Deposition of Nickel-Based Superalloy Claddings on Low Alloy Structural Steel by Direct Laser Deposition

In this study, direct laser deposition (DLD) of nickel-based superalloy powders (Inconel 625) on structural steel (42CrMo4) was analysed. Cladding layers were produced by varying the main processing conditions: laser power, scanning speed, feed rate, and preheating. The processing window was established based on conditions that assured deposited layers without significant structural defects and a dilution between 15 and 30%. Scanning electron microscopy, energy dispersive spectroscopy, and electron backscatter diffraction were performed for microstructural characterisation. The Vickers hardness test was used to analyse the mechanical response of the optimised cladding layers. The results highlight the influence of preheating on the microstructure and mechanical responses, particularly in the heat-affected zone. Substrate preheating to 300 °C has a strong effect on the cladding/substrate interface region, affecting the microstructure and the hardness distribution. Preheating also reduced the formation of the deleterious Laves phase in the cladding and altered the martensite microstructure in the heat-affected zone, with a substantial decrease in hardness.


Introduction
The laser-based additive manufacturing (LBAM) technologies applied in the production and repair of industrial components emerged in the late 1990s. Their use continues to extend to many industrial sector applications for components that operate in extreme conditions. LBAM technologies are unique and versatile in the manufacturing of parts with complex geometry, functionally graded or customised, producing an improvement in properties that can be used for a variety of industrial applications, such as within the aerospace, metallurgy, energy, and automotive industries [1,2]. Direct laser deposition (DLD) is an LBAM technology used for the additive manufacturing of metal parts, reconstructions, and repairs. DLD consists of the supply, through a nozzle, of metallic powder (or wire) processed by a focused laser, creating a melt pool on the surface of a metallic substrate. Several processing variables directly or indirectly affect the quality and structural integrity of components, dictated by solidification and metallurgical bonding [3].
DLD involves interactions between the laser beam, powder, and substrate in an environment with local protection from inert gas. Laser power, scanning speed, beam size, and powder feed rate are parameters that play a dominant role in cladding geometry (height, width, and length), dilution, and metallurgical properties. Clad overlapping, gas flow rate, powder flow profile, powder quality (size, shape, and density), and preheating are important secondary parameters [4,5].

DLD System Setup
A laser system, LDF 3000-100, was used to produce DLD claddings. The system has a high-power fibre-coupled laser diode (wavelength 900-1030 nm, depending on the power), with a nominal beam power of 6000 W. The machine concept is a KUKA KR90 R3100 industrial robot, based on a 6-axis industrial robot. All axes are connected to the robot and laser control units, which command the temperature of the melt pool as well as the laser power. The powder was fed during the deposition process by a coaxial feeding system, as illustrated in Figure 1. The substrate was preheated (PHT) to 300 • C with a manual gas system. The temperature control of the preheated substrate was done by a digital pyrometer, for verification of the uniformity of the substrate surface temperature distribution. Preheating is intended to decrease the cooling rate in the melt pool and HAZ regions as well as eliminate moisture. Tests were performed on substrates after the production of clads, with and without PHT, to evaluate susceptibility to cracking and eventual formation of metastable phases. cladding quality, was evaluated considering the absence of cracks and structural imperfections. Preheating was performed on an optimised cladding condition in order to moderate the microstructure and mechanical responses. Microhardness profiles of claddings were obtained and correlated with the microstructures.

DLD System Setup
A laser system, LDF 3000-100, was used to produce DLD claddings. The system has a high-power fibre-coupled laser diode (wavelength 900-1030 nm, depending on the power), with a nominal beam power of 6000 W. The machine concept is a KUKA KR90 R3100 industrial robot, based on a 6-axis industrial robot. All axes are connected to the robot and laser control units, which command the temperature of the melt pool as well as the laser power. The powder was fed during the deposition process by a coaxial feeding system, as illustrated in Figure 1. The substrate was preheated (PHT) to 300 °C with a manual gas system. The temperature control of the preheated substrate was done by a digital pyrometer, for verification of the uniformity of the substrate surface temperature distribution. Preheating is intended to decrease the cooling rate in the melt pool and HAZ regions as well as eliminate moisture. Tests were performed on substrates after the production of clads, with and without PHT, to evaluate susceptibility to cracking and eventual formation of metastable phases.

