Grain Refinement and Mechanical Enhancement of Titanium Matrix Composites with Nickel-Coated Graphene Nanoflakes: Influence of Particle-Size Mismatch

Abstract

A novel type of titanium matrix composite (TMC) with a uniform network microstructure has been successfully fabricated by adjusting particle-size mismatch (Φ). This study can also improve the understanding of the effects of particle size on microstructure and mechanical properties, particularly in titanium matrix composites reinforced with graphite flakes (GNFs). Microstructural analysis reveals the absence of noticeable defects, and significant grain refinements have been realized. The experimental results indicate that the yield strength of the mismatched composite is improved by 24.75% compared to that of normal composites. The micro-hardness also exhibits a 10.3% increase. These enhancements can be attributed to the introduction of particle-size mismatch, the refinement of the microstructure, and the deflection of interface cracks. The presence of distorted GNF lattices in the interface micro-region of the composites primarily results from the appropriate sizing of different particles.

1. Introduction

Titanium matrix composites (TMCs) have garnered significant attention in various industries, including aerospace, petrochemical engineering, and automobile manufacturing, due to their exceptional properties [1,2,3]. These composites exhibit remarkable characteristics, such as high specific strength, low density, excellent corrosion resistance, and resistance to abrasion. These attributes offer numerous advantages, such as reducing the weight of structures, improving aircraft performance and fuel efficiency, and minimizing exhaust emissions. Additionally, the exceptional corrosion resistance of titanium matrix composites plays a crucial role in extending the service life of components and ensuring aircraft safety, making them indispensable in the aviation industry [4,5,6]. The combination of carbon materials’ unmatched thermal conductivity [7] and titanium’s exceptional mechanical properties [8] further enhances the potential of TMCs [9,10,11,12,13]. This unique synergy holds great promise for the development of lightweight aircraft components that can meet stringent performance requirements while providing excellent heat transfer capabilities.

Graphene is widely recognized as an exceptional reinforcement material, primarily due to its remarkable strength, surpassing that of other conventional fillers, and its ability to achieve good dispersion [14,15]. Furthermore, graphene exhibits the highest thermal conductivity, exceeding 5000 W/mK. Consequently, graphene has garnered significant attention in the past five years for its potential to enhance the performance of lightweight matrices [16,17,18]. Notably, Yun et al. have successfully fabricated titanium composites reinforced with nickel-coated graphene, which have exhibited superior strength and ductility [19]. The introduction of graphene into titanium matrices holds considerable promise and has the potential to serve as a next-generation aviation material [20,21,22].

The incorporation of a two-scale structure has proven to be an effective approach in enhancing toughness. In the case of composites, such as laminate composites, the addition of a titanium layer has demonstrated a substantial improvement in toughness values, surpassing monolithic composites by more than an order of magnitude [23,24,25]. In the context of two-scale TMCs, the utilization of Ti6Al4V powders with varying sizes has exhibited a remarkable combination of specific strength and toughness. Shen et al. successfully fabricated graphene-reinforced titanium composites through laser melting deposition, resulting in a significant increase in specific strength. The tensile strength witnessed an approximate 20% improvement, reaching 1250 MPa. However, it is important to note that improper processing conditions can lead to interface defects within the composites.

To address these challenges, the development of alternative methods for producing multi-scale titanium matrix composites (TMCs) becomes crucial. Powder metallurgy has emerged as an ideal technique for fabricating TMCs with a sandwich structure, utilizing discontinuous reinforcement [26,27,28]. In a study by Mu et al. [29], an immersion reduction method was proposed to introduce nickel-coated graphite flakes into titanium alloys, establishing a mechanical–metallurgical synergy between the graphite nanofibers (GNFs) and titanium. The significant improvement in composite strength can be attributed to the uniform distribution of nickel-coated graphite nanofibers within the titanium matrix, as well as the formation of special interfaces. In line with the successful development of the discontinuously reinforced titanium matrix composites (DRTMCs) mentioned in the literature, the present experiment aims to utilize powder metallurgy to create a two-scale structure with favorable interface characteristics. The objective of this research is to achieve superior strength and ductility in the resulting TMCs.

