Examination of the Thermo-mechanical Properties of E-Glass / Carbon Composites

Eight-ply E-glass, carbon and E-glass/carbon fabric-reinforced polyester based hybrid composites were manufactured in this study. A vacuum infusion system was used as the production method. Dynamic mechanical analysis, thermogravimetric analysis and diff erential scanning calorimetry analysis were conducted to examine the thermo-mechanical properties of composite samples. The eff ect of reinforcement type and different stacking sequences of fabric plies on the thermo-mechanical properties of composite samples were also investigated. Results showed that the type and alignment of reinforcement material has a signifi cant eff ect on the dynamic mechanical properties of composite samples.


Introduction
Fibre-reinforced composites are one of the most favoured materials in many sectors owing to their properties, such as high speci c strength and stiness, low thermal expansion, good fatigue performance and processing advantages at a low cost [1].Hybrid composites include two or more types of reinforcement materials.
ey possess much better performance characteristics than traditional compo-sites, and are designed for speci c applications.By hybridising low-modulus, high-elongation bres (e.g.E-glass) with high-modulus, low-elongationbres (e.g.carbon), more damage-tolerant and coste ective advanced composite materials can be manufactured [2,3].E-glass and carbon bres are the most preferred textile bres in composite structures.Due to the ease of orientation/manufacture and their excellent mechanical properties, woven carbon fabric-reinforced composites can be used in many applications [4].

Izvleček
Osemslojni hibridni kompoziti iz poliestra, ojačeni s tkaninami iz E-stekla, ogljika in kombinacije E-steklo/ogljik, so bili v tej študiji izdelani s pomočjo metode vakuumske infuzije.Termomehanske lastnosti kompozitnih vzorcev so bile proučevane z dinamično mehansko in termogravimetrično analizo in dinamično kalorimetrijo.Raziskani so bili tudi učinki vrste ojačitve in različne razporeditve zlaganja tkanin na termomehanske lastnosti kompozitnih vzorcev.Rezultati so pokazali, da vrsta in razporeditev ojačitvenega materiala pomembno vplivata na dinamične mehanske lastnosti kompozitnih vzorcev.Ključne besede: tekstilni kompoziti, dinamična mehanska analiza, termogravimetrična analiza, dinamična kalorimetrija e advantageous properties of E-glass bres that make them preferable materials for composite structures are their high damping characteristics, lightweight, low cost, high strength, good resistance to corrosion and excellent insulation properties [5][6][7][8].Dynamic mechanical analysis (DMA) is a technique that measures both the mechanical and viscoelastic properties of materials [9].DMA is a method that applies sinusoidal force to samples to study their viscoelastic properties and structures.It measures the modulus and damping properties of materials [10][11][12].Fibre-reinforced composites are exposed to di erent types of dynamic stresses during handling, which demonstrates the importance of analysing the viscoelastic properties of composite materials [13].Di erential scanning calorimetry (DSC) is a thermal analysis method that measures heat ow rate as a function of time and temperature [14].By using a reference material, it computes the required amount of heat to make the temperature di erence between sample and reference material zero. is heat provides information regarding the physical and chemical transformations of material [15,16].
ermogravimetric analysis (TGA) shows the degradation temperature and amount of weight materials lose as temperature is increased [14,16].
e thermo-mechanical properties of E-glass/carbon fabric-reinforced composite samples were determined in this study using DMA, TGA and DSC analysis methods, while the e ect of reinforcement type and di erent stacking sequences of fabric plies on the thermo-mechanical properties of composite samples were also investigated.

Materials
E-glass and carbon fabrics were used as reinforcement material, while unsaturated polyester resin was used as matrix material.Fabric parameters are given in Table 1.e matrix system included an accelerator (cobalt) and hardener (methyl ethyl ketone peroxide-MEKP), together with polyester resin.e ratio of polyester:cobalt:MEKP was 1:0.00175:0.002by weight.

Methods
A vacuum infusion system was used as the production method for composite laminates (Figure 1).Fabrication was carried out at the room temperature (20 °C ±2 °C).Four composite samples, one of which had eight plies of E-glass fabric (GGGGGGGG), one of which had eight plies of carbon fabric (CCCCCCCC) and two of which had both E-glass and carbon fabrics with di erent stacking sequences (GGCCCCGG, CCGGGGCC), were manufactured.A CNC milling machine was used for cutting the samples to the required dimensions.An RMI DX04T dynamic mechanical analyser with a three-point bending con guration was used for the dynamic mechanical analysis of composite samples.Samples measuring 10 mm x 50 mm were cut from composite plates.Tests were performed at a frequency of 1 Hz, while temperature programs were run from 30 to 150 °C under a controlled sinusoidal strain, at a heating rate of 3 °C/min.A Mettler Toledo TGA/SDTA851e analyser was used for thermogravimetric analysis.Samples weighing between 6 and 7 mg were heated from 25 °C to 600 °C with a heating rate of 5 °C/min in a nitrogen atmosphere.3 Results and discussion

