Effect of cenosphere size and loading on silicone rubber composites

Abstract

This study investigated the influence of fly ash-derived aluminosilicate cenospheres on the structure and performance of silicone rubber composites. The silicone matrices were filled with cenospheres of two representative particle sizes (average diameters of 46 and 106 µm) at concentrations of 10, 25, and 50 phr. A comprehensive characterization was conducted, including SEM analysis, mechanical and thermal testing, ESR spectroscopy, FTIR analysis, TGA/DTG, specific heat capacity evaluation, and swelling resistance in various media. The results showed that the incorporation of cenospheres at low to moderate loadings (10–25 phr) led to improved mechanical retention and thermal stability. In particular, composites filled with smaller cenospheres (46 µm) at 10 phr retained high elongation at break (up to 650%) while exhibiting enhanced tear resistance (up to 32.8 kN·m^-1) compared to the neat silicone rubber. Thermogravimetric analysis revealed delayed thermal degradation and a significant increase in inorganic residue, exceeding 65% at 800°C for filled composites. Accelerated thermal aging at 250°C for 72 h demonstrated that cenosphere-filled systems preserved their elastomeric character, whereas the unfilled rubber became brittle. Overall, the quantitative results confirm that aluminosilicate cenospheres act as effective functional fillers, enhancing the mechanical durability and thermal-protection performance of silicone rubber composites.

Keywords

silicone rubber composites cenospheres particle size filler concentration thermal stability mechanical properties swelling resistance ESR spectroscopy FTIR analysis thermal aging specific heat capacity

Introduction

Silicone materials have been commercially produced since the early 1940s. Over the past decades, they have grown into a billion-dollar industry and are now widely used in the civil engineering, construction, electronics, transportation, aerospace, defense, textiles, and cosmetics industries.¹ Owing to their unique chemical structure, polysiloxanes exhibit several exceptional properties such as high elasticity, electrical insulation, biocompatibility, and stable physical and mechanical characteristics over a wide temperature range. They also possess high resistance to heat and environmental factors.¹ Contemporary studies on aging evaluation of silicone rubber composites discuss multiple thermal aging mechanisms and characterization methods that provide insights into interface evolution under prolonged thermal exposure.^2–5 However, the pristine silicone rubber displays relatively poor tracking and erosion resistance.

To enhance its performance, extend its service life, and reduce production costs, fillers are commonly incorporated into the silicone matrix.⁶ Fillers are generally classified into two categories: reinforcing and extending. Reinforcing fillers enhance tensile strength, modulus, tear resistance, and abrasion resistance, and can also confer functional properties, such as electrical conductivity, as observed with carbon black. Common reinforcing fillers include fumed silica, aerogel silica,^7,8 and carbon black.^9–11 Extending fillers, which are either semi- or non-reinforcing, are incorporated to reduce costs and sometimes to provide specific properties, such as improved thermal or electrical conductivity. Typical extending fillers include ground quartz, titanium dioxide,¹² clay, whiting, alumina trihydrate (ATH), and zinc oxide. Recently, fly ash-derived cenospheres (CS) have emerged as promising fillers for rubber and plastic composites due to their low density, spherical shape, inertness, smooth surface, low thermal conductivity, and nanoscale size.¹³ In silicone rubber composites, fly ash-derived aluminosilicate cenospheres are best described as functional fillers, as their primary contribution arises from structural and thermal effects rather than classical mechanical reinforcement.¹⁴ Being a by-product of coal-fired power plants, FA is a cost-effective, readily available, and environmentally friendly alternative.

Cenospheres have been successfully incorporated into various polymer matrices, including polystyrene,¹⁵ polyester,¹⁶ epoxy,^17–20 polyethylene,²¹ polypropylene,^22,23 nylon,²⁴ and acrylonitrile butadiene styrene (ABS).²⁵ In elastomeric systems, they have been explored in natural rubber,^26,27 styrene-butadiene rubber (SBR),²⁸ chlorinated SBR,²⁹ natural rubber (NR),³⁰ polyester thermoplastic elastomer,³¹ acrylonitrile butadiene styrene rubber,²⁵ polyurea elastomer,³² and carboxylated nitrile butadiene rubber. These studies generally report that increasing the filler content tends to reduce the tensile strength and elongation at break while enhancing the thermal stability, shear viscosity, and dielectric strength. A filler loading of up to 30 phr is often cited as the optimal concentration.^29,32 Few studies have addressed the use of cenospheres, specifically in silicone matrices. Rybiński et al. reported a significant enhancement in the thermal stability of silicone rubber composites filled with cenospheres, compared to those containing conventional mineral fillers.¹ However, the simultaneous influence of cenosphere particle size and concentration on the overall performance of silicone composites has not yet been sufficiently explored.

Despite the growing interest in fly ash-derived cenospheres as functional fillers for polymeric systems, the existing literature remains fragmented in terms of structure–property correlations. Most reported studies focus either on the effect of filler loading at a fixed particle size or on particle size variations at a single concentration, which limits the possibility of a quantitative assessment of size–loading synergy. As a result, the combined influence of particle size and concentration on key performance parameters such as rheology, mechanical integrity, thermal stability, and aging resistance remains insufficiently understood.

Furthermore, many studies primarily address macroscopic properties, while providing limited insight into interfacial mechanisms, including filler-induced restrictions of molecular mobility, changes in curing behavior, and the role of filler surface area in governing matrix–filler interactions. In silicone-based systems, where thermal stability and long-term performance are critical, such interfacial effects are particularly important but are rarely analyzed using complementary spectroscopic or molecular-dynamics-sensitive techniques.

Despite the growing interest in cenosphere-filled elastomer systems, most available studies focus on either filler content or particle size as isolated parameters and primarily evaluate short-term mechanical performance. The combined influence of cenosphere size and loading on the structure–property–performance relationships of silicone rubber composites, particularly under thermal aging conditions, remains insufficiently addressed. Moreover, the reduction in tensile strength and elongation at break commonly reported for filled rubbers highlights the need to identify optimal trade-offs rather than pursuing maximum reinforcement.

In this context, the present study aims to systematically investigate the effects of representative cenosphere size fractions and filler loadings on the morphological, rheological, mechanical, and thermal behavior of silicone rubber composites. The work focuses on balancing mechanical retention with enhanced thermal stability, aging resistance, and dimensional stability, which are critical for thermally demanding elastomer applications. By employing a comprehensive multi-technique approach, this study seeks to establish clear structure–property–performance relationships and to define a practical performance window for cenosphere-filled silicone rubber systems.

Materials and methods

Materials

The silicone rubber used in this study was polydimethylsiloxane diol (Elastosil R 401/60 S) supplied by Wacker Chemie AG (Munich, Bavaria, Germany). This compound contains approximately 60 wt % of highly active fumed silica. The key physical properties of the samples are listed in Table 1.

Table 1.

Physical properties of silicone rubber.

Property	Value
Density at 20°C	1.15 g/cm³
Shore A hardness	56
Mooney viscosity	42 MU
Tensile strength	11 N/mm²
Tear strength	24 N/mm

Fly ash cenospheres (CS) were obtained from the Ekibastuz GRES-2 power station (Pavlodar Region, Kazakhstan) and supplied by INOTECK (Moscow Region, Russia). Their true densities were measured using a helium pycnometer (AccuPyc 1340, Micromeritics, Norcross, GA, USA). The specific surface area (SSA) was determined using the Brunauer–Emmett–Teller (BET) method on an ASAP 2020 MP analyzer (Micromeritics, Norcross, GA, USA) following standard degassing at 300°C under a 0.5 mmHg vacuum for 12 h. BET analysis was performed at 77 K with nitrogen as the adsorbate.

The particle size distribution was measured by laser diffraction (Bettersize analyzer) in ethanol using ultrasonic dispersion (intensity 5) and mechanical stirring (1200 rpm). The Mie scattering model was applied with a refractive index of 1.52—0.1i (particles) and 1.361 (medium). Particle size data (D10, D50, D90 and D[4,3]) and SSA (m²/kg) were recorded. Figure 1 illustrates the size distributions of the CS50 and CS100 cenospheres. Hereafter, the cenospheres are denoted as CS50 and CS100, corresponding to average particle diameters of approximately 46 µm and 106 µm, respectively.

Figure 1.

Particle size distributions of aluminosilicate cenospheres: (a) CS50, (b) CS100.

Information on the chemical composition and physical properties was provided by the supplier INOTECK (Moscow Region, Russia) and is presented in Tables 2 and 3.

Table 2.

Physical properties of cenospheres.

Property	CS50	СS100
True density (g/cm³)	0.9281	0.8178
BET surface area (m²/g)	1.102	0.507
SSA from granulometry (m²/kg)	75.17	25.6
Average particle size (µm)	46	105.9
D10 (µm)	16.76	67.69
D50 (µm)	45.45	103.6
D90 (µm)	74.9	151.6

Table 3.

Chemical composition of cenospheres.