Graphene has been introduced as a reinforcement material using powder metallurgy techniques for the high-performance TMCs [30]. The properties of these composites are influenced by various factors, including the volume fraction, distribution, and size of the reinforcement, as well as the density and bonding between the matrix and reinforcement [31,32,33,34]. Among these factors, achieving a uniform dispersion of graphene within the metal matrix is a crucial requirement in the manufacturing process of titanium matrix composites. Several researchers have developed models to describe the impact of these factors on the composite properties. Previously, Dong et al. [35] successfully dispersed 0.3 wt% graphene oxide nanosheets (GONs) into pure Ti powders using a hydrothermal synthesis method. The resulting GON/Ti composites were sintered via spark plasma sintering (SPS). Their findings revealed a 9.7% increase in tensile strength compared to that of the sintered pure Ti matrix. However, despite significant increases in graphene content, the mechanical properties of Ti-based composites often reach a plateau and cannot be further improved [36,37,38].

This study aimed to optimize various milling and sintering parameters for the preparation of GNF(Ni)/Ti-6Al-4V composites. The GNF-reinforced titanium composites were successfully fabricated through the mechanical mixing of graphene nanoflakes, followed by hot-press sintering. The main objective of this study is to investigate the effect of particle-size mismatch on the microstructure and mechanical properties of GNF(Ni)/Ti-6Al-4V composites. Aiming at the problem of the inhomogeneous microstructure of additive manufacturing titanium matrix composites, this paper innovatively proposes a new method to precisely control the microstructure based on particle-size mismatch. The study investigates the effects of particle-size mismatch on nucleation, grain growth and phase transition behavior. The study also explores the evolutionary mechanisms of microstructural features such as grain size, flake width, elemental distribution and phase composition. In addition, this paper compares the mechanical properties of different grain size mismatches and elucidates the intrinsic metallurgical mechanism of microstructural modification through grain size mismatches.

2. Experimental Procedures

2.1. Raw Materials

The graphene nanoflakes (GNFs) used in this study were obtained from Jiangsu XFNANO Materials Tech Co. Ltd. (Nanjing, China). Figure 1 presents the scanning electron microscope (SEM) image of the pristine GNFs, while the energy-dispersive X-ray spectroscopy (EDS) mapping results confirm the presence of carbon (C) elements within the GNFs. Raman spectroscopy analysis reveals three characteristic peaks, indicating the high crystallinity of the GNFs’ surfaces [39]. Additionally, the images in Figure 1c showcase the original atomic force microscope (AFM) images of the GNFs, highlighting an average thickness of 1–3 μm.

The Ti-6Al-4V (TC4) powders utilized in this study were commercially obtained from Xi’an Sino-Euro Materials Technologies Co. Ltd. (Xi’an, China). These powders were manufactured using a plasma rotating electrode process (PREP), starting from machined forging processed bars. The morphology and size distribution of the powders are depicted in Figure 2. The chemical composition of the TC4 powder was determined using an inductively coupled plasma atomic emission spectrometer (ICP-AES), and the results are presented in

Table 1. Chemical composition (wt.%) of as-atomized Ti-6Al-4V powder.

Diameter/μm	Al	V	Fe	O	N	C	Ti
0–15	6.15	4.12	0.04	0.23	0.034	<0.01	Bal.
15–53	6.01	3.91	0.16	0.11	0.007	<0.01	Bal.
53–105	6.23	4.14	0.06	0.05	0.006	<0.01	Bal.
53–150	6.1	4.05	0.04	0.06	0.006	<0.01	Bal.

