Mechanical Performance of Diamine Silane Modified Carbon Nanotubes Reinforced Epoxy Resin Composites

Abstract

The addition of unmodified carbon nanotubes (CNTs) to epoxy resin will cause a decrease in the initial thermal decomposition temperature of the EP/CNT composite material, likely due to the weak interfacial adhesion between the nanofiller and its surrounding matrix. As such, functionalized drug carriers using CNTs could overcome this; for example, after silane modification, the diameter of CNTs is increased from 32 nm to 38 nm. The fracture cross-section of EP/CNT composite material is rough on the surface and exhibits ductile fracture, while the pure EP material presents a brittle fracture cross-section with a smooth fracture cross-section. It has also been proven that the dispersibility of CNTs is improved, along with an enhancement in the degree of dispersion. Thus, as compared to pure EP, after surface treatment of the CNTs, the tensile strength and elastic modulus of the EP/CNT composite material were improved up to a value of 134.6% and 32.9%, respectively, while the elongation at break decreased to 60.09%.

1. Introduction

Owing to the huge specific surface area and aspect ratio, carbon nanotubes (CNTs) have been widely used as mechanical reinforcement materials for polymeric matrices like epoxy resin (EP). But adding CNTs as an additive alone does not significantly improve the mechanical properties of composite materials, which has an even likelier chance to lead to significant attenuation [1], thereby affecting the further possible application of the CNTs-based composite materials. This is mainly ascribed to the tendency of CNTs to be easily enveloped by the EP surface, which hinders establishing a strong interfacial contact between CNTs and the EP surface, limiting their expansion and, thus, the optimal load transfer function [2,3]. Therefore, the uniformity of dispersing additives like CNTs in EP materials can be improved [4] through the surface functionalization of CNTs.

In industrial regimes, the surface functionalization of CNTs, such as carboxylation, amination, and silanization processes, are commonly used in the production and processing to improve on dispersing CNTs well uniform through EP [5]. To date, the carboxylation modification of CNTs is most often conducted via the strong acid oxidation and free radical addition methods [6]. For example, the mechanical properties of single-walled (SW) CNTs/EP composites could be significantly enhanced by acidification and fluorination treatment of SWCNTs [7]. The mixture of both concentrated sulfuric acid and nitric acid was explored to oxidize multi-walled CNTs (MWCNTs) plus further sonication (i.e., 3 h in a 40 °C water bath) to produce carboxyl groups on the surface of MWCNTs [8], and the activation energy of carboxylates MWCNTs/EP composites was reported to be significantly improved. In addition, ultrasonication pretreatment technology not only increases the content of functional groups greatly but also contributes to further functionalization improvement [9].

The number of active groups on the surface of CNTs could be increased by surface grafting functional groups onto the surface, effectively improving the dispersibility and interfacial adhesion between CNT additives and the EP matrix [10,11,12]. Furthermore, the amide groups, through grafting onto the surface of CNTs, can noticeably reduce the phenomena of CNTs self-aggregation in EP, and then the overall mechanical properties of CNTs/EP could be significantly enhanced [13,14]. In addition, surface modification of MWCNT with amino-silane reagent like γ-Aminopropyltriethoxysilane (APTES) exhibited significant improvements in aspects of bending stress, modulus, and impact strength of the corresponding amino functionalized MWCNTs/EP composites [15]. Due to the strong van der Waals forces and π-π stacking interactions between CNTs [16], they exhibit chemical inertness and are not easily mixed or reacted with organic solutions [17,18]. Therefore, it is a relatively difficult task to modify CNTs [19] to achieve desired functionalities.

To achieve a decent mechanical enhancement of the CNTs/EP materials, carbonylated CNTs were used as the raw additives in this study, and further amination treatment was applied to attach new chemical functional groups to the CNTs’ macromolecular chain. Then, the improvements in both the thermal and mechanical properties of the composites were summarized, which is likely due to the enhanced binding ability between these modified CNTs and EP.