Silica (SiO₂)	50-68%
Alumina (Al₂O₃)	18-38%
Iron (Fe₃O₂)	1.4 - 6%
CaO	1.8%
K₂O + Na₂O	1.5%

Preparation of rubber compounds

The rubber formulations are listed in Table 4. Mixing was carried out on a laboratory two-roll mill (MonTech, Germany) under controlled conditions. The roll temperature was maintained at 25 ± 2°C to prevent premature vulcanization. The friction ratio between the front and rear rolls was set to 1:1.25, with roll speeds of 20 and 25 rpm, respectively. The initial nip gap was adjusted to 1.5 mm and gradually reduced to 0.8 mm during mixing to ensure effective filler incorporation and dispersion. The total mixing time did not exceed 10 min. Cenospheres were added gradually to the silicone rubber matrix to avoid particle agglomeration, followed by incorporation of curing additives at the final stage of mixing.

Table 4.

Formulations of rubber blends (phr).

Ingredients		Formulations of rubber blends (phr^a)
Ingredients		0	50–10	50–25	50–50	100–10	100–25	100–50
1.	Silicone rubber	100.0	100.0	100.0	100.0	100.0	100.0	100.0
2.	Dicumylperoxide	1.3	1.3	1.3	1.3	1.3	1.3	1.3
3.	Vulcanization inhibitor (ethynyl cyclohexanol)	0.83	0.83	0.83	0.83	0.83	0.83	0.83
4.	Platinum catalyst	1.56	1.56	1.56	1.56	1.56	1.56	1.56
5.	Cenospheres CS50	–	10	25	50	–	–	–
6.	Cenospheres CS100	–	–	–	–	10	25	50
	Total	103.69	113.69	128.69	153.69	113.69	128.69	153.69

^aparts per hundred parts of rubber.

Vulcanization was performed in a hydraulic press (LP 3000, MonTech) at 150°C for 15 min, followed by post-curing at 200°C for 4 h in a convection oven. The sample coding includes the cenosphere fraction designation (CS50, CS100) and its content in phr (e.g., CS50–10 denotes a composite filled with the CS50 fraction at a loading of 10 phr). The numerical labels CS50 and CS100 correspond to cenosphere fractions with average particle diameters of 46 and 106 µm, respectively. The unfilled silicone rubber is designated as “0.”

Filler/matrix morphology

The morphologies of neat cenospheres and silicone rubber composites were studied using a scanning electron microscope (Tescan Vega 3, Czech Republic) equipped with a tungsten thermionic cathode. SEM observations were performed under high-vacuum conditions at an accelerating voltage of 10–15 kV. Image acquisition and processing were carried out using TESCAN Essence™ software. Cryogenic fracture surfaces of the composites were examined to preserve the native structure of the filler/matrix interface. Prior to imaging, the fractured surfaces were sputter-coated with a thin platinum layer (∼20 nm) using a JEOL JFC-1600 sputtering system to ensure sufficient surface conductivity.

Hardness testing

Hardness measurements were performed using an HT 3000 Shore A hardness tester (MonTech, USA). Five measurements were taken at different positions for each sample and the average value was reported.

Rheological testing

Mooney viscosity (ML 1 + 4 @ 100°C) was determined using an MDR V-MV 3000 Mooney viscometer (MonTech, Germany). The cure characteristics were analyzed using a Rubber Process Analyzer RPA D-MDR 3000 (MonTech, Germany) at 1.67 a frequency of 0.5° strain.

Abrasion resistance

The abrasion resistance was assessed using an ABR 3000 abrasion resistance tester (MonTech, Germany). Cylindrical vulcanized samples (10 ± 0.5 mm in diameter and height) were tested. The drum rotated at (40 ± 1) rpm with the sample undergoing linear displacement of (420.0 ± 4.0) mm per 100 drum revolutions. The applied normal force was 5.0 and 10.0 N with an accuracy of ±0.2 N.

Thermogravimetric analysis (TGA)

The thermal stability was investigated using a TGA/DSC3 + analyzer (Mettler Toledo, Switzerland). Samples (4–5 mg) were heated from 25°C to 800°C at a constant rate of 10°C/min in air.

FT-IR spectroscopy

FT-IR spectra of the composites were recorded using a Lumos BRUKER spectrometer (Berlin, Germany) equipped with an ATR diamond crystal. The spectra were collected at room temperature (22 ± 2°C) over a range of 4000-600 cm⁻¹.

Mechanical properties

Tensile properties were measured in accordance with ASTM D412 using a GOTECH universal testing machine (Turkey) equipped with a 5 kN load cell, at a crosshead speed of 500 mm/min. Dumbbell-shaped specimens were used in this study. Each reported value represents the mean of five replicates.

Specific heat capacity

The specific heat capacity was determined using a DSC 214 Polyma (NETZSCH, Germany). The samples (10 ± 1 mg) were sealed in aluminum pans. The heating program consisted of an isothermal segment, followed by a linear ramp at 10°C/min. The calculations were based on the following equation

c_{p} = \frac{H F * Δ t}{m * Δ T},

Where HF is the heat flow, Δt is the time interval, ΔT is the temperature change, and m is the mass of the sample. Cp was calculated in the range of 30–150°C.

Accelerated aging

Accelerated aging tests were conducted in an AO 3000 Rubber Aging Oven (MonTech, Germany) for 72 h at 250°C.

Chemical resistance

The chemical resistance was assessed based on the degree of swelling according to ASTM D471. The samples were immersed in various liquids, including quaternary ammonium disinfectant (QAS), food-grade Vaseline oil, acetic acid (CH₃COOH), petrochemical solvents (Nefras), and industrial oil (I-40A). The samples were weighed every 24 h. The swelling degree was calculated as follows

Swelling degree (%) = (W 2 - W 1) / W 1 \times 100,

Where W1 is the initial weight and W2 is the weight after immersion.

ESR-spectroscopy

Electron spin resonance (ESR) measurements were performed using a Bruker EMX spectrometer (Germany) at 22°C. Modulation amplitude was <0.5 G; microwave power was 2.1 mW; sweep width was 100 G (1024 points recorded). The spin probe used was 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO, <1 × 10⁻³ mol·dm⁻³), which was introduced into the vulcanized samples in the vapor phase at 60°C. The probe rotation correlation time was calculated as follows

τ c = Δ H^{+ 1} \times [(I^{+ 1} / I^{- 1}) 0.5 - 1] \times 6.65 \times 10^{- 10} [s]

The equation includes the width (from peak to peak) of the low-field spectral component (ΔH+1) and intensities of high/low-field components following (I ± 1). The experimental uncertainty was within ±7%. WINEPR™ and SIMFONIA™ (Bruker) software packages were used for data analysis.

Results and discussion

Morphology of filler dispersion

The dispersion of aluminosilicate cenospheres in the silicone matrix was examined using SEM micrographs of cryo-fractured surfaces. Figure 2 shows the morphology of the pristine cenospheres at different magnifications. Figure 3 presents the morphologies of the composites with varying filler loadings (10, 25, and 50 wt%) for two distinct cenosphere fractions: CS50 (average particle size 46 µm) and CS100 (average size 105.9 µm).

Figure 2.

SEM micrographs of cenospheres (CS50) with different magnifications (а) - at 2000× magnification, (b) - at 200× magnification.

Figure 3.

SEM micrographs of cryogenically fractured silicone/cenospheres composites: (а) 50–10; (с) 50–25; (e) 50–50; (g) 100–10; (i) 100–25; (k) 100–50 at 500× magnification; (b) 50–10; (d) 50–25; (f) 50–50; (h) 100–10; (j) 100–25; (l) 100–50 at 2000× magnification.

At a low filler content (10 wt%), both CS50- and CS100-filled composites exhibited relatively uniform dispersion and good adhesion to the silicone matrix. In the CS50 sample, the smaller cenospheres with higher specific surface area (BET = 1.102 m²/g) were well embedded in the polymer, and most of the hollow particles retained their structural integrity. Similarly, CS100 cenospheres at 10 wt% largely preserved their spherical shape, with only limited signs of particle damage or matrix detachment. With increasing filler content, notable changes in the microstructure were observed for both systems. In the CS50 composite containing 25 wt% cenospheres, the number of hollow particles increased substantially, while dispersion remained relatively homogeneous. However, more frequent interparticle contacts became evident, indicating the onset of particle–particle interactions. At 50 wt%, pronounced agglomeration and extensive fracturing of cenospheres were observed. The compact packing of smaller particles and their high specific surface area (SSA = 75.17 m²/kg) likely promoted increased particle–particle interactions, leading to insufficient matrix wetting and the formation of mechanical interlocking defects. For the CS100 series, good dispersion was maintained at 10 wt%, whereas at 25 wt% partial matrix detachment around individual cenospheres and occasional particle crushing were observed. At the highest filler loading (50 wt%), the fracture surfaces were dominated by broken hollow spheres, thin-walled fragments, and large interfacial voids. Owing to their lower BET surface area (0.507 m²/g) and larger average particle size (105.9 µm), CS100 cenospheres appear more susceptible to mechanical damage under processing stresses and exhibit weaker interfacial bonding with the silicone matrix.

