Influence of Scanning Speed on the Microstructure and Performance of a TiAl Coating Formed by Selective Laser Melting on a Titanium Alloy Surface

Abstract

In order to improve the surface performance of aviation titanium alloy and enhance the safety level and reliability of the overall structure during the service phase, selective laser melting (SLM) technology was used to prepare TiAl coatings on the surface of TC4 titanium alloy. The influence of laser scanning speed on the coating performance was systematically investigated. The microstructure, element distribution, phase composition, and microhardness of the coatings at different scanning speeds were characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD), and energy-dispersive X-ray spectroscopy (EDS). The results show that when the scanning speed is 1000 or 1100 mm/s, multiple penetrating longitudinal and transverse cracks are formed at the interface between the coating and the substrate; at 1200 mm/s, the coating and substrate interface is flat and crack-free, with a stable Ti/Al mass percentage of 1:12.9, and the Sc element is fully dissolved to form an Al₃Sc strengthening phase, with Ti–Al phases mainly composed of Ti₃Al and Ti₂Al, showing uniform element distribution, stable phase composition, and a high-quality structure with a maximum hardness of 565.8 HV_0.2. When the scanning speed is increased to 1300 mm/s, the number of cracks is reduced compared with 1000 and 1100 mm/s, mostly short longitudinal cracks, with the appearance of delamination on the outer surface of the coating and poor flatness. The control law of scanning speed on the process, structure, and performance of TiAl coatings formed by SLM is clarified, providing a theoretical basis and practical guidance for the preparation of high-performance TiAl coatings using SLM technology.

1. Introduction

Titanium alloys are the preferred choice for core components such as aircraft engine blades and casings due to their excellent high strength-to-weight ratio and superior corrosion resistance, holding a significant position in the aerospace, aviation, and maritime fields. − However, traditional titanium alloy components experience noticeable performance degradation when operating in high-temperature oxidation environments and facing multiple physical effects such as high-cycle fatigue and erosion, severely limiting their long-term stable service under extreme conditions. − The formation of functional coatings with specific properties on the surface of titanium alloy substrates not only fully retains the mechanical properties of the substrate but also provides special protective functions to effectively address harsh environments. − Therefore, surface engineering technology has become an important approach to enhancing the reliability of service for aerospace titanium alloys and has gained widespread attention from the academic and industrial sectors. Common technologies for forming coatings on titanium alloy surfaces include physical vapor deposition coatings, thermal spraying, laser cladding technology, and micro-arc oxidation coatings. − Hein et al. applied coatings via physical vapor deposition (PVD) on Ti-6Al-7Nb materials, suggesting that PVD coatings can improve high-cycle fatigue performance. PVD technology typically requires long deposition times, relatively low production efficiency, and high equipment costs. Thirumalvalavan et al. used high-speed oxygen fuel thermal spraying to apply nano-coatings on titanium alloy surfaces, gradually increasing wear resistance with increasing powder spray distance. The main drawbacks of thermal spraying coatings include insufficient bond strength and high porosity. , Kumari et al. utilized fiber laser cladding technology to prepare composite coatings on titanium alloys, with crack-free composite coating structures beneficial for enhancing the performance and long-term durability of orthopedic implants. However, during the laser cladding process, even with a certain extension of the protective effect of the gas protection device, some areas of the coating may be exposed to air cooling, leading to the formation of brittle oxides in the coating. − Muhaffel et al. prepared micro-arc oxidation coatings on a cp-Ti substrate using a bipolar pulse power supply, generating a coating of mixed crystalline and amorphous compounds during the anodic cycling process, where an increase in applied voltage results in increased coating thickness and crack formation. The preparation process of titanium alloy micro-arc oxidation coatings requires strict process parameter requirements, with slight deviations in parameter control easily leading to defects such as uneven coating thickness and excessively high porosity. Although micro-arc oxidation coatings have high hardness, they exhibit significant brittleness and poor impact resistance, making them prone to cracking under dynamic loads or impact conditions. −