2. Materials and Methods

2.1. Materials

To further improve the reinforcement effect of carbon nanotubes (CNTs), carbonylated CNTs were subjected to salinization treatment by grafting the surface of CNTs with silane gradients and then used for preparing the EP-based reinforcement composites as listed in

Table 1. Raw material specifications.

Chemical Name	Specification	Manufacturer
Hydroxylated multi-walled carbon nanotubes (MWCNTs)	≥98%, L ≈ 6 nm	Beijing Deke Island Gold Technology Co., Ltd., Beijing, China.
Epoxy resin	E51	Huntsman New Materials (Guangdong) Co., Ltd., Guangzhou, China
Anhydrous ethanol	AR, 99%	Huzhou Shuanglin Chemical Technology Co., Ltd., Huzhou, China
Bis [3-(triethoxysilicon)propyl] amine	97.04%	Bijiasuo Biochemical Technology Co., Ltd., Guangzhou, China
Curing Agent	Industrial Grade	Huntsman New Materials (Guangdong) Co., Ltd., Guangzhou, China

, all of which are used without further purification until stated otherwise.

2.2. Silane Grafting Modification

A simple and effective silane grafting modification was used as detailed in the following. Briefly, 1.0 g of hydroxylated multi-walled carbon nanotubes (MWCNTs) was dispersed in ethanol of 94 mL, stirred well, and then sonicated for 30 min in a water bath at room temperature to obtain a stable MWCNT suspension in EtOH. Afterward, the silane solution with varied mass fractions of 1.0%wt, 2.0%wt, and 3.0%wt was added into the above MWCNT dispersions. Then, 5.0 g of deionized water was added to start the silane hydrolysis process and covered with parafilm to prevent EtOH evaporation. This mixture was further heated at 40 °C in a water bath for 4 h, followed by washing with excessive water and ethanol in a filtration bottle to remove unreacted and physically absorbed silane molecules on the surface of the CNTs. Finally, the modified CNT particles were dried in an oven at 80 °C for 12 h to obtain silane-modified CNTs labeled as BCNTs-x (x = 1, 2, and 3, which denotes the silane solution mass fraction as mentioned above).

2.3. Fabrication of CNT–Reinforced Epoxy Resin Composites

A mixture of 10 g curing agent in 0.25 g (0.5%wt) of silane-modified CNTs (BCNTs-x) was magnetically stirred at room temperature for 1 h, followed by sonication in a water bath for 40 min at ambient conditions. Afterward, 40 g of dry epoxy resin was added into the above-prepared mixture composed of the curing agent and silane-modified CNTs, which was further mixed as evenly as possible via magnetically stirring at room temperature. This modified CNT/epoxy resin-based mixture was added to a vacuum filtration bottle for 40 min to eliminate air bubbles trapped during the stirring process involved. Lastly, the liquid mixture was carefully poured into a polyvinylidene fluoride (PVDF) membrane mold and placed in a flatbed machine for a hot-pressing procedure maintained at 100 °C for 45 min to obtain CNT/EP reinforced composite materials for further analysis and to obtain EP/BCNT composite material samples; this preparation procedure is portrayed in Figure 1. Samples with a dimension of about 5 cm in length and 1 cm in width were taken from the two bottom edges, and the middle portion of the samples was used for composition analysis and mechanical testing. The samples were named EP/BCNT-1, EP/BCNT-2, and EP/BCNT-3, respectively.

2.4. Characterization

The chemical structure of CNTs before and after modification was evaluated using Fourier transform infrared spectroscopy (FTIR, TENSOR 27 IR, Bruker, Saarbrucken, Germany). The powder pattern X-ray diffraction (XRD, AXS diffractometer D8, Bruker, Germany) of the nanoparticles was measured using an irradiation source of Cu Kα at a scanning rate of 5°/min in the range of 5–80°. Then, the software Jade 6 was utilized to perform baseline calibration on the data and determine the diffraction peak crystal plane.

