Slurry Erosion–Corrosion Characteristics of As-Built Ti-6Al-4V Manufactured by Selective Laser Melting

Erosion and erosion–corrosion tests of as-built Ti-6Al-4V manufactured by Selective Laser Melting were investigated using slurries composed of SiO2 sand particles and either tap water (pure water) or 3.5% NaCl solution (artificial seawater). The microhardness value of selective laser melting (SLM)ed Ti-6Al-4V alloy increased as the impact angle increased. The synergistic effect of corrosion and erosion in seawater is always higher than erosion in pure water at all impact angles. The seawater environment caused the dissolution of vanadium oxide V2O5 on the surface of SLMed Ti-6Al-4V alloy due to the presence of Cl− ions in the seawater. These findings show lower microhardness values and high mass losses under the erosion–corrosion test compared to those under the erosion test at all impact angles.


Introduction
Ti-6Al-4V is the most broadly utilized titanium alloy as it currently covers about 50% of the worldwide production of titanium alloys [1][2][3]. It is a duplex ((hcp, α-Ti) + (bcc, β-Ti))-type microstructure alloy at room temperature. The major advantages of Ti-6Al-4V are its high strength and hardness, low specific weight (roughly 45% lighter than steel), superior corrosion resistance, biocompatibility, high creep resistance, low maintenance cost, and long service life [4][5][6]. Thanks to these advantageous characteristics, Ti-6Al-4V has emerged as a powerful alloy for a wide field of applications. In addition to its wide use in surgical instruments and medical implants, it is currently used as a material for aerospace components (i.e., engine compressor blades and disks and helicopter rotor blades) [7][8][9], marine and offshore components (i.e., marine ship hulls, propellers, tubes, and shells) [1,10,11], oil and wastewater systems components (i.e., valves, pipelines, pipe fittings, and pumps) [12][13][14], hydropower plant components (i.e., turbine blades, pumps) [15][16][17], and architectural cladding and roofing [11]. These components are typically exposed to slurry erosion, where the working surfaces are continuously impacted by a stream of a slurry mixture. So, central to the entire discipline of Ti-6Al-4V industrial slurry carrier applications is the investigation of its tribological behaviors, especially slurry erosion and corrosion.
Slurry erosion is a type of mechanically induced wear process that occurs as a result of relative motion and/or impact between a stream of liquid containing suspended abrasive particles such as sand and the target surface, causing material degradation. It is a life-limiting challenge in the design of Ti-6Al-4V components exposed to a slurry environment leading to a drop in efficiency and increased morphology of Ti-6Al-4V powder particles. Overall, particles have a smooth, spherical shape indicating a good flowability. The SLM processing parameters are shown in Table 2. Figure 2 shows the sample size and orientation on the building platform. After fabrication, the as-built SLMed samples were cut, and sample surfaces were subjected to standard metallographic procedures using the following protocol: grinding with 240, 600, 1200, and 2500 Grit SiC papers and then mirror polishing using a polishing cloth with suspensions containing 6, 3, and 1 µm diamond particles, followed by cleaning in distilled water. Next, to reveal the microstructure, selected polished samples were etched using a solution comprising 10% hydrofluoric acid (HF), 5% nitric acid (HNO3), and 85% distilled water for an etching time of 5-10 s. All samples were then degreased in ethanol using an ultrasound cleaner for 10 min each and dried using compressed hot air.   Schematic of the test sample. The x-z surfaces were exposed to the slurry stream. morphology of Ti-6Al-4V powder particles. Overall, particles have a smooth, spherical shape indicating a good flowability. The SLM processing parameters are shown in Table 2. Figure 2 shows the sample size and orientation on the building platform. After fabrication, the as-built SLMed samples were cut, and sample surfaces were subjected to standard metallographic procedures using the following protocol: grinding with 240, 600, 1200, and 2500 Grit SiC papers and then mirror polishing using a polishing cloth with suspensions containing 6, 3, and 1 µm diamond particles, followed by cleaning in distilled water. Next, to reveal the microstructure, selected polished samples were etched using a solution comprising 10% hydrofluoric acid (HF), 5% nitric acid (HNO3), and 85% distilled water for an etching time of 5-10 s. All samples were then degreased in ethanol using an ultrasound cleaner for 10 min each and dried using compressed hot air.   Schematic of the test sample. The x-z surfaces were exposed to the slurry stream. Figure 2. Schematic of the test sample. The x-z surfaces were exposed to the slurry stream.

