Mechanical,thermo-mechanical and thermal characteristics of multi-walled carbon nanotubes-added textile-reinforced composites

Abstract

The influence of adding different types of multi-walled carbon nanotubes on mechanical (tensile strength and impact strength), thermo-mechanical (storage modulus, loss modulus and damping factor) and thermal properties (thermogravimetric and differential scanning calorimetry analysis) of fabric-reinforced polyester-based composite structures are analyzed in this study. Jute, E-glass and carbon fabrics are preferred as the reinforcement materials. Four-plied fabric-reinforced composites are fabricated using vacuum-assisted resin transfer molding technique. Results indicate that adding different types of multi-walled carbon nanotubes have increasing effect on mechanical and thermo-mechanical properties of composite structures; however, they have barely effect on thermal properties. Pristine multi-walled carbon nanotube-added specimens show higher mechanical and thermo-mechanical properties compared to functionalized multi-walled carbon nanotube-added samples.

Keywords

Textiles carbon nanotubes mechanical properties thermal properties

Introduction

Polymer matrix composites are mostly reinforced with textile structures. The utilization of textile materials in composite structures is increasing day by day, thanks to their relative low costs and high performances [1 –4]. The first example of fiber-reinforced composite was created after Second World War by embedding glass fibers in polymeric resin for petrochemical industry [5].

In composite structures, textile materials are used in different forms such as unidirectional continuous fibers, bidirectional continuous fibers, multidirectional continuous fibers, unidirectional discontinuous fibers and random discontinuous fibers [6,7]. Fabric-reinforced composites have superior properties than the fiber-reinforced composites due to their orderly alignment, high load-bearing capacity and good entirety concept [8].

The properties of the composite material such as strength and stiffness are mostly associated with the reinforcement material, while corrosion, temperature and chemical resistance of composite materials are directly related to the properties of the polymer matrix material. However, most of the mechanical properties of the composites are provided by both the reinforcement and the matrix material [9].

Due to the fact that interfacial strength is the primary factor that influences the mechanical properties, the properties of composites are mostly enhanced by interfacial treatments [10]. Incorporation of filler materials and fiber surface treatments are the foremost interfacial treatment methods [11].

Nano-fillers are privileged filler materials that are used to develop the properties of composite structures. They have very high specific surface areas (more than 1000 m²/g) and aspect ratios compared to micro-scaled fillers [12]. Nanoclays, nanofibers and nanotubes are the forms of nano-fillers that are used in composite industry [13]. Carbon nanotubes (CNTs) are upcoming nanoparticles used in composite industry to enhance their properties [14 –19]. They belong to a specific material group that shows an excellent combination of mechanical, electrical and thermal properties that no other material group has showed until now [14,20]. They have two times higher thermal conductivity than diamond, and 1000 times higher electric current-carrying capacity than copper. Their strength is 10 to 100 times greater than steel, their thermal stability is up to 2800℃ and they have elastic modulus more than 1 TPa [21]. The major properties that are improved by addition of very few amounts of CNTs to composite structures are electrical conductivity, thermo-mechanical properties, tensile strength, stiffness, fracture toughness, elastic modulus and impact strength [11,12,17,19,22 –24].

Zhu et al. [19] were the first who integrated single-wall carbon nanotubes (SWCNT) to glass fiber-reinforced vinyl ester composites and analyzed its effect on the interlaminar shear strength on the composite. Glass fabrics were coated with SWCNTs, and vacuum assisted resin transfer molding (VARTM) process was used to manufacture composite structures. It was concluded that 45% increase of shear strength was procured by coating midplane ply of composite with 0.015 wt.% SWCNT. Zhao et al. [25] added CNTs to carbon fiber-reinforced epoxy composites by two methods (mixing directly with matrix and attaching on the fiber) to compare the effects of two different methods. VARTM method was used. It was observed that matrix-treated samples have 10.41%, 10.22% and 15.14% higher tensile strength, flexural strength and flexural moduli compared to carbon fiber-reinforced epoxy sample, while the fabric-treated samples have 24.42%, 18.43% and 27.01% higher tensile strength, flexural strength and flexural moduli compared to carbon fiber-reinforced epoxy sample, respectively. Besides, fabric-treated samples reached slightly higher T_g values than matrix-treated samples. Zhang et al. [26] designed and fabricated glass fabric-reinforced epoxy-based composite structures for sport equipment. They added multi-walled carbon nanotubes (MWCNTs) (0.4 wt.%, 0.75 wt.% and 1.1 wt.%) to the resin and examined their effects on the mechanical properties. 0.75 wt.% MWCNT-added sample had higher tensile fracture work with fewer damages, whereas 0.4 wt.% MWCNT-added sample had the highest flexural strength. Shokrieh et al. [23] added different amounts of MWCNTs (0.05, 0.1 and 0.5 wt.%) to unsaturated polyester resin to investigate their effects on mechanical properties. It was noticed from the results that better mechanical properties were achieved with 0.05 wt.% MWCNT-added sample. The tensile and flexural strengths of 0.05 wt.% MWCNT-added sample increased 6% and 20%, respectively. Besides, scanning electron microscope (SEM) images indicated that when the amount of MWCNT increases to 0.05 wt.%, a large amount of agglomeration was procured. Shen et al. [18] incorporated CNTs (up to 4 wt.%) into glass fabric-reinforced polyamide 6 composite and investigated its influence on mechanical and thermal properties. Hot press technique was used. It was understood that incorporation of CNTs up to 2 wt.% has a positive effect on the thermal stability, thermal conductivity and mechanical features of the sample. The flexural stress was improved up to 17%, 32% and 36% by incorporation of 0.5 wt.%, 1 wt.% and 2 wt.% CNTs, respectively. In another study, Seyhan et al. [17] added CNTs to the polyester resin to examine their influence on the viscosity and the tensile strength of the composite material. They utilized four different CNTs (MWCNT, MWCNT-NH₂, DWCNT, DWCNT-NH₂) with three different amounts (0.1 wt.%, 0.3 wt.% and 0.5 wt.%). It was seen that NH₂ functionalized CNT-added samples displayed higher tensile strength than the others. In addition, the highest tensile strengths were achieved with 0.5 wt.% CNT-added samples. Qiu et al. [27] integrated MWCNTs to the epoxy resin to enhance the mechanical properties of glass fabric-reinforced epoxy composites. Pristine and strong acid functionalized MWCNTs were used. VARTM technique was utilized as the composite production method. The tensile strength, Young’s modulus, shear strength and short beam modulus of composite samples improved by 14%, 20%, 5% and 8%, respectively, with 1 wt.% loading of functionalized MWCNTs. In another study, Fan et al. [28] analyzed the interlaminar shear strength of glass fiber-reinforced MWCNT-treated epoxy composites. Three different amounts of MWCNTs (0.5 wt.%, 1 wt.% and 2 wt.%) were added to epoxy resin by high-speed mechanical stirring, ultrasonic agitation and acid oxidation methods. They used traditional VARTM technique for production of samples that include 0.5 wt.% MWCNT and a novel injection and double vacuum-assisted resin transfer molding method for manufacturing of other samples, as the insufficient permeability across the fiber layers causes VARTM method to break down with concentrations more than 0.5 wt.%. It was concluded that interlaminar shear strength (ILLS) increased by 9.7%, 20.5% and 33.1% with addition of 0.5, 1 and 2 wt.% of MWCNT to the composite samples, respectively. Bekyarova et al. [29] deposited MWCNTs and SWCNTs to the surface of carbon fabric by electrophoresis method to fabricate composite structures with enhanced interlaminar shear strengths. VARTM method was preferred to produce epoxy-based composites. Results demonstrated that CNT integrated samples reached 30% higher interlaminar shear strengths than untreated sample.