The morphology of all the epoxy resin composites was characterized using field emission scanning electron microscopy (FE-SEM, SU8010, Hitachi, Tokyo, Japan). The curing behavior of epoxy resin in the presence of CNT was characterized by differential scanning calorimetry (DSC, DSC4000, PerkinElmer, Waltham, MA, America) at a heating rate of 10 °C/min under a nitrogen atmosphere (ASTM E1356, “Standard Test Method for Assignment of the Glass Transition Temperature by Differential Scanning Calorimetry”). The grafting efficiency of the modified CNTs was measured using a thermogravimetric analyzer (TGA, TG209F3, Tarsus-NETZSCH, Bruker, Saarbrucken, Germany), heating at a rate of 10 °C/min from 25 °C to 600 °C under a nitrogen atmosphere.

The mechanical tests were conducted using the universal testing machine (Instron5943, Canton, Massachusetts, America) with 10 specimens evaluated for each sample. The fracture strength refers to the load per unit area that a fiber can withstand when subjected to external forces, measured in MPa as calculated from Equation (1).

$$P={\displaystyle \frac{N}{S}}$$

(1)

where P denotes the fracture strength (MPa), N is the load borne by the fiber when it breaks (N), and S stands for the cross-sectional area of the fiber (mm²).

Elongation at break refers to the ratio of the displacement value of fiber from the moment it is subjected to force to the length before stretching, measured in% as implied from Equation (2):

$$\epsilon ={\displaystyle \frac{L-L_0}{L_0}}\times 100\mathrm{\%}$$

(2)

where ε is the elongation at break (%), and L and L₀ are the length at fracture and the original measurement of the fibers (mm), respectively.

3. Results and Discussion

3.1. Chemical Structural Analysis of CNT Modification

The images of the as-prepared composite samples made of hydroxylated carbon nanotubes (CNTs–OH) and hydroxylated carbon nanotubes (BCNTs) modified with diamino silane (3-(triethoxysilane) propyl) as a coupling agent in epoxy resin are presented in Figure 2a, and the bare difference could be perceived by the naked eye. Unique and noticeable changes could be implied from the IR spectrum, as shown in Figure 2b, since the broadband at 3434 cm⁻¹ is generated by the hydroxyl vibration on its surface of CNTs–OH. As for the BCNT, a hydroxyl peak also appears at 3434 cm⁻¹, a vibration peak of C–H appeared at 2716 cm⁻¹, N–H appeared at 1627 cm⁻¹, a stretching vibration peak of the BCNT characteristic functional group N–C appeared at 1350 cm⁻¹, and a Si–OH peak without hydrolysis reaction appeared at 766 cm⁻¹. All the characteristic IR bands indicated that diamino silane was successfully modified on the surface of CNTs–OH [20].

The similar XRD spectrum pattern for both CNTs–OH and BCNT–3 samples, as illustrated in Figure 2c, further confirms that the functionalization and silane grafting has no effect on the crystal size of nanotubes [21]. The diameter of CNTs–OH increased from 32 nm to 38 nm after silicification, which is likely due to the grafting of larger silane molecules onto the functionalized CNTs–OH surface. In addition, the slight shift in the peak toward a higher angle confirms the successful grafting of silane molecules on the functionalized CNTs–OH surface.

The as-received CNTs–OH has a smaller relative particle size and aggregation, as shown in Figure 2d, while silane-treatment-obtained BCNT has a significantly larger particle diameter and obvious agglomeration phenomenon, as shown in Figure 2e, possibly due to the surface modification of CNTs–OH by silane enhancing the hydrogen bond attractions among CNTs–OH, as well as the chemical crosslinking of silane itself, and thus improves the crosslinking between CNTs–OH. Furthermore, it also indicates that silane modification on the surface of CNTs–OH is beneficial for the interfacial bonding of the resin matrix.

