The Influence of Filler Morphology and Loading Level on the Properties of Light-Curing Dental Composites

Abstract

Background/Objectives: Light-curing dental resin composites remain limited by high polymerization shrinkage, inadequate wear resistance, and elevated water sorption. The combined influence of filler shape, size, and loading level on mechanical performance and hydrolytic stability remains insufficiently understood. This study aimed to systematically investigate the effects of filler morphology and particle size distribution on the key properties of dental composites. Methods: Spherical silica (SiO₂) nanoparticles (D50 = 0.50 μm) were synthesized via the Stöber method, while irregular aluminosilicate glass was used in coarse (D50 = 3.71 μm) and fine (D50 = 1.98 μm) fractions. Three composite groups were formulated: Group 1 (72 wt.% filler with 0–30% SiO₂), Group 2 (maximum filler loading 76–80 wt.% with 10–30% SiO₂), and Group 3 (74.5 wt.% filler with varying coarse/fine glass ratios). Flexural strength, flexural modulus, Vickers microhardness, depth of cure, water sorption, and solubility were evaluated according to ISO 4049:2019. Results: Incorporation of spherical SiO₂ nanoparticles significantly reduced composite viscosity, enabling maximum filler loading to increase from 72 to 80 wt.%. All composites exceeded ISO requirements for flexural strength (80.54–118.11 MPa), depth of cure (3.01–5.65 mm), water sorption (14.61–22.87 μg/mm³), and solubility (1.20–5.90 μg/mm³). The highest flexural strength (118.11 ± 10.54 MPa) and modulus (9.26 ± 1.12 GPa) were achieved at 78 wt.% filler loading. Bimodal glass systems (50/50 ratio) demonstrated optimal mechanical properties, while higher fine fractions reduced strength. Conclusions: Spherical SiO₂ nanoparticles effectively reduce viscosity and enable higher filler loading. The optimal balance between filler loading, particle shape, and size distribution should be tailored to clinical requirements, with high-strength formulations suited for posterior restorations and bimodal formulations for universal applications.

1. Introduction

The widespread use of polymer-based dental resin composites is attributed to their aesthetic appearance, good processing characteristics, and improved mechanical properties [1]. However, their life cycle in the aggressive oral environment is limited by such factors as high polymerization shrinkage, low wear resistance, a propensity for bulk/marginal fractures, and high-water sorption [1,2]. Furthermore, concerns exist regarding potential health risks associated with the release of components, such as Bisphenol-A, from traditional monomer systems like Bis-GMA [1,2].

To overcome these limitations, significant attention is being devoted to the development of new inorganic fillers and the optimization of their characteristics. Fillers are considered the most critical component of dental composites, as they are responsible for the improvement of mechanical and physical properties, such as strength, modulus of elasticity, hardness, and fracture toughness, while also influencing polymerization shrinkage, aesthetics, and radiopacity [3,4]. Today, aluminosilicate glass fillers are one of the most widely used, well-established, and widely used class of fillers for dental composites. Although new modern fillers are being introduced, aluminosilicate glasses containing BaO/SrO remain a “workhorse” in mid-range and economy hybrid composites due to their favorable physical and mechanical properties and radiopacity. They provide reliable and predictable clinical behavior, which is confirmed by long-term observations.

Spherical SiO₂ particles are of particular interest as a filler owing to their superior optical properties, high mechanical strength, low abrasiveness, and ability to reduce volumetric shrinkage [3,5,6,7]. The use of SiO₂ nano- and microspheres has been shown to improve filler dispersion, increase its content level, and ensure good mechanical properties of the resulting composite [8,9,10].

The size and shape of filler particles are considered critical parameters which substantially influence various properties of dental composites, including the maximum filler content, mechanical characteristics, and polymerization shrinkage [11]. Research indicates that optical properties and the depth of cure are improved as the filler size is reduced [4,12]. Furthermore, a reduction in filler particle size generally leads to an increase in wear resistance [13]. Regarding mechanical properties, at identical filler content levels, for instance 60% by weight, different filler particle sizes can yield comparable values for flexural strength and modulus of elasticity [11]. However, if composites are loaded with the maximum possible amount of filler, larger particle sizes can provide superior mechanical properties, which is explained by their ability to achieve higher loading volumes [1]. The most frequently encountered filler particle morphologies are spheroidal and irregular, although several others have also been investigated [14,15]. Spherical filler particles are synthesized by the sol–gel method, which results in a very narrow particle size distribution [16]. It has been demonstrated that spheroidal fillers can sustain a higher loading, leading to superior mechanical properties [8]. Moreover, although spherical fillers demonstrate wear characteristics similar to those of irregular fillers, they maintain superior gloss and smoothness throughout the process [10].

The filler loading level is also a key factor that determines the properties of composites, since mechanical reinforcement is generally increased with a higher content of inorganic filler [11,17]. This strengthening is particularly noticeable when the fillers form a percolated network structure at high concentrations, thereby facilitating the effective distribution and dissipation of stress within the composite [18,19]. A higher filler loading, achieved through its optimal (bimodal and trimodal) distribution, can significantly improve compressive strength and the modulus of elasticity. [20,21,22,23]. However, this improvement is dependent upon the maintenance of quality adhesion between the filler and the matrix and the absence of particle agglomeration. It has also been well established that increasing filler content reduces volumetric polymerization shrinkage [1,24,25], which is a critical factor affecting marginal adaptation, secondary caries risk, and long-term clinical success of resin composite restorations.

It is known that characteristics such as size and shape have a significant impact on the maximum filling level achievable in a composite: for example, spherical fillers can provide denser packing and higher loading levels compared to irregularly shaped particles. However, a systematic understanding of the combined influence of particle size and morphology on percolated network formation and the development of key physical and mechanical properties, including strength, stiffness, and water sorption, at high filling levels remains limited in the open literature [11]. Furthermore, the distinct roles of filler shape and size in maximum loading capacity have not been clearly disentangled, as most investigations have predominantly centered on size-related effects.

Given the complexity of the interrelationship between particle size, their content level, and the properties, the objective of the present work is to investigate the dependence of the flexural strength, flexural modulus, Vickers microhardness, depth of cure and water sorption and solubility of dental composites on (a) the size and shape of the particles and (b) the magnitude of their loading within the composite volume. A systematic study of these parameters is expected to provide a deeper understanding of the optimal filler design for the development of dental composites with enhanced characteristics and extended service life.

2. Materials and Methods

2.1. Reagents and Fillers

The starting materials used for the polymer matrix were Bisphenol A glycidyl methacrylate (BisGMA), Urethane dimethacrylate (UDMA), Triethylene glycol dimethacrylate (TEGDMA), Camphorquinone (CQ), and Ethyl 4-(dimethylamino) benzoate (4-EDMAB), all purchased from Alfa Chemical Co., Ltd. (Zhengzhou, China). The inorganic fillers consisted of two components. The first component was amorphous aluminosilicate glass micropowder particles modified with barium and strontium oxides (with a composition of 15 wt.% Al₂O₃, 8 wt.% BaO, and 2 wt.% SrO, also manufactured by Alfa Chemical Co., Ltd. (Zhengzhou, China). The second component was colloidal SiO₂ nanopowder, which was synthesized via the sol–gel method.

To obtain a smaller fraction of the aluminosilicate glass particles, the as-received powder was initially subjected to milling for 24 h in distilled water using an ML1 ball mill (Promstroymash Ltd., Moscow, Russia) with zirconia balls (d = 3 mm). The resultant material was subsequently dried in a VO400 vacuum oven (Memmert GmbH, Büchenbach, Germany) and sieved using a VP-30 sieve analyzer (Vibrotekhnik Ltd., Saint Petersburg, Russia) on sieves with a mesh size of 63 μm.