. It should be noted that the oxygen content on the powder surface complies. The chemicals used in the experiment were sourced from Nanjing WANQING Chemical Glassware & Instrument Co. Ltd. (Nanjing, China).

To investigate the microstructure evolution, physical properties, and mechanical properties of composite materials with different particle-size mismatches, four different particle-size mismatches (Φ = 0, 0.15, 0.4, 0.75) were designed. These mismatches were used to prepare 0.5 wt.% GNF(Ni)/TC4 composites through a combination of high-energy ball-milling (HEBM) and hot-pressing sintering (HPS). The particle size range and mismatch value (Φ) for the two different particle sizes used in the mismatches are provided in

Table 2. Mismatch coefficient Φ and size distribution of as-atomized Ti-6Al-4V powder.

D50Fine TC4/μm	D50/μm	D50Corse TC4/μm	D50/μm	Φ
0–15	15	53–150	100	0.15
15–53	40	53–150	100	0.4
53–105	75	53–150	100	0.75
/	/	53–150	100	0

.

2.2. Surface Coating

In the paper, the Ni-P coated GNFs (noted as GNF(Ni)) were prepared by impregnation reduction. Firstly, 0.2 g GNFs were immersed in CH3COCH3 solution under ultrasonic treatments for 30 min to improve CNF dispersion. The roughened GNFs were immersed in the sensitization solution (SnCl2 10 g/L + HCl 40 mL/L) and activation solution (PdCl2 0.5 g/L + HCl 25 mL/L) in turn and stirred for 30 min. Remarkably, GNFs must be cleaned completely with deionized water after each transfer. Pre-GNFs were immersed in solution with 25 g/L NiSO₄, 45 g/L C₆H₅Na₃O₇ (sodium citrate) and 13 g/L NaH₂PO₃ (sodium hypophosphite monohydrate), stirred for 10 min and adjusted to pH 7–10 with ammonium hydroxide solution. Ni-P coated GNFs were obtained by drying at a low temperature (60 °C, 24 h).

2.3. Preparation of TC4/GNF(Ni) Composites

The ball-milling with alcohol was used for mixing the powders, and HEBM with ethyl alcohol as a process control agent was used for dispersion GNFs uniformly into the TC4 powders, the 0.05 g GNF(Ni) and 20 g TC4 powders of three sizes (0–15 μm, 15–53 μm, 53–150 μm) were charged into an agate jar. The powder was ball-milled at 300 rpm for 2 h, with the agate milling ball-to-powder weight ratio as 5:1. Finally, GNF(Ni)/TC4 was prepared by reduction at 900 °C for 2 h under the Ar atmosphere.

2.4. Characterization of TC4/GNF(Ni) Composites

The microstructures of both the powders and composites were meticulously examined using a state-of-the-art scanning electron microscope (SEM, FEI, Scios, Waltham, MA, USA), which was coupled with energy-dispersive X-ray spectroscopy (EDS) for further analysis. Prior to observation, the SEM specimen underwent polishing using SiC paper and colloidal silica (OP-S) and then etching with Kroll solution (5 mL HF, 10 mL HNO₃, and 70 mL H₂O). The phase composition of the composites was determined using X-ray diffraction (XRD, DMAX-RB, Rigaku, Japan), which utilized Cu (Kα) monochromatic radiation. In order to gain a better understanding of the structure of the GNFs, Raman spectrometry was conducted using a Renishaw inVia at a wavenumber range of 1200 to 3000 cm⁻¹ with an excitation laser of 514 nm at a laser power of 5.63 mW. Finally, the transmission electron microscopy (TEM, TecnaiG22, FEI, Boston, MA, USA) technique was applied to obtain a clear image of the interface between the GNFs and TC4 matrix. The compression specimens had dimensions of Ø 6 mm × 6 mm and were compressed at room temperature using a universal tensile testing machine (ETM205D, Vance, Shenzhen, China) at a constant strain rate of 0.5 mm/min.