Selective laser melting (SLM) is currently the most widely used metal additive manufacturing technology, which has the advantages of complex component forming and short manufacturing cycles. − In this technology, metal powders are formed in a completely oxygen-free environment, and the generation of brittle oxide is effectively avoided in a 99.99% argon protective atmosphere. Meanwhile, in the rapid solidification process of SLM technology, solute atoms are difficult to achieve long-range diffusion and rearrangement, which can inhibit the precipitation of some unfavorable equilibrium phases and generate new phases that do not exist in the equilibrium state, providing the possibility of improving the high-temperature mechanical properties of alloys. − Laser power, laser scanning speed, and scanning pattern are the main process parameters in the SLM preparation process, among which laser power plays a key role in the coating performance. Therefore, it is of great significance to study the performance of TiAl coatings prepared by titanium alloy laser selective melting formation and explore the influence of different scanning speeds on coating performance. In this study, the surface modification coating of titanium alloy was designed and prepared using laser selective melting technology, and the mechanism of scanning speed affecting the performance of SLM coatings on titanium alloy surfaces was explored through multi-scale experiments and characterization methods, further deepening the understanding of the structural safety and reliability of titanium alloys.

2. Materials, Equipment and Methods

The powder AlMgSc prepared by vacuum atomization method was used as raw material in the experiment. The powder has excellent spherical and smooth surface, effectively optimizing the fluidity and bulk density during the powder distribution process, specific chemical composition is shown in Table . The precise proportion of each alloy element provides an important basis for the performance control of subsequent coatings.

1. Chemical Compositions of AlMgSc (Mass Fraction, %).

main component/%			impurity/%, not more than
Al	Mg	Sc	Zr	Si	Fe	Cu
Bal.	5.09	0.84	0.34	0.03	0.07	0.01

The experiment was carried out using the DMP Flex350 selective laser melting device. Its main components include a 500 W fiber laser, with a maximum scanning speed of 7 m/s, precise control of spot diameter at 100 μm, and powder distribution thickness ranging from 10 μm to 100 μm. In addition, the built-in atmospheric monitoring system supports an oxygen concentration in the working environment ≤0.0025%, effectively preventing oxidation of metal powders at high temperatures and ensuring the cleanliness of the formed parts. The substrate size of 275 mm × 275 mm provides sufficient space for the preparation of various samples, supporting the high-quality formation of complex geometric structures. The corresponding experimental equipment and coating formation principles are shown in Figure , which displays the calculation formula of laser specific energy E, where P is the laser power, v is the scanning speed, and D is the spot diameter. The thermal chemical reaction formula for TiAl coating formation is xTi + yAl = TixAly.

1.

Experimental equipment and working principle.

The TC4 base is installed on the workbench of the selective laser melting equipment. With the help of the metal powder AlMgSc, argon gas is introduced for protection, and thecoating is prepared by layer-by-layer melting deposition. The geometric dimensions of the coating are set to 15 mm × 15 mm × 1 mm, and the main parameters of the forming process are shown in Table . The micro-morphology of the coating and base interface is analyzed using a JSM-IT500 scanning electron microscope (SEM) equipped with an energy dispersive spectrometer (EDS), and the distribution characteristics of elements such as Ti, Al, Mg, and Sc in the interface transition zone are explored through line scanning analysis. The phase composition of the coating and interface areas is determined using a Rigaku Ultima IV X-ray diffractometer (XRD). An HV-1000A Vickers hardness tester is selected to continuously indent the coating and base interface area at 50 μm intervals, with a load of 1.96 N (corresponding to 0.2 kgf), a hold time of 10 s, and 5 valid data points are taken for each test group to obtain the average value as the hardness value of that area.

2. SLM Main Test Parameters.

sample no.	laser power/W	scanning speed/mm/s	spot diameter/mm	layer thickness/mm	strip overlap/mm
1	370	1000	0.1	0.03	0.02
2	370	1100	0.1	0.03	0.02
3	370	1200	0.1	0.03	0.02
4	370	1300	0.1	0.03	0.02

