Glass Fiber Matrix Composite
Piyush K.Vaghela1, James O. Stoffer2 and Oliver C. Sitton3 1,3
Department of Chemical Engineering, University of Missouri-Rolla, Rolla, MO 65401. 2Chemistry and Materials Research Center, University of Missouri-Rolla, Rolla, MO 65401.
Introduction Severe windstorms such as hurricanes result in extensive damage to building envelopes and the interior due to the effects of flying debris and fluctuating wind pressures. Developing and manufacturing a new transparent laminated glass window would greatly reduce the breakage of the glass panels under adverse conditions. A high strength, low weight composite would be suitable for this purpose. Although there are many reinforcement and matrix materials available to produce such composites, the choices for selecting these materials are limited since optical transparency is vital. This work used the Christiansen method to reduce the loss of transparency in composites by matching the refractive indices of the reinforcement and matrix phases to eliminate light scattering and reflection.1 This paper focuses on the criteria necessary to produce and characterize an improved polyester/glass fiber transparent composite containing 15 volume percent glass fibers. The Young’s modulus, modulus of rupture, work of fracture, toughness and laminar shear strength are measured for composites using two types of silane agents. The relationship of optical transmission of the polyester composite with respect to volume percent fiber and fiber diameter were also determined.
Experimental Procedures Materials. The unsaturated polyester resin (Part 811-191) and a peroxide mixture (Part 811-194) were purchased from the Leco Corporation. The coupling agents were 3-(trimethoxysilyl)propyl methacrylate (3TPM) and N-( 3-[trimethoxysilyl]propyl )- ethylenediamine (N3TPED) from Aldrich Chemical Co. The refractive index of the polyester is 1.554 ± 0.001 at the sodium line. The refractive index of the glass fibers is smaller than that of the bulk glass because their rapid cooling reduces the glass density.2 Thus, larger diameter fibers have a higher refractive index than smaller diameter fibers since the larger diameter fibers cool more slowly and have a higher density. Drawing bulk glass into fiber or ribbon lowered the refractive index of the glass by approximately 0.006, so a bulk glass with a refractive index of 1.560 was desired. NSK-11 glass (Schott Glass Technologies, Inc.) has a refractive index of 1.56388, which is suitable for this work. This glass was tested to determine the conditions where its refractive index dropped to 1.554. Glass Fiber Production. A fiber puller was constructed at the Materials Research Center at the University of Missouri-Rolla to be used for glass fiber production. Glass fragments were placed in a electrically heated platinum container fitted with a platinum/10% rhodium bushing. The temperature ranged between 1150 and 1300°C, causing the glass to melt and continuous fibers were pulled and wound on a rapidly spinning drum. Glass fibers with a diameter between 20 and 150 µm can be produces. The fibers were then chopped by hand with scissors to the desired length. A digital micrometer (Mitutoyo Corporation) measured the glass fiber diameter, which had a maximum error of just ± 1 µm. The Becke line method was used to measure the refractive index of the glass fibers and polymer matrix. Composite Preparation and Testing. Unidirectionally oriented glass fiber reinforced polyester composite was prepared. A dip coat procedure was used to apply a coating to the glass fibers. The dip coat procedure removes the by-products of the coupling reaction from the interface and produces composites with quality optical surfaces. A 600 ml solution of 95 percent (w/v) ethanol in water was adjusted to pH 4.5-5.5 with acetic acid3,4 3TPM was added (12.5 ml) with stirring to yield a 2% final weight concentration. Five minutes were allowed for hydrolysis and silanol formation. Glass fibers of 44, 54, 70, 71, 121 and 123µm diameters were dipped briefly in ethanol. The glass fibers were rinsed for fifteen minutes with ethanol. The curing of the silane layer took 24 hours at room temperature (< 60 % relative humidity). The same procedure was repeated using N3TPED as a coupling agent. The water absorbed on the fiber surface hydrolyzes the trimethoxy silane to the