The observed differences in interfacial behavior between CS50- and CS100-filled composites can be attributed to a combination of geometric, mechanical, and interfacial factors. Larger cenospheres possess lower surface curvature and a reduced effective contact area per unit volume, which limits physical adhesion and mechanical interlocking with the polymer matrix. In addition, variations in shell thickness of larger cenospheres increase their susceptibility to fracture under shear and compressive stresses during two-roll milling. The collapse or partial breakage of hollow particles generates local voids and stress concentrators, further weakening interfacial integrity. In contrast, smaller cenospheres with higher specific surface area provide a greater number of contact points with polymer chains, promote more homogeneous stress transfer, and are more effectively embedded within the matrix, resulting in improved interfacial bonding and structural stability at low filler loadings.

Overall, the morphological evolution demonstrates that both filler particle size and loading play a critical role in dispersion quality and filler–matrix interactions. Smaller cenospheres (CS50) provide better dispersion at low filler contents but are more prone to agglomeration at high loadings, whereas larger cenospheres (CS100) are more susceptible to shell fracture and matrix debonding as the filler concentration increases. The assessment of filler dispersion was based on qualitative analysis of SEM micrographs of representative fracture surfaces. The comparison between different composites was performed by visual inspection, focusing on characteristic morphological features such as particle integrity, matrix–filler adhesion, particle clustering, and overall homogeneity of particle distribution. While SEM analysis does not provide quantitative dispersion metrics, the observed morphological trends were consistent across multiple micrographs for each composition and correlated well with the rheological, mechanical, and ESR results discussed in subsequent sections.

It should be emphasized that the cenospheres used in this work were not surface-functionalized. While unmodified aluminosilicate cenospheres contribute to density reduction and thermal stabilization, their limited chemical affinity toward the silicone matrix may result in interfacial debonding at high filler loadings, as evidenced by local matrix detachment and void formation observed for the 50 phr composites, particularly in the CS100-filled system. Surface modification using silane coupling agents represents a well-established strategy to enhance filler–matrix interactions, improve dispersion, and reduce particle aggregation. Although such modifications were not explored in the present study, they constitute a promising direction for future work.³³

Rheological and curing properties

One of the key technological parameters for rubber compounds is their processability, which is commonly evaluated using the Mooney viscosity. The rheological behavior of silicone rubber composites filled with aluminosilicate cenospheres was studied using an MDR 3000 rheometer. The results are summarized in Table 5. Mooney viscosity values are reported as representative measurements obtained according to ASTM D1646. The repeatability of the method was within ±1 MU.

Table 5.

Mooney viscosity values of silicone rubber and silicone/cenosphere composites.

Rubber blend	Mooney initial (MI)Δ ± 1 MU	Mooney viscosity (ML (1 + 4) 100°C)Δ ± 1 MU
0	53	37
50–10	55	40
50–25	65	50
50–50	77	61
100–10	56	41
100–25	69	50
100–50	86	69

At a low filler loading of 10 phr, the addition of cenospheres resulted in only a slight increase in Mooney viscosity, by approximately 3–4 units compared to the unfilled silicone rubber, irrespective of the cenosphere size fraction. This behavior indicates that at low concentrations the cenospheres are well accommodated within the silicone matrix and do not significantly hinder polymer chain mobility. As the filler content increased to 25 phr, a pronounced rise in viscosity was observed for both cenosphere fractions. In this case, the Mooney viscosity increased by 13 units for both the CS50- and CS100-filled composites relative to the neat rubber. This result highlights the dominant role of filler loading at intermediate concentrations, while the influence of particle size remains secondary. At the highest filler concentration of 50 phr, the effect of particle size became significantly more pronounced. The Mooney viscosity of the composite filled with CS50 reached 61 units (+24 units compared to the neat rubber), whereas the CS100-based composite exhibited the highest resistance to flow, with a viscosity of 69 units (+32 units). These values clearly demonstrate that increasing both filler concentration and particle size cumulatively contributes to the stiffening of the rubber compound.

Overall, the rheological behavior of silicone/cenosphere composites is highly sensitive to filler loading, while particle size becomes a critical factor affecting processability at concentrations above 25 phr. At low filler contents, the cenospheres are well dispersed and readily accommodated by the matrix. In contrast, at higher loadings, larger cenospheres (such as CS100) impose greater mechanical constraints, restrict polymer chain mobility, and increase internal friction, resulting in elevated viscosity.

The curing behavior was analyzed by measuring torque development during vulcanization. The results are shown in Figure 4 and Table 6.

Figure 4.

Cure curves of the as-received silicone rubber material (sample 0) and silicone/cenospheres composites (50–10, 50–25, 50–50, 100–10, 100–25, 100–50).

Table 6.

Cure properties of silicone/cenosphere composites.

Rubber blend	Minimum torque (ML), dN⸳m	Maximum torque (MH), dN⸳m	ΔTorque(MH−ML), dN⸳m	Scorch time (ts₁) (min)	Cure time t90 (min)
0	1.24	16.34	15.1	82.42	0.59
50–10	1.42	20.04	18.62	97.61	0.64
50–25	3.25	25.42	22.17	157.39	0.37
50–50	2.99	29.37	26.38	179.0	0.35
100–10	1.62	19.62	18.0	110.78	0.5
100–25	2.15	27.95	25.80	181.76	0.41
100–50	5.7	46.35	40.65	369.11	0.31

The increase in both minimum (ML) and maximum (MH) torques with increasing cenosphere content reflects enhanced stiffness of the composite system and increased resistance to deformation during curing. The torque difference (ΔTorque = MH − ML), which is commonly associated with the effective rigidity of the cured network and rubber–filler interactions, increases markedly at higher filler loadings, particularly for CS100-filled composites. This behavior is consistent with the SEM observations, indicating restricted matrix mobility and pronounced filler-induced rigidity at high cenosphere contents. At a low filler loading of 10 phr, a moderate increase in scorch time (ts₁) was observed for all cenosphere-filled composites, indicating a slight delay in the onset of vulcanization due to the physical presence of the filler. As the filler content increased to 25–50 phr, a pronounced extension of scorch time was observed, especially for the CS100-filled systems. This trend suggests that higher cenosphere loadings significantly retard the initiation of curing. The retardation of scorch time at high filler contents may be attributed to several factors, including restricted diffusion of peroxide curing agents within the highly filled matrix and possible adsorption of curing species on the aluminosilicate surface. In addition, the presence of metal oxide components in cenospheres (such as Al₂O₃ and Fe₂O₃) may alter the local chemical environment of the peroxide curing system, leading to delayed radical generation and reduced curing efficiency at the early stages of vulcanization.^34,35 Similar retardation effects induced by mineral fillers on peroxide curing kinetics have been reported in the literature.^36,37 This observation is in line with previous reports suggesting that mineral fillers can influence peroxide vulcanization kinetics by modifying the chemical environment of the curing system.³⁸

Mechanical properties

Table 7 summarizes the mechanical properties of the composites, including the tensile strength, tear resistance, elongation at break, and Shore A hardness, as a function of the cenosphere particle size and concentration.

Table 7.

Mechanical properties of silicone/cenosphere composites.

Properties	Rubber blend
Properties	0	50–10	50–25	50–50	100–10	100–25	100–50
Tensile strength (MPa)	9.83SD ± 0.26	6.52SD ± 0.29	4.47SD ± 0.27	3.60SD ± 0.25	5.43SD ± 0.21	3.22SD ± 0.23	2.5SD ± 0.22
Tear resistance (kN/m)	23.8 ± 1.1	32.8 ± 2.0	22.4	16	26.6	19.7	14.4
Elongation at break (%)	666SD ± 15	650SD ± 26	504SD ± 25	295	600	384	209
Shore A hardness	56	63	69	75	63	71	80

Tensile strength

The addition of cenospheres led to a pronounced reduction in tensile strength for all filled composites. The neat silicone rubber exhibited the highest tensile strength (9.83 MPa). Upon the introduction of 10 phr cenospheres, the tensile strength decreased to 6.52 MPa for the CS50-filled composite (−34%) and to 5.43 MPa for the CS100-filled composite (−44.8%). This indicates that even low filler contents significantly weaken the load-bearing capacity of the matrix, with larger cenospheres exerting a more pronounced detrimental effect. As the filler concentration increased to 25 and 50 phr, the tensile strength continued to decline for both cenosphere fractions. This trend is consistent with the SEM observations, which revealed increasing particle clustering, cenosphere fracture, and interfacial discontinuities at higher filler loadings, all of which reduce effective stress transfer across the filler–matrix interface.

