MAX phase ternary carbide derived 2-D ceramic nanostructures [CDCN] as chemically interactive functional fillers for damage tolerant epoxy polymer nanocomposites

Abstract

A 2-dimensional ceramic nanostructure was successfully processed out of nanolamellar 312 MAX phase ternary carbide, titanium silicon carbide, Ti₃SiC₂ (TSC), via a simple shear-induced delamination technique. It has been explored as a functional nanofiller for obtaining chemically homogeneous, low-friction, self-lubricating epoxy nanocomposites. The structural characterization of the MAX phase Carbide Derived Ceramic Nanostructure (CDCN) was carried out using Dynamic Light Scattering (DLS), Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) analysis. Subsequently, CDCN was mixed with Araldite CY 225 (DGEBA) at different percentages and thermally cured using Aradur HY 925 hardener at 130 °C to make epoxy–Ti₃SiC₂ nanocomposites. The effect of CDCN-nanofiller was studied on epoxy rheology, glass transition temperature (T_g), thermal stability, flexural and compressive strengths, microhardness, dry sliding wear and friction properties. It was found that, unlike other ceramic fillers, CDCN chemically interacts with epoxy and readily dispersed in a polymer matrix without any deleterious structural defects. It resulted in the formation of physico-chemically homogeneous microstructures. Epoxy composites prepared with CDCN filler attained 50% more mechanical strength and hardness. Wear analysis trends indicate Ti₃SiC₂ nano reinforcement possibly formed a lubricating tribo-chemical film that decreases the wear rate and coefficient of friction. This work is significant in such a way that a novel nanofiller has been identified from MAX phase carbide family which offers a self-lubricating interface and produces mechanically reliable, damage tolerant epoxy composite for state-of-the-art engineering applications.

---

1. Introduction

Nanomaterials capable of producing ‘smart-microstructures’ that seemingly offer noteworthy beneficial properties (e.g. faster thermal dissipation, enhanced wear resistance, self-lubrication, crack healing and damage tolerance) have tremendous potential as multifunctional nanofillers for fabricating thermo-mechanically stable structural nanocomposites. Design of intricate structural parts using epoxy nanocomposites is steadily growing in the aerospace and automobile fields. It is also emerging as a dielectric material to substitute porcelain and glass insulators.^1,2 A major drawback of the epoxy polymer is its low thermal dissipation property due to poor thermal conductivity (λ = 0.2 W mK⁻¹), high thermal expansion coefficient (α = 66.5 μm m⁻¹ °C⁻¹) together with weak thermal shock resistance and profusely poor crack propagation resistance that often leads to catastrophic mechanical failures. These undesired thermo-mechanical properties ultimately restrict the use of this economically benign glassy polymer for engineering applications. It calls for the design of new multifunctional fillers to make strong, tough and reliable epoxy composites.

A range of nanoparticles such as carbon,³ graphene,^4–6 Al₂O₃,⁷ SiO₂,^8,9 SiC,¹⁰ AlN¹¹ and BaTiO₃ ^12,13 were widely attempted in epoxy matrix and studied. Recently, trend is drifting to non-conventional, multifunctional nanostructures in order to transform epoxy a highly competent material for engineering applications. Chemically cross-linked inorganic–organic hybrid nanofillers are one of the examples that have been tried in epoxy in order to impart function properties. Nano SiO₂ filler simply pretreated with polysiloxane introduced hydrophobic character that reduced the surface wettability and increased the water, moisture and humid repellency. In the series of functional-fillers, carbon nanostructures, e.g. graphene,^14,15 carbon nano tube (CNT),^16,17 BN nanotubes,¹⁸ graphene oxide^19–21 are extensively studied and well explored. Incorporation of these fillers was found successful to achieve increased thermal and electrical conductivities.^22–25 Studies on CNT nanofiller is centrally focused on production of mechanically reliable epoxy composites. The effect of CNT volume fraction and its directional orientation in enhancing the bulk elastic modulus and compressive strength were studied in detail. Interestingly, addition of single walled carbon nano tubes (SWCNT) is found to increase the mechanical strength from 88 to 128 MPa.²⁶ Zhang et al. studied the effect of graphene oxide [GO] modified carbon fiber and reported that GO–carbon fiber fillers also resulted in improved mechanical strength.²⁷ This work was also demonstrated via solvent-free approach to have high strength epoxy–SWCNT composites.²⁸ Li Chen et al. has attempted similar work with GO coated SiO₂ nano fillers.²⁹ Studies concerned with CNT and graphene fillers confirm that greater interfacial interaction of CNT, graphene nanostructure was the reason to have enhanced mechanical performance. Though, they are prospective fillers, it is inevitable that an appropriate surface treatment is needed to achieve strong interfacial interaction.^30,31 Even with better interfacial interaction, the ultimate mechanical strength fluctuates strongly with the orientation and hence reproducibility pose a technical challenge.