After the examination of the preliminary studies on MWCNT-reinforced textile-based composites, it is seen that there are not many studies utilizing polyester resin as the matrix material. It is also noteworthy that in general, pristine MWCNTs are used instead of functionalized MWCNTs and when it comes to composites with functional group-treated MWCNTs, the studies using MWCNT-OH and MWCNT-COOH are rarely found. In this study, three types of MWCNTs are added to the fabric-reinforced polyester-based composites to investigate their effects on mechanical, thermo-mechanical and thermal properties of textile-reinforced composites.

Experimental study

Materials

Jute (supplied by Ege Izmir Cuval, Izmir, Turkey), E-glass (supplied by Omnis Kompozit, Istanbul, Turkey) and carbon (supplied by Spinteks, Denizli, Turkey) plain weave fabrics which have basis weights of 200 g/m² are used as reinforcement materials, while unsaturated orthophthalic polyester resin (Poliya), cobalt and methyl ethyl ketone peroxide are utilized as matrix system. Fabric parameters are given in Table 1.

Table 1.

Fabric parameters.

	Jute	E-glass	Carbon
Thickness (±SD) (mm)	0.99 (±0.10)	0.41 (±0.01)	0.29 (±0.005)
Fabric count (e.p.c. × p.p.c)	4 × 4	4 × 3	5 × 5
Warp yarn count (±SD) (Tex)	375 (±36.3)	288 (±10.3)	206 (±6.2)
Weft yarn count (±SD) (Tex)	234 (±25.8)	409 (±11.5)	203 (±5.9)

Three different types of MWCNTs (MWCNT, MWCNT-OH and MWCNT-COOH) supplied by US Research Nanomaterials are added to the matrix material. They have purity of 95%, their wall length is 10–30 µm and the average outside diameter is less than 7 nm.

Methods

Resin preparation and composite manufacturing

It is stated in the literature that agglomeration of CNTs increases with ascending amount of CNT [23]. In this study, 0.5 g of MWCNT is added to 1 kg of polyester resin (0.05%). A very few amount of CNT is added to the resin to investigate the effect of such few amounts of CNTs on mechanical properties of composite materials and also to minimize the agglomeration. Ultrasonic mixer is used to mix polyester resin with CNTs.

Vacuum-assisted resin transfer molding technique is used for the preparation of composite samples. Before starting the process, a releasing agent is applied to the surface. Then fabric plies are placed and then peel ply and perforated film are laid on the fabric layers. Finally, the vacuum sealant tape is pasted around the sample and a vacuum bag is attached on it. A small hole is also opened on the vacuum bag to enable the manifold tube to pass through it. Then, the vacuum pump is opened to suck the resin (Figure 1(a)).

Figure 1.

(a) VARTM system and (b) JJJJ sample in CNC milling machine.

Using these materials and methods, 12 composite specimens (40 × 50 cm) are produced. Each sample comprises four plies of fabric reinforcement. J, G and C represent jute, E-glass and carbon fabric plies, respectively.

Mechanical analysis

Shimadzu AG-IS test machine (Figure 2(a)) is utilized to evaluate the tensile strength of the composite samples according to ASTM D638-10 standard. A load cell of 10 kN at cross-head speed of 6 mm/min is used. Impact testing is performed by Devotrans Charpy Impact Test Machine (Figure 2(b)) according to BS EN ISO 179:1997 (ISO 179/1 eA) standard. Before testing, a notch is opened at the middle of each sample by a notching apparatus.

Figure 2.

(a) Tensile strength and (b) impact strength tester.

For the mechanical analysis, samples are cut from both warp and weft directions of the composite structure by the CNC milling machine (Figure 1(b)). For each test, three warp and three weft specimens are prepared.

Thermo-mechanical and thermal analysis

Dynamic mechanical analysis, thermogravimetric analysis and differential scanning calorimetry analysis are used to analyze the thermo-mechanical and other thermal properties of polyester matrix and composite structures. RMI DX04T dynamic mechanical analyzer with a three-point bending configuration is used for the dynamic mechanical analysis of composite samples. Samples measuring 10 mm × 50 mm are cut from warp and weft directions of composite plates. Tests are performed at a frequency of 1 Hz, while temperature program runs from 30℃ to 150℃ under a controlled sinusoidal strain, at a heating rate of 3℃/min. Mettler Toledo TGA/SDTA851e analyzer is used to conduct the thermogravimetric analysis. Samples that have weight between 6 and 7 mg are cut and put into aluminum oxide crucibles. Samples are heated from 25℃ to 600℃ with a heating rate of 5℃/min in a nitrogen atmosphere. DSC-6 Perkin–Elmer differential scanning calorimeter is used for differential scanning calorimetry analysis. Samples weighing between 8 and 10 mg are cut from composite panels and put into aluminum crucible with a tweezer and a lid is put on it. Samples are heated from 25℃ to 400℃ at a heating rate of 10℃/min, with a pause of 1 min at 25℃. They are then cooled to 25℃ at a rate of 10℃/min.