Feedstock Powder and Substrate
A nickel-based superalloy (MetcoClad 625 from Oerlikon), similar to Inconel 625, produced by the gas-atomised process, was used in this study. This powder was developed specifically for laser processing and presents a spherical morphology as well as particle sizes ranging between 45 and 90 µ m. Figure 2 shows scanning electron microscopy (SEM) images of the morphology of the MetcoClad 625 powder.

Feedstock Powder and Substrate
A nickel-based superalloy (MetcoClad 625 from Oerlikon), similar to Inconel 625, produced by the gas-atomised process, was used in this study. This powder was developed specifically for laser processing and presents a spherical morphology as well as particle sizes ranging between 45 and 90 µm. 42CrMo4 steel, in the quenched and tempered condition, was used as a substrate for the DLD deposition. Specimens with dimensions of 100 mm × 120 mm × 15 mm were prepared for depositions. This steel is classified as a low alloy structural steel with high mechanical strength and toughness as well as good fatigue resistance and machinability, 42CrMo4 steel, in the quenched and tempered condition, was used as a substrate for the DLD deposition. Specimens with dimensions of 100 mm × 120 mm × 15 mm were prepared for depositions. This steel is classified as a low alloy structural steel with high mechanical strength and toughness as well as good fatigue resistance and machinability, being widely used in the manufacturing of critical industrial components, such as gears, automotive components, wing generators, and drilling joints for deep wells [27]. Its mechanical and chemical properties are described in standard EN 10,269 [28]. The chemical composition of the MetcoClad 625 (M625) powder and 42CrMo4 steel are shown in Table 1.

Process Parameters
The main process parameters (laser power, scanning speed, and feed rate) were considered to evaluate the effect of processing conditions on clads. The evaluation of the clad quality depends on clad characteristics, namely, absence of cracks or pores, good metallurgical bond (interfaces without microcracks, pores, and fragile phases, exhibiting good wettability), and dilution between 15 and 30%. The values of the tested parameters are shown in Table 2. All results are representative of single-layer samples. The terminology ILP_SS_FR was used to identify the samples, being I-M625 powder (Inconel powder); LP-laser power (kW); SS-scanning speed (mm/s); FR-feed rate (g/min). Eighteen combinations of different processing conditions were tested, with and without preheating (see Table 3). Before deposition, the substrates were cleaned with pure acetone.  In all tests, a spot size of 2.5 mm and an offset in the Z-axis of 0.2 mm were used. High-purity argon (99.99%), with a 5.5 L/min flow rate, was used as the shielding gas for minimising contamination of the melt pool during the DLD process. Samples with and without PHT were cooled in air.

