Enhanced mechanical performance and damping behavior of CFRP composites through exfoliated MWCNT functionalization

Abstract

In evaluating high-performance multiwall carbon nanotube (MWCNT) based carbon fiber composites, effective chemical functionalization of fillers for sufficient exfoliation and precise interface interactions with resin matrix are challenging to attain due to significant interlayer cohesive energy and inactive surfaces of composites. Herein, we demonstrate an effective way to produce carbon fiber-reinforced polymer (CFRP) composite with amended interfacial characteristics via incorporation of MWCNT through oxidation followed by silane functionalization with their exfoliation levels (0/48/60/72 h). The surface functional elements, morphological changes, elemental groups, quantitative information, and thermal degradation of silane functionalized MWCNTs were characterized by FESEM/HR-TEM, XRD, UV, and TG-DTA analysis, respectively. The influence of silane functionalized MWCNT reinforcement on CFRP's mechanical properties were examined using tensile, flexural, ILSS, and damping behavior was analyzed through impulse excitation technique (IET). The resulting silane-modified MWCNTs with 60 h oxidation (OAC-60) infused CFRP demonstrated significant enhancement in flexural strength (73.8%), tensile strength (58.2%), interlaminar shear strength (28.5%) compared to pure-CFRP.

GRAPHICAL ABSTRACT

Keywords:

• CFRP composite
• multiwall carbon nanotubes
• APTES silane functionalization
• mechanical performance
• oxidation
• flexural strength

1. Introduction

Carbon fiber-reinforced polymer (CFRP) composites have recently been used in a wide range of industries, including aerospace, automotive, and naval, due to their lightweight, high strength/weight ratio, good corrosion resilience, excellent thermal resistance, low manufacturing cost, etc. [1,2]. Research on CFRP composites is actively expanding, with a greater emphasis on creating and upgrading the composite properties [3]. Despite these exceptional capabilities, CFRPs were restricted by unavoidable fiber matrix delamination caused by weak matrix/fiber interfacial interaction and matrix cracking caused by poor fracture toughness [4,5]. Due to the chemical immobilization caused by the high-temperature graphitization/carbonization phase during manufacturing and the graphitic crystallized basal patterns with non-polar surfaces, carbon fibers cannot link strongly with the polymer matrix in the CFRP composite structure [6,7]. Weak bonding of the composite material leads to poor interface and fiber breakage challenges. These issues resulted in the formation of microfractures and lead to catastrophic structural damage [8]. It is widely known that improving the epoxy matrix fracture toughness helps in mitigating intra-laminar and interlaminar delamination at the laminated composite interfacial zone [9,10].

Generally, matrix transfers the imparted load to fibers and distributes the stress among the fibers [11]. To explain it another way, the CFRP’s mechanical properties can be determined by matrix specifications, carbon fiber surface treatment, and interfacial adhesion characteristics [12]. Strong connections between matrix and carbon fibers at the CFRP interface are required for effective shear transmission under load. The adhesion of matrix and fibers influences the interface load-bearing capability [13]. From the previous studies [11,13,14], it was identified that adding nanoparticles to matrix structure will increase the fracture toughness and will be a significant step towards overcoming above-said issues. Multiwall carbon nanotubes (MWCNTs) offer outstanding mechanical characteristics due to its extremely high surface energy, surface area, and aspect ratio [15]. Nevertheless, one of the most significant issues for particle-infused laminates is that during production, these nanoparticles form aggregations in the composite and cannot be dispersed uniformly [16,17]. The interface implications impact the load transfer between the reinforcement and the matrix. Fulmali et al. [18] compared the effect of pristine and carboxyl functional carbon nanotube (FCNT) reinforcement on the durability of glass fiber-reinforced epoxy (GE) composite under repeated hydrothermal cycling (HC) and compared with that of control GE composites. They reported that cyclic changes in water bath temperature caused interfacial de-bonding at the CNT/matrix interface resulting in accelerated water absorption and reduced flexural performance. The studies conducted by Mashhadzadeh et al. [19] believe that if MWCNT is exfoliated as a one atom thick walls, it can be used as an efficient reinforcer in high-performance CFRP composite with superior interconnection characteristics [19]. However, only a few researchers have shown such uniformly dispersed MWCNTs in the epoxy matrix and remarkably enhanced the mechanical characteristics of the resulting MWCNTs/epoxy nanocomposite [20,21]. Li et al. [22] strengthened the weak epoxy–amine network, by incorporating the functionalized MWCNTs along with acrylic tri-block copolymer (BMG). It was noted from the results that the maximum tensile and impact strength of epoxy-based composites with the addition of 7.1 wt.% BMG and 0.4 wt.% acid-treated MWCNTs were 89 MPa and 19.5 kJ/m², respectively. According to Lahi et al. [23] reinforcing 0.1, 0.5, and 1.0 weight percent of MWCNTs to epoxy enhanced the tensile strength of the samples but resulted in agglomeration of MWCNT. Kumar et al. [24] investigated epoxy nanocomposites that contained MWCNTs at concentrations of 0.25–1.0 wt.%, and several techniques were used to ensure uniform dispersion of the CNTs within the polymer matrix. The investigations mentioned above show that the homogeneous dispersion of pristine nanotube particles as reinforcement in hybrid CFRP composite is relatively low. The literature review on the MWCNTs hybridized CFRP composites showed that only few studies were conducted on the exfoliation rates to observe highly resilient CFRP composite laminates for industrial applications.

