Effects of Aluminum Addition on Microstructure and Properties of TiC-TiB2/Fe Coatings In Situ Synthesized by TIG Cladding

This study focuses on the synthesis of TiC-TiB2/Fe coatings with varying amounts of aluminum (Al) using tungsten inert gas (TIG) cladding and investigates the impact of Al addition on microstructure refinement and performance enhancement of the coatings. The coatings were prepared on a mild steel substrate using TIG cladding. X-ray diffraction (XRD) analysis revealed the presence of TiC, TiB2, AlxTi, and AlxFe phases in the coatings. Scanning electron microscopy (SEM) images showed that the addition of Al improved the microstructure, reducing defects and enhancing the distribution of reinforcing phases within the coatings. The particle size of the reinforcing phases was significantly refined by the addition of Al. The micro-hardness of the coatings was significantly higher than that of the substrate, with the maximum micro-hardness of the coating reaching 955.5 ± 50.7 HV0.1, approximately six times that of the substrates. However, excessive Al addition led to a reduction in hardness due to a decrease in the quantity of hard phases. The wear tests showed that all the coatings had lower wear loss compared to the substrate material, with the wear loss initially decreasing and then increasing with the increasing Al content. Samples with a 28.57 wt.% Al addition exhibited the best wear resistance, with approximately 16.8% of the wear volume loss compared to mild steel under the same testing conditions, attributed to the optimal combination of reinforcement phase quantity and matrix properties.


Introduction
By combining metal matrices with ceramic reinforcements, metal matrix composites (MMCs) provide a promising approach for developing advanced components with improved mechanical strength, hardness, and wear resistance [1], making them suitable for wear-resistant and anti-friction applications [2]. In the form of coatings, MMCs can effectively safeguard components from wear [3], erosion [4], and other forms of degradation [5], thereby significantly improving the lifespan and performance of the underlying material [6].
The volume fraction and size of the reinforcing phase are key factors influencing the performance of MMCs. Rahmani et al. [7] found that the volume fractions of SiC reinforcement significantly affect the relative density, micro-hardness, and strength of Mg-SiC MMCs. Increasing the SiC content improves the hardness and strength of the material but decreases the relative density [7]. Additionally, grain refinement plays a beneficial role in further enhancing the performance of MMCs. Rahmani et al. [8] also reported that magnesium composites reinforced with five volume percent of ZrO 2 and TiO 2 nanoparticles exhibited tensile strengths 2.5 and 2.1 times higher than that of unreinforced magnesium. Majzoobi et al. [9] investigated the effect of pre-compaction on the mechanical properties of However, to the best of our knowledge, there is a lack of relevant reports on the effects of Al addition on the microstructure and properties of TiC-TiB 2 /Fe coatings prepared by tungsten inert gas (TIG) cladding. Therefore, the aim of this study was to in situ synthesize a series of TiC-TiB 2 /Fe coatings with varying amounts of Al addition using TIG cladding from a FeTi-B 4 C system and investigate how the addition of Al affects the microstructure refinement and performance enhancement of the coatings.

Materials and Methods
The TiC-TiB 2 /Fe coatings were prepared on the surface of Q235 steel (a commonly used mild steel in China, equivalent to ASTM A36 steel in the United States) by TIG cladding. TIG cladding was performed using a tungsten electrode and argon gas as the heat source to melt the pre-placed powder and form the coating. The cladding was carried out using a YC-300WX4 N-type TIG welding machine with direct current electrode negative (DCEN) polarity. A tungsten electrode with a diameter of 2.0 mm was used, and the welding current was set to 60 A. Air cooling was employed for the cooling process. Argon gas with a purity of 99.9% and a flow rate of 10 L/min was used to shield and stabilize the arc. The schematic diagram of TIG cladding is shown in Figure 1 [13]. The scan speeds were set to 2.0 mm/s. Before cladding, the surface of the substrates was polished with 800-mesh emery paper and cleaned with acetone. The surface roughness (Ra) was approximately 0.6 µm. Mixtures of Al, FeTi70, and B 4 C powders were used as precursors for coating preparation. The main chemical components of the powders are shown in Table 1, and the particle sizes of all powders were −300 mesh. The component ratios of the precursor powders and the corresponding sample numbers are listed in Table 2. The mass ratio of FeTi70 and B 4 C powders was fixed at 4:1, which is approximately equal to the molar ratio of complete reaction of 3Ti + B 4 C = 2TiB 2 + TiC. Based on this ratio, the ratio of Al powder was increased. Firstly, the powders were mixed using a ball milling technique for three hours. Subsequently, the blended powders were pre-placed on the surface of Q235 steel with organic binder to give a thickness of approximately 1.0-1.5 mm. The samples were then set in a ventilated place at room temperature for 24 h and subsequently placed into a vacuum drying oven at 80°C for 1 h. density and compressive strength of these composites increased [25].