A novel variety of nano material, namely ‘carbide derived carbon nanostructure’ (CDCN), a tunable nano porous carbon, is up-coming as functional nano fillers. It is carefully synthesized from SiC, TiC and ternary carbides. Ternary carbides fall in MAX phase family. Michael Barsoum's group in Drexel University, USA^32–35 has studied these materials in-depth to expose the various beneficial properties for structural applications. The MAX phase compounds are crystalline materials with a laminar microstructure of hexagonal symmetry. Titanium silicon carbide (Ti₃SiC₂) is a nano-laminate ternary ceramics. This MAX phase received tremendous attention because it exhibits both metals and ceramics properties.³⁶ It possess high thermal and electrical conductivities like metals and shows high temperature stability like ceramics. A unique nature in Ti₃SiC₂ is its soft nano laminate sheets like microstructure which are stalked together through covalent interactions and can offer ductility, super plasticity and damage tolerance via kink-band formation upon any mechanical stress.^37,38 It is reasonably low dense (4600 kg m⁻³) and exhibits high thermal conductivity (54 W mK⁻¹). Till date, the beneficial properties of Ti₃SiC₂ nano laminates as fillers have not been examined in epoxy composites. Our group has already explored the multifunctional properties of Ti₃SiC₂ for enhancing the performance of polyaryletherketone (PAEK) nanocomposites.³⁹

It is expected that Ti₃SiC₂ can improve both mechanical and tribology behavior of epoxy polymers. In this work, nano lamellar Ti₃SiC₂ was first delaminated via mechanical shear to form 2-D nano layered Ti₃SiC₂, which was then explored as functional filler for epoxy composites. The role of Ti₃SiC₂ as nano filler is addressed with respect to thermal, mechanical, dielectrical and tribological properties. The wear and friction analysis has been carried out at various loads to understand the role of Ti₃SiC₂ nanostructures. The capability of this filler in rendering multifunctional benefits over microstructural homogeneity, chemical interaction, mechanical strength, hardness, tribology and dielectric permittivity was studied and reported.

2. Experimental

2.1 Materials

Ti₃SiC₂ particles with an average diameter D₅₀ = 13 μm was purchased from M/s 3-One-2 LLC, USA. Epoxy resin (Araldite CY 225, DGEBA) and aromatic anhydride (Aradur HY 925) hardener were procured from Huntsman Resins, Switzerland. N,N-Dimethyl formamide (DMF, 99% purity) and Triton-X 100 were purchased from Nice Chemicals, India. Nano alumina from Aldrich chemicals.

2.2 Preparation of Ti₃SiC₂ nanostructures

Initially, bulk Ti₃SiC₂ powder was washed in 2 N hot HNO₃ and then sonicated in Triton–water mixture for 30 minutes in order to remove the associated surface impurities formed due to surface oxidation. During this step, de-agglomeration of the particles was envisaged. After vacuum drying at 60 °C, the particles were mildly stirred in dimethyl formamide (DMF). The powder slurry was then transferred into Retzch mechanical planar grinder, rotating at constant speed of 150 rpm. It has a stationary dense ceramic bowl made up of cubic zirconia having a capacity of 200 mL and a freely rotating ceramic pestle. There was no extra grinding force given for delamination and maintained the free-rotation of pestle. However, the slurry was frequently sonicated during the course of grinding. The time given for grinding is 10 min to 24 h. The DMF medium was screened by the periodical examination of the particle morphology under SEM and the dispersion stability in water, poly ethylene glycol, silicone oil and DMF. A highly stable (up to 3 months shelf-life) Ti₃SiC₂ colloid suspension is obtained in DMF. It was finally tested for laser light scattering, morphology and particle size distribution analyses to confirm its physical nature.

2.3 Fabrication of epoxy–Ti₃SiC₂ composites

Epoxy resin (100 parts) was weighed and mechanically stirred in a stainless steel container at a temperature of 50 °C. Ti₃SiC₂ nano filler were progressively introduced into easily flowing, low-viscous epoxy resin. The stirring speed was maintained at 750 rpm. After 15 min of stirring, hardener (80 parts) was added slowly and continued stirring until a homogeneous epoxy composite melt was formed. An intermediate degassing was performed at 60 °C for removing the trapped air before starting the thermal-curing. The de-gassed epoxy melt was then transferred into a pre-heated metallic mould kept at 150 °C in order to prevent any preferential settling of the fillers, if any. After casting, the epoxy composite was cured at 100 °C for 5 h followed by post curing at 130 °C for 8 h. The epoxy–Ti₃SiC₂ nano composites prepared with different wt% Ti₃SiC₂ CDCN were designated as ET5, ET10, ET20 and ET30 which correspond to 5 wt%, 10 wt%, 20 wt% and 30 wt% of filler addition respectively. ‘E’ stands for neat epoxy without any CDCN.

2.4 Characterizations

Phase purity, particle size distribution and particle morphology were assessed using powder X-ray diffraction (Philips X'pert pro, CuKα, λ = 1.54 Å, The Netherlands) and scanning electron microscopy. JEOL 5600 SL scanning electron microscope, Japan and JEOL 200 CX, transmission electron microscope, Netherlands were used for morphological studies. Both polished and fractured surfaces were examined for better morphological analyses. Thermal analysis of Ti₃SiC₂ and epoxy–Ti₃SiC₂ nanocomposites was carried out in TGA-50, SHIMADZU thermal analyzer, Japan at a heating rate of 10 °C min⁻¹ up to 1000 °C. Glass-transition temperature (T_g) of the composites were measured using differential scanning calorimetry (DSC, 200F3, Netzsch, Germany) from room temperature to 170 °C with a heating rate of 10 °C min⁻¹. In order to overcome the mismatch due to thermal history, the sample was subjected to a second heating and then the T_g was calculated. The visco-elastic properties of the resin in presence of fillers were studied using Anton Paar Physica modulated Compact Rheometer 150 at 80 °C with varying shear rates in disposable cups of size 50 mm diameter.