hydroxy species, which can bond to the glass surface.5 Most silaceous substrates have 4-12 silanols per mµ2 . Thus, an average of 7500m2 should be covered by one mole of evenly distributed silane.4 Approximately 3-8 monolayers (< 10 nm) of trimethoxy silane deposits onto the fiber surface using this dip coat technique. The ethanol rinse washes the methanol and water by-products from the interface. The trimethoxy silane bonds to the adjacent silane molecules and to the glass surface resulting in a polymeric coating. After rinsing and curing the glass fibers for at least 24 hours, the coated glass fibers were placed in a box-like steel mold. The inner dimensions of the mold were 170 mm x 61 mm x 17 mm. The interior vertical sides of the mold were lined with Teflon® (Teflon is a registered trademark of Dupont). Polyester resin and the peroxide initiator were then poured into the prepared mold. The mold was degassed (8-10 inch Hg) for 45 minutes while simultaneously heating at 60°C to remove dissolved gases in the resin and gases adsorbed on the fibers. All composites used in this paper were tested with glass fibers placed at the bottom of the composites. After polymerization, all composites were removed from the mold and heated at 50°C for four hours to ensure complete polymerization and that all samples had a uniform thermal history. The composites were surface machined to reduce the sample thickness to 5 mm. Individual mechanical test bars 120 mm in length by 17 mm wide were cut from the full sample with a band saw. The volume percent fiber for each test bar containing 15 volume percent glass fibers was measured by a burn-out method. A known weight of sample, wt, was placed in a furnace kept at 600°C for two hours, to burn-off the polyester. The final weight of the glass fiber, wg, was determined. The weights were converted to volumes using the density of the polyester (ρp = 1.17 g/ml) and the density of the glass (ρg = 3.08 g/ml) by Equation (1).
Fiber Volume % =
Vg V p + Vg
=
wg ρ g
(1)
(w t − wg ) ρ g + wg ρ g
where Vp and Vg are the volumes of the polymer and the glass respectively. Flexural properties were determined by the ASTM three-point bending test method D790-66, using an Instron 4204. Optical Transmission Measurements. Optical transmission of the transparent composite was measured relative to air with a UV-Vis-NIR Spectrophotometer from Varian Analytical Instruments at 589.3 nm wavelength. The spectrophotometer was calibrated to 100 percent transmission with no sample in the cell holder. Once calibrated, a polished sample was attached to the cell holder using double stick tape and the transmission measured. The transmission was normalized for 1 cm sample thickness by calculating an absorption coefficient for the sample.
Results and Discussion Evaluation of Glass Fiber Diameter. The diameter of NSK-11 glass fibers produced in a platinum/10% rhodium boat varied with the melt temperature and pull speed. As seen in Figure 1, as the pull speed increased, the fiber diameter decreased and reached a near constant value. The fiber puller can produce fiber diameters of 40-70 µm using a wide range of melt temperatures and pull speeds. As the melt temperature increased, the fiber diameter also increased. The higher temperature lowers the viscosity of the glass, thereby allowing the molten glass to flow through the nozzle more easily. Pull velocity between 1 and 4.5 m/s at 1200°C produced glass fibers with a range of diameters with the same refractive index as that of the polyester resin. Fiber Diameter, (µm)
FABRICATION AND MECHANICAL PROPERTIES ANALYSIS OF OPTICALLY TRANSPARENT POLYESTER/GLASS FIBER MATRIX COMPOSITES
160 140
NSK-11 Glass Fibers
120 100
1150 °C 1200°C 1250 °C
80 60 40 20 0 0
2
4
6
8
10
12
Pull Speed, (m/s) Figure 1. Fiber diameter at various pull speeds and temperatures of 1150, 1200, and 1250°C.