3. Results

3.1. Characterization of Powder Surface

Powder metallurgy is a highly effective method for manufacturing titanium matrix composites reinforced with particles. The process involves several key steps. Initially, graphene is mixed with titanium powder of varying sizes using ball-milling. This allows for precise control over the content ratio of the reinforcing particles. Vacuum sintering is then employed to consolidate the mixture and form composite blanks. Additional processing techniques such as hot extrusion or rolling are subsequently applied to further refine and shape the titanium matrix composites.

To cater to specific requirements, three distinct ball-milling processes were developed and are shown in Figure 3. Process A enables the adjustment and regulation of the content ratio of reinforcing particles, making it suitable for producing graphene-reinforced titanium matrix composites with a uniform particle size. Process B, on the other hand, employs the same ball-milling time and speed as Process A but incorporates only low-particle-size titanium powder. This approach facilitates the production of two-scale composite powder. As for Process C, it provides effective control over the reinforcing phase content and powder proportion. However, this process is more complex and necessitates secondary processing steps to enhance the uniformity of the composite powder.

SEM images of the 0.5 wt% GNF(Ni)/TC4 powders processed with ball-milling are presented in Figure 4. These images reveal notable differences in powder distribution compared to the powders shown in Figure 2a. Specifically, a larger amount of GNF(Ni) is observed to be dispersed and adhered to the surfaces of TC4 powders after 3 h of ball-milling (Figure 4). While the powders in Figure 2a exhibit good sphericity, the powders in Figure 4 display a more regular shape, with CNFs uniformly grown on the surface of the powders due to the ball-milling process. It is worth noting that even when the content of the reinforcing agent is the same, the distribution of powders can vary depending on the mixing method employed. The effectiveness of the mixing method determines whether the desired level of mixing is achieved.

Based on the aforementioned analysis, the mismatch coefficient of the sintered specimen was calculated by analyzing the changes in the mixed powders. The calculation was done using the following equation:

Φ = D50_{FINE TC4}/D50_{CORSE TC4}

(1)

Mismatch coefficient Φ equals the ratio of the median particle size D50 of the titanium alloy. For this study, we selected Ti-6Al-4V alloy powders with varying particle sizes (preliminarily determined as 0, 0.15, 0.4, 0.75). The mismatch coefficient of the configuration was determined to be Φ = 0.15, 0.4, and 0.75 for the mixed powders. After sintering, a mixed material with a GNF mass fraction of 0.5 wt.% was acquired.

3.2. Characterization of the GNF(Ni)/TC4 Composites

In Figure 5, the surface morphology of GNF(Ni)/TC4 composites that underwent hot-press sintering at different mismatches and sintering temperatures is presented. Figure 5a,d show that GNF(Ni) are uniformly and flawlessly distributed within the Ti matrix, indicating the absence of visible holes or defects in the composite’s microstructure. This uniform distribution suggests the occurrence of an interface reaction between graphene and the powder, as particle sizes (0–15 μm) have already melted and agglomerated. Interestingly, at a mismatch of Φ0.4, shown in Figure 5b,e, isolated large particles connect through the powder with particle sizes ranging from 15 to 53 μm, resulting in the formation of a high-density, ideal network structure. Similarly, at a mismatch of Φ0.75, shown in Figure 5c,f, a continuous, evenly sized network structure emerges. This indicates the possible difference of purity and porosity within these network structure.

Figure 5g illustrates the density of 0.5 wt% GNF(Ni)/TC4 composites at different mismatches and sintering temperatures. As the mismatch coefficient increases, both the relative density and actual density of the composites decrease. Notably, the composite with a mismatch coefficient of Φ0.75 experiences a significant drop in relative density, which can be attributed to the presence of defects and particle clustering within the structure. Conversely, the density of the composites increases with higher sintering temperatures. It is important to highlight that the relative density of the composites exceeds 98%, indicating a minimal presence of defects. Additionally, the spectral line of the composites is significantly higher compared to other alloys, suggesting superior density characteristics.