Microstructure and Microhardness Investigation
The surface morphology and microstructure of samples before and after erosion and erosion-corrosion tests were observed using both scanning electron microscopy (SEM, JEOL JSM 5400, Japan) and optical microscopy (PME OLYMPUS, Tokyo, Japan). Microhardness was measured on the eroded surface at a load of 200 g and an indentation duration of 15 s using a Vickers microhardness tester (MICROMET ® , ADOLPH I. BUEHLER INC, Lake Bluff, IL, USA). An average of at least five readings for each sample was reported in this study. The microhardness results are reported as the microhardness versus the impact angle.

Erodent
Natural SiO 2 sand with a nominal size ranging from 355 to 500 µm was employed as an erodent to form a solid-liquid slurry with tap water and artificial seawater (tap water + 3.5% NaCl). A scanning electron micrograph of the erodent is shown in Figure 3, and the statistical values of the used SiO 2 particles are listed in Table 3. The SEM image in Figure 3 shows that the particles' shapes were relatively block-like and regular. corrosion tests were observed using both scanning electron microscopy (SEM, JEOL JSM 5400, Japan) and optical microscopy (PME OLYMPUS, Tokyo, Japan). Microhardness was measured on the eroded surface at a load of 200 g and an indentation duration of 15 s using a Vickers microhardness tester (MICROMET ® , ADOLPH I. BUEHLER INC, Lake Bluff, IL, USA). An average of at least five readings for each sample was reported in this study. The microhardness results are reported as the microhardness versus the impact angle.

Erodent
Natural SiO2 sand with a nominal size ranging from 355 to 500 µm was employed as an erodent to form a solid-liquid slurry with tap water and artificial seawater (tap water + 3.5% NaCl). A scanning electron micrograph of the erodent is shown in Figure 3, and the statistical values of the used SiO2 particles are listed in Table 3. The SEM image in Figure 3 shows that the particles' shapes were relatively block-like and regular.

Slurry Erosion and Erosion-Corrosion Tests Procedure
Erosion and erosion-corrosion tests were performed using a whirling arm slurry erosion test rig (WASET) [52], as shown in Figures 4 and 5. The rig was composed of 3 main units: a slurry mixing unit, a slurry test chamber, and a vacuum unit. The details of the used experimental test rig design and performance are provided in [53,54]. In the first set of experiments (erosion tests), the procedure started with a slurry mixing unit consisting of a 25 L cylindrical tank containing 1% sand particles that were added to tap water and mixed via a stirrer, before being passed through a pipe to the slurry test chamber. In the slurry test chamber, the slurry mixture flowed to a funnel with a 3 mm diameter orifice equipped with a stirrer to keep the slurry under suspension. The funnel provided a falling homogenous stable slurry stream at the center of the sample surface. Test specimens were placed on two holders mounted at the ends of two horizontal arms fitted at 180° apart from each other to balance the dynamic forces. The diametric holder-to-holder distance was 248 mm, and samples were

Slurry Erosion and Erosion-Corrosion Tests Procedure
Erosion and erosion-corrosion tests were performed using a whirling arm slurry erosion test rig (WASET) [52], as shown in Figures 4 and 5. The rig was composed of 3 main units: a slurry mixing unit, a slurry test chamber, and a vacuum unit. The details of the used experimental test rig design and performance are provided in [53,54]. In the first set of experiments (erosion tests), the procedure started with a slurry mixing unit consisting of a 25 L cylindrical tank containing 1% sand particles that were added to tap water and mixed via a stirrer, before being passed through a pipe to the slurry test chamber. In the slurry test chamber, the slurry mixture flowed to a funnel with a 3 mm diameter orifice equipped with a stirrer to keep the slurry under suspension. The funnel provided a falling homogenous