Microstructure and Mechanical Properties of Hybrid Pure Al/B₄C/Microsilica Composites Produced by Ultrasonically Assisted Stir Casting

Abstract

This study explores the fabrication and characterization of hybrid aluminum matrix composites reinforced with boron carbide (B₄C) and microsilica, produced via ultrasonically assisted stir casting followed by T6 heat treatment. Pure aluminum was selected as the base matrix to evaluate the combined effects of B₄C and microsilica reinforcements. Microstructural analyses showed that ultrasonic treatment effectively dispersed nanoparticles, reduced agglomeration, and enhanced particle–matrix interfacial bonding. T6 heat treatment further refined the grain structure through Zener pinning and promoted the formation of reaction layers at particle interfaces. Mechanical testing revealed that Al/B₄C composites provided the highest strength and hardness, while Al/microsilica systems retained superior ductility. The hybrid Al/B₄C/microsilica composites demonstrated a balanced combination of yield strength (38.6 MPa), ultimate tensile strength (82.6 MPa), and elongation (35.2%), confirming a synergistic strengthening–toughening effect. These results highlight the potential of Al/B₄C/microsilica hybrid reinforcements to optimize the trade-off between strength and ductility in aluminum-based composites.

1. Introduction

Aluminum matrix composites (AMCs) are attractive for lightweight structural applications because they combine low density with corrosion resistance and can achieve a favorable strength–ductility balance through particle strengthening, grain refinement, and dislocation storage mechanisms [1,2]. Compared with unreinforced alloys, AMCs enable load transfer to stiff ceramic particulates [3], Zener pinning of grain boundaries [4], and constraint of plastic flow [5], which collectively improve microstructural stability and mechanical performance [6,7,8]. This combination has motivated their use in transportation [9], defense [10], and energy devices [11] where weight reduction directly translates into improved efficiency and performance margins. Within this broader class, hybrid composites—where two or more reinforcement types are combined—are pursued to exploit complementary strengthening and toughening mechanisms beyond what single reinforcements can deliver [12,13,14].

From a processing perspective, multiple fabrication methods have been explored for AMCs, including powder metallurgy, squeeze casting, and stir casting with mechanical agitation, along with variants that introduce auxiliary energy to improve dispersion and wettability. Powder metallurgy routes provide fine control of reinforcement content and size distributions but often entail higher cost and multistep consolidation [15,16]. Squeeze casting can reduce porosity but requires specialized tooling and precise process windows [17]. In contrast, stir casting is attractive for scalability [18,19]; however, it faces well-known challenges—particle clustering [20], floating/sedimentation due to density mismatch [21], interfacial porosity [22], and inadequate wetting [23]—especially acute for ultrafine particles prone to agglomeration. To mitigate these issues, ultrasonic assistance superimposed on stir casting introduces acoustic streaming and transient cavitation [24] that fragment clusters, promote infiltration of liquid Al into particle interstices, and scrub oxide films [25,26], thereby improving particle–melt wetting and homogenizing distributions. Prior reports [27,28] point to reduced porosity, sharper reinforcement–matrix interfaces, and narrower spatial variance in particle number density when ultrasonics are properly coupled into the melt.

The choice of reinforcement is pivotal. Boron carbide (B₄C) is widely used thanks to its very high stiffness and hardness [29,30]; its angular morphology promotes mechanical interlocking and efficient load transfer [31], typically raising strength and hardness in Al/B₄C systems [32,33,34]. At the same time, high-volume B₄C can accentuate strain localization if the matrix–particle interface or spatial distribution is suboptimal [21,35]. Microsilica (silica fume), by contrast, consists of ultrafine, near-spherical SiO₂ particles [36]. When well dispersed, such particles act as effective dispersoids and grain-growth inhibitors [37]; they can stabilize substructures and help preserve ductility at a given overall particle loading by delaying necking and homogenizing plastic flow. These contrasting attributes suggest a hybrid pairing—using B₄C as a primary load-bearing reinforcement and microsilica as a fine dispersoid to refine grains, stabilize dislocation structures, and mitigate strain localization. Realizing this synergy depends sensitively on dispersion, interface quality, and defect control—factors strongly influenced by melt treatment and solidification conditions [38].

A further practical driver for selecting microsilica is its sustainability profile. Microsilica is an industrial by-product captured during silicon/ferrosilicon alloy production [39,40]. Leveraging this secondary stream valorizes a waste product, reduces landfill burden, and aligns with circular-material strategies [41] while avoiding additional high-temperature unit operations needed to synthesize purpose-made ceramics. Because the particles are recovered from ongoing metallurgical processes, the incremental embodied energy and CO₂ burden associated with their use as reinforcement is comparatively low relative to many primary powders. These advantages are particularly relevant for transportation applications where cumulative life-cycle emissions matter, provided standard industrial hygiene is observed for handling fine powders [42,43]. In short, a B₄C/microsilica hybrid in Al offers not only a pathway to balance strength and ductility but also an opportunity to reduce the environmental footprint of reinforcement sourcing.

