation free energy for nucleation, which leads to an increase in the quantity of activated germs and results in a higher value for the starting crystallization temperature. Adding more glass fibers means adding more nucleating agent, leading to higher T^sub c^. It also can be seen in Table 1 that the grafted glass fiber composites have higher T^sub c^ than USGF composites, indicating that nylon- 6,6 sizing on the surface of glass fibers further reduced the activation free energy for nucleation and resulted in a higher T^sub c^. This result implies that the grafted nylon-6,6 molecules on the surface of glass fibers induce a faster crystallization than regular glass fiber surface. Further work is under way to clarify the effect of grafted glass fibers on the crystallization kinetics of nylon- 6,6 composites.

Interfacial Characterization

One of the simplest ways to characterize interfacial behavior is the single fiber composite (SFC) fragmentation test. In this test, a continuous tensile stress is applied to the SFC specimen. The applied stress will induce fiber breakage. As the test goes on, fiber fragments get shorter and shorter, down to a length equivalent to the critical fiber length, l^sub c^, for a given system. Al this point fiber breakage stops and the shear stress along the fiber is considered to be constant. Kelly and Tyson (23) proposed the following equation relating l^sub c^ and the interfacial shear strength [tau]

where d^sub f^ the diameter of glass fibers and [sigma]^sub f^ (l^sub c^) is the fiber tensile strength at its critical length. On a first approximation, [sigma]^sub f^(l^sub c^) can be considered to be

equal to the fiber ultimate strength, [sigma]^sub f^. A higher [tau] and therefore a smaller l^sub c^ characterize a better interface. The l^sub c^ can be calculated using the average length of the fragments, l . The following approximation will be used:

Table 2 summarizes the data obtained for the fragmentation test of SFCs containing the two types of fibers: USGF and SGF-20. The lower l^sub c^ and the higher [tau] show that surface treatment of glass fibers increased the interaction between fiber and matrix. About 33% increase in interfacial shear strength was found after surface treatment, in agreement with the results reported by Salehi- Mobarakeh, Brisson and Ait-Kadi (13).

The interaction between glass fibers and matrix was also observed by means of SEM analysis of cryofractured nylon-6,6 composites. The scanning electron micrographs shown in Fig. 2 indeed support SFC results. Figure 2a shows the fractured surface of USGF composite. A void is present around the fiber, indicating poor fiber-matrix adhesion. Fiber breakage occurred far from the matrix surface, indicative of fiber pull-out. This micrograph indicates that there is no sign of contact or interaction between the solid substrate and the polymeric matrix, the poor interfacial adhesion existed between the fiber and matrix. Figure 2b shows the micrograph for the equivalent composite but this time, with SGF-20. A large contact area exists between the matrix and fiber surface and fiber breakage occurred at the matrix surface. These results suggest that improved interfacial adhesion occurred in the presence of grafted glass fibers.

Fiber Length Distribution

Typical fiber length distributions are presented in Fig. 3a for an unsized glass fiber composite, USGF/70-30, and in Fig. 3b for nylon-6,6 treated glass fiber composites, SGF-20/70-30. The results clearly show that fiber attrition is very important. At the same loading, the average fiber length has dramatically decreased from 4 mm for the original glass fibers to 0.25 mm for unsized fibers and 0.30 mm for nylon-6,6, treated fibers. Distribution width is also different, a slight but noticeable decrease being observed upon use of treated fibers. A slight improvement is therefore seen in fiber attrition for this composition. Similar results were found for the other compositions measured. Several factors affect the fiber length during the processing (14), such as frictional forces from the rotating screws, which can be assumed to be same for USGF and grafted glass fiber at the same loading. A second factor involved is the frictional force between fibers. Grafted glass fibers are expected to have better wetting with nylon-6,6, which will increase their dispersion in the matrix, and therefore frictional forces between fibers should be weaker for grafted glass fibers than for unsized glass fibers. This could explain the improvement in fiber attrition. A third factor that must be taken into account is the presence of shear forces originating from fiber-matrix interaction. Grafted glass fibers are expected to show stronger interactions with the matrix, and this factor should result in glass fiber breakage. But in fact, as in this specific case the grafted polymer is the same as the matrix, interactions between glass fibers and matrix are mainly of a physical type, i.e. van der Waals electrostatic forces and possibly hydrogen bonds. These become relatively weak at processing temperature, as will be testified by the rheological behavior of composites shown later in this work.

