Mechanical And Thermal Behavior of Carbon Nanotubes/ Vinyl Ester Nanocomposites

ADRIAN COTEŢ, MARIAN BAŞTIUREA, GABRIEL ANDREI*, ALINA CANTARAGIU, ANTON HADĂR Dunărea de Jos University of Galaţi, Faculty of Engineering, 47 Domneasca Str., Galati, 800008, Romania University Politehnica of Bucharest, Department of Materials Strength, Splaiul Independenţei 313, Bucharest, 060042, Romania Academy of Romanian Scientists, 3 Ilfov Str., Bucharest, 050045, Romania Technical Sciences Academy of Romania, 26 Dacia Blvd, Bucharest, 010413, Romania

MWCNT was used as additive for vinyl ester composites in order to increase flexural stress with 48% [39], 24% [40] or 29% [41] compared to neat resin. Aligned MWCNs were used to study the influence on glass transition temperature. Depending on the orientation of nanotubes, the glass transition temperature increased from 93 0 C, for neat vinyl ester, to 107 0 C for aligned nanotube with DC electric field [42,43].
In the present article, the mechanical (bending and compression) and thermal properties (DSC and TMA) of neat vinyl ester and its nanocomposites have been investigated.
Also, the effect of the addition of carbon nanotube into the vinyl ester matrix on the overall properties of nanocomposites has been discussed, considering the serious drawback in obtaining a good dispersion of nanoparticles into polymer matrix that is the natural tendency of the carbon nanotubes to agglomerate [44,45].

Experimental part Materials and methods
Vinyl ester resin POLIMAL VE 11M was supplied by SC PROFESIONAL SRL and was used as such. According to the quality certificate provided by the supplier, the bending stress for vinyl, as per ISO 178, was 120 MPa. The composite specimens studied in this work were prepared by reinforcing vinyl ester resin with multiwall and single wall CNTs. The amount of MWCNTs and SWCNTs in the composite has been set to 1.0 wt.%, 0.15%wt and 0.2%wt. In order to obtain SWCNT/ vinyl ester (SW) and MWCNT/ vinyl ester (MW) nanocomposites respectively, a certain amount of carbon nanotubes was introduced into resin using the analytical balance Kern, model EG4200 -2 NM, precision 0,001g. The mixture was placed in a mortar and stirred for 1h using a magnetic stirrer (600 rpm). Afterwards, the mixture was degassed under vacuum (1-2 torr) for 1 minute. The catalyst, 1% cobalt octoate was added under continuously stirring for 5 minute, followed by last degassing. The composites were cured at room temperature, in rubber molds with samples dimension of Φ6×6 mm×mm, for compression test. The composite samples were further heated in an oven, at 70C, for 8 hours.

Compression and bending tests
Bending samples were prepared in rubber molds with dimensions 40 mm x 8 mm x 4 mm. Compression and three point bending tests were performed according to ISO 604 and ISO 178, respectively, using an INSTRON 8872 Servohydraulic Fatigue Testing System (Fig.1a). The compression samples were tested at a crosshead speed of 1 mm/min, 5 mm/min, 10 mm/min, 25 mm/min and 50 mm/min. Five tests of each type of composites were conducted.

Differential Scanning Calorimetry (DSC)
The test was performed according to the ASTM 1269 standard, using DSC equipment from Mettler Toledo (Fig. 1b). The samples were maintained at 30℃ for 3 minutes, then they were heated from 30℃ to 90℃ with heating rate of 10℃/min, maintained for 3 minutes at 90℃ and cooled from 90℃ to 30℃ with cooling rate of 10℃/min. The mass of samples was of 31 mg. Five samples of each material were used for measurements. The glass transition was evaluated from the second run to erase the thermal history of the sample.

Thermo-mechanical analysis (TMA)
Thermo-mechanical analysis (TMA) was conducted according to ASTM E831 using a TMA/ SDTA 840 equipment from Mettler Toledo (Fig. 1c). Coefficient of thermal expansion (CTE) was measured on specimens of nominal dimensions of 8x10x2mm. Five specimens of each material were tested for CTE. The temperature range was 30 0 C -220 0 C at a constant rate of 10 0 C/min and the modification of the specimen width was recorded.