2.4.1 Mechanical strength and hardness measurements. Flexural and compression tests were performed by the aid of INSTRON 5500 R Universal Testing Machine, USA, at a constant cross head speed of 2 mm min⁻¹. Test specimens for the flexural tests were prepared as per ASTM D6109 standard with a dimensions of 80 mm length, 5 mm width and 5 mm thickness. For compression tests, cylindrical samples with 12 mm diameter and 18 mm height were prepared as per ASTM D6641 standard. The samples were cut according to the ASTM standards using Buhler Low Speed Saw diamond cutter. Micro hardness of the composites were analysed using CLEMEX Vickers microhardness tester, Canada, with a Vickers diamond pyramidal indenter having a square base and pyramidal angle of 136°. The specimens were subjected to a load of 100 g with a dwell time duration of 15 s.

2.4.2 Tribology measurements. Wear and friction analyses were conducted using Tribometer model Macro PoD-TRI-201 LE, DUCOM instruments (India). A standard pin-on-disc method was followed in which the cylindrical samples having 10 mm diameter and 15 mm height were tightly rotated against a stainless steel plate (EN 31 hardened upto 62 HRC), with a track diameter of 70 mm at an rpm of 300 for 15 minutes. Tribological testings were done at different loads, viz. 1, 3, 5 and 7 kg at 27 °C and relative humidity of 76%. The coefficient of friction (COF) was calculated using the equation COF (μ) = F/mg, where F is the force acting on the object, m the mass of the object under motion, g = acceleration due to gravity (g = 9.81 m s⁻²). Specific Wear Coefficient (SWC) was calculated using the relation where V_i is the wear volume (mm³), F = normal load acting during testing (N), S = sliding distance (m).

2.4.3 Dielectric measurements. Dielectric permittivity of the composites were measured by LCR meter (Hioki 3532-50LCR HiTESTER, Nagano, Japan) from the frequency range 1 kHz to 1 MHz using the parallel plate capacitor method by applying silver paste on both sides of the samples having thickness of 1.5 mm and diameter of 11 mm.

3. Results and discussions

3.1 Morphology analysis of Ti₃SiC₂ nano platelets and its effect in epoxy matrix

The sequence of morphology changes in nanolamellar Ti₃SiC₂ during shear-induced delamination process was systematically analyzed using SEM, TEM and DLS measurements to confirm the nano Ti₃SiC₂ laminates. Fig. 1a shows the original morphology of a single, bare Ti₃SiC₂ particle. As a dry powder, Ti₃SiC₂ appears as a large size distribution of particles which agglomerate all together. Most of the volume of the powder is composed of large platelets of several microns large with the thickness in the nanometer range (about 500 nm thick and 5–30 microns in plane dimensions).

Fig. 1 (a) SEM image of as received Ti₃SiC₂ (b) SEM image of Ti₃SiC₂ after acid boiling (c) stable colloidal suspension of Ti₃SiC₂ in DMF after shear induced de-lamination associated with sonication (d) laser scattering of suspension indicating Tyndall effect (e) particle size analysis of delaminated Ti₃SiC₂ (f) TEM images of delaminated Ti₃SiC₂ and (g) TEM image at its higher magnification.

In Ti₃SiC₂, the Ti–Si inter-atomic bond length is approximately 0.269 nm.⁴⁰ The existence of such short length indicates the strong bond energy of the as-synthesized material. The boiling of material in hot HNO₃ under stirring (60 rpm) for extended period of time (24 h) chemically weaken the Ti–Si inter layer bonds. Fig. 1b shows the effect of both acid treatment and mechanical stirring on the particle morphology. Interestingly, small particles (<500 nm) with dark and small spherical morphology were observed. These are agglomerates of very small flakes stack on each other (about 5–250 nm in plane dimensions and several nm or less in thickness). The bond existing between the interlayers of the nanoplatelets of Ti₃SiC₂ weakens to a greater extent which explains the transition of nanoplatelets to nanometer flakes morphology. At this stage, the application of mechanical force induce shear on the surface of the small flakes staked on each other and delaminate the material that stand freely in dispersed state as single or with very few layers.

Fig. 1c shows the dispersion of Ti₃SiC₂ nanolayers in DMF medium. It was prepared with solids having concentration in the ratio 1 : 1 and even at this higher concentration, the fluid obeys Tyndall effect which was confirmed by the laser light scattering, a typical quality control describing the nano size colloidal quality of the suspension particulates (Fig. 1d).

The nano nature of delaminated Ti₃SiC₂ was further examined for the particle size distribution analysis by DLS technique (Fig. 1e). The size distribution curve confirms that the particles fall in 20 to 250 nm range. It indicates the formation of 2-dimensional particles having nano sized thickness, but has micron dimensions length, physically resemble sheet-like structure. The morphology of delaminated Ti₃SiC₂ nanolayer was seen under TEM wherein the sheet structure was more clear (Fig. 1f). However, it was observed that the employed shear-induced delamination is not very effective because the delamination has not occurred layer-by-layer. The magnified image in Fig. 1g indicates Ti₃SiC₂ nanolaminate assembly contains stacking of more than one layer.

The delamination or exfoliation of MAX phase materials to produce 2-D MXenes and MAXenes, similar to graphene, was earlier reported.⁴¹ For MXenes, the MAX phase Ti₃AlC₂ was employed and exfoliation was done via HF leaching. Works related to MAXenes was reported only once where the authors have used impurity assisted (e.g. Li ion) thermal-exfoliation through solid-solution route. In this work, a simple technique is found to yield nanostructures similar to MAXenes and MXenes. Till date such MAX phase carbide derived ceramic nanostructures (CDCN) have not been studied and reported as functional nano fillers for epoxy composite preparation. The unique benefit noticed with Ti₃SiC₂ based CDCN was its excellent chemical interaction with epoxy resin compared to other carbon nanostructures, e.g. CNT and graphene.