However, to the best of our knowledge, there is a lack of relevant reports on the effects of Al addition on the microstructure and properties of TiC-TiB2/Fe coatings prepared by tungsten inert gas (TIG) cladding. Therefore, the aim of this study was to in situ synthesize a series of TiC-TiB2/Fe coatings with varying amounts of Al addition using TIG cladding from a FeTi-B4C system and investigate how the addition of Al affects the microstructure refinement and performance enhancement of the coatings.

Materials and Methods
The TiC-TiB2/Fe coatings were prepared on the surface of Q235 steel (a commonly used mild steel in China, equivalent to ASTM A36 steel in the United States) by TIG cladding. TIG cladding was performed using a tungsten electrode and argon gas as the heat source to melt the pre-placed powder and form the coating. The cladding was carried out using a YC-300WX4 N-type TIG welding machine with direct current electrode negative (DCEN) polarity. A tungsten electrode with a diameter of 2.0 mm was used, and the welding current was set to 60 A. Air cooling was employed for the cooling process. Argon gas with a purity of 99.9% and a flow rate of 10 L/min was used to shield and stabilize the arc. The schematic diagram of TIG cladding is shown in Figure 1 [13]. The scan speeds were set to 2.0 mm/s. Before cladding, the surface of the substrates was polished with 800-mesh emery paper and cleaned with acetone. The surface roughness (Ra) was approximately 0.6 μm. Mixtures of Al, FeTi70, and B4C powders were used as precursors for coating preparation. The main chemical components of the powders are shown in Table 1, and the particle sizes of all powders were −300 mesh. The component ratios of the precursor powders and the corresponding sample numbers are listed in Table 2. The mass ratio of FeTi70 and B4C powders was fixed at 4:1, which is approximately equal to the molar ratio of complete reaction of 3Ti + B4C = 2TiB2 + TiC. Based on this ratio, the ratio of Al powder was increased. Firstly, the powders were mixed using a ball milling technique for three hours. Subsequently, the blended powders were pre-placed on the surface of Q235 steel with organic binder to give a thickness of approximately 1.0-1.5 mm. The samples were then set in a ventilated place at room temperature for 24 h and subsequently placed into a vacuum drying oven at 80 ℃ for 1 h.

Powders
Chemical Component (wt.%) The microstructure and chemical composition of the samples were analyzed by a JSM-5610 LV scanning electron microscope (SEM), which was attached with an energy dispersive spectrometer (EDS). Phase constituents of the coatings were identified by a D8-ADVANCE X-ray diffractometer (XRD) with an accelerating voltage of 40 KV and a current of 40 mA. The XRD scanning speed was set at 2 • /min in the range of 20-90 • . The Vickers microhardness of the coatings along the depth of the cross-section was measured using an MHV-2000 microhardness tester (Beijing TIME Shuncheng Technology Co., Ltd., Beijing, China) with a load of 100 g and a loading time of 10 s [26]. The microhardness value was from the average of 5 measurements. According to the national standard of the People's Republic of China, GB 12444.2-90 [27], "Metallic materials-Wear tests Block-onring wear test," dry sliding wear tests were carried out at room temperature. The tests were conducted using a block-on-ring wear tester (MM-200, Zhangjiakou, China) without the use of any lubrication [26]. A quenched bearing steel ring with a hardness of 65 HRC was used as the wear couple. The ring had an outer diameter of 40 mm and a width of 10 mm. Prior to the tests, the samples were machined to a size of 15 mm × 7 mm × 7 mm and were ground using silicon carbide sandpapers. The wear tests were conducted using a normal load of 98 N, a sliding speed 0.31 m/s, and a sliding distance of 2260 m. After the tests, the width of the wear track b was measured using a vernier caliper with a precision of 0.01 mm. The volume of wear volume loss V can be obtained using the following formula: where, V is the volume of wear loss, mm 3 ; D is the outer diameter of the ring, mm; t is the width of the specimen, mm; b is the width of the wear track, mm.
