Cellulose nanocrystal functionalized aramid nanofiber reinforced epoxy nanocomposites with high strength and toughness

ANF/DMSO solution (500 ml of 0.2 wt%) was prepared using the Yang's modified process [56]. KM2 + Kevlar fibers (1 g, CS-800), potassium hydroxide (KOH, 1.5 g, Fisher Scientific), and deionized (DI) water (20 ml) were added to DMSO (500 ml, Fisher Scientific), and the solution was stirred for 4 h at 23 °C. CNCs (1 g, CelluForce) were dispersed in N,N-dimethylformamide (20 ml, Fisher Scientific), and the suspension was heated at 80 °C for 12 h to activate the CNCs. Thionyl chloride (3.5 ml, SOCl₂, MilliporeSigma) was added dropwise into the suspension, and the suspension was stirred and heated at 80 °C for 24 h to form a chlorinated CNC (CNC-Cl) solution. Then, the CNC-Cl solution and ANF/DMSO solution were mixed and heated at 80 °C overnight, followed by adding and mixing 10 wt% of 3-glycidoxypropyltrimethoxysilane (GPTMS, MilliporeSigma) at 80 °C for 24 h to prepare functionalized ANF/CNCs (fACs) with covalent linkages between the nanomaterials (figure 1(a)). In addition to the preparation of fACs, CNCs were added to ANF/DMSO solution, followed by GPTMS functionalization to fabricate ANF/CNCs without covalent bonding (ACs) for the purpose of comparison with fACs.

Zoom In Zoom Out Reset image size

DI water was used to precipitate the fACs from the solution, and a vacuum-assisted filtration set was used to wash the nanofibers thoroughly with DI water and acetone (MilliporeSigma). After the filtration, the fACs were dispersed in acetone using horn sonication (Fisher scientific, Model 500) before drying them to yield fAC fillers with 50 wt% acetone content. GPTMS modified ANFs, CNCs, and ACs were also prepared to fabricate epoxy nanocomposites.

The fACs were imaged using an AFM (XE-70) to characterize any changes in the morphology and dimensions due to covalent bonding between ANFs and CNCs. Additionally, ATR-FTIR (Nicolet iS50 spectrometer) and XRD (Rigaku Ultima IV) were used to identify the changes of the chemical and crystalline structure of the nanofillers.

The fAC reinforced epoxy samples were prepared by adding 0.2 wt% to 1.5 wt% of fACs to the epoxy resin based on Epon862 (diglycidyl ether of bisphenol-F epoxy, Hexion) and Epikure W (diethyltoluenediamine, Hexion). The fAC paste was dispersed into the Epon862 using horn sonication and centrifugal mixing (FlackTek Speedmixer) for 20 min. After the dispersion, the mixture was dried under vacuum at 70 °C for 24 h to remove acetone from the fAC paste. Then, Epikure W (26.4 part per hundred resin)was added to the fAC mixture and mixed for 20 min, followed by degassing at 70 °C under vacuum prior to being cast into silicone molds. The mixture was subsequently cured in a convection oven with a 30 min temperature ramp from 25 °C to 121 °C followed by a 240 min isothermal hold at 121 °C (figure 1(b)). ANF, CNC, and AC reinforced epoxy nanocomposites were also fabricated using the same process described above.

Tensile tests were conducted using a universal testing machine (Instron model 5982) following ASTM D638. Type V specimens were prepared and tested using a crosshead speed of 1 mm min⁻¹. Three-point bend specimens were also fabricated and tested to conduct fracture toughness tests following ASTM D5045-99. After testing, the critical stress intensity factor (K_Ic ) and fracture toughness (G_Ic ) were obtained using the following equations

$$\begin{eqnarray}&&{K}_{Ic}=\displaystyle \frac{{P}_{{\rm{\max }}}}{B{W}^{\tfrac{1}{2}}}f\left(x\right)\end{eqnarray} \tag{ 1 }$$

$$\begin{eqnarray}&&\begin{array}{l}f\left(x\right)=6{x}^{1/2}\displaystyle \frac{\left[1.99-x\left(1-x\right)\left(2.15-3.93x+2.7{x}^{2}\right)\right]}{\left(1+2x\right){\left(1-x\right)}^{3/2}},\\ x=a/W\end{array}\end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}&&{G}_{Ic}=\displaystyle \frac{{{K}_{Ic}}^{2}}{E}\left(1-{\nu }^{2}\right),\end{eqnarray} \tag{ 3 }$$

where, P_max is the maximum load, B is the thickness of specimen, W is the width of specimen, a is the crack length, E is the Young's modulus of samples, and ν is the Poisson's ratio (assumed to be 0.353). A dynamic mechanical analyzer (DMA TA 800) was used to examine the viscoelastic properties of the epoxy nanocomposites in tensile mode. Tests were conducted at 1 Hz with a heating rate of 3 °C min⁻¹ in a temperature range from 30 °C to 190 °C. The fractured surfaces of epoxy samples were examined using optical microscope and scanning electron microscopy (SEM, JSM-7800FLV, JEOL).