Fulmali et al. [25] improved flexural properties, interlaminar shear strength (ILSS), mode-I and II interlaminar fracture toughness (ILFT) by incorporating FCNTs in the GE composite. They reported 82% improvement in critical energy release rate of 0.1 FCNT-GE over the control GE composite. In a study conducted by Chou et al. [26], 4,4-methylene diphenyl diisocyanate (MDI) was grafted and prepared MWCNT/CFRP composites, to improve the dispersivity of MWCNTs in the matrix. They reported ∼28% increase in ILSS for functionalized MWCNT/CFRP over controlled specimen. Tefera et al. [27] prepared CFRP laminates reinforced with acid-treated MWCNTs and reported an increase in the flexural strength by 17.4%, 15.3%, and interlaminar shear failure stress by 14%, 7% at mixing times of 24 and 96 h, respectively. Fulmali et al. [28] prepared functionalized MWCNTs/GFRP composites and reported an exponential improvement of ∼47% in flexural strength, and ∼30% in ILSS over the control GE composite at elevated temperature conditions (ET). Zhao et al. [29] prepared oxidized MWCNTs reinforced CFRP composite and reported 10.22%, 15.14% increase in flexural strength, flexural modulus, respectively, compared to the control CF/EP composite. Numerous fabrication techniques, such as vacuum bagging, hand lay-up, and vacuum-aided resin infusion, have produced fiber-infused epoxy laminated composites. Due to ease of manufacturing, low cost, customizable mold sizes and shapes, vacuum bagging method is most frequently employed. Abdelal et al. [30] fabricated epoxy/carbon fiber composite laminates with various proportions of CNT. and sizing agent using hand lay-up assisted vacuum bagging process and reported 20% improvement in Young’s modulus.

Many attempts have been made to functionalize CNT through covalent connections using silane agents and the esterification process [31]. Silanization, also known as functionalization with a silane coupling agent, necessitates oxygenated chemical moieties named hydroxyl, carboxylic, and epoxide connected to the nanoparticles surface to aid condensation [20,32]. The oxygenated MWCNT is reported to be less prone to aggregation and, if effectively exfoliated, promotes homogenous distribution in the matrix materials. However, studies are found missing on the degree of oxidation to produce exfoliated MWCNTs from their aggregates, which is having sufficient oxygenated functional groups to attach silane functional moieties. At this point, we firmly believed that oxidation of the aggregated MWCNTs could serve our purpose of exfoliating MWCNTs and then attaching the silane moieties on them to use as a hybridizing agent in CFRP composite.

In order to fill the research gaps mentioned above and to synthesize unaggregated MWCNTs, in the present study MWCNTs underwent a one-step oxidization method with variable oxidation times. These oxidized MWCNTs later underwent amine-based silane moiety functionalization through an Aminopropyl triethoxysilane (APTS) silane agent. Topographical examination and spectroscopic studies were conducted to determine the exfoliation degree and functional group adhesion on the resulting functionalized MWCNTs. To assess the impact of functionalization on laminated hybrid CFRP, the amine-modified MWCNTs were initially diffused in a hardener/epoxy matrix using an ultrasonic-mechanical dual stirring process and then hybrid CFRP composite was prepared using vacuum bagging technique and further assisted in autoclave curing. The impact of oxidation time and amine-functionalized MWCNT infusion on CFRP composite were investigated using in-lane/out-of-plane flexural and tensile characteristics. The interaction behavior between the epoxy matrix/filler/fiber was evaluated using ILSS testing. Further, the damping characteristics of CFRP were investigated using the impulse excitation technique (IET). This study aimed to report on the most durable hybrid CFRP composite and to analyze the effect of different oxidation times on mechanical and damping behavior of composites.

2. Experimental section

2.1. Materials

The epoxy DGEBA resin of density 1.16 g/cc (Araldite, LY 556), accelerator (DY070), and MHHPA anhydride hardener with ρ-0.98 g/cc (HY906) were utilized as matrix materials and supplied by Huntsman Corporation. Nanoshell Intelligence Material Ltd provided multi-walled carbon nanotubes with 2–10 nm diameter range and 10–20 μm length. The silane coupling agent aminopropyl triethoxysilane (APTS) was supplied by Alfa Aesar India. HNO₃ (95% concentrated), toluene, ethanol, and acetone were provided by Merck Corporation, India. Plain weave unidirectional carbon fiber with 200 GSM and 1.82 g/cc density was used as reinforcing fibers, supplied by Composites Tomorrow India.

2.2. Amine-based silane functionalization of MWCNTs

Covalent amine-modified MWCNTs through APTS were carried out in two steps, oxidation and silane grafting, as shown in Figure 1. The pristine MWCNTs were oxidized through concentrated HNO₃ at 80 °C for various oxidation durations: 48, 60, and 72 h through continuous magnetic stirring and reflux condensation. The obtained resultant was washed using DI water until pH 7 and subsequent ethanol and acetone cleaning through a moisture trap. Then, the resultant was placed in a vacuum chamber for 12 h at an 80 °C temperature and named O-MWCNTs. Further, O-MWCNTs were silanized with APTS (1 wt.% of O-MWCNT) in toluene using an ultrasonic probe sonicator, followed by reflux condensation at 110 °C for 8 h.

Figure 1. Representation of amine-based covalently bonded MWCNTs.

APTS is a silane coupling agent used to attach amine functional groups through hydrolysis and condensation reaction between silane oligomers and the hydroxyl groups of oxidized MWCNTs. The samples were diluted and washed using toluene, ethanol, and acetone, followed by curing at 70 °C for 24 h in a vacuum atmosphere. The resultant surface silanized MWCNTs (48, 60, and 72 h) were coded as OAC-48, OAC-60, and OAC-72, respectively.