The following reagents were used for the sol–gel synthesis of the nanometer-sized silica particles: Ammonium hydroxide (NH₃, 25%, SigmaTek LLC, Khimki, Russia), Tetraethyl orthosilicate (TEOS) (98.5%, EKOS-1 JSC, Moscow, Russia), Isopropyl alcohol (99.7%, Selkor, Saint Petersburg, Russia), 3-Methacryloxypropyltrimethoxysilane (γ-MPS) (Beijing Chemical Works, Beijng, China), Cyclohexane (99.7%, Component-Reactive, Moscow, Russia), and n-Propylamine (99%, Sisco Research Laboratories, Mumbai, India). All reagents were of Analytical Reagent (AR) grade.

2.2. Synthesis of SiO₂ Nanospheres by the Sol–Gel Method

Spherical SiO₂ particles were synthesized via the Stöber method [26], involving the hydrolysis and condensation of Tetraethyl orthosilicate (TEOS) in an alcoholic solvent in the presence of ammonium hydroxide. The reactions were conducted with a TEOS concentration of [TEOS] = 0.75 M and under low concentrations of ammonia in isopropanol, specifically using molar ratios of [NH₃] = 0.46[TEOS] and [H₂O] = 4.7[TEOS].

First, predetermined amounts of deionized water, NH₃, and isopropanol were mixed for 10 min. Subsequently, a known volume of TEOS was added dropwise to this solution at a temperature of 20 °C under continuous stirring using a magnetic stirrer for 5 h. Following the reaction, the silica particles were separated from the suspension by centrifugation and subsequently washed four times with water and ethanol. The purified particles were then dried in an oven at 150 °C for 5 h. All concentrations were calculated based on the final concentration in the reaction mixture. Prior to use, all reaction vessels were cleaned with distilled water using sonication at a frequency of 80 kHz. The vessels were sealed with a plastic membrane to prevent the volatilization of NH₃.

2.3. Silanization

The amount of silane required for the silanization process was set at 10 wt.% relative to the mass of the filler powder samples.

The sufficiency of this selected silane quantity for the complete coverage of the silica and aluminosilicate glass used in this study was verified using the following relationship [27]:

X = (A/ω)·f,

(1)

where X is the required mass of silane (g), A is the specific surface area of the powder being coated (m²/g), f is the mass of the filler (g), and ω is the surface area that 1 g of the silane can cover (2525 m²/g, based on the premise that one silane molecule covers 1.11 nm² [28]).

The silanization procedure involved several steps [29,30]. First, 5 g of the filler powder were placed into a 250 mL round-bottom flask. Next, 100 mL of cyclohexane, 0.1 g of N-propylamine, and the calculated amount of silane were added to the flask. The mixture was stirred at room temperature for 30 min, followed by stirring at 60 °C for an additional 30 min under atmospheric pressure. The flask was then placed in a rotary evaporator and rotated under a moderate vacuum (≈2.7 kPa) and a temperature of 60 ± 5 °C until the solvent and all volatile byproducts were completely removed. Finally, the resulting powder was dried in a vacuum drying oven at 80 °C for 18 h.

2.4. Composite Preparation

The polymer matrix was formulated using: 50 wt.% BisGMA, 25 wt.% UDMA, and 25 wt.% TEGDMA. Camphorquinone (0.2 wt.%) and Ethyl 4-(dimethylamino) benzoate (0.8 wt.%) were incorporated as the photoinitiator and accelerator, respectively.

The silanized silica nanosphere powders were mixed with the aluminosilicate glass and the pre-mixed polymer matrix. Initial blending was performed in a SK-300SII Kakuhunter high-speed mixer (Shashin Kagaku, Kyoto, Japan) at a maximum speed of 2000 rpm. Final homogenization of the mixture was achieved using a three-roll mill (MT-SG50Z, Changsha Mitr Instrument Equipment Co., Ltd., Changsha, China) until a uniform composite material was obtained [31,32].

For testing purposes, the composite mixtures were packed into silicone rubber molds, using cylindrical molds for the depth of cure tests and rectangular molds for the three-point bending tests. The composites were light-cured using a Turbodent dental curing lamp (Dental ABC, Shanghai, China), employing light irradiation (420–480 nm) for 2 min on each side at a minimum of three distinct points per specimen.

2.5. Flexural Strength and Flexural Modulus

Three-point bending tests were conducted on a Instron ElectroPuls E10000 universal testing machine (Instron, Norwood, MA, USA) in accordance with ISO 4049:2019 [33] standards. For each composite material, five specimens were prepared with dimensions of 25 mm × 2 mm × 2 mm. The specimens were loaded to failure at a crosshead speed of 0.75 mm/min.

Flexural strength (FS) and flexural modulus (FM) were calculated according to the following equations:

$$\mathrm{F}\mathrm{S}={\displaystyle \frac{3·\mathrm{F}·\mathrm{l}}{2·\mathrm{b}·{\mathrm{h}}^2}}$$

(2)

$$\mathrm{F}\mathrm{M}={\displaystyle \frac{{\mathrm{F}}_1{·\mathrm{l}}^3}{4·\mathrm{b}·{\mathrm{d}·\mathrm{h}}^3}}$$

(3)

where F is the maximum load (N), F₁ represents the load exerted on the specimen (N), l is the distance between the supports (mm), b is the width (mm), h is the height of the sample (mm) and d is the deflection generated by the load F₁ (mm).

2.6. Vickers Microhardness

Vickers microhardness tests were performed using a microhardness tester (Qness, Salzburg, Austria) with a load of 9.8 N applied for 10 s. For these tests, disc-shaped specimens of each material were prepared with a diameter of 6 mm and a height of 3 mm.

The Vickers hardness number (HV) was automatically calculated by the instrument software using the formula:

$$\mathrm{H}\mathrm{V}=1.854\times {\displaystyle \frac{\mathrm{F}}{{\mathrm{d}}^2}}$$

(4)

where F is the applied load (N) and d is the mean diagonal length of the indentation (mm). The reported hardness value for each composite represents the mean of five specimens, with five indentations per specimen (n = 25 total measurements per composite formulation).

2.7. Depth of Cure

The resin-based composite was packed into a mold (4 mm diameter × 8 mm height) and cured for 10 s from one end. The uncured material was subsequently removed using a spatula. The thickness of the hardened specimen was then measured at five points using a CD-6 CSX caliper (Mitutoyo Corp., Kawasaki, Japan) with an accuracy of up to 0.01 mm. This entire procedure was carried out in accordance with ISO 4049:2019 [33].

2.8. Water Sorption and Solubility

The determination of water sorption and solubility was performed according to the methodology described in the ISO 4049:2019 [33]. For these tests, disc-shaped specimens were prepared with a diameter of 15 mm and a thickness of 1 mm. Five specimens were prepared for each group of materials. After polymerization, the specimens were dried in a vacuum desiccator at a temperature of 37 ± 2 °C or in an air-circulating oven at 37 °C until a constant mass was achieved. This mass was recorded as the initial mass (M₁) with an accuracy of up to 0.01 mg using an AND GR-300 laboratory balance (A&D Company Ltd., Japan). The initial dimensions (thickness and diameter) were also measured to calculate the initial specimen volume (V₀).