3. Results and Discussion

3.1. Macroscopic and Microscopic

Figure shows the dynamic forming process and macroscopic morphology features of TiAl coating on TC4 substrate surface by SLM under different laser scanning speeds. In Figure a, the layer-by-layer construction process of the coating on the TC4 substrate is intuitively presented. The laser beam precisely acts on the pre-placed AlMgSc powder, and the coating is metallurgically bonded to the substrate through selective melting and solidification. In Figure b, the coatings labeled ①–④ correspond to scanning speeds of 1000, 1100, 1200, and 1300 mm/s, respectively. The outer surface of the coatings presents a uniform bright white color, consistent with the color of aluminum alloy, and exhibits excellent gloss and high cleanness. This is because within this power range, the input laser energy and material thermodynamic response reach dynamic equilibrium: the laser energy is efficiently absorbed by the AlMgSc powder, which rapidly heats the powder particles to the melting temperature. In the melting pool, the molten metal spreads uniformly and solidifies stably under the combined action of surface tension and convection. During solidification, the nucleation rate matches the crystal growth rate, promoting the formation of a dense continuous coating structure. The obvious patterns on the outer surface of the coating can be attributed to during the scanning process, the laser moves at a certain rate, causing the melting pools to periodically form and dissipate along the laser scanning trajectory, with thermal overlap zones between adjacent melting pools. During solidification, the molten metal undergoes uneven contraction due to temperature gradients and solute redistribution. Simultaneously, excessive melting causes some already solidified areas to undergo remelting due to secondary heat input. The remelted region and the non-remelted region exhibit differences in solidification shrinkage rate and crystal orientation evolution. These differences accumulate through multiple layers, ultimately forming observable patterns on the outer surface of the coating. −

2.

Coating forming process and macroscopic morphology at different laser powers. a. Coating forming process; b. Macroscopic morphology of the coating.

The organizational characteristics of TiAl coatings formed by SLM at different scanning speeds observed under a 200× microscope are shown in Figure , reflecting the regulatory rules of coating formation by scanning speed. The direct influence of laser beam energy density on the behavior of the molten pool, and the scanning speed, as the key control parameter of the beam energy density, is calculated according to formula (1), directly determines the melting and solidification behavior of the molten pool, and thereby affects the forming quality of the coating.

$$\mathrm{E}=\mathrm{P}/(\mathrm{s}\times \mathrm{v}\times \mathrm{t})$$ 1

3.

Microstructure of coatings formed at different scanning speeds (magnification: 200×). a. 1000 mm/s; b. 1100 mm/s; c. 1200 mm/s; d. 1300 mm/s.

E represents the laser energy density (J/mm³), P denotes the laser power (W), t indicates the powder layer thickness (mm), s stands for the scanning interval (mm), and v is the scanning speed (mm/s). When the scanning speed is 1000 mm/s, as shown in Figure a, the laser beam acts on the pre-placed powder at a lower rate, the powder absorbs a higher energy density per unit time, the molten pool exists for a long time, the melt flow is sufficient, and a coating with a thickness of about 714 μm can be formed during solidification. However, the high energy input causes a large heat-affected zone at the edge of the molten pool, and the volume shrinkage and thermal stresses are prone to accumulate inside the coating during solidification. With the scanning speed increased to 1100 mm/s, as shown in Figure b, the time window of interaction between the laser and powder is shortened, the energy input changes, dynamic balance between convection and heat diffusion in the molten pool is adjusted, leading to an increase in the coating thickness to 785 μm. At this point, the match between thermal input and material solidification contraction is better than at low speeds, and the continuity of coating growth is preliminarily improved. , At a scanning speed of 1200 mm/s, as shown in Figure c, the dynamic process of laser scanning is further matched with the material’s thermodynamic response, the formation and solidification period of the molten pool is more in line with the alloy’s crystallization kinetics, the molten metal spreads evenly and the solidification interface advances steadily, not only allowing the coating thickness to reach 807 μm, but also achieving a relatively flat interface between the coating and the substrate due to the synergy between solute redistribution at the solidification front and crystal growth direction. This is attributed to the balance between thermal stresses during solidification and metallurgical bonding forces at an appropriate scanning speed. However, when the scanning speed increases to 1300 mm/s, as shown in Figure d, the uneven absorption of energy by the powder due to the excessively fast laser moving rate disrupts the stability of the temperature and flow fields in the molten pool. During solidification, layer-by-layer crystallization occurs due to insufficient heat input: the upper layer solidifies quickly, while the lower layer fails to compensate timely, and the disorderly layer-by-layer stacking of the coating occurs under high-speed scanning, resulting in a final thickness of 862.50 μm and a layered structure with poor surface quality and low flatness.