Mechanical and Microstructural Characterisation
Samples from each deposition were cut for microstructural and mechanical characterisation using a metallographic cut-off machine with refrigeration in order to avoid substrate and cladding overheating. Samples were mounted in resin and polished down to a 1 µm diamond suspension. Kalling's N • . 2 chemical etching (CuCl 2 -5 g, hydrochloric acid-100 mL, and ethanol-100 mL) was used to reveal the microstructures.
The measurements of the height, depth, and width of the claddings produced by the DLD technique were performed using a Leica DVM6 A 2019 digital microscope (DM) (Wetzlar, Germany). The Leica DM 4000 M optical microscope (OM) (Wetzlar, Germany) allowed for a microstructural analysis at low magnifications to evaluate, for example, the size of the heat-affected zone. A scanning electron microscope, FEI Quanta 400 FEG (ESEM, Hillsboro, OR, USA), equipped with energy-dispersive X-ray spectroscopy (EDX) (EDAX Genesis X4M, Oxford Instrument, Oxfordshire, UK) and electron backscatter diffraction (EBSD) (EDAX-TSL OIM EBSD, Mahwah, NJ, USA) was used for higher magnification observation and phase identification. For EBSD evaluation, the samples went through an additional polishing step, using a 0.06 µm silica colloidal suspension, for a superior surface finish and to remove polishing-induced plastic deformations, allowing Kikuchi patterns to be obtained [29]. EBSD allows for the obtaining of information on microstructural characteristics with a small interaction volume and a high resolution, for which TSL OIM Analysis 5.2 software was used. For all raw data obtained by EBSD, a dilatation clean-up routine was performed, with a grain tolerance angle of 15 • and a minimum grain size of 10 points.
Quantitative image analysis was employed on optical images using the ImageJ software, version 1.51p (National Institutes of Health, Bethesda, MD, USA).
Vickers microhardness tests gave the mechanical characterisation. The tests were performed using a test force of 300 g for 15 s in a Struers Duramin 5 (Struers Inc., Cleveland, OH, USA) Vickers hardness tester. Each hardness value corresponds to the average of three indentations.