Physical analysis (void fraction)

Void fraction is calculated from the theoretical and experimental densities of the composite through equation (1), according to ASTM D2734-94 standard

Δ v = \frac{ρ_{ct} - ρ_{ex}}{ρ_{ct}}

ρ_ex and ρ_ct represent, respectively, the experimental density and theoretical density of the composite laminate. Experimental density of composite laminate is calculated by dividing the measured weight (g) by the measured volume (cm³). While the dimensions of the samples are measured by micrometer, the samples are weighed by a precision balance.

Equation (2) is used to calculate the theoretical density of the composites

ρ_{ct} = \frac{1}{(W_{f} / ρ_{f}) + (W_{m} / ρ_{m})}

W_f and W_m represent weight fractions of fabric and matrix, and ρ_f and ρ_m are the densities of fabric and matrix, respectively.

Morphological analysis

The interfacial surface morphologies of composite samples are observed using an SEM TS5130 Vega-Tescan at 20 kV acceleration voltage.

Results and discussion

Mechanical analysis

Tensile strength

The tensile strength results of fabric-reinforced composite samples that are taken from both warp and weft directions of samples are given in Figure 3. The tensile strength values of jute fabric-reinforced composite specimens are between 30 and 60 MPa, while the tensile strength of E-glass fabric-reinforced samples and carbon fabric-reinforced specimens are between 200 and 300 MPa and 300 and 450 MPa, respectively. Carbon, E-glass and jute fibre have tensile strengths of 3.5–5 GPa, 1.8–2.7 GPa and 0.45–0.55 GPa, respectively [30]. The diﬀerences between these values also support the tensile strength results of composite samples and reveals the fact that the type of reinforcement material has a huge eﬀect of the tensile strength of the composite material. It is obvious that samples that are taken from warp direction show higher tensile strength than the weft direction samples.

Figure 3.

Tensile strength results of one type of fabric-reinforced composite samples.

After the examination of the effects of MWCNTs on the tensile strength of samples, it is found out that all types of MWCNTs have increasing effect on samples. Owing to the low surface density and short length of CNTs, polymer molecules easily percolate into the gaps between CNTs and this brings on a strong interface between matrix and reinforcement material [31]. And this strong bonding strength at the interface allows materials to achieve higher tensile strength. Besides, it is seen that the highest tensile strength values are found in pristine MWCNT-treated specimens in all three fabric-reinforced composites. When the literature about effects of MWCNTs on tensile strength is examined, it is seen that the strength increased between 5% and 25% [17,23,25,27], while an increment of 3% to 61% is achieved in this study.

Impact strength

Figure 4 shows the impact strength results of fabric-reinforced composite samples. Jute fabric-reinforced samples have impact strengths between 20 and 30 kJ/m², whereas E-glass and carbon fabric-reinforced samples have impact strengths between 190 and 290 kJ/m² and 120 and 145 kJ/m², respectively.

Figure 4.

Impact strength results of one type of fabric-reinforced composite samples.

It is seen that MWCNTs have increasing effect on impact strengths of samples. It is stated in the literature that addition of MWCNTs to unsaturated polyester resin improves its coefficient of friction value [32]. It can be said that the increase in the coefficient of friction value of the resin may be the reason of increment of the impact strength of the MWCNT-added composite structures.

Results show that the addition of MWCNT-OH and MWCNT-COOH to the composite structures affects the impact characteristics of the samples barely; however, the addition of pristine MWCNT has slightly higher influence on the impact strength results of jute, E-glass and carbon fabric-reinforced composites.

It is noted that pristine MWCNT shows better results than the other MWCNTs in both mechanical (tensile and impact) tests. Jute fiber has an anionic structure due to free hydroxyl groups in its amorphous region of cellulose, lignin and hemicellulose regions [33]. Since anionic materials cannot easily bond together and MWCNT-OH and MWCNT-COOH have anionic natures, we achieve lower mechanical strength with aforementioned MWCNTs-added samples. For the case of the E-glass, it also contains hydroxyl groups and it is stated in the literature that acids and alkalis causes brittleness in E-glass fibers [34 –36]. This fact could be the reason of achieving much better mechanical properties with pristine MWCNT.

Warp direction samples show higher mechanical properties than the weft direction samples. This is an expected result owing to the inherent characteristics of the biaxial woven fabric structures. Moreover, either thicker yarns or higher yarn densities used at the warp directions create higher mechanical strength values in this direction.

Thermo-mechanical and other thermal analysis

Dynamic mechanical analysis

Storage modulus: Storage modulus demonstrates the elastic behavior of a material and is proportional to the energy stored in one cycle [37]. Figure 5 shows the variation of storage moduli of fabric-reinforced composite samples (warp and weft directions) as a function of the temperature. It can be seen that due to the transition from glassy region to the rubbery region, the storage moduli of samples decreased with increase in temperature. Materials are extremely immobile, highly packed and have strong intermolecular forces in glassy region, whereas after glassy region they become mobile and loosely packed which results in decrease in storage modulus [38].

Figure 5.

The variation of storage moduli of one type of fabric-reinforced samples.

After the examination of the storage moduli differences of warp direction samples at starting temperature (25℃), it is seen that carbon fabric-reinforced sample has a storage modulus of 30 GPa, whereas E-glass and jute fabric-reinforced samples have storage moduli of approximately 10 GPa and 3 GPa, respectively. This is also related to the tensile strengths of carbon, E-glass and jute fibers [30]. Moreover, the storage moduli of carbon-reinforced (6 GPa) and E-glass-reinforced (2.2 GPa) samples become stable at about 80℃; however, jute-reinforced (0.4 GPa) sample become stable at about 40℃. The fact that E-glass and carbon fibers are very resistant to heat compared to jute fiber has caused this situation. Similarly, at weft direction, carbon fabric-reinforced sample reach the highest storage modulus value (24 GPa at 25℃). It is observed that storage moduli of carbon, E-glass and jute fabric-reinforced samples become stable at about 6 GPa, 1 GPa and 0.2 GPa with ascending temperature, respectively. Lastly, the storage moduli of samples that are taken from warp direction are higher than those of weft direction.