The microstructure of TiAl coating under different scanning speeds at a magnification of 500 times is shown in Figure . Scanning speed is a core parameter for regulating the energy input and solidification rate of the melt pool, and its variation directly affects the internal stress and crack initiation and propagation in the coating. Figure a shows the formed coating at a scanning speed of 1000 mm/s. The lower scanning speed results in a higher laser energy density received by the melt pool per unit time, significantly increasing the temperature gradient of the melt pool. During solidification, due to the insufficient release of thermal stress through sufficient plastic deformation caused by the difference in thermal expansion coefficients, obvious cracks are formed at the coating junctions. These cracks include vertical cracks perpendicular to the interface and horizontal cracks parallel to the interface. Among them, the vertical cracks extend to the base alloy, with a total length of about 200 μm, possibly due to the slow solidification rate of the melt pool at this speed, inadequate element diffusion at the coating-base interface, low bonding strength, continuous accumulation of thermal stress along the interface normal, breaking the interface bonding strength threshold, eventually leading to cracks penetrating the coating and extending to the base. − Figure b shows the formed coating at a scanning speed of 1100 mm/s, with a total crack length of about 530 μm, significantly larger than the total number of cracks formed in the coating at a scanning speed of 1000 mm/s. The increase in scanning speed slightly reduces the energy input to the melt pool, but the fluidity of the melt pool is weakened due to the faster cooling rate, reducing the shrinkage capacity of molten metal in the melt pool. Additionally, Sc element is prone to local segregation during rapid solidification, leading to further increase in the number of cracks at the coating junction, primarily in the form of vertical cracks. The low melting point phase of the segregated Sc element tends to crack under stress, forming longitudinally crack channels along grain boundaries. Figure c presents the formed coating at a scanning speed of 1200 mm/s, where the laser energy input and melt pool cooling rate reach a dynamic balance: moderate energy density controls the temperature gradient of the melt pool within a reasonable range, avoiding the generation of excessive thermal stress. A faster solidification rate suppresses macroscopic element segregation, promoting the formation of uniform fine equiaxed crystal structure in the coating, reducing stress concentration points at grain boundaries. Thus, besides the smooth coating-base interface, no obvious cracks are found in both the overall coating and the interface area, demonstrating excellent forming quality. Figure d depicts the formed coating at a scanning speed of 1300 mm/s, with a total crack length of about 170 μm. However, compared to 1100 mm/s, the number of cracks at this speed is significantly reduced, with most cracks being shorter in length. Further increase in scanning speed results in a substantial decrease in melt pool energy density, reduced melt pool depth, and significantly accelerated solidification rate. Although the thermal stress generated during coating solidification decreases to some extent, the excessively fast cooling rate reduces the material’s plastic deformation ability. Hence, residual thermal stress may still generate microcracks along grain boundaries in the coating, leading to the presence of several shorter vertical cracks.

4.

Microstructure of coatings formed at different scanning speeds (magnification: 500×). a. 1000 mm/s; b. 1100 mm/s; c. 1200 mm/s; d. 1300 mm/s.