The wear resistance of the coating specimens is measured by the wear rate, denoted as W r . The calculation formula is as follows: where, W r is the wear rate of the specimen, mm 3 /Nm; V is the wear volume loss, mm 3 ; F is the applied load, N; S is the sliding distance, m.
During the wear test, the friction torque is recorded, and the coefficient of friction (COF) of the coating can be calculated based on the friction torque. To minimize random errors, the final values of wear volume loss and COF were obtained by conducting three tests under the same conditions. Subsequently, the worn surface of the samples was examined using scanning electron microscopy.

Results and Discussion
3.1. X-ray Results of Coatings Figure 2 shows the XRD patterns of the coatings with varying Al contents. The patterns revealed that TiC and TiB 2 peaks were detected in all samples, indicating that the TiC and TiB 2 particles were in situ synthesized from the FeTi70 and B 4 C powders during the cladding process. When Al powders were added, Al x Ti and Al x Fe phases were detected in the coatings. As the Al content increased from 16.67% to 44.44%, the Al x Fe phases changed from Al 2 Fe 3 (PDF#45-0982) to Al 13 Fe 4 (PDF#47-1420) phases. Sample S3 had a higher relative diffraction peak intensity of TiB 2 than samples S2 and S4, indicating that the content of TiB 2 in sample S3 was relatively higher than that in samples S2 and S4. Additionally, intermediate phase TiB appeared in the coatings, likely due to the molten pool undergoing an incompletely non-equilibrium reaction under the extremely rapid solidification process. Although the diffraction peaks were not salient, small amounts of Fe x (C,B) phases were also detected in all samples as reaction products. Additionally, the peaks of TiB 2 and the matrix (α-Fe, AlxFe) overlap significantly, which is also observed in the references [21,28,29]. This situation prevents us from performing certain XRD quantitative analyses, such as estimating the content ratio of different phases based on peak intensities or using Scherrer's equation to calculate the grain size by analyzing the peak broadening of XRD peaks.  Figure 2 shows the XRD patterns of the coatings with varying Al contents. The pat terns revealed that TiC and TiB2 peaks were detected in all samples, indicating that th TiC and TiB2 particles were in situ synthesized from the FeTi70 and B4C powders during the cladding process. When Al powders were added, AlxTi and AlxFe phases were de tected in the coatings. As the Al content increased from 16.67% to 44.44%, the AlxFe phase changed from Al2Fe3 (PDF#45-0982) to Al13Fe4 (PDF#47-1420) phases. Sample S3 had higher relative diffraction peak intensity of TiB2 than samples S2 and S4, indicating tha the content of TiB2 in sample S3 was relatively higher than that in samples S2 and S4 Additionally, intermediate phase TiB appeared in the coatings, likely due to the molten pool undergoing an incompletely non-equilibrium reaction under the extremely rapid so lidification process. Although the diffraction peaks were not salient, small amounts o Fex(C,B) phases were also detected in all samples as reaction products. Additionally, th peaks of TiB2 and the matrix (α-Fe, AlxFe) overlap significantly, which is also observed in the references [21,28,29]. This situation prevents us from performing certain XRD quanti tative analyses, such as estimating the content ratio of different phases based on peak in tensities or using Scherrer's equation to calculate the grain size by analyzing the peak broadening of XRD peaks.   Figure 3 shows the SEM micrographs near the interface between the substrate and the coatings in sample S1 (a) and sample S3 (b) at a lower magnification (500×). The microstructures at a lower magnification in samples S2, S3, and S4 were similar. Therefore, sample S3 was chosen as a representative sample with the addition of Al element to compare with sample S1, which did not have Al element addition.