Despite the breadth of literature on AMCs, several knowledge gaps remain pertinent to hybrid systems in pure Al matrices. First, many studies investigate either a single reinforcement or employ alloyed matrices [44,45] whose solute content complicates attribution of strengthening to the particles versus solid-solution or precipitation effects. Pure Al provides a stringent baseline in which dispersion strengthening, grain refinement, and load transfer are the dominant mechanisms, allowing cleaner interpretation of reinforcement roles [46]. Second, direct side-by-side comparisons of B₄C-only [47], microsilica-only [48,49], and B₄C + microsilica hybrids fabricated via ultrasonic-assisted stir casting under matched conditions and matched total reinforcement content remain limited. Without identical processing histories and equal total loading, it is difficult to isolate true “hybridization” effects from confounding variables such as porosity differences, particle size distributions, or melt treatment time. Third, there is a need to correlate microstructural statistics with tensile properties to clarify how microsilica dispersoids interact with angular B₄C to shape the strength–ductility response at low overall particle contents—regimes most relevant to castability and downstream forming [50].

To address these points, we adopt a processing framework that integrates mechanical stirring with ultrasonic treatment to enhance wetting and dispersion, followed by controlled solidification to limit buoyancy-driven segregation. By keeping the total reinforcement content constant across three material states—(i) B₄C-only, (ii) microsilica-only, and (iii) B₄C + microsilica hybrid—we can (a) quantify the degree to which the hybrid suppresses particle clustering relative to its single-phase counterparts, (b) evaluate grain refinement and porosity suppression attributable to fine oxide dispersoids under otherwise identical melt histories, and (c) measure the consequent shifts in the strength–ductility trade-off. Within this design, B₄C contributes primary load transfer and indentation resistance, while microsilica is expected to pin boundaries and homogenize the microstructure; ultrasonic cavitation/streaming is leveraged to realize these roles by improving particle–melt contact and breaking up agglomerates.

2. Materials and Methods

2.1. Experimental Overview

This study compares four material conditions—pure Al, Al reinforced with B₄C, Al reinforced with microsilica, and a hybrid Al reinforced with both B₄C and microsilica—produced by ultrasonically assisted stir casting.

Table 1. Nomenclature of the fabricated materials (HT—heat-treated).

Nomenclature	Designation	Pure Al (wt%)	B4C (wt%)	Microsilica (wt%)
Pure Al	Al	100	0	0
Pure Al–HT	Al–HT	100	0	0
Pure Al/1 wt% B4C	Al/B4C	99	1	0
Pure Al/1 wt% B4C–HT	Al/B4C–HT	99	1	0
Pure Al/1 wt% Microsilica	Al/Microsilica	99	0	1
Pure Al/1 wt% Microsilica–HT	Al/Microsilica–HT	99	0	1
Pure Al/0.5 wt% B4C/0.5 wt% Microsilica	Al/B4C/Microsilica	99	0.5	0.5
Pure Al/0.5 wt% B4C/0.5 wt% Microsilica–HT	Al/B4C/Microsilica–HT	99	0.5	0.5

summarizes the sample nomenclature and reinforcement mass fractions used consistently throughout the work, whereas Figure 1 illustrates the overall sequence from raw materials to processing, heat treatment, and characterization so that the reader can connect each material designation to its path through the workflow.

Figure 1a shows the starting materials used in this study. High-purity industrial aluminum (~99.5%) was used as the matrix material. Although it has relatively low strength, it offers high ductility. For the experiments, two ingots weighing approximately 14.5 kg each were supplied by Kazakhstan Metal Industrial Company LLP (Astana, Kazakhstan). Prior to melting, the ingots were cut into smaller blocks and placed in a cylindrical graphite crucible (working dimensions Ø84 × 161 mm), which provides high thermal resistance and chemical inertness with respect to aluminum. In this study, B₄C (supplier: Dongguan Sat Nano Technology Material Co., Ltd., Dongguan, China; purity ~93–94%) and microsilica powders were used as reinforcing phases; the microsilica is an industrial by-product (silica fume) from ferrosilicon production at YDD Corporation (Karaganda, Kazakhstan).

As shown in Figure 1b, the ultrasonically assisted stir-casting sequence proceeds in two stages: mechanical stirring followed by ultrasonic treatment of the melt. In the first stage, the crucible was placed in a vertical resistance furnace (type 1) capable of heating above 750 °C, i.e., well beyond the melting point of aluminum (660 °C). The furnace was equipped with a chromel–alumel (K-type) thermocouple and an electronic control unit, ensuring temperature stability within ±5 °C. A steel stirring rod with four blades, driven by an electric motor, was positioned at the center of the crucible; stirring was conducted at ~600 rpm for ~15 min to homogenize the melt and disperse B₄C and microsilica particles within the aluminum matrix. To minimize temperature gradients between the melt and the reinforcing phases, the powders were preheated to ~500 °C in a PM-5 muffle furnace (Plavka.Pro, Korolev, Russia). Immediately after mechanical stirring, the graphite crucible was rapidly transferred to a second resistance furnace (type 2) for ultrasonic treatment. A JH-LRT30 system (Hangzhou Precision Machinery Co., Ltd., Hangzhou, China) operating at ~20.6 kHz and 1.7 kW was employed, with a titanium probe partially immersed in the melt for ~15 min. The resulting cavitation and acoustic streaming disrupted particle agglomerates and enhanced particle–matrix wettability, leading to a more uniform distribution of the reinforcing phases in the aluminum melt.