The dried specimens were then immersed in vials containing the storage medium, which was 10 mL of distilled water. The specimens were stored at a temperature of 37 ± 2 °C for 30 days. At the end of the storage period, the specimens were retrieved, excess surface water was removed, and they were air-dried for 20 s before being weighed. This value was recorded as the mass after saturation (M₂). The specimens were then returned to the desiccator and dried at 37 °C until a final constant mass (M₃) was reached.

The Water Sorption (WS, μg/mm³) was calculated as follows:

$$\mathrm{W}\mathrm{S}={\displaystyle \frac{{\mathrm{M}}_2-{\mathrm{M}}_3}{{\mathrm{V}}_0}}.$$

(5)

The Solubility (SL, μg/mm³), was calculated as:

$$\mathrm{S}\mathrm{L}={\displaystyle \frac{{\mathrm{M}}_1-{\mathrm{M}}_3}{{\mathrm{V}}_0}}.$$

(6)

2.9. Composites and Fillers Characterization

The size, morphology, and elemental composition of the synthesized particles were determined using a Vega 3 LMH Scanning Electron Microscope (SEM) (Tescan, Brno, Czech Republic), which was equipped with an Energy Dispersive X-ray Spectroscopy (EDS) module (Oxford Instruments, Abingdon, UK). Filler samples were prepared by sprinkling powder onto double-sided adhesive carbon conductive tape, which was mounted on a holder. As all specimens were electrically non-conductive, they were initially coated with a thin layer of gold using a Quorum Q150ES sputter coater (Quorum Technologies, East Sussex, UK).

Particle size distribution of the powders was measured using laser diffraction Analysette 22 NanoTec plus (Fritsch GmbH, Amberg, Germany).

The phase composition of the fillers was determined using an Empyrean X-ray diffractometer (PANalytical, Almelo, The Netherlands) equipped with a Cu-Kα X-ray source, wavelength 1.5405981 Å, with a shooting angle 2θ from 10° to 70°.

The specific surface area of the particles was determined by the Brunauer–Emmett–Teller (BET) method using an ASAP 2020 MP surface area and porosimetry analyzer (Micromeritics Instrument Corp., Norcross, GA, USA).

2.10. Statistical Analysis

One-way analysis of variance (ANOVA) was conducted to determine statistically significant differences among the composites within each group, using Minitab 21.4.1 software. When significant differences were detected (p < 0.05), Tukey’s test was applied for pairwise comparisons. The sample size of five specimens per test group was determined based on ISO 4049:2019 [33] requirements.

3. Results

3.1. Characterization of Fillers

SEM images at various magnifications (Figure 1a,b,d) show that the SiO₂ powder has a regular spherical morphology. The resulting EDS spectra (Figure 1c) and elemental mapping (Figure 1e,f) confirmed the presence of only silicon (Si) and oxygen (O) in the synthesized powder, in a ratio of 30.9:69.1 wt.%. The presence of characteristic gold lines in the spectrum is due to the preliminary deposition of a conductive coating. The Au peak was excluded from all EDS quantifications, and the elemental composition was recalculated based on the normalization of the remaining elements to 100%.

Figure 1. SEM images of synthesized SiO₂ particles at magnifications 20,000× (a), 60,000× (b) and 4000 (d); EDS analysis (c) and elemental distribution maps for Si, O (e,f).

Figure 2 and Figure 3 show SEM images of the raw aluminosilicate glass powder and the ball-milled aluminosilicate glass powder used in this study, along with the results of EDS analysis and element distribution maps.

Figure 2. SEM images of aluminosilicate glass at magnifications 6000× (a) and 20,000× (b); EDS analysis (c) and elemental distribution maps for Si, O, Al, Sr, Ba, Ca, Mg (d–j).

Figure 3. SEM images of aluminosilicate glass (ball-milled) at magnifications 6000× (a) and 20,000× (b); EDS analysis (c) and elemental distribution maps for Si, O, Al, Sr, Ba, Ca, Mg and Zr (d–k).

Figure 2a,b show that the raw aluminosilicate glass powder consists of large, irregularly shaped particles. EDS analysis (Figure 2c) revealed the presence of the following elements in the powder: Si (22.9 wt%), O (61.1 wt%), Al (8.0 wt%), Ca (3.5 wt%), Ba (1.7 wt%), Sr (1.5 wt%), and Mg (1.2 wt%). Figure 2d–j show the distribution maps of all detected elements. The characteristic Lα line of strontium (Sr) overlaps the Kα line of silicon (Si), so the strontium distribution on the map mirrors that of silicon. It is worth noting that the manufacturer did not indicate the presence of Mg and Ca in the original powder.

Figure 3a,b show that after grinding in a ball mill with zirconium oxide balls, the aluminosilicate glass powder has decreased in size. The ground glass fraction exhibits a similar elemental composition to the raw powder, but a small content of 0.7 wt.% zirconium (Zr) appears in the spectrum (Figure 3c), most likely due to the milling of ZrO₂ from the grinding ball material. Figure 3d–k show the distribution maps of all detected elements, confirming the presence of Zr.

Figure 4 shows the X-ray diffraction pattern obtained for the synthesized SiO₂ and the initial coarse-phase aluminosilicate glass. Analysis of the diffraction patterns did not reveal the presence of sharp peaks characteristic of crystalline structures. Both samples demonstrate broad diffraction maxima typical of amorphous materials, centered near 2θ = 25°. The absence of intense crystalline reflections clearly indicates the glassy state of the studied materials. The amorphous nature of the synthesized SiO₂ sample is confirmed by the absence of sharp peaks and the presence of wide humps. The broad diffraction peak centered near 25° (black line in Figure 4) corresponds to the standard X-ray diffraction pattern of amorphous silica (JCPDS No. 29-0085). This correspondence serves as objective evidence of the successful synthesis of SiO₂ by the Stöber method.

Figure 4. X-ray diffraction spectra of fillers.

Table 1 shows the measured particle size distribution parameters for three types of fillers.

Table 1. Particle size distribution parameters for the three filler types.

Filler Type	D10 (μm)	D50 (μm)	D90 (μm)	Span
SiO2	0.38 ± 0.09	0.50 ± 0.08	0.62 ± 0.12	0.48
Aluminosilicate glass (raw)	1.18 ± 0.32	3.71± 0.58	14.06± 1.11	3.49
Aluminosilicate glass (ball-milled)	0.78 ± 0.25	1.98 ± 0.41	3.56 ± 0.64	1.80

The narrow size distribution of synthesized SiO₂ (span = 0.48) confirms the effectiveness of the Stöber method in producing monodisperse spherical particles. The narrower distribution of the fine glass fraction (span = 1.80) compared to the coarse fraction (span = 3.49) is explained by the ball milling process, which breaks the coarse fraction into small pieces.

The measured specific surface area for the raw coarse aluminosilicate glass powder was determined to be 2.603 ± 0.053 m²/g, while that of the fine fraction was 4.325 ± 0.061 m²/g, and that of the SiO₂ was 8.361 ± 0.099 m²/g.

All presented particles were subsequently utilized for the creation of dental composites using the same polymer matrix.

3.2. Characterization of the Obtained Dental Composites

3.2.1. The Composition and Loading Level of the Composites

The filler content in modern commercial composites varies from 70 to 80% by weight. This study examined the following three groups of composites containing different filler concentrations and differing in composition, shape, and size.

Group 1: Composites with a filler concentration of 72 wt.% and varying content of spherical silicon dioxide.