Dynamic mechanical analysis
Figure 2 shows the variation of storage modulus results of the samples as a function of temperature.
Storage modulus demonstrates the elastic behaviour of a material and is proportional to the energy stored in one cycle [17].It was shown that the storage modulus values of composites are found to decrease as temperature increases as the result of a loss in the sti ness of bres at high temperature [9].e E-glass-reinforced composite sample and the hybrid sample with E-glass fabric plies at the outer layers demonstrated the highest storage modulus values.It was shown while studying the storage modulus values of samples at 30 °C that while the storage modulus values of GGGGGGGG and GGCCCCGG samples were approximately 14 GPa, the storage modulus values of CCCCCCCC and CCGGGGCC samples were between 6 and 8 GPa.Although there was a decrease in the storage modulus values of all samples as temperature was increased, GGGGGGGG and GGCCCCGG samples maintained higher storage modulus values (approximately 4 GPa at between 100 and 150 °C) than the other samples (approximately 1 GPa at between 100 and 150 °C) at higher temperatures.Figure 3 shows the loss modulus results of the samples.Loss modulus demonstrates the viscous behaviour of a material and is proportional to the energy dissipated in one cycle [17].Similar to storage modulus results, it was shown that the E-glass-reinforced composite sample and the hybrid sample with E-glass fabric plies at the outer layers had the highest loss modulus values (2 GPa).Moreover, it was shown that the peak point of the loss modulus curve of the carbon fabric-reinforced sample also reached 2 GPa, but at a lower temperature.of the of the lines show the glass transition temperatures (T g ) of the materials [17].We observed that GGGGGGGG and GGCCCCGG samples had higher T g values than the other samples.It is a well-known fact that E-glass is very durable to heat.us, by reinforcing polyester resin with Eglass bre, the temperature point at which the composite changes from a glassy state to a rubbery state becomes higher than that of the carbon brereinforced sample.

Thermogravimetric analysis
TGA results (weight loss and onset temperature) are given in Table 2.It can be seen from Table 2 that the onset temperatures of all samples were approximately the same (around 310 °C).It is a wellknown fact that E-glass and carbon bres begin to degrade at 850 °C and 300 °C, respectively [18].Because the samples were heated to 600 °C, we can conclude that E-glass bre does not begin to degrade during the measurement process.It can be observed, however, from TGA results that the degradation temperature of the E-glass-reinforced sample was also around 310 °C. is fact shows that polyester resin also begins to degrade at about same temperature.When the weight losses of composite samples were taken into account, it was shown that hybrid samples had lower weight losses than the other samples.It can thus be concluded that hybrid samples are more durable than the other samples.

Diff erential Scanning Calorimetry
e results of the DSC analysis of samples are given in Figure 5.Samples were heated from 25 °C to 200 °C.It was shown that none of the samples demonstrated any endothermic or exothermic reactions between these temperatures.is can be explained by the thermal characteristics of matrix and reinforcement materials.e polyester that was used in this study is a thermoset polyester that does not have a melting point.Moreover, as previously mentioned, E-glass and carbon bres are very durable to heat.Although it cannot be seen precisely from the above gure, the glass transition temperatures of samples were obtained from DSC graphs.Table 3 shows the glass transition temperatures of samples that were

Conclusıon
In conclusion, the DSC analysis indicated that composite samples did not demonstrate any phase changes until 200 °C.Moreover, the TGA results demonstrated that hybrid samples had a lower weight loss than the other samples and that all samples began to degrade at around 310 °C.We also observed that the stacking sequence had a signi cant e ect on the dynamic mechanical properties of E-glass/carbon hybrid polyester composites.e E-glass-reinforced sample and hybrid sample with E-glass fabric at the outer layers had higher modulus values than the other samples.We can conclude that E-glass fabric-reinforced samples displayed higher viscosity and elasticity than the carbon fabric-reinforced samples.In addition, higher viscosity and elasticity in hybrid structures were obtained by placing E-glass fabric plies at the outer layers of the structure.

Figure 2 :
Figure 2: Storage modulus results of samples

Figure 5 :
Figure 5: Di erential scanning calorimetry results of samples

Table 1 :
Fabric parameters Perkin-Elmer di erential scanning calorimeter was used for di erential scanning calorimetry.Samples weighing approximately 8 mg were heated from 25 °C to 200 °C at a heating rate of 10 °C/min, with a pause of one minute at 25 °C.ey were then cooled to 25 °C at a rate of 10 °C/min.

Table 2 :
ermogravimetric analysis results of samples

Table 3 :
Glass transition temperatures of samples from loss modulus, tangent delta and di erential scanning calorimetry curves values were achieved with the eight-ply E-glassreinforced samples.It was also seen that the T g values obtained from tan delta curves were higher than other values. g