Table 2. Fragmentation Test Data: Critical Fiber Length (l^sub c^) and Interfacial Shear Strength ([tau]).

Fig. 2. SEM micrographs of nylon-6,6 composites. Fig. 2a: USGF/ 70-30, Fig. 2b: SGF-20/70-30.

Fig. 3. Histograms of the fiber length distribution for USGF and SGF-20 composites. Fi\g. 3a: USGF/70-30, Fig. 3b: SGF-20/70-30.

The increased l ^sub f^ indicates that grafted nylon-6,6 chains enhanced wetting and dispersion in the matrix and increased interactions between glass fibers and matrix. The better length to diameter ratio of the grafted glass fibers after melt blending should contribute to the improvement of mechanical properties of resulting composites.

Tensile Properties of Composites and Their Prediction

Experimental Results

Tensile modulus and tensile strength of pure nylon-6,6 and USGF and SGF-20 composites are shown in Fig. 4 and 5, respectively, as a function of glass fiber content. Fiber incorporation in nylon-6,6 results in increased tensile modulus and tensile strength, as expected. Reinforcing was better in materials containing grafted glass fibers. The increase in mechanical properties of grafted fiber composites can be attributed partly to better length to diameter ratio of the fibers, as discussed in the previous paragraph, and partly to an improvement in interfacial adhesion between fibers and matrix. This could be related to physical entanglement of nylon-6,6 chains of the matrix with nylon-6,6 molecules grafted on the fiber surface. The effect of grafted glass fibers on reinforcement is, as expected, more pronounced for high fiber concentrations.

Fig. 4. Measured and predicted tensile modulus as a function of fiber contents.

Fig. 5. Measured and predicted tensile strength as a function of fiber contents.

Prediction of Tensile Modulus of Composites

Tensile modulus of the composites with USGF and SGF-20 were calculated at different loadings using Halpin-Tsai equations (24) as follows:

Longitudinal modulus:

and for three-dimensional randomly oriented discontinuous fiber composites, tensile modulus can be calculated from the following equation (25):

Material constants of both glass fibers and nylon-6,6 matrix are given in Table 3 and predicted values are shown in Fig. 4. As can be seen, the predicted tensile modulus of SGF-20 composites match very well experimental results. However, a large discrepancy is found between predicted and measured values for USGF composites. This can be explained by the fact that the Halpin-Tsai equation is derived for a perfect interaction between reinforcement and matrix. This seems to be the case for the SGF-20 fibers. However, for USGF, imperfect or weak interactions result in large differences between calculated and observed properties.

Table 3. Engineering Constants of Glass Fibers and Nylon-6,6.

Table 4. Calculated and Experimental Values of Tensile Strength of Nylon-6,6 and Its Composites.

Prediction of Tensile Strength of Composites

Tensile strength of the composites were estimated using the modified Kelly and Tyson model as given by the following equation (14):

where the contribution of fibers with the length shorter and longer than l^sub c^ are considered in the first and second term of the equation, respectively. V^sub i^ is the volume subfraction of fibers with lengths l^sub i^, which are shorter than l^sub c^, V^sub j^ the volume subfraction of fibers with lengths l^sub j^ that are longer than l^sub c^ and V^sub m^ is the matrix volume fraction ([Sigma]^sub i^V^sub i^ + [Sigma]^sub j^V^sub j^ + V^sub m^ = 1). [sigma]'^sub m^ is the matrix strength at the composite failure strain.