2.3. Preparation of CFRP composite with silane functionalized MWCNTs

Initially, 0.5 wt.% of pristine MWCNTs were dispersed in the HY906 (anhydride hardener) via ultrasonic-mechanical dual mixing. The double mixing process was carried out using an ultrasonic probe sonicator with subsequent mechanical stimulation at 600 rpm. The ultra-sonication (with titanium alloy probe) was carried at 69% amplitude for 10s pulse on and 10s pulse off time for 30 min. A flowing water chiller bath was used during the dual mixing process to keep the temperature below 50 °C. After dual mixing, DGEBA resin and DY070 accelerator were infused by ensuring a weight mixing ratio of 100:95:2 among DGEBA/HY906/DY070, mixed at 1200 rpm using a high-speed mechanical stirrer. The resultant solution was then degassed to release the trapped air that had been trapped during mixing under a vacuum desiccator for 40 min. The mixture was infused into the 8-layer stack of unidirectional carbon fiber mats using the vacuum bagging process. The pristine MWCNT-infused hybrid CFRP composite underwent pre-curing at 110 °C/2 h and 150 °C/8 h 70 °C for 24 h post-curing using autoclave vacuum pressure at 2 bar and graded as CFRP-Pr(0.5). The preparation of neat CFRP and functionalized MWCNT-infused CFRP with OAC variations used the same procedure mentioned above. Table 1 describes the CFRP composite codes and their respective volume fractions.

Table 1. Pure and functionalized MWCNT reinforced CFRP composite with various oxidation times and their respective volume fraction.

Composite type	Oxidation duration	Nano filler oxidation	Nano filler (wt.%)	Volume fraction (%)
				Fiber	Matrix	Nano filler content
Pure-CFRP	–	–	–	38.80	61.2	0
CFRP-Pr(0.5)	0	OAC-0	0.5	38.42	61.19	0.384
CFRP-Pr(1.0)	0	OAC-0	1.0	38.27	60.94	0.779
CFRP-Pr(1.5)	0	OAC-0	1.5	38.0	60.76	1.17
CFRP-Si48(0.5)	48/APTS	OAC-48	0.5	38.42	61.19	0.384
CFRP-Si48(1.0)	48/APTS	OAC-48	1.0	38.27	60.94	0.779
CFRP-Si48(1.5)	48/APTS	OAC-48	1.5	38.0	60.76	1.17
CFRP-Si60(0.5)	60/APTS	OAC-60	0.5	38.42	61.19	0.384
CFRP-Si60(1.0)	60/APTS	OAC-60	1.0	38.27	60.94	0.779
CFRP-Si60(1.5)	60/APTS	OAC-60	1.5	38.0	60.76	1.17
CFRP-Si72(0.5)	72/APTS	OAC-72	0.5	38.42	61.19	0.384
CFRP-Si72(1.0)	72/APTS	OAC-72	1.0	38.27	60.94	0.779
CFRP-Si72(1.5)	72/APTS	OAC-72	1.5	38.0	60.76	1.17

2.4. Mechanical performance assessment

The mechanical characteristics like tensile (ASTM D 3039, l × w − 160 × 150 mm), flexural (ASTM D 790, l × w − 60 × 12 mm) and short beam shear (SBS) (ASTM D2344, l × w − 15 × 5 mm) specimens of plain and MWCNT-Si (different oxidation durations) grafted composite materials were investigated using advanced equipment UTM a rate of speed 1 mm/min, 50 kN load cell. Three identical samples were examined for each result, and the average was considered for reporting. Equation (1)(1) $$\mathrm{\text{wof}}=\frac{A}{\mathit{\text{bh}}},$$(1) was used to determine CFRP wof (work of fracture) from a flexural test (1) $$\mathrm{\text{wof}}=\frac{A}{\mathit{\text{bh}}},$$(1) where b is the sample width, h is its thickness, and A is expressed in Joules to indicate the work required to deform/rupture the flexural specimen. The wof is measured in kJ/m². Equation (2)(2) $$\mathrm{\text{ILSS}}=\frac{3P}{4\mathit{\text{bh}}}$$(2) was used to determine the (2) $$\mathrm{\text{ILSS}}=\frac{3P}{4\mathit{\text{bh}}}$$(2) of the resulting hybrid CFRP composites. Where b – the width, h – the thickness of the SBS sample, and P –the maximum breaking load.

The IET and flexural vibration mode were considered to explore the CFRP composite’s damping performance. According to ASTM E1876, IET is a nondestructive impulse-based method for determining material dynamic elastic characteristics. The elastic modulus of the specimen is determined using IET resonant frequency, which also makes it possible to evaluate the specimen damping behavior. Figure 2 shows the experimental setup and schematics used in the current study. Equations (3)(3) $$\left(E\right)=0.9465\mathrm{ }\frac{{m\mathrm{ }f}_f^2}{b}\left(\frac{L^3}{t^3}\right)T_1.$$(3) and (5)(5) $$\left(Q^{-1}\right)=\frac{K}{{\pi f}_f}.$$(5) calculate the specimen’s dynamic elastic properties and damping factor. Dynamic elastic modulus (3) $$\left(E\right)=0.9465\mathrm{ }\frac{{m\mathrm{ }f}_f^2}{b}\left(\frac{L^3}{t^3}\right)T_1.$$(3)

Figure 2. Impulse excitation technique (IET) setup (a) schematic diagram (b) experimental setup.

For samples with L/t > 20, T₁ can be figured out using (4) $$T_1=\left[1.000+6.585\mathrm{ }\left(\frac{t}{L}\right)\right];$$(4) damping factor (5) $$\left(Q^{-1}\right)=\frac{K}{{\pi f}_f}.$$(5)

The relationship between the damping ratio and the damping factor (ζ) can be given by (6) $$\left(Q^{-1}\right)=2\mathrm{\zeta }.$$(6)

2.5. Characterization techniques

The morphological studies and EDX analysis of chemically functionalized MWCNTs were examined through (JSM-7600F FEG-SEM) field emission scanning gun electron microscope at five kV acceleration voltage. JEM2100F HR-TEM (high-resolution transmission electron microscope) was used to study the effectiveness of silane functionalization on MWCNTs surface.

3. Results and discussion

3.1. Morphological transformations of functionalized MWCNTs under FESEM study

A bifunctional molecule that reacts with the fillers and the polymer is needed to connect inorganic nanofillers and the CFRP matrix. To study the influence of different oxidation durations (48/60/72 h) and APTS silanization on pristine MWCNTs, we initially concentrated on morphological analysis using FE-SEM, as shown in Figure 3.