3.2. Phase Analysis

The EDS elemental analysis of the bonding areas of TiAl coatings formed by SLM at different scanning speeds is shown in Figure and Table . The test areas 1–3 are respectively the outer layer of the coating, the middle part of the coating, and the bonding area of the coating. The atomic values of each element are marked with ±5% standard deviation in the figure. The scanning speed affects the diffusion of elements at the interface between the coating and the substrate, resulting in differences in the element distribution characteristics. In test area 1, only the Ti element was detected when the scanning speed was 1000 mm/s. At other scanning speeds, there was no Ti element in this area, indicating that as the scanning speed increases, it is difficult for the Ti element in the substrate to diffuse to the outermost layer of the coating. In test area 2, the Ti element was only detected at scanning speeds of 1000 and 1100 mm/s, which further illustrates this issue. The EDS analysis of the coating at a scanning speed of 1000 mm/s shows that the lower scanning speed causes the laser to act on the powder for a longer time, allowing the Al element to obtain sufficient diffusion driving force and migrate significantly into the coating interior, resulting in a Ti/Al mass percentage of 1:4.8 in test area 3. On the other hand, the excessively high melt pool temperature intensifies the volatilization loss of Mg element, with its atomic percentage being only 1.87%, while the Sc element, due to its lower diffusion coefficient, undergoes certain diffusion over a long period of high temperature, but its overall solubility is limited by the slow solidification process, with a mass fraction of only 0.31%. The EDS analysis of the coating at a scanning speed of 1100 mm/s shows that with the increase in scanning speed, the laser energy input intensity decreases, the melt pool cooling rate accelerates, the diffusion time of Ti element from the substrate to the coating is shortened, and the diffusion distance decreases, resulting in a Ti/Al mass percentage of 1:9.0 at the bonding area of the coating. The Ti element interface enrichment degree is significantly weakened. Meanwhile, the duration of the melt pool at high temperature is shortened, and the volatilization loss of Mg element is effectively suppressed, with its atomic percentage reaching 2.95%, which is the highest among all scanning speeds. Thanks to the reasonable melt pool temperature gradient, the vapor pressure loss of low-boiling-point elements is reduced. Additionally, the appropriate cooling rate promotes the sufficient solidification of Sc element in the Al matrix, reducing the grain boundary segregation phenomenon, and the atomic percentage reaches 0.37%, further optimizing the alloying degree of the coating. The EDS analysis of the coating at a scanning speed of 1200 mm/s shows that the melt pool energy input and cooling rate reach the optimal state. The moderate energy density ensures the good fluidity of the molten metal, promoting the moderate diffusion of the Al element at the interface, and the Ti/Al mass ratio at the bonding area of the coating is 1:12.9. The volatilization of Mg element is effectively controlled, with its atomic percentage rising to 4.73%, the highest among all scanning speeds. Thanks to the reasonable melt pool temperature gradient, the loss of low-boiling-point elements due to vaporization is reduced. Moreover, the appropriate cooling rate promotes the sufficient solidification of Sc element in the Al matrix, reducing the grain boundary segregation phenomenon, and the atomic percentage reaches 0.37%, further optimizing the alloying degree of the coating. The EDS analysis of the coating at a scanning speed of 1300 mm/s further increases the scanning speed, reducing the melt pool depth and accelerating the cooling rate. The diffusion of Al element is significantly inhibited, and the Ti and Al atoms at the bonding area are further adjusted to 1:10.3, with the distribution of Ti element at the interface becoming more uniform, but the overall content decreases. Although the volatilization loss of Mg element theoretically should continue to decrease due to the further shortening of the high-temperature duration, due to the insufficient melt pool energy, the fluidity of the molten metal deteriorates, and some Mg element is quickly encapsulated by the solidified grains, forming local enriched or depleted regions. , The actual detected atomic percentage is reduced to 4.11%. Sc element, due to insufficient atomic diffusion time, has a slightly decreased atomic solubility, with the atomic percentage adjusted to 0.38%, but still higher than the content at the low scanning speed. This indicates that although there is a problem of insufficient element diffusion at high scanning speeds, compared to the volatilization loss at low speeds, its retention effect for Sc element is better.

5.

EDS analysis of coatings formed at different scanning speeds. a. 1000 mm/s; b. 1100 mm/s; c. 1200 mm/s; d. 1300 mm/s.

3. EDS Spectrum Comparison of the Coatings Formed at Different Scanning Speeds.

	test area-1				test area-2				test area-3
test area element (Wt% ± 5%)	Ti	Al	Mg	Sc	Ti	Al	Mg	Sc	Ti	Al	Mg	Sc
1000 mm/s	0.55	91.72	4.36	0.54	0.29	92.14	4.05	0.40	15.74	75.17	1.87	0.31
1100 mm/s	/	89.20	4.24	0.46	0.18	91.89	4.71	0.44	6.93	62.51	2.99	0.30
1200 mm/s	/	91.90	3.71	0.22	/	90.84	4.52	0.46	6.58	84.76	4.73	0.37
1300 mm/s	/	88.61	3.83	0.48	/	91.98	4.62	0.58	8.02	82.24	4.11	0.38

