Preparation and Characterization of In Situ (TiC-TiB2)/Al-Cu-Mg-Si Composites with High Strength and Wear Resistance

This study involved the preparation and characterization of in situ (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites with excellent mechanical and abrasive wear properties. The composites were synthesized in an Al-Ti-B4C system by combining combustion reaction synthesis with hot-pressed sintering and hot extrusion. The in situ TiB2 and TiC particles were of multi-scaled sizes ranging from 20 nm to 1.3 μm. The TiB2 and TiC particles effectively increased the yield strength (σ0.2), ultimate tensile strength (σUTS), hardness (HV), and abrasive wear resistance of the composites. The 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite exhibited the highest σ0.2 (569 MPa), σUTS (704 MPa) and hardness (286 HV), which were 74%, 51% and 110% higher than those of the matrix alloy, respectively. Compared with the matrix alloy, the abrasive wear resistance of the 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite was increased by 4.17 times under an applied load of 5 N and Al2O3 abrasive particle size of 13 µm. Micro-ploughing and micro-cutting were the main abrasive wear mechanisms for the Al-Cu-Mg-Si alloy and the composites.

Micron-sized ceramic particles in Al matrix composites have been reported to increase the strength and hardness of the composites [16,17]. Ceramic particles also effectively resist the penetration of abrasives and reduce the abrasive wear of the composites [18,19]. However, the strength and plasticity of Al matrix composites reinforced by nano-sized ceramic particles are higher than for composites reinforced by micron-sized particles [20]. Nanosized ceramic particles have been shown to significantly increase the plastic deformation resistance of the composites [6,21]. Based on our previous investigations, Al matrix composites reinforced with bi-modal-sized ceramic particles showed an improved distribution of ceramic particles and exhibited excellent elevated temperature mechanical properties compared to composites reinforced with single-sized particles [22,23]. Zhang et al. reported that the σ 0.2 (358 MPa) and σ UTS (585 MPa) for hybrid-sized SiC/Al-Cu matrix composite (40 nm + 15 µm) were higher than those for single nano-sized SiC/Al-Cu composite (312 MPa, 512 MPa) and single micron-sized SiC/Al-Cu (302 MPa, 503 MPa) composite [22].  Figure 2 shows the process flowchart for the preparation of in situ (TiC-TiB2)/Al-Cu-Mg-Si composite. As shown in Figure 2b, the powders were ball-milled for 30 h at 50 rpm. The grinding bodies used were ZrO2 balls. A ball-to-powder weight ratio of 50:1 was used. The ball-milling treated powders used can be seen in Figure 1d. The mixed powders were   Figure 2 shows the process flowchart for the preparation of in situ (TiC-TiB 2 )/Al-Cu-Mg-Si composite. As shown in Figure 2b, the powders were ball-milled for 30 h at 50 rpm. The grinding bodies used were ZrO 2 balls. A ball-to-powder weight ratio of 50:1 was used. The ball-milling treated powders used can be seen in Figure 1d. The mixed powders were cold-pressed into cylinders (ϕ45 mm × 30 mm), as shown in Figure 2c. The combustion synthesis process of the cylinder was performed in a vacuum furnace at a heating rate of 30 K/min, as shown in Figure 2d. When the measured temperature suddenly increased, the combustion synthesis reaction started, and the reactant cylinder was rapidly pressed by an axial stress of 50 MPa for 20 s. As the furnace temperature decreased, the sintered (TiC-TiB 2 )/Al-Cu-Mg-Si composites were obtained. When the content of the Ti-B 4 C reactants increased from 10 wt.% to 40 wt.