Figure 3. FE-SEM image of (a) pristine MWCNT (agglomeration of nanotubes), (b) OAC-48 showing nano-level ruptures (shown by arrow marks), (c) OAC-60 showing sufficient lateral and transverse exfoliation, (d) OAC-72 showing tiny transverse agglomeration.

In line with expectations, the FESEM image of the MWCNT-Pr (Figure 3a) showed an aggregation between several hundred nanotubes connected by a weak van der waals attraction. As MWCNT-Pr oxidization with nitric acid (HNO₃), a nanotube exfoliation is intended, resulting in the bonding of oxygenated chemical compounds such as O–H (hydroxylic), C = O (carbonyl), and COOH (carbocylic) on the surface. The oxidizing effect allows the MWCNT-Pr to benefit from de-aggregation by reducing the interfacial attraction forces [33]. Although the MWCNTs remained aggregated after 48 h of oxidation (OAC-48), attractively, nano-scale fissures are produced on the nanotube surface (Figure 3b shown by arrows). As shown in Figure 3c and d, these nano-level fissures continue to accelerate the lateral exfoliation of MWCNT after 60 h (OAC-60) as well as 72 h (OAC-72). Additionally, transverse separation is seen for OAC-60 under FESEM along with the lateral exfoliation of the MWCNTs, and the combination of these exfoliations creates a 3D networked nanotube structure with tiny openings. This 3D networking structure is anticipated to have two benefits: first, it will allow for the homogeneous silane oligomer grafting onto the nanotube edges and surfaces through silanization, and second, it provides greater access for epoxy resin crosslink to build a robust interface [34]. Significant transverse agglomeration is observed in the case of OAC-72, probably as a result of additional lateral dispersion of MWCNTs, which may have enhanced the surface energy of MWCNTs (Figure 3d). The sieving function of the fibers is expected to prevent infusion of CFRP intra-laminar zones that exhibit either or both transverse and lateral aggregation. In the hybrid CFRPs, the network of distinctive OAC-60s is predicted to be a more favorable interaction with the epoxy structure. It may show a greater ability to lay close to the carbon fibers effectively.

3.2. Analysis of HNO₃ oxidation followed by APTS silane functionalized MWCNTs

3.2.1. HR-TEM

To determine the efficiency of silane moiety infusion on the surface of MWCNTs, HR-TEM micrograph analysis of OAC (48/60/72 h) was performed, and the results are shown in Figure 4. With a minimal amount of layer exfoliation, the homogeneous layer of silane functional moieties adhering to the surface of MWCNT was observed for OAC-48, as depicted in Figure 4a. However, transverse direction exfoliation and the lateral separation of the carbon nanotube occur after oxidation for a period of more than 48 h. These phenomena are distinguishable in the OAC-60 and OAC-72 samples (Figure 4b, c).

Figure 4. HR-TEM micrographs of (a) OAC 48 h, (b) OAC 60 h, (c) OAC 72 h.

For OAC-60, there may be a noticeable amount of exfoliation of the silane layer-infused nanotube was observed. Furthermore, the OAC-60 surface roughness is more prominent than that of OAC-48 (shown by arrows). Therefore, the roughness can be ascribed to uniform silane functional moiety presence [35]. The MWCNTs multilayer structure was mainly preserved during the oxidation and silanization process. However, in the case of OAC-72, the lateral splitting of the MWCNT and the exfoliation of nanotubes have formed a step-like pattern discovered as aggregation after washing and drying. It may be concluded that effective adsorption of several oxygenated groups to the surface of MWCNTs during oxidation with HNO₃ for at least 60 h, followed by silanization, can lead to successful exfoliation of nanotubes.

3.2.2. UV spectroscopy

Investigating the optimum oxidation duration of MWCNT that leads to improved dispersion is necessary.

As an illustration, Figure 5 shows the impact of oxidation times on the absorption spectra of an MWCNT-Si dispersion. Each nanotube is functional in the UV–visible range and displays typical van Hove singularities. A strong absorption peak for MWCNT-Si can be seen at 230 nm due to the aromatic C–C bond π–π transition, while a smaller shoulder peak can be seen at about 260 nm due to the C = O in sp³ hybrid area transition [36]. The peak intensity increases with the oxidation time, indicating a higher dispersion quality at 60 h. The MWCNTs are probably not effectively dispersed by 48 h oxidation, while 72 h oxidation may result in water evaporation and unreliable concentration findings.

Figure 5. UV spectroscopy of functionalized MWCNTs.

3.2.3. EDX analysis and elemental mapping

The mapping regions of silanized MWCNT (OAC-60) with C, O, and Si elemental investigations along with EDX was studied, as shown in Figure 6.

Figure 6. (a–d) Elemental mapping of C, Si, and O. (e) EDX analysis of functionalized MWCNTs.

In the case of OAC-60, ‘C’ was the principal element. The oxygen particle was detected in the Si–O–Si bond of APTS and was equally dispersed throughout the mapping zone. The formation of covalent bonds is the fundamental reason for the silane moieties’ adsorption to the MWCNT surface [37], and Figure 6(a) reveals uniformly attached silane functional moieties on the OAC-60 surface. The energy dispersion spectroscopy (EDS) examination revealed additional proof of the covalent modification of functionalized MWCNTs. The spectrum and elemental composition of OAC-60 are shown in Figure 6(e). Regarding the chemical composition of the samples, purity was asserted, no other contaminants were noted, and the carbon was predominant over 95% of the total. The atomic percentage ratio of (O:C) was found to be 0.0225. The data support the uniform adhesion of silane oligomers (O–H, C = O, and COOH) to MWCNTs surface.

