Mechanical properties of epoxy nanocomposites filled with melamine functionalized molybdenum disulfide

In this work, a melamine functionalized molybdenum disulfide (M-MoS2) was prepared and used as fillers to form epoxy (EP)/MoS2 nanocomposites. The effects of molybdenum disulfide (MoS2) and melamine functionalized molybdenum disulfide (M-MoS2) loading on the mechanical properties of epoxy composites were investigated and compared. With only addition of 0.8 wt% M-MoS2, the tensile strength and modulus of EP/M-MoS2 nanocomposites showed 4.5 and 4.0 times increase over the neat epoxy. Interestingly, the elongation at break value of EP was also increased with the introduction of M-MoS2 fillers. These properties could result from the good dispersion and strong interfacial adhesion of M-MoS2 fillers and the EP matrix. Therefore, this work provides a facile way to produce of high-performance EP nanocomposites.


Introduction
EP resin is an important thermoset material extensively used in a wide variety of applications such as coatings, 1 adhesives, 2 laminate, 3 semiconductor encapsulate, 4 and resin matrix composites, 5 because of its excellent mechanical stiffness and toughness, low shrinkage, good chemical resistance and superior adhesive force to many substrates. [6][7][8][9][10] Up to now, great efforts have been conducted to improve the properties of EP resin through addition of nanollers, such as montmorillonite, polyhedral oligomeric silsesquioxanes, carbon nanotube, graphene. [11][12][13][14][15][16][17][18][19] Recently, transition metal dichalcogenides (TMDCs) have attracted great interest in a wide range of research elds. [20][21][22][23][24][25][26][27] MoS 2 is one of the most typical TMDC. [28][29][30][31] A monolayer of MoS 2 reportedly has an extraordinarily high breaking strength ($23 GPa) and Young's modulus ($300 GPa), which are greater than those of chemically reduced graphene. 32,33 Derived from these remarkable properties, the MoS 2 sheets may hold considerable potential as a new EP resin reinforcement nanoller. It has recently been reported that incorporation of MoS 2 sheets into polymers at extraordinarily low ller content resulted in remarkable impact on the mechanical properties of the polymer, such as polystyrene, poly(methyl methacrylate), poly(vinylidene uoride), polyvinyl alcohol, polyethylene and polypropylene. [34][35][36][37][38][39] The resultant MoS 2 -lled polymer nanocomposites exhibited enhanced thermal stability, ame retardance and mechanical properties. With regards to EP resin, Y. Hu and Z. Gui et al. reported a MoS 2 -carbon nanotube reinforced EP composites. 40 With the introduction of 2 wt% MoS 2 -carbon nanotube, the organic volatiles and carbon monoxide was suppressed, while the mechanical properties were improved. N. Koratkar et al. also found the addition of exfoliated MoS 2 could enhance the mechanical properties of EP resin. 41 Although the MoS 2 well dispersed in the EP matrix, the interfacial adhesion between the MoS 2 and the EP matrix are less considered.
Therefore, in this research, we report a melamine functionalized MoS 2 and further used as llers to reinforce the EP resin. The functionalization of MoS 2 with melamine can prevent the agglomeration of MoS 2 , which can improve the dispersibility of MoS 2 in EP resin. Additionally, the amine group of melamine could promote the ring-opening reaction of the EP ring and lead to form a cross-linked structure. Thus, the attached melamine enhanced the interfacial interaction between the MoS 2 and EP matrix. Simultaneously, because the melamine and MoS 2 were widely used ame retardant, the ame retardance of EP resin will be improved. Therefore, this work provides a facile way to produce of high-performance EP nanocomposites.

Preparation of organophilic MoS 2
MoS 2 was rst exfoliated according to a literature method. 42 Typically, 5 g MoS 2 was placed in an autoclave, and 20 mL nbutyllithium (2.5 M in hexanes) was added. The autoclave was heated at 90 C for 12 h under stirring. Aer that, the product was ltered and washed with anhydrous hexane (5 Â 100 mL). The resultant lithium intercalated MoS 2 was vacuum-dried and immersed in melamine aqueous solution (3 g in 1000 mL H 2 O) under ultrasonication for 4 h to produce a colloidal suspension of melamine-functionalized MoS 2 (M-MoS 2 ). The suspension was neutralized with 1 M HCl, and the products were washed with distilled water (3 Â 1 L) and then washed with methanol to remove unreacted melamine. Subsequently, M-MoS 2 was obtained by freeze-drying.

Preparation of EP/MoS 2 nanocomposites
The desired amount of MoS 2 or M-MoS 2 powder was dispersed in appropriate amount of acetone and then stirred with a magnetic stirrer at room temperature for 30 min, followed by sonicated for 60 min. Aer that the mixture was mixed with desired amount of EP(m(EPON828)/m(PPGDGE) ¼ 55/45) and sonicated for another 60 min at 50 C. The acetone was then removed by vacuum distillation under stirring with a magnetic stirrer at 80 C. When the mixture was cooled down to 50 C, a stoichiometric amounts of curing agent (D230) corresponding to 100% of EP resin content was added and stirred for some time. The resulting mixture was then outgassed in a vacuum oven at 60 C for a short period of time and then cast into a Teon mold with special size. The sample was cured at 75 C for 2 hours and post-cured at 120 C for 8 hours.