%, the maximum combustion temperature of the reaction system increased from 1580 K to 1800 K. The Al-4.7Cu-0.32Mg-0.44Si matrix alloy used for comparison was prepared using a hot-pressed sintering method (873 K for 2 h). Before hot extrusion, the Al-4.7Cu-0.32Mg-0.44Si matrix alloy and composites were homogenized at 758 K for 12 h. The Al matrix alloy and composites were hot-extruded at 833 K with an extrusion ratio of 19:1, as shown in Figure 2e. The extruded sheets of matrix alloy and composites were subjected to T6 heat treatment, which involved solution treatment (778 K, 2 h), quenching in water, and artificial aging (433 K, 18 h), as shown in Figure 2f.  XRD (D/Max 2500PC, Rigaku, Tokyo, Japan), an SEM (Tescan vega3 XM, Tescan, Brno, Czech Republic) equipped with energy dispersive spectroscopy (EDS) and Oxford NordlysMax EBSD detector, a field emission scanning electron microscope (FESEM, JSM 6700F, JEOL, Tokyo, Japan), and TEM (JEM 2100F, JEOL, Tokyo, Japan) were used for the detailed characterizations. The step size in EBSD characterization was 0.2 μm. The sam- XRD (D/Max 2500PC, Rigaku, Tokyo, Japan), an SEM (Tescan vega3 XM, Tescan, Brno, Czech Republic) equipped with energy dispersive spectroscopy (EDS) and Oxford NordlysMax EBSD detector, a field emission scanning electron microscope (FESEM, JSM 6700F, JEOL, Tokyo, Japan), and TEM (JEM 2100F, JEOL, Tokyo, Japan) were used for the detailed characterizations. The step size in EBSD characterization was 0.2 µm. The samples were first polished with SiC grinding paper and then electro-polished with 10% HClO 4 -90% ethanol solution at 253 K. About 1000 particles in the FESEM images were randomly selected; the maximum sizes of the particles were used to calculate the size distribution of the prepared TiC-TiB 2 particles.
The samples used for tensile tests and abrasive wear tests were taken from the Al-4.7Cu-0.32Mg-0.44Si alloy and composite plates along the extrusion direction (ED). Tensile mechanical tests were performed at room temperature using an MTS-810 machine (MTS Systems Corporation, Minneapolis, MN, USA) with a strain rate of 3 × 10 −4 s −1 . Dog-bone samples (10 mm in gauge length, 4 mm in width, and 2 mm in thickness) were used for tensile tests. The abrasive wear experiments were carried out on a wheeled abrasion tester (MLH-30, Zhangjiakou Chengxin Test Equipment Manufacturing Co., Ltd., Zhangjiakou, China) at 298 K [27]. The dimensions of the sample used for wear-testing were 30 mm in ED, 5 mm in the transverse direction (TD), and 5 mm in the normal direction (ND), respectively. Applied loads of 5, 15 and 25 N and Al 2 O 3 abrasive papers with grits of 360 (~40 µm), 600 (~23 µm) and 1000 (~13 µm) were selected to characterize the abrasive wear behaviors of the Al-4.7Cu-0.32Mg-0.44Si alloy and composites. The total sliding distance was 120 m. The volume wear rates (WR) were calculated by dividing the mass loss by the actual density (ρ actual ) of the samples. The ρ actual was measured by Archimedes' method. The mass loss of the sample was measured by a high-precision electronic balance (0.0001 g). The micro-hardness tests were performed using a Vickers hardness tester (1600-5122VD Buehler, Feasterville, PA, USA), using an applied load of 5 N and a dwell time of 10 s for 10 times. The abrasive surface roughness (Rt) of samples was measured using a laser scanning confocal microscope (OLYMPUS LEXT OLS3000, Olympus, Tokyo, Japan). The Rt value represents the height difference between the highest point and the lowest points of the worn surface. Figure 3 shows the XRD analysis results of the prepared in situ (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites with different TiC and TiB 2 particle content. The reaction products of the Al-Ti-B 4 C systems mainly consisted of Al, CuAl 2 , TiC and TiB 2 phases. An intermediate Al 3 Ti phase was observed in the 10 wt.% and 20 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites due to insufficient reaction of the Al-Ti-B 4 C systems. The intensity of the detected TiC and TiB 2 phases increased with increasing TiC and TiB 2 particle content from 10 wt.% to 40 wt.%.

