Effect of hybrid nano-fillers on the mechanical,thermal and viscoelastic properties of epoxy-based polymer composites for progressive applications

Materials

Properties and processing details of EP, hardener and fillers are listed in Supplemental Table 1. Hybrid fillers composed of silica, alumina and carbon black have been used to overcome these issues. The use of carbon black as a conductive filler may boost the electrical conductivity, mechanical strength and stiffness of the resin. Epoxy composites may significantly improve their longevity, stiffness and strength with the use of alumina particles sized between 2 and 10 nm.

Silica is a prevalent filler that enhances the thermal and mechanical reinforcement of the resin matrix. The epoxy system comprised Huntsman LY 556, a bisphenol-A-based diglycidyl ether (DGEBA) resin, and HY 951, an aliphatic amine curing agent supplied by Covai Seenu & Company, Coimbatore, Tamil Nadu. The resin exhibits a viscosity of 10,000–12,000 mPa·s and a density of 1.15–1.20 g/cm³, while the hardener has a lower viscosity (100–200 mPa·s) and density (0.95–0.99 g/cm³), ensuring good miscibility and homogeneous network formation at a 10:1 (resin: hardener) weight ratio.

The resin–hardener mixture was cured following a two-stage schedule of 80 °C for 2 h followed by 120 °C for 1 h, resulting in a cross-linked matrix with a Shore D hardness of approximately 80–85 and thermal conductivity around 0.22 W/m·K. The inorganic fillers were obtained from globally reputed suppliers to ensure uniformity in particle quality. Amorphous silica (AEROSIL 200, Evonik/Sigma-Aldrich) with particle sizes between 10 and 20 nm and a surface area of 300–400 m²/g was used to improve thermal stability and dimensional integrity. Gamma-alumina (Merck/Sigma-Aldrich, 20–200 nm), characterised by high hardness (Mohs 9), high modulus (350 GPa) and thermal conductivity (25–30 W/m·K), served as a mechanical reinforcement phase. Conductive carbon black (Ketjenblack^® EC-300J, Cabot/Sigma-Aldrich), having a density of 1.8 g/cm³ and particle sizes in the range 20–300 nm, was incorporated to enhance electrical conductivity and energy dissipation capacity.

Prior to mixing, all fillers were surface-functionalised to strengthen their interaction with the polymer matrix. Silica nanoparticles were silanised using APTES (2 wt%) at 60 °C for 2 h. Alumina nanoparticles were initially hydroxylated using dilute nitric acid (1 M) at 60 °C for 1 h, followed by washing, drying and subsequent functionalisation with γ-methacryloxypropyltrimethoxysilane (γ-MPS, 1–2 wt% in ethanol) at 70 °C for 3 h to improve interfacial compatibility and dispersion stability within the epoxy matrix. Carbon black underwent nitric acid oxidation followed by TEPA grafting to enhance dispersion and matrix adhesion. The hybrid filler-reinforced systems were processed under controlled vacuum and temperature conditions to minimise porosity and ensure reproducible mechanical and thermal behaviour. The composition details of neat epoxy, single-filler and hybrid-filler epoxy nanocomposites are summarised in Supplemental Table 2.

The fillers were included in the EP at a weight concentration of 1% to assess their collective impact on the composite's properties. We used both a 1:1 and a 4:1 system to evaluate the impact of varying ratios of carbon black to alumina, with a particle size of 20–200 nm, on the performance of the resin. To get further insight into the interaction of these fillers in enhancing the composite's overall performance, we also assessed the carbon black-silica system at two additional ratios: 1:1 and 1:4. The SiO₂ used in this study was procured from Sigma-Aldrich (Product No. 637238) with a specified particle size of 10–20 nm and 99.5% purity (trace metals basis). A list of the filler mixes and their corresponding particle sizes is listed in Supplemental Table 3.