Polymer Preprints 2003, 44(2), 699
Table 1. Fiber Diameter and Refractive Index of NSK-11 Glass Fibers at Different Temperatures Drawn at Different Speeds. Temp. of glass ( °C )
Pull Speed (m / s)
Fiber Dia. (µm)
Ref.Index (nd)
1150 1150 1150 1200 1200 1200 1250 1250 1250
1.13 5.67 11.33 1.13 5.67 11.33 1.13 5.67 11.33
108 36 23 121 39 26 145 41 29
1.552 1.552 1.550 1.554 1.552 1.550 1.554 1.552 1.552
Effects of Volume Percent Fiber on Mechanical Properties. All mechanical properties of composites having 15 volume percent glass fibers were used for comparison between 3TPM and N3TPED. The glass fiber volume percent is not an accurately controlled variable and it can be determined only at the end of the burn–out procedure. Composites with 15.16 volume percent glass fibers coated with 3TPM and 14.77 volume percent glass fibers coated with N3TPED were used for comparison purpose as they were very close to 15 volume percent glass fiber composites. Table 2 shows Young’s modulus increased 2.6 times for composites coated with 3TPM and 1.85 times for composites coated with N3TPED compared to pure polyester composite. The modulus of rupture increased 4.5 times for composites coated with 3TPM and 2.84 times for composites coated with N3TPED compared to pure polyester composite. Work of fracture increased 11 times composites coated with 3TPM and 8 times for composites coated with N3TPED compared to pure polyester composite. Toughness also increased 11 times for composites coated with 3TPM and 8.4 times for composites coated with N3TPED compared to pure polyester composite. Laminar shear strength increased 4.7 times for composites coated with 3TPM and 3.3 times for composites coated with N3TPED compared to pure polyester composite. Thus, all mechanical properties of transparent glass fiber composites increased for 15 volume percent of unidirectionally oriented NSK-11 fibers having a fiber diameter between 44 and 123 µm as compared to pure polyester composite. Also, all mechanical properties tested are better when the NSK-11 glass fibers are coated with 3TPM than with N3TPED. The inability of the amino group in N3TPED to copolymerize with polyester resin may be the primary reason for it to be weaker than composites with glass fibers coated with 3TPM, which showed higher strengths as the methacrylate forms strong covalent bonds with polyester resin during the final polymerization. Table 2. Comparison between 3TPM and N3TPED at 15 vol % fiber. Properties
Polyester
Uncoated Fiber
N3TPED
3TPM
YM
3.0 GPa
5.23 GPa
5.55 GPa
7.89 GPa
MOR
45.3 MPa
115 MPa
129 MPa
207 MPa
WOF
3.76 KJ/m2
14.41 KJ/m2
31.76 KJ/m2
41.9 KJ/m2
0.18 MPa
0.397 MPa
0.524 MPa
3.73 MPa
4.488 MPa
6.486 MPa
Toughness 0.047 MPa LSS
1.36 MPa
Optical Properties. As the volume percent fiber in the composite increased, the optical transmission decreased. Figure 2 shows the curves to start below the expected 90 percent transmission measured for pure polyester. Optical transmission decreased with increasing volume percent fiber and decreasing diameter. Each glass fiber in a transparent composite can potentially scatter a portion of the incident light. Scattering of the incident light increases as the number of fibers increases, thereby causing the objects to be indistinct when viewed through the composite. Thus, as the number of glass fibers increased, optical interfaces in the composite also increased and there was a reduction in optical transmission. The covalent bonding of the glass fiber is important to light transmission in that silane coupling agents can prevent voids at the matrix fiber interface. The voids reduce optical transmission. For composites containing less than one volume percent fiber content, the optical transmission was very high and comparable to 90 percent transmission for pure polyester matrix. This indicated that the dip-coating procedure for applying the silane-coupling agents was useful in eliminating any unwet fibers, which are sites for air gaps and voids. 100
Percent Transmission at 589.3 nm
Evaluation of Refractive Index. Due to the differences in the thermal history (quench rate) of the fiber, the refractive index of the NSK-11 glass varies with the fiber diameter.6 Table 1 lists the refractive index of glass fibers with various fiber diameters. The refractive index of NSK-11 glass decreases from 1.56388 for bulk annealed glass to 1.554 for glass fibers from 40-123 µm at 1200°C.
Composites containing unidirectional NSK-11 glass fibers in polyester
90 80
Group I : 121 or 123 µm Fibers Group II : 70 or 71 µm Fibers Group III : 54 µm Fibers Group IV : 44 µm Fibers Pure Polyester : 90 % Transmission
70 60
I
50 40
IV
III
30
II
20 10 0 0
5
10
15
20
25
30
Volume Percent Fiber
Figure 2. Percent optical transmission vs. volume percent fiber for composites using either 3TPM or N3TPED as a coupling agent.