ATR-FTIR was performed to characterize the chemical structure alterations before and after the GPTMS treatment and CNC functionalization on the ANFs. The FTIR spectra of ANFs, CNCs, ACs, and fACs are displayed in figure 2. After the surface modification using GPTMS, new peaks are shown in the spectra of ANFs and CNCs in figure 2(a). The peaks at 2923, 1100, 1008, and 988 cm⁻¹ correspond to the C–H stretching vibration of methylene, Si–O, Si–O–C, and Si–O stretching vibration from GPTMS, respectively. Thus, the alterations of FTIR spectra before and after the treatment show that ANFs and CNCs were functionalized using GPTMS. The chemical structure changes after functionalizing ANFs using CNCs and combining ANFs and CNCs are shown in figure 2(b). Both AC and fAC displayed all the characteristic peaks of both ANFs and CNCs, however, fAC showed two new characteristic peaks in the FTIR spectra. The peaks at 1720 and 3460 cm⁻¹ are attributed to ester C=O stretching and a secondary amine, respectively, and they can be shown from the reaction between the functionalities of ANFs and CNC-Cl. The results confirm that ANFs and CNCs are linked through covalent bonding. Therefore, the FTIR spectra of ANFs, CNCs, ACs, and fACs demonstrate the modification of nanofillers and the successful preparation of fACs with chemical covalent linkages.