Processing Effects
The quality of cladding was first evaluated by inspection with a digital microscope (DM). The microstructural analysis of all the claddings produced did not detect cracks, pores, or inclusions of significant dimensions, an essential requirement for obtaining highperformance deposits.
SEM characterisation confirmed the observations made by the DM. Figure 3 shows an SEM image of the cross-section of an M625 clad deposited on 42CrMo4 steel, representing the geometric aspects of cladding: height (h), width (w), depth (d), clad area (AC), melting area (AM), and wetting angle (θ). These geometric aspects were measured on all claddings using the ImageJ software. The results are shown in Table 3. This analysis of the geometry of the single-track deposits is critical as it provides information on process yield and cladding performance. For example, the contact angle is an essential parameter in assessing the quality of the cladding [30,31]. Higher beads can promote low wettability during the deposition of multi-tracks, hindering overlapping and generating discontinuities in the hatch spacing of the overlapping deposits, thus facilitating crack propagation. Typically, a contact angle greater than 90 • is associated with lowerquality claddings [32]. Table 3 shows that depositions with higher heights have a high contact angle, as samples I1_2_15, I2_2_15, and I3_2_15 demonstrate. This analysis of the geometry of the single-track deposits is critical as it provides information on process yield and cladding performance. For example, the contact angle is an essential parameter in assessing the quality of the cladding [30,31]. Higher beads can promote low wettability during the deposition of multi-tracks, hindering overlapping and generating discontinuities in the hatch spacing of the overlapping deposits, thus facilitating crack propagation. Typically, a contact angle greater than 90° is associated with lower-quality claddings [32]. Table 3 shows that depositions with higher heights have a high contact angle, as samples I1_2_15, I2_2_15, and I3_2_15 demonstrate. The results presented in Table 3 show that the analysed processing parameters (laser power, scanning speed, powder feed rate, and preheating) strongly influence the production of the cladding and its bonding to the substrate. The selection of a processing window that guarantees a metallurgically bonded clad with good material yield is a fundamental task. This selection is difficult since the mutual interaction of the various parameters is complex. Thus, it is common to apply combined parameters in the DLD process to obtain a more accurate relationship between processing parameters and the clad characteristics [33].
One of the most used complex parameters is powder deposition density (PDD), which expresses the combined effect of feed rate, scanning speed, and laser spot size (ϕ) (Equation (1)) [34][35][36]. Figure 4 exposes the linear growth of the cladding area with the increase in the PDD parameter.
This parameter shows that by increasing the feed rate or decreasing the scanning speed, we can obtain claddings with a larger area, which was expected since both situations result in more powder supply in the same period of time. PDD is a valuable parameter, but this relationship is only valid for the cladding area and not for the total area of the deposit, including the area of the substrate that has been melted. This last area is vital because it affects the quality of the cladding.  This parameter shows that by increasing the feed rate or decreasing the scanning speed, we can obtain claddings with a larger area, which was expected since both situations result in more powder supply in the same period of time. PDD is a valuable parameter, but this relationship is only valid for the cladding area and not for the total area of the deposit, including the area of the substrate that has been melted. This last area is vital because it affects the quality of the cladding.
Additionally, with the measured areas (Table 3), it was possible to calculate the dilution, which quantifies the relative amount of melted substrate during laser processing, according to Equation (2). Dilution (%) = AM/(AC + AM) × 100 (2) A dilution ratio between 15 and 30% is sufficient to allow a good metallurgical bond between the substrate and the cladding. Higher dilution, which means a greater melting of the substrate, is undesirable since it reduces the deposition yield and induces considerable changes in the chemical composition of the deposited material, modifying the expected properties of the cladding.
The results obtained, and presented in Table 3, indicate that dilution increases with increased laser power, keeping the other processing variables constant. A laser power of 1.5 kW is enough to guarantee a dilution higher than 15% for almost all conditions (the only exception is the I1.5_10_15 sample).
Despite this apparent direct relation between laser power and dilution, the effect of other critical processing variables, namely the scanning speed and the feed rate, makes the establishment of relationships between processing condition and dilution complex. To overcome this difficulty, it is common to apply complex parameters, empirically adjusted, to the clad/substrate set under analysis to define the processing window [33,37,38]. Figure 5 shows a process window map, which associates laser power with the scanning speed and feeding rate ratio, as well as the dilution that is correlated with the laser power through complex parameter LP (SS/FR) 0.5 ; it is represented by two curves, one for 15% and the other for 30%. Additionally, as was also considered in the map, a vertical line that corresponds to the acceptable limit for the wetting angle is present. Additionally, with the measured areas (Table 3), it was possible to calculate the dilution, which quantifies the relative amount of melted substrate during laser processing, according to Equation (2).
A dilution ratio between 15 and 30% is sufficient to allow a good metallurgical bond between the substrate and the cladding. Higher dilution, which means a greater melting of the substrate, is undesirable since it reduces the deposition yield and induces considerable changes in the chemical composition of the deposited material, modifying the expected properties of the cladding.
The results obtained, and presented in Table 3, indicate that dilution increases with increased laser power, keeping the other processing variables constant. A laser power of 1.5 kW is enough to guarantee a dilution higher than 15% for almost all conditions (the only exception is the I1.5_10_15 sample).
Despite this apparent direct relation between laser power and dilution, the effect of other critical processing variables, namely the scanning speed and the feed rate, makes the establishment of relationships between processing condition and dilution complex. To overcome this difficulty, it is common to apply complex parameters, empirically adjusted, to the clad/substrate set under analysis to define the processing window [33,37,38]. Figure 5 shows a process window map, which associates laser power with the scanning speed and feeding rate ratio, as well as the dilution that is correlated with the laser power through complex parameter LP (SS/FR) 0.5 ; it is represented by two curves, one for 15% and the other for 30%. Additionally, as was also considered in the map, a vertical line that corresponds to the acceptable limit for the wetting angle is present.
As seen in Figure 5, the shaded area, delimited by the previous conditions (dilution range, acceptable wetting angle, and process parameters), reveals the desirable practical manufacturing processing window of Inconel 625. In addition, the dilution increases proportionally with the increase in laser power and decreases with the ratio between scanning speed and feeding rate [39]. The flow of liquid metal in the melt pool is dominated by Marangoni's convection effect, caused by the surface tension gradient. As the temperature of the substrate increases, this effect becomes more evident. However, in practice, the surface tension gradient (γ) dγ/dT depends on both temperature (T) and composition. In this case, the most significant influence factor is the thermal gradient promoted between laser beam and substrate, as Le et al. [40] demonstrated, where the increase in the substrate temperature becomes more evident. As seen in Figure 5, the shaded area, delimited by the previous conditions (dilution range, acceptable wetting angle, and process parameters), reveals the desirable practical manufacturing processing window of Inconel 625. In addition, the dilution increases proportionally with the increase in laser power and decreases with the ratio between scanning speed and feeding rate [39]. The flow of liquid metal in the melt pool is dominated by Marangoni's convection effect, caused by the surface tension gradient. As the temperature of the substrate increases, this effect becomes more evident. However, in practice, the surface tension gradient (γ) dγ/dT depends on both temperature (T) and composition. In this case, the most significant influence factor is the thermal gradient promoted between laser beam and substrate, as Le et al. [40] demonstrated, where the increase in the substrate temperature becomes more evident.
Laser power control allows the required good metallurgical integrity and dilution to be achieved [41], producing a cladding with a good metallurgical bond and uniform properties. Increasing the power of the laser promotes increased energy density as well as greater dilution and mixing between the substrate and the deposited powder, which is characteristic of the laser alloy process [42].
Considering the analyses performed, the I2_2_15, I2_4_15, I2_6_15, I2_6_20, I2_10_10, I3_2_15, and I3_4_15 conditions showed dilutions within the established range, between 16.6% and 29.9%, but a lower dilution value is preferable. Nevertheless, the I2_2_15, I2_4_15, and I3_2_15 samples presented poor wettability angles (107°, 88°, and 99°, respectively) that may lead, in future, to claddings with overlapping defects between strands. On the other hand, the four remaining conditions presented good wetting angles, but taking into account not only the quality of the process but also its efficiency, condition I2_6_20 ( Figure 5 red point) is the only one that allows for the possibility of manufacturing a larger cladding area and consequently a higher cladding high, which has an inverse linear relationship with the SS/FR ratio. Considering this evaluation, the I2_6_20 condition will be used to perform the analyses throughout the following sections. Laser power control allows the required good metallurgical integrity and dilution to be achieved [41], producing a cladding with a good metallurgical bond and uniform properties. Increasing the power of the laser promotes increased energy density as well as greater dilution and mixing between the substrate and the deposited powder, which is characteristic of the laser alloy process [42].
Considering the analyses performed, the I2_2_15, I2_4_15, I2_6_15, I2_6_20, I2_10_10, I3_2_15, and I3_4_15 conditions showed dilutions within the established range, between 16.6% and 29.9%, but a lower dilution value is preferable. Nevertheless, the I2_2_15, I2_4_15, and I3_2_15 samples presented poor wettability angles (107 • , 88 • , and 99 • , respectively) that may lead, in future, to claddings with overlapping defects between strands. On the other hand, the four remaining conditions presented good wetting angles, but taking into account not only the quality of the process but also its efficiency, condition I2_6_20 ( Figure 5 red point) is the only one that allows for the possibility of manufacturing a larger cladding area and consequently a higher cladding high, which has an inverse linear relationship with the SS/FR ratio. Considering this evaluation, the I2_6_20 condition will be used to perform the analyses throughout the following sections. Figure 6 shows the cladding microstructure formed adjacent to the substrate by the deposition of M625 on 42CrMo4 steel. This microstructure consists of columnar grains, mainly dendrites, and cellular morphologies in a few regions.