Conclusions Composites of polyester and NSK-11 glass fibers fabricated by this procedure produced higher mechanical properties than pure polyester. The refractive index of the NSK-11 glass fiber increased with the increase in fiber diameter. Glass fiber diameter decreased with increase in the pull speed of the rotating drum. As the melt temperature was increased, the fiber diameter also increased. Mechanical properties like Young’s modulus, modulus of rupture, work of fracture, toughness and laminar shear strength for composites at 15 volume percent glass fibers increased 2.6, 4.5, 11, 11 and 4.7 times respectively that of pure polyester when coated with 3TPM. It increased 1.85, 2.84, 8, 8.4 and 3.3 times respectively that of pure polyester when coated with N3TPED. As is observed in the conclusions above, all mechanical properties tested are better when the NSK-11 glass fibers are coated with 3TPM than with N3TPED. This is because the methacrylate group in 3TPM readily copolymerizes with polyester to form strong covalent bonding with the polyester resulting in stronger composites that have higher mechanical properties. On the other hand amino-silane exhibits a stronger interaction with the glass surface than either the methacryloxy or the epoxy groups and does not have any interactions with polyester, and hence the glass fibers coated with N3TPED have a weaker bond with polyester. Optical transmission decreased with increasing volume percent fiber and decreasing diameter.
References (1) Christiansen, C. Ann. Phys. Chem. 1884, 23(10), 298. (2) Watt, W. Handbook of Composites, Vol.1- Strong Fibres, 1985, 327, Elsevier Science Publishers, New York, N.Y. (3) Arkles, B. Silicon Compounds Register and Review, 1987, 54, Petrarch Systems Inc., Bristol, P.A. (4) Plueddemann, E. P. J. Adhesion, 1970, 2, 184. (5) Daley, L. R.; Rodriguez, F. Polym. Eng. Sci. 1969, 9(6), 428. (6) Olson, J. R.; Day, D. E.; Stoffer, J. O. J Compos. Mater. 1992, 26(8), 1181.
Polymer Preprints 2003, 44(2), 700
Department of Chemical Engineering, University of Missouri-Rolla, Rolla, MO 65401. 2Chemistry and Materials Research Center, University of Missouri-Rolla, Rolla, MO 65401.
Introduction Severe windstorms such as hurricanes result in extensive damage to building envelopes and the interior due to the effects of flying debris and fluctuating wind pressures. Developing and manufacturing a new transparent laminated glass window would greatly reduce the breakage of the glass panels under adverse conditions. A high strength, low weight composite would be suitable for this purpose. Although there are many reinforcement and matrix materials available to produce such composites, the choices for selecting these materials are limited since optical transparency is vital. This work used the Christiansen method to reduce the loss of transparency in composites by matching the refractive indices of the reinforcement and matrix phases to eliminate light scattering and reflection.1 This paper focuses on the criteria necessary to produce and characterize an improved polyester/glass fiber transparent composite containing 15 volume percent glass fibers. The Young’s modulus, modulus of rupture, work of fracture, toughness and laminar shear strength are measured for composites using two types of silane agents. The relationship of optical transmission of the polyester composite with respect to volume percent fiber and fiber diameter were also determined.