Predicted values are shown in Fig. 5 and listed in Table 4. As can be seen, these are slightly higher than experimental data. As this is the case both for USGF and grafted fibers, interfacial interactions cannot be invoked, and a systematic factor must be sought. Several approximations affect the predicted results in this study. One is the value of the glass fiber tensile strength, [sigma]^sub f^, which was taken as 1725 MPa to calculate interfacial shear strength, [tau], used in Eq 7. However, in order to remove sizing, the fibers were burnt. Salehi-Mobarakeh et al. (14) reported that, after burning, the tensile strength of glass fibers decreased and the tensile modulus slightly increased. An additional source of error is [sigma]'^sub m^, which is the matrix strength at the composite failure strain. It is known that the deformation of the matrix in composites during tensile testing is inhomogeneous. The use of composite elongation at break to determine the [sigma]'^sub m^ can therefore be a source of systematic error. Therefore, the present predicted values are considered to be in reasonable agreement with experimental results.

As the average fiber length varies with glass fiber content, a linear relationship is not necessarily expected for Fig. 4 and 5, which report tensile modulus and strength versus glass fiber content. Nevertheless, as can be seen in these figures, experimental results are very close to such a linear relationship.

Dynamic Mechanical Analysis

Temperature dependence of the dynamic mechanical properties for materials investigated in this study is presented in Figs. 6 to 8. Figure 6 shows the storage modulus as a function of temperature for USGF and grafted glass fibers composites at the same loading. The introduction of glass fibers in the nylon-6,6 matrix had profound effects on the stiffness of the composites. As expected, increases in storage modulus values of grafted glass fiber composites are significantly higher than those of unsized glass fiber composites. This can be attributed to the stronger interfacial adhesion of nylon- 6,6 to the glass fiber and to higher average fiber lengths of treated glass fibers versus USGF. It also can be seen in Fig. 6 that the storage modulus increases with increasing molecular weight of grafted nylon-6,6 on the glass fiber surface. The SGF-50 composite was found to have the highest storage modulus in the tested temperature range, suggesting that the longer the grafted nylon-6,6 molecular chains, the stronger the interactions between fibers and matrix. The interaction between grafted glass fibers and matrix might therefore be related to physical entanglement of nylon-6,6 molecules from the matrix with those present at the glass fiber surface, which can be improved by increasing the grafted nylon-6,6 molecular weight.

Figure 7 shows the storage modulus as a function of temperature for USGF and SGF-20 composites with different fiber concentrations. Filled symbols are for the SGF-20 and unfilled symbols are for the composites at the same composition (same symbol) with USGF. Results clearly indicate that, for all composites over the whole temperature range, storage moduli of SGF-20 composites are higher than those of USGF. The tan [delta] are reported in Fig. 8 and 9 and in Table 5 as a function of temperature. From Fig. 8, it is clear that increasing USGF content in the composites results, as expected for an increase in fiber concentrations, in a decrease in the maximum of tan [delta], whereas the position of tan [delta] does not change. However, for the SGF-20 composites, it was observed that the position of tan [delta] increases with increasing glass fiber loading. Inspection of the results in Fig. 8, 9 and Table 5 reveals that the peak value (intensity) of tan [delta] decreases and the glass transition temperature, T^sub g^ (the position of the maximum in tan [delta]) increases with grafting. These two parameters are related to the mobility of the molecular chains. The decrease of the peak value and the increase in T^sub g^ indicate that polymer chain mobility decreases in the presence of grafted glass fibers. This is attributed to enhanced interactions between the matrix and glass fibers, interactions promoted by grafting. These enhanced interactions contribute to a better stress transfer, which explains the enhanced stiffness obtained for composites with SGF-20 (Fig. 7).

Fig. 6. Storage modulus of composites with different glass fibers at 30 wt% loading.