Figure shows the EDS element line scanning results of SLM for preparing TiAl coatings and the substrate at different scanning speeds. The test scanning path covers the outer layer of the coating, the coating-substrate bonding zone, and the substrate area. The change in scanning speed directly alters the laser thermal input and the residence time of the molten pool, thereby regulating the element diffusion behavior at the coating-substrate interface. Figure a corresponds to a scanning speed of 1000 mm/s. At this speed, the laser thermal input is relatively sufficient, and the solidification rate of the molten pool is gentle. The Ti, Al, and Mg elements are only stably distributed in their respective areas, and the element content boundaries at the coating-substrate interface are clearly distinct. The element compositions in both regions have not undergone significant loss or migration. However, the Sc element is not constrained by the interface boundary and is uniformly dispersed throughout the coating and substrate regions, without any obvious distribution boundaries. Figure b corresponds to a scanning speed of 1100 mm/s. After a slight increase in the scanning speed, the cooling and solidification rate of the molten pool accelerates, and the activity of interface element diffusion slightly increases. The overall Ti, Al, and Mg elements still maintain a clear interface content boundary, but there is a slight fluctuation in the local element distribution. The content of Ti elements inside the coating increases locally, indicating that a small amount of substrate Ti elements have diffused and migrated to the coating side. The local content of Al and Mg elements in the coating decreases, corresponding to the diffusion of some Al and Mg elements from the coating to the interface bonding zone. At this speed, the Sc element still maintains a uniformly dispersed distribution characteristic throughout the coating and substrate regions, without any local enrichment or segregation phenomena. Figure c corresponds to a scanning speed of 1200 mm/s. With further increase in the scanning speed, the laser action time shortens, and the driving force for element migration across the interface decreases. The Ti, Al, and Mg elements are basically stable in their original distribution areas. The interface element boundary characteristics at the coating-substrate interface are significant, and the element compositions in each region remain stable. The Sc element is still uniformly dispersed in the coating and substrate regions, and the distribution pattern has not changed. Figure d corresponds to a scanning speed of 1300 mm/s. At the highest scanning speed of this test, the laser thermal input action time is the shortest, and the rapid solidification characteristic of the molten pool causes local redistribution of interface elements. Compared with other scanning speed samples, Ti, Al, and Mg elements appear significantly localized enrichment at the coating-substrate interface region. The uniformity of element distribution decreases, while the distribution characteristics of the Sc element are not affected by the gradient change in the scanning speed, and it remains uniformly dispersed throughout the coating and substrate regions.

6.

Energy spectrum line scanning analysis of the coatings formed under different scanning speeds. a. 1000 mm/s; b. 1100 mm/s; c. 1200 mm/s; d. 1300 mm/s.

The phase composition of the coating at different scanning speeds is shown in Figure . Different scanning speeds further result in changes in the types of phases. Figure a shows the XRD analysis results of the coating at a scanning speed of 1000 mm/s, where the lower scanning speed allows for a longer laser energy interaction time and insulation time, providing sufficient thermodynamic conditions for phase transformation. At high temperatures, deep diffusion of Ti elements in the substrate and Al elements in the coating leads to the formation of various Ti–Al intermetallic compounds: TiAl₂, Ti₅Al₁₁, TiAl, TiAl₃. Among them, the formation of Ti₅Al₁₁ originates from the sufficient proportion reaction of Ti and Al at high temperatures, while the formation of TiAl depends on equilibrium phase transformation at a relatively slow cooling rate. , Due to the increased diffusion coefficient at high temperatures, Sc elements combine with Al elements to form stable Al₃Sc strengthening phase. Additionally, the relatively fast cooling rate at the edge of the melt retains a small amount of unreacted Al elements. Figure b shows the XRD analysis of the coating at a scanning speed of 1100 mm/s, where with the increase in scanning speed, the elemental diffusion time shortens, leading to significant adjustments in the phase evolution of the Ti–Al series: the high-temperature stable phase Ti₅Al₁₁ is difficult to fully nucleate due to insufficient insulation time. Conversely, Ti₂Al and Ti₃Al₅, which are easier to nucleate in the medium-temperature region, disappear. With the accelerated cooling rate, TiAl phase disappears, resulting in insufficient phase transformation driving force, and only the more stable TiAl₂, Ti₃Al, TiAl₃ and other intermetallic compounds are retained. Due to the weakening diffusion kinetic conditions, some Sc atoms do not fully combine with Al, leading to a small amount of unreacted elemental Sc. Furthermore, due to inadequate supply of Sc elements, Al₃Sc phase does not appear in the XRD pattern, only leaving elemental Ti and Al. Figure c presents the XRD analysis of the coating at a scanning speed of 1200 mm/s, where the moderate peak temperature not only ensures moderate reaction between Ti and Al elements to form Ti₃Al, Ti₂Al, Ti₅Al₁₁ and other TiAl compounds, which are both stable and have strong bonding strength, but also prevents elements from over-volatilizing due to high temperatures. Sc elements combine with Al to form Al₃Sc, interface element reaction is sufficient, and no elemental Sc appears, only a small amount of elemental Ti and Al are retained. The diffusion coefficient of Sc in Al directly affects the efficiency of its reaction with aluminum. The higher the diffusion coefficient, the greater the probability of scandium reacting with aluminum and the easier it is to form Al₃Sc. The cooling rate corresponding to a scanning speed of 1200 millimeters per second regulates the formation process of precipitates. An appropriate cooling rate can inhibit excessive diffusion of scandium, promote the stable precipitation of Al₃Sc, and avoid the formation of other impurity phases. This is consistent with the result in XRD analysis, which only detected Al₃Sc and no free scandium elements. Figure d shows the XRD analysis of the coating at a scanning speed of 1300 mm/s. The higher scanning speed suppresses the diffusion of elements and phase transformation processes. The main phases in the TiAl series are TiAl₂, Ti₂Al, and Ti₃Al₅. The detection of Ti₃Al₅ is due to the non-equilibrium phase transformation under rapid cooling, while Ti₃Al decreases in nucleation due to insufficient diffusion of Ti elements, only retaining a small amount. Due to the short duration of high temperatures in the melt pool, incomplete reaction occurs between Mg and Al elements, resulting in elemental Mg. At the interface, a small amount of elemental Ti and Al are still retained, and due to non-equilibrium solidification, the phase composition shows obvious metastable characteristics.