Figure 1c illustrates the molding stage using a seven-channel split steel mold (each Ø25 mm × 220 mm) preheated to ~300 °C. Immediately after ultrasonic treatment, the melt was poured into the mold and allowed to cool to room temperature. The cast billets then underwent a T6 heat treatment cycle (Figure 1d): solutionizing at 530 ± 5 °C for 2 h in a PM-5 muffle furnace (Plavka.Pro, Korolev, Russia), water quenching at room temperature, and artificial aging at 170 ± 2 °C for 10 h. This sequence promotes the precipitation of strengthening phases within the aluminum matrix and enhances particle–matrix interfacial bonding, thereby improving the composite’s mechanical performance [51].

Figure 1e presents the geometries of the tensile and hardness specimens, as well as the samples prepared for light microscopy (LM), scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), and X-ray diffraction (XRD). The procedures for these specimens and samples are described in greater detail in Section 2.3 of the article.

2.2. Powders: SEM/EDS Characterization

SEM provides direct evidence of particle morphology and size—both critical for wetting, dispersion, and interfacial bonding in liquid Al—while EDS verifies chemical identity and reveals surface oxides or trace impurities that can affect reactivity during processing. SEM observations were performed on a Crossbeam 540 (Carl Zeiss, Oberkochen, Germany) operated at 5 kV. Representative micrographs are shown in Figure 2. Elemental analyses are summarized in Figure 3.

Figure 2a shows that B₄C particles are predominantly polyhedral with sharp edges, ranging from ~0.5 to 2 μm, in agreement with the supplier’s specification (Dongguan Sat Nano Technology Material Co., Ltd., Dongguan, China; purity ~93–94%). This size range favors uniform dispersion within the aluminum matrix when adequate stirring is applied. Figure 2b presents the microsilica powder, which consists of spherical primary particles ~50–200 nm in diameter; however, dense agglomerates up to several hundred nanometers are evident in the SEM images. Such agglomeration is typical for ultrafine silica and necessitates ultrasonic treatment to achieve effective dispersion in molten aluminum. Particle-size distribution descriptors are defined as follows: D10 is the diameter below which 10% of particles are found; D50 is the median diameter; D90 is the diameter below which 90% lie [52]. Particle size distributions were quantified from microscopy images using ImageJ version 1.54d (Wayne Rasband, National Institute of Health, Bethesda, MD, USA). For B₄C: D10 ≈ 0.6 μm, D50 ≈ 1.2 μm, D90 ≈ 1.9 μm; mean size 1.1 ± 0.4 μm. For microsilica: D10 ≈ 0.08 μm, D50 ≈ 0.17 μm, D90 ≈ 0.40 μm; mean size 0.18 ± 0.07 μm. The angular morphology of B₄C promotes mechanical interlocking with the Al matrix, whereas microsilica—owing to its high specific surface area and tendency to agglomerate—requires ultrasonic dispersion to achieve a uniform distribution.

For the B₄C powder (Figure 3a), the EDS spectra revealed two major elements—boron (B) and carbon (C)—consistent with the expected B₄C stoichiometry. Elemental mapping showed a homogeneous co-distribution of boron and carbon within the particles, confirming their carbide nature. The oxygen (O) signal was attributed to the formation of a thin oxide film during storage. Quantitative analysis yielded B ≈ 66.9 ± 0.3 wt% and C ≈ 26.3 ± 0.4 wt%, with oxygen and trace impurities accounting for <7 wt%. Despite the well-known limitations of EDS in detecting light elements, these results agree well with the theoretical composition and confirm the relatively high purity of the B₄C powder (~93–94%). For the microsilica powder (Figure 3b), the EDS spectra displayed dominant peaks for silicon (Si) and oxygen (O). The average values were Si ≈ 41.0 ± 0.1 wt% and O ≈ 52.5 ± 0.1 wt%, consistent with the composition of silica. Several impurity elements were also detected: K ≈ 2.5 ± 0.0 wt%, Fe ≈ 2.5 ± 0.1 wt%, and Na ≈ 1.6 ± 0.0 wt%. These impurities are most likely associated with the production of ferrosilicon at YDD Corporation or residual mineral contaminants. The distribution maps showed these impurities to be uniformly dispersed, suggesting that they were incorporated into the material structure rather than present as separate inclusions. Overall, the elemental analysis confirmed that the powders correspond to their declared characteristics: the B₄C powder is of high purity, while the microsilica consists mainly of SiO₂ with minor alkali and transition metal impurities.