Results and Discussion
As shown in Figure 4a-d, the TiC-TiB 2 particles in (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites showed a river-like distribution, which improved as the TiC and TiB 2 particle content increased. The 30 wt.% and 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites showed more uniform distribution of TiC and TiB 2 particles. Figure 4e-l show the morphologies and the size-distribution maps of the extracted TiC and TiB 2 particles. The synthesized TiC and TiB 2 particles exhibited a large size span from 20 nm to 1.3 µm. With TiC and TiB 2 particle content increasing from 10 wt.% to 40 wt.%, the average particle size exhibited obvious increments (from 95 nm to 223 nm), while the percentage of nano-sized TiC-TiB 2 particles decreased significantly (from 82.1% to 3.0%). In the combustion reaction synthesis process of the Al-Ti-B 4 C system, an Al-Ti liquid phase formed initially, followed by an Al-Ti-B-C liquid phase that formed with the diffusion of [ [28][29][30]. The temperature for the generation of TiC and TiB 2 particles was the maximum combustion temperature of the reaction system. As the content of the Ti-B 4 C reactants in the Al-Ti-B 4 C system increased from 10 wt.%, to 20 wt.%, to 30 wt.% to 40 wt.%, the maximum combustion temperature of the reaction system increased from 1580 K, to 1642 K, to 1702 K to 1800 K, respectively. Due to the exponential relationship between the crystal growth and the combustion temperature [30], the size of the precipitated TiC-TiB 2 particles increased with increasing combustion temperature. This caused an increase in Ti-B 4 C reactant content resulting in an increase in the size of the precipitated TiC-TiB 2 particles. It is of note that the lower diffusion rate of [B] in the molten Al was not favorable for generation of homogenously distributed [B] rich regions [28,29]. The violent reaction between [Ti] and [B] released a large amount of heat, resulting in substantial growth of precipitated TiB 2 particles. Hexagonally shaped TiB 2 particles of micron and sub-micron size were identified, as shown in Figure 4e Figure 3 shows the XRD analysis results of the prepared in situ (TiC-Ti 0.32Mg-0.44Si composites with different TiC and TiB2 particle content. The r ucts of the Al-Ti-B4C systems mainly consisted of Al, CuAl2, TiC and TiB2 p termediate Al3Ti phase was observed in the 10 wt.% and 20 wt.% (TiC-Ti 0.32Mg-0.44Si composites due to insufficient reaction of the Al-Ti-B4C system sity of the detected TiC and TiB2 phases increased with increasing TiC and content from 10 wt.% to 40 wt.%. As shown in Figure 4a-d, the TiC-TiB2 particles in (TiC-TiB2)/Al-4.7Cu-0 composites showed a river-like distribution, which improved as the TiC and content increased. The 30 wt.% and 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0 sites showed more uniform distribution of TiC and TiB2 particles. Figure 4 morphologies and the size-distribution maps of the extracted TiC and TiB2 synthesized TiC and TiB2 particles exhibited a large size span from 20 nm to TiC and TiB2 particle content increasing from 10 wt.% to 40 wt.%, the averag exhibited obvious increments (from 95 nm to 223 nm), while the percentage TiC-TiB2 particles decreased significantly (from 82.1% to 3.0%). In the combu synthesis process of the Al-Ti-B4C system, an Al-Ti liquid phase formed initi   Figure 5b,c show the IPF and recrystallized microstructure maps of area A in Figure 5a. The high-angle grain boundaries (HAGBs, ≥15 • ) and low-angle grain boundaries (LAGBs, 2-15 • ) are marked by black lines and gray lines, respectively. Figure 5c shows the deformed grains, sub-grains and recrystallized grains of the composite, which are marked by red, yellow and blue colors, respectively. The composite had a bimodal microstructure containing large-sized sub-grains and fine recrystallized grains. The α-Al grain sizes near the TiC-and TiB 2-particle-rich regions (0.3 µm) were much smaller than those near the lean regions (34.5 µm). Dynamic recrystallization could not be avoided during hot extrusion. The nano-sized and sub-micro-sized TiC and TiB 2 particles distributed on the grain boundaries (GBs) may inhibit the rotation and merging of GBs and the growth of α-Al grains [9,30]. Accumulated dislocations in the inner α-Al grains were caused by the TiC and TiB 2 particles and then LAGBs formed, as seen in Figure 5b. It is possible that multi-sized α-Al grains result in superior mechanical properties of the composites [15].