stable slurry stream at the center of the sample surface. Test specimens were placed on two holders mounted at the ends of two horizontal arms fitted at 180 • apart from each other to balance the dynamic forces. The diametric holder-to-holder distance was 248 mm, and samples were Materials 2020, 13, 3967 5 of 16 positioned 40 mm from the tip of the orifice. An impact angle from 0 • to 90 • could be adjusted by rotating the sample holder around the arm axis, as shown in Figure 5. The two arms were attached to a brass sleeve firmly tightened to the top end of a vertical whirling shaft which provided balance under high-speed operation and was driven by a variable speed motor. Two samples were tested at the same time, and a single surface with dimensions of 23 mm × 10 mm (see Figure 1) for each sample was exposed to the slurry stream at impact angles of 30 • , 45 • , 60 • , and 90 • . To eliminate aerodynamic effects on the slurry stream, the slurry test chamber was evacuated by a vacuum system (up to 28 cm Hg). The second set of experiments (erosion-corrosion tests) followed the same procedure, but samples were exposed to a slurry with seawater (tap water + 3.5% NaCl) instead of tap water. To replenish the consumed slurry during any set of experiments, predetermined amounts of pure water (or seawater) and SiO 2 sand flowed and were stirred continuously in the slurry tank.
Materials 2020, 13, x FOR PEER REVIEW 5 of 16 positioned 40 mm from the tip of the orifice. An impact angle from 0° to 90° could be adjusted by rotating the sample holder around the arm axis, as shown in Figure 5. The two arms were attached to a brass sleeve firmly tightened to the top end of a vertical whirling shaft which provided balance under high-speed operation and was driven by a variable speed motor. Two samples were tested at the same time, and a single surface with dimensions of 23 mm × 10 mm (see Figure 1) for each sample was exposed to the slurry stream at impact angles of 30°, 45°, 60°, and 90°. To eliminate aerodynamic effects on the slurry stream, the slurry test chamber was evacuated by a vacuum system (up to 28 cm Hg). The second set of experiments (erosion-corrosion tests) followed the same procedure, but samples were exposed to a slurry with seawater (tap water + 3.5% NaCl) instead of tap water. To replenish the consumed slurry during any set of experiments, predetermined amounts of pure water (or seawater) and SiO2 sand flowed and were stirred continuously in the slurry tank.   Materials 2020, 13, x FOR PEER REVIEW 5 of 16 positioned 40 mm from the tip of the orifice. An impact angle from 0° to 90° could be adjusted by rotating the sample holder around the arm axis, as shown in Figure 5. The two arms were attached to a brass sleeve firmly tightened to the top end of a vertical whirling shaft which provided balance under high-speed operation and was driven by a variable speed motor. Two samples were tested at the same time, and a single surface with dimensions of 23 mm × 10 mm (see Figure 1) for each sample was exposed to the slurry stream at impact angles of 30°, 45°, 60°, and 90°. To eliminate aerodynamic effects on the slurry stream, the slurry test chamber was evacuated by a vacuum system (up to 28 cm Hg). The second set of experiments (erosion-corrosion tests) followed the same procedure, but samples were exposed to a slurry with seawater (tap water + 3.5% NaCl) instead of tap water. To replenish the consumed slurry during any set of experiments, predetermined amounts of pure water (or seawater) and SiO2 sand flowed and were stirred continuously in the slurry tank.   At each test condition, five measurement intervals were carried out, and the weight of the specimen was very carefully measured before and after each interval using a precision balance with an accuracy of ±0.1 mg. Two specimens were tested in each test condition, and the average mass loss value was reported. All of the slurry erosion tests were carried out at an ambient temperature of about 25 • C and a relative humidity of 30-40%.