Elongation at break

The elongation at break decreased systematically with increasing filler content and, to a lesser extent, with increasing particle size. For the CS50 series, the elongation decreased from 650% at 10 phr to 504% at 25 phr and further to 295% at 50 phr. A similar trend was observed for the CS100-filled composites, where the elongation decreased from 600% at 10 phr to 384% at 25 phr and 209% at 50 phr. These reductions indicate that cenospheres restrict polymer chain mobility and deformability, particularly at higher filler concentrations. The formation of rigid filler networks, particle–particle contacts, and interfacial voids limits large-scale chain rearrangements during deformation, leading to premature failure. Such behavior is consistent with literature reports describing the creation of localized immobilization zones around mineral fillers in elastomeric matrices.³⁹

Tear resistance

In contrast to tensile strength, the tear resistance exhibited an initial increase at low filler loading. At 10 phr, the tear resistance increased from 23.8 kN/m for the neat rubber to 32.8 kN/m for the CS50-filled composite and to 26.6 kN/m for the CS100-filled composite. This enhancement can be attributed to crack deflection and energy dissipation mechanisms introduced by well-dispersed rigid inclusions, which hinder straight crack propagation. However, with increasing filler content, the tear resistance gradually decreased. At 50 phr, the tear resistance dropped to 16.0 kN/m for CS50 and 14.4 kN/m for CS100, reflecting the adverse effects of filler agglomeration, reduced matrix continuity, and the presence of filler-induced defects that facilitate crack initiation and propagation.

Hardness

Shore A hardness increased monotonically with both filler concentration and particle size. The neat rubber exhibited a Shore A hardness of 56, whereas composites containing 50 phr cenospheres reached values of 75 for CS50 and 80 for CS100. This progressive increase reflects the stiffening effect of the inorganic filler phase, which limits polymer chain flexibility and increases the overall rigidity of the composite. The hardness trends are consistent with the observed increases in Mooney viscosity and cure torque and agree with previous reports on mineral-filled silicone and elastomeric systems.⁴⁰ Overall, the mechanical properties of silicone/cenosphere composites are governed by a coupled effect of filler loading and particle size. At low filler loading (10 phr), SEM observations reveal predominantly isolated cenospheres with limited interfacial defects, enabling relatively uniform dispersion and effective stress redistribution. Under these conditions, tear resistance and hardness improve while high elongation is largely preserved, despite a reduction in tensile strength. As the filler content increases to 25–50 phr, the microstructure progressively evolves toward increased particle–particle contacts, clustering, and a higher fraction of fractured hollow spheres. These features disrupt matrix continuity and create interfacial discontinuities, which simultaneously increase rheological resistance and reduce tensile strength and elongation due to stress concentration around agglomerates and voids. Particle size further modulates this structure–property pathway. Smaller cenospheres (CS50) provide a larger number of contact points and are more effectively embedded in the matrix at low loadings, whereas at high loadings their high surface area promotes clustering. Larger cenospheres (CS100), in contrast, are more susceptible to mechanical damage during processing; particle fracture and larger interfacial voids lead to more severe losses in tensile performance at high filler contents. These mechanical trends are consistent with the rheological, morphological, and curing data discussed in previous sections and support a unified interpretation of the size–loading effects on composite performance. The observed mechanical trends are consistent with previous reports on mineral-filled elastomer composites, where increased filler content results in reduced tensile performance and extensibility accompanied by increased stiffness and hardness.^31,41,42

Accelerated thermal aging

Accelerated thermal aging was performed on both the neat silicone rubber (sample 0) and silicone/cenosphere composites (50–10, 50–25, 50–50, 100–10, 100–25, 100–50) in an AO 3000 Rubber Aging Oven (MonTech, Germany) at 250°C for 72 h. Following aging, visual inspection, mechanical testing (tensile strength and hardness), FTIR analysis, and molecular mobility assessments were conducted to evaluate thermal stability and structural integrity.

As shown in Figure 5, thermal aging has a significant impact on unfilled rubber. The neat sample (0) became brittle and glass-like, crumbling under minimal stress, and thus impractical for mechanical testing. In contrast, all filled composites remained elastic and flexible without signs of embrittlement, even after exposure to 250°C for 72 h.

Figure 5.

Digital image of (a) as-received silicone rubber (rubber blend 0) and (b) silicone/cenospheres composites after thermal aging (72 h, 250°C).

The Hardness profiles of the materials across the three processing stages (after the first curing, second curing, and post-aging steps) are shown in Figure 6. A minor increase in Shore A hardness (by 2–4 units) was observed after the second curing step, which is typical of thermal post-curing. However, after thermal aging, the unfilled rubber displayed a dramatic increase in hardness by 32 units, corresponding to a complete loss of elastomeric character. In contrast, the hardness of the composites remained stable or increased only slightly, particularly in samples with 10 and 25 phr of cenospheres. In highly filled composites (50 phr), the increase in hardness was more pronounced; however, the materials maintained their integrity. These results suggest that the presence of cenospheres stabilizes the polymer network, likely by limiting the chain mobility and promoting a more thermodynamically stable structure.

Figure 6.

Shore A hardness of unfilled silicone rubber (sample 0) and silicone/cenosphere composites (50–10, 50–25, 50–50, 100–10, 100–25, 100–50) measured after the first curing stage (first column), the second curing stage (second column), and after accelerated thermal aging (third column).

The tensile strength results before and after thermal aging are shown in Figure 7. The unfilled silicone rubber could not be tested after aging due to catastrophic embrittlement. In contrast, all cenosphere-filled composites retained sufficient mechanical integrity to allow tensile testing after aging. As shown in Figure 7, a pronounced reduction in tensile strength after aging was observed for composites containing 10 and 25 phr of cenospheres for both particle sizes. In comparison, composites with a higher filler loading of 50 phr exhibited a relatively smaller relative decrease in tensile strength after aging, despite their lower absolute strength values. This behavior suggests that although high filler contents lead to reduced initial tensile strength, the rigid filler network formed at 50 phr may partially suppress further degradation during thermal exposure. Importantly, even at 50 phr, all composites remained structurally intact and testable after aging, in contrast to the unfilled silicone rubber.

Figure 7.

The tensile strengths of the as-received silicone rubber (sample 0) and silicone/cenosphere composites (50–10, 50–25, 50–50, 100–10, 100–25, 100–50) before (first column) and after (second column) accelerated thermal aging (250°C, 72 h).

ATR/FT-IR analysis of the composites before and after accelerated thermal aging

The infrared spectra of neat cenospheres and silicone/cenosphere composites before and after accelerated thermal aging are presented in Figures 8 and 9, covering the spectral range from 800 to 3500 cm^-1.

Figure 8.

ATR/FT-IR spectra of cenospheres in the wavelength range 800–3500 cm^-1.

Figure 9.

ATR/FT-IR spectra of silicone/cenospheres composites: 0, 50–10, 50–25, 50–50, 100–10, 100–25, 100–50 before thermal aging (a) and after thermal aging (b) in the wavelength range 800-3500 cm^-1.

Figure 8 displays the ATR/FT-IR spectrum of neat cenospheres, which shows two distinct absorption bands at 1021 cm^-1 and 772 cm^-1, corresponding to asymmetric and symmetric stretching vibrations of Si-O-Si, respectively. These signals are consistent with previous reports on the structural fingerprints of aluminosilicate materials.⁴³

Figure 9 shows the spectra of silicone/cenosphere composites (0, 50–10, 50–25, 50–50, 100–10, 100–25, 100–50) before and after thermal aging. The characteristic absorption bands of the silicone matrix are summarized in Table 8. These include Si-CH₃ bending (784 cm^-1), Si-O-Si stretching (1005 cm^-1), and CH₃ stretching (2960 cm^-1), along with other bands indicative of the siloxane backbone and methyl side groups.

Table 8.

Characteristic absorption bands of silicone rubber.

Functional group	Wavenumber (cm^-1)
Si(CH₃)₃ (bending)	700–800 (784 см^-1)
O-Si(CH₃)₂-O	790–840
Si–O–Si (asymmetric)	1000–1100 (1005 см^-1)
Si–CH₃ (stretching)	1255–1270 (1258 см^-1)
CH₃ (deformation)	1410–1440
CH₃ (stretching)	2960–2963

Because the chemical bonds in the cenospheres and silicone are similar (primarily based on Si-O-Si linkages), no new absorption bands appear upon filler incorporation. Instead, the cenosphere-related signals were overlapped by strong silicone matrix absorptions. Comparison of the FT-IR spectra before and after thermal aging reveals no significant appearance of carbonyl bands in the 1715–1730 cm^-1 region, which would indicate oxidative degradation. Similarly, no broadening in the 3200–3700 cm^-1 region was observed, implying a lack of -OH accumulation on the polymer surface. This suggests that thermal aging at 250°C for 72 h does not induce significant oxidation or hydrolytic degradation in the filled composites. However, slight changes in the intensity and shape of the 1005 cm^-1 band were observed, particularly in the unfilled sample (sample 0), which may reflect a minor degradation of the siloxane backbone. In contrast, samples filled with cenospheres showed minimal spectral changes, indicating better structural stability during thermal aging. To further assess the structural evolution of the silicone matrix, the peak height ratio (H-value), defined as the absorbance ratio A₇₈₄/A₁₀₀₅, was calculated for three representative samples (neat rubber, CS50–10, and CS100–10). The corresponding values before and after aging are presented in Table 9.

Table 9.

H-values (A₇₈₄/A₁₀₀₅) before and after aging.