Results and Discussion
15, x FOR PEER REVIEW 6 of 16 the precipitated TiC-TiB2 particles increased with increasing combustion temperature. This caused an increase in Ti-B4C reactant content resulting in an increase in the size of the precipitated TiC-TiB2 particles. It is of note that the lower diffusion rate of [B] in the molten Al was not favorable for generation of homogenously distributed [B] rich regions [28,29]. The violent reaction between [Ti] and [B] released a large amount of heat, resulting in substantial growth of precipitated TiB2 particles. Hexagonally shaped TiB2 particles of micron and sub-micron size were identified, as shown in Figure 4e-h.  Figure 5a shows an SEM image for the 10 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite after T6 heat treatment. Figure 5b,c show the IPF and recrystallized microstructure maps of area A in Figure 5a. The high-angle grain boundaries (HAGBs, ≥15°) and low-angle grain boundaries (LAGBs, 2-15°) are marked by black lines and gray lines, respectively. Figure 5c shows the deformed grains, sub-grains and recrystallized grains of the composite, which are marked by red, yellow and blue colors, respectively. The composite had a bimodal microstructure containing large-sized sub-grains and fine recrystallized grains. The α-Al grain sizes near the TiC-and TiB2-particle-rich regions (0.3 μm) were much smaller than those near the lean regions (34.5 μm). Dynamic recrystallization could not be avoided during hot extrusion. The nano-sized and sub-micro-sized TiC and TiB2 particles distributed on the grain boundaries (GBs) may inhibit the rotation and merging of GBs and the growth of α-Al grains [9,30]. Accumulated dislocations in the inner α-Al grains were caused by the TiC and TiB2 particles and then LAGBs formed, as seen in Figure 5b. It is possible that multi-sized α-Al grains result in superior mechanical properties of the composites [15]. The tensile mechanical properties, hardness and actual density of the Al-4.7Cu-0.32Mg-0.44Si alloy and in situ (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites are presented in Figure 6 and Table 2. The hot-extrusion process effectively consolidated the composites to produce a high relative density (100%). As has been reported previously, the dense composites showed excellent strength and hardness [15,19,[31][32][33]. The composites The tensile mechanical properties, hardness and actual density of the Al-4.7Cu-0.32Mg-0.44Si alloy and in situ (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites are presented in Figure 6 and Table 2. The hot-extrusion process effectively consolidated the composites to produce a high relative density (100%). As has been reported previously, the dense composites showed excellent strength and hardness [15,19,[31][32][33]. The composites exhibited superior σ 0.2 , σ UTS and HV values to those of the matrix alloy. The σ 0.2 , σ UTS and HV values of the composite increased with increasing TiC-TiB 2 content. Orowan strengthening effects and grain-boundary pinning effects were the main strengthening mechanisms for the prepared (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites. As the TiC and TiB 2 particle content increased and the particle distribution improved, the plastic deformation resistance of the composites increased. The 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite showed the highest σ 0.2 (569 MPa), σ UTS (704 MPa) and HV (286 HV) values, approximately 74%, 51% and 110% higher, respectively, than the corresponding values for the Al-4.7Cu-0.32Mg-0.44Si alloy.

treatment.