At the same test time, specimens at different impact angles were not subjected to the same mass of erodent. Therefore, in this test rig, a comparison between the different impact angles was carried out depending upon the mass of the erodent, i.e., all samples at each impact angle should be impacted by the same amount of erodent and not the same test-time. Table 4 illustrates the mass of the erodent that impacts the surface of the sample and the corresponding test time at each impact angle. If the interval mass of the erodent was 28.2 g of SiO 2 particles, then the test time was 15, 17.5, 21.6, and 30.9 min at impact angles 90 • , 60 • , 45 • , and 30 • , respectively. The mass of the erodent and the corresponding test time at each impact angle was determined according to Equation (1) [52]: where θ o is the angle between the surface plane of the specimen and the horizontal plane; l is the length of wear specimen surface in m; A n is the area of the orifice in m 2 ; C w is the weight fraction of solid particles in the water; ρ w is the water density in kg/m 3 ; D is the rotational diameter of the wear specimen in m; Q is the volume flow rate of slurry in m 3 /min; N is the rotational speed of the wear specimen in rpm.  Figure 6 shows an SEM image of the morphology of the external surface of as-built SLMed Ti-6Al-4V specimens. The image reveals the presence of un-melted and partially melted powder particles adhering to the surface due to the high scan speed and rapid cooling, causing increased surface roughness which is a common characteristic in the SLM process.

As-Built Microstructure
The microstructure of as-built SLMed Ti-6Al-4V is shown in Figure 7. It can be seen in Figure 7 that the microstructure consisting of a fine acicular α' martensite originated from the prior β grain boundaries. This microstructure evolved due to the martensitic transformation of the bcc β-phase as a consequence of the inherent rapid heating and cooling of the SLM process. These findings are in accordance with reported work [38][39][40][41][42][43]. As a consequence, the as-built Ti-6Al-4V possesses high hardness and strength, giving it greater wear resistance compared to its as-cast counterpart [55]. SLM process parameters, including the laser type and energy, scan parameters and strategy, powder characteristics, and layer thickness have strong influences on the surface quality and microstructure [41]. The microstructure of as-built SLMed Ti-6Al-4V is shown in Figure 7. It can be seen in Figure 7 that the microstructure consisting of a fine acicular α' martensite originated from the prior β grain boundaries. This microstructure evolved due to the martensitic transformation of the bcc β-phase as a consequence of the inherent rapid heating and cooling of the SLM process. These findings are in accordance with reported work [38][39][40][41][42][43]. As a consequence, the as-built Ti-6Al-4V possesses high hardness and strength, giving it greater wear resistance compared to its as-cast counterpart [55]. SLM process parameters, including the laser type and energy, scan parameters and strategy, powder characteristics, and layer thickness have strong influences on the surface quality and microstructure [41].  Figures 8a,b and 9a,b show the accumulative mass loss of specimens as a result of slurry erosion tests using pure water and artificial seawater, respectively, at impact angles of 30°, 45°, 60°, and 90°. By increasing the striking mass of the erodent (increasing the test time) on the samples, the mass loss increased dramatically with an almost constant rate for all impact angles-30°, 45°, 60°, and 90°-and for both mediums (pure water and seawater). Also, from Figures 8b and 9b, it is clear that the amount of mass loss decreased as the impact angle increased. The highest mass loss value due to slurry erosion of both mediums (pure water, and seawater) was recorded at an impact angle of 30°, while the lowest value was recorded at an impact angle of 90°.  