2.3. Material Characterization

A comprehensive suite of methods was used to evaluate the microstructure and properties of the fabricated materials, including LM, SEM with EDS, XRD, and mechanical testing.

Polished metallographic sections were prepared as follows: grinding with SiC papers (320–2000 grit), primary polishing with polycrystalline diamond suspensions (9 → 3 → 1 μm), and final polishing with Eposil M suspension (ATM Qness GmbH, Mammelzen, Germany), 0.06 μm, pH 9.5. Prior to imaging, surfaces were etched for 15–20 s using an active Eposil F suspension (ATM Qness GmbH, Mammelzen, Germany) containing ~0.1 μm particles and diluted with an aluminum etchant. This treatment removed micro-irregularities, improved surface reflectivity, and revealed microstructural features. LM observations were conducted at ×250 magnification (Axioscope 5 microscope, Carl Zeiss, Oberkochen, Germany). Surface morphology and matrix–reinforcement interfaces were examined using a Crossbeam 540 SEM (Carl Zeiss, Oberkochen, Germany) operated at 10–20 kV (working distance 3.4–6.6 mm). Elemental composition and distributions were assessed with the integrated EDS system, enabling mapping of Al, B, C, Si, and O.

Phase analysis was carried out by XRD on a Rigaku SmartLab diffractometer (Rigaku Corporation, Tokyo, Japan) equipped with a monochromatic Cu-Kα source (λ = 1.540 Å). Scans were performed over 2θ = 5–75° with a step size of 0.1°, allowing detection of characteristic peaks of aluminum and reinforcing phases. The acquisition time per sample was ~10 min, ensuring reliable phase identification and evaluation of relative peak intensities.

Tensile tests were conducted in accordance with ASTM E8 [53] using a WDW-100 universal testing machine (100 kN; Jinan Xinluchang Testing Machine Co., Ltd., Jinan, China) at a crosshead speed of 1 mm/min. Cylindrical specimens had a gauge diameter of 9 mm and a gauge length of 60 mm. At least three replicates were tested for each condition.

Microhardness was measured on cylindrical specimens (Ø16 × 30 mm) following ASTM E384 [54] using an HVT-1000A microhardness tester (Laizhou Laihua Testing Instrument Factory, Laizhou, China) with a Vickers diamond indenter (apex angle 136°). A load of 0.2 kgf (~1.96 N) was applied with a dwell time of 15 s. To improve statistical reliability, at least five indentations were made at different surface regions of each specimen, and the average value was reported.

3. Results and Discussion

3.1. Microstructural Observations

Representative microstructures (LM) of the fabricated materials, both in the as-cast condition and after T6 heat treatment, are shown in the optical micrographs (Figure 4).

Figure 4a shows the optical microstructure of pure Al (as-cast), exhibiting a coarse dendritic, cast-type structure with an average grain size of ~80–120 µm, typical for high-purity aluminum that lacks precipitation hardening. Figure 4b illustrates pure Al–HT. After T6, no substantial microstructural change is observed: the structure remains coarse and dendritic because the matrix contains no alloying elements that could form fine precipitates to impede dislocation motion during artificial aging. In contrast, aluminum alloys such as Al–Si, Al–Cu, or Al–Mg typically exhibit precipitation of fine phases (e.g., Mg₂Si or Al₂Cu), which hinder dislocation motion and enhance thermal strengthening [55,56]. In high-purity aluminum (~99.5% Al), such processes do not occur; thus, even after solution treatment and aging, the structure remains coarse-grained and the strengthening effect is minimal [57,58]. Figure 4c shows Al/1 wt% B₄C (as-cast). The B₄C particles are generally well distributed, with occasional small clusters. Their angular morphology promotes strong mechanical interlocking with the Al matrix, and even in the as-cast state the composite displays a finer structure than pure Al because the particles restrict unrestricted dendritic growth. Figure 4d illustrates Al/1 wt% B₄C–HT. T6 treatment leads to improved interfacial bonding and pronounced grain refinement of aluminum to ~40–60 µm. This behavior is consistent with the Zener pinning mechanism [59,60], in which dispersed inclusions limit grain-boundary migration, lowering the likelihood of microcrack initiation and enhancing strength. Figure 4e shows Al/1 wt% microsilica (as-cast). The initial microstructure is less homogeneous: coarse-grained regions form near nanoscale SiO₂ agglomerates, yielding an average grain size of ~70–90 µm. Figure 4f illustrates Al/1 wt% microsilica–HT. Heat treatment promotes a more uniform distribution of microsilica and further grain refinement to ~60–70 µm, reflecting the ability of ultrafine/nanoscale particles to suppress grain growth during aging [61,62,63]. Figure 4g shows the hybrid Al/0.5 wt% B₄C/0.5 wt% microsilica (as-cast). The structure already exhibits a more uniform distribution of both reinforcements and an average grain size of ~70–80 µm, finer than in pure Al. Figure 4h illustrates the hybrid composite–HT. After T6, the aluminum grain size is further reduced to ~50–70 µm, and the microstructure becomes notably more homogeneous. This outcome arises from the synergistic action of the reinforcements: angular B₄C particles serve as grain-boundary pinning sites, while ultrafine microsilica suppresses coarsening and coalescence [64,65].