3.2.4. X-ray diffraction analysis

X-ray diffraction of unprocessed (pristine), silane functionalized MWCNTs with different oxidation times were carried, and the results are shown in Figure 7. MWCNT exhibited an intense sharp diffraction peak at 2θ = 26.29° attributed to diffraction plane (0 0 2). This peak demonstrates that the MWCNTs graphite structure underwent acid oxidation followed by salinization without considerable damage. When the MWCNTs are treated with an acids/silane group, their crystallinity order weakens, and the diffraction pattern moves to lower angles [38]. The other distinctive diffraction patterns of MWCNT were observed at 2θ = 44.38° and 54.48°, which correspond to the diffraction planes (1 0 1) and (0 0 4), respectively. The functionalized MWCNT (OAC-60) diffraction peak intensity was significantly higher than other MWCNTs, showing that bi-product removal resulted in pronounced graphitic structure and distinctive MWCNTs along with precise exfoliation. According to the standard spectral data, the MWCNT's characteristic peaks are successfully confined (JCPDS/41-1487). MWCNTs mean crystal size was estimated using Debye–Scherrer’s formulae: (7) $$D=\frac{0.9\mathrm{ }\lambda }{\beta \mathrm{\text{cos}}\hspace{0.17em}\theta }\mathrm{ }\mathrm{\text{nm}},$$(7) where β – Braggs peak/FWHM = 2.24; D – mean crystal domain size; θ = diffraction reflection angle (radians); λ – wavelength of X-ray (0.154056 nm) for (Cu/Kα tube). Lattice strain results from crystal flaws and is a fraction of dispersed lattice constants. The ε-lattice strain was exposed from Stokes–Wilson (8) $$\mathrm{\epsilon }=\frac{\beta \mathrm{\text{cos}}\hspace{0.17em}\theta }{4}.$$(8)

Figure 7. XRD analysis of pristine and functionalized MWCNTs.

Dislocation densities (δ) were calculated using X-ray line enlargement, (9) $$\delta =\frac{1}{D^2}.$$(9)

The mean crystal size and lattice strain of MWCNT-Si(60) were found to be 4.05 nm and 0.0522, respectively, and the values are listed in Table 2. Carolina Rodríguez et al. [39] performed the XPS analysis of oxidized MWCNTs to quantify the surface elemental composition of MWCNTs and to understand the adsorption mechanism. They reported a significant increase in oxygen content due to the formation of oxygen-containing functional groups, C–O, C = O O–C = O. In line with the XPS data, the same oxygenated functional groups are confirmed with our earlier FTIR results of functionalized MWCNTs [20].

Table 2. Mean crystal size (nm), dislocation density (δ), and lattice strain (ε).

Sample code	Crystal domain size (nm)	Dislocation density (δ)	Lattice strain (ε)
MWCNT-Pr	3.08	0.1054	0.0173
MWCNT-Si(48)	3.47	0.0711	0.0259
MWCNT-Si(60)	4.05	0.0407	0.0522
MWCNT-Si(72)	3.89	0.0643	0.0396

3.2.5. TG–DTA analysis

TG/DTA was used to evaluate the thermal stability of pristine and silanized MWCNT (OAC-48/60 h) under an N₂ environment. The results are shown in Figure 8. Pristine MWCNTs hardly degrade at below 600 °C temperature, leaving a residue of about 96.8 wt.% due to moisture elimination at 100 °C. The silanized MWCNT (OAC-48/60 h) exhibits a progressive weight loss of around 12–14 wt.% throughout the temperature range [40]. At 330 °C, significant mass degradation was observed due to the evaporation of hydroxylic acid/carbonyl/carboxylic functional groups, followed by thrixosilane layer decomposition on the MWCNT-Si surface. At temperatures between 600 and 700 °C, the carbon skeleton underwent pyrolysis, a thermo-chemical process that further degrades MWCNT-Si. The increasing mass decomposition of the functionalized MWCNT provides evidence that the silane functional group was successfully infused onto the MWCNT surface. Furthermore, the DTA curves demonstrate that, particularly for OAC (48/60 h), the temperature associated with the maximal degradation rate is higher due to the stronger covalent connections generated between the silane moiety molecules and the surface of the MWCNTs.

Figure 8. Weight loss percentages of functionalized MWCNTs through TG–DTA analysis.

3.3. In/out-of-plane mechanical and fracture characteristics of functionalized MWCNT grafted CFRP composite

A higher possibility of their oxidation level exfoliation and silane functionalization of MWCNT to improve mechanical strength and delamination resilience in hybrid CFRPs has been theorized as a result of the large number of morphological alterations that were seen for functionalized MWCNTs under high definition microscope.

3.3.1. Tensile characteristics

To verify those above, the tensile characteristics of the hybrid CFRP laminates have been assessed, and characteristic change as a factor of oxidization duration, amine surface modification, and nanofiller reinforcement quantity are related to pure-CFRP. The obtained tensile responses are shown in Figure 9. The variation in stiffness among matrix, carbon fiber/epoxy matrix interphase, filler/matrix interface, and individual fiber stiffness are the factors responsible for the non-linear response of the stress/strain curve. A large portion of the applied load is initially handled by the matrix with excellent stiffness and minimal distortion, increasing the slope of the curve. Once the micro–nano crack initiation and propagation begin via the fiber/matrix interphase, the composite total stiffness will be reduced due to stiffness variation in the filler/matrix/fiber interphase [41].

Figure 9. (a) Tensile strength variation in CFRP composites. (b) Bar graph with tensile strength vs. modulus with oxidizing time and filler content variation. FE-SEM micrographs of tensile fracture surface (c) pure carbon fiber surface, (d) fiber pullout, (e) epoxy debris on the surface of composite, (f) fiber/epoxy de-bonding, (g) agglomerated MWCNT in epoxy.