The microstructure of as-built SLMed Ti-6Al-4V is shown in Figure 7. It can be seen in Figure 7 that the microstructure consisting of a fine acicular α' martensite originated from the prior β grain boundaries. This microstructure evolved due to the martensitic transformation of the bcc β-phase as a consequence of the inherent rapid heating and cooling of the SLM process. These findings are in accordance with reported work [38][39][40][41][42][43]. As a consequence, the as-built Ti-6Al-4V possesses high hardness and strength, giving it greater wear resistance compared to its as-cast counterpart [55]. SLM process parameters, including the laser type and energy, scan parameters and strategy, powder characteristics, and layer thickness have strong influences on the surface quality and microstructure [41].  Figures 8a,b and 9a,b show the accumulative mass loss of specimens as a result of slurry erosion tests using pure water and artificial seawater, respectively, at impact angles of 30°, 45°, 60°, and 90°. By increasing the striking mass of the erodent (increasing the test time) on the samples, the mass loss increased dramatically with an almost constant rate for all impact angles-30°, 45°, 60°, and 90°-and for both mediums (pure water and seawater). Also, from Figures 8b and 9b, it is clear that the amount of mass loss decreased as the impact angle increased. The highest mass loss value due to slurry erosion of both mediums (pure water, and seawater) was recorded at an impact angle of 30°, while the lowest value was recorded at an impact angle of 90°.   Although titanium alloys have good resistance to corrosion, seawater is considered a severe medium for most commonly used materials, and it influences the amount of material lost due to corrosion. As the corrosion of specimens in pure water is negligible [56], the mass loss due to slurry erosion using pure water is referred to as erosion mass loss (WE). The mass loss due to slurry erosion using seawater is referred to as the erosion-corrosion mass loss (WEC). The erosion-corrosion mass loss (WEC) was found to vary similarly to the erosion mass loss (WE) at impact angles of 30°, 45°, 60°, and 90°. At all studied impact angles-30°, 45°, 60°, and 90°-the mass loss due to slurry erosion in seawater was always higher than that due to slurry erosion in pure water, i.e., WEC was higher than WE at all impact angles (see Figure 10). Also, it is clear from Figure 10 that when the impact angle increased, the WEC and WE curves and their error bars approached each other. This indicates that the effect of corrosion decreases as the impact angle increases.  Although titanium alloys have good resistance to corrosion, seawater is considered a severe medium for most commonly used materials, and it influences the amount of material lost due to corrosion. As the corrosion of specimens in pure water is negligible [56], the mass loss due to slurry erosion using pure water is referred to as erosion mass loss (WE). The mass loss due to slurry erosion using seawater is referred to as the erosion-corrosion mass loss (WEC). The erosion-corrosion mass loss (WEC) was found to vary similarly to the erosion mass loss (WE) at impact angles of 30°, 45°, 60°, and 90°. At all studied impact angles-30°, 45°, 60°, and 90°-the mass loss due to slurry erosion in seawater was always higher than that due to slurry erosion in pure water, i.e., WEC was higher than WE at all impact angles (see Figure 10). Also, it is clear from Figure 10 that when the impact angle increased, the WEC and WE curves and their error bars approached each other. This indicates that the effect of corrosion decreases as the impact angle increases. Although titanium alloys have good resistance to corrosion, seawater is considered a severe medium for most commonly used materials, and it influences the amount of material lost due to corrosion. As the corrosion of specimens in pure water is negligible [56], the mass loss due to slurry erosion using pure water is referred to as erosion mass loss (W E ). The mass loss due to slurry erosion using seawater is referred to as the erosion-corrosion mass loss (W EC ). The erosion-corrosion mass loss (W EC ) was found to vary similarly to the erosion mass loss (W E ) at impact angles of 30 • , 45 • , 60 • , and 90 • . At all studied impact angles-30 • , 45 • , 60 • , and 90 • -the mass loss due to slurry erosion in seawater was always higher than that due to slurry erosion in pure water, i.e., W EC was higher than W E at all impact angles (see Figure 10). Also, it is clear from Figure 10 that when the impact angle increased, the W EC and W E curves and their error bars approached each other. This indicates that the effect of corrosion decreases as the impact angle increases. The previous results can be