In addition to light (optical) microscopy of the grain structure, a detailed examination of the surface and the distribution of reinforcing phases was carried out using SEM (Figure 5). The resulting images provided clearer insights into the interaction between the aluminum matrix and the B₄C and microsilica particles, as well as the quality of the matrix–particle interfaces. Figure 5a shows pure Al (as-cast) with a relatively featureless matrix, sparse casting pores, and interdendritic traces; the secondary dendrite arm spacing is ~5–10 µm. Figure 5b (pure Al–HT) reveals no meaningful morphological change, consistent with the limited efficacy of T6 for high-purity Al lacking precipitation. Figure 5c displays Al/B₄C (as-cast): dispersed B₄C particles are visible within the matrix together with localized agglomerates and micropores adjacent to particles. In Figure 5d (Al/B4C–HT), the particle–matrix interfaces sharpen and the count of interfacial defects decreases, indicating improved wettability/diffusion bonding after T6 [66,67,68]. Figure 5e presents Al/microsilica (as-cast): clusters of nanoscale SiO₂ appear as rounded inclusions; microvoids/porosity are frequently observed near these regions, evidencing weaker interfacial bonding. After T6, Figure 5f shows a more uniform dispersion and a denser Al–SiO₂ interfacial zone, although residual agglomerates remain. For the hybrid system, Figure 5g (as-cast) reveals a more balanced co-distribution of angular B₄C (≈1–2 µm) and fine microsilica (≈100–200 nm), forming combined reinforcement zones alongside isolated particles. Post-treatment, Figure 5h exhibits well-defined interfaces and fewer pores/defects; any remaining interfacial gaps are limited to ~200–400 nm, consistent with strong bonding and reduced weak regions.

Full-field EDS of pure aluminum is shown in Figure 6a for the as-cast state and in Figure 6b after T6. Aluminum is the primary element at ~94–95 wt%. Minor signals of carbon (4–5 wt%) and oxygen (~1 wt%) are present. The presence of carbon (C) is attributed to the limitations of the EDS method in detecting light elements, as well as potential background contributions from the carbon substrate or surface contamination during sample preparation. The oxygen (O) corresponds to a thin native Al₂O₃ film formed on air exposure. After T6, the quantitative ratios are essentially unchanged, and no new peaks or precipitated phases are detected, indicating compositional stability of pure aluminum. Elemental maps show a uniform Al distribution without localized impurity clusters, confirming a homogeneous matrix. The lack of notable chemical change after T6 aligns with microstructural observations: in the absence of alloying elements, T6 mainly induces limited recrystallization without altering composition.

EDS elemental mapping from selected regions of interest (ROIs) is shown for the Al/B₄C composite in Figure 7a and for the heat-treated state in Figure 7b. In both panels, Al dominates the matrix (>74.0 wt%). In Al/B₄C, B and C are co-localized at the particle in the field center, while matrix O remains low (≈0.7 wt%). After T6, B–C co-localization persists, and a slightly higher interfacial O signal appears near particle boundaries at the upper right; across the analyzed ROIs O ≈ 0.7 wt%. No new peaks or secondary phases are resolved after T6, indicating a stable B₄C distribution and no change in matrix composition.

EDS elemental mapping from selected ROIs is shown for Al/microsilica in Figure 8a and for Al/microsilica–HT in Figure 8b. In the as-cast state (Al/Microsilica), the Al matrix (~89.2 wt%) contains discrete inclusions with co-localized Si (~1.4 wt%) and O (~3.8 wt%), confirming SiO₂ particles. Carbon (~4.9 wt%) and trace Fe and Ti (~0.5 wt% each) occur sporadically, consistent with contamination or residual inclusions. After T6 (Al/Microsilica–HT), Si ≈ 0.6 wt%, O ≈ 2.0 wt%, and C ≈ 3.7 wt%. Local Si–O co-distribution persists within inclusion regions, but the microsilica particles appear more sharply defined, showing cleaner boundaries, reduced haloing, and less apparent dispersion into the matrix. The narrower Si and O intensity profiles across particle–matrix interfaces suggest interfacial stabilization after heat treatment. A thin Al–Si–O interfacial layer is plausible, yet no new peaks or secondary phases are resolved in the ROI spectra. Taken together, the maps indicate that T6 reduces background O and Si, preserves SiO₂ identity, and limits apparent solute bleed-out, yielding sharper inclusions embedded in a chemically uniform Al matrix. The outcome is consistent with diffusion-limited interfacial adjustment rather than bulk phase transformation, and it aligns with the absence of measurable compositional change in the surrounding matrix.