Microstructure of Coatings
PEER REVIEW 6 of 16 Figure 3 shows the SEM micrographs near the interface between the substrate and the coatings in sample S1 (a) and sample S3 (b) at a lower magnification (500×). The microstructures at a lower magnification in samples S2, S3, and S4 were similar. Therefore, sample S3 was chosen as a representative sample with the addition of Al element to compare with sample S1, which did not have Al element addition. Several pores and slag inclusion defects can be observed in sample S1. However, in sample S3, the base metal and coating exhibited a good metallurgical bond, with no apparent cracks, pores, or other defects. This indicates that the addition of Al element improved the microstructure and reduced the coating defects. This improvement can be attributed to the low melting point of Al (660 °C). During the cladding process, Al melted first, forming a molten pool. The longer holding time of the molten pool, resulting from the addition of Al, enhanced the metallurgical reaction [30]. As a result, gases and slag inclusions had sufficient time to rise to the top of the molten pool. Figure 4 presents a higher magnification cross-section micrograph of the coatings with varying Al contents. In Figure 4a, it is evident that the reinforcing phases of sample S1 were not uniformly distributed within the matrix. However, the addition of Al element improved the distribution of reinforcing phases in the coatings, as illustrated in Figure  4b-d. The coatings exhibited three main types of reinforcing phases. One type was rectangular in shape (point "1" in Figure 4a), another was granular (point "2" in Figure 4b), and the third type was network-shaped (point "3" in Figure 4). Similar typical microstructural features have also been observed in the literature [21]. Several pores and slag inclusion defects can be observed in sample S1. However, in sample S3, the base metal and coating exhibited a good metallurgical bond, with no apparent cracks, pores, or other defects. This indicates that the addition of Al element improved the microstructure and reduced the coating defects. This improvement can be attributed to the low melting point of Al (660 • C). During the cladding process, Al melted first, forming a molten pool. The longer holding time of the molten pool, resulting from the addition of Al, enhanced the metallurgical reaction [30]. As a result, gases and slag inclusions had sufficient time to rise to the top of the molten pool. Figure 4 presents a higher magnification cross-section micrograph of the coatings with varying Al contents. In Figure 4a, it is evident that the reinforcing phases of sample S1 were not uniformly distributed within the matrix. However, the addition of Al element improved the distribution of reinforcing phases in the coatings, as illustrated in Figure 4b-d. The coatings exhibited three main types of reinforcing phases. One type was rectangular in shape (point "1" in Figure 4a), another was granular (point "2" in Figure 4b), and the third type was network-shaped (point "3" in Figure 4). Similar typical microstructural features have also been observed in the literature [21]. Figure 5 presents the EDS analysis results of these three reinforcing phases. In Figure 5a, the main elements at point "1" were Ti and B, suggesting that the rectangularshaped particles were TiB 2 or TiB. Figure 5b indicates that the granular particles were composed of Ti and C, implying that they were likely TiC. Additionally, Figure 5b reveals that the network-shaped particles were rich in Fe, C, and B elements, indicating the formation of Fe x (C,B) compound in the coating. The elements of the matrix (point "4" in Figure 4) in the coatings were also examined, as shown in Figure 5d, and mainly comprised Fe, Al, and Ti. Combined with the XRD analysis results, it can be inferred that the matrix of the coatings consists of intermetallic compounds such as Al x Ti and Al x Fe.    Figure 5 presents the EDS analysis results of these three reinforcing phases. In Figure  5a, the main elements at point "1" were Ti and B, suggesting that the rectangular-shaped particles were TiB2 or TiB. Figure 5b indicates that the granular particles were composed of Ti and C, implying that they were likely TiC. Additionally, Figure 5b reveals that the network-shaped particles were rich in Fe, C, and B elements, indicating the formation of Fex (C,B) compound in the coating. The elements of the matrix (point "4" in Figure 4) in and Ti. Combined with the XRD analysis results, it can be inferred that the matrix of the coatings consists of intermetallic compounds such as AlxTi and AlxFe. The results depicted in Figure 4 demonstrate that the particle size of the reinforcing phase was significantly refined by the addition of Al. In particular, sample S3 exhibited a uniformly distributed reinforcing phase with smaller particle size compared to the other samples. This can be primarily attributed to the significant effects of Al on the reaction behaviors and phase compositions [23].