explained by considering the slurry erosion mechanism at each impact angle. At relatively small impact angles (e.g., θ = 30°), the dominant slurry erosion mechanism is plowing, whereas the impacted solid particle forms a relatively long crater on the surface of the sample with some depth, causing chips to be accumulated on both sides of this crater and in front of it, as shown in Figure 11. When the impact angle increased, the length of the slurry erosion craters decreased. The impacting force has two components, the horizontal impacting force component and the vertical impacting force component, and both of these components work simultaneously to form the slurry erosion crater. At low impact angles, the high horizontal force component causes long craters, while at high impact angles (e.g., θ = 90°) the vertical force component causes penetrations, micro-forging, and extrusions on the surface of the specimens [54,57]. In between, at medium impact angles (e.g., θ = 45°, and θ = 60°) the predominant mechanisms are a mixture of plowing and microcutting mechanisms (see Figure 11). Furthermore, because of the high strength and ductility of the SLMed Ti-6Al-4V alloy compared to other materials, such as low alloy steel, the SLMed Ti-6Al-4V alloy has a high resistance for penetration but a weakness for scraping. Therefore, the effect of the vertical impacting force component is insignificant compared with the effect of the horizontal impacting force component. Accordingly, at an impact angle of 30° the mass loss was found to be the highest, while at an impact angle of 90°, the mass loss was the lowest. This suggests that the erosion mechanism is similar to slurry erosion using pure water slurry or sodium chloride water (seawater) slurry [58]. The previous results can be explained by considering the slurry erosion mechanism at each impact angle. At relatively small impact angles (e.g., θ = 30 • ), the dominant slurry erosion mechanism is plowing, whereas the impacted solid particle forms a relatively long crater on the surface of the sample with some depth, causing chips to be accumulated on both sides of this crater and in front of it, as shown in Figure 11. When the impact angle increased, the length of the slurry erosion craters decreased. The impacting force has two components, the horizontal impacting force component and the vertical impacting force component, and both of these components work simultaneously to form the slurry erosion crater. At low impact angles, the high horizontal force component causes long craters, while at high impact angles (e.g., θ = 90 • ) the vertical force component causes penetrations, micro-forging, and extrusions on the surface of the specimens [54,57]. In between, at medium impact angles (e.g., θ = 45 • , and θ = 60 • ) the predominant mechanisms are a mixture of plowing and micro-cutting mechanisms (see Figure 11). Furthermore, because of the high strength and ductility of the SLMed Ti-6Al-4V alloy compared to other materials, such as low alloy steel, the SLMed Ti-6Al-4V alloy has a high resistance for penetration but a weakness for scraping. Therefore, the effect of the vertical impacting force component is insignificant compared with the effect of the horizontal impacting force component. Accordingly, at an impact angle of 30 • the mass loss was found to be the highest, while at an impact angle of 90 • , the mass loss was the lowest. This suggests that the erosion mechanism is similar to slurry erosion using pure water slurry or sodium chloride water (seawater) slurry [58]. Additionally, through measuring the microhardness along with the slurry eroded pathway in the middle of the surface, it was found that the microhardness value increased when the impact angle increased (Figure 12), unlike the unworn areas that were not exposed to any strikes, which showed the lowest microhardness value (412 HV). The measured microhardness values at impact angles of 30°, 45°, 60°, and 90° were 490, 608.5, 650.8, and 757.6 HV, respectively, for pure water slurry and 442.8, 495.3, 522.3, and 755.7 HV, respectively, for seawater slurry. It is clear from the microhardness results at all impact angles that the increased surface hardness resulted in lower mass losses for both erosive and corrosive tests. A higher microhardness can effectively block plowing, micro-cutting, penetration, and the separation of formed chips.