Experimental Procedures Materials. The unsaturated polyester resin (Part 811-191) and a peroxide mixture (Part 811-194) were purchased from the Leco Corporation. The coupling agents were 3-(trimethoxysilyl)propyl methacrylate (3TPM) and N-( 3-[trimethoxysilyl]propyl )- ethylenediamine (N3TPED) from Aldrich Chemical Co. The refractive index of the polyester is 1.554 ± 0.001 at the sodium line. The refractive index of the glass fibers is smaller than that of the bulk glass because their rapid cooling reduces the glass density.2 Thus, larger diameter fibers have a higher refractive index than smaller diameter fibers since the larger diameter fibers cool more slowly and have a higher density. Drawing bulk glass into fiber or ribbon lowered the refractive index of the glass by approximately 0.006, so a bulk glass with a refractive index of 1.560 was desired. NSK-11 glass (Schott Glass Technologies, Inc.) has a refractive index of 1.56388, which is suitable for this work. This glass was tested to determine the conditions where its refractive index dropped to 1.554. Glass Fiber Production. A fiber puller was constructed at the Materials Research Center at the University of Missouri-Rolla to be used for glass fiber production. Glass fragments were placed in a electrically heated platinum container fitted with a platinum/10% rhodium bushing. The temperature ranged between 1150 and 1300°C, causing the glass to melt and continuous fibers were pulled and wound on a rapidly spinning drum. Glass fibers with a diameter between 20 and 150 µm can be produces. The fibers were then chopped by hand with scissors to the desired length. A digital micrometer (Mitutoyo Corporation) measured the glass fiber diameter, which had a maximum error of just ± 1 µm. The Becke line method was used to measure the refractive index of the glass fibers and polymer matrix. Composite Preparation and Testing. Unidirectionally oriented glass fiber reinforced polyester composite was prepared. A dip coat procedure was used to apply a coating to the glass fibers. The dip coat procedure removes the by-products of the coupling reaction from the interface and produces composites with quality optical surfaces. A 600 ml solution of 95 percent (w/v) ethanol in water was adjusted to pH 4.5-5.5 with acetic acid3,4 3TPM was added (12.5 ml) with stirring to yield a 2% final weight concentration. Five minutes were allowed for hydrolysis and silanol formation. Glass fibers of 44, 54, 70, 71, 121 and 123µm diameters were dipped briefly in ethanol. The glass fibers were rinsed for fifteen minutes with ethanol. The curing of the silane layer took 24 hours at room temperature (< 60 % relative humidity). The same procedure was repeated using N3TPED as a coupling agent. The water absorbed on the fiber surface hydrolyzes the trimethoxy silane to the
hydroxy species, which can bond to the glass surface.5 Most silaceous substrates have 4-12 silanols per mµ2 . Thus, an average of 7500m2 should be covered by one mole of evenly distributed silane.4 Approximately 3-8 monolayers (< 10 nm) of trimethoxy silane deposits onto the fiber surface using this dip coat technique. The ethanol rinse washes the methanol and water by-products from the interface. The trimethoxy silane bonds to the adjacent silane molecules and to the glass surface resulting in a polymeric coating. After rinsing and curing the glass fibers for at least 24 hours, the coated glass fibers were placed in a box-like steel mold. The inner dimensions of the mold were 170 mm x 61 mm x 17 mm. The interior vertical sides of the mold were lined with Teflon® (Teflon is a registered trademark of Dupont). Polyester resin and the peroxide initiator were then poured into the prepared mold. The mold was degassed (8-10 inch Hg) for 45 minutes while simultaneously heating at 60°C to remove dissolved gases in the resin and gases adsorbed on the fibers. All composites used in this paper were tested with glass fibers placed at the bottom of the composites. After polymerization, all composites were removed from the mold and heated at 50°C for four hours to ensure complete polymerization and that all samples had a uniform thermal history. The composites were surface machined to reduce the sample thickness to 5 mm. Individual mechanical test bars 120 mm in length by 17 mm wide were cut from the full sample with a band saw. The volume percent fiber for each test bar containing 15 volume percent glass fibers was measured by a burn-out method. A known weight of sample, wt, was placed in a furnace kept at 600°C for two hours, to burn-off the polyester. The final weight of the glass fiber, wg, was determined. The weights were converted to volumes using the density of the polyester (ρp = 1.17 g/ml) and the density of the glass (ρg = 3.08 g/ml) by Equation (1).
Fiber Volume % =
Vg V p + Vg
=
wg ρ g
(1)
(w t − wg ) ρ g + wg ρ g
where Vp and Vg are the volumes of the polymer and the glass respectively. Flexural properties were determined by the ASTM three-point bending test method D790-66, using an Instron 4204. Optical Transmission Measurements. Optical transmission of the transparent composite was measured relative to air with a UV-Vis-NIR Spectrophotometer from Varian Analytical Instruments at 589.3 nm wavelength. The spectrophotometer was calibrated to 100 percent transmission with no sample in the cell holder. Once calibrated, a polished sample was attached to the cell holder using double stick tape and the transmission measured. The transmission was normalized for 1 cm sample thickness by calculating an absorption coefficient for the sample.