The storage modulus variations of samples are examined after the addition of MWCNTs to the composite structures. Table 2 presents the storage modulus values of polyester resin and fabric-reinforced samples at 25℃ and 100℃. The storage moduli of samples started to decrease at about 60–70℃ and stabilized at about 100℃. Therefore, the storage moduli of the composite specimens are given at 25℃ and 100℃ representing the initial and final moduli of these samples. As the pure polyester sample begins to soften at 35℃, the dynamic mechanical analysis (DMA) analysis cannot be continued over 35℃.

Table 2.

Storage modulus values of polyester and one type of fabric-reinforced composites at 25℃ and 100℃.

	Storage modulus at 25℃ (GPa)		Storage modulus at 100℃ (GPa)
	Warp	Weft	Warp	Weft
Polyester	2.8	2.8	–	–
JJJJ	3.0	2.7	0.23	0.20
JJJJ-MWCNT	3.3	3.0	0.24	0.18
JJJJ-MWCNT-OH	3.1	2.9	0.22	0.18
JJJJ-MWCNT-COOH	2.4	2.6	0.16	0.22
GGGG	10.1	7.1	1.99	1.48
GGGG-MWCNT	13.2	11.9	4.01	3.57
GGGG-MWCNT-OH	13.0	11.0	4.20	3.72
GGGG-MWCNT-COOH	15.3	13.8	6.00	5.32
CCCC	28.1	24.2	5.46	4.71
CCCC-MWCNT	27.2	20.0	4.65	3.32
CCCC-MWCNT-OH	25.4	21.1	4.22	3.76
CCCC-MWCNT-COOH	22.1	20.9	3.95	3.84

The storage modulus analysis of jute fabric-reinforced samples shows that these samples have slightly higher storage modulus values than the polyester resin except JJJJ-MWCNT-COOH sample. It is understood that the nano-filled composites (except MWCNT-COOH-reinforced sample) reach higher storage modulus than the jute fabric-reinforced composite without nano-filler, and the highest value at the initial temperature is obtained with the pristine MWCNT-added sample. Furthermore, likewise, it is seen that the moduli of the samples taken from the warp direction are higher than those taken from the weft direction due to the thicker yarns or higher yarn densities used at the warp directions. The highest storage modulus value at 100℃ is also obtained with pristine MWCNT-added sample (JJJJ-MWCNT). The highest values obtained with pristine MWCNT at both temperatures can be due to the non-compatibility of anionic jute fiber with MWCNTs with anionic functional groups. Moreover, it is known that the excellent load-bearing capabilities of nanotubes are related to their high aspect ratio [39,40]. However, the functionalization of CNTs reduces their aspect ratio [41]. This can be also illustrated as the reason for obtaining a lower storage modulus with functionalized MWCNTs.

In E-glass fabric-reinforced specimens, storage moduli at both temperature values increase with addition of all three types of MWCNTs and also as distinct from jute fabric-reinforced composites, the highest values are obtained with MWCNT-COOH-added composite structures.

Surprisingly, MWCNT supplements decrease the storage modulus of the carbon fabric-reinforced samples. Among nano-filler-added samples (CCCC-MWCNT, CCCC-MWCNT-OH and CCCC-MWCNT-COOH), the highest storage modulus is achieved with the pristine MWCNT-added specimen. Moreover, samples taken from the warp direction reach higher modulus values than the weft direction samples.

Loss modulus: Loss modulus is related to the viscous behavior of a material and is proportional to the energy dissipated during a deformation period. The peak value of the loss modulus curve represents the glass transition temperature [38].

The variations of the loss modulus of composite samples (warp and weft direction) as a function of temperature is shown in Figure 6. The results demonstrate that there is a big difference between the loss moduli of the three types of fabric-reinforced composite samples. Carbon fabric-reinforced samples have the highest loss modulus, followed by E-glass fabric-reinforced composites and jute fabric-reinforced composites, respectively. As it is mentioned before, the peak value of the loss modulus curve indicates the T_g value of the material. It is noteworthy that the curves of the E-glass-reinforced composite samples do not have a peak; however, the other two samples have peak points in the loss modulus curves. Moreover, samples taken from warp direction reach higher values than the weft direction samples.

Figure 6.

Loss modulus curves of one type of fabric-reinforced composite samples.

Since the T_g can also be obtained from the tan delta curves, the T_g values obtained from the loss modulus curves are given in the same table to compare both transition temperatures.

Table 3 demonstrates the peak values of loss modulus curves of polyester resin and fabric-reinforced composites. Since the DMA of polyester resin cannot be measured over 35℃, it does not have a peak point. The examination of the specimens reinforced with different types of fabrics shows that the highest values are obtained with the carbon fabric-reinforced sample; however, the lowest values are obtained with the jute fabric-reinforced samples. It is also seen that the addition of different types of MWCNTs reduces the loss modulus values at all samples. The low tan delta value, which is an indicator of the high bonding strength between the matrix and the reinforcement material, is the ratio of loss modulus to the storage modulus of the material. The obtained nano-filled composite materials have lower loss modulus values than the non-treated samples, which means that tan delta value also decreases. This demonstrates that these nano-fillers (MWCNT, MWCNT-OH and MWCNT-COOH) positively affect the bonding strength between matrix and the reinforcement material. It is known that the reinforcing materials in the nanoscale size prevent the formation of microscale defects in the composite materials [27]. Based on the results obtained, the literature is supported by the fact that the bond between the matrix and reinforcement material is strengthened by the added MWCNT.

Table 3.

Peak values of loss modulus curves of polyester and one type of fabric-reinforced composite samples.