7.

XRD analysis of coatings formed at different scanning speeds. a. 1000 mm/s; b. 1100 mm/s; c. 1200 mm/s; d. 1300 mm/s.

3.3. Microhardness of the Coating

Figure shows the microhardness analysis of the coating-interface region at different scanning speeds. Intermetallic compounds are mostly formed at the coating-substrate interface. Therefore, the maximum hardness of the coating is concentrated near the interface, gradually decreasing toward the outer layer of the coating. The outer layer of the coating is the original aluminum alloy that did not participate in the heat reaction, so its hardness is similar to that of the aluminum alloy itself. At a scanning speed of 1000 mm/s, the maximum hardness of the coating is only 459.9 HV_0.2, which is only 24.3% higher than the 370.1 HV_0.2 of the TC4 alloy base. When the scanning speed is too slow, the input heat accumulation in the SLM process is too high, the cooling rate is insufficient, interfacial intermetallic compound grains tend to coarsen, excessive element diffusion leads to uneven phase composition, and the effects of grain refinement strengthening and dispersion strengthening are significantly weakened. Increasing the scanning speed to 1100 mm/s, the maximum hardness of the coating increases to 523.8 HV_0.2, still concentrated near the interface. At this point, with the increased scanning speed, the heat input increases and the cooling rate accelerates, promoting an increase in the nucleation rate of intermetallic compounds and grain refinement, gradually enhancing the strengthening effect. When the scanning speed reaches 1200 mm/s, the maximum hardness of the coating reaches 565.8 HV_0.2, the peak value at all scanning speeds. This speed range represents the optimal match between heat input and cooling rate in the SLM process, with the best dynamic conditions for element diffusion, uniform nucleation of intermetallic compounds at the interface, fine grains and dense phase distribution, and the full play of the effects of grain refinement, dispersion strengthening, and interface strengthening, achieving maximum hardness. Further increasing the scanning speed to 1300 mm/s, the maximum hardness of the coating decreases to 485.8 HV_0.2. This may be due to the excessive scanning speed leading to insufficient heat input, shortened pool existence time, inadequate element diffusion, restricted nucleation of intermetallic compounds, suppression of the interface strengthening mechanism, and a significant drop in hardness. To verify the reproducibility of the above results, in this study, 5 valid data points will be selected from each test group to calculate the average value of that area as the hardness value of that region. The hardness distribution curves of all repeated samples have the same trend, and the relative deviation of the peak hardness is within ±5%, indicating that the correspondence between the process parameters and the hardness results is stable and reliable. The hardness comparison between the titanium–aluminum coating formed by SLM-formed TiAl coating and other coating processes is shown in Table . Through our comparison, we found that the hardness performance of the titanium–aluminum coating produced by SLM is more advantageous. It not only ensures the required hardness level of the coating but also takes into account the adaptability during use. Compared with other coating processes, this hardness performance better meets the actual needs of advanced titanium alloy surface engineering and demonstrates the technical rationality of the SLM process in controlling coating hardness.