Zoom In Zoom Out Reset image size

To investigate the crystalline structure of fACs, XRD patterns were obtained. Figure 3 displays the XRD patterns of ANFs, CNCs, ACs, and fACs. As shown from previous research [31, 57], ANFs showed their diffraction peaks at 20.6° and 22.9°, which are attributed to the (110) and (200) planes of aramid fibers. The characteristic peaks of ANFs are also shown from the XRD patterns of ACs and fACs due to the presence of ANFs in these hybrid fillers. The XRD patterns of CNCs showed peaks of the crystalline cellulose structure at 14.6°, 16.7°, 22.8°, and 34.7° [58, 59]. While these diffraction peaks of CNCs appeared in the XRD pattern of ACs after the combination of ANFs and CNCs, they were not shown in the XRD pattern of fACs, which were prepared through the chemical covalent linkages. The disappearance of the peaks could be due to the attenuated crystalline structure of CNCs following the reaction between ANFs and CNCs [60]. The crystallinity indexes of the samples were also calculated and shown in table 1. It is shown that both combination of ANFs and CNCs and the functionalization of ANFs using CNCs decrease the degree of crystallinity. This could be due to the disorientation of the crystal lattice of fillers when they were functionalized. Additionally, AFM was used to examine the changes in the morphology and dimensions of ACs and fACs. Figure 4 displays the AFM images of ANFs, CNCs, ACs, and fACs. The diameters of ANFs and CNCs are measured to be in the range of 4–12 nm and 3–8 nm, respectively, which is in accordance with the results from previous research [44, 61]. In the image of ACs, CNCs are observed to be separated from ANFs. However, in the image of fACs, CNCs are shown on the ANFs, and the shape of CNCs is likely changed due to the chemical reaction between the ANFs and CNCs. Therefore, the AFM images show the absence of chemical interaction between ANFs and CNCs in ACs, and the establishment of covalent linkages in fACs. Drawing from the characterization results in this section, the preparation of fACs with covalent bonding between ANFs and CNCs is confirmed. The new hybrid materials based on ANFs and CNCs are expected to contribute to improved mechanical interlocking interaction between the polymer and nanofillers due to the extensive branching and network structures of the nanofillers, in addition to the enhanced chemical interaction due to the modification using a silane coupling agent.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Tensile tests were performed to examine the effect of CNC functionalization on the ANFs for the mechanical properties of epoxy composites. The concentration of nanofillers in the samples was limited to 1.5 wt% due to the formation of micro voids in samples with a higher concentration of nanofillers even after the degassing procedure. The surface modification using GPTMS increased the viscosity of the epoxy mixture significantly during the mixing process [62]. The average Young's modulus and tensile strength of ANF, CNC, AC, and fAC reinforced epoxy nanocomposites are displayed in figures 5 and 6 shows stress versus strain curves of epoxy samples reinforced with 1.5 wt% nanofillers. As shown from previous studies [29, 44], ANFs and CNCs were found to be beneficial for reinforcing epoxy nanocomposites due to their high aspect ratio and surface area. The Young's modulus and tensile strength of ANF reinforced samples improved up to 7.5% and 9.1%, and those of CNC reinforced epoxy resin improved by 8.1% and 6.9%, respectively, compared to the reference epoxy samples. When ACs were incorporated into the epoxy matrix, tensile properties were in the middle of ANF and CNC reinforced epoxy nanocomposites. This means that there is no synergetic effect of ACs when they are simply combined without covalent linkages. However, fAC reinforced samples outperformed other samples in tensile properties. The Young's modulus of fAC reinforced samples was observed to increase up to 3.29 ± 0.06 GPa with 1.5 wt% fACs, showing a 15.7% improvement relative to the reference sample. The samples reinforced with 1.5 wt% fAC also displayed the highest tensile strength, 87.2 ± 1.5 MPa, which is 10.1% higher than that of the neat epoxy samples. Moreover, the lowest tensile strength of fAC reinforced samples at 0.2 wt% was even higher than the highest tensile strength of all the ANF, CNC, and AC reinforced epoxy samples. The average Young's modulus of epoxy samples reinforced with 1.5 wt% fAC was found to be 7.6%, 6.0%, and 5.5% higher than that of the 1.5 wt% ANF, CNC, and AC reinforced nanocomposites, respectively. These results show the effect of CNC functionalization on the ANFs for the tensile properties of epoxy nanocomposites in addition to the individual properties from both ANFs and CNCs. The ANFs maintain the excellent mechanical properties of p-phenylene terephthalamide fibers after the dissolution, and their high aspect ratio and large specific area are beneficial for the reinforcement of the epoxy matrix. CNCs also have a high aspect ratio and large surface area with excellent mechanical properties. In addition to the characteristics of the nanomaterials, CNCs functionalized at the various reaction sites of ANFs can improve the mechanical interlocking of nanofillers in the epoxy matrix through branched network structures. Moreover, the surface modification of fACs using GPTMS improves the chemical interaction between the epoxy matrix and fACs [44]. Therefore, the enhanced chemical and mechanical interaction improved the stress transfer between the epoxy matrix and nanofillers, yielding improvement in the overall mechanical properties of epoxy composites [17, 23, 63].