A series of experiments was initially conducted with the coarse fraction of aluminosilicate glass. The value of 72 wt.% was empirically determined to be the maximum loading level for the coarse aluminosilicate glass particles in the selected polymer matrix, at which the composite maintained cohesive paste consistency and could be effectively handled. Consequently, to further study the influence of shape and size, the first group of dental composites was prepared with proportions of 72 wt.% filler and 28 wt.% polymer matrix. All composites in this group were filled with filler at an identical mass content. Spherical SiO₂ nanoparticles were added to the coarse aluminosilicate glass fraction at varying mass contents, while the total filler concentration was maintained. The resulting consistency of the compositions after mixing in a high-speed mixer and a three-roll mill was visually inspected. The samples of dental composites from the first group were designated as a/bSiO₂, where ‘a’ is the mass content of the filler in the composite, and ‘b’ is the percentage content of silica within the filler. The filler compositions for the first group are presented in Table 2.

Table 2. Filler content in the composites of the first group, wt.%.

Composite	Filler	Polymer Matrix	SiO2	Aluminosilicate Glass
72/0SiO2	72	28	0	100%
72/1SiO2	72	28	1	99
72/10SiO2	72	28	10	90
72/20SiO2	72	28	20	80
72/30SiO2	72	28	30	70

During the preparation of the composites, it was observed that the viscosity of the dental composite was reduced even with the addition of 1% spherical SiO₂ nanoparticles to the coarse aluminosilicate glass fraction. A continued trend of viscosity reduction was noted with a further increase in the SiO₂ content. However, when more than 30% was added, the composite lost its properties as a thick paste, becoming highly stringy, sticky, and unable to retain its shape. Consequently, it was decided to limit the SiO₂ concentration to this level for the characterization of the resulting composites.

Group 2: Composites with Maximum Filler Loading.

Since it was observed that the addition of 10–30% SiO₂ substantially reduced the viscosity of the composite based on the coarse aluminosilicate glass fraction, a decision was made to create a second group of composites with a selected concentration of silicon dioxide, but with the maximum possible mass fraction of filler. To evaluate the maximum loading, the inorganic filler was added to the polymer matrix in 1 wt.% increments, followed by mixing in a high-speed mixer. When the composite lost its coherence and disintegrated, it was considered to be overloaded; the value prior to the last addition was accepted as the maximum filler loading. The resulting dental composites of Group 2 with maximum loading are presented in Table 3.

Table 3. Filler content in composites of the second group, wt.%.

Composite	Filler	Polymer Matrix	SiO2	Coarse Aluminosilicate Glass
76/10SiO2	76	24	10	90
78/20SiO2	78	22	20	80
80/30SiO2	80	20	30	70

Addition of 10 wt.% spherical SiO₂ enabled a maximum filler concentration of 76 wt.%. Furthermore, when the SiO₂ content was increased to 20 and 30 wt.%, the maximum filler loading was elevated to 78 and 80 wt.%, respectively. All resulting composites were characterized by an extremely dense structure.

Group 3: composites with a filler content of 74.5 wt.% of different shapes and sizes.

During the creation of the composites with maximum loading, it was noted that the composite with 10% SiO₂ content exhibited more thixotropic behavior compared to the materials with higher content. However, its maximum content made it suboptimal for clinical handling. The viscosity of dental materials is critical as it determines the ease of clinical application of composites [32,34,35,36]. In this context, it was decided to reduce the total filler content to achieve more suitable handling properties. Similarly to Group 2, a filler content of 74.5 wt.% was empirically determined. This composite was characterized by high plasticity, combined with thixotropic behavior and low adhesion to instruments.

Subsequently, to investigate how the properties of the selected composite would be altered by the addition of the two fractions of aluminosilicate glass (coarse and fine), a third composite group was fabricated. The samples of dental composites from the third group were designated as a/bSiO₂/c, where ‘a’ is the mass content of the filler in the composite, ‘b’ is the percentage content of SiO₂ within the filler, and ‘c’ is the content of the fine glass fraction. The list and composition of the fabricated composites in the third group are presented in Table 4.

Table 4. Filler content in composites of the third group, wt.%.

Composite	Filler	Polymer Matrix	SiO2	Coarse Aluminosilicate Glass	Fine Aluminosilicate Glass
74.5/10SiO2	74.5	25.5	10	100	0
74.5/10SiO2/50	74.5	25.5	10	50	50
74.5/10SiO2/60	74.5	25.5	10	40	60
74.5/10SiO2/70	74.5	25.5	10	30	70
74.5/10SiO2/100	74.5	25.5	10	0	100

3.2.2. Microstructure of the Composites

The characteristic fracture surfaces of light-cured composites at 6000× (scale bar 10 μm) and 20,000× (scale bar 2 μm) magnifications are shown in Figure 5. The first group: 72/0SiO₂ (Figure 5a) and 72/10SiO₂ (Figure 5b), the second group: 80/30SiO₂ (Figure 5d) and the third group: 74.510SiO₂50 (Figure 5c).

Figure 5. Characteristic microstructure of fracture surfaces of composites with different fillers content at 6000× and 20,000× magnifications: (a) 72/0SiO₂; (b) 72/10SiO₂; (c) 74.5/10SiO₂/50; (d) 80/30SiO₂.

For the composite 72/0SiO₂ from the first group (Figure 5a) coarse glass fraction, it can be observed that the glass particles are well distributed within the matrix. However, the potential exists to further fill the composite with a finer phase in the polymer only matrix areas. Following the addition of 10 wt.% of spherical SiO₂ nanoparticles (Figure 5b) to this composite, the microstructure has become denser and the nanoparticles filled the gaps between the coarse glass fraction and were uniformly dispersed throughout the entire volume of the polymer matrix. Nevertheless, such a microstructure is still capable of a further increase in filler content putting them in interstitial space, which is supported by composite fluidity during the mixing stage.

In the composite 74.5/10SiO₂/50 (Figure 5c), a uniform distribution of the coarse and fine glass fraction particles and the SiO₂ nanoparticles is observed, which resulted in a denser packing of the composite.

Volume fraction of free polymer matrix is clearly in negative correlation with filler content. For the 80/30SiO₂ composite (Figure 5d), the low proportion of resin present in the interparticle spaces is evident, which prevented a further increase in particle concentration. The described differences in the microstructures are expected to influence the physical, mechanical and adsorption properties of the investigated composites.

3.3. Properties of Experimental Dental Composites

The test results of flexural strength, flexural modulus, depth of cure, and Vickers microhardness obtained for all groups of the composites studied are presented in Table 5, Table 6 and Table 7, respectively.

Table 5. Flexural strength, flexural modulus, depth of cure, and Vickers microhardness of composites Group 1 (mean ± standard deviation).

Group	Composite	Flexural Strength, MPa (p = 0.151)	Flexural Modulus, GPa (p = 0.021)	Hardness, HV (p < 0.001)	Depth of Cure, mm (p < 0.001)
1	72/0SiO2	86.19 ± 5.91	5.91 ± 0.54 AB	56.67 ± 1.71 A	5.65 ± 0.18 A
	72/1SiO2	84.24 ± 4.70	5.72 ± 0.95 AB	55.10 ± 1.88 AB	5.34 ± 0.22 B
	72/10SiO2	88.49 ± 5.14	6.25 ± 0.36 A	58.26 ± 1.68 A	4.32 ± 0.07 C
	72/20SiO2	82.47 ± 4.51	5.24 ± 0.80 AB	49.58 ± 4.09 B	4.15 ± 0.13 C,D
	72/30SiO2	80.54 ± 4.38	4.82 ± 0.35 B	43.48 ± 3.51 C	3.85 ± 0.16 D

Different superscript letters indicate statistical differences within a column (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between groups within a column for that parameter.