Characterisation techniques

Flexural test analysis

The flexural strength, fatigue life and retained flexural strength of the developed epoxy nanocomposites are summarised in Supplemental Table 7 and illustrated in Supplemental Figure 2. PE exhibited an ultimate flexural stress of 43.9 ± 1.2 MPa, fatigue life of 12.5 × 10³ cycles and retained strength of 78.2%, reflecting its inherently brittle nature and limited capacity to resist crack propagation. Hybrid filler systems displayed markedly divergent behaviours depending on dispersion quality and synergistic interactions. PE + C achieved the highest performance, with ultimate stress of 67.45 ± 1.8 MPa, fatigue life of 27.6 × 10³ cycles and 92.3% retained strength, representing a 54% increase in strength and more than a twofold enhancement in fatigue resistance relative to neat epoxy.

The improvement is attributed to crack deflection induced by rigid alumina particles, which force propagating cracks along tortuous paths, and energy dissipation through carbon black pull-out at the matrix–filler interface, which delays catastrophic failure. PE + G exhibited a balance of moderate initial strength (53.2 ± 1.5 MPa) with durable fatigue resistance (24.3 × 10³ cycles, 90.4% retained strength), where silica nanoparticles contributed to crack pinning while carbon black pull-out absorbed significant fracture energy during cyclic loading.

The comparatively lower flexural performance observed in PE + A and PE + E is due to localised filler agglomeration and less effective stress transfer within the epoxy matrix. Such localised particle clustering can act as stress concentration regions that facilitate earlier crack initiation under bending loads, thereby reducing flexural durability. This interpretation is supported by the comparatively lower retained strength and fatigue resistance of these formulations relative to PE + C and PE + G. In contrast, PE + B, despite containing alumina-based reinforcement, exhibited improved mechanical response, suggesting that filler distribution and interfacial interactions were more effective at this composition ratio. These observations indicate that flexural performance is governed not only by filler type, but also by dispersion uniformity, hybrid filler ratio, and the efficiency of stress transfer across the filler–matrix interface.

Hybrid systems with more homogeneous dispersion and stronger interfacial interactions demonstrated improved resistance to crack propagation and cyclic loading. The flexural behaviour observed in the hybrid systems is strongly supported by the microstructural and interfacial evidence presented in sections ‘FTIR analysis’ and ‘SEM analysis’. FTIR analysis confirmed enhanced interfacial interactions through Si–O–Si, carbonyl and hydroxyl-related bonding features, while SEM observations revealed crack deflection, reduced void density and more tortuous fracture paths in the optimally dispersed hybrid systems. These combined structural and chemical effects contributed to improved stress transfer efficiency and delayed crack propagation under bending loads.

Indentation test analysis

The indentation behaviour of the hybrid epoxy nanocomposites reveals the influence of filler type, dispersion quality and matrix–filler interactions on load-bearing and energy absorption under localised stress, as shown in Supplemental Table 8. The cyclic indentation resistance of the hybrid epoxy nanocomposites is presented in Supplemental Figure 2(c). PE exhibited ultimate forces of 124, 208 and 315 N across three successive cycles, demonstrating a baseline resistance to localised deformation. Incorporation of hybrid fillers markedly enhanced indentation resistance, with PE + C achieving 207, 340 and 491 N, corresponding to increases of approximately 67%, 63% and 56% over PE, respectively. The improvement arises from rigid alumina particles acting as micro-scale barriers that hinder indenter penetration, while carbon black facilitates interfacial energy dissipation and accommodates stress through controlled micro-yielding.