The interaction of Ti₃SiC₂ based CDCN was further evident from the viscosity analysis of epoxy–Ti₃SiC₂ composites. For comparison, epoxy resin was first examined without any nano fillers. It was then added with 10 wt% of nano Al₂O₃. The viscosity measurement was taken at 80 °C. The neat epoxy has the viscosity of 0.2 Pa s at shear rate upto 20 s⁻¹. At identical shear rates, the epoxy-nano Al₂O₃ composite showed viscosity >1.0 Pa s. In case of 10 wt% nano Ti₃SiC₂ additions, the viscosity was well controlled and it is retained as equal to that of neat epoxy showing that Ti₃SiC₂ based CDCN is well dispersed in the epoxy. The rheology curves prepared with nano Al₂O₃ and de-laminated Ti₃SiC₂ nano filler are presented in ESI Fig. S1.† The high surface area associated with nano Al₂O₃ produce agglomerates with high macroscopic viscosity, where as in Ti₃SiC₂ CDCN, the thin-sheet structure, its delamination during mechanical shear and stability as dispersed phase in resin suspension controls the viscosity.

Usually conventional ceramic nano fillers (e.g. SiO₂, Al₂O₃, ZnO and TiO₂) may chemically interact via surface –OH groups and produce homogeneous microstructures. However, CNT and graphene nanostructures need additional surface treatments with –OH functional groups and silanes providing a route for bonding between epoxy and nano particles. In Ti₃SiC₂, the Si-interlayer is found to interact with epoxy –OH groups which is originally exposed during epoxy-hardener reaction at elevated temperature. The finger print of Si–O and –OH bonds is clearly seen in FTIR spectrum given in Fig. 2. The appearance of respective characteristic peaks correspond to –OH, Si–O and –CH₂ bonds at wavelengths 3300, 1100 to 798 and 2850 cm⁻¹ confirms the Ti₃SiC₂ interaction with epoxy. The schematic chemical interaction is given in Fig. 3.

Fig. 2 FTIR spectra of bare Ti₃SiC₂ (TSC), neat epoxy (E) and epoxy with Ti₃SiC₂ (ET).

Fig. 3 Schematic representation of delamination of Ti₃SiC₂ and its chemical interaction with epoxy resin.

3.2 Effect of Ti₃SiC₂ on thermal stability and glass transition temperature

Such chemically interactive nano filler contributes marginally to enhance the thermal stability of epoxy. The TG analysis in Fig. 4a shows the weight loss due to thermal heating between neat epoxy and Ti₃SiC₂ CDCN blended epoxy. The neat epoxy has thermal stability up to 190 °C. Later, the thermal decomposition is gradually picked up and in general, the decomposition occurs in two stages. Stage I is accounted between 190 to 400 °C, where the neat epoxy has peak decomposition at 335 °C which is marginally improved to 366 °C. Compared to neat epoxy, the composite has significantly slow decomposition which is understood from the change of slope in TG curves. These trends conclusively indicated that the nanostructured Ti₃SiC₂ has advantages in increasing the molecular scale stability. The decomposition was found to be completed roughly at 550 °C. A weight gain was also seen in composites after 550 °C probably due to the oxidation of Ti₃SiC₂. The TG analysis of bare Ti₃SiC₂ shown in ESI Fig. S2,† clearly indicates the weight gain after 500 °C.

Fig. 4 (a) Thermo gravimetric analysis of neat epoxy and epoxy–Ti₃SiC₂ composites (b) DSC analysis and glass transition temperature variation of composites with respect to Ti₃SiC₂ filler loading.

The DSC curves in Fig. 4b shows the effect of Ti₃SiC₂ CDCN on the glass transition temperature (T_g) of epoxy composites. In epoxy polymer, the T_g value strongly varies with curing mechanism, cross linking density due to filler phase interaction, its volume fraction and orientation in the matrix. In this work, curing was performed at identical conditions. Hence, the filler interaction and its microstructure modification were expected to influence the epoxy T_g value. The DSC analysis showed that the neat epoxy has T_g value at 97 °C. It has been increased to 111 °C when 30 wt% Ti₃SiC₂ was incorporated in epoxy matrix. The increment in glass transition temperature with respect to filler loading is monotonian which provides an insight to enhanced cross linked networks. It reveals that both good dispersion of Ti₃SiC₂ and the consequent development of a great quantity of interfaces would ultimately hamper molecular movements.

3.3 Effect of Ti₃SiC₂ CDCN on the epoxy mechanical properties

Fig. 5a shows the effect of Ti₃SiC₂ CDCN on the flexural strength and modulus of epoxy composites at different reinforcement levels. The neat epoxy was found to have flexural strength and modulus values as 47 MPa and 3.4 GPa respectively. Upon reinforcement, these values were found to increase up to 69 MPa and 4.75 GPa for ET30 composites, which corresponds to 26% and 40% enhancement in flexural strength and modulus. The effect of Ti₃SiC₂ CDCN upon the compressive strength profile is depicted in Fig. 5b. Just like flexural property enhancement, the compressive strength was also steadily increased with increasing Ti₃SiC₂ CDCN. The neat epoxy has a compressive strength of 132 MPa. It was increased to 148, 182 and 225 MPa for the composites ET10, ET20 and ET30 respectively. These values are accounted for 12%, 38% and 70% improvement. In line with strength, the Ti₃SiC₂ CDCN was also found to increase the bulk hardness in the composites [Fig. 5c]. Being a glassy polymeric material, the neat epoxy has poor resistance to plastic deformation.