Absorption band (cm^-1)	0		50-10		100-10
Absorption band (cm^-1)	Before aging	After aging	Before aging	After aging	Before aging	After aging
784	1.1459	1.1300	1.1413	1.0561	1.0796	1.0632
1005	1.1447	1.1159	1.1187	1.0440	1.0669	1.0582
H-value	1.0010	1.0126	1.0202	1.0116	1.0119	1.0047

In the neat silicone sample, the H-value increased after aging (from 1.0010 to 1.0126), suggesting a higher relative concentration of methyl groups, possibly due to the thermal degradation of the main siloxane backbone accompanied by post-curing or crosslinking reactions (so-called “over-curing” or network densification). In contrast, the H-values of the cenosphere-filled composites (CS50–10 and CS100–10) remained nearly constant after aging, suggesting that the silicone network structure was largely preserved. This behavior is consistent with the mechanical and hardness results and further confirms the stabilizing effect of cenospheres during thermal aging. Overall, ATR/FT-IR analysis demonstrates that cenosphere-filled silicone composites undergo minimal chemical changes under accelerated thermal aging, whereas the neat rubber exhibits signs of structural modification.

Molecular dynamics of silicone/cenosphere composites before and after accelerated thermal aging

The molecular dynamics of the silicone matrix were investigated using electron spin resonance (ESR) spectroscopy with a stable nitroxyl radical (TEMPO) as a spin probe. The rotational correlation time (τ), which reflects the local microviscosity and segmental mobility of polymer chains, was used as a key parameter to evaluate the influence of cenosphere loading, particle size, and thermal aging on molecular mobility.^44–46 The τ-values obtained for the unfilled silicone rubber and for composites containing cenospheres of different sizes and concentrations before and after thermal aging are summarized in Table 10.

Table 10.

Rotational correlation times (τ × 10^-10 s) of TEMPO radical in silicone/cenosphere composites.

Rubber blend	Rotational correlation times of TEMPO spin probe (τ·10^-10)
Rubber blend	Before aging	After aging
0	1.25	1.77
50–10	0.77	1.02
50–25	1.46	1.16
50–50	1.49	1.15
100–10	0.79	1.16
100–25	1.63	1.05
100–50	1.81	1.03

The neat silicone rubber exhibited a rotational correlation time of 1.25 × 10^-10 s prior to aging. Interestingly, the introduction of a low amount (10 phr) of cenospheres led to a noticeable decrease in τ to 0.77 × 10^-10 s for CS50–10 and 0.79 × 10^-10 s for CS100–10. This decrease indicates a slight increase in segmental mobility, which may be associated with local rearrangements of the polymer network in the vicinity of well-dispersed filler particles. Such behavior has been reported previously for lightly filled elastomer systems, where low filler contents do not yet form a rigid filler network and may locally perturb chain packing.

As the filler content increased to 25 and 50 phr, the τ-values increased markedly. For the CS50 series, τ rose to 1.46–1.49 × 10^-10 s, while for the CS100-filled composites, τ reached even higher values of 1.63–1.81 × 10^-10 s. These increases indicate restricted molecular mobility and increased microviscosity of the silicone matrix. The more pronounced effect observed for the CS100 composites can be attributed to the larger particle size, which promotes stronger topological constraints, increased particle–particle contacts, and enhanced restriction of polymer chain motion at higher filler loadings.

Thermal aging at 250°C for 72 h significantly affected the molecular dynamics of the unfilled silicone rubber, with τ increasing from 1.25 to 1.77 × 10^-10 s. This behavior reflects densification of the polymer network, which may result from thermal relaxation, post-curing reactions, or oxidative crosslinking processes. In lightly filled composites (10 phr), a similar increase in τ was observed after aging, indicating reduced chain flexibility due to aging-induced structural rearrangements.

In contrast, for composites containing higher filler loadings (25–50 phr), the effect of thermal aging on τ was less pronounced. In some cases, a slight decrease in τ was observed after aging, which may be associated with stress relaxation and structural reorganization within the constrained polymer regions surrounding rigid filler domains. Because molecular mobility in these systems is already strongly restricted by the high filler content, additional thermal exposure induces comparatively smaller changes in chain dynamics.

Overall, the ESR results demonstrate that aluminosilicate cenospheres significantly modify the molecular dynamics of the silicone matrix in a manner that depends on both filler loading and particle size. At low filler contents, cenospheres may locally enhance chain mobility, whereas at higher loadings they act as rigid constraints that suppress segmental motion. The reduced sensitivity of highly filled composites to thermal aging further confirms the stabilizing effect of cenospheres on the silicone network, in agreement with the trends observed in mechanical performance, hardness evolution, and FTIR analysis.

TG/DTG analysis of the silicone/cenospheres composites

Thermogravimetric analysis (TGA) was performed to investigate the thermal stability and decomposition behavior of the silicone rubber composites filled with cenospheres. The results are presented in Figure 10.

Figure 10.

The residual solid mass versus temperature (TG) curves and derivative residual mass versus temperature (DTG) curves of as-received silicone and silicone/cenospheres composites: (a) - as-received cenospheres; (b) - 0; (c) - 50-10; (d) - 50-25; (e) - 50-50; (f) - 100-10; (g) - 100-25; (h) 100-50 under 20°C/min heating rate.

The TG curve of the neat cenospheres (Figure 10(a)) showed no mass loss up to 800°C, confirming their excellent thermal stability. In contrast, the TG and DTG curves of the silicone-based composites (Figures 10(b)–(h)) exhibit thermal degradation typical of cross-linked polysiloxanes, with a multi-step decomposition mechanism that depends on the filler type and content. The TG curve of as-received silicone revealed that when the temperature reached 411°C, decomposition reaction of the silicone rubber material began at a heating rate of 20°C/min, and this decomposition reaction completes at approximately 800°C. The dominant peak in the DTG curves indicates that the thermal decomposition of the as-received silicone rubber material is a single-stage reaction.

The decomposition of the unfilled silicone rubber begins at approximately 411°C, with the major mass loss occurring between 411 and 600°C, and the process is completed near 800°C. The DTG curve for the unfilled rubber reveals a two-stage decomposition pattern, which is consistent with the earlier reports by Nair et al.⁴⁷ and Rybinski et al.¹ The first stage is attributed to the depolymerization of the siloxane network into cyclic oligomers, predominantly trimers (D₃), tetramers (D₄), and pentamers (D₅), facilitated by the presence of oxygen. The second stage, occurring between 450 and 600°C, corresponds to the continued release of these volatiles, with D₃ remaining the dominant product throughout. Several distinct changes were observed upon the introduction of cenospheres at concentrations above 10 phr. First, the temperature of the maximum decomposition rate (Tmax) in the second degradation stage shifted to higher values, suggesting retardation of thermal degradation. Second, a third DTG peak temperature appears between 670 and 750°C, which was absent in the unfilled sample. It should be noted that the primary thermal degradation of polysiloxanes typically occurs in the temperature range of approximately 300–500°C and is associated with depolymerization of the siloxane backbone into cyclic oligomers (D3–D5) and the formation of silica-rich residues. In the present study, this behavior corresponds to the main DTG peaks observed for the unfilled silicone rubber.

The additional DTG feature appearing in the range of 670–750°C for cenosphere-filled composites should therefore not be interpreted as primary polysiloxane decomposition. Instead, this high-temperature peak is attributed to secondary thermal processes, including oxidation and restructuring of residual carbonaceous species, further condensation of silica networks, and high-temperature reactions within the inorganic-rich char layer formed in the presence of aluminosilicate cenospheres. Similar high-temperature DTG features have been reported for silicone rubber composites containing mineral fillers and are commonly associated with ceramization-related transformations and stabilization of the inorganic residue.

Although direct analysis of gaseous decomposition products was not performed in this study, the emergence of the high-temperature DTG peak exclusively in filled systems, together with the increased final residue yield, supports its assignment to secondary oxidation and inorganic phase reorganization rather than to primary polysiloxane degradation. The presence of carbonaceous gases (CO and CO₂) and volatile hydrocarbons in this temperature region, as reported in the literature, supports this interpretation.⁴⁸ These changes indicate that the cenospheres enhance the thermal resistance of the composite by creating a more robust structure at high temperatures. The final residue (P₈₀₀), primarily pyrogenic silica, is significantly higher in cenosphere-filled composites (up to 79.85%) compared to the neat rubber (55.29%), as shown in Table 11. This increase correlates with the flame-retardant effect of the filler because a higher residue content implies a lower release of combustible volatiles into the flame zone. The initial decomposition temperatures (T₅) remain above 400°C for all samples, which can be attributed to the platinum catalyst used during vulcanization. Platinum inhibits early depolymerization of PDMS by preventing chain scission and cyclomer formation at lower temperatures.⁴⁷ The resultant white product, or residue after heating to 800°C, was pyrogenic silica formed during the combustion process. This thermal decomposition residue (parameter Pw) significantly affects the flammability of the polymeric materials. An increase in the Pw value indicates that fewer destruction products enter the flame. The flame-retardant effect of the cenospheric filler is linked to its beneficial effect on the boundary layer structure, which significantly reduces the segmental mobility of the polymeric chains, thereby decreasing the decomposition and chain transfer efficiencies.⁴⁸ The presence of aluminosilicate cenospheres promotes ceramization-related behavior in silicone rubber composites under high-temperature exposure. In silicone-based systems, ceramization is generally associated with the transformation of the polymer matrix into a silica-rich inorganic framework accompanied by the incorporation of thermally stable mineral fillers. In the present study, this behavior is indirectly evidenced by the significantly increased inorganic residue yield after heating to 800°C, the appearance of an additional high-temperature DTG feature (670–750°C) observed only in filled composites, and the enhanced thermal stability and structural integrity of the materials after prolonged thermal aging. While the formation of a continuous ceramic surface layer was not directly characterized in this work, the combined TG/DTG results indicate progressive consolidation and reorganization of the inorganic phase at elevated temperatures. Similar behavior has been reported for silicone rubber composites containing mineral fillers and is commonly interpreted as a precursor to ceramization, contributing to improved fire resistance and thermal protection by limiting volatile release and oxygen diffusion. Therefore, the obtained results suggest that cenospheres facilitate ceramization-related transformations rather than conclusively demonstrating a fully developed ceramic coating. In silicone rubber containing mineral fillers, silica bridges may form with adhesion forces between the silica from polymer degradation, and the thermally stable filler plays a critical role.