3.2. Thermal Analysis

Thermal analysis refers to a variety of techniques in which a property of a sample is continuously measured as the sample is programmed through a predetermined temperature profile [22]. Among the most common techniques are thermal gravimetric analysis (TGA) and differential scanning calorimetry (DSC) [23]. TGA is a technique in which changes in sample mass are measured as a function of time and temperature, gaining valuable insights into a wide range of material properties and behaviors, including thermal and oxidative stability, decomposition kinetics, moisture content, volatile components, the presence of multiple constituents, and it also serves as a crucial tool for determining the decomposition temperature of substances. The melting point, glass transition temperature (T_g), and crystallization behavior of the composite material can be determined by DSC to understand its thermal stability and phase transition characteristics [24]. Here, the thermal stability tests of EP, EP/CNTs–OH, and EP/BCNT materials under a nitrogen atmosphere via TGA are illustrated in Figure 3 and

Table 2. Summary of thermal properties of EP, EP/CNTs–OH, and EP/BCNT composite materials.

Sample	EP	EP/CNT	EP/BCNTs–1	EP/BCNTs–2	EP/BCNTs–3
TGA @ T5% (°C)	337.6	307.0	336.0	342.2	340.0
DTG @ Tmax (°C)	372.3	374.8	368.3	371.1	374.7
DSC @ Tg (°C)	45.1	51.1	45.8	37.7	37.22

. As can be seen from Figure 3a, the initial decomposition temperature (T_5%) at a weight loss of 5% of pure EP is 337.6 °C, while that of the modified EP/CNTs–OH composite was decreased to a certain extent of about 307.0 °C. As compared to that of pure EP, the T_5% of EP/BCNT–1 composite material exhibits a bare change at about 336.0 °C. As the loading of BCNT–x increases, the T_5% of composites slightly changed from 342.2 °C for EP/BCNT–2 and 340.0 °C for EP/BCNT–3, as compared to that of the EP/BCNT–1. The enhancement of EP by unmodified CNTs–OH will lead to a decrease in the thermal stability of the material, while the enhancement of EP by surface-modified CNTs–OH, that is, BCNT, has a lesser impact on the thermal stability of the material.

On the other hand, the temperature (T_max) at the maximum thermal decomposition rate of the materials via the derivative thermogravimetry (DTG), as exhibited in Figure 3b, and the T_max of pure EP itself is 372.3 °C, and that of EP/CNTs–OH, EP/BCNT–1, EP/BCNT–2, and EP/BCNT–3 composites is 374.8 °C, 368.3 °C, 371.1 °C, and 374.7 °C, separately, implying the influence of pure unmodified and modified CNT on max decomposition rate at the T_max of the composites is not as significant as expected.

The glass transition temperature (T_g) traces of all the samples via DSC measurement are portrayed in Figure 3c, and the obtained T_g peak of pure EP is clearly labeled as 45.1 °C, while that of the EP/CNTs–OH slightly increases up to 51.1 °C after treatment. This is further lowered and broadened at 45.8 °C, 37.7 °C and 37.2 °C for EP/BCNT–1, EP/BCNT–2, and EP/BCNT–3, respectively. This observation indicates that pure CNTs–OH could increase the enthalpic relaxation temperature of the composite material and be of some help in enhancing the thermal properties of EP/CNTs–OH composites. The exothermic behavior is most likely due to the significant increase in the mobility of the polymer chains, which is facilitated by introducing relatively soft Bis [3-(triethoxysilicon)pro-pyl] amine between the EP matrix and the additive CNT. However, the silane-modified CNTs–OH reinforced EP composites (i.e., EP/BCNT) exhibit a reduction in the enthalpic relaxation temperature. This is likely due to the influence of CNT on the movement of molecular segments in composite materials [25], which makes the molecular movement more sluggish, resulting in a certain increase in the mobility of the polymer chains of composite materials. The surface modification of CNTs–OH with the silane coupling agent improves the interfacial compatibility between the CNTs and EP, also reducing the hindrance to the movement of molecular chain segments, further leading to a decrease in the glass transition temperature of the composite material.