In Figure 6, the XRD patterns of the GNF(Ni)/TC4 composites with different mismatches are presented. The patterns reveal the presence of Ti, TiC, and α-Ti phases in both composites. The TiC phase is typically formed during the hot-pressing process [19]. The composites exhibit diffraction peaks corresponding to Ti, with a notable peak observed at 42°, corresponding to the TiC (200) plane. Furthermore, the two-scale composites exhibit distinct α-Ti phase peaks, indicating a well-organized material structure. The α-Ti and GNFs bands display progressive broadening as the composite mismatch coefficient increases. It is important to note that the intensity of the Ni peaks in the GNF(Ni)/TC4 composites is too low to be detected, suggesting a low concentration of Ni in the composites.

The presence of TiC and GNFs in the GNF(Ni)/TC4 composites was further confirmed through XPS (X-ray photoelectron spectroscopy) and Raman spectroscopy analysis.

Table 3. The detailed Raman spectra results of the composites.

Φ	D Band	G Band	2D	ID/IG
0	1355.36	1579.77	2728.51	0.214
0.15	1355.56	1582.54	2728.57	0.274
0.4	1344.44	1580.95	2723.81	0.238
0.75	1358.73	1580.95	2711.11	0.547

provides the strengths of the D bands, G bands, and 2D bands of GNFs in the composites, as observed in the Raman spectra. In Figure 2d, the Raman spectra of the 0.5 wt% GNF(Ni)/TC4 composites with different mismatches (Φ = 0.15, 0.4, and 0.75) are shown, along with the spectra of the raw composite (mismatch 0). The ID/IG ratios of GNF(Ni)/TC4 composites are commonly used to assess the defects and quality of carbon materials. As the mismatch coefficient increases from 0 to 0.75 (Figure 3c), the ID/IG ratios of the GNF(Ni)/TC4 composites increase from 0.214 to 0.546, indicating the severe damage and amorphization of the GNFs. However, the ID/IG ratio of the GNF(Ni)/TC4 composite with a mismatch of 0.214 suggests that the structure of GNFs has not changed significantly at this mismatch. XPS analysis was employed to investigate the influence of different mismatch coefficients on the GNF(Ni)/TC4 composites and to provide information on the interface. In addition, high-resolution spectra of Ti-C bonding are presented in Figure 6. It is important to note that the Ni peak can be detected in the Ti spectral line, indicating the presence of Ni-P coated GNFs and the successful control of interfacial reactions. This is a critical aspect of fabricating high-performance titanium matrix composites for elevated temperature applications.

XPS analysis was performed to determine the composition of the 0.5 wt% GNF(Ni)/TC4 composites and validate the presence of defects in the GNF structure under different mismatch coefficients. In Figure 7, the C 1s characteristic peaks of the composites are observed after sintering. The fitting results revealed specific peaks at 284.8 eV, 286.2 eV, and 288.2 eV, which corresponded to the C 1s levels of C=C sp2, C–C sp3, and C–O, respectively. These findings were consistent with the XPS data of carbide and carbon-related functional groups.

Furthermore, oxygen-containing species were detected, indicating the presence of oxygen-containing polar groups adsorbed on the surface of the composites. The sp2/sp3 ratio is often used to evaluate the defects and quality of carbon materials in GNF(Ni)/TC4 composites. The hybrid strength of sp2 and sp3 varied with the increasing mismatch coefficient in the composites. The sp2/sp3 ratios of the GNF(Ni)/TC4 composites decreased from 4.72 to 3.21 as the mismatch coefficient increased from 0 to 0.75, as shown in

Table 4. The detailed XPS spectra of the GNF(Ni)/TC4 composites.