Table 6. Flexural strength, flexural modulus, depth of cure, and Vickers microhardness of composites Group 2 (mean ± standard deviation).

Group	Composite	Flexural Strength, MPa (p = 0.187)	Flexural Modulus, GPa (p = 0.028)	Hardness, HV (p = 0.490)	Depth of Cure, mm (p < 0.001)
2	76/10SiO2	104.04 ± 9.84	7.31 ± 0.85 b	66.42 ± 4.28	3.93 ± 0.07 a
	78/20SiO2	118.11 ± 10.54	9.26 ± 1.12 a	67.23 ± 3.27	3.79 ± 0.09 a
	80/30SiO2	110.42 ± 13.28	7.58 ± 0.94 ab	69.13 ± 2.63	3.01 ± 0.15 b

Different superscript letters indicate statistical differences within a column (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between groups within a column for that parameter.

Table 7. Flexural strength, flexural modulus, depth of cure, and Vickers microhardness of composites Group 3 (mean ± standard deviation).

Group	Composite	Flexural Strength, MPa (p = 0.047)	Flexural Modulus, GPa (p < 0.001)	Hardness, HV (p = 0.013)	Depth of Cure, mm (p < 0.001)
3	74.5/10SiO2	91.19 ± 3.59 αβ	6.84 ± 0.32 αβ	61.54 ± 2.52 αβ	4.23 ± 0.08 α
	74.5/10SiO2/50	94.56 ± 5.78 α	6.92 ± 0.28 α	63.46 ± 1.32 α	4.09 ± 0.13 αβ
	74.5/10SiO2/60	83.56 ± 3.31 β	5.26 ± 0.74 γ	55.22 ± 4.47 β	3.84 ± 0.12 βγ
	74.5/10SiO2/70	86.76 ± 5.54 αβ	5.91 ± 0.35 βγ	59.38 ± 4.26 αβ	3.72 ± 0.18 γ
	74.5/10SiO2/100	88.47 ± 7.25 αβ	6.43 ± 0.53 αβ	60.88 ± 2.21 αβ	3.61 ± 0.17 γ

Different superscript letters indicate statistical differences within a column (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between groups within a column for that parameter.

All the obtained composites met the requirements of the ISO 4049:2019 [33] standard for flexural strength (>80 MPa) and depth of cure (>1.5 mm).

3.3.1. Flexural Strength and Flexural Modulus

A review revealed that the typical flexural strength and flexural modulus of conven-tional dental composites filled with hybrid fillers are in the range from 70 MPa to 130 MPa and from 5 GPa to 10 GPa correspondingly [37,38,39,40,41]. The test results obtained for all groups of the composites are presented in Figure 6 and Table 5, Table 6 and Table 7.

Figure 6. The flexural strength values for groups 1 (A), 2 (B), and 3 (C). Different superscript letters indicate statistical differences between columns in groups (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between columns for that parameter.

The flexural strength of the composites demonstrated a complex dependence on filler composition and loading. In Group 1 (Table 5), the addition of spherical SiO₂ nanoparticles at concentrations of 1 to 30 wt.% within the filler did not lead to statistically significant differences in strength (p = 0.151), with values ranging from 80.54 ± 4.38 MPa to 88.49 ± 5.14 MPa.

Maximizing the total filler content in Group 2 by utilizing SiO₂ yielded a substantial improvement in mechanical performance (Table 6). All composites in this group exceeded 100 MPa, with the composition 78/20SiO₂ (78 wt.% total filler, 20 wt.% SiO₂ within filler) achieving the highest mean strength of 118.11 ± 10.54 MPa, although the differences within the group were not statistically significant (p = 0.187).

In Group 3, flexural strength was highly sensitive to the ratio of coarse to fine aluminosilicate glass fractions (Table 7). A statistically significant optimum (p = 0.047) was found for the 74.5/10SiO₂ 50 composite (50 wt.% fine fraction), which exhibited the highest strength of 94.56 ± 5.78 MPa. A deviation from this optimum, specifically an increase in the fine fraction to 60 wt.%, resulted in a significant drop in strength to 83.56 ± 3.31 MPa. A further increase to 70 and 100 wt.% fines led to a partial recovery to 86.76 ± 5.54 MPa and 88.47 ± 7.25 MPa, respectively, but these values remained statistically lower than the optimum.

The flexural modulus for traditional resin-based dental composites containing hybrid inorganic fillers, according to the literature, ranges from 5 to 10 GPa, while for the modern commercial composites this value is typically around 10 GPa. The obtained values of the modulus of elasticity for all composite groups are presented in Figure 7 and Table 5, Table 6 and Table 7, respectively.

Figure 7. Flexural modulus values for groups 1 (A), 2 (B), and 3 (C). Different superscript letters indicate statistical differences between columns in groups (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between columns for that parameter.

The flexural modulus showed clear and statistically significant trends across groups. In Group 1 (Table 5), the modulus was significantly affected by SiO₂ content (p = 0.021). The highest value of 6.25 ± 0.36 GPa was recorded for the composite with 10 wt.% SiO₂ (72/10SiO₂), while the modulus decreased to 4.82 ± 0.35 GPa at 30 wt.% SiO_2.

In Group 2 (Table 6), the modulus increased with total filler loading, peaking at 9.26 ± 1.12 GPa for the 78/20SiO₂ composite, which was significantly higher (p = 0.028) than that of the 76/10SiO₂ composite (7.31 ± 0.85 GPa).

For Group 3 (Table 7), the modulus was strongly dependent on the fine glass fraction content (p < 0.001). The highest modulus of 6.92 ± 0.28 GPa was again observed for the optimal 74.5/10SiO₂/50 blend. Increasing the fine fraction to 60 wt.% caused a significant decrease to the lowest group value of 5.26 ± 0.74 GPa. Subsequent increases to 70 and 100 wt.% fines resulted in a gradual recovery to 5.91 ± 0.35 GPa and 6.43 ± 0.53 GPa, respectively.

3.3.2. Vickers Hardness

In particular, the hardness of light-curing dental composites depends greatly on the fraction (volume fraction) and size of the filler particles. Vickers microhardness of various dental composites ranges from 48 to 125 HV. The obtained hardness values of the studied composites are within the specified limits (Figure 8).

Figure 8. Vickers hardness testing results for the 1 (A), 2 (B) and 3 (C) group. Different superscript letters indicate statistical differences between columns in groups (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between columns for that parameter.

Microhardness was significantly influenced by filler composition. In Group 1 (Table 5), it decreased progressively and significantly (p < 0.001) with increasing SiO₂ content, from 56.67 ± 1.71 HV (0 wt.% SiO₂) to 43.48 ± 3.51 HV (30 wt.% SiO₂).

No statistically significant differences in hardness were found within Group 2 (p = 0.490) (Table 6), with all composites showing high values between 66.42 ± 4.28 HV and 69.13 ± 2.63 HV.

In Group 3 (Table 7), a significant effect of the fine glass fraction was observed (p = 0.013). The composite with the optimal 50/50 blend (74.5/10SiO₂/50) showed the highest hardness of 63.46 ± 1.32 HV. The composition with 60 wt.% fine fraction exhibited significantly lower hardness (55.22 ± 4.47 HV), while other formulations showed intermediate values.

3.3.3. Depth of Cure

All resulting composites demonstrated values exceeding the threshold specified by the standard [33]. The depth of cure values for each composite group are presented in Figure 9.