	Loss modulus peak (GPa)
	Warp	Weft
Polyester resin	–	–
JJJJ	0.34	0.24
JJJJ-MWCNT	0.31	0.30
JJJJ-MWCNT-OH	0.30	0.21
JJJJ-MWCNT-COOH	0.26	0.23
GGGG	1.40	1.10
GGGG-MWCNT	1.00	0.80
GGGG-MWCNT-OH	1.00	0.65
GGGG-MWCNT-COOH	1.20	0.85
CCCC	3.10	2.60
CCCC-MWCNT	2.70	2.20
CCCC-MWCNT-OH	2.80	2.30
CCCC-MWCNT-COOH	2.60	2.45

Damping factor (tan delta): The damping factor (tan δ) gives information about the correlation of the viscous and elastic components of a viscoelastic material [42]. It is the ratio of loss modulus to storage modulus [35]. T_g value can also be obtained from the peak point of a tan delta curve but it is stated in the literature that T_g value obtained from loss modulus curve is more accurate than T_g value obtained from tan delta curve. In the literature, it is reported that if the matrix and reinforcement material have high bonding strength, the mobility of molecular chains decreases and this causes reduction in damping factor [43].

The variation of the tan delta curves of composites as a function of temperature is shown in Figure 7. After the examination of the samples taken from both directions, it is found out that the jute fabric-reinforced composites have the highest tan delta values (0.35), whereas the tan delta values of the E-glass and carbon fabric-reinforced samples (0.28–0.29) are so close to each other. This situation can be explained by the fact that adhesion of natural fibers to matrices is insufficient compared to traditional fibers [44,45].

Figure 7.

Tan delta curves of one type of fabric-reinforced composites.

The peak values of tan delta curves of non-treated and nano-filler-added fabric-reinforced samples are given in Table 4. The tan delta value of unreinforced polyester resin plate cannot be obtained because DMA cannot be measured above 35℃. After the examination of the values in the table, it is seen that the addition of three types of MWCNTs reduces the tan delta values of the material, except carbon fabric-reinforced samples. In the carbon fabric-reinforced samples, addition of nano-filler materials (MWCNT, MWCNT-OH and MWCNT-COOH) has an increasing effect on the tan delta value. As stated earlier, the low tan delta value in the composite materials indicates the high bonding strength between matrix and the reinforcement material. The addition of carbon-based MWCNTs to the jute and E-glass fabric-reinforced composites reduces the tan delta value (improves polymer bonding with the textile material) of the material. However, the combination of carbon-based MWCNTs and the carbon fabric shows the opposite effect. Among the three types of MWCNTs, pristine MWCNT is the one that greatly reduces the tan delta values which can be also related to the higher aspect ratio compared to functionalized MWCNTs.

Table 4.

Peak values of tan delta curves, T_g values from loss modulus and tan delta curves of one type of fabric-reinforced composites.

	Peak value of tan delta curve (GPa)		T_g from loss modulus curve		T_g from tan delta curve
	Warp	Weft	Warp	Weft	Warp	Weft
Polyester resin	–	–	–	–	–	–
JJJJ	0.34	0.37	69.9	65.2	78.7	76.3
JJJJ-MWCNT	0.30	0.34	72.3	68.9	79.9	77.7
JJJJ-MWCNT-OH	0.32	0.35	71.8	69.2	78.1	77.9
JJJJ-MWCNT-COOH	0.31	0.34	70.7	68.7	79.0	76.8
GGGG	0.29	0.30	–	–	68.9	70.3
GGGG-MWCNT	0.23	0.25	–	–	74.1	73.6
GGGG-MWCNT-OH	0.25	0.27	–	–	72.7	72.9
GGGG-MWCNT-COOH	0.26	0.27			72.5	72.5
CCCC	0.28	0.28	72.0	70.2	77.9	82.2
CCCC-MWCNT	0.32	0.31	67.8	65.8	76.7	78.6
CCCC-MWCNT-OH	0.29	0.29	68.0	67.9	76.9	79.1
CCCC-MWCNT-COOH	0.29	0.28	67.7	65.4	75.5	78.4

The T_g values obtained from the loss modulus curves and tan delta curves are given in Table 4. The DMA measurements are carried out in two directions (warp and weft directions), the T_g values are given in the table as well, even though the glass transition temperature of the materials is not influenced by the orientation of the reinforcement material. The T_g values of the jute fabric-reinforced samples obtained from the loss modulus curves are around 65–70℃, while the ones that obtained from the tan delta curves are between 76℃ and 80℃. The T_g values of E-glass fabric-reinforced samples cannot be obtained from loss modulus curves since the loss modulus curves do not have peak points. The values obtained from tan delta curves varied between 69℃ and 74℃. Finally, for carbon fabric-reinforced samples, the T_g values obtained from the loss modulus curves range between 68℃ and 72℃, while the T_g values from tan delta curves range between 76℃ and 82℃. It is found out that the addition of MWCNTs do not significantly change the T_g value of the materials.

Thermogravimetric analysis

Table 5 demonstrates the amounts of weight losses and onset temperatures of polyester resin and composite samples. Polyester resin lost 97% of its weight by heating from 25℃ to 600℃ and its degradation starts at 305℃. When degradation temperatures of jute fabric-reinforced composite samples are compared, it is observed that pure jute fabric-reinforced composite has an onset temperature of 300℃, whereas pristine MWCNT, MWCNT-OH-treated and MWCNT-COOH-treated samples have onset temperatures of 321℃, 320℃ and 316℃, respectively. The addition of MWCNTs to jute fabric-reinforced composites increases the degradation temperatures at an amount of 20℃. Moreover, it is observed that the weight loss percentage of jute fabric-reinforced composite decreased from 92% to 89% by the addition of MWCNTs.

Table 5.

The percentage of weight loss and onset temperatures of polyester resin and fabric-reinforced composites.

Sample code	Initial weight (mg)	Final weight (mg)	Weight loss (%)	Onset temperature (℃)
Polyester	6.53	0.22	97	305
JJJJ	6.53	0.50	92	300
JJJJ-MWCNT	6.74	0.69	89	321
JJJJ-MWCNT-OH	6.65	0.70	89	320
JJJJ-MWCNT-COOH	6.26	0.69	88	316
GGGG	6.57	5.08	23	306
GGGG-MWCNT	6.68	4.91	26	327
GGGG-MWCNT-OH	6.91	4.68	32	327
GGGG-MWCNT-COOH	6.91	5.22	24	334
CCCC	6.27	4.27	23	320
CCCC-MWCNT	6.61	4.04	39	323
CCCC-MWCNT-OH	6.43	4.10	36	324
CCCC-MWCNT-COOH	6.40	4.06	36	330

The onset temperatures of neat polyester resin and pure E-glass fabric-reinforced composites are approximately at same level due to the high temperature resistance of E-glass fiber. Addition of MWCNTs to E-glass fabric-reinforced composites results in ascending onset temperatures. It is observed that highest onset temperature (334℃) is achieved with MWCNT-COOH-added sample. When weight loss percentages of samples are taken into consideration, it is unexpectedly found out that addition of MWCNTs increased the amount of weight loss of E-glass fabric-reinforced composite. The onset temperature of polyester resin increased from 305℃ to 320℃ after reinforcing with carbon fabric. Also, it is obtained from the results that MWCNT-COOH-added sample has the highest onset temperature (330℃) among carbon fabric-reinforced samples. When weight loss percentages of samples are considered, it is seen that the addition of MWCNTs increased the amount of weight loss of carbon fabric-reinforced composite, like the E-glass fabric-reinforced composites.