Results and discussions Compression test
A typical compression response of the tested MWCNT/ vinyl ester and SWCNT/ vinyl ester is shown in Fig.2. Results from Fig. 2 clearly indicate that there is no significant variation in compression stress between the composites and vinyl ester. It can be seen that the compressive stress of vinyl ester and its carbon nanotubes composites significantly increase with increasing test speed, as presented in Fig. 2. The MW 0.1 featured the best compression properties among the three nanocomposites considered. Very little improvement was observed when stress is increased by 11 MPa (8%) for MW 0.1 under test speed of 10 mm/min and 25 mm/min (Fig.2a). Upon incorporation of only 0.15 wt. % SWNTs, the compression stress of vinyl ester is significantly improved by about 14% from 135 to 155 MPa, in case of test performed at a speed of 5 mm/min (Fig.2b). Moreover, it is worth comparing the mechanical properties of MWCNT/ vinyl ester with SWCNT/ vinyl ester nanocomposites studied here. Upon reaching the test speed of 5 mm/min, SWCNT / vinyl ester takes the higher value of compression stress. If the test speed increases above 5 mm/min, the influence of MWCNT on compression stress become more pronounced (Fig.2c). The gap between SW 0.15 and MW 0.15 increases from 6% up to 13% in case of a test speed of 5 mm/min. With the exception of the nanocomposites MW 0.2 and SW 0.2, it can be noticed that the compression stress has improved, compared to the neat vinyl ester resin.    Fig.4b, SW 0.1 exhibited better bending stress than SW 0.15, SW 0.2 and VE. It can be clearly seen that bending stress of the SW 0.1 is nearly 2.3 times higher than neat vinyl ester for the test performed with 10 mm/min. This may be due to the efficient load transfer between the SWCNTs and vinyl ester matrix. An overall estimation revealed that addition of MWCNTs and SWCNTs into vinyl ester matrix improved the bending stress, in all the cases. For the same content of additive, SWCNTs had a higher influence on the bending stress of vinyl ester resin than MWCNT. Fig.4d shows the bending stress of 0.1 wt. % SWCNT and 0.1 wt. % MWCNT/ vinyl ester composites. SW 0.1 showed a higher bending stress than MW 0.1 for tests speed higher than 5 mm/min. The bending stress of SW 0.1 reaches the maxim of 81.5 MPa for test conducted with 10 mm/min, and increases by 25% to 56% in comparison with MW 0.  Fig.5 shows bending modulus as a function of tests speed for SWCNT and MWCNT/ vinyl ester nanocomposite. In all tests, the bending modulus of MWCNT/ vinyl ester composites exhibits a higher value than neat vinyl ester (Fig.5a). The bending modulus of vinyl ester was improved by 26%, 47% and 27%, respectively, for tests performed with 1mm/min, 5 mm/min and 25 mm/min, in case of MW 0.1. Fig.5b shows the bending modulus variation of nanocomposites SW 0.1, SW 0.15, SW 0.2 and neat vinyl ester. In case of SW 0.1 nanocomposite, the bending modulus value increases by 33% and 45%, respectively, for test speeds of 1mm/min and 5 mm/min (Fig. 5c). A careful examination of results reveals that MWCNTs have a strong effect on flexural modulus of vinyl ester than SWCNTs (Fig.5d).  Adding CNTs into vinyl ester resin leads to mechanical properties improvement for all the nanocomposites being tested. The presence of MWCNTs and SWCNTs may hinder the mobility of polymer chains, during the curing reaction, as well as modifying the final structure of cross-linked network of the vinyl ester resin. The strengthening of interfacial bonding between the vinyl ester matrix and carbon nanotubes enables an effective transfer of internal stresses between vinyl ester and CNTs, leading to the enhancement of the mechanical properties of CNTs/vinyl ester nanocomposites.
Differential Scanning Calorimetry (DSC) DSC analysis for both SWCNT and MWCNT / vinyl ester composites was performed for all types of samples investigated. Fig.6 shows DSC thermograms of SWCNT and MWCNT/ vinyl ester composites. Results demonstrate the increase in specific heat for MW and SW nanocomposites compared to neat vinyl ester. DCS graphs reveal that, in case of MW 0.15 and SW 0.15, the specific heat takes higher values along the temperature range. Furthermore, DSC thermograms highlight an endothermic melting process with peak temperatures between 80-90C. a b Fig.6. Specific heat vs. temperature for: a) VE and MW; b) VE and SW DSC results clearly suggest the good thermal stability of SWCNT and MWCNT/ vinyl ester nanocomposites. It was observed that the glass transition temperature (Tg) increased with addition of CNTs, due to the restriction in polymer chain mobility caused by interaction with carbon nanotubes. The increase in Tg of CNT/ vinyl ester nanocomposites was explained by the prevention of the motion of the vinyl ester macromolecular chains due to the presence of the carbon nanotubes. For contents above 0.15%, the Tg is found to decrease due to the formation of nanoparticles agglomerates which are not very effective in the restriction of chain mobility. Glass transition temperature increases from 91.1C for neat vinyl ester to 96.3C for 0.15 wt. % SWCNT/ vinyl ester and to 97.91C for 0.15 wt. % MWCNT/ vinyl ester.