Microstructures and Mechanical Characterisation of the DLD Samples
As shown in Figure 6B, on solidification a continuous thin layer, < 10 µm, consisting of planar grains formed in the vicinity of the substrate. This morphology evolves into columnar grains with continued solidification of the cladding. This microstructure is consistent with the results of a similar study in which martensitic stainless steel is deposited [43]. The very high thermal gradient in the contact zone of the melt pool with the cold substrate contributed to the formation of planar grains. The microstructure evolves into a colum-nar/dendritic structure due to a rapid decrease in the thermal gradient when more material solidifies. Moreover, columnar grains grow perpendicular to the substrate/solidified material, i.e., in the opposite direction of the primary heat dissipation source, as is usual in DLD solidification [44]. As seen in Figure 7, this columnar/dendritic structure is the characteristic microstructure of the cladding.  Figure 6 shows the cladding microstructure formed adjacent to the substrate by the deposition of M625 on 42CrMo4 steel. This microstructure consists of columnar grains, mainly dendrites, and cellular morphologies in a few regions. As shown in Figure 6B, on solidification a continuous thin layer, < 10 µ m, consisting of planar grains formed in the vicinity of the substrate. This morphology evolves into columnar grains with continued solidification of the cladding. This microstructure is consistent with the results of a similar study in which martensitic stainless steel is deposited [43]. The very high thermal gradient in the contact zone of the melt pool with the cold substrate contributed to the formation of planar grains. The microstructure evolves into a columnar/dendritic structure due to a rapid decrease in the thermal gradient when more material solidifies. Moreover, columnar grains grow perpendicular to the substrate/solidified material, i.e., in the opposite direction of the primary heat dissipation source, as is usual in DLD solidification [44]. As seen in Figure 7, this columnar/dendritic structure is the characteristic microstructure of the cladding.   As shown in Figure 6B, on solidification a continuous thin layer, < 10 µ m, consisting of planar grains formed in the vicinity of the substrate. This morphology evolves into columnar grains with continued solidification of the cladding. This microstructure is consistent with the results of a similar study in which martensitic stainless steel is deposited [43]. The very high thermal gradient in the contact zone of the melt pool with the cold substrate contributed to the formation of planar grains. The microstructure evolves into a columnar/dendritic structure due to a rapid decrease in the thermal gradient when more material solidifies. Moreover, columnar grains grow perpendicular to the substrate/solidified material, i.e., in the opposite direction of the primary heat dissipation source, as is usual in DLD solidification [44]. As seen in Figure 7, this columnar/dendritic structure is the characteristic microstructure of the cladding.    Table 4 shows the chemical composition obtained by EDX analysis of the zones indicated in Figure 8.