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The fracture properties of fAC reinforced epoxy samples were evaluated using three-point bending tests following the ASTM standard. After testing, the average critical stress intensity factor, K_Ic , and fracture toughness, G_Ic , of epoxy samples reinforced with ANF, CNC, AC, and fAC were calculated and displayed in figure 7. From the test results, both K_Ic and G_Ic of epoxy nanocomposites improved as the weight fraction of the nanofillers in the samples increased. The improvement of K_Ic and G_Ic of CNC reinforced samples compared to the neat epoxy resin was slightly better than that of ANF reinforced samples, and both K_Ic and G_Ic of AC reinforced samples were in the middle of ANF and CNC reinforced samples. These results indicate that ANFs and CNCs can enhance the fracture properties, but the non-covalent interaction between the ANFs and CNCs is not significant for the further reinforcement of composites. However, the fracture properties of fAC reinforced epoxy composites were higher than those of all the other samples. The test results show that the 1.5 wt% fAC reinforced samples can improve K_Ic by 104.4% and G_Ic by 249.6% compared to the reference samples. Both K_Ic and G_Ic of epoxy composites reinforced with 1.5 wt% fACs were 11.5% and 17.5% higher than those of 1.5 wt% CNC reinforced epoxy composites, respectively. The recently reported works regarding the improvement of mechanical properties of epoxy nanocomposites are summarized in table 2. It was shown that nanofillers such as CNTs, ANFs, and GOs contributed to improving mechanical properties, including Young's modulus, tensile strength, and fracture toughness. The fracture toughness improvement by novel fAC nanofillers was remarkable compared to the recently reported results of epoxy nanocomposites reinforced with various nanofillers. The improved fracture properties can be attributed to the characteristics of CNC functionalized ANFs and the chemical covalent interaction between the epoxy and fACs [44]. The sliding of large network structured fACs in the epoxy matrix can contribute to the energy dissipation at the interphase region and impede crack propagation in the nanocomposites. The fACs with a high aspect ratio and CNC functionalization on the surface, which provides mechanical interlocking in the matrix, can also contribute to crack bridging during crack propagation. Moreover, chemical covalent interaction between the epoxy matrix and fACs through surface modification using a silane coupling agent can enhance crack bridging in epoxy nanocomposites. In order to investigate the reinforcement mechanisms of fAC reinforced epoxy nanocomposites, optical microscope and SEM were used to examine the fractured surfaces of samples and the images are displayed in figure 8. The fractured surface of neat epoxy sample looked smooth and very few ripples were observed, which indicate a lack of energy dissipation mechanism during the crack propagation. However, novel nanofillers reinforced epoxy samples exhibit very rough surfaces with ripples and crazes due to the network structured nanofillers, which usually are not observed from the neat epoxy resins. Formation of such crazes and ripples contributes to absorbing energy during the fracture, which leads to improved fracture toughness. Drawing from the test results, CNC functionalization on the ANFs is found to be beneficial for improving both the tensile strength and fracture resistance simultaneously.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Dynamic mechanical properties of epoxy composites were obtained to investigate the effect of CNC functionalization on the ANFs over a broad temperature range. Representative storage modulus and tan δ curves versus temperature are displayed in figure 9, and storage modulus at 40 °C and 160 °C, and glass transition temperature (T_g) of epoxy nanocomposites are summarized in table 3. The value of T_g was obtained from the maximum tan δ peak value of each sample. As the temperature increased during the test, the storage modulus decreased for all the samples and showed a relaxation above 128 °C, which is attributed to the glass transition of polymeric materials. The ANF, CNC, AC, and fAC reinforced epoxy nanocomposites exhibited improved storage modulus at 40 °C and 160 °C. The storage modulus below and above T_g increased as the nanofillers weight fraction increased, although a slight decrease was observed for some samples which is likely due to non uniform dispersion. The 1.0 wt% fAC reinforced epoxy composites exhibited a maximum of 25.1% and 68.0% improvement in storage modulus at 40 °C and 160 °C, respectively, compared to the reference sample. The results show that fAC can improve the storage modulus of epoxy samples both in a glassy state and rubbery state. This could be due to the enhanced chemical and mechanical interaction between the epoxy matrix and fACs through the CNC functionalization on the ANFs and surface modification using GPTMS. The chain flexibility of the epoxy resin can be restricted by the incorporation of fACs and covalent linkages between them [63, 68, 69]. Moreover, the higFh thermal stability and T_g of aramid polymer chains can contribute to the dynamic mechanical properties over the broad temperature range, as shown from the results. However, the contribution of ACs for improving the storage modulus was less than that of fACs due to the relatively weak interfacial interaction between the polymer molecules and ACs.

Zoom In Zoom Out Reset image size

The glass transition behavior of epoxy nanocomposites was observed using the T_g values. As displayed in table 3, the T_g value of ANF and fAC reinforced epoxy composites shifted to a higher temperature compared to the neat samples, while the improvement of T_g value was not significant for the AC reinforced nanocomposites. The T_g of nanocomposites shifted 6.7 °C higher when 1.0 wt% of fACs was incorporated into the epoxy resin. The T_g is known to represent the chain mobility of the polymer matrix in polymer nanocomposites and shifts to a higher temperature when the chain mobility is hindered. The factors affecting the glass transition behavior include the morphology of nanosized fillers, chemical structure and flexibility of the polymer matrix [70, 71]. The fACs with attached CNCs on the ANFs can restrict the movement of the epoxy matrix by mechanical interlocking between them. Additionally, covalent bonding formed between the fACs and epoxy chains can hinder the mobility of the crosslinked epoxy network, resulting in an increase of T_g. The restriction of mobility of a polymer matrix by modified nanofillers has also been reported by previous research [23, 68]. In addition to the shifted T_g value, tan δ peak value also indicates the mobility of polymer chains and interfacial energy dissipation of polymer nanocomposites. The tan δ peak value decreased when nanofillers were incorporated into the epoxy nanocomposites, and the fAC reinforced samples exhibited the lowest tan δ peak value. The results can be explained by the limitation of polymer chain movement and reduced energy dissipation by the incorporated fACs [28, 72]. Therefore, the DMA tests of nanocomposites confirm the improved chemical and mechanical interaction between the epoxy matrix and CNC functionalized ANFs.