EDS elemental mapping from selected ROIs in the hybrid composite is shown in Figure 9a (Map 6) and Figure 9b (Map 7). For Map 6, the ROI-averaged composition is Al = 88.2 ± 1.5 wt%, C = 6.1 ± 0.3 wt%, B = 2.9 ± 1.6 wt%, O = 2.1 ± 0.5 wt%, and Si = 0.4 wt%. Local maps show Si–O co-distribution, confirming SiO₂, and overlapping B–C, consistent with B₄C. The spatial separation of these two inclusion types within the same field indicates a two-population microstructure where SiO₂-rich and B₄C-rich regions coexist but remain chemically distinct at the map scale. The oxygen signal, while modest, intensifies at inclusion rims and along narrow particle–matrix contact zones, which is consistent with the presence of thin oxide interlayers or limited interfacial reaction products. For Map 7, a B₄C-rich inclusion is observed. The reported composition is Al = 56.3 ± 1.0 wt%, C = 33.6 ± 0.6 wt%, B = 8.6 ± 1.4 wt%, O = 0.6 wt%. The B:C ratio approaches boron-carbide stoichiometry, and the very low O and Si values indicate a nearly “pure” B₄C domain with minimal microsilica admixture in this ROI. Element maps exhibit strong B–C overlap with steep intensity gradients at the inclusion boundary and little to no O enrichment at the rim, suggesting that any oxide layer present is below EDS detectability or is discontinuous here. The adjacent Al matrix shows no detectable secondary phases within the ROI.

EDS elemental mapping from selected ROIs in the Al/B₄C/microsilica–HT composite is shown in Figure 10a (Map 1) and Figure 10b (Map 2). In Map 1, the ROI composition is Al = 63.7 ± 0.6 wt%, C = 29.7 ± 0.3 wt%, B = 3.8 ± 0.9 wt%, O = 2.0 wt%, with Fe ≈ 0.6 wt%. Element maps show overlapping B and C at the particle (center-left), consistent with B₄C. A modest O signal concentrates along the particle–matrix contact, consistent with a thin oxide-rich interfacial zone or limited reaction layer; Fe appears as trace, spatially discrete spots without forming a continuous phase. In Map 2 (Figure 10b), the ROI composition is Al = 87.9 ± 0.5 wt%, C = 6.6 ± 0.5 wt%, Si = 2.9 ± 0.1 wt%, B = 1.6 ± 3.1 wt%, with O = 1.1 wt%. The maps show co-localized Si and O at the field center, confirming a microsilica (SiO₂) inclusion. Overall, after T6, the composite retains two distinct inclusion types—B₄C particles (occasionally with localized interfacial oxygen) and SiO₂ particles embedded in Al. The data indicate no complete dissolution of reinforcements; instead, the maps support interface-limited adjustment dominated by oxygen at particle boundaries, without the emergence of new bulk phases in the mapped regions.

XRD patterns for the reference and composite conditions are shown in Figure 11. All investigated samples exhibited three characteristic aluminum peaks in the diffractograms: the (111) plane at ~38.5°, the (200) plane at ~44.7°, and the (220) plane at ~65.1° (2θ, Cu Kα). Comparative analysis of peak intensities for different compositions revealed trends associated with the dispersion of secondary phases and the effects of heat treatment (Figure 11). For as-cast aluminum, the (111) peak exhibited relatively low intensity. After T6, the Al peaks show higher integrated intensity and narrower breadths, consistent with reduced microstrain and defect density due to recovery; we do not attribute these changes to crystallite or grain growth. Differences in the (111), (200), and (220) relative intensities between Al and Al-HT are most likely due to mild preferred orientation (texture) and are not used for size inference. With the addition of B₄C, the (111) peak intensity decreased compared with pure Al–HT, reflecting lattice distortion caused by the incorporation of the refractory phase. The addition of microsilica also reduced the (111) peak intensity, though less markedly than B₄C. The most pronounced reduction occurred when both B₄C and microsilica were introduced, especially after HT, reflecting increased microstrain and greater structural refinement.

For Al and Al–HT, the (200) peak was well defined and intensified after HT. The addition of B₄C caused a slight decrease in intensity, suggesting partial texture disruption. A similar but less pronounced effect was observed in samples containing microsilica. In the combined systems (B₄C + microsilica), the (200) peak was significantly suppressed, indicating the cumulative influence of both phases on crystallite orientation. After HT, the (200) intensity in these systems remained subdued, reflecting stabilization of a defect-rich structure. For pure Al, the (220) peak was clearly visible and intensified after HT. With B₄C addition, this peak weakened, consistent with local lattice distortions. Microsilica produced a similar but milder reduction. In ternary systems (B₄C + microsilica), the (220) peak retained moderate intensity in the as-cast state but became less pronounced after HT, suggesting stress redistribution and increased dislocation density.