Microstructure of Coatings
Upon adding a small amount of Al to the precursor, the formation of AlxTi and AlxFe intermetallic compounds occurred initially. These intermetallic compound layers surrounded the B4C particles, facilitating their contact with AlxTi and reducing the atomic diffusion distance for subsequent reactions. According to references [23,31], the diffusion of Al into the B4C crystal structure may have played a role in breaking the bonds between boron and carbon, thereby promoting the dissociation of B4C. Furthermore, some researchers have reported that the AlxTi phase acts as a master alloy in the Al-Ti-C system, leading to the refinement of TiC and TiB2 particle sizes [32,33]. However, when the addition of Al powder increased to 44.4%, the amount of strengthening phases in the coatings was noticeably reduced, as observed in Figure 4d.

Micro-Hardness Analysis of the Coatings
The micro-hardness measurements along the cross-section of the specimens are presented in Figure 6. The micro-hardness of the coatings was significantly higher than that of the substrate, with the maximum micro-hardness of the coating reaching 955.5 ± 50.7 HV0.1 (S1), approximately six times that of the substrates. This increase in hardness can be attributed to the presence of hard reinforcement phases synthesized in situ during the cladding process. From Figure 6, it is evident that the average micro-hardness of sample S1 was the highest, followed by sample S3 (826.1 ± 59.9 HV0.1) and S2 (769.5 ± 51.3 HV0.1), while the hardness of sample S4 was relatively lower (582.5 ± 39.7 HV0.1). The hardness The results depicted in Figure 4 demonstrate that the particle size of the reinforcing phase was significantly refined by the addition of Al. In particular, sample S3 exhibited a uniformly distributed reinforcing phase with smaller particle size compared to the other samples. This can be primarily attributed to the significant effects of Al on the reaction behaviors and phase compositions [23].
Upon adding a small amount of Al to the precursor, the formation of Al x Ti and Al x Fe intermetallic compounds occurred initially. These intermetallic compound layers surrounded the B 4 C particles, facilitating their contact with Al x Ti and reducing the atomic diffusion distance for subsequent reactions. According to references [23,31], the diffusion of Al into the B 4 C crystal structure may have played a role in breaking the bonds between boron and carbon, thereby promoting the dissociation of B 4 C. Furthermore, some researchers have reported that the Al x Ti phase acts as a master alloy in the Al-Ti-C system, leading to the refinement of TiC and TiB 2 particle sizes [32,33]. However, when the addition of Al powder increased to 44.4%, the amount of strengthening phases in the coatings was noticeably reduced, as observed in Figure 4d

Micro-Hardness Analysis of the Coatings
The micro-hardness measurements along the cross-section of the specimens are presented in Figure 6. The micro-hardness of the coatings was significantly higher than that of the substrate, with the maximum micro-hardness of the coating reaching 955.5 ± 50.7 HV0.1 (S1), approximately six times that of the substrates. This increase in hardness can be attributed to the presence of hard reinforcement phases synthesized in situ during the cladding process. From Figure 6, it is evident that the average micro-hardness of sample S1 was the highest, followed by sample S3 (826.1 ± 59.9 HV0.1) and S2 (769.5 ± 51.3 HV0.1), while the hardness of sample S4 was relatively lower (582.5 ± 39.7 HV0.1). The hardness values of these coatings were comparable to the TiC-TiB2-reinforced MMC coatings reported in the literature for TIG cladding and some laser cladding processes [14,19], significantly higher than the hardness of various common