8.

Cross-sectional hardness analysis of coatings formed with different scanning speeds. (a) Coating hardness distribution area; (b) Coating hardness area analysis.

4. Comparison of the Performance of Ti–Al Coatings Formed by SLM with Those Obtained by Other Processes.

Forming process	Microstructure	Hardness (HV 0.2 )	Thickness
SLM-formed Ti–Al coating	No obvious cracks	565.8	Controllable
Pressure-assisted Ti–Al coating	Too much pressure, causing the coating to peel off	378.4	Thicker
Electromagnetic field-assisted Ti–Al coating	The defect has been suppressed	433.4	Thicker
Magnetron sputtering coating	No obvious defects	484	-

3.4. Discussion and Analysis

This study reveals the control mechanism of scanning speed on the microstructure, performance of formed TiAl coatings through multidimensional characterization of macroscopic morphology, microscopic structure, element distribution, phase composition, and microhardness. The essence is that the scanning speed changes the thermodynamic state and kinetic behavior of the melt pool, thus dominating the coating from macroscopic construction to microscopic structure. The formation principles of coatings formed at different scanning speeds are shown in Figure .

9.

Forming principle of coatings formed at different scanning speeds. a. 1000 mm/s; b.1200 mm/s.

The core issue in the range of 1000–1100 mm/s is the mismatch between energy and cooling: high energy at low speed induces excessive thermal stress and element volatilization, while fast cooling at medium speed leads to compositional segregation and structural defects, both of which weaken the bond strength between the coating and the substrate, and the disordered distribution of Ti–Al compounds hinders excellent mechanical properties; the advantage of 1200 mm/s as the optimal process parameter lies in achieving crack-free structure, uniform composition, and stable phases through precise control of the melt pool behavior: a flat interface ensures metallurgical bonding reliability, the homogeneous distribution of Ti₃Al and Ti₂Al compounds enhances interface bonding stability, laying the foundation for the comprehensive performance of the coating; the layered structure and metastable phases at 1300 mm/s, although with fewer cracks than at 1100 mm/s, still cannot match the coating performance under 1200 mm/s conditions. Compared with coatings formed by processes like cold spray, SLM coatings exhibit higher interface bonding strength resulting from the metallurgical bonding formed by the fully melted solidified melt pool, surpassing the mechanical bonding and plastic deformation bonding of cold spray. Although cold spray does not cause thermal damage, it cannot achieve strengthening through phase control like SLM. The analysis of cooling rate, thermal stress-induced phase transformation and crack initiation is mainly based on the microscopic observation results of the specimens, and lacks quantitative data to support the mechanism derivation. In the subsequent research, numerical simulation and thermodynamic calculations will be introduced to improve the theoretical proof of the related mechanisms.

4. Conclusions

This study systematically investigates the influence of scanning speed on the microstructure and properties of TiAl coatings fabricated by SLM, where scanning speed governs the evolution process of the coating from microstructure to macroscopic performance by altering the melt pool energy input, cooling rate, and element diffusion kinetics.

• 1When the scanning speed is 1000 mm/s, longitudinal and transverse cracks that penetrate into the substrate are formed at the coating interface, the Ti/Al atomic ratio to decrease to 1:4.8. Although Ti₅Al₁₁, Al₃Sc and other phases are formed, the crack defects seriously degrade the integrity of the structure. At 1100 mm/s, the number of cracks further increases, the Ti–Al phases shift to Ti₂Al, Ti₃Al₅, and the uniformity and stability of the structure remain poor.
• 2When the scanning speed is 1200 mm/s, the energy of the molten pool reaches the optimal balance with the cooling rate. The coating-binding interface is smooth without cracks, the retention rate of Mg element is the highest, the Ti/Al mass percentage stabilizes at 1:12.9, Sc element fully dissolves to form Al₃Sc strengthening phase, Ti–Al phases are mainly Ti₃Al and Ti₂Al, and the maximum hardness of the coating reaches 565.8 HV_0.2.
• 3When the cooling rate is too high at 1300 mm/s, leading to a reduction in the number of cracks, mostly shorter longitudinal cracks. The surface of the coating shows stratification and low smoothness.

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

The authors declare no competing financial interest.