Φ	SP2	SP3	C–O	ISP2/ISP3
0	284.8	286.2	288.2	4.034
0.15	284.8	286.1	288.2	4.72
0.4	284.8	286.2	288.2	4.58
0.75	284.8	285.8	288.2	3.21

. This decrease in the sp2/sp3 ratio suggests that the GNPs were severely damaged and underwent amorphization due to higher mismatch coefficients.

In Figure 8, an image of a GNF(Ni)/TC4 composite with a network structure is presented. The image clearly shows the presence of flaky or equiaxial α-Ti, along with an equiaxial network structure of the composite where the reinforced phase graphene is distributed at the corners. This network structure is a result of powder extrusion and particle agglomeration during the fabrication process. Figure 8c reveals that the network structure in the 0.5 wt% GNF(Ni)/TC4 composites becomes smaller when the network structure is introduced. The width at the boundary narrows, leading to an irregular network structure due to the uneven force exerted during the formation of the double-scale composites.

Moreover, high-magnification photography shows that the size of the reinforced phase is at the nanometer level. The EDS analysis in Figure 8d indicates that the carbon element is primarily distributed in the graphene, and the sediment on its surface only contains Ti and C. Based on the XRD results, it can be identified that the enhanced phase at the boundary of the network structure is TiC, which is generated in situ during the fabrication process. Lastly, the EDS data reveal that the atomic percentages of the Ni element and P element are 1% and 0.31%, respectively, in the composite.

Based on the analysis, it can be concluded that the 0.4 GNF(Ni)/TC4 composites exhibit good strength–plasticity compatibility. The interface micro-regions of the composite are further characterized in Figure 9. The TEM image shows the interface between two adjacent Ti crystals (matrix) with some particles distributed at the interface. The in situ generated TiC crystals and the interface of GNFs were closely bonded. Figure 9c presents TEM images along with the corresponding high-resolution TEM (HRTEM) image and fast Fourier transform (FFT) results, confirming the presence of GNFs. The HRTEM image reveals contorted lattice edges of the GNFs, indicating the presence of amorphous carbon at the interface. These findings are attributed to the mismatch in the coefficients of thermal expansion (CTEs) between the GNFs and Ti matrix. Moreover, characteristic atom fringes of GNFs are still observable. The fringes in the marked region in Figure 9c exhibit a crystal plane spacing of 0.34 nm, consistent with previous reports. Additionally, lattice distortion is observed in the crystalline graphite near the interface, mainly due to the mismatch. Furthermore, several amorphous regions can be seen in the graphite region, as depicted in Figure 9. The high temperature, acid base, and ultrasonic treatment of the GNFs result in the disordering and reconstruction of carbon atoms on the GF surface, which is consistent with the Raman spectroscopy results.

3.3. Mechanical Properties of the GNF(Ni)/TC4 Composites

Figure 10a illustrates the engineering properties of GNF(Ni)/TC4 composites with varying coefficients of mismatch. As the mismatch coefficient increases, the yield strength and fracture strain of the composites exhibit an initial increase followed by a decrease. The 0.4 GNF(Ni)/TC4 composite demonstrates the highest compressive strength at 1316.45 ± 3.4 MPa, which is 24.75% greater than the 1055.29 ± 1.1 MPa of the Φ0 GNFs (Ni)/TC4 composite. Figure 10b depicts the compressive strength and strain of the GNF(Ni)/TC4 composites with mismatch coefficients of Φ0, Φ0.15, Φ0.4, and Φ0.75. The compressive strength of the GNF(Ni)/TC4 composites are determined as 2051.5 ± 0.8, 2195.9 ± 1.0, and 1917.4 ± 0.9 MPa, respectively. Meanwhile, the strains of the Φ0.15, Φ0.4, and Φ0.75 GNF(Ni)/TC4 composites are 30.38%, 37.27%, and 30.21%, respectively. It is worth noting that the compressive strength of the mismatched composite is higher than that of the single-scale GNF(Ni)/TC4 composite with the same volume fraction of the reinforcement phase.