steel materials. However, they were lower than Wang's laser-cladded TiC-TiB 2 /Fe coatings (approximately 1100-1300 HV), which can be attributed to the inclusion of a considerable amount of Al 2 O 3 as an additional reinforcing phase in Wang's samples [21]. values of these coatings were comparable to the TiC-TiB2-reinforced MMC coatings reported in the literature for TIG cladding and some laser cladding processes [14,19], significantly higher than the hardness of various common steel materials. However, they were lower than Wang's laser-cladded TiC-TiB2/Fe coatings (approximately 1100-1300 HV), which can be attributed to the inclusion of a considerable amount of Al2O3 as an additional reinforcing phase in Wang's samples [21]. This difference in hardness among the four coatings can primarily be attributed to the addition of aluminum, which influenced the relative amount of reinforcement phases. However, an appropriate aluminum content resulted in the refinement and uniform distribution of the reinforcement phase particles, as observed in Figure 4. Simultaneously, the matrix of the coatings transformed from α-Fe into FexAl intermetallic compounds, further contributing to the improved hardness of the coatings. Conversely, when too much aluminum was added, as seen in sample S4, the quantity of hard phases decreased significantly, leading to a noticeable reduction in hardness.
Furthermore, we observed a gradual increase in micro-hardness from the base metal to the coatings. This phenomenon can be attributed to the gradient distribution of TiC and TiB2 particles. The density of TiC (4.92 g/cm 3 ) and TiB2 (4.52 g/cm 3 ) is lower than that of Fe (approximately 7.86 g/cm 3 ) [21]. As a result, these reinforcing phases have a natural tendency to migrate towards the upper zone during the cladding process.
However, the rapid solidification of the molten pool captures these reinforcing phases and incorporates them into the coatings. Consequently, the combined effects of floating and capturing lead to the gradient distribution of these hard particles. This gradient micro-hardness distribution helps reduce stress concentration at the interface, thereby improving the service life of the workpieces during operation. Figure 7 illustrates the wear volume loss of the coatings and the substrate material (Q235 steel). It is evident that the wear loss of all the coatings is significantly lower than that of the Q235 steel. Among the coatings, S3 coating exhibits the best wear resistance, with a wear loss approximately 16.8% that of Q235 steel. This can be attributed to the high hardness of the in situ synthesized hard reinforcement particles within the coatings. These This difference in hardness among the four coatings can primarily be attributed to the addition of aluminum, which influenced the relative amount of reinforcement phases. However, an appropriate aluminum content resulted in the refinement and uniform distribution of the reinforcement phase particles, as observed in Figure 4. Simultaneously, the matrix of the coatings transformed from α-Fe into Fe x Al intermetallic compounds, further contributing to the improved hardness of the coatings. Conversely, when too much aluminum was added, as seen in sample S4, the quantity of hard phases decreased significantly, leading to a noticeable reduction in hardness.

Wear Resistance Analysis of the Coatings
Furthermore, we observed a gradual increase in micro-hardness from the base metal to the coatings. This phenomenon can be attributed to the gradient distribution of TiC and TiB 2 particles. The density of TiC (4.92 g/cm 3 ) and TiB2 (4.52 g/cm 3 ) is lower than that of Fe (approximately 7.86 g/cm 3 ) [21]. As a result, these reinforcing phases have a natural tendency to migrate towards the upper zone during the cladding process.