Similarly, PE + F displayed the highest absolute forces (291, 420, 553), indicating superior hardness and energy absorption, attributable to optimised filler packing and synergistic interaction between multiple nanophases, which retard crack nucleation beneath the indenter. PE + B and PE + G also exhibited substantial gains, while PE + A showed moderate enhancement, and PE + E, with poor dispersion, offered minimal improvement over neat epoxy, highlighting the detrimental effect of agglomeration in local stress concentration zones. These results confirm that indentation resistance is governed not only by filler addition but by the combination of particle-induced obstruction, energy dissipation through pull-out or interfacial sliding and uniform filler distribution, which collectively enhance hardness, retard local cracking and improve the cyclic load-bearing capacity of epoxy nanocomposites. High indentation resistance in PE + C and PE + F systems makes them suitable for applications requiring surface durability and impact resistance, such as protective coatings, tooling, electronic casings and wear-resistant components. The enhanced indentation resistance of the hybrid systems is consistent with the DMA and SEM observations. Increased Tg values indicate restricted polymer chain mobility and improved network rigidity, while SEM micrographs revealed improved filler embedding and reduced microvoid formation in the optimised hybrid systems. These factors collectively contributed to improved localised load-bearing capacity and resistance to permanent deformation.

DMA test results

DMA analysis was performed on epoxy nanocomposites containing different hybrid nano-filler combinations and ratios. The results, plotted below, illustrate the material's behaviour under temperature sweep testing, where the complex modulus was measured at a constant frequency while varying the sample temperature. The temperature-dependent viscoelastic behaviour and glass transition characteristics of the epoxy nanocomposites are illustrated in Supplemental Figure 3.

This reflects the material's energy dissipation due to molecular rearrangements and internal friction. PE exhibits a Tg of 155 °C. Carbon black-modified composites (PE + A and PE + B) show Tg values of 158 °C and 160 °C, respectively, indicating partial restriction of polymer chain mobility due to filler–matrix interactions. Alumina- and silica-containing systems (PE + C, PE + D, PE + E) display Tg values of 162 °C, 165 °C and 168 °C, reflecting strong interfacial adhesion. Hybrid systems, including CB–Al₂O₃ (4:1) and CB–SiO₂ (1:1), exhibit the highest Tg values of 170 °C and 173 °C, demonstrating synergistic filler effects that further constrain polymer chain mobility and improve both thermal stability and mechanical performance.

The loss modulus, which relates to energy loss due to viscous flow, reveals that the glass transition temperature is lower for the pure epoxy and silica-modified epoxy systems. However, for the hybrid filler-modified resin systems, the glass transition range is slightly higher. The temperature sweep plots for each specimen can be found in the appendices. The increase in Tg observed in the hybrid composites directly correlates with the improved mechanical and thermal behaviour reported in sections ‘Flexural test analysis’ and ‘TGA analysis’. Stronger filler–matrix interactions and reduced polymer chain mobility contribute to enhanced stress transfer, improved dimensional stability and delayed thermal degradation. These observations are further supported by FTIR evidence of interfacial bonding and SEM evidence of homogeneous filler dispersion.

TGA analysis

The thermal degradation behaviour and derivative weight-loss profiles of the epoxy nanocomposites are presented in Supplemental Figure 4. A comparative evaluation of the thermal and mechanical performance of single- and hybrid-filler epoxy composites is presented in Supplemental Table 9. The PE exhibits a degradation onset (Tonset) at 300 °C and a peak decomposition temperature (Tmax) around 350 °C, with a residual mass of 8% at 800 °C, consistent with rapid polymer backbone scission and minimal energy dissipation. Upon addition of carbon black (PE + A/B), Tonset increases to 320–322 °C and Tmax to 370–372 °C, with residual mass rising to 12–13%.

This enhancement arises from interfacial interactions between oxidised CB surface groups and epoxy chains, facilitating partial stress transfer and slowing thermal decomposition, although minor particle agglomeration can create localised micro-stress regions. PE + C/D display further improvements, with Tonset of 335–340 °C, Tmax of 395–400 °C and residual masses of 16–22%. The observed synergistic effects in CB–Al₂O₃ hybrids are attributed to uniform dispersion, strong filler–matrix adhesion via hydroxyl–epoxy interactions and nanoparticle-mediated energy dissipation, which deflect heat flow and restrict polymer chain mobility. The 4:1 CB–Al₂O₃ system exhibits the highest thermal resistance, whereas excessive alumina in the 1:4 ratio leads to minor agglomeration, slightly reducing stability.