Results and Discussion Evaluation of Glass Fiber Diameter. The diameter of NSK-11 glass fibers produced in a platinum/10% rhodium boat varied with the melt temperature and pull speed. As seen in Figure 1, as the pull speed increased, the fiber diameter decreased and reached a near constant value. The fiber puller can produce fiber diameters of 40-70 µm using a wide range of melt temperatures and pull speeds. As the melt temperature increased, the fiber diameter also increased. The higher temperature lowers the viscosity of the glass, thereby allowing the molten glass to flow through the nozzle more easily. Pull velocity between 1 and 4.5 m/s at 1200°C produced glass fibers with a range of diameters with the same refractive index as that of the polyester resin. Fiber Diameter, (µm)
FABRICATION AND MECHANICAL PROPERTIES ANALYSIS OF OPTICALLY TRANSPARENT POLYESTER/GLASS FIBER MATRIX COMPOSITES
160 140
NSK-11 Glass Fibers
120 100
1150 °C 1200°C 1250 °C
80 60 40 20 0 0
2
4
6
8
10
12
Pull Speed, (m/s) Figure 1. Fiber diameter at various pull speeds and temperatures of 1150, 1200, and 1250°C.
Polymer Preprints 2003, 44(2), 699
Table 1. Fiber Diameter and Refractive Index of NSK-11 Glass Fibers at Different Temperatures Drawn at Different Speeds. Temp. of glass ( °C )
Pull Speed (m / s)
Fiber Dia. (µm)
Ref.Index (nd)
1150 1150 1150 1200 1200 1200 1250 1250 1250
1.13 5.67 11.33 1.13 5.67 11.33 1.13 5.67 11.33
108 36 23 121 39 26 145 41 29
1.552 1.552 1.550 1.554 1.552 1.550 1.554 1.552 1.552
Effects of Volume Percent Fiber on Mechanical Properties. All mechanical properties of composites having 15 volume percent glass fibers were used for comparison between 3TPM and N3TPED. The glass fiber volume percent is not an accurately controlled variable and it can be determined only at the end of the burn–out procedure. Composites with 15.16 volume percent glass fibers coated with 3TPM and 14.77 volume percent glass fibers coated with N3TPED were used for comparison purpose as they were very close to 15 volume percent glass fiber composites. Table 2 shows Young’s modulus increased 2.6 times for composites coated with 3TPM and 1.85 times for composites coated with N3TPED compared to pure polyester composite. The modulus of rupture increased 4.5 times for composites coated with 3TPM and 2.84 times for composites coated with N3TPED compared to pure polyester composite. Work of fracture increased 11 times composites coated with 3TPM and 8 times for composites coated with N3TPED compared to pure polyester composite. Toughness also increased 11 times for composites coated with 3TPM and 8.4 times for composites coated with N3TPED compared to pure polyester composite. Laminar shear strength increased 4.7 times for composites coated with 3TPM and 3.3 times for composites coated with N3TPED compared to pure polyester composite. Thus, all mechanical properties of transparent glass fiber composites increased for 15 volume percent of unidirectionally oriented NSK-11 fibers having a fiber diameter between 44 and 123 µm as compared to pure polyester composite. Also, all mechanical properties tested are better when the NSK-11 glass fibers are coated with 3TPM than with N3TPED. The inability of the amino group in N3TPED to copolymerize with polyester resin may be the primary reason for it to be weaker than composites with glass fibers coated with 3TPM, which showed higher strengths as the methacrylate forms strong covalent bonds with polyester resin during the final polymerization. Table 2. Comparison between 3TPM and N3TPED at 15 vol % fiber. Properties
Polyester
Uncoated Fiber
N3TPED
3TPM
YM
3.0 GPa
5.23 GPa
5.55 GPa
7.89 GPa
MOR
45.3 MPa
115 MPa
129 MPa
207 MPa
WOF
3.76 KJ/m2
14.41 KJ/m2
31.76 KJ/m2
41.9 KJ/m2
0.18 MPa
0.397 MPa
0.524 MPa
3.73 MPa
4.488 MPa
6.486 MPa
Toughness 0.047 MPa LSS
1.36 MPa