Eroded Surface Characteristics and Erosion Mechanisms
Despite the excellent corrosion resistance of the Ti-6Al-4V alloy, its slurry erosion-corrosion resistance in a corrosive environment such as seawater is weaker than that in a pure water environment at all impact angles. The superior corrosion resistance of Ti-6Al-4V alloy is due to the passive oxide film that protects it from corrosive agents such as seawater. However, under the consecutive impacts of sand particles on the surface of the alloy, the formed passive film is destroyed and removed, causing subsequent exposure of fresh active material to seawater.
Natively formed TiO2 films on titanium alloys, in general, have poor mechanical properties and they can be easily fractured under slurry erosion conditions or any other mechanical actions. Due to the successive slurry impacts, the distortion of the oxide layer and the reformation of the passive oxide layer processes, which cause a sustained dissolution of the underlying metal, result in gradual consumption of the material [56,[59][60][61][62]. Furthermore, the native passive film resistance can change due to the film's structural changes or changes in the electrical conductivity or ionic components of the film. The formed vanadium oxide V2O5 on the surface of the Ti-6Al-4V alloy strongly dissolves due to the presence of Clions in the seawater [62,63]. Dissolution of vanadium oxide V2O5 results in the creation and diffusion of vacancies in the oxide layer of the SLMed Ti-6Al-4V alloy. Therefore, the change in the passive behavior of SLMed Ti-6Al-4V alloy with slurry erosion testing using seawater (contain Cl − ions) may be the cause of the lower microhardness values and high mass losses compared to those of the same material tested in pure water. Additionally, through measuring the microhardness along with the slurry eroded pathway in the middle of the surface, it was found that the microhardness value increased when the impact angle increased (Figure 12), unlike the unworn areas that were not exposed to any strikes, which showed the lowest microhardness value (412 HV). The measured microhardness values at impact angles of 30 • , 45 • , 60 • , and 90 • were 490, 608.5, 650.8, and 757.6 HV, respectively, for pure water slurry and 442.8, 495.3, 522.3, and 755.7 HV, respectively, for seawater slurry. It is clear from the microhardness results at all impact angles that the increased surface hardness resulted in lower mass losses for both erosive and corrosive tests. A higher microhardness can effectively block plowing, micro-cutting, penetration, and the separation of formed chips.
Despite the excellent corrosion resistance of the Ti-6Al-4V alloy, its slurry erosion-corrosion resistance in a corrosive environment such as seawater is weaker than that in a pure water environment at all impact angles. The superior corrosion resistance of Ti-6Al-4V alloy is due to the passive oxide film that protects it from corrosive agents such as seawater. However, under the consecutive impacts of sand particles on the surface of the alloy, the formed passive film is destroyed and removed, causing subsequent exposure of fresh active material to seawater.
Natively formed TiO 2 films on titanium alloys, in general, have poor mechanical properties and they can be easily fractured under slurry erosion conditions or any other mechanical actions. Due to the successive slurry impacts, the distortion of the oxide layer and the reformation of the passive oxide layer processes, which cause a sustained dissolution of the underlying metal, result in gradual consumption of the material [56,[59][60][61][62]. Furthermore, the native passive film resistance can change due to the film's structural changes or changes in the electrical conductivity or ionic components of the film. The formed vanadium oxide V 2 O 5 on the surface of the Ti-6Al-4V alloy strongly dissolves due to the presence of Cl − ions in the seawater [62,63]. Dissolution of vanadium oxide V 2 O 5 results in the creation and diffusion of vacancies in the oxide layer of the SLMed Ti-6Al-4V alloy. Therefore, the change in the passive behavior of SLMed Ti-6Al-4V alloy with slurry erosion testing using seawater (contain Cl − ions) may be the cause of the lower microhardness values and high mass losses compared to those of the same material tested in pure water. The slurry erosion test at a normal impact angle is similar to the wet shot peening process. For both