Characterization of EP/MoS 2 nanocomposites
X-ray diffraction (XRD) patterns were obtained on a Germany BRUKER D8 ADVANCE diffractometer with Cu Ka X-ray radiation. The scanning range was 2-50 with a scanning speed of 1 min À1 . The sample specimens for transmission electron microscopy were microtomed with a diamond knife using a Leica Ultramicrotome at liquid N 2 atmosphere. TEM images were obtained from a Japan JEOL JEM-1011 microscope operating at 200 kV in bright eld mode. Differential scanning calorimeter (DSC) was conducted on Perkin-Elmer Diamond thermal analyzer. The samples were rst heated from room temperature to 50 C under N 2 atmosphere at heating rate of 10 C min À1 and was then cooled to À50 C. Finally, the sample was reheated to 50 C at 10 C min À1 . Tensile properties were performed on a USA INSTRON 5869 electronic testing instrument according to ASTM D638 at a cross-head speed of 5 mm min À1 at room temperature. The dumbbell like specimens (20 Â 4 Â 2 mm 3 ) were cut from the above cured sample casted on Teon mold. Each tensile value reported is the average of 5 tests. Optical microscope photos were obtained from an optical microscope (ANA-006, Leitz, Germany) and recorded using a charge-coupled device camera.

Results and discussion
MoS 2 was exfoliated according to a literature method and functionalized by melamine. The exfoliation and functionalization process were given in Scheme 1. To conrm the successful functionalization of MoS 2 , Fourier transform infrared (FTIR) spectroscopy, XRD analysis and thermogravimetric analysis (TGA) were conducted. Fig. 1(a) shows the FTIR spectra of the bulk MoS 2 , melamine and M-MoS 2 . No characteristic peaks appear in the spectrum of bulk MoS 2 , whereas several sharp peaks are observed for the M-MoS 2 sample. The two new peaks at 3340 and 3120 cm À1 resulting from -NH 2 stretching of the melamine, imply the existence of the melamine on M-MoS 2 . Furthermore, the characteristic peaks of melamine (CN stretching and NH 2 bending vibrations) at 1078 cm À1 and 1300-1700 cm À1 were apparently shied, indicating the formation of covalent bonds between melamine and MoS 2 . These results indicate that melamine was successfully graed onto the MoS 2 surface.
The bulk MoS 2 and M-MoS 2 samples were characterized by XRD. As shown in Fig. 1(b), bulk MoS 2 shows a single (002) diffraction peak at 2q ¼ 14.3 , which corresponds to a d-spacing of 0.6 nm. Upon exfoliation and functionalization, this peak becomes dramatically smaller and broader, and several new peaks are observed at lower 2q values. M-MoS 2 shows two new and very broad diffraction peaks at 2q ¼ 7.7 and 9.3 (corresponding to d-spacings of 1.1 and 0.9 nm, respectively), which indicate an increase in the layer distance of MoS 2 owing to the gra of melamine. Additionally, the weak and broad diffraction peak also indicate the crystallinity of the M-MoS 2 ller is low.
Thermal stabilities of the bulk MoS 2 and M-MoS 2 were investigated by TGA under nitrogen with a temperature range from the room temperature to 800 C. As presented in Fig. 1(c), bulk MoS 2 is clearly very thermally stable, as the mass loss is only 0.8 wt% upon heating to 800 C. In contrast, for M-MoS 2 sample, three degradation steps are observed. In the rst step (<200 C), the weight loss is due to the evaporation of physically adsorbed water; in the second step (220-380 C), the weight loss is caused by decomposition of melamine that is functionalized on the MoS 2 surface; and in the third step (>380 C), the weight loss may be attributed to decomposition of the carbon formed on the MoS 2 surface as a result of carbonization of melamine. The weight content of melamine in M-MoS 2 nanollers was calculated from the char yield of TGA measurement to $52 wt%. However, the content of melamine calculated from TGA is not exact, because the C 3 N 4 will formed during the carbonization process.
In order to investigate the dispersion of MoS 2 and M-MoS 2 in the EP matrix, the resultant thin lms of EP, EP/MoS 2 and EP/M-MoS 2 nanocomposites were prepared. The thin lms were observed under an optical microscope in transparent mode; the obtained micrographs are shown in Fig. 2. It was found that the M-MoS 2 llers are well dispersed in the EP matrix, while MoS 2 aggregation was observed in EP/MoS 2 nanocomposite with 1 wt% MoS 2 addition.