Optical Properties. As the volume percent fiber in the composite increased, the optical transmission decreased. Figure 2 shows the curves to start below the expected 90 percent transmission measured for pure polyester. Optical transmission decreased with increasing volume percent fiber and decreasing diameter. Each glass fiber in a transparent composite can potentially scatter a portion of the incident light. Scattering of the incident light increases as the number of fibers increases, thereby causing the objects to be indistinct when viewed through the composite. Thus, as the number of glass fibers increased, optical interfaces in the composite also increased and there was a reduction in optical transmission. The covalent bonding of the glass fiber is important to light transmission in that silane coupling agents can prevent voids at the matrix fiber interface. The voids reduce optical transmission. For composites containing less than one volume percent fiber content, the optical transmission was very high and comparable to 90 percent transmission for pure polyester matrix. This indicated that the dip-coating procedure for applying the silane-coupling agents was useful in eliminating any unwet fibers, which are sites for air gaps and voids. 100
Percent Transmission at 589.3 nm
Evaluation of Refractive Index. Due to the differences in the thermal history (quench rate) of the fiber, the refractive index of the NSK-11 glass varies with the fiber diameter.6 Table 1 lists the refractive index of glass fibers with various fiber diameters. The refractive index of NSK-11 glass decreases from 1.56388 for bulk annealed glass to 1.554 for glass fibers from 40-123 µm at 1200°C.
Composites containing unidirectional NSK-11 glass fibers in polyester
90 80
Group I : 121 or 123 µm Fibers Group II : 70 or 71 µm Fibers Group III : 54 µm Fibers Group IV : 44 µm Fibers Pure Polyester : 90 % Transmission
70 60
I
50 40
IV
III
30
II
20 10 0 0
5
10
15
20
25
30
Volume Percent Fiber
Figure 2. Percent optical transmission vs. volume percent fiber for composites using either 3TPM or N3TPED as a coupling agent.
Conclusions Composites of polyester and NSK-11 glass fibers fabricated by this procedure produced higher mechanical properties than pure polyester. The refractive index of the NSK-11 glass fiber increased with the increase in fiber diameter. Glass fiber diameter decreased with increase in the pull speed of the rotating drum. As the melt temperature was increased, the fiber diameter also increased. Mechanical properties like Young’s modulus, modulus of rupture, work of fracture, toughness and laminar shear strength for composites at 15 volume percent glass fibers increased 2.6, 4.5, 11, 11 and 4.7 times respectively that of pure polyester when coated with 3TPM. It increased 1.85, 2.84, 8, 8.4 and 3.3 times respectively that of pure polyester when coated with N3TPED. As is observed in the conclusions above, all mechanical properties tested are better when the NSK-11 glass fibers are coated with 3TPM than with N3TPED. This is because the methacrylate group in 3TPM readily copolymerizes with polyester to form strong covalent bonding with the polyester resulting in stronger composites that have higher mechanical properties. On the other hand amino-silane exhibits a stronger interaction with the glass surface than either the methacryloxy or the epoxy groups and does not have any interactions with polyester, and hence the glass fibers coated with N3TPED have a weaker bond with polyester. Optical transmission decreased with increasing volume percent fiber and decreasing diameter.
References (1) Christiansen, C. Ann. Phys. Chem. 1884, 23(10), 298. (2) Watt, W. Handbook of Composites, Vol.1- Strong Fibres, 1985, 327, Elsevier Science Publishers, New York, N.Y. (3) Arkles, B. Silicon Compounds Register and Review, 1987, 54, Petrarch Systems Inc., Bristol, P.A. (4) Plueddemann, E. P. J. Adhesion, 1970, 2, 184. (5) Daley, L. R.; Rodriguez, F. Polym. Eng. Sci. 1969, 9(6), 428. (6) Olson, J. R.; Day, D. E.; Stoffer, J. O. J Compos. Mater. 1992, 26(8), 1181.
Polymer Preprints 2003, 44(2), 700