Figure 9. Depth of cure values for the 1st (A), 2nd (B), and 3rd (C) composite groups. Different superscript letters indicate statistical differences between columns in groups (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between columns for that parameter.

The depth of cure was the parameter most consistently and significantly reduced by the incorporation of spherical SiO₂ nanoparticles and increased filler loading across all groups (p < 0.001 for each group). In Group 1 (Table 5), it decreased from 5.65 ± 0.18 mm for the SiO₂-free control to 3.85 ± 0.16 mm for the composite with 30 wt.% SiO_2. In Group 2 (Table 6), the highest-loaded composite 80/30SiO₂ had a significantly lower depth of cure (3.01 ± 0.15 mm) compared to the 76/10 SiO₂ and 78/20 SiO₂ composites (~3.8–3.9 mm). Similarly, in Group 3 (Table 7), a progressive and significant decrease was recorded with an increasing proportion of the fine glass fraction, from 4.23 ± 0.08 mm (0 wt.% fines) to 3.61 ± 0.17 mm (100 wt.% fines).

3.3.4. Water Sorption and Solubility

To evaluate the hydrolytic stability, WS and SL were measured. The results for all experimental groups are presented in Table 8, Table 9 and Table 10, respectively.

Table 8. Water sorption (WS) and solubility (SL) values of composites Group 1 (mean ± standard deviation, n = 5).

Group	Composite	WS (μg/mm3) (p < 0.001)	SL (μg/mm3) (p < 0.001)
1.	72/0SiO2	14.61 ± 0.85 b	1.77 ± 0.17 d
	72/1SiO2	14.75 ± 1.31 b	1.83 ± 0.13 d
	72/10SiO2	16.07 ± 1.72 b	3.03 ± 0.21 c
	72/20SiO2	16.65 ± 1.54 b	3.77 ± 0.22 b
	72/30SiO2	22.87 ± 1.41 a	5.90 ± 0.25 a

Different superscript letters indicate statistical differences within a column (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between groups within a column for that parameter.

Table 9. Water sorption (WS) and solubility (SL) values of composites group 2 (mean ± standard deviation, n = 5).

Group	Composite	WS (μg/mm3) (p = 0.062)	SL (μg/mm3) (p = 0.343)
2.	76/10SiO2	20.53 ± 0.45	4.68 ± 0.22
	78/20SiO2	21.29 ± 0.72	4.79 ± 0.15
	80/30SiO2	21.86 ± 1.08	4.56 ± 0.32

Table 10. Water sorption (WS) and solubility (SL) values of composites group 3 (mean ± standard deviation, n = 5).

Group	Composite	WS (μg/mm3) (p = 0.120)	SL (μg/mm3) (p < 0.001)
3.	74.5/10SiO2	15.66 ± 0.56	3.75 ± 0.12 A
	74.5/10SiO2/50	15.37 ± 0.41	1.30 ± 0.23 B
	74.5/10SiO2/60	16.01 ± 0.74	1.33 ± 0.21 B
	74.5/10SiO2/70	15.95 ± 0.49	1.2 ± 0.13 B
	74.5/10SiO2/100	16.33 ± 0.57	1.22 ± 0.15 B

Different superscript letters indicate statistical differences within a column (p < 0.05, Tukey’s test). Absence of letters indicates no significant differences between groups within a column for that parameter.

Analysis of the obtained data revealed that both water sorption and solubility for all experimental specimens were maintained below the established regulatory limits. According to ISO 4049:2019 [33], the maximum permissible value for WS is 40 μg/mm³, and for SL is 7.5 μg/mm³.

Water Sorption was significantly influenced by the composition of the filler. In Group 1, WS (Table 8) increased progressively with higher content of spherical SiO₂ nanoparticles (p < 0.001). The composite with the maximum SiO₂ content (72/30SiO₂) exhibited a significantly higher sorption value (22.87 ± 1.41 μg/mm³) compared to all other formulations in this group, where WS ranged from 14.61 ± 0.85 to 16.65 ± 1.54 μg/mm³. In contrast, for composites in Group 2 with maximized filler loading (Table 9), no statistically significant differences in WS were found (p = 0.062), with values ranging narrowly from 20.53 ± 0.45 to 21.86 ± 1.08 μg/mm³ and showing a slight upward trend with increased total filler content. Similarly, in Group 3, variations in the ratio of coarse to fine aluminosilicate glass (Table 10) did not lead to statistically significant differences in water sorption (p = 0.120). The WS values for all composites in this group remained within a close range of 15.37 ± 0.41 to 16.33 ± 0.57 μg/mm³.

Solubility demonstrated distinct trends across the experimental groups. In Group 1 (Table 8), SL increased sharply and significantly with the content of spherical SiO₂ (p < 0.001). The 72/30SiO₂ composite showed the highest solubility (5.90 ± 0.25 μg/mm³), forming a distinct statistical group, while values for composites with 0–20 wt.% SiO₂ increased progressively from 1.77 ± 0.17 to 3.77 ± 0.22 μg/mm³. For Group 2 composites (Table 9), solubility values were consistent, showing no statistically significant variation (p = 0.343) and averaging around 4.68 ± 0.22 μg/mm³, regardless of the total filler loading.

A significant effect on solubility was observed in Group 3 (p < 0.001) (Table 10). The baseline composite without a defined fine fraction (74.5/10SiO₂) exhibited the highest solubility in this group (3.75 ± 0.12 μg/mm³). The introduction and variation in the fine aluminosilicate glass fraction (from 50 to 100 wt.%) led to a pronounced and statistically significant reduction in solubility, stabilizing at a consistently low level of approximately 1.25 ± 0.23 μg/mm³ for all corresponding composites.

4. Discussion

4.1. Effect of Spherical SiO₂ on Filler Loading

The unexpected viscosity reduction observed upon addition of spherical SiO₂ nano-particles, despite their higher specific surface area (8.361 ± 0.099 m²/g) compared to aluminosilicate glass (2.603 ± 0.053 m²/g), highlights the key role of particle morphology in composite rheology. Contrary to the established notion that a larger surface area enhances interactions with the matrix and increases viscosity, our results demonstrate that geometric factors may override this effect. The previous reports [4,32] have indicated that the viscosity of composites increases exponentially with filler content, a phenomenon which is more pronounced for smaller particles. Typically, the addition of a nanoscale filler increases adhesion within the polymer matrix due to an increased particle surface area, which in turn leads to an increase in the composition’s viscosity. Studies [3,4] noted that dental resin composites with similarly sized spherical amorphous SiO₂ fillers had a maximum loading of 71 wt%. In our case, the addition of spherical 500 nm SiO₂ particles to the large irregular aluminosilicate glass particles produced reduction in the viscosity of the composites and allowed to increase the filler loading to 80 wt.%.

The smooth, spherical surface minimizes mechanical interlocking between particles, in contrast to the irregular aluminosilicate glass particles, which exhibit higher friction due to their angular morphology and rough surface texture.

It should be noted here that as the content of the fine glass fraction was increased, the composites of group 3 became denser and lost their viscosity. This observation is consistent with the theory that smaller particles increase the total surface area, which requires a greater amount of resin to coat the particles. This, in turn, influences the viscosity and reduces the achievable filler loading in the material. Evidently, for the fine glass fraction particles, which possess a specific surface area (4.325 ± 0.061 m²/g) nearly two times smaller than that of the SiO₂ (8.361 ± 0.099 m²/g), their irregular shape is the predominant factor. This irregular shape provides greater friction between the particles themselves and the matrix material, prevents them from packing more densely.