Microstructures and Mechanical Characterisation of the DLD Samples
The γ dendrites are formed during the solidification of nickel-based superalloys processed by DLD, segregating Nb and Mo into the liquid, thus creating the local conditions for forming the Laves phase. The final stage of non-equilibrium solidification thus gives rise to this microstructure consisting mainly of the γ matrix and Laves phase. A similar microstructure has been found in other nickel-based superalloy solidification studies [45,46,47,48,49,50,51].  Laves phase formation promotes the initiation and propagation of cracks, with a detrimental effect in mechanical response, reducing ductility, ultimate tensile strength, fracture resistance, and fatigue life [52,53]. Thus, this phase reduces the performance of Inconel 625, requiring control of morphology and distribution in the cladding.
A similar analysis was performed at the cladding/substrate interface, Figure 9. In the continuous thin layer adjacent to the substrate, characterized by planar grains (see Figure  6), the thermal gradient and growth rate are significantly different from those of the dendritic region, and the Laves phase was not detected. Figure 9 shows that DLD deposition of M625 on the 42CrMo4 substrate led to cladding with a flat interface, with a thin continuous layer of planar γ grains, followed by γ dendrites and a dispersed Laves phase. As already mentioned, the appearance of the Laves phase in this region has a detrimental effect and should be minimized. Therefore,  Table 4. Chemical composition (%) of the zones indicated in Figure 8. The γ dendrites are formed during the solidification of nickel-based superalloys processed by DLD, segregating Nb and Mo into the liquid, thus creating the local conditions for forming the Laves phase. The final stage of non-equilibrium solidification thus gives rise to this microstructure consisting mainly of the γ matrix and Laves phase. A similar microstructure has been found in other nickel-based superalloy solidification studies [45][46][47][48][49][50][51].