Fig. 5 Ti₃SiC₂ effect in epoxy composites on (a) flexural strength and modulus (b) compressive strength and % of lateral strain at constant stress of 80 MPa (c) Vicker's microhardness at 5 N load. Micro hardness indentation images of neat epoxy and epoxy with 30 wt% of Ti₃SiC₂ (d & e). SEM images of epoxy–Ti₃SiC₂ composites at (f) polished mode and (g) fracture mode.

When Ti₃SiC₂ CDCN introduced, a progressive gain in the bulk hardness was noticed. The neat epoxy possessed Vicker's micro hardness of 13 Hv at 5 N load which was increased to 19 Hv for composites containing 30 wt% Ti₃SiC₂. It has been calculated and found to be 44% more than the neat epoxy value. To understand the deformation pattern, the micro hardness indentation images obtained for neat epoxy and ET30 composites were given in Fig. 5d & e. The bulk Ti₃SiC₂ ceramic has hardness in the range 3–4 GPa. Since nanocrystalline Ti₃SiC₂ CDCN has a layered nature, the rise in hardness was the net effect of resistance against plastic deformation through pile-up mechanism due to preferential stacking of geometrically platy nanolayers.⁴² In fact, the increase in mechanical strength is usually correlated with the orientation, homogeneous distribution and interfacial bonding offered by the reinforcement phase. In functional fillers like CNT and graphene, the extent of their dimensionalities, directional alignment, and also the kind of agglomerate free dispersion were claimed as responsible factors.^15,17 The MAX phase Ti₃SiC₂ CDCN is unique in its mechanical ductility that results in kink-band formation.⁴³ In addition to good interfacial interaction offered by Ti₃SiC₂ nanolamella, these inherent qualities of Ti₃SiC₂ CDCN also contributes to absorb the excessively large stresses originating in the vicinity of the defects. The steady increase in strength and modulus with increasing Ti₃SiC₂ CDCN also revealed that it acts as an absolute damage tolerant second phase. The thermoplastic epoxy composites obtained with conventional fillers like micronic Al₂O₃ do not show any brittle to ductile transition.⁴⁴ interestingly, in presence of Ti₃SiC₂ CDCN, epoxy transforms to a semi-plastic and deforms permanently without any fracture. The percentage lateral strain upon compression at a load of 80 MPa was continuously observed for 20 s in order to understand the plasticity of Ti₃SiC₂ CDCN reinforced epoxy. The neat epoxy has maximum lateral to axial strain ratio which was decreased with increasing wt% of Ti₃SiC₂ CDCN [Fig. 5b]. It implies that the MAX phase Ti₃SiC₂ CDCN can accommodate higher magnitude of force that acts on it without much lateral strain. This trend leaves an idea that the chemically interactive Ti₃SiC₂ CDCN resulted in damage tolerance microstructures in epoxy composites. The fractured microstructures of epoxy–Ti₃SiC₂ nanocomposite were shown in Fig. 5f & g. The SEM images showed that the composite is free from processing defects like cavities, pits and air traps, supporting that CDCN is uniformly dispersed. Microstructures further showed that the composite undergoes ductile fractures. Presence of CDCN was seen on the vicinity of fractured surface. They also held a firm interaction with the matrix and fracture occurred along the interface indicating that the crack is deflected. A typical epoxy fracture in Al₂O₃ reinforced composite was also shown for better comparison (see Fig. S3 in ESI†).

3.4 Effect of MAX phase Ti₃SiC₂ CDCN on tribology of epoxy nanocomposites

Data pertaining to the specific wear rate and coefficient of friction as determined through dry sliding wear analysis were presented in Fig. 6 for various epoxy–Ti₃SiC₂ CDCN composites. The pin-on-disc wear test was conducted at a load range of 10 to 70 N, where other parameters like sliding distance, time and speed were kept constant. Fig. 6a shows the dependence of specific wear rate (SWR) and coefficient of friction (COF) with loads for neat epoxy. It shows that the wear rate of thermoplastic epoxy decreases with increasing load. In case of friction, the coefficient of friction increases from 0.7 to 0.83 up to a load 49 N and further increase in load decreases the coefficient of friction to 0.7 at 68.7 N. Softening of epoxy polymer during dynamic, mechanical abrasion conditions appears to be the reason for decreasing wear rate with respect to the increase in load.⁴⁵ At higher loads, heat generation between polymer work-piece and steel disc is exorbitantly high that causes softening of the epoxy polymer.

Fig. 6 (a) Dependence of specific wear rate and coefficient of friction of neat epoxy on the normal load applied (b) dependence of Ti₃SiC₂ loading in epoxy composites on (b) coefficient of friction and (c) specific wear rate at 2 different loads.