This mechanism is consistent with rheological studies demonstrating that particle shape plays a dominant role in determining the flow behavior of highly filled polymer systems [21,42]. Recent fundamental research on composite fluids further confirms that tailoring the microscopic filler network structure through particle design is the key to controlling macroscopic rheological properties [43]

Optimal rheological properties and maximum filler loading are achieved not by simply reducing particle size, but through the rational combination of fillers with different shapes and sizes. Creating bimodal or multimodal particle systems, where spherical particles effectively fill the voids between larger irregular fragments, minimizes friction, improves packing, and overcomes the traditional limitation imposed by viscosity increase.

4.2. Mechanical Properties

For the first group of composites, the addition of SiO₂ initially led to a slight reduction in strength (Figure 6A), followed by an increase, and subsequently a further decrease. Such results are explained by the addition of nanoscale fillers, which may negatively affect mechanical properties of a composite as described previously in [41]. It should also be noted that the composites with 20% and 30% content exhibited reduced viscosity. Consequently, during the preparation of specimens for strength testing, they adhered to the instruments and molds, which could have contributed to uneven layering and the formation of voids in the cured materials, negatively affecting the flexural strength properties. Despite this, the highest flexural strength of 88.49 ± 5.14 MPa was shown by the specimen containing 10% spherical SiO₂ nanoparticles. This concentration likely allows for the initial formation of a stress-transferring filler network without significantly impeding the percolation of the primary, irregular aluminosilicate filler [44]. Apparently, adding 10% improves filler distribution within the polymer matrix, which improves adhesion and more effective stress dissipation. The microstructure photograph (Figure 5b) shows that the filler is uniformly distributed throughout the matrix, with no pores or voids detected.

An increased filler content leads to a noticeable improvement in flexural strength (Figure 6B) and flexural modulus (Figure 7B). Such effect was earlier described in [4]. The specimens of the 2nd group outperformed the rest by flexural strength. It is attributed to the number of contacts between the particles that facilitate distribution and dissipation of the stress throughout the whole volume. This corresponds to the formation of a rigid percolating network, where stress is effectively distributed via connected particle pathways [44]. The stress is distributed among a multitude of adjacent particles instead of being concentrated at a single point where failure is likely to initiate. A reduction in filler loading decreases the number of these contacts, which, in turn, lowers the strength limit. The highest strength value of 118.11 ± 10.54 MPa was demonstrated by the composite with 78 wt.% filler. When the loading was increased to 80%, the strength decreased; this is most likely attributed to the limiting value of the filler in the polymer matrix (Figure 5d) and the appearance of minor pores and voids.

For the third group of composites (Figure 6C), the highest strength value of 94.56 ± 5.78 MPa was demonstrated by the composite with a 50/50 ratio of coarse and fine glass fractions. This is the highest value in the group, statistically significantly exceeding the composition with 60% fines. This effect is directly related to the formation of the most efficient particle packing. Large irregular fragments create a structural framework, while fine particles optimally fill the interparticle voids, as confirmed by SEM images (Figure 5c). This dense, homogeneous structure ensures uniform distribution of mechanical load across the polymer matrix and minimizes stress concentrations.

Increasing the fines content to 60 wt.% caused a sharp strength decrease to 83.56 ± 3.31 MPa (Figure 6C), indicating a disruption of optimal particle packing. Excess small, angular particles raise surface area and friction, impairing flow, promoting agglomeration, and leading to incomplete resin wetting and a defective microstructure. A further increase to 70 and 100 wt.% fines yielded a slight strength recovery to 86.76 ± 5.54 MPa and 88.47 ± 7.25 MPa, respectively. These values, while above the 60 wt.% minimum, remain statistically similar to the base composite and below the 50/50 optimum. This trend suggests a shift toward a more homogeneous but suboptimal monodisperse structure, where high friction is constant but the synergistic density benefits of bimodal packing are lost.

In conclusion, strength depends nonlinearly on the glass fraction ratio, with a clear maximum at a balanced 50/50 blend. This optimum synergy from efficient packing is disrupted by an excess of fines, which worsens processing and creates a less perfect microstructure, ultimately compromising mechanical strength.

The flexural modulus is significantly improved by the addition of micro- and nano-particles to the polymer matrix, since the solid particles possess greater stiffness. The higher percentage of inorganic filler the higher flexural modulus. The hardness of light-curing dental composites is determined by a complex of parameters including the quality and content of the filler, the degree and uniformity of polymerization, as well as the characteristics of the polymer matrix and the curing technology [45].

The observed trends for flexural modulus and microhardness are intrinsically linked to the efficiency of stress transfer from the polymer matrix to the rigid filler particles and the overall packing density of the composite. According to the generalized model for spherical fillers, reinforcement is most effective above a percolation threshold where a connected network forms [46]. In Group 1, the initial increase in modulus (Figure 7A) and hardness (Figure 8A) with the addition of 10 wt.% spherical SiO₂ can be attributed to improved particle mobility and packing, leading to a more homogeneous microstructure with fewer defects. The spherical morphology likely facilitates better initial dispersion and packing, as shown in comparative studies with irregular particles [47]. However, the subsequent decline at higher SiO₂ concentrations (20–30 wt.%) suggests a critical threshold is surpassed. While spherical particles reduce viscosity, an excessive amount may disrupt the percolating network of the stronger, albeit irregular, aluminosilicate glass, creating regions with sub-optimal load-bearing capacity.

The flexural modulus and hardness in Group 2 directly result from achieving maximum filler loading (76–80 wt.%). The higher volume fraction of rigid inorganic phase provides a stiffer scaffold, directly increasing the composite’s resistance to elastic deformation (modulus (Figure 7B) and surface indentation (hardness Figure 8B). At such extreme loadings, the system likely approaches a jammed state, where particle mobility is severely restricted, leading to the observed peak in stiffness [48].

The non-linear relationship in Group 3, where both properties peaked at a 50/50 ratio of coarse to fine glass fractions, provides strong evidence for the “optimal packing” theory. At this ratio, the fine particles efficiently fill the voids between the coarse ones, maximizing density and minimizing resin-rich regions. This dense, interlocked structure enhances stress distribution and improves resistance to deformation and penetration. This synergistic effect of bimodal packing facilitates the formation of a more continuous and rigid mechanical network within the composite [48]. Deviating from this optimum leads to the significant drop in both modulus (Figure 7C) and hardness (Figure 8C).

4.3. Depth of Cure

The depth of cure of dental composites is a critically important parameter that directly influences the strength, durability, and biocompatibility of a restoration. The size, shape, and content of the inorganic filler affect the light scattering within the material [3,36,49]. Highly filled composites may slightly reduce the depth of cure due to greater light scattering on the filler particles. For the first composite group, a clear trend is observed: the depth of cure decreases with an increase in SiO₂ content (Figure 9A). This is associated with the fact that this type of filler, with a size of 500 nm, possesses a low transmittance coefficient for visible light [4,50]. When irradiated by a dental curing light, it scatters the light. The light, upon encountering the particles, is repeatedly deviated and undergoes multiple scattering within the material, being largely dispersed backwards or to the sides. Consequently, the direct component of the light beam, which is capable of reaching deep layers and initiating polymerization, is rapidly attenuated. The best result was shown by the composite without SiO₂ in its composition, with a depth of cure of 5.65 mm. The lowest value of 3.85 mm was possessed by the composite with a high (30 wt.%) content of silica.