3.3. Mechanical Properties of EP/CNT Composite Materials

Enhanced mechanical properties of composite materials can be achieved by adding reinforcing materials [26], and this could also be validated from the measured stress–strain curves of the EP and different proportions of EP/modified CNT cloth composite materials (i.e., EP/BCNT–x composites), as shown in Figure 4a. Basic parameters are summarized in

Table 3. Summary of mechanical properties of EP, EP/CNTs–OH, and EP/BCNT composite materials.

Sample	EP	EP/CNT	EP/BCNTs–1	EP/BCNTs–2	EP/BCNTs–3
Stress @ peak(MPa)	27.2 ± 1.41	53.4 ± 2.14	58.5 ± 1.87	62.0 ± 2.35	65.9 ± 2.56
Strain @ peak (%)	4.7 ± 0.24	5.0 ± 0.32	7.4 ± 0.42	6.6 ± 0.35	6.1 ± 0.29
Stress @ rupture (MPa)	8.0 ± 1.24	19.2 ± 1.96	30.0 ± 1.53	33.4 ± 1.69	24.7 ± 1.82
Strain @ rupture (%)	54.6 ± 0.37	6.7 ± 0.27	8.6 ± 0.33	8.4 ± 0.48	24.7 ± 0.45

, which implies that the tensile strength gradually increases, the elastic modulus slightly increases, and the fracture elongation changes significantly. It also tells us that EP composites become more brittle with either the unmodified or modified CNTs, showing noticeably less ductile EP/BCNT–x as the fraction BCNT increases. With the increase in silane content, the mechanical properties of EP/BCNT are improved, with the tensile strength of EP/BCNT–3 material being 65.9 MPa, which is much higher than the value of 27.2 MPa for the pure EP material. The variation in elongation at break varies, decreasing from 54.6% for pure EP material to 4.7% for EP/BCNT–3. This is due to the improved compatibility of the material, which reduces the fracture elongation of the composite material.

The fracture section of the composite materials after undergoing a tensile test was further investigated by SEM observation, as shown in Figure 4b. It turns out that the fracture section of the pure EP material is flat and appears smooth overall, indicating a clearly brittle fracture of the epoxy resin itself, which is also evidenced by the suddenly dropped-off-like tail in the stress-strain curves, as shown in Figure 4a. Furthermore, the cross-section generated during the fracture of the EP/CNTs–OH composite material exhibits rugged and rough characteristics, implying that the fracture of the EP/CNTs–OH composite is similar to a ductile fracture (based on the support shown in the sluggish-like stress-strain curves in Figure 4a). This is because the addition of CNTs–OH not only enhances the EP material [27] but also changes the compatibility of the EP matrix due to the possible interfacial hydrogen bonding between the –OH from CNTs–OH and the epoxy ring in the resin matrix. There may also be possible π-π stacking interactions between the aromatic ring motifs of the epoxy resin and the sp²-C ring of the CNTs–OH. These interactions may cause the material to crack from different defect points during stretching. The fracture section of EP/BCNT–3 is observed to be less smooth and a bit more rough compared to that of the pure EP; however, it is noticeably smoother than that of EP/CNTs–OH, as shown in Figure 4b. This is because the obtained BCNT after the surface modification of CNTs–OH significantly enhances the compatibility between BCNT additives and the EP matrix by increasing the numbers of possible hydrogen bonding or van der Waal force type interactions, which arise from motifs like Si···OH and N···H from the BCNT. These possible weak interactions could help reduce the defect points of the material and make its fracture surface relatively smooth and thus also help to improve the mechanical properties of the composite material.