In composites, Fig. 6b shows the influence of Ti₃SiC₂ loading on the composites tribology at two different loads, 29.4 N and 68.6 N respectively. The friction coefficient continuously decreased with increasing percentage of Ti₃SiC₂ fillers. This trend is predominant at high loads. At a lower load of 29.4 N, neat epoxy had COF of 0.82 which reduced to a value of 0.48 for the composites containing 30 wt% of Ti₃SiC₂. The composite was again found to reduce the friction coefficient of the epoxy from 0.69 to 0.43 when the load was increased to 68.6 N. The reduction in COF appears to be an excellent proof for the formation of a third-body interface by Ti₃SiC₂ between the base plate and softened composite work piece. The presence of Ti₃SiC₂ interface was also verified by examining the wear rate at high and low loads with varying the amount of Ti₃SiC₂ fillers [Fig. 6c]. The wear rate was found to be increased with filler contents, as expected, due to an increase in the filler erosion. However, it was comparatively lesser at higher loads. In neat epoxy, primarily cause for material removal is adhesive wear.

However, in composite, in addition to the adhesive wear, the third-body abrasion also contributes.⁴⁶ The erosion of layered Ti₃SiC₂ ceramic reinforcement from the epoxy matrix occurs simultaneously during wear. Initially, the Ti₃SiC₂ debris freely moves and later it gets embedded on the softened polymer base. At low loads, freely moving Ti₃SiC₂ debris resulted in high wear rate. When the load was increased, the softened polymer along with the picked-up nanolayered Ti₃SiC₂ ceramic produced an interface that possibly reduces the rate of material removal resulting in a lower wear rate. The reduction in coefficient of friction and wear comparatively at higher loads indicated that the interface offers a sort-of lubricating effect,⁴⁷ due to the formation of a tribo-chemical film.

Microstructure analysis of epoxy and its composite surfaces after wear test was carried out under SEM and the representative SEM images were given in Fig. 7. It showed a clear difference on the erosion patterns between neat epoxy and its composites. Neat epoxy has a broad but shallow wear marks without much deep pull outs where as in composites, the erosion of Ti₃SiC₂ particles is apparent. At high loads and wt%, the erosion occurred more deeply. SEM image exhibits macro-pits on the surface, which clearly indicates a transition from two body abrasion to three body abrasion causing a change from adhesive wear to abrasive wear. Since Ti₃SiC₂ is characterized as a soft ceramic material, abrasion due to Ti₃SiC₂ interface is not strong, leading to less friction and wear rates even there are more pull outs at high loads.

Fig. 7 SEM images of the worn surfaces of neat epoxy and epoxy–Ti₃SiC₂ composites after tribological analysis at a load of 68.6 N for 15 minutes.

3.4.1 Wear mechanism. To confirm tribo-chemical reactions, the debris produced after the wear tests at given loads were also characterised using powder X-ray, TEM and EDX. The results are given in Fig. 8a–d.

Fig. 8 XRD analysis of debris collected during the wearing of ET30 composites (a) 29.4 N load and (b) 68.6 N load. (c) TEM image of the debris obtained at 68.6 N load (d) EDX analysis of debris obtained at 2 different loads.

The X-ray analysis confirmed the presence of intense Ti₃SiC₂ characteristic peaks in the debris formed at the load 29.4 N. A broad peak at the diffraction angle 2θ = 18.3° indicated the presence of amorphous epoxy. The debris collected at high load (68.6 N) also confirms the Ti₃SiC₂ phase but with less intensity indicating that the crystal plane will be preferably eroded faster. The debris was also viewed through TEM. Elemental analysis has also been carried out using EDX. Elemental analysis provides an insight for a possible tribo-chemical reaction. EDX data shows that the debris contains no oxygen and iron elements at low loads. However, intense peaks of oxygen and iron are seen in EDX profile at high load. It indicates the formation of oxides at the interface as a result of tribo-chemical reactions.⁴² The possible tribo-chemical reaction is represented as,

Ti₃SiC₂ + 6O₂ → 3TiO₂ + SiO₂ + 2CO₂

This tribo-chemical film can effectively reduce the wear rate, by creating a hindrance to 3-body abrasion mechanisms.⁴⁸

4. Dielectric permittivity

The MAX phase derived Ti₃SiC₂ nanostructure has another beneficial property compared to CNT and graphene functional fillers. When unmodified CNT and graphene nanostructures are incorporated with epoxy, they often produce electrically conducting epoxy, damaging the technically important insulating nature of the epoxy. The reinforcement of Ti₃SiC₂ CDCN marginally enhances the dielectric permittivity, unlike CNT and graphene. Fig. 9a shows the effect of Ti₃SiC₂ on the dielectric permittivity of epoxy composites. The epoxy has an average permittivity of 3.7 at 1 MHz. It is gradually increased with increasing Ti₃SiC₂ contents due to the increased inter-particle networking and better particle–matrix interactions. However, only a limited increase was seen with Ti₃SiC₂ even at high loadings. A dielectric permittivity of only 7.5 was seen in epoxy composites having 30% of Ti₃SiC₂. In fact, Ti₃SiC₂ is electrically conducting which undergoes inherent polarization in presence of electric field. These fillers cause a dipole-orientation and ionic polarization so as to increase the dielectric permittivity. However, a gradual decrease in dielectric permittivity was noticed [Fig. 9b] on increasing the ac frequency. It was due to the ceasing of orientation polarization, because the reversible switching of dipoles at higher frequencies becomes difficult. At higher frequencies, only electronic and ionic polarizations predominate, and as a result, the dielectric permittivity decreases.

Fig. 9 (a) Dielectric permittivity of epoxy composites at 1 MHz frequency. (b) Dielectric permittivity profile of composites at varying ac frequency.