As the filler content was increased to its maximum values, the depth of cure was observed to decrease, as expected (Figure 9B). For the second group of composites, the lowest result of 3.01 mm was recorded for the specimen with the highest filler loading and SiO₂ content, while the highest value was observed for the specimen with the lowest loading and silica content.

For the third composite group (Figure 9C, a slight reduction in the depth of cure values was noted with an increase in the content of the fine aluminosilicate glass fraction. The modest reduction in depth of cure with increasing fine glass fraction content (from 4.23 mm for 74.5/10SiO₂ to 3.61 mm for 74.5/10SiO₂/100) has dual origins. Firstly, increased surface area: fine fraction (4.325 m²/g) provides more filler-matrix interfaces for light scattering compared to coarse fraction (2.603 m²/g). Secondly, ZrO₂ contamination: a minor contamination with ZrO₂ from the grinding media used for milling the glass. Zirconium (Zr) was detected by energy dispersive analysis of the powder after milling (Figure 3c,k). Zirconium dioxide also possesses a low transmittance coefficient for visible light [51,52], which contributes to the reduction in depth of cure values. However, it should be noted that ZrO₂ is biocompatible, chemically inert, exhibits high radiopacity, and enhances the mechanical properties of dental composites [53,54]. This factor must be taken into account during the preparation of fillers by the milling method.

4.4. Water Sorption and Solubility

Composite hydrolytic stability depends on resin matrix hydrophilicity and filler-matrix interface integrity. The sharp increase in both water sorption and solubility with higher spherical SiO₂ content in Group 1 (Table 8) is a critical finding. Spherical SiO₂ nanoparticles possess an extremely high specific surface area and a hydrophilic, silanol-rich surface. As their concentration rises, the total interfacial area between the filler and the hydrophobic resin matrix increases exponentially (Table 9). This creates a percolating network of pathways for water diffusion along the imperfectly bonded interfaces. In Group 3 (Table 10), the dramatic and statistically significant reduction in solubility upon introducing the fine glass fraction (from ~3.75 μg/mm³ to ~1.25 μg/mm³) is particularly noteworthy. While WS remained relatively constant, the drastic drop in SL indicates a fundamental improvement in the stability of the polymer network or the interfacial region.

4.5. Clinical Implications and Material Selection Guidelines

This part provides evidence-based guidance for selecting composites based on their filler characteristics for different clinical scenarios.

For high-stress posterior restorations (Class I/II cavities, cores), the 78/20SiO₂ composite demonstrates optimal performance, offering superior flexural strength (118.11 ± 10.54 MPa) and modulus (9.26 ± 1.12 GPa). Its lower depth of cure (~3.8 mm) necessitates careful incremental placement with layers ≤ 2 mm.

For bulk-fill applications, the 74.5/10SiO₂ composite presents a balanced profile: adequate strength (>91 MPa), a favorable depth of cure (4.23 mm), and excellent thixotropic handling. It is suitable for moderate-depth cavities (3–4 mm), Class V restorations, and pediatric dentistry.

A universal option for both anterior and posterior use is the 74.5/10SiO₂/50 composite. Its optimized bimodal filler system provides high strength (~95 MPa), good cure depth (~4.1 mm), and a dense, void-free microstructure, making it ideal for general use in low-to-moderate stress situations.

Regarding hydrolytic stability, all composites meet ISO 4049:2019 [33] for water sorption. However, for long-term restorations in high-moisture environments, Group 1 or Group 3 composites (WS: 14–16 μg/mm³) are preferable over the maximum-load Group 2 composites (WS: 21–22 μg/mm³) to minimize potential plasticization and marginal degradation.

Clinicians should adapt their technique to material viscosity: high-filler composites are packable for condensation, while balanced formulations offer ideal thixotropy.

4.6. Limitations of the Study

A notable limitation of the present work is the absence of direct polymerization shrinkage measurements. Although the literature consistently demonstrates that increased filler loading reduces volumetric shrinkage [1,24,25], and our composites achieved high filler contents (up to 80 wt.%), quantitative shrinkage data would have strengthened the clinical relevance of our findings.

Based on established relationships in the literature, we can predict the following trends for our composite systems.

Filler content effect: Group 2 composites (76–80 wt.% filler) would be expected to exhibit lower polymerization shrinkage compared to Group 1 (72 wt.%), with the 80/30SiO₂ composite likely showing the lowest shrinkage values. Studies have shown that each 10 wt.% increase in filler content can reduce volumetric shrinkage by approximately 1–2% [24].

The incorporation of spherical SiO₂ nanoparticles, which enables higher filler loading while maintaining workability, would be expected to provide dual benefits: (a) direct shrinkage reduction through increased inorganic content, and (b) improved filler-matrix stress transfer reducing shrinkage stress development [3].

The Group 3 composites, particularly the 74.5/10SiO₂/50 composition with optimized particle packing, would be expected to show reduced shrinkage compared to monomodal systems at equivalent filler loadings, due to more efficient space filling and reduced matrix content [55].

Comprehensive shrinkage characterization using standardized methods [56] is planned for future studies.

5. Conclusions

This study investigated the properties of light-cured dental composites utilizing aluminosilicate fillers of various sizes and spherical silica nanoparticles synthesized via the Stöber method.

It was found that the incorporation of spherical SiO₂ particles of 500 nm (up to 30 wt.%) significantly decreased the viscosity of the aluminosilicate glass-based dental composite. This reduction in viscosity, in turn, allowed for a higher maximum filler loading within the composite. The maximum filler content was successfully raised to 76 wt.% with the addition of 10 wt.% SiO₂, to 78 wt.% with 20 wt.% SiO₂, and reached 80 wt.% with 30 wt.% SiO₂. All developed composites demonstrated values for mechanical properties (flexural strength >80 MPa), depth of cure (>1.5 mm), and water sorption and solubility (<40 and 7.5 μg/mm³, respectively), which fully complied with the requirements of the ISO 4049:2019 [33].

In the study of mechanical properties, it was noted that the increased filler loading led to a marked improvement in flexural strength and elastic modulus. The highest flexural strength value (118.11 ± 10.54 MPa) was observed in the composite with 78 wt.% filler content, while the highest hardness value (69.13 ± 2.63 HV) was recorded for the composite with the maximum 80 wt.% filler loading. This can be explained by the maximized dense packing provided by the hard filler.

Nevertheless, an inverse relationship was observed: an increase in the SiO₂ content led to a decrease in the depth of cure. This is because the 500 nm SiO₂ filler exhibits a low visible light transmittance coefficient and causes light scattering. Furthermore, the increase in SiO₂ nanoparticles also increased water sorption and solubility due to the larger total surface area, which promotes easier water diffusion.

Although direct polymerization shrinkage measurements were not performed in this study, the achieved high filler loadings (up to 80 wt.%) are expected to provide clinically beneficial shrinkage reduction based on well-established literature trends. Future work will include comprehensive shrinkage characterization to validate these expectations.

In conclusion, when formulating hybrid dental composites, it is essential to consider not only the size but also the shape of the filler, as it significantly influences the filler’s behavior within the matrix and modifies the overall composite properties. The optimal balance between filler loading, particle shape, and size distribution must be carefully tailored to specific clinical requirements, with high-strength formulations (78–80 wt.% filler) suited for posterior load-bearing restorations and balanced formulations (74.5 wt.% with bimodal glass) offering versatility for universal applications. The results of this study contribute to a deeper understanding of how filler composition influences the characteristics of the final materials, which will enable the precise tuning of composite properties to meet specific parameters while achieving compliance with all necessary clinical requirements in the future.