The tensile strength (σ_s), elastic modulus (E) of the pure-CFRP were 790 MPa and 15.30 GPa, respectively, after reinforcing with MWCNT-Pr (1.0 wt.%), they were improved to σ_s = −1020 MPa and E = −25.80 GPa, respectively. Fractographic examinations with FE-SEM show extensive intra/interlaminar deformation; this was caused by the defect propagation and nucleation at the matrix/fiber junction, which enabled matrix/fiber de-bonding and ultimately resulted in fiber pullout [42]. By reinforcing MWCNT-Pr (1.0 wt.%) into the CFRP, reduced matrix/fiber de-bonding and fiber pullout are seen in Figure 9d (due to failure energy absorption at the interlaminar zone) from fractographic investigation compared to pure-CFRP. Still, it is insufficient to prevent the matrix/fiber de-bonding. This is attributed to the matrix cracking with micro crack development caused by the aggregated MWCNT-Pr that ultimately results in matrix/fiber de-bonding, as seen in the FE-SEM micrograph in Figure 9f. However, the tensile strength and modulus were further increased to 1090 MPa and 26.20 GPa, respectively, for CFRP-Si48(1.0) due to the effective hybridization of matrix/fiber with silanized MWCNT, which can transfer the load uniformly throughout the composite.

For CFRP-Si60(1.0), the tensile properties (σ_s, E) were improved to 1250 MPa and 130.60 GPa, respectively, concerning pure-CFRP. This can be attributed to the delamination resistance offered by the matrix/fiber interlaminar junction by forming a significant amount of resin debris both on the sides and surfaces of fibers, shown by the fractographic image (Figure 9e). Furthermore, threads are typically seen hidden by the epoxy resin matrix, emphasizing the limiting of defect propagation and nucleation at the matrix/fiber junction. This is due to improved grafting of the special OAC-60s among sieves of CFRP interlayer laminates, which led to an effective interfacial connection between matrix/fibers/nanofillers and improved the bonding strength of the CFRP laminate. Sharma et al. [43] prepared carbon fiber-reinforced composite with graphene oxide (GO) nanosheet reinforcement and reported 750 MPa tensile strength, but our work demonstrated much improvement in tensile strength due to MWCNTs functionalization. Furthermore, for CFRP-Si72(1.0), the tensile strength (σ_s) and elastic modulus (E) were decreased to 1120 MPa and 28.70 GPa concerning CFRP-Si60(1.0). From the CFRP-Si72(1.0) FE-SEM micrographs, fiber pullout and matrix cracking were observed due to transverse agglomeration of MWCNTs and inadequate interfacial contact between the matrix/fibers (Figure 9g). In addition to the preceding, when the filler percentage was increased beyond 1.5 wt.%, all the hybridized CFRP versions had decreased tensile characteristics due to agglomerated MWCNT. The surface morphology of CFRP-Si60 with more excellent filler content exhibits bare fiber surfaces, minimal debris, weakly integrated bulk resin matrix/fibers, and aggregated MWCNTs. Even after silanization and probe sonication, the viscosity increase in epoxy increases the stress-raising impact in the interlaminar zone, resulting in diminished tensile characteristics. From the data, CFRP-Si60(1.0) exhibited superlative tensile properties than other composites.

3.3.2. ILSS through short beam shear test

ILSS of the hybrid CFRP was further evaluated using the three-point short beam strength (SBS) test to understand the impact of silanized MWCNTs on the delamination criteria. Shear stress in the SBS test causes interlaminar shearing cracks when it reaches the threshold at mid-thickness. The abrupt decrease in peak load can be used to estimate this maximal shear stress. However, because of the unidirectional FRP's inadequate load-bearing capacity in the transverse direction, intense compressive stress concentration at the loading point might cause fiber breakage and resin micro-cracking. We examined the load–deformation curves better to comprehend the ILSS failure behavior of hybrid CFRP. Figure 10 depicts the difference in the load–deformation angle of the combination CFRP; at least three test samples were used for each. The ILSS graph exhibits a mixed mode of fracture due to repetitive rise and fall in load, predominantly caused by matrix/fiber cracking due to weak transverse load capacity [44]. On the other hand, all the hybrid CFRP curves exhibit a rapid load reduction, followed by a slow decline, which is a sign of the interlaminar mode of shear failure. This behavior demonstrates the benefit of every MWCNT type on crack inhibition in a transverse orientation [45]. The ILSS is therefore calculated based on the maximum load of the graphs and shown in Figure 10a, b; the values are shown in Table 2.

Figure 10. (a) SBS sample load vs. displacement curves of pure and functionalized MWCNTs (1 wt.%). (b) Bar graph of interlinear shear stress with varied filler content.

The ILSS of the pure-CFRP was 42.32 MPa, which is enhanced by 14.29% (48.37 MPa) for CFRP-Pr(1.0). The CFRP-Pr may have been unable to infuse due to large-scale aggregation since they remained clustered at the lamina interlayers. Furthermore, the ILSS of CFRP-Si48(1.0) and Si60(1.0) show a significant improvement by 19.30% (50.49 MPa) and 28.18% (54.29 MPa), respectively. The ILSS strength, however, exhibits a decreasing trend due to a high silanization time of 72 h and a high filler content (1.5 wt.%). The minor or lack of ILSS deterioration for the CFRP with OAC-60 and OAC-72 at a larger reinforcement ratio is due to the lower stress-raising impact produced by silanization in the intra-laminar region.

Pathak et al. [46] revealed that 0.3 weight percent of GO mixed with epoxy matrix increased the CFRP ILSS by about 23%. Our result thus exhibits a significantly better enhancement in ILSS strength than previously reviewed literature, which justifies a stronger interlaminar shear resistance. At the matrix/fibers intersection, we anticipate networks of distinct OAC-60 that responsibly increase the shear resilience by stiffness modification caused by the amine-based silane oligomers. The reinforced MWCNTSi (silane oligomers) positively impact the epoxy surface energy via H-bonding and improve wetting by strengthening their chemical functional group interaction with the fiber surface.