The tensile mechanical properties, hardness and actual density of the Al-4.7 0.32Mg-0.44Si alloy and in situ (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites are p sented in Figure 6 and Table 2. The hot-extrusion process effectively consolidated the co posites to produce a high relative density (100%). As has been reported previously, dense composites showed excellent strength and hardness [15,19,[31][32][33]. The compos exhibited superior σ0.2, σUTS and HV values to those of the matrix alloy. The σ0.2, σUTS HV values of the composite increased with increasing TiC-TiB2 content. Orowan streng ening effects and grain-boundary pinning effects were the main strengthening mec nisms for the prepared (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites. As the TiC TiB2 particle content increased and the particle distribution improved, the plastic de mation resistance of the composites increased. The 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32M 0.44Si composite showed the highest σ0.2 (569 MPa), σUTS (704 MPa) and HV (286 HV) ues, approximately 74%, 51% and 110% higher, respectively, than the corresponding ues for the Al-4.7Cu-0.32Mg-0.44Si alloy.    Figure 7a shows a TEM microstructure image of the 10 wt.% (TiC-TiB 2 )/Al-Cu-Mg-Si composite following the hot-extrusion process and heat treatment. Multi-sized TiC and TiB 2 particles, marked by circles, were located at the inner and grain boundaries (GBs) of α-Al grains. Hexagonally shaped submicron-sized TiB 2 particles of approximately 800 nm diameter are shown in Figure 7b. Figure 7c shows the corresponding HRTEM image of zone A in Figure 7b. The clean and continuous TiB 2 -Al interface helped to increase the strength of the composites [23,30]. Figure 7d shows an image of a cubic-shaped submicron TiC particle of about 300 nm width. An HRTEM image of zone B in Figure 7d is shown in Figure 7e, illustrating a clean TiC-Al interface with mismatched low lattice (2.5%). Figure 7f,g show the TEM image and corresponding nano-sized TiB 2 particle. Figure 7h,i show nano-sized TiC particles with spherical shapes. Figure 7g,i show the HRTEM images of zone C and zone D in Figure 7f,h, respectively. The TiC and TiB 2 particles, located in the interior of α-Al grains effectively restricted the dislocation movements, generated dislocation accumulations, and increased the strength (σ 0.2 , σ UTS ) of the composites. TiC and TiB 2 particles occurring at the GBs could restrain the grain boundary rotations of the α-Al grains and the plastic deformations of the composites, contributing to increased strength and hardness of the composites. zone A in Figure 7b. The clean and continuous TiB2-Al interface helped to increase the strength of the composites [23,30]. Figure 7d shows an image of a cubic-shaped sub-micron TiC particle of about 300 nm width. An HRTEM image of zone B in Figure 7d is shown in Figure 7e, illustrating a clean TiC-Al interface with mismatched low lattice (2.5%). Figure 7f,g show the TEM image and corresponding nano-sized TiB2 particle. Figure 7h,i show nano-sized TiC particles with spherical shapes. Figure 7g,i show the HRTEM images of zone C and zone D in Figure 7f,h, respectively. The TiC and TiB2 particles, located in the interior of α-Al grains effectively restricted the dislocation movements, generated dislocation accumulations, and increased the strength (σ0.2, σUTS) of the composites. TiC and TiB2 particles occurring at the GBs could restrain the grain boundary rotations of the α-Al grains and the plastic deformations of the composites, contributing to increased strength and hardness of the composites. composite after room-temperature tensile test. TEM images and corresponding HRTEM of (b,c) a sub-micron TiB2 particle, (d,e) a sub-micron TiC particle, (f,g) a nano-TiB2 particle, and (h,i) a nano-TiC particle.  composite after room-temperature tensile test. TEM images and corresponding HRTEM of (b,c) a sub-micron TiB 2 particle, (d,e) a sub-micron TiC particle, (f,g) a nano-TiB 2 particle, and (h,i) a nano-TiC particle. 0.32Mg-0.44Si composite was 2.43, 3.29, and 4.17 times higher than the Al-4.7Cu-0.32Mg-0.44Si alloy, respectively. Compared with the matrix alloy, the composites showed lower WRs and higher abrasive wear resistance. With increasing TiC and TiB2 particle content, the WRs of the composites clearly decreased. The 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite showed the lowest wear rates and best abrasive wear resistance at all the tested conditions. The relative abrasive resistance of 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite was 4.17 times higher than that of the Al-4.7Cu-0.32Mg-0.44Si alloy under a load of 5 N and an abrasive size of 13 μm.  It is of note that, with increasing load from 5 N, to 15 N to 25 N, with an Al 2 O 3 abrasive size of 23 µm, the WRs of the 31.9 wt.% (20 vol.%) nano-TiC/Al-4.7Cu-0.32Mg-0.44Si composite increased from 5.84, to 10.73 to 11.21 [34], while the WRs of the 30 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite increased from 2.11, to 5.86 to 9.60. This means that the dual-phased (TiC-TiB 2 )/Al-Cu-Mg-Si composite showed better abrasive wear resistance than the single-phased nano-TiC/Al-Cu-Mg-Si composite.