Figure 10b displays the mechanical properties of composites with varying mismatch coefficients. Notably, the Φ0.15, Φ0.4, Φ0.75, and 0 GNF(Ni)/TC4 composites exhibited higher hardness values of 383 HV, 382 HV, 364.1 HV, and 346.05 HV, respectively. This can be attributed to the smaller Ti grains present in the composites when compared to traditional titanium alloys. However, as the gradation ratio increased, the hardness of the two-scale composite decreased due to the presence of defects and particle aggregation, ultimately resulting in performance variances within the composite network.

3.4. Fracture Morphology and Crack Propagation

Figure 11 displays the compressive fractographies of GNF(Ni)/TC4 composites. The GNF(Ni)/TC4 composites with a network microstructure exhibit brittle fracture behavior, primarily through particle shedding, which can be attributed to particle aggregation and interlocking. On the other hand, the GNFs/TC4 composites show interface detachment between the Ti matrix and the GNF(Ni) particles. The compressive fractographies of GNF(Ni)/TC4 composites in Figure 11c,d reveal the presence of fractured GNF(Ni) particles, marked by yellow arrows, which is consistent with previous studies.

Fractures with more ductile characteristics, such as tendon-like features, indicate greater plasticity compared to cleavage fracture surfaces of brittle fractures. Upon closer examination, it can be observed that the dimple depth on the surface of the Φ0.75 GNF(Ni)/TC4 composite is deeper than that of other samples. The decreased mismatch in the Φ0.75 composite results in increased brittleness, causing some areas of the fracture morphology to exhibit large, flat surfaces resembling cleavage steps.

4. Discussion

The investigation presented above sheds light on the relationship between the microstructure and mechanical properties of GNF(Ni)/TC4 composites, particularly in relation to the coefficient of mismatch. To further explore the strengthening mechanism of these composites, an in-depth analysis was conducted on the 0.5 wt.% GNF(Ni)/TC4 composite. Figure 12 provides EBSD-IPF plots and grain size distribution data for composites with mismatch coefficients of Φ0 and Φ0.4. Notably, the EBSD-IPF diagram reveals a transformative shift in the matrix grains from a coarsened α-Ti morphology to a finely isometric structure. This transformation plays a vital role in reducing particle size, increasing specific surface area, and decreasing the depleted area of the reinforcement phase, thereby facilitating significant grain refinement. It is noteworthy that despite maintaining consistent overall reinforcement-phase content, the reduction in the depleted area of the reinforcement phase effectively restricts the growth of coarse grains [40].

Additionally, it is observed that the width of the mesh interface remains relatively unaffected by changes in the reinforcing phase content, as it is primarily governed by the length of the reinforcing phase. Moreover, when analyzing the average grain size distribution, it becomes evident that the reduction in the depletion region of the reinforcement phase leads to a substantial refinement of the grain size from 19.32 μm to 9.45 μm, underscoring the profound influence of grain refinement on the mechanical properties of the composite.

The microstructure of titanium composites is significantly influenced by the coefficient of mismatch. Non-mismatched composites exhibit coarse and directional α-Ti morphology, with a predominance of non-oriented equiaxed grains and the effective elimination of original β grains. The presence of TiC particles distributed throughout the microstructure and their encapsulation of GNFs contribute to enhanced load-bearing capacity under stress. Conversely, an increase in powder particle size leads to grain growth, adversely affecting the mechanical properties of the composite. This comparison highlights the importance of managing the fine-grain effect resulting from powder particle-size mismatch. This underscores the potential advantages of well-designed internal grain structure and interface in enhancing the mechanical properties of sintered titanium matrix composites (TMCs). Additionally, the agglomeration of the reinforcement phase at the interface is observed, which can be attributed to the presence of pores between the powders during the sintering process, leading to the agglomeration of graphene in these pores. However, as the size of the matrix particles increases, the depleted zone of the reinforcing phase inside the matrix grows in the prepared mesh structure, causing the grains to enlarge, as analyzed using electron backscatter diffraction (EBSD) techniques.