PE + E and PE + F/G exhibit Tonset of 338–342 °C, Tmax of 398–405 °C and residual mass of 18–25%, with broad, reduced DTG peaks (0.028–0.031 %/ °C), indicating gradual, distributed decomposition. The synergistic filler effect in these hybrids arises from CB-assisted silica dispersion, micro-crack pinning and the formation of tortuous thermal pathways, which collectively slow heat propagation. Surface-modified silica contributes to chemical grafting via silane groups, enhancing filler–matrix wetting and promoting network densification. As a result, the epoxy network exhibits reduced free epoxy groups, stronger interfacial adhesion and higher energy absorption, which collectively delay decomposition.

Hybrid fillers improve thermal stability through multiple synergistic processes: (i) physical barrier effects, where well-dispersed nanoparticles impede heat and mass transfer; (ii) chemical interactions, including hydrogen bonding, π–π stacking and covalent grafting, enhancing interfacial adhesion; (iii) microstructural reinforcement, with fillers pinning micro-cracks and creating tortuous pathways that inhibit thermal chain scission; and (iv) energy dissipation, where stress and heat are redistributed across the filler network.

Collectively, these mechanisms explain the enhanced Tonset, reduced DTG peak rates and higher residual char observed in hybrid composites, clearly correlating with improved mechanical resilience and demonstrating the critical importance of optimised filler selection, surface modification and hybrid ratios for advanced epoxy-based systems. The improved thermal stability of the hybrid composites is closely associated with enhanced interfacial bonding and optimised filler dispersion. FTIR results confirmed stronger chemical interactions between the functionalised fillers and epoxy matrix, while SEM analysis revealed reduced porosity and more homogeneous microstructures in the hybrid systems. These combined effects restricted polymer chain mobility, delayed thermal decomposition and improved residual char formation, particularly in the CB–Al₂O₃ and CB–SiO₂ hybrid formulations.

FTIR analysis

The FTIR spectra of neat and filler-modified epoxy composites (Supplemental Figure 5, a–f) provide compelling molecular-level evidence of interfacial phenomena governing the structure–property relationships in these systems. The neat epoxy (Supplemental Figure 5a) displays its characteristic functional group signatures: a broad O–H stretching band (3400 cm⁻¹), aromatic C–H stretching (2875 cm⁻¹), aromatic C = C stretching (1610 cm⁻¹), C–O–C asymmetric stretching (1255 cm⁻¹), and the epoxy ring vibration (915 cm⁻¹). The persistence of the latter confirms the presence of unreacted epoxide moieties, which serve as potential reactive sites for filler–matrix interactions. Upon incorporation of CB (Supplemental Figure 5b), notable spectral modifications emerge.

The decrease in C–H stretching intensity (2975/2850 cm⁻¹) indicates partial masking of the epoxy backbone by dispersed CB nanoparticles. More importantly, the emergence of a distinct carbonyl peak at 1747cm⁻¹ suggests surface oxidation of CB (introduced during nitric acid treatment), which introduces oxygenated functionalities capable of forming hydrogen bonds and polar interactions with the epoxy matrix. Concurrent minor shifts in the aromatic region (1485 cm⁻¹) further point to π–π stacking interactions between the conjugated CB structure and the aromatic epoxy backbone, thus providing dual physical–chemical anchoring mechanisms.