Melt Viscosity

Figure 10 shows the complex viscosity versus frequency relationship at 280[degrees]C for pure nylon-6,6 and its composites with 30 wt% of USGF or grafted glass fibers. The pure nylon-6,6 shows a newtonian behavior in the frequency range tested. The addition of glass fibers increases the melt viscosity, as generally observed upon addition of a filler, and the composites exhibit non- newtonian character. At the same glass fiber loading level, grafted glass fiber composites exhibit slightly higher melt viscosities as compared to USGF composites. This higher viscosities may be attributed to the nylon-6,6 layer coating on the surface of glass fibers. This nylon-6,6 layer enhances dispersion of glass fibers in the matrix, increases fiber-matrix interactions and offer a protection against fiber breakage, which all increase the resistance to flow and result in higher viscosities. It also can be seen that SGF-50/70-30 exhibits higher melt viscosity than SGF-10/70-30. This increased viscosity resulted from the longer molecular chains of grafted nylon-6,6 on the SGF-50 compared with SGF-10, which further increases the interactions between fibers and matrix.

Fig. 7. Storage modulus of USGF and SGF-20 composites as a function of loading.

Fig. 8. Tan [delta] of nylon-6,6 and its composites with USGF.

Fig. 9. Tan [delta] of nylon-6,6 and its composites with SGF-20.

CONCLUSIONS

It was shown that chemical modification of short glass fibers by grafting of nylon-6,6 molecules led to a higher crystallization temperature and an increase in tensile modulus and strength as compared to untreated fiber composites. Increased interactions between the glass fibers and matrix were studied through single fiber composite tests and scanning electron microscopy. Smaller l^sub c^ were also obtained for treated glass fibers. The treatments slightly decreased fiber attrition during fiber-matrix mixing, resulting in higher l ^sub [function of]^, which in turn affected the mechanical properties of the resulting composites. Dynamic mechanical analysis of grafted glass fiber composites showed an increase in glass transition temperat\ure and a decrease in intensity of tan [delta] as compared to the unsized glass fiber composites. This was attributed to the reduced chain mobility at the interfacial zone resulting from the increased interactions between glass fibers and matrix. A slight increase in melt viscosity after treatment also implied that new interactions between glass fibers and matrix occur.

Table 5. Tan [delta] of Investigated Materials as Determined From DMA at 1 Hz.

ACKNOWLEDGMENTS

The authors acknowledge the financial support provided by the Natural Sciences and Engineering Research Council (NSERC) of Canada and le Fonds pour la Formation de chercheurs et l'aide a la recherche (FCAR) of the province of Quebec. We also wish to thank Steve Pouliot for his help in testing the OH groups at the surface of glass fibers, Marlaine Rousseau for her help in rheology measurements and Marc Choquette for SEM measurements.

Fig. 10. Complex melt viscosity ([eta]*) of Nylon-6,6 and its composites as a function of frequency.

ABBREVIATIONS AND SYMBOLS

nylon-6,6: Polyamide 66

USGF: Unsized short glass fibers

DSC: Differential Scanning Calorimetry

SEM: Scanning Electron Microscopy

DMA: Dynamic Mechanical Analysis

wt%: Weight Percentage

tan [delta]: Loss tangent

SGF-N: Grafted short glass fiber with N rounds nylon-6,6 molecules

SFC: Single fiber composite

l^sub c^: Critical fiber length

l ^sub [function of]^: Average fiber length in composites

E': Storage modulus

T^sub c^: Crystallization temperature

T^sub m^: Melt temperature

[Delta]H^sub [function of]^: Heat of fusion

X^sub c^: Degree of crystallization

[Delta]H^sup 0^^sub [function of]^: Heat of fusion of perfent crystal in PA66

[tau]: Interfacial shear strength

[sigma]^sub [function of]^: Tensile strength of glass fiber

d^sub [function of]^: Diameter of glass fiber

l : Average fiber length in SFC test

E^sub [function of]^: Tensile modulus of glass fiber

E^sub m^: Tensile modulus of nylon-6,6

E^sub 11^: Longitudinal modulus of composite

E^sub 22^: Transverse modulus of composite

E^sub random^: Tensile modulus of three-dimensional randomly oriented discontinuous fiber composite