In the abrasive wear process, an increase in applied load would contribute to an increase in the contact zone between the specimens and the abrasives [35]. The penetration ability of the Al 2 O 3 abrasives increased as well as the micro-ploughing efficiency, which caused the increases in the WRs for the matrix alloy and composites. This was confirmed by the worn surface, as shown in Figure 9; with severe plastic deformation, the grooves became deeper and wider in the matrix alloy and the composites when the applied load increased. It can be inferred that micro-ploughing was the main abrasive wear mechanism for the Al-4.7Cu-0.32Mg-0.44Si alloy and the (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites at an Al 2 O 3 abrasive size of 40 µm. As indicated in Figure 9, the worn surface of the composites became smoother with less debris and formed shallower and narrower grooves as TiC-TiB2 particle content increased. As shown in Figure 10, with increase in TiC-TiB2 particle content, the measured maximum value of Rt decreased from 16.34 μm in the Al matrix alloy, 10.58 μm in the 10 As indicated in Figure 9, the worn surface of the composites became smoother with less debris and formed shallower and narrower grooves as TiC-TiB 2 particle content increased. As shown in Figure 10, with increase in TiC-TiB 2 particle content, the measured maximum value of Rt decreased from 16.34 µm in the Al matrix alloy, 10.58 µm in the 10 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite, 9.27 µm in the 20 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite, 8.24 µm in the 30 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite, to 7.53 µm in the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite. The hardness of the composites largely determines the penetration ability of the abrasive particles [19]. The TiC and TiB 2 particles effectively increased the hardness of the composites (from 192 HV to 286 HV), which reduced the penetration ability as well as the microploughing and micro-cutting ability of the Al 2 O 3 abrasives. The greater hardness of the composites contributed to a reduction in the wear rates and improvement in the abrasive wear resistance. The TiC and TiB 2 particles showed superior interfacial bonding with the Al matrix and effectively resisted the plastic deformation of the matrix. As discussed above, the distribution of the TiC and TiB 2 particles improved significantly, which also contributed to resistance to the non-homogenous plastic deformation of the Al matrix. Then, the micro-ploughing efficiency of the Al 2 O 3 abrasives decreased. The (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites with higher TiC-TiB 2 content showed greatly superior abrasive wear resistance and lower wear rates. As indicated in Figure 9a, micro-cutting and micro-ploughing occurred simult ously for the Al matrix alloy when tested under a 5 N load using a 13 μm Al2O3 abras With increase in the abrasive particle size, the micro-ploughing and micro-cutting ciency of the Al2O3 abrasive particles increased as well as the WRs of the specimens. W As indicated in Figure 9a, micro-cutting and micro-ploughing occurred simultaneously for the Al matrix alloy when tested under a 5 N load using a 13 µm Al 2 O 3 abrasive. With increase in the abrasive particle size, the micro-ploughing and micro-cutting efficiency of the Al 2 O 3 abrasive particles increased as well as the WRs of the specimens. When the abrasive size increased from 13 µm, to 23 µm to 40 µm, grooves and ridges with severe plastic deformation appeared, as shown in Figure 9k,p,u. This result indicated an abrasive wear mechanism transition from micro-cutting and micro-ploughing to micro-ploughing for the Al matrix alloy. The WRs increased from 0.69, to 1.80 to 2.51 for the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite with increase in the Al 2 O 3 abrasive sizes from 13 µm, to 23 µm to 40 µm. The abrasive wear resistance of the composite was 4.17 times higher than for the Al matrix when tested under a 5 N load with 13 µm abrasive. As shown in Figure 9o,t,y, the worn surface of the composites showed increase in the groove width and depth with increase in abrasive particle size. When the tested abrasive sizes increased from 13 µm, to 23 µm to 40 µm under a 5 N load, the abrasive wear mechanism of the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite changed from micro-cutting to micro-ploughing. Figure 11a,b show cross-sectional SEM images of the worn scar for the Al-4.7Cu-0.32Mg-0.44Si alloy and the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite, respectively, at 5 N load and with 13 µm Al 2 O 3 abrasive size. Abrasive damage was confined to the subsurface layer of the Al-4.7Cu-0.32Mg-0.44Si alloy and the composite specimens. As shown in Figure 11, the thickness of the cutting layer of the composite was significantly smaller than for the Al alloy.   