However, the rapid solidification of the molten pool captures these reinforcing phases and incorporates them into the coatings. Consequently, the combined effects of floating and capturing lead to the gradient distribution of these hard particles. This gradient microhardness distribution helps reduce stress concentration at the interface, thereby improving the service life of the workpieces during operation. Figure 7 illustrates the wear volume loss of the coatings and the substrate material (Q235 steel). It is evident that the wear loss of all the coatings is significantly lower than that of the Q235 steel. Among the coatings, S3 coating exhibits the best wear resistance, with a wear loss approximately 16.8% that of Q235 steel. This can be attributed to the high hardness of the in situ synthesized hard reinforcement particles within the coatings. These hard particles are capable of withstanding the alternating load during the dry wear test, thereby reducing the contact between the matrix and the grinding ring [34].

Wear Resistance Analysis of the Coatings
Materials 2023, 16, x FOR PEER REVIEW hard particles are capable of withstanding the alternating load during the dry w thereby reducing the contact between the matrix and the grinding ring [34].   [21] and lower than that of Q235 steel. As shown in Figures 7 and 8, the wear loss of the coatings initially decreases a   [21] and lower than that of Q235 steel. hard particles are capable of withstanding the alternating load during the dry wear test thereby reducing the contact between the matrix and the grinding ring [34].   [21] and lower than that of Q235 steel. As shown in Figures 7 and 8, the wear loss of the coatings initially decreases and then increases with the increase in Al addition. Interestingly, although sample S1 has the highest micro-hardness value, its wear rate is not the lowest. This discrepancy can be explained by considering factors beyond hardness, such as the microstructure of the coatings. In Figure 3a, it can be observed that sample S1 exhibits some slag inclusions in the coating As shown in Figures 7 and 8, the wear loss of the coatings initially decreases and then increases with the increase in Al addition. Interestingly, although sample S1 has the highest micro-hardness value, its wear rate is not the lowest. This discrepancy can be explained by considering factors beyond hardness, such as the microstructure of the coatings. In Figure 3a, it can be observed that sample S1 exhibits some slag inclusions in the coating, which can contribute to increased wear loss. However, proper aluminum addition refines the particles of the reinforcement phases and ensures their uniform distribution, as mentioned earlier. Additionally, the addition of Al element alters the matrix of the coatings and increases the relative content of the matrix alloy, as depicted in Figure 4. The matrix material plays a crucial role in supporting the reinforcing phases. Specifically, the matrix of sample S1 is composed of α-Fe, whereas Al x Fe and Al x Ti intermetallic compounds appear in the matrix of samples S2-S4 due to the addition of Al element, as seen in Figure 2. The hardness of α-Fe is lower than that of Al x Fe and Al x Ti intermetallic compounds. Furthermore, the addition of Al element enhances the supporting effect of the matrix on the reinforcement particles. Consequently, the wear resistance of the coatings improves with the appropriate amount of Al addition. Figure 9 presents the morphology of the worn surfaces of the coatings after dry wear tests. In Figure 9a, numerous irregular pits can be observed on the worn surface of coating S1. However, in Figure 9b-d, the worn surfaces of samples S2-S4 show minimal to no presence of small pits. This can be attributed to the addition of the appropriate amount of aluminum (Al) in the precursors, which increased the matrix metal content and facilitated the formation of Al x Fe and Al x Ti intermetallic compounds in the matrix. These compounds effectively hindered the spalling of hard particles. Additionally, grooves were observed on the worn surfaces of samples S2-S4. Notably, the grooves on the worn surface of sample S4 were deeper than those of samples S2 and S3, and traces of plastic deformation were also evident. The excessive addition of Al reduced the relative content of the reinforcing phases, leading to these observations. Additionally, the addition of Al element alters the matrix of the coatings and increases the relative content of the matrix alloy, as depicted in Figure 4. The matrix material plays a crucial role in supporting the reinforcing phases. Specifically, the matrix of sample S1 is composed of α-Fe, whereas AlxFe and AlxTi intermetallic compounds appear in the matrix of samples S2-S4 due to the addition of Al element, as seen in Figure 2.