The abovementioned results show that the addition of MWCNTs increases the degradation temperatures of all the samples. It is stated in the literature that nanotubes have barrier labyrinth effect which slows down the diffusion of degradation from the bulk of the polymer to the gas phase [46]. The increased degradation temperature that is obtained in this study can be related to this barrier effect.

Differential scanning calorimetry analysis

Figure 8 shows the DSC curves of polyester resin and fabric-reinforced composite samples. From DSC graph of polyester resin, it is observed that an exothermic reaction starts at about 260℃ which is the degradation temperature of polyester resin. Moreover, an endothermic reaction is observed at about 390℃, which can be a sign of a change in the degradation process. Also, it is observed that these exothermic and endothermic reactions are compatible with the DSC analysis of unsaturated polyester resin in the literature [47].

Figure 8.

DSC curves of (a) polyester resin and (b, c, d) one type of fabric-reinforced composite samples.

The DSC curves of four-plied jute fabric-reinforced composites (Figure 8(b)) show endothermic peaks centered at around 52℃. These peaks are characterized by the removal of adsorbed moisture from the jute fabric. Jute (62.6% cellulose, 23.2% hemi-cellulose, 15.86% lignin) is majorly made up of cellulose. Therefore, its affinity to moisture is high since cellulose is hydrophilic in nature [48]. At all samples, an exothermic reaction starts at around 260℃ which is the degradation temperature of the polyester resin. Moreover, it is seen that at around 350℃, the exothermic reactions have peaks. As it is mentioned before, jute contains 62.6% cellulose and cellulose starts to degradate at around 315℃. It is thought that these peaks show the degradation of cellulose.

Figure 8(c) and (d) demonstrates the DSC graphs of E-glass and carbon fabric-reinforced samples. It is seen from graphs that the exothermic reaction starts at about 260℃. As it is known, E-glass and carbon are so durable to heat and it can be said that exothermic reaction which starts at around 260℃ indicates the degradation of only polyester resin. It has been understood that the addition of different types of MWCNTs does not have significant effects on the phase changes of the composite materials.

Physical analysis (void fraction)

Table 6 displays the experimental/theoretical densities and void fractions of pure and MWCNT-treated composite samples. While pure E-glass and carbon fabric-reinforced composite laminates have void fractions of about 8%, pure jute fabric-reinforced sample has a void fraction of 2.4%. Sreekala et al. [49] mention in their study that void content increased with the increment of fiber content. When the fiber weight and volume ratios of the samples are taken into consideration, it can be said that such void fraction amounts are reasonable since the fiber weight and volume ratios of jute-reinforced composites (weight ratio: 19%, volume ratio: 16%) are far less than those of E-glass-reinforced (weight ratio: 61%, volume ratio: 44%) and carbon-reinforced (weight ratio: 62%, volume ratio: 52%) ones. Besides, owing to the hydrophilic property of jute fiber [50], jute fabric absorbs more resin than the other fabrics and thus results in less amount of void content.

Table 6.

Experimental densities, theoretical densities and void fractions of one type of fabric-reinforced samples.

	Experimental density (g/cm³)	Theoretical density (g/cm³)	Void fraction (%)
JJJJ	1.22	1.25	2.4
JJJJ-MWCNT	1.24	1.25	0.8
JJJJ-MWCNT-OH	1.23	1.25	1.6
JJJJ-MWCNT-COOH	1.22	1.25	2.6
GGGG	1.64	1.78	7.86
GGGG-MWCNT	1.73	1.78	2.81
GGGG-MWCNT-OH	1.66	1.78	6.74
GGGG-MWCNT-COOH	1.73	1.78	2.81
CCCC	1.39	1.51	7.94
CCCC-MWCNT	1.42	1.51	5.96
CCCC-MWCNT-OH	1.40	1.51	7.28
CCCC-MWCNT-COOH	1.37	1.51	9.25

The examination of the effects of different types of MWCNTs on the void content of samples shows that at all fabric types, pristine MWCNT-treated samples have the lowest void content.

Morphological analysis

The morphological analysis of the composite samples is carried out using SEM images. In this section, SEM images of composite materials are examined by taking into consideration of the mechanical properties of the composite materials and the void ratios they possess.

Figure 9 demontrates the SEM images of JJJJ and JJJJ-MWCNT and JJJJ-MWCNT-COOH samples. In the samples JJJJ and JJJJ-MWCNT-COOH, the weak bonding areas of the matrix and reinforcement materials (voids) are seen; however, the composite sample of JJJJ-MWCNT is more tightly packed (almost no void area). The obtained void ratio values of these samples also support the results. The void ratios of the JJJJ and JJJJ-MWCNT-COOH samples are 2.4% and 2.6%, respectively, whereas the void ratio of the JJJJ-MWCNT sample is 0.8%.

Figure 9.

SEM images of (a) JJJJ, (b) JJJJ-MWCNT and (c) JJJJ-MWCNT-COOH samples.

The SEM images of GGGG, GGGG-MWCNT, GGGG-MWCNT-OH and GGGG-MWCNT-COOH samples are given in Figure 10. The void ratio measurement results show that the GGGG sample has the highest void ratio (7.86%); however, the GGGG-MWCNT sample (2.81%) has the lowest. After the examination of the SEM images, it is seen that in GGGG sample, the matrix and reinforcement materials are bonded weakly and the amount of voids are higher than the other samples. This is also reflected in the mechanical properties of the materials. The GGGG-MWCNT specimen shows the highest tensile (294.16 MPa–222.84 MPa) and impact strengths (284.12 kJ/m²–209.55 kJ/m²) in warp and weft directions; however, the GGGG specimen shows the lowest tensile (262.46 MPa–199.08 MPa) and impact strength (262.11 kJ/m²–190.42 kJ/m²).