Table 11.

Thermal properties of silicone/cenosphere composites. T₅, T₂₀—temperatures of sample at 5 and 20% mass loss; P800/%—residue after heating to 800°C, T_max (temperature of the maximal rate of decomposition).

Sample	T_onset (°С)	T₅/°C	T₂₀/°C	P₈₀₀/%	T_max (DTG peak (°C))
Sample	T_onset (°С)	T₅/°C	T₂₀/°C	P₈₀₀/%	1 DTG peak	2 DTG peak	3 DTG peak
0	411	454	534	55.29	456	520
50–10	409	453	563	71.9	433	520
50–25	433	467	577	69.45	481	603	752
50–50	431	479	552	72.42	436	539	724
100–10	411	451	556	66.94	462	555	675
100–25	422	473	700	79.85	461	-	673
100–50	429	456	585	77.16	415	557

Specific heat

The influence of aluminosilicate cenospheres on the thermal storage characteristics of silicone rubber was evaluated by measuring the specific heat capacity (Cp) of the composites with different filler loadings and particle sizes. The results summarized in Table 12 demonstrate a systematic decrease in Cp with increasing cenosphere content for both CS50 and CS100-filled systems.

Table 12.

Specific heat capacity (Cp, J·g^-1 K^-1) of silicone rubber and silicone/cenosphere composites.

Rubber blend	Cp, $\frac{J}{k g ℃}$
0	1.57
50–10	1.45
50–25	1.38
50–50	1.28
100–10	1.47
100–25	1.44
100–50	1.36

This behavior is primarily associated with the partial replacement of the silicone matrix, which possesses a relatively high heat capacity, by hollow aluminosilicate particles characterized by lower intrinsic Cp values. In addition, the incorporation of cenospheres reduces the effective volume fraction of the polymer matrix capable of storing thermal energy. Similar reductions in Cp have been reported for polymer composites filled with thermally inert mineral fillers and hollow particles, where the overall heat storage capacity is governed by both the filler fraction and the polymer–filler interfacial interactions.

At a low filler loading of 10 phr, a modest particle size effect can be observed. Composites filled with smaller cenospheres (CS50) exhibit a slightly lower Cp value (1.45 J·g^-1 K^-1) compared to those containing larger cenospheres (CS100, 1.47 J·g^-1 K^-1). This difference may be attributed to the higher specific surface area of CS50 particles, which increases the interfacial contact between the filler and the silicone matrix and imposes stronger constraints on polymer chain mobility in the interfacial regions. As a result, the effective contribution of polymer segments to heat storage is reduced.

With increasing filler content (25–50 phr), the Cp values continue to decrease for both particle sizes, and the influence of particle size becomes less pronounced compared to the dominant effect of filler loading. At high cenosphere contents, the reduction in Cp is mainly governed by the volumetric replacement of the polymer phase and the formation of a rigid filler-dominated structure, rather than by subtle interfacial effects. These observations are consistent with the rheological and ESR results discussed in the corresponding sections, which indicate progressively restricted chain mobility at higher filler loadings.

In addition to thermal capacity, the incorporation of hollow cenospheres significantly affects the density of the silicone rubber composites. As shown in Table 12, the composite density decreases monotonically with increasing filler content due to the low density and hollow morphology of the aluminosilicate cenospheres. This density reduction is more pronounced at higher filler loadings and represents a key advantage of using cenospheres compared to conventional solid mineral fillers.

The combined reduction in density and specific heat capacity suggests that cenosphere-filled silicone rubber composites may offer improved thermal insulation efficiency on a mass basis, as less material is required to achieve comparable thermal resistance. Although direct measurements of thermal conductivity were not performed in the present study, the observed trends in Cp and density are consistent with literature reports on silicone-based syntactic foams and hollow-particle-filled elastomers, where reduced density and heat capacity contribute to enhanced potential thermal insulating performance.

Overall, the results indicate that the incorporation of aluminosilicate cenospheres with particle sizes CS50 and CS100 enables simultaneous reduction of density and thermal storage capacity in silicone rubber composites. Moderate filler loadings provide a balanced combination of reduced Cp, lower density, and preserved mechanical integrity, whereas higher filler contents lead to more pronounced thermal effects accompanied by increased stiffness and reduced extensibility, as discussed in previous sections.

Swelling resistance

The swelling behavior of silicone/cenosphere composites was evaluated in various polar and non-polar liquid environments over a 96-h exposure period. The swelling degree as a function of time for unfilled silicone rubber and filled composites (50–10, 50–25, 50–50, 100–10, 100–25, 100–50) is presented in Figure 11 for five representative liquids: petroleum solvent, Vaseline oil, disinfectant solution, 10% NaOH, and acetic acid.

Figure 11.

Swelling as a function of time of the as-received silicone rubber material (sample 0) and silicone/cenosphere composites (50–10, 50–25, 50–50, 100–10, 100–25, 100–50) in (a) petroleum solvent, (b) Vaseline oil, (c) disinfectant, (d) NaOH (10%), and (e) acetic acid.

In non-polar environments, such as petroleum-based solvents and Vaseline oil, the addition of cenospheres led to a pronounced decrease in the degree of swelling. For instance, in petroleum solvent, the unfilled silicone rubber swelled by 205% over 96 h. Composites with higher filler loadings (50 phr) generally exhibited reduced swelling compared to lower-filled systems; however, the swelling degree exceeded 96% for the 50–10 and 50–25 composites and dropped below this value only at the highest filler loading. This behavior is attributed to the polar nature of the cenospheres (composed primarily of SiO₂, Al₂O₃, and Fe₂O₃), which hinders the absorption of non-polar solvents due to interfacial incompatibility. The greater the filler content, the more pronounced the reduction in solvent uptake, whereas the particle size had a negligible influence in non-polar media. These findings confirm that cenospheres act as barriers to oil and fuel penetration, enhancing the oil- and solvent-resistance of the composites.⁴⁹ A different trend was observed in polar aqueous systems (e.g., disinfectant solution, NaOH, and acetic acid). For acetic acid, all composite samples exhibited low overall swelling (below 3%), suggesting the good inherent acid resistance of the silicone matrix. However, a modest increase in swelling with increasing filler content was noted, which was likely due to the partial interaction between the hydrophilic cenospheres and the polar solvent. Composites incorporating larger cenospheres exhibited marginally higher degrees of swelling at a constant filler concentration of 10 phr. This observation suggests that the surface area and reactivity of the cenospheres contribute to the increased liquid uptake in polar environments. In the disinfectant solution, the degree of swelling increased with the filler content. This effect is likely due to the interactions between the polar filler surface and polar solvent, facilitating limited water penetration. In contrast, in the 10% NaOH solution, all the samples exhibited a slight mass loss after immersion, indicating partial leaching or dissolution of some filler components. The most pronounced mass loss occurred in composites containing 50 phr of the smallest cenospheres (50–50, 50–25), likely because finely dispersed particles are less effectively bound within the matrix and are more susceptible to being washed out.

These findings were further supported by the ATR/FTIR analysis conducted on the dried composites after 96 h of immersion (Figure 12).

Figure 12.

ATR/FT-IR spectra of silicone/cenosphere composites: 0, 50–10, 50–25, 50–50, 100–10, 100–25, 100–50 after 96 h immersion in petroleum solvent, Vaseline oil, disinfectant, NaOH (10%), and acetic acid.

No new absorption bands were observed, confirming the absence of any detectable chemical degradation or functional group formation. However, a general reduction in peak intensity was observed, particularly in the Si-O-Si absorption region around 1005 cm^-1. This reduction was more significant in composites exposed to disinfectant, NaOH, and acetic acid media, where swelling or mass loss was noted. Importantly, the intensity of the C-H absorption band (∼1410 cm^-1), corresponding to the side-chain methyl groups, remained stable, suggesting that the changes in the Si-O-Si region were primarily linked to filler interactions or partial filler loss rather than matrix degradation. The ratio of the CH₃ (784 cm^-1) to Si-O-Si (1005 cm^-1) absorption peaks (H-value), commonly used to assess the surface polarity of silicone composites, was also analyzed (Table 13). Higher H-values correlate with increased hydrophilicity and decreased surface non-polarity. In non-polar solvents, the H-values remained relatively stable or increased slightly, whereas in polar media (e.g., NaOH, disinfectant), the H-value increased more significantly, indicating a decrease in surface hydrophobicity after exposure.