In this work, a systematic validation has been carried out with Ti₃SiC₂ filler for the development of a new epoxy based structural composites. The property enhancement with MAX phase derived Ti₃SiC₂ nano fillers was compared with neat epoxy and are summarized in Table 1.

Table 1 Comparison of property enhancement in epoxy–Ti₃SiC₂ composites with respect to neat epoxy

Property	Neat epoxy (E)	Epoxy + 30% Ti3SiC2 (ET30)	% improvement
Thermal stability (°C)	335	366	9.25
Glass transition temperature (°C)	97	111	14.4
Flexural strength (Mpa)	47	69	25.5
Flexural modulus (Gpa)	2.3	4.75	106.5
Compressive strength	132	225	70
Vicker's microhardness (Hv)	13	19	46
Specific wear rate (10^−5 mm^3 N^−1 m^−1)	2.1	6.6	314
Coefficient of friction	0.7	0.4	−43
Dielectric permittivity@1 MHz	3.4	7.5	120

---

5. Conclusions

MAX phase derived Ti₃SiC₂ microstalks has been modified to 2 dimensional carbide derived ceramic nanostructure (CDCN) through shear-induced delamination method and subsequently incorporated in epoxy matrix. CDCN readily dispersed in epoxy resin via chemical interaction to form structurally homogeneous nanocomposites. As a result, epoxy nanocomposites acquire enormous quantity of interfaces which can hamper molecular movements of the polymeric segments during the course of temperature variations. These resulted in an increase of the thermal stability and the glass transition temperature of the nanocomposite compared to pure epoxy resin material. Moreover it provides a progressive increase in strength, hardness and relatively low lateral to axial strain under compression. 26% improvement in flexural strength and 70% in compressive strength was noticed. Compared to neat epoxy, 46% improvement was observed for micro hardness. The increased strength retention and hardness at higher loads confirms that CDCN fillers act as damage tolerant interfaces. The tribology studies revealed a low wear rate at higher loads and reduction of friction coefficient from 0.7 to 0.4. A tribo-chemical film formed by delaminated Ti₃SiC₂ interface in epoxy melt was found to be a reason for wear and friction control. The dielectric permittivity of the composites monotonically increased with enhanced filler loading due to the electrically conducting nature of Ti₃SiC₂.

Acknowledgements

Authors acknowledge the support received from Dr J. D. Sudha, Dr P. P. Rao, Mr A. Peer Mohammed, Ms. Lucy Paul and Mr Kiran Kumar in CSIR-NIIST for extending characterization facilities for rheology and microstructure analysis. This work is financially supported by Schneider Electric, France.