The peak positions remained unchanged across all series, indicating no significant alteration of the aluminum lattice parameter and confirming that the reinforcements did not dissolve in the aluminum matrix as a solid solution. Differences between the samples were manifested primarily as variations in peak intensities: in pure aluminum, heat treatment enhanced peak intensities due to reduced defect density and crystallite growth, whereas the addition of B₄C and microsilica reduced peak intensities and caused partial broadening. These effects are associated with the development of microstrain and a more refined structure, particularly evident in composites containing both phases.

Distinct peaks of the reinforcing phases themselves were either not detected or were extremely weak. Several factors explain this: the relatively low content of dispersed particles, close to the XRD detection limit; the light-element composition of B₄C with a low scattering factor and reflections overlapping with intense aluminum peaks (particularly at 38–39° and 65–67°); and the predominantly amorphous character of microsilica, which produces only a broad diffuse maximum in the 20–25° range, nearly indistinguishable from aluminum reflections. Furthermore, fine particle dispersion and the presence of interfacial reaction layers further reduced the diffraction contribution of the reinforcements.

Despite the absence of distinct B₄C and microsilica peaks, their presence in the composites is confirmed indirectly and directly. Indirect evidence lies in the changes in intensity and broadening of the aluminum peaks, reflecting increased microstrain and incorporation of hard particles into the matrix. Direct confirmation is provided by microstructural observations, which revealed inclusions enriched in B, C, Si, and O, consistent with the composition of the reinforcements. Thus, although the diffractograms do not display strong signals from B₄C and microsilica, the combined structural and microstructural evidence confirms their successful incorporation into the aluminum matrix.

3.2. Mechanical Properties

Figure 12a shows representative stress–strain curves; Figure 12b and Figure 13 summarize YS, UTS, elongation to fracture, and microhardness HV0.2. The baseline Al is soft and very ductile: YS 25.4 MPa, UTS 37.4 MPa, elongation 62.2%, 21.6 HV0.2. T6 heat treatment raises strength and hardness with only a modest loss of ductility (Al–HT): YS 30.5 MPa, UTS 42.4 MPa, elongation 53.3%, 25.7 HV0.2. The curve shape confirms limited work-hardening capacity typical of high-purity Al; T6 mainly reduces defects and redistributes solute so the uniform strain remains high. Introducing 1 wt% B₄C changes the balance toward strength. Al/B₄C reaches YS 35.3 MPa and UTS 63.1 MPa, while elongation drops to 28.5% and HV0.2 increases to 38.1 HV0.2. The stress–strain curve shows a short uniform strain and an early necking event, consistent with strong load transfer to stiff B₄C and strain localization near particle interfaces. After T6, Al/B₄C–HT achieves the highest single-reinforcement UTS at 88.6 MPa with YS 40.5 MPa, elongation 33.7%, and 51.4 HV0.2. Compared with the as-cast B₄C composite, UTS rises by ~40% and hardness by ~35%, while elongation recovers by ~18%. The slight YS decrease (−15%) indicates that heat treatment mainly improves work-hardening and interfacial integrity rather than the onset of yielding. Mechanistically, dislocation storage around hard B₄C, improved metallurgical bonding, and reduced porosity increase the post-yield slope and the true fracture stress. Microsilica produces a different signature. In the as-cast state Al/Microsilica shows YS 33.4 MPa, UTS 72.3 MPa, elongation 47.5%, and 29.5 HV0.2. Relative to Al (as-cast), YS increased by ~32%, UTS by ~93%, and microhardness HV0.2 by ~37%, with a moderate ~24% reduction in ductility. The curve exhibits an extended stage of stable plastic flow before necking, which points to strain homogenization by fine SiO₂ dispersoids. After T6, Al/Microsilica–HT records YS 34.3 MPa, UTS 72.9 MPa, elongation 63.5%, and 34.5 HV0.2. Versus the as-cast state (Al/Microsilica), YS and UTS are nearly unchanged, but elongation jumps by ~34% and hardness increases by ~17%. The gain in ductility with modest hardness growth suggests coarsening and redistribution of dispersoids, relaxation of residual stresses, and cleaner interfaces that delay void coalescence, thereby shifting failure to higher strains [69,70]. Among all single-phase systems, Al/Microsilica–HT delivers the best ductility (72.3%) at intermediate strength. The hybrid composite balances these tendencies. As-cast Al/B₄C/Microsilica yields YS 36.2 MPa, UTS 70.4 MPa, elongation 26.4%, 39.5 HV0.2. Against Al this is +43% YS, +88% UTS, +83% HV0.2, and −58% elongation. T6 unlocks the intended synergy: Al/B₄C/Microsilica–HT reaches YS 38.6 MPa, UTS 82.6 MPa, elongation 35.2%, and 48.8 HV0.2. Relative to its as-cast state (Al/B₄C/Microsilica) this is +7% YS, +17% UTS, +33% elongation, and +24% hardness. Compared with Al/B₄C–HT, the hybrid sacrifices ~6% UTS (82.6 vs. 88.6 MPa) but gains ductility (35.2 vs. 33.7%). The stress–strain curve shows higher work-hardening than pure Al and a delayed instability relative to as-cast B₄C material, consistent with mixed mechanisms: B4C supplies load-bearing and Orowan strengthening; microsilica dispersoids refine grains and distribute strain; T6 reduces porosity and sharpens particle–matrix interfaces so the composite hardens more uniformly before necking.