Table 13.

Ratio of absorption peak heights A₇₈₄/A₁₀₀₅ after solvent immersion.

Sample	Peak absorption ratio A784 cm^-1/A1005 cm^-1 (H)
Sample	0	50–10	50–25	50–50	100–10	100–25	100–50
As-received	0.98	1.02	1.01	1.03	1.02	1.02	1.03
Petroleum solvent	0.98	1.03	1.10	1.11	1.03	1.08	1.09
Vaseline	0.98	1.03	1,02	1.03	1.02	1.01	1.03
Disinfectant	0.98	1.05	0,93	1.00	1.10	1.04	1.15
Acetic acid	1.02	1,05	1,15	1.09	1.09	0.9	1.09
NaOH (10%)	1.14	1.15	1.15	1.24	1.10	1.09	1.11

These results suggest that cenospheres are chemically active in aqueous environments and can undergo partial dissolution, releasing silica and metal cations (e.g., Al³⁺, Fe³⁺, and Ca²⁺). This process may lead to the formation of secondary compounds such as silicic acid, calcium hydroxide, or carbonate precipitates. The interaction between the cenospheres and aqueous media explains the observed swelling and weight changes in polar environments, as well as the variations in IR absorption behavior. The incorporation of cenospheres into silicone rubber significantly improves the resistance to non-polar oils and solvents, while in polar media, the response is more complex and influenced by the filler content and particle size. These findings provide important insights into the chemical stability and environmental durability of silicone-based composites under real-world exposure conditions.

Conclusion

This study systematically investigated the combined influence of aluminosilicate cenosphere particle size (CS50 and CS100) and filler loading (10–50 phr) on the structural, mechanical, rheological, thermal, and environmental resistance properties of silicone rubber composites. Morphological analysis revealed that at low filler loadings (10 phr) both CS50 and CS100 cenospheres were relatively well dispersed within the silicone matrix, with limited interfacial defects. Increasing the filler content to 25–50 phr resulted in progressive particle clustering, shell fracture, and interfacial debonding, particularly for larger cenospheres (CS100), which was attributed to their lower specific surface area and higher susceptibility to mechanical damage during processing. Rheological measurements demonstrated a strong filler-loading-dependent increase in Mooney viscosity. At 10 phr, viscosity increased only slightly (by approximately 3–4 units), indicating good processability. At 25 phr, the viscosity increased by 13 units for both particle sizes, while at 50 phr it reached 61 units for CS50 and 69 units for CS100, reflecting pronounced restriction of polymer chain mobility and reduced processability at high filler contents. Mechanical testing showed a clear trade-off between stiffness and extensibility. Tensile strength and elongation at break decreased systematically with increasing filler content, whereas Shore A hardness and tear resistance increased at low filler loadings. Notably, the composite containing 10 phr of CS50 retained a high elongation at break (∼650%) while exhibiting enhanced tear resistance (∼32.8 kN·m^-1) and only a moderate increase in hardness compared to the neat silicone rubber, indicating effective reinforcement without severe embrittlement. At higher filler loadings (50 phr), mechanical performance was significantly reduced due to filler agglomeration and interfacial discontinuities. Accelerated thermal aging at 250°C for 72 h demonstrated a substantial improvement in thermal durability for all cenosphere-filled composites. While the unfilled silicone rubber became brittle and untestable, composites containing cenospheres preserved their elastomeric integrity. ESR spectroscopy confirmed a filler-loading-dependent reduction in molecular mobility, whereas FTIR analysis revealed no evidence of oxidative degradation after aging. Thermogravimetric analysis further showed delayed thermal decomposition, increased inorganic residue yields exceeding 65% at 800°C, and the appearance of an additional high-temperature DTG feature in the range of 670–750°C, indicating secondary high-temperature stabilization and ceramization-related transformations in filled systems. The incorporation of hollow cenospheres led to a systematic decrease in specific heat capacity and composite density. In addition, swelling resistance in non-polar solvents was significantly enhanced with increasing filler content due to the barrier effect of the inorganic filler. In polar media, partial swelling and leaching phenomena were observed at high filler loadings, highlighting interfacial limitations of unmodified cenospheres. Overall, the results demonstrate that aluminosilicate cenospheres at moderate loadings of 10–25 phr, particularly with smaller particle size (CS50), provide the most favorable balance between mechanical retention, rheological processability, thermal stability, and environmental resistance. In contrast, high filler contents (50 phr), especially when using larger cenospheres (CS100), lead to pronounced interfacial degradation and reduced mechanical performance despite further gains in thermal residue. These findings establish a clear structure–property–performance relationship and define a practical design window for cenosphere-filled silicone rubber composites intended for thermally demanding elastomer applications. Future work will focus on surface functionalization of cenospheres using silane coupling agents to improve interfacial adhesion and enable higher filler loadings without compromising mechanical integrity.

Footnotes

ORCID iD

Ksenia Sukhareva

Author contributions

Conceptualization, supervision, project administration, writing—original draft preparation, writing—review and editing: Sukhareva K.V.; investigation: Besedina V.O., Kolesnikov E., Kiselev N.V., Karpova S.G.; formal analysis: Lyusova L.R.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.*

References

Rybiński

Żukowski

Bradło

. Influence of cenosphere particles on thermal properties composites of silicon rubber. J Therm Anal Calorim 2015; 122: 1307–1318. https://doi.org/10.1007/s10973-015-4829-0

Zeng

Guo

Zhang

, et al. Review of aging evaluation methods for silicone rubber composite insulators. Polymers 2023; 15: 1141. https://doi.org/10.3390/polym15051141

Morawska-Chochół

Szumera

Młyniec

, et al. The effect of aging process conditions on the thermal properties of Poly(Dimethylsiloxane)-Based silicone rubber. Materials 2024; 17: 5608. https://doi.org/10.3390/ma17225608

Zeng

Gao

, et al. Aging analysis of HTV silicone rubber under coupled Corona discharge, humidity and cyclic thermal conditions. Electronics (Basel) 2025; 14: 4071. https://doi.org/10.3390/electronics14204071

. Silicone rubber thermal aging performance for cables and accessories. J Mater Sci Mater Electron 2024; 35: 328. https://doi.org/10.1007/s10854-024-12074-w

Hamdani

Longuet

Lopez-Cuesta

J-M

, et al. Calcium and aluminium-based fillers as flame-retardant additives in silicone matrices. I. Blend preparation and thermal properties. Polym Degrad Stabil 2010; 95: 1911–1919. https://doi.org/10.1016/j.polymdegradstab.2010.04.013

Hosseini‐Nasab

saber‐Tehrani

Haghgoo

, et al. Design and evaluation of lightweight <scp>EPDM</scp> composites using hydrophobic silica aerogel for thermal management systems. Polym Int 2025, Epub ahead of print 14. https://doi.org/10.1002/pi.70045

Kucukesmen

Sarac

. Preparation and characterization of fumed silica added PMMA denture composites. Epub ahead of print 22 February 2024. https://doi.org/10.21203/rs.3.rs-3972161/v1

Rwei

S-P

F-H

Cheng

K-C

. Dispersion of carbon black in a continuous phase: electrical, rheological, and morphological studies. Colloid Polym Sci 2002; 280: 1110–1115.

10.

Moronkeji

Das

Lee

, et al. Local electrical conductivity of carbon black/PDMS nanocomposites subjected to large deformations. J Compos Mater 2023; 57: 507–519. https://doi.org/10.1177/00219983231156253

11.

Chen

Botef

Guduri

, et al. Thermal and bonding properties of nano size carbon Black filled PDMS. AIP Conf Proc. 2010; 1315: 243–248. https://doi.org/10.1063/1.3504305

12.

Madidi

Momen

Farzaneh

. Dielectric properties of TiO 2/Silicone rubber Micro- and nanocomposites. Adv Mater Sci Eng 2018; 2018: 1–7. https://doi.org/10.1155/2018/4682076

13.

Wadatkar

Shende

Wasewar

. Fly ash cenosphere - formation, separation, and applications in diverse fields. Journal of Environmental Nanotechnology 2024; 13: 456–469. https://doi.org/10.13074/jent.2024.06.242601

14.

Barrera

Cornish

. Fly ash as a potential filler for the rubber industry. In Handbook of Fly Ash. Elsevier, 2022, pp. 763–792.

15.

Mohamad Altaweel

AMA

Ranganathaiah

Siddaramaiah . Siddaramaiah. Erratum: Positron lifetime spectroscopy and differential scanning calorimetric study of polystyrene-based composites with fly ash, cenospheres, and calcium aluminosilicate as fillers. J Appl Polym Sci 2011; 119: 1249. https://doi.org/10.1002/app.33066

16.