Figure 12a,c, scratches, deep grooves and bridges were observed simultaneously on the worn surface of the Al-4.7Cu-0.32Mg-0.44Si alloy, representing indicators of micro-cutting and micro-ploughing abrasive wear mechanisms. Figure 12d shows the uniform distribution of the Ti element, reflecting the homogenous distribution of TiC and TiB2 particles in the composite. The TiC-TiB2 particles significantly constrained the plastic deformation of the Al matrix and increased the wear resistance of the composite. The composite presented a much smoother worn surface than the Al alloy, as indicated in Figure 12.   Figure 12a,c, scratches, deep grooves and bridges were observed simultaneously on the worn surface of the Al-4.7Cu-0.32Mg-0.44Si alloy, representing indicators of micro-cutting and micro-ploughing abrasive wear mechanisms. Figure 12d shows the uniform distribution of the Ti element, reflecting the homogenous distribution of TiC and TiB 2 particles in the composite. The TiC-TiB 2 particles significantly constrained the plastic deformation of the Al matrix and increased the wear resistance of the composite. The composite presented a much smoother worn surface than the Al alloy, as indicated in Figure 12.
representing indicators of micro-cutting and micro-ploughing abrasive wear mechanisms. Figure 12d shows the uniform distribution of the Ti element, reflecting the homogenous distribution of TiC and TiB2 particles in the composite. The TiC-TiB2 particles significantly constrained the plastic deformation of the Al matrix and increased the wear resistance of the composite. The composite presented a much smoother worn surface than the Al alloy, as indicated in Figure 12.  Figure 13 shows an abrasive wear behavior schematic view of the Al-4.7Cu-0.32Mg-0.44Si alloy and (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite. In the abrasive wear process for the Al-4.7Cu-0.32Mg-0.44Si alloy, the Al2O3 abrasives penetrated the Al matrix, moved on the surface and generated the grooves (stage a2), subsequently converting the materials into debris and chips, as shown in Figure 13a. The abrasive wear loss of the Al matrix alloy occurred during the reciprocating movements of the Al2O3 abrasives (stage a3). Grooves and wear debris formed on the worn surfaces, as shown in Figures 9 and 12. As shown in Figure 13b, the TiC and TiB2 particles effectively increased the micro-hard-  Figure 13 shows an abrasive wear behavior schematic view of the Al-4.7Cu-0.32Mg-0.44Si alloy and (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite. In the abrasive wear process for the Al-4.7Cu-0.32Mg-0.44Si alloy, the Al 2 O 3 abrasives penetrated the Al matrix, moved on the surface and generated the grooves (stage a2), subsequently converting the materials into debris and chips, as shown in Figure 13a. The abrasive wear loss of the Al matrix alloy occurred during the reciprocating movements of the Al 2 O 3 abrasives (stage a3). Grooves and wear debris formed on the worn surfaces, as shown in Figures 9 and 12. As shown in Figure 13b, the TiC and TiB 2 particles effectively increased the micro-hardness of the composites, decreased the contact area between the matrix and the Al 2 O 3 abrasives, and weakened the penetration and micro-cutting efficiency of the Al 2 O 3 abrasives (stage b2). The TiC and TiB 2 particles also played a role in the load-bearing elements, which effectively prevented the penetration of the abrasives and then increased the abrasive resistance of the composites [27,36]. Therefore, the composites with higher TiC-TiB 2 particle content were able to show better abrasive resistance and a lower wear rate. As indicated in Figure 11, the thickness of the plastic deformation zone on the wear surface of (TiC-TiB 2 )/Al-Cu-Mg-Si composite was significantly smaller than that of the Al matrix alloy. Since the penetrating depth and micro-cutting efficiency of the abrasive particles increased with increase in the applied load and abrasive size [37,38], the micro-cutting efficiency of the abrasive particles was improved, leading to an increase in the WRs. When tested with a 13 µm abrasive size and under a 5 N load, the abrasive wear mechanisms were micro-cutting and micro-ploughing for the Al-4.7Cu-0.32Mg-0.44Si alloy, while micro-cutting was the main abrasive wear mechanism for the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite. The composite showed much superior abrasive wear resistance to that of the Al alloy (up to 4.17 times higher). As discussed above, with increase in the applied load (5 N to 25 N) or in the Al 2 O 3 abrasive size (13 µm to 40 µm), the abrasive wear mechanism of the (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites changed from micro-cutting and micro-ploughing to micro-ploughing, with formation of much rougher and deeper furrows and much debris. The main abrasive wear mechanisms for the Al-4.7Cu-0.32Mg-0.44Si alloy and the (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites were micro-ploughing and micro-cutting.