Figure 10.

SEM images of (a) GGGG, (b) GGGG-MWCNT, (c) GGGG-MWCNT-OH and (d) GGGG-MWCNT-COOH samples.

It is stated that the addition of different MWCNTs to the composite material reduced the void ratio in almost all the samples. In some samples, however, the addition of MWCNTs negatively affects the void ratio of the composite structures. CCCC-MWCNT-COOH is an example of this situation. The void ratio of the CCCC sample is 7.94%, after the addition of MWCNT-COOH, this value increases to 9.25%. The SEM image of the CCCC-MWCNT-COOH sample is given in Figure 11, along with the SEM image of the CCCC sample. Although there is not a big difference between these two images, the matrix and the resin look more tightly packed in the CCCC sample.

Figure 11.

SEM images of (a) CCCC and (b) CCCC-MWCNT-COOH samples.

Conclusion

In this study, the mechanical, thermo-mechanical and thermal properties of textile-reinforced polyester-based composites are investigated. Three types of nano-filler materials (MWCNT, MWCNT-OH and MWCNT-COOH) are added to the structures to improve the properties of the composite materials.

Results show that:

Nano-filler materials improve the mechanical properties of the composite structures, but it has been found out that pristine MWCNT shows better results than functionalized MWCNTs due to the incompatibility of anionic materials. Besides improving the mechanical properties, MWCNTs also decrease the void ratio in the fabric-reinforced composite materials by providing highly packed structures.

Nano-filler materials decrease the storage moduli and increase the loss moduli of composite materials. And so, the damping value (tan delta), which is the ratio of the loss modulus to the storage modulus, decreases. Low tan delta value, which is an indicator of the high adhesion of the matrix and reinforcing material in the composite structures, shows that MWCNT addition strengthens the bonds in the composite structures.

The results of TGA show that the addition of nano-filler materials increase the degradation temperature of the composite structure to some extent by improving the bonding strength, whereas DSC analysis results show that the addition of MWCNTs does not lead to a change in the degradation temperature of the material.

It is considered that composite specimens developed within the scope of this thesis can be used in the automotive sector. Since the jute fabric-reinforced specimens have lower mechanical properties than the others, it is thought that these composite structures can be used in automobile interior door panels, floor panels and glove compartments, while E-glass and carbon fabric-reinforced composites are considered to be suitable for use in automobile exterior panels, drive shafts and leaf springs.

Footnotes

Acknowledgements

The authors are thankful to TUBITAK (The Scientific and Technological Research Council of Turkey) for 2211 PhD Scholarship Program and 2214 A International Doctoral Research Fellowship Program.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by the Istanbul Technical University Scientific Research Projects Fund (grant number 40030).

References

Bodaghi

Cristóvão

Gomes

, et al. Experimental characterization of voids in high fibre volume fraction composites processed by high injection pressure RTM. Composites Part A 2016; 82: 88–99.

Ding

Yan

McIlhagger

, et al. Comparison of the fatigue behaviour of 2-D and 3-D woven fabric reinforced composites. J Mater Process Technol 1995; 55: 171–177.

Hasan

Islam

Morshed

, et al. Comparison of the fatigue behaviour of 2-D and 3-D woven fabric reinforced composites. Am J Appl Phys 2016; 1: 1–12.

Mishra

Biswas

. Physical and mechanical properties of bi-directional jute fiber epoxy composites. Procedia Eng 2013; 51: 561–566.

Bakis

Bank

Brown

, et al. Fiber-reinforced polymer composites for construction – state of the art review. J Compos Construct 2002; 6: 73–87.

Mallick

. Fiber-reinforced composites: materials, manufacturing, and design, 3rd ed. Boca Raton: CRC Press, 2007.

Misnon MI, Islam MM, Epaarachchi JA, et al. Textile Material Forms for Reinforcement Materials – A Review. In: Third Malaysian postgraduate conference (MPC), Sydney, Australia, 2013, vol. 105, 2013.

Zhang

Wang

, et al. Friction and wear properties of carbon fabric composites filled with nano-Al2O3 and nano-Si3N4. Compos Part A 2006; 37: 1351–1357.

Friedrich

Almajid

. Manufacturing aspects of advanced polymer composites for automotive applications. Appl Compos Mater 2013; 20: 107–128.

10.

Sapuan

Maleque

. Design and fabrication of natural woven fabric reinforced epoxy composite for household telephone stand. Mater Des 2005; 26: 65–71.

11.

Seyhan

Gojny

Tanoglu

, et al. Critical aspects related to processing of carbon nanotube/unsaturated thermoset polyester nanocomposites. Eur Polym J 2007; 43: 374–379.

12.

Gojny

Wichmann

MHG

Fiedler

, et al. Influence of nano-modification on the mechanical and electrical properties of conventional fibre-reinforced composites. Compos Part A 2005; 36: 1525–1535.

13.

Chandrasekaran

VCS

Advani

Santare

. Influence of nano-modification on the mechanical and electrical properties of conventional fibre-reinforced composites. Carbon 2010; 48: 3692–3699.

14.

Chen

Liu

. Square representative volume elements for evaluating the effective material properties of carbon nanotube-based composites. Comput Mater Sci 2004; 29: 1–11.

15.

Dichiara

Zha

, et al. Square representative volume elements for evaluating the effective material properties of carbon nanotube-based composites. Compos Sci Technol 2014; 103: 36–43.

16.

Liao

Wang

Fang

, et al. Interface enhancement of glass fiber reinforced vinyl ester composites with flame-synthesized carbon nanotubes and its enhancing mechanism. ACS Appl Mater Interfaces 2011; 3: 534–538.

17.

Seyhan

Gojny

Tanoğlu

, et al. Critical aspects related to processing of carbon nanotube/unsaturated thermoset polyester nanocomposites. Eur Polym J 2007; 43: 374–379.

18.

Shen

Bateman

, et al. The effects of carbon nanotubes on mechanical and thermal properties of woven glass fibre reinforced polyamide-6 nanocomposites. Compos Sci Technol 2009; 69: 239–244.