Phases
Laves phase formation promotes the initiation and propagation of cracks, with a detrimental effect in mechanical response, reducing ductility, ultimate tensile strength, fracture resistance, and fatigue life [52,53]. Thus, this phase reduces the performance of Inconel 625, requiring control of morphology and distribution in the cladding.
A similar analysis was performed at the cladding/substrate interface, Figure 9. In the continuous thin layer adjacent to the substrate, characterized by planar grains (see Figure 6), the thermal gradient and growth rate are significantly different from those of the dendritic region, and the Laves phase was not detected. Figure 9 shows that DLD deposition of M625 on the 42CrMo4 substrate led to cladding with a flat interface, with a thin continuous layer of planar γ grains, followed by γ dendrites and a dispersed Laves phase. As already mentioned, the appearance of the Laves phase in this region has a detrimental effect and should be minimized. Therefore, another cladding was performed with the substrate preheated to 300 • C. This procedure leads to a decrease in the thermal gradient of the deposited layer, influencing the microstructure. Preheating (PHT) increases the interdiffusion of constituents of M625 and 42CrMo4 at the bonding interface and, by decreasing the substrate cooling rate, can also affect the microstructure and properties in the heat-affected zone.
another cladding was performed with the substrate preheated to 300 °C . This procedure leads to a decrease in the thermal gradient of the deposited layer, influencing the microstructure. Preheating (PHT) increases the interdiffusion of constituents of M625 and 42CrMo4 at the bonding interface and, by decreasing the substrate cooling rate, can also affect the microstructure and properties in the heat-affected zone. The influence of PHT application on the cladding/substrate interface is shown in Figure 10. The thin layer of planar grains was not formed. Substrate heating significantly reduced the thermal gradient in the initial solidification phase of the melt pool, leading to the formation of dendrites throughout the cladding. Furthermore, it appears that PHT slightly increases the interdendritic spacing from 5-7 µ m to 6-9 µ m, measured by ImageJ software in Figures 6 and 10, respectively. This observation confirms the decrease in the cooling rate allowing the growth of the interdendritic spacing. This feature is consistent with a study that indicates that lower cooling rates promote dendrite growth and decrease cellular grain formation in the initial stages of solidification [54].
Besides, the application of PHT has led to the reduction in Laves phases, this effect highlighted at the interface zone, as seen in Figure 11. A lower cooling rate effectively maintained the Nb and Mo in the γ matrix, avoiding segregation. Moreover, the elemental mapping and EBSD image in Figure 11 confirm a more intense interdiffusion at the interface, which is not as plane as it was without PHT, Figure 9. Diffusion of nickel, which is a gamma-phase stabilizer, to steel is associated with a greater amount of residual austenite in this region. Thus, the application of preheating seems helpful for reducing the deleterious Laves phase and for enhancing metallurgical bonding.
The HAZ of 42CrMo4 steel is also a critical region, as substrate heating by the laser followed by rapid cooling leads to martensite formation, which can create cracks and allow for rapid crack propagation. Microstructural differences in HAZ caused by PHT were analysed by EBSD, as illustrated in Figure 12. This figure shows that PHT affects the HAZ microstructure, with larger (longer and wider) martensite laths caused by a slower cooling rate. Martensite with wider laths is associated with lower mechanical strength The influence of PHT application on the cladding/substrate interface is shown in Figure 10. The thin layer of planar grains was not formed. Substrate heating significantly reduced the thermal gradient in the initial solidification phase of the melt pool, leading to the formation of dendrites throughout the cladding. Furthermore, it appears that PHT slightly increases the interdendritic spacing from 5-7 µm to 6-9 µm, measured by ImageJ software in Figure 6 and Figure 10, respectively. This observation confirms the decrease in the cooling rate allowing the growth of the interdendritic spacing. This feature is consistent with a study that indicates that lower cooling rates promote dendrite growth and decrease cellular grain formation in the initial stages of solidification [54].   Besides, the application of PHT has led to the reduction in Laves phases, this effect highlighted at the interface zone, as seen in Figure 11. A lower cooling rate effectively maintained the Nb and Mo in the γ matrix, avoiding segregation. Moreover, the elemental mapping and EBSD image in Figure 11 confirm a more intense interdiffusion at the interface, which is not as plane as it was without PHT, Figure 9. Diffusion of nickel, which is a gammaphase stabilizer, to steel is associated with a greater amount of residual austenite in this region. Thus, the application of preheating seems helpful for reducing the deleterious Laves phase and for enhancing metallurgical bonding.  substrate. An EBSD image shows the phase distribution in this regio FCC phases in green and BCC phases in red. Figure 12 also shows that no preferential crystalline orientation was formed in th HAZ of claddings without or with PHT. The HAZ of 42CrMo4 steel is also a critical region, as substrate heating by the laser followed by rapid cooling leads to martensite formation, which can create cracks and allow for rapid crack propagation. Microstructural differences in HAZ caused by PHT were analysed by EBSD, as illustrated in Figure 12. This figure shows that PHT affects the HAZ microstructure, with larger (longer and wider) martensite laths caused by a slower cooling rate. Martensite with wider laths is associated with lower mechanical strength which, together with the more significant amount of residual austenite determined when using PHT, can reduce the brittleness of this region. Figure 12 also shows that no preferential crystalline orientation was formed in the HAZ of claddings without or with PHT. Metals 2021, 11, x FOR PEER REVIEW 14 of 17