The hardness of α-Fe is lower than that of AlxFe and AlxTi intermetallic compounds. Furthermore, the addition of Al element enhances the supporting effect of the matrix on the reinforcement particles. Consequently, the wear resistance of the coatings improves with the appropriate amount of Al addition. Figure 9 presents the morphology of the worn surfaces of the coatings after dry wear tests. In Figure 9a, numerous irregular pits can be observed on the worn surface of coating S1. However, in Figure 9b-d, the worn surfaces of samples S2-S4 show minimal to no presence of small pits. This can be attributed to the addition of the appropriate amount of aluminum (Al) in the precursors, which increased the matrix metal content and facilitated the formation of AlxFe and AlxTi intermetallic compounds in the matrix. These compounds effectively hindered the spalling of hard particles. Additionally, grooves were observed on the worn surfaces of samples S2-S4. Notably, the grooves on the worn surface of sample S4 were deeper than those of samples S2 and S3, and traces of plastic deformation were also evident. The excessive addition of Al reduced the relative content of the reinforcing phases, leading to these observations.  Based on a comparison with previously published literature, it can be concluded that the main wear mechanisms of MMC coatings include spalling or fracture of reinforcing phase particles [26], as well as plastic erasing and removal of the matrix phase [34], such as micro-cutting or micro-plowing. In Figure 9a, the presence of irregular pits can be attributed to the spalling of hard particles. Comparing it with Figure 4a, it can be observed that coating S1 had a relatively high content of hard phases but a low content of matrix metal, which resulted in the spalling of hard particles without sufficient support and in the formation of these small pits. The grooves on the worn surfaces of samples S2-S4 are typical features of micro-cutting or micro-plowing [34]. Especially in sample S4, due to the decrease in hardness, the worn surface exhibited deeper grooves and traces of plastic deformation. Therefore, we can infer that the wear mechanisms of the coatings undergo certain changes with the quantity of reinforcing phases and the properties of the matrix materials. When the content of the binder phase is low, combined with the presence of some microstructural defects (see Figure 3), the hard particles in coating S1 lack sufficient peripheral bonding support, leading to a wear mechanism primarily characterized by spalling of hard particles followed by abrasive wear. However, due to the high hardness, the overall wear rate is still relatively low. With the addition of aluminum (Al), the observed phenomena in coating S1 are improved, and the surfaces of coatings S2 and S3 show almost no small pits. Some shallow grooves indicate the presence of abrasive wear mechanisms. Wang's research has shown that when the content of hard particles is high, they carry a larger portion of the load and reduce the real contact area during the wear process, resulting in severe plastic deformation and adhesion [21]. However, when the content of hard particles is relatively low, such as in sample S4, there is a significant decrease in hardness, and the surface's resistance to plastic deformation is reduced, leading to prominent features of plastic removal and a higher wear rate compared to the other three coating samples. Based on the above analysis, we can conclude that S3 exhibits the best wear resistance, benefiting from a combination of factors including reduced defects, matrix strengthening, appropriate content of reinforcing phases, and uniform refinement of the microstructure.

Conclusions
(1) TiC-TiB 2 particles-reinforced Fe-based composite coatings were successfully synthesized in situ through TIG cladding using FeTi70, B 4 C, and Al powders. The addition of Al powders facilitated the synthesis of Al x Fe and Al x Ti phases in the coatings. (2) The inclusion of Al element in the precursors reduced the occurrence of defects in the coatings. Although the relative amount of reinforcement phases decreased with increasing Al powders, the particles of reinforcement phases were refined and exhibited uniform distribution. (3) The micro-hardness of the coatings was significantly higher than that of the substrate, with the maximum micro-hardness of the coating reaching 955.5 ± 50.7 HV0.1, approximately six times that of the substrates. However, excessive Al addition led to a reduction in hardness due to a decrease in the quantity of hard phases. (4) The appropriate addition of Al powders improved the wear resistance of the TiC-TiB 2 /Fe coatings. Sample S3, with a 28.57 wt.% Al addition, demonstrated the best wear resistance due to the optimal combination of reinforcement phase quantity and matrix properties. The wear rate of sample S3 was approximately 16.8% that of the mild steel under the same testing conditions.