References

1. K. C. Yung, B. L. Zhu, T. M. Yue and C. S. Xie, Appl. Polym. Sci., 2009, 116, 518.
2. E. Amendola, A. M. Scamardella, C. Petrarca and D. Acierno, Appl. Polym. Sci., 2011, 122, 3686.
3. E. Bekyarova, E. T. Thostenson, A. Yu, H. Kim, J. Gao, J. Tang, H. T. Hahn, T. W. Chou, E. Itkis and R. C. Haddon, Langmuir, 2007, 23, 3970.
4. X. Huang, X. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 2012, 41, 666.
5. L. C. Tang, Y. J. Wan, D. Yan, Y. B. Pei, L. Zhao, Y. B. Li, L. B. Wu, J. X. Jiang and G. Q. Lai, Carbon, 2013, 60, 16.
6. Y. J. Wan, L. C. Tang, D. Yan, L. Zhao, Y. B. Li, L. B. Wu, J. X. Jiang and G. Q. Lai, Compos. Sci. Technol., 2013, 82, 60.
7. S. S. Vaisakh, M. Hassanzadeh, R. Metz, S. Ramakrishnan, D. Chappelle, J. D. Sudha and S. Ananthakumar, Polym. Adv. Technol., 2014, 25, 240.
8. M. K. Umboh, T. Adachi, K. Oishi, M. Higuchi and Z. Major, J. Mater. Sci., 2013, 48, 5148.
9. Y. Zheng, K. Chonung, G. Wang, P. Wei and P. Jiang, Appl. Polym. Sci., 2009, 111, 917.
10. A. Nassar and E. Nassar, Nanosci. Nanoeng., 2013, 1, 89.
11. B. L. Zhu, J. Wang, J. Ma, J. Wu, K. C. Yung and C. S. Xie, J. Appl. Polym. Sci., 2013, 127, 3456.
12. H. A. Ávila, L. A. Ramajo, M. S. Góes, M. M. Reboredo, M. S. Castro and R. Parra, ACS Appl. Mater. Interfaces, 2013, 5, 505.
13. Y. Song, Y. Shen, H. Liu, Y. Lin, M. Li and C. W. Nan, J. Mater. Chem., 2012, 22, 16491.
14. S. H. Song, K. H. Park, B. H. Kim, Y. W. Choi, G. H. Jun, D. J. Lee, B. Kong, K. W. Paik and S. Jeon, Adv. Mater., 2013, 25, 732.
15. X. Sun, H. Sun, H. Li and H. Peng, Adv. Mater., 2013, 25, 5153.
16. J. N. Coleman, U. Khan and Y. K. Gun'ko, Adv. Mater., 2006, 18, 689.
17. M. T. Byrne and Y. K. Gun'ko, Adv. Mater., 2010, 22, 1672.
18. X. Huang, C. Zhi, P. Jiang, D. Golberg, Y. Bando and T. Tanaka, Adv. Funct. Mater., 2013, 23, 1824.
19. Y. J. Wan, L. C. Tang, L. X. Gong, D. Yan, Y. B. Li, L. B. Wu, J. X. Jiang and G. Q. Lai, Carbon, 2014, 69, 467.
20. L. Z. Guan, Y. J. Wan, L. X. Gong, D. Yan, L. C. Tang, L. B. Wu, J. X. Jiang and G. Q. Lai, J. Mater. Chem. A, 2014, 2, 15058.
21. Y. J. Wan, L. X. Gong, L. C. Tang, L. B. Wu and J. X. Jiang, Composites, Part A, 2014, 64, 79.
22. J. Jia, X. Sun, X. Lin, X. Shen, Y. Mai and J. Kim, ACS Nano, 2014, 8, 5774.
23. M. Bozlar, D. He, J. Bai, Y. Chalopin, N. Mingo and S. Volz, Adv. Mater., 2010, 22, 1654.
24. W. Song, W. Wang, L. M. Veca, C. Y. Kong, M. S. Cao, P. Wang, M. J. Meziani, H. Qian, G. E. Lecroy, L. Cao and Y. P. Sun, J. Mater. Chem., 2012, 22, 17133.
25. X. Huang, C. Zhi and P. Jiang, J. Phys. Chem. C, 2012, 116, 23812.
26. M. R. Yadienka, A. Behnam, G. Jingwen, K. Christopher, J. Andrew, S. Benoit, M. Vahid, H. Pascal, D. Libo and J. Y. Robert, ACS Appl. Mater. Interfaces, 2011, 3, 2309.
27. X. Zhang, X. Fan, C. Yan, H. Li, Y. Zhu, X. Li and L. Yu, ACS Appl. Mater. Interfaces, 2012, 4, 1543.
28. J. M. Gonzalez-Domínguez, A. Anson-Casaos, A. M. Díez-Pascual, B. Ashrafi, M. Naffakh, D. Backman, H. Stadler, A. Johnston, M. Gomez and M. T. Martínez, ACS Appl. Mater. Interfaces, 2011, 3, 1441.
29. C. Li, S. Chai, K. Liu, N. Ning, J. Gao, Q. Liu, F. Chen and Q. Fu, ACS Appl. Mater. Interfaces, 2012, 4, 4398.
30. E. Bekyarova, E. T. Thostenson, A. Yu, M. E. Itkis, D. Fakhrutdinov, T. Chou and R. C. Haddon, J. Phys. Chem. C, 2007, 111, 17865.
31. J. M. Gonzalez-Domınguez, A. M. Dıez-Pascual, A. Anson-Casaos, M. A. Gomez-Fatoub and M. T. Martınez, J. Mater. Chem., 2011, 21, 14948.
32. M. Radovic, M. W. Barsoum, T. El-Raghy, J. Seidensticker and S. Wiederhorn, Acta Mater., 2000, 48, 453.
33. M. W. Barsoum, T. El-Raghy, C. J. Rawn, W. D. Porter, H. Wang, E. A. Payzant and C. R. Hubbard, J. Phys. Chem. Solids, 1999, 60, 429.
34. C. J. Gilbert, D. R. Bloyer, M. W. Barsoum, T. El-Raghy, A. P. Tomsia and R. O. Ritchie, Scr. Mater., 2000, 42, 761.
35. M. W. Barsoum and T. El-Raghy, J. Am. Ceram. Soc., 1996, 79, 1953.
36. Y. C. Zhou, Z. M. Sun, J. H. Sun, Y. Zhang and J. Zhou, Z. Metallkd., 2000, 91, 329.
37. S. Li, J. Xie, J. Zhao and L. Zhang, Mater. Lett., 2002, 57, 119.
38. N. F. Gao, Y. Miyamoto and D. Zhang, Mater. Lett., 2002, 55, 61.
39. K. V. Mahesh, S. Balanand, R. Raimond, A. Peer Mohamed and S. Ananthakumar, Mater. Des., 2014, 63, 360.
40. M. Magnuson, J. P. Palmquist, M. Mattesini, S. Li, R. Ahuja, O. Eriksson, J. Emmerlich, O. Wilhelmsson, P. Eklund, H. Högberg, L. Hultman and U. Jansson, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 245101.
41. M. Naguib, V. N. Mochalin, M. W. Barsoum and Y. Gogotsi, Adv. Mater., 2014, 26, 992.
42. T. El-Raghy, A. Zavaliangos, M. W. Barsoum and S. R. Kalidindi, J. Am. Ceram. Soc., 1997, 80, 513.
43. S. Li, J. Xie, J. Zhao and L. Zhang, Mater. Lett., 2002, 57, 119.
44. S. H. Lima, K. Y. Zenga and C. B. Heb, Mater. Sci. Eng., A, 2010, 527, 5670.
45. M. Naffakh, A. M. Dıez-Pascual, M. Remskar and C. Marco, J. Mater. Chem. C, 2012, 22, 17002.
46. M. Z. Rong, M. Q. Zhang, H. Liu, H. Zeng, B. Wetzel and K. Friedrich, Ind. Lubr. Tribol., 2001, 53, 72.
47. S. Myhra, J. W. B. Summers and E. H. Kisi, Mater. Lett., 1999, 39, 6.
48. D. Sarkar, B. Basu, S. J. Cho, M. C. Chu, S. S. Hwang and S. W. Park, J. Am. Ceram. Soc., 2005, 88, 3245.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra16518g

---