[sigma]'^sub m^: Tensile strength of nylon-6,6 at the composite failure strain

V^sub m^: Matrix volume fraction

REFERENCES

1. D. G. Brady and J. L. Kardos, in Thermoplastics Composite Materials, L. A. Carlsson ed., Elsevier, New York (1991).

2. E. P. Plueddemann, Silane Coupling Agent, Plenum Press, New York (1982).

3. K. Hashimoto, T. Fujisawa, M. Kobayashi, and R. Yosomiya, J. Appl. Polym. Sci., 27, 4529 (1982).

4. T. Fujisawa, M. Kobayashi, K. Hashimoto, and R. Yosomiya, J. Appl. Polym. Sci., 27, 4849 (1982).

5. K. Fujiki, M. Sakamoto, A. Yoshida, and H. Maruyama, J. Polym. Sci.: Polym., Chem., 37, 2121 (1999).

6. R. R. N. Sailaja and A. Kumar, Polym. Compos., 17, 637 (1996).

7. N. Tsubokawa, H. Hamada, and K. Fujiki, Polymer, 35, 1084 (1994).

8. S. E. Barbosa, M. L. Ferreira, D. E. Damiani, and N. J. Capiati, J. Appl. Polym. Sci., 81, 1266 (2001).

9. E. C. Botelho, N. Scherbakoff, M. C. Rezende. A. M. Kawamoto, and J. Sciamareli, Macromolecules, 34, 3367 (2001).

10. S. Idemura and H. Haraguchi, U.S. Patent 6,063,862 (2000).

11. Z. Yu, A. Ait-Kadi, and J. Brisson, Polym. Eng. Sci., 31, 1222 (1991).

12. Z. Yu, A. Ait-Kadi, and J. Brisson, Polym. Eng. Sci., 31, 1228 (1991).

13. H. Salehi-Mobarakeh, J. Brisson, and A. Ait-Kadi, J. Mater. Sci. 32, 1297 (1997).

14. H. Salehi-Mobarakeh, J. Brisson, and A. Ait-Kadi, Polym. Compos., 19, 264 (1998).

15. H. Salehi-Mobarakeh, A. Ait-Kadi, and J. Brisson, Polym Eng. Sci., 36, 778 (1996).

16. A. Ait-Kadi and C. Dubois, "Rheology of Polymer-Coated Fumed Silica Suspensions," PPS-17, Montreal (2001).

17. W. Feng, A. Ait-Kadi, and B. Riedl, accepted by Polym. Eng. Sci., 42 (2002).

18. Z. Wu, C. U. Pittman, and S. D. Gardner, Carbon, 34, 59 (1996).

19. M. Bera-Aberem, W. Feng, A. Ait-Kadi, B. Riedl, and J. Brisson, in preparation.

20. N. Klein, D. Selivansky, and G. Marom, Polym. Compos., 16, 189 (1995).

21. J. Brandrup, E. H. Immergut, and E. A. Grulke, eds., Polymer Handbook, 4th Ed., V-121, John Wiley and Sons, Inc., New York (1999).

22. A. Amash and P. Zugenmaier, J. Appl. Polym. Sci., 63, 1143- 1154 (1997).

23. A. Kelly and W. R. Tyson, J. Mech. Phys. Solid., 13, 329 (1965).

24. J. C. Halpin and J. L. Kardos, Polym. Eng. Sci., 16, 244 (1976).

25. P. K. Mallick, in Comprehensive Composite Materials, Volume 2, Polymer Matrix Composites, R. Talreja and J.-A. E. Manson, eds., Elsevier, Amsterdam (2000).

WEI FENG1, ABDELLATIF AIT-KADI1,**, JOSEE BRISSON2,*, and BERNARD RIEDL3

1CERSIM/Department of Chemical Engineering

2CERSIM/Department of Chemistry

3CERSIM/Department of Wood Sciences Laval University

Quebec City, Canada, G1K 7P4

*To whom correspondence should be addressed. E-mail: josee.brisson@chm.ulaval.ca

**Deceased December 3, 2001.

Copyright Society of Plastics Engineers Aug 2003

More News in this Category