The Al₂O₃-filled composite (Supplemental Figure 5c) confirms nanoparticle incorporation through a sharp Al–O band (750 cm⁻¹), coupled with attenuation of the epoxy ring band (2015–915 cm⁻¹ region). This indicates strong filler–matrix adhesion mediated by surface hydroxyl–epoxy interactions and partial consumption of free epoxide groups. Such interfacial hydrogen bonding effectively constrains crack initiation sites and enhances stress transfer. For silica-filled epoxy (surface-modified) (Supplemental Figure 5d), the strong Si–O–Si asymmetric (1080 cm⁻¹) and symmetric (790 cm⁻¹) bands are observed, alongside broadening of the O–H band (3400 cm⁻¹). These features confirm successful silane modification of silica, which improves wettability and compatibility with the polymer matrix.

The enhanced spectral intensity of the Si–O–Si network indicates the formation of a robust inorganic substructure within the organic matrix, supporting effective crack deflection and reduced porosity. In CB–Al₂O₃ hybrid systems, the 4:1 ratio (Supplemental Figure 5e) shows simultaneous retention of Al–O (670–750 cm⁻¹) and C–O–C (1255 cm⁻¹) peaks, along with suppression of the epoxy ring (915 cm⁻¹). This provides direct evidence of cooperative filler–epoxy bonding, which rationalises the superior mechanical performance of this hybrid formulation. The 1:4 CB–Al₂O₃ composite exhibits weaker epoxy suppression, suggesting excessive alumina loading leads to particle agglomeration, reduced dispersion and less efficient chemical synergy.

The CB–SiO₂ hybrids (Supplemental Figure 5f) reveal intense Si–O–Si bands and a broadened O–H stretching profile (3400 cm⁻¹), signifying a silica-rich interconnected network assisted by CB dispersion. Suppression of the epoxy ring (915 cm⁻¹) and ether linkage (1255 cm⁻¹) signals highlights chemical grafting and reduced unreacted epoxy groups, confirming the creation of a more chemically integrated and defect-resistant matrix. The FTIR results establish that each filler contributes a distinct chemical pathway for interfacial reinforcement: carbonyl groups from CB enable polarity-driven interactions, hydroxylated alumina promotes hydrogen bonding and silane-functionalised silica ensures covalent-like network integration.

In the hybrids, these mechanisms act cooperatively, reducing free epoxy groups and promoting uniform filler dispersion. Such molecular-level chemical anchoring translates directly into the observed improvements in fracture toughness, flexural strength, hardness and thermal stability, underscoring the critical role of functional group chemistry in dictating macro-scale composite performance. The FTIR observations provide molecular-level evidence supporting the mechanical and thermal performance trends discussed in earlier sections. Improved interfacial interactions between the functionalised fillers and epoxy matrix contributed to enhanced stress transfer, restricted segmental mobility and improved crack resistance, which collectively explain the higher flexural strength, Tg, indentation resistance and thermal stability observed in the optimised hybrid systems.

SEM analysis

SEM micrographs of the fabricated epoxy composites (Supplemental Figure 6, a–h) reveal a clear evolution of microstructural features from neat epoxy to hybrid nano-filler systems, providing insight into the mechanisms governing mechanical performance. The pure epoxy (Supplemental Figure 6a) exhibits a smooth, featureless fracture surface, indicative of brittle failure, minimal energy dissipation and weak crack resistance, with cracks propagating linearly through the polymer matrix due to the absence of reinforcing fillers.

The lack of surface roughness and stress-arresting sites highlights the inherent susceptibility of neat epoxy to catastrophic fracture under load. Incorporation of carbon black (Supplemental Figure 6b) leads to partial nanoparticle dispersion, resulting in increased surface roughness and localised stress transfer sites. However, regions of agglomeration act as micro-stress concentrators, potentially initiating micro-crack formation under mechanical load. These observations explain the moderate improvement in toughness, as carbon black partially arrests crack propagation but does not fully inhibit brittle failure.