to that of the Al alloy (up to 4.17 times higher). As discussed above, with increase in the applied load (5 N to 25 N) or in the Al2O3 abrasive size (13 μm to 40 μm), the abrasive wear mechanism of the (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites changed from micro-cutting and micro-ploughing to micro-ploughing, with formation of much rougher and deeper furrows and much debris. The main abrasive wear mechanisms for the Al-4.7Cu-0.32Mg-0.44Si alloy and the (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites were micro-ploughing and micro-cutting.

Conclusions
In situ multi-sized (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composites were synthesized in an Al-Ti-B4C reaction system by combining combustion reaction synthesis with hotpressed sintering and hot extrusion. With increase in TiC-TiB2 particle content from 10 wt.% to 40 wt.%, the TiC-TiB2 average particle size increased from 95 nm to 223 nm, while the percentage of nano-sized particles reduced from 82.1% to 3.0%. The multi-sized particles effectively improved the tensile properties and abrasive wear resistance of the composites. The mechanical and abrasive wear properties increased with TiC-TiB2 particle content. The yield strength, tensile strength, and hardness of the 40 wt.% (TiC-TiB2)/Al-4.7Cu-0.32Mg-0.44Si composite (569 MPa, 704 MPa, and 286 HV) were, respectively, 74%, 51%, and 110% higher than those of the matrix alloy (327 MPa, 466 MPa, and 136 HV). The

Conclusions
In situ multi-sized (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composites were synthesized in an Al-Ti-B 4 C reaction system by combining combustion reaction synthesis with hot-pressed sintering and hot extrusion. With increase in TiC-TiB 2 particle content from 10 wt.% to 40 wt.%, the TiC-TiB 2 average particle size increased from 95 nm to 223 nm, while the percentage of nano-sized particles reduced from 82.1% to 3.0%. The multi-sized particles effectively improved the tensile properties and abrasive wear resistance of the composites. The mechanical and abrasive wear properties increased with TiC-TiB 2 particle content. The yield strength, tensile strength, and hardness of the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite (569 MPa, 704 MPa, and 286 HV) were, respectively, 74%, 51%, and 110% higher than those of the matrix alloy (327 MPa, 466 MPa, and 136 HV). The abrasive wear resistance of the 40 wt.% (TiC-TiB 2 )/Al-4.7Cu-0.32Mg-0.44Si composite was 4.17 times greater than for the Al matrix under a 5 N load and with 13 µm Al 2 O 3 abrasive size. Micro-cutting and micro-ploughing were the main abrasive wear mechanisms for the Al alloy and composites under various applied loads (5 N to 25 N) and for various Al 2 O 3 abrasive particle sizes (13 µm to 40 µm). The improved abrasive wear performance of the composites was attributed to the excellent bonding of the TiC-Al and TiB 2 -Al interfaces, and improvement in the TiC-TiB 2 particle distribution.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.