3.3.3. Flexural

A three-point bend test has been carried out to examine the out-of-plane characteristics of the hybrid CFRP relative to pure bending conditions. The covalent bonds that form between the polymer and MWCNT provide more significant interfacial interaction and total exfoliation, significantly amplifying their impact on the characteristics of the composite. It is generally known that the interface between polymer matrix/fiber, depth stress gradient, and intrinsic stiffness significantly influences flexural characteristics. The variance in flexural traits is shown in Figure 11a–d and the mechanical properties are shown in Table 3. The flexural strength, flexural modulus, and work of rupture for the neat and CFRP-Pr(1.0) are 720 MPa, 48.10 GPa, 34.23 kJ/m², and 903 MPa, 59.45 GPa, 44.62 kJ/m², respectively.

Figure 11. (a–b) MWCNT reinforced CFRP composite, variation in flexural strength. Bar graphs: (c) flexural strength and flexural modulus, (d) work of fracture (kJ/m²).

Table 3. Mechanical properties of oxidation followed by silane functionalized MWCNT reinforced CFRP composites.

Sample grade	Tensile strength (MPa)	Tensile strain (%)	Tensile modulus (GPa)	Flexural strength (MPa)	Flexural modulus (GPa)	Work of fracture (kJ/m^2)	ILSS (MPa)
Pure-CFRP	790 ± 35	10.15 ± 0.67	15.30 ± 1.07	720 ± 32	48.10 ± 2.60	34.23 ± 1.20	42.32 ± 1.15
CFRP-Pr(0.5)	940 ± 51	8.31 ± 0.69	24.27 ± 1.14	812 ± 47	54.66 ± 4.30	38.79 ± 1.40	46.27 ± 2.33
CFRP-Pr(1.0)	1020 ± 59	9.21 ± 0.85	25.80 ± 1.63	903 ± 58	59.45 ± 4.60	44.62 ± 1.50	48.37 ± 1.76
CFRP-Pr(1.5)	780 ± 49	8.05 ± 0.74	23.15 ± 1.82	846 ± 52	58.86 ± 3.70	39.69 ± 1.10	45.29 ± 1.85
CFRP-Si48(0.5)	980 ± 55	7.25 ± 0.52	25.74 ± 1.02	891 ± 56	60.66 ± 3.90	42.35 ± 1.80	49.87 ± 1.93
CFRP-Si48(1.0)	1090 ± 60	7.98 ± 0.67	26.20 ± 1.48	1016 ± 61	66.19 ± 4.10	47.14 ± 1.90	50.49 ± 2.45
CFRP-Si48(1.5)	910 ± 57	9.95 ± 0.89	24.89 ± 1.72	902 ± 59	58.48 ± 3.20	41.66 ± 1.40	48.62 ± 1.66
CFRP-Si60(0.5)	1050 ± 61	9.82 ± 0.91	28.63 ± 1.89	1025 ± 62	67.55 ± 5.60	48.64 ± 2.10	53.70 ± 2.05
CFRP-Si60(1.0)	1250 ± 68	11.16 ± 1.03	30.60 ± 2.15	1252 ± 67	74.27 ± 6.20	54.85 ± 2.30	54.25 ± 3.11
CFRP-Si60(1.5)	1010 ± 59	8.04 ± 0.75	27.24 ± 1.73	1072 ± 61	65.65 ± 4.50	47.91 ± 1.60	52.40 ± 2.10
CFRP-Si72(0.5)	1030 ± 63	7.68 ± 0.58	26.61 ± 1.57	933 ± 60	64.45 ± 5.20	46.82 ± 1.40	50.41 ± 2.35
CFRP-Si72(1.0)	1120 ± 65	9.34 ± 0.80	28.70 ± 1.95	1114 ± 64	68.61 ± 4.60	50.05 ± 1.80	51.94 ± 2.84
CFRP-Si72(1.5)	980 ± 56	6.82 ± 0.63	24.90 ± 1.48	908 ± 58	63.10 ± 3.90	46.24 ± 1.20	50.15 ± 2.50

These values have increased greatly by 41.1%, 37.6%, and 37.7%, respectively, for CFRP-Si48 and 73.8%, 54.4%, and 53.8%, respectively, for CFRP-Si60 about pure-CFRP. Sanchez et al. [47] demonstrated the impact of CNT on CFRP flexural strength, and instead, our work highlights how OAC-60 has a distinctive advantage in improving flexural properties. Furthermore, a small decrease in properties was observed for CFRP-Si72 when compared to CFRP-Si60. From the above, it was evident that the difficulty of distributing nanofillers into polymeric matrix often poses a significant barrier to using high concentrations. Due to their inherent propensity to aggregate, MWCNTs. have a viable mixing weight percent of less than 1.5 wt.%.

According to the flexural graph, the matrix containing OAC-60 can enhance the composite’s load-bearing capacity and minimize deformation at the matrix/fiber interface. As a result, the biggest improvement in flexural characteristics for CFRP-Si60 provides validation that the 60 h oxidation time followed by silanization enhances the uniform dispersion throughout carbon fibers and improves CFRP stiffness due to the hydrophobic nature of silane moieties as shown in Table 4.

Table 4. Comparison of current work with earlier published studies.

Composite type	Oxidation time (h)	Property considered	Improvement in our work (%)	Reference
FMWCNTs/CFRP	60	Flexural strength − 1252 MPa	–	Current study
FMWCNTs/CFRP	60	ILSS − 54.85 MPa	–	Current study
FMWCNTs/CFRP	24	Flexural strength − 948.0 MPa	∼31	[27]
FMWCNTs/CFRP	24	Flexural strength − 987.0 MPa	∼27	[29]
FMWCNTs/CFRP	24	Flexural strength − 860.0 MPa	∼45	[11]
FMWCNTs/CFRP	12	ILSS − 48.79 MPa	∼12	[26]
FMWCNTs/GFRP	24	ILSS − 36.78 MPa	∼49	[45]

3.3.4. Damping behavior of hybrid CFRP composite

Figure 12 shows the vibration signals and corresponding frequency spectra and Table 5 shows the damping factor and elastic modulus of pure and functionalized MWCNT-infused CFRP composite through flexural and IET test. The findings reveal that the resonance frequency of CFRP was increased with silanized MWCNT reinforcement. This can be associated with reducing the specimen’s vibrational response by an increase in the dissipation of vibration energy throughout the structure [11]. The resonant frequency of pure-CFRP was lower than silanized MWCNT-infused CFRP can be attributed to a reduction in vibrational energy absorption.