3.4. Surface Wettability Analysis

To assess the hydrophilicity of CNT films before and after silicification, silane solutions with varied concentrations of 1.0%wt, 2.0%wt, and 3.0%wt (i.e., 1.0 g, 2.0 g, and 3.0 g) were added into 94 mL of ethanol (99%). Then, they were evenly mixed and poured into a dish containing 1.5 g of carbon nanocloth. Afterward, 5.0 g of deionized (DI) water was added to begin the silane hydrolysis process and then wrapped with parafilm to prevent ethanol evaporation. The treated carbon nanocloth was further heated in a water bath at 40 °C for 4 h, followed by washing with excessive DI water and ethanol in sequence to remove unreacted and attached silane molecules on the surface of the nanotubes.

It can be considered a non-wetting surface or material when the measured water contact angle (WCA) is greater than 90° [28], and it can be considered a wettable and hydrophilic surface when the WCA is less than 90°. The surface wettability of the modified CNT–treated carbon nanocloth was evaluated by WCA measurement, as presented in Figure 5. The WAC on the surface of untreated CNT cloth before silicification is 120°, and the cloth is classified as hydrophobic, with less affinity for water. Compared to the unmodified CNT, silane–treated CNTs are reduced slightly to 105° for the sample of CNTC–1 with a silane solution of 1.0%wt, still possessing hydrophobic features. This is noticeably lowered to 58° and 50° for samples treated with silane solutions composed of concentrations at 2.0%wt and 3.0%wt, namely CNTC–2 and CNTC–3, respectively. All these findings imply that the modified material CNTC–2 has good hydrophilicity, which improves even further after grafting with a higher concentration of 3.0%wt silane (CNTC–3), resulting in better wettability. There is still not much of a decrease in WCA as the silane concentration increases from 1.0 to 2.0%wt, suggesting that the cost-effectiveness of CNTC–2 is better than that of CNTC–3, although the latter has overall good hydrophilicity. Overall, the hydrophilicity of CNT fabrics improves via silane grafting modification, and its hydrophilicity gradually improves as the percentage of silane content increases.

The noticeable transition from the hydrophobic untreated CNTCs to hydrophilic silane-modified CNTCs further implies that silane treatment improves the interface adhesion between fibers and the matrix [29]. This is especially significant for the latter, which are composed of many functional groups that tend to be hydrophilic, such as –OH, –NH, and –Si–OH. This is evident from the IR analysis of BCNTs compared to that of the untreated CNT–OH (Figure 2b). This will further increase the surface adhesion between the treated CNTs and the epoxy matrix, which is, in turn, manifested and observed from the mechanical performance as backed up by the stress–strain curve (Figure 4a). Thus, suitable surface treatment is needed for producing outstanding properties composites [30].

4. Conclusions

Larger silane molecules or a silane coupling agent was successfully grafted onto carbon nanotubes (CNTs) to achieve a surface-functionalized BCNT composite as backed up by thorough IR and XRD characterizations; specifically, the diameter of CNT increased from 32 nm to 38 nm after silicification. Then, this BCNT composite was used as an additive into the EP as a matrix for improving the mechanical properties since the pure EP material is well known for illustrating a smooth brittle fracture cross-section. While the fracture cross-sections of EP/CNT and EP/BCNT composite materials are noticeably rough, fluctuating, and could be regarded as ductile fractures, they still maintain as good thermal properties as EP itself. These findings prove that the dispersibility of CNTs has been improved, and the degree of dispersion has also been enhanced. This, in turn, implies that the tensile strength and elastic modulus of EP/CNT–3, especially, have been improved compared to EP after treatment, with specific values of 134.6% and 32.9%, respectively, while exhibiting a decrease in elongation at break of about 60.09%. In addition, these silane-treated CNTs are of great importance for significantly modifying a surface from hydrophobic to hydrophilic. This change might be due to the enhanced interfacial surface created by these treated CNT additives in the matrix, which can significantly improve the mechanical properties of composites.