Figure 13 presents a schematic diagram illustrating the 0.5 wt.% GNF(Ni)/TC4 composites. To further enhance the optimal reinforcement-phase content and consequently increase the strength level of the mesh composites, it is necessary to increase the content of the reinforcing phase by reducing the particle size of the matrix particles, thus increasing their specific surface area. However, it should be noted that as the size of the matrix particles continues to decrease, the size of the depleted zone of the reinforcing phase inside the matrix gradually diminishes in the prepared mesh composites. This reduction in the depleted zone has implications for the plasticity of the composite [41]. Therefore, in addition to employing a single particle-size approach to fabricate the composites, TC4 titanium alloy powder with different particle sizes, specifically 15–53 μm and 53–150 μm, respectively, was utilized as the raw material to prepare dual particle-size-mismatched mesh composites, namely 0.5 wt.% GNF(Ni)/TC4 (Φ0.4) composites. By comparing these results with those presented in Figure 10, it becomes apparent that the dual particle-size-mismatched composites exhibit the largest reinforcing-phase depletion zone while also displaying a smaller grain organization. Hence, for this titanium matrix composite with a mesh structure, the mechanical properties are not solely governed by the reinforcing phase but are also influenced by the matrix microstructure.

5. Conclusions

In this study, the dispersion of graphene nanoflakes (GNFs) in a titanium (Ti) matrix was improved by electroless nickel plating, forming a high-quality nickel-plating layer on the GNF surfaces. The GNF(Ni)/Ti composites were fabricated using milling and hot-pressing sintering (HPS) processes, with varying coefficients of particle-size mismatch. The objective was to investigate the interfacial structure evolution and mechanical properties of the composites. The main findings and conclusions can be summarized as follows:

(1) At a sintering temperature of 900 °C, the composite samples exhibited favorable densification and a strong bond between the matrix and the reinforcement phase. Increasing particle-size mismatch led to a more uniform mesh-like structure, indicating enhanced local reinforcement but uneven distribution of the reinforcement phase.

(2) Microstructure analysis revealed the presence of GNFs near the interface, as confirmed by diffraction patterns. Lattice distortion was observed at the edge of crystalline graphene due to particle-size mismatch, inducing thermal stress during sintering and resulting in the distortion of the carbon lattice structure. In situ formation of titanium carbide (TiC) particles occurred within the composite material, and the presence of nickel (Ni) and phosphorus (P) elements was detected near the interface. These results indicated that the addition of Ni reduced the interface reaction in the composite.

(3) Among the investigated compositions, the 0.5 wt.% GNF(Ni)/TC4 (Φ0.4) composite exhibited the best performance, with significantly improved yield strength, compressive strength, and strain. The measured values were 1316.45 ± 3.4 MPa, 2195.9 ± 1.0 MPa, and 42.1 ± 0.3%, respectively, representing increases of 24.75%, 5.9%, and 22.8% compared to the base composite.

(4) Electron backscatter diffraction (EBSD) analysis demonstrated that particle-size mismatch caused a transformation of the matrix grains from a rough α plate-like morphology to a fine equiaxed microstructure. This transformation synergistically enhanced the mechanical and toughness properties of the composite material. Fracture morphology observations indicated that initial fractures occurred along the interface, maintaining a high reinforcement effect. With increasing coefficients of mismatch, the depletion area of the reinforcement phase also increased, leading to enhanced resistance against crack bending exceeding transcrystalline resistance.

Overall, these findings provide valuable insights into the evolution of interfacial structure and the mechanical properties of GNF(Ni)/Ti composites. The results highlight the significance of particle-size mismatch and the incorporation of Ni coating in optimizing the performance of the composites.