Figure 12. Frequency domain and time domain spectrums of (a) pure and (b–d) functionalized MWCNT (1 wt.%) reinforced CFRP composites.

Table 5. Damping factor and elastic modulus of pure and functionalized MWCNT-infused CFRP composite through flexural and IET test.

Composite type	Resonant frequency (Hz)	Damping factor	Elastic modulus (GPa)
			Through IET	Through flexural
Pure-CFRP	4348.15	0.0029	47.52 ± 2.15	48.10 ± 2.60
CFRP-Si48(1.0)	4252.17	0.01324	64.82 ± 3.74	66.19 ± 4.10
CFRP-Si60(1.0)	4150.98	0.0614	72.55 ± 5.01	74.27 ± 6.20
CFRP-Si72(1.0)	4128.79	0.01532	64.59 ± 3.95	68.61 ± 4.60

Equations (3)(3) $$\left(E\right)=0.9465\mathrm{ }\frac{{m\mathrm{ }f}_f^2}{b}\left(\frac{L^3}{t^3}\right)T_1.$$(3) and (5)(5) $$\left(Q^{-1}\right)=\frac{K}{{\pi f}_f}.$$(5) are used to calculate the CFRP's elastic modulus and damping factor using their respective frequency domain data. The damping factor of CFRP-Si is shown in Figure 12b–d. The findings demonstrate that the inclusion of silanized MWCNTs increases the damping factor of CFRP composite. A noticeable improvement in the damping factor was observed for CFRP-Si60(1.0) because of adequate crack energy, which improves the composite’s energy absorption capability.

In addition to MWCNT-Si wt.%, the viscoelastic properties of the epoxy resin and the interfacial connection between the matrix and the MWCNT-Si also affect the damping factor. In general, flaws such as micro voids, dislocations, and different interfaces spread during vibrations or are vulnerable to slippage, which promotes to dissipation of vibration energy. The lower damping factor of pure-CFRP, compared to MWCNT reinforced, can be attributed to CFRPs natural frequency because it is more significant than pure-CFRPs natural frequency.

4. Conclusions

In summary, we have fabricated and characterized a hybrid CFRP composite with exfoliated MWCNTs through various oxidation times followed by amine functionalization as reinforcement. The influence of oxidation level followed by amine-based silane functionalized MWCNTs (various wt.%) on the mechanical properties (tensile, flexural, and ILSS) and damping behavior through IET was evaluated. It was observed that the oxidation level significantly affected the de-aggregation of MWCNTs. The FE-SEM and HR-TEM analysis showed that the aggregated MWCNTs experienced sufficient transverse and lateral exfoliation over 60 h of oxidation (OAC-60) and created a 2D networked nanotube structure. The FTIR, XRD, and UV confirm that the hydroxylic, carbocylic, and carbonyl functional elements in OAC-60 justify the effective silane moiety grafting on the MWCNTs surface, which promotes uniform dispersion. With the incorporation of 1 wt.% OAC-60 in CFRP laminates, using sonication and vacuum bagging, significant improvement in tensile strength (1250 MPa) and modulus (30.60 GPa) about pure-CFRP was observed. The improved flexural strength (1252 MPa), modulus (74.27 GPa), and wof (54.85 GPa) with regard to pure-CFRP were observed for CFRP-Si60(1.0) because of the high stiffness, load-bearing capability of the composite, which resulted in less deformation at the fiber/matrix/filer interface. The interlaminar shear response of CFRP-Si60(1.0) composite exhibited transverse damage inhibition because of high shear resistance induced by the hydrophobic nature of silane moieties. The resonant frequency of CFRP-Si60(1.0) resulted in high damping factor because of increased vibrational energy absorption throughout the composite. Thus, this research provides a comprehensive analysis of the OAC-60 exfoliation and potential usage by industries as a nanofiller reinforcement to regulate the stiffness at the fiber/matrix interface of CFRPs in high-performance applications.

Author’s contribution

The authors confirm their contribution to the article as follows:

Conceptualization, funding acquisition: Goteti Dhanaraju, B. Satish Ben; Data curation: Goteti Dhanaraju, B. Satish Ben; Investigation: Vinay Atgur, Goteti Dhanaraju, M. A. Umarfarooq; Methodology: Goteti Dhanaraju, B. Satish Ben, Raj Kumar Pittala; Validation: Goteti Dhanaraju, B. Satish Ben, Tabrej Khan, Balbir Singh; Visualization: M. A. Umarfarooq, Tabrej Khan, Vinay Atgur, Balbir Singh; Roles/writing – original draft: Goteti Dhanaraju, B. Satish Ben, Raj Kumar Pittala; Writing – review & editing: Goteti Dhanaraju, Raj Kumar Pittala, N. R. Banapurmath; Supervision: Goteti Dhanaraju, B. Satish Ben, N. R. Banapurmath. All authors reviewed the results and approved the final version of the manuscript.

Acknowledgments

The authors would also like to acknowledge I.I.T. Bombay for supplying TG-DTA, HR-TEM, and FE-SEM data and CRIF NIT Warangal for providing XRD and other essential equipment. The authors would also like to thank Prince Sultan University for their support.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

Additional information

Funding

The authors thank the financial assistance given by DST-SERB under DST-SERB Program 2020 against Grant No. E.E.Q./2020/000105.