19.

Zhu

Imam

Crane

, et al. Processing a glass fiber reinforced vinyl ester composite with nanotube enhancement of interlaminar shear strength. Compos Sci Technol 2007; 67: 1509–1517.

20.

Coleman

Khan

Blau

, et al. Small but strong: A review of the mechanical properties of carbon nanotube–polymer composites. Carbon 2006; 44: 1624–1652.

21.

Thostenson

Ren

Chou

. Advances in the science and technology of carbon nanotubes and their composites: A review. Compos Sci Technol 2001; 61: 1899–1912.

22.

Alexopoulos

Bartholome

Poulin

, et al. Structural health monitoring of glass fiber reinforced composites using embedded carbon nanotube (CNT) fibers. Compos Sci Technol 2010; 70: 260–271.

23.

Shokrieh

Saeedi

Chitsazzadeh

. Mechanical properties of multi-walled carbon nanotube/polyester nanocomposites. J Nanostruct Chem 2013; 3: 20–25.

24.

Zhang

Zhuang

Liu

, et al. Functional interphases with multi-walled carbon nanotubes in glass fibre/epoxy composites. Carbon 2010; 48: 2273–2281.

25.

Zhao

Teng

, et al. Mechanical, thermal and interfacial performances of carbon fiber reinforced composites flavored by carbon nanotube in matrix/interface. Compos Struct 2017; 159: 761–772.

26.

Zhang

Wang

Neo

, et al. Design of glass fiber reinforced plastics modified with CNT and pre-stretching fabric for potential sports instruments. Mater Des 2016; 92: 621–631.

27.

Qiu

Zhang

Wang

, et al. Carbon nanotube integrated multifunctional multiscale composites. Nanotechnol 2007; 18: 1–11.

28.

Fan

Santare

Advani

. Interlaminar shear strength of glass fiber reinforced epoxy composites enhanced with multi-walled carbon nanotubes. Compos Part A 2008; 39: 540–554.

29.

Bekyarova

Thostenson

, et al. Multiscale carbon nanotube−carbon fiber reinforcement for advanced epoxy composites. Langmuir 2007; 23: 3970–3974.

30.

Alves

Dias

APS

Diogo

, et al. Eco-composite: The effects of the jute fiber treatments on the mechanical and environmental performance of the composite materials. J Compos Mater 2011; 45: 573–589.

31.

Agnihotri

Basu

Kar

. Effect of carbon nanotube length and density on the properties of carbon nanotube-coated carbon fiber/polyester composites. Carbon 2011; 49: 3098–3106.

32.

Prawitasari

Ujianto

Juwono

. Effect of multi walled carbon nanotube (MWCNT) and mixing conditions on unsaturated polyester (UP) physical properties. IOP Conf Ser Mater Sci Eng 2018; 432: 1–6.

33.

Ammayappan

Nayak

Ray

, et al. Functional finishing of jute textiles – an overview in India. J Nat Fibers 2013; 10: 390–413.

34.

Norström

Watson

Engström

, et al. Treatment of E-glass fibres with acid, base and silanes. Colloids Surf A 2001; 194: 143–157.

35.

Plotnichenko

Sokolov

Dianov

. Hydroxyl groups in high-purity silica glass. J Non-Cryst Solids 2000; 261: 186–194.

36.

Tzounis L. Glass and Jute fibers modified with CNT-based functional coatings for high performance composites. PhD Dissertation, Technical University of Dresden, 2014.

37.

Tejyan

Patnaik

Singh

. Effect of fibre weight percentage on thermo-mechanical properties of needle punched nonwoven reinforced polymer composites. Int J Res Mech Eng Technol 2013; 3: 41–44.

38.

Jabbar

Militký

Wiener

, et al. Nanocellulose coated woven jute/green epoxy composites: Characterization of mechanical and dynamic mechanical behavior. Compos Struct 2017; 161: 340–349.

39.

Thostenson

Chou

. Aligned multi-walled carbon nanotube-reinforced composites: Processing and mechanical characterization. J Phys D Appl Phys 2002; 35: 77–80.

40.

Vera-Agullo

Glória-Pereira

Varela-Rizo

, et al. Comparative study of the dispersion and functional properties of multiwall carbon nanotubes and helical-ribbon carbon nanofibers in polyester nanocomposites. Compos Sci Technol 2009; 49: 1521–1532.

41.

Moisala

Kinloch

, et al. Thermal and electrical conductivity of single- and multi-walled carbon nanotube-epoxy composites. Compos Sci Technol 2006; 66: 1285–1288.

42.

Yang

Taha-Tijerina

Serrato-Diaz

, et al. Dynamic mechanical and thermal analysis of aligned vapor grown carbon nanofiber reinforced polyethylene. Compos Part B 2007; 38: 228–235.

43.

Pothan

Oommen

Thomas

. Dynamic mechanical analysis of banana fiber reinforced polyester composites. Compos Sci Technol 2003; 63: 283–293.

44.

Gassan

Bledzki

. The influence of fiber-surface treatment on the mechanical properties of jute-polypropylene composites. Compos Part A 1997; 28: 1001–1005.

45.

Tripathy

Di Landro

Fontanelli

, et al. Mechanical properties of jute fibers and interface strength with an epoxy resin. J Appl Polym Sci 2000; 75: 1585–1596.

46.

Barus

Zanetti

Bracco

, et al. Influence of MWCNT morphology on dispersion and thermal properties of polyethylene nanocomposites. Polym Degrad Stab 2010; 95: 756–762.

47.

Patel

, et al. Thermal and mechanical properties of modified polyester resin and jute composite. Der Chem Sin 2014; 5: 47–54.

48.

Biermann

Hädicke

Koltzenburg

, et al. Hydrophilicity and lipophilicity of cellulose crystal surfaces. Angew Chem 2001; 40: 3822–3825.

49.

Sreekala

George

Kumaran

, et al. The mechanical performance of hybrid phenol-formaldehyde-based composites reinforced with glass and oil palm fibres. Compos Sci Technol 2002; 62: 339–353.

50.

Karmaker

Hoffmann

Hinrichsen

. Influence of water uptake on the mechanical properties of jute fiber‐reinforced polypropylene. Appl Polym Sci 1994; 54: 1803–1807.