Epoxy composites reinforced with 20 nm alumina (Supplemental Figure 6c) show uniform particle embedding and strong filler–matrix adhesion, with cracks exhibiting deflection at particle sites. The alumina particles effectively occupy void spaces, reduce porosity and promote mechanical interlocking, which collectively enhance energy absorption and delay catastrophic crack propagation. The microstructural evidence supports the observed improvements in flexural strength and hardness, demonstrating that nano-sized ceramic fillers act as efficient crack arresters. Surface-modified silica composites (Supplemental Figure 6d) display well-integrated filler networks and synergistic dispersion, with clear micro-crack pinning and tortuous crack paths. This behaviour reflects enhanced chemical bonding via silane functional groups, improved filler–matrix wetting and efficient stress transfer across the interface. The resulting morphology underlines the role of surface modification in reinforcing fracture resistance by promoting crack deflection and bridging mechanisms.

In hybrid filler systems, SEM observations directly correlate with the mechanical behaviour discussed in section ‘Flexural test analysis’. The CB–alumina (4:1) composite (Supplemental Figure 6e), which exhibited the highest flexural strength and fatigue resistance, showed relatively homogeneous filler distribution, strong filler–matrix adhesion and rough fracture morphology characterised by crack deflection and localised crack-bridging features. These microstructural characteristics increase the crack propagation path and promote energy dissipation during bending, thereby delaying catastrophic failure under cyclic loading conditions. In contrast, the 1:4 CB–alumina hybrid (Supplemental Figure 6f) exhibited localised particle agglomeration and interfacial debonding regions that can act as stress concentration sites, facilitating earlier crack initiation and reduced flexural durability. This observation is consistent with the comparatively lower retained strength observed for this formulation in flexural testing.

Similarly, the CB–silica hybrid systems demonstrated that improved filler dispersion and interfacial compatibility contribute significantly to enhanced mechanical response. The 1:4 CB–silica composite (Supplemental Figure 6g) displayed reduced micro-void density, effective filler embedding and localised crack-path deviation, while the 1:1 CB–silica hybrid (Supplemental Figure 6h) exhibited balanced dual-filler distribution with rougher fracture surfaces and more tortuous crack propagation paths. These features indicate improved stress redistribution and greater resistance to crack growth, which correlate with the enhanced indentation resistance and improved fatigue behaviour measured experimentally.

The SEM analysis suggests that the improved mechanical performance of the hybrid epoxy nanocomposites is governed by the combined influence of dispersion uniformity, interfacial adhesion, crack deflection and localised energy-dissipation mechanisms. Conversely, localised agglomeration and weak interfacial regions promote stress concentration and accelerate crack propagation, thereby reducing mechanical durability. Overall, the SEM observations establish direct structure–property relationships between filler dispersion, interfacial adhesion, crack propagation behaviour and the experimentally observed mechanical and thermal performance of the hybrid epoxy nanocomposites.

Implications for long-term durability

Although long-term environmental aging tests were not conducted in the present study, the observed improvements in thermal stability, glass transition temperature, filler–matrix interfacial integrity and residual char formation provide useful insight into the potential durability behaviour of the developed hybrid epoxy nanocomposites under thermally demanding conditions. Surface-functionalised silica and alumina nanoparticles improve interfacial adhesion and may restrict localised polymer chain mobility, thereby reducing the likelihood of premature microcrack formation during mechanical or thermal loading. In addition, the enhanced thermal resistance observed from TGA analysis and the increased Tg values measured through DMA suggest improved structural stability of the epoxy network at elevated temperatures.

The hybrid filler systems also exhibited comparatively rougher fracture morphologies and improved stress-transfer characteristics in SEM analysis, indicating greater resistance to crack propagation relative to neat epoxy. Carbon black may additionally contribute to improved stress redistribution within the matrix because of its fine particulate morphology and high surface area. Collectively, these observations suggest that appropriately dispersed hybrid fillers could contribute to improved resistance against thermally induced damage and mechanical degradation. However, dedicated long-term aging studies involving environmental exposure, moisture absorption, UV stability and hygrothermal cycling are required to conclusively establish the durability performance of these epoxy nanocomposites.