of these tests, the specimens are impacted in the vertical direction with high-speed solid particles. Shot peening increases the strength and durability of the surface and makes it more resistant to many wear types, which may be the cause of the low mass loss of the SLMed Ti-6Al-4V alloy at a normal impact angle (i.e., θ = 90°); however, reducing the impact angle from the normal angle causes the strength of the sample surfaces to decrease, which leads to greater mass loss. This further illustrates the behavior of slurry erosion at different impact angles of the SLMed Ti-6Al-4V alloy. Figures 13 and 14 show the optical microscopy images of the solid particles impacting the specimen surfaces at impact angles of 30°, 45°, 60°, and 90° using pure water slurry and seawater slurry, respectively. Figure 13 also shows the large size of micro-distortion that occurred on the surface of the samples as a result of the successive solid particle strikes. These micro-distortions increased slurry erosion resistance and surface roughness. It can also be seen from Figure 14 that small pits over the SLMed Ti-6Al-4V alloy surfaces were observed at all impact angles; therefore, the mode of degradation under the seawater environment is pitting corrosion. The slurry erosion test at a normal impact angle is similar to the wet shot peening process. For both of these tests, the specimens are impacted in the vertical direction with high-speed solid particles. Shot peening increases the strength and durability of the surface and makes it more resistant to many wear types, which may be the cause of the low mass loss of the SLMed Ti-6Al-4V alloy at a normal impact angle (i.e., θ = 90 • ); however, reducing the impact angle from the normal angle causes the strength of the sample surfaces to decrease, which leads to greater mass loss. This further illustrates the behavior of slurry erosion at different impact angles of the SLMed Ti-6Al-4V alloy. Figures 13 and 14 show the optical microscopy images of the solid particles impacting the specimen surfaces at impact angles of 30 • , 45 • , 60 • , and 90 • using pure water slurry and seawater slurry, respectively. Figure 13 also shows the large size of micro-distortion that occurred on the surface of the samples as a result of the successive solid particle strikes. These micro-distortions increased slurry erosion resistance and surface roughness. It can also be seen from Figure 14 that small pits over the SLMed Ti-6Al-4V alloy surfaces were observed at all impact angles; therefore, the mode of degradation under the seawater environment is pitting corrosion.

Conclusions
In this study, the erosion and erosion-corrosion behaviors of as-built SLMed Ti-6Al-4V alloy were examined using pure water slurry and seawater slurry, respectively, at different impact angles. The microstructure of as-built SLMed Ti-6Al-4V appeared as a needle-like martensitic α' phase. The

Conclusions
In this study, the erosion and erosion-corrosion behaviors of as-built SLMed Ti-6Al-4V alloy were examined using pure water slurry and seawater slurry, respectively, at different impact angles. The microstructure of as-built SLMed Ti-6Al-4V appeared as a needle-like martensitic α' phase. The

Conclusions
In this study, the erosion and erosion-corrosion behaviors of as-built SLMed Ti-6Al-4V alloy were examined using pure water slurry and seawater slurry, respectively, at different impact angles. The microstructure of as-built SLMed Ti-6Al-4V appeared as a needle-like martensitic α' phase. The impact angle was found to dramatically affect the erosion and erosion-corrosion behaviors of SLMed Ti-6Al-4V alloy. The SLMed Ti6Al4V alloy showed ductile erosion behavior, and the maximal mass losses for erosion and erosion-corrosion occurred at an impact angle of 30 • . The minimal mass losses for erosion and erosion-corrosion occurred at an impact angle of 90 • . SEM investigations of the eroded surfaces of the SLMed Ti6Al4V alloy specimens revealed the dominant erosion and erosion-corrosion mechanisms, such as ploughing, microcutting, and penetrations, at each impact angle. The microhardness increased as the impact angle increased in both erosion and erosion-corrosion tests. The vanadium oxide V 2 O 5 formed on the surface of the Ti-6Al-4V alloy strongly dissolved due to the presence of Clions in the seawater. Therefore, the mass losses of slurry erosion-corrosion was higher than that of slurry erosion at all impact angles.