Thermo-mechanical analysis (TMA)
The thermal expansion coefficient below and above glass transition temperature of neat vinyl ester and its nanocomposites was calculated from the slopes of thermograms using STAR e software. Resulted values are given in Table  1 for three temperature ranges (30-80C, 30-220C and 100-220C). It is clear that the CTE decreases with the content of carbon nanotubes in all cases. The addition of 0.15 and 0.2 wt. % of MWCNTs produces the higher decrease in CTE below Tg (in this case the fall was about 11.9 %). In temperature range above Tg, SW 0.1 and SW 0.15 recorded the lower CTE, by a diminution of 17.8%. Furthermore, there is no direct ratio between CNTs type, CNTs content and CTE values. That may be explained by admitting the existence of an optimal threshold for carbon nanotubes content into vinyl ester resin, taking into account that carbon nanotubes tend to agglomerate thus affecting the dispersion process during samples formation.
As for glass transition temperature, it can be noticed that Tg increases as the carbon nanotubes are being incorporated into the vinyl ester composites. Tg was higher than those of vinyl ester resin for 0.1 and 0.15 wt. % content of carbon nanotubes. An increasing of Tg, with 5C and 6C, was measured for SW 0.15 and MW 0.15.
In Table 2, the Tg values, as determined by TMA and DSC methods, are shown. It is worth noting that Table 2 reveals significant difference between glass transition temperature measured by DSC and TMA which may be due to the use of two different measurement methods and the instrument sensitivity.

SEM Images
Carbon nanotubes, particularly the functionalized ones, have the ability to form a network throughout the polymer matrix. When the content of CNTs is higher, the network formed in the matrix is more developed. Content close to the percolation threshold has a beneficial effect on the physical properties of the composites. Thus, the thermal conductivity, the dimensional stability and the mechanical strength are improved. Figure 7 suggests the presence of carbon nanotubes into vinyl ester matrix of the nanocomposites. SEM images were obtained on fracture surface of CNT/vinyl ester samples, after bending tests. As seen in Figure 7d, a proper dispersion of the carbon nanotubes within the polymer resin was achieved. Also, SEM analysis reveals the dimensions of CNTs diameters, which are different for MWCNTs and SWCNTs, i.e. around 100 nm and 50 nm, respectively ( Fig. 7a and 7c, for 0.1 wt. % content), or about 60 nm and 30 nm, respectively, for 0.2 wt. % content (Fig. 7b and 7d).
A well-defined and homogeneous SWCNTs network can be seen in Figure 7d. Certainly, this relatively dense and uniform CNTs network ensures a reinforcement of the composite material and an improvement of the thermal conductivity and dimensional stability.

Conclusions
The mechanical and thermal properties of SWCNT and MWCNT/ vinyl ester nanocomposites have been studied. Compression, three point bending test, differential scanning calorimetry and thermo-mechanical analysis were performed over nanocomposite samples. In case of SW 0.15 composite, a maximum increase of 14% and 9%, respectively, for compression strength and elastic modulus have been noticed, as compared to the vinyl ester resin. After bending test of MW 0.1 and SW 0.1 nanocomposites, a maximum increase of 47% and 46%, respectively, have been obtained for elastic modulus and bending strength, as compared to the neat resin. In case of compression tests performed at speed up to 5 mm/min, SWCNTs induced better mechanical behavior of the nanocomposites than MWCNTs. As for MW 0.15 and MW0.2 nanocomposites, the coefficient of thermal expansion has shown a decrease of about 12% as compared to neat vinyl ester resin. Enhancement of mechanical and thermal properties of CNTs/ vinyl ester nanocomposites is due to the proper dispersion and interfacial interaction between MWCNTs and vinyl ester resin. Addition of carbon nanotubes into vinyl ester resin lead to structural changes of the nanocomposite. Thus, the increase in compression strength, elastic modulus, specific heat, glass transition temperature and coefficient of thermal expansion may be accounted for by the excellent mechanical and thermal properties of carbon nanotubes introduced into the vinyl ester nanocomposites.