Hardness trends (Figure 13) follow the tensile data and validate the microstructural mechanisms. HV0.2 ascends from 21.6 (Al) → 25.7 (Al–HT) → 38.1 (Al/B₄C) → 51.4 (Al/B₄C–HT), and from 29.5 (Al/Microsilica) → 34.5 (Al/Microsilica–HT); the hybrid tracks near the B4C systems at 39.5 (as-cast) and 48.8 (T6). The large hardness increments with B₄C indicate strong resistance to localized plasticity, whereas the smaller but steady gains with microsilica reflect dispersion strengthening without severe embrittlement.

Ultrasonically assisted stir casting and subsequent T6 act as the two main levers that shape the strength–ductility balance across all variants. During casting, high-intensity ultrasound generates cavitation and acoustic streaming that deagglomerate microsilica, break B₄C clusters, scrub oxide films, and expel gas, which reduces porosity and refines the grain structure. The cleaner, finer matrix and the improved particle wetting/anchoring raise load-transfer efficiency and delay void nucleation, so even the as-cast composites show large gains over Al. T6 then locks in these ultrasonic benefits: solutionizing and quench remove residual segregation, interfaces heal, and the dislocation/dispersion substructure stabilizes; after ageing, work-hardening capacity increases and strain localization weakens. In the hard-particle route, Al/B₄C–HT reaches the peak single-reinforcement strength (UTS 88.6 MPa, 51.4 HV0.2) with partial recovery of elongation to 33.7%, showing that better interfaces and lower defect content convert early yielding into sustained hardening. In the dispersoid route, Al/Microsilica–HT achieves plasticity (47.5%) with intermediate strength (UTS 72.3 MPa), indicating effective strain homogenization after ultrasonic dispersion and ageing. The hybrid profits from both mechanisms: Al/B₄C/Microsilica–HT delivers a practical compromise of UTS 82.6 MPa, YS 38.6 MPa, elongation 35.2%, and 48.8 HV0.2. Overall, ultrasound minimizes casting defects and maximizes particle effectiveness, while T6 consolidates these microstructural advantages into stable interfaces and higher work-hardening, producing stronger yet more damage-tolerant composites.

4. Conclusions

This study shows that ultrasonically assisted stir casting followed by T6 heat treatment yields aluminum matrix composites with clean particle–matrix interfaces, low porosity, and a uniform dispersion of both B₄C and microsilica. The processing route reliably wets the reinforcements and suppresses agglomeration, which in turn produces a refined and homogenized matrix microstructure suitable for load-bearing applications. The two reinforcements play complementary roles. B₄C particles pin grain boundaries and act as strong, stiff load carriers, raising the resistance to plastic flow. Microsilica behaves as a fine, well-distributed dispersoid that stabilizes the substructure, delays strain localization, and helps preserve ductility after ageing. These mechanisms operate under the same thermal schedule and do not require additional alloying or multi-step processing, which supports scalability.

Mechanical data are consistent with these roles. The reference Al shows the following: YS of 25.4 MPa, UTS of 37.4 MPa, elongation to failure of 62.2%, and hardness of 21.6 HV0.2. With single reinforcements, the Al/B₄C–HT condition attains the highest strength (YS 40.5 MPa, UTS 88.6 MPa, elongation 33.7%, hardness 51.4 HV0.2), while Al/Microsilica–HT maintains superior ductility at intermediate strength (YS 34.3 MPa, UTS 72.9 MPa, elongation 63.5%, hardness 34.5 HV0.2). The hybrid Al/B₄C/Microsilica–HT achieves a balanced property set—YS of 38.6 MPa, UTS of 82.6 MPa, elongation of 35.2%, and hardness of 48.8 HV0.2—indicating a synergistic strengthening–toughening effect that narrows the usual strength–ductility trade-off at low total particle content.

Taken together, the results confirm that coupling ultrasound-aided dispersion with a standard T6 schedule is an efficient and industrially practical route to damage-tolerant hybrid AMCs. B₄C primarily delivers strength, microsilica primarily preserves ductility, and their combination provides a controllable balance without sacrificing manufacturability. These findings provide a clear basis for tailoring property targets through reinforcement selection and heat treatment, and they motivate future work on fatigue and corrosion performance under service-relevant conditions.