Trino

ASM

Costa

CSMF

Fonseca

, et al. Novel composites from green unsaturated polyesters and fly ashes: preparation and characterization. React Funct Polym 2016; 106: 24–31. https://doi.org/10.1016/j.reactfunctpolym.2016.07.004

17.

Sathishkumar

Rajkumar

Srinivasan

, et al. Structural analysis and mechanical properties of lignite fly-ash-added jute–epoxy polymer matrix composite. J Reinforc Plast Compos 2018; 37: 90–104. https://doi.org/10.1177/0731684417735183

18.

Gupta

Brar

Woldesenbet

. Effect of filler addition on the compressive and impact properties of glass fibre reinforced epoxy. Bull Mater Sci 2001; 24: 219–223. https://doi.org/10.1007/bf02710105

19.

Hegde

Padmaraj

Siddesh

, et al. Experimental investigation of mechanical sustainability and acoustic performance of fly ash cenosphere/epoxy polymer composites. J King Saud Univ Eng Sci 2021, Epub ahead of print. https://doi.org/10.1016/j.jksues.2021.11.002

20.

Dochev

Ublekov

Dimova

, et al. Development of composites based on thermosetting polymers (epoxy, polyester and vinylester resins) with cenosphere fillers in different concentrations – 5 wt.%, 10 wt.% and 15 wt. Environment technology resources: Proceedings of the International Scientific and Practical Conference 2025; 4: 81–84. https://doi.org/10.17770/etr2025vol4.8431

21.

Satapathy

Bihari Nando

. Mechanical, dynamic mechanical, and thermal characterization of fly ash and nanostructured fly ash-waste polyethylene/high-density polyethylene blend composites. Polym Compos 2016; 37: 3256–3268. https://doi.org/10.1002/pc.23524

22.

Besco

Brisotto

Gianoncelli

, et al. Processing and properties of polypropylene-based composites containing inertized fly ash from municipal solid waste incineration. J Appl Polym Sci 2013; 130: 4157–4164. https://doi.org/10.1002/app.39692

23.

Besco

Bosio

Brisotto

, et al. Structural and mechanical characterization of sustainable composites based on recycled and stabilized fly ash. Materials 2014; 7: 5920–5933. https://doi.org/10.3390/ma7085920

24.

Bose

Mahanwar

. Effect of flyash on the mechanical, thermal, dielectric, rheological and morphological properties of filled nylon 6. J Miner Mater Char Eng 2004; 3: 65–72. https://doi.org/10.4236/jmmce.2004.32007

25.

Kulkarni

Bambole

Mahanwar

. Effect of particle size of fly ash cenospheres on the properties of acrylonitrile butadiene styrene-filled composites. J Thermoplast Compos Mater 2014; 27: 251–267. https://doi.org/10.1177/0892705712443253

26.

Panitchakarn

Wikranvanich

Phisalaphong

. Synthesis and characterization of natural rubber/coal fly ash composites via latex aqueous microdispersion. J Mater Cycles Waste Manag 2019; 21: 134–144. https://doi.org/10.1007/s10163-018-0774-x

27.

Sombatsompop

Thongsang

Markpin

, et al. Fly ash particles and precipitated silica as fillers in rubbers. I. Untreated fillers in natural rubber and styrene–butadiene rubber compounds. J Appl Polym Sci 2004; 93: 2119–2130. https://doi.org/10.1002/app.20693

28.

Ren

Sancaktar

. Use of fly ash as eco-friendly filler in synthetic rubber for tire applications. J Clean Prod 2019; 206: 374–382. https://doi.org/10.1016/j.jclepro.2018.09.202

29.

Ramesan

. Effect of fly ash on thermal stability, flammability, oil resistance and transport properties of chlorinated styrene butadiene rubber composites. J Elastomers Plast 2013; 46: 303–324. https://doi.org/10.1177/0095244312468440

30.

Surya

Sudibyo

Khatami

, et al. The tear and tensile strengths of silica fly ash reinforced natural rubber vulcanizate. ASEAN Journal of Chemical Engineering 2025; 25: 458–470. https://doi.org/10.22146/ajche.18182

31.

Sreekanth

Joseph

Mhaske

, et al. Effects of mica and fly ash concentration on the properties of polyester thermoplastic elastomer composites. J Thermoplast Compos Mater 2011; 24: 317–331. https://doi.org/10.1177/0892705710389293

32.

Qiao

Amirkhizi

Schaaf

, et al. Dynamic mechanical analysis of fly ash filled polyurea elastomer. J Eng Mater Technol; 133. https://doi.org/10.1115/1.4002650, Epub ahead of print 3 December 2010.

33.

Dai

Wang

, et al. Effect and enhancement mechanism of silane coupling agent modified fly ash cenospheres on the mechanical and tribological properties of epoxy composites. J Appl Polym Sci; 143. https://doi.org/10.1002/app.70048. Epub ahead of print 15 February 2026.

34.

Smejda-Krzewicka

Olejnik

Strzelec

. The effect of metal oxide on the cure, morphology, thermal and mechanical characteristics of chloroprene and butadiene rubber blends. Polym Bull 2020; 77: 4131–4146. https://doi.org/10.1007/s00289-019-02964-z

35.

Smejda-Krzewicka

Mrozowski

Kobędza

, et al. The use of Iron(II,III) oxide (Fe3O4) as a cross-linking agent for unfilled and filled chlorosulfonated polyethylene (CSM) and study of the vulcanizates properties. Materials 2022; 15: 7276. https://doi.org/10.3390/ma15207276

36.

Kantala

Wimolmala

Sirisinha

, et al. Reinforcement of compatibilized NR/NBR blends by fly ash particles and precipitated silica. Polym Adv Technol 2009; 20: 448–458. https://doi.org/10.1002/pat.1293

37.

Sombatsompop

Wimolmala

Sirisinha

. Fly ash particles and precipitated silica as fillers in rubbers. III. Cure characteristics and mechanical and oil‐resistance properties of acrylonitrile‐butadiene rubber. J Appl Polym Sci 2008; 110: 2877–2883. https://doi.org/10.1002/app.28842

38.

Thongsang

Vorakhan

Wimolmala

, et al. Dynamic mechanical analysis and tribological properties of NR vulcanizates with fly ash/precipitated silica hybrid filler. Tribol Int 2012; 53: 134–141. https://doi.org/10.1016/j.triboint.2012.04.006

39.

Joseph

Bambola

Sherhtukade

, et al. Effect of flyash content, particle size of flyash, and type of silane coupling agents on the properties of recycled poly(ethylene terephthalate)/flyash composites. J Appl Polym Sci 2011; 119: 201–208. https://doi.org/10.1002/app.32449

40.

Nguyen

. Effects of the amount of fly ash modified by stearic acid compound on mechanical properties, flame retardant ability, and structure of the composites. Int J Chem Eng 2020; 2020: 1–6. https://doi.org/10.1155/2020/2079189

41.

Maurya

Singh

Amanulla

, et al. Mechanical, electrical and thermal properties of Nylon-66/Flyash composites: effect of flyash. Organic Polymer Material Research; 4, Epub ahead of print 2022. https://doi.org/10.30564/opmr.v4i2.5233

42.

Dasaesamoh

Osotchan

Subannajui

. Mechanical strength of natural rubber filled fly ash. IOP Conf Ser Mater Sci Eng; 625, Epub ahead of print 2019. https://doi.org/10.1088/1757-899X/625/1/012013

43.

Gohatre

Biswal

Mohanty

, et al. Study on thermal, mechanical and morphological properties of recycled poly(vinyl chloride)/fly ash composites. Polym Int 2020; 69: 552–563. https://doi.org/10.1002/pi.5988

44.

Sokolova

Nepomnyashchiy

Tatarinov

. Rotational mobility of the radical probe in butadiene-nitrile rubbers. Polym Sci 2017; 59: 27–32. https://doi.org/10.1134/s0965545x17010151

45.

Naveed

K-R

Wang

, et al. Recent progress in the electron paramagnetic resonance study of polymers. Polym Chem 2018; 9: 3306–3335. https://doi.org/10.1039/c8py00689j

46.

Wasserman

Barashkova

Vasil’ev

, et al. EPR spin probe study of molecular mobility and structure of aqueous solutions and gels of polydiphenylenesulfophthalide. Appl Magn Reson 2015; 46: 1409–1420. https://doi.org/10.1007/s00723-015-0702-3

47.

Nair

Aswathy

Mathew

, et al. Studies on the thermal properties of silicone polymer based thermal protection systems for space applications. J Therm Anal Calorim 2017; 128: 1731–1741. https://doi.org/10.1007/s10973-016-6025-2

48.

Rybiński

Żukowski

Bradło

. Effect of cenospheric fillers on the flammability and fire hazard of silicone rubber composites. J Therm Anal Calorim 2016; 125: 1373–1386. https://doi.org/10.1007/s10973-016-5741-y

49.

Padhi

Priyadarsini Jena

Nayak

. Dielectric behaviour of ethylene vinyl acetate/cenosphere composites. Mater Today Proc 2020; 30: 355–359. https://doi.org/10.1016/j.matpr.2020.02.693