In order to investigate the dispersion and interaction of MoS 2 and M-MoS 2 in the EP matrix, the resulted samples were   characterized by SEM and TEM. As shown in Fig. 3, the SEM image of the EP/MoS 2 nanocomposite exhibited a smooth fractured surface and some MoS 2 sheet could be clear observed on the surface of the fractured surface without interaction with EP matrix. With regards to EP/M-MoS 2 nanocomposites, the wrapped structure was observed, implying the strong interaction between M-MoS 2 and the EP matrix. In order to fully characterized the dispersion of llers in the nanocomposites, TEM of the microtomed section of compression molded samples was examined (Fig. 4). It was found that the M-MoS 2 well dispersed in the EP matrix, while MoS 2 aggregation was observed in EP/MoS 2 nanocomposite. This morphology is correlated with the morphology obtained by optical micrographs. We therefore expected the EP/M-MoS 2 nanocomposites will exhibit better mechanical properties than EP nanocomposite with MoS 2 llers.
The effect of MoS 2 and M-MoS 2 on the glass transition temperature (T g ) of EP was investigated by DSC; the typical DSC curves are shown in Fig. 5 The T g of the virgin EP was 5.1 C. Upon introduction of MoS 2 , the T g value tend to decrease with increasing MoS 2 content, while the T g gradually increased with the M-MoS 2 content increasing. With only 1 wt% M-MoS 2 addition, the T g of EP rises up to 7.3 C. The increment of T g refers to the reduction of matrix chain mobility by the presence of M-MoS 2 . During fabrication, melamine molecules bridged MoS 2 with EP matrix and a strong interface was thus produced. The strong interface restricts the motion of polymer chain and thus gives rise to the increase in T g . While, with regards to EP/ MoS 2 nanocomposites, the reduced T g are probably caused by two reasons (i) agglomeration when neat MoS 2 llers added. (ii) Reduction of the EP matrix's cross-linking density due to the barrier effect of MoS 2 .  Thermal stabilities of the EP, EP/MoS 2 and EP/M-MoS 2 were evaluated by TGA under nitrogen atmosphere. As can be observed in Fig. 6, all the nanocomposites present similar degradation behaviors, suggesting that the existence of MoS 2 and M-MoS 2 did not signicantly affect the degradation mechanism of the matrix polymers. For the EP/MoS 2 and EP/M-MoS 2 nanocomposites, their degradation temperature is lower than that of pure EP, which could be attributed to the earlier thermal degradation of melamine functional groups on MoS 2 surface and/or the high thermal conductivity of MoS 2 . However, the addition of MoS 2 or M-MoS 2 llers exhibited higher char residues compared to neat EP. Additionally, it can also be seen that the weight loss rates of the EP/M-MoS 2 nanocomposites was lower than the EP and EP/MoS 2 nanocomposites. This phenomenon played an important role in improving the ame retardancy of the EP resins. When increasing the temperature, the melamine degraded at rst and form char on the MoS 2 surface. The formed char can provide a protective shield of mass and heat transfer, which slow down the heat release rate during the thermal degradation process.
The inuence of MoS 2 and M-MoS 2 on the mechanical properties of the EP nanocomposite is evaluated using a Universal Testing Machine (UTM). The reinforcing effects of the MoS 2 and M-MoS 2 on the tensile properties of the EP composites are summarized in Fig. 7 and Table 1. Clearly, the tensile strength and modulus of the resultant EP/M-MoS 2 nanocomposites were signicantly enhanced, even at very low M-MoS 2 nanoller loadings. The tensile modulus of EP/M-MoS 2 nanocomposites increased from 3.7 to 18.6 MPa (approximately a 400% increase over neat EP), and the tensile strength increased from 1.5 to 8.3 MPa (approximately a 450% increase over neat EP) when the M-MoS 2 content increased from 0 to 0.8 wt%. When the M-MoS 2 content higher than 0.8 wt%, the tensile strength and modulus value barely changed. Upon introduction of MoS 2 , as the MoS 2 content increased from 0 to 0.4 wt%, the tensile strength and modulus of EP/MoS 2 nanocomposites slightly increased, but a reducing trend was observed for the further increasing MoS 2 content. This phenomenon could be attributed to the agglomerate of MoS 2 llers and reduced the effective contact area between the MoS 2 surface and the EP matrix, and thus reduced the reinforcement efficiency. At the same time, melamine as a modier not only improves the interfacial compatibility between molybdenum disulde and EP, but also acts as a co-curing agent to promote cross-linking of EP (Scheme 2). Therefore, the addition of M-MoS 2 greatly improves the mechanical properties of EP.

Conclusions
In summary, M-MoS 2 llers were successfully prepared through exfoliation of MoS 2 followed by reaction with melamine. The effects of M-MoS 2 on the thermal and mechanical properties of EP were investigated. Because the good dispersion and strong interfacial adhesion of M-MoS 2 llers and the EP matrix, the mechanical properties of EP were signicantly improved, even with very low M-MoS 2 addition. Therefore, this work provides a facile way to produce of high-performance EP nanocomposites.

Conflicts of interest
There are no conicts to declare.