Microhardness Measurements
The microhardness profile shown in Figure 13 shows that the hardness distribution in the cladding zone is uniform and independent of PHT, with an average hardness of 258 ± 2 HV and 253 ± 2 HV for the samples with and without preheating, respectively. A sharp transient zone with a pronounced hardness increase is measured in the HAZ of the sample without PHT (maximum hardness of 491 ± 23 HV). The application of PHT to the substrate before deposition produced a more uniform distribution of hardness in the HAZ (368 ± 25 HV), with a less sharp transient near the interface. Furthermore, the hardness peak has been eliminated, and the hardness values show less dispersion. These differences in hardness are attributed to changes in the microstructure induced by PHT, as seen in Figure 12, and its influence on chemical composition distribution, seen in Figures 9 and  11, and indicate that the HAZ region is less prone to cracking.

Microhardness Measurements
The microhardness profile shown in Figure 13 shows that the hardness distribution in the cladding zone is uniform and independent of PHT, with an average hardness of 258 ± 2 HV and 253 ± 2 HV for the samples with and without preheating, respectively. A sharp transient zone with a pronounced hardness increase is measured in the HAZ of the sample without PHT (maximum hardness of 491 ± 23 HV). The application of PHT to the substrate before deposition produced a more uniform distribution of hardness in the HAZ (368 ± 25 HV), with a less sharp transient near the interface. Furthermore, the hardness peak has been eliminated, and the hardness values show less dispersion. These differences in hardness are attributed to changes in the microstructure induced by PHT, as seen in Figure 12, and its influence on chemical composition distribution, seen in Figures 9 and 11, and indicate that the HAZ region is less prone to cracking.

Microhardness Measurements
The microhardness profile shown in Figure 13 shows that the hardness distribution in the cladding zone is uniform and independent of PHT, with an average hardness of 258 ± 2 HV and 253 ± 2 HV for the samples with and without preheating, respectively. A sharp transient zone with a pronounced hardness increase is measured in the HAZ of the sample without PHT (maximum hardness of 491 ± 23 HV). The application of PHT to the substrate before deposition produced a more uniform distribution of hardness in the HAZ (368 ± 25 HV), with a less sharp transient near the interface. Furthermore, the hardness peak has been eliminated, and the hardness values show less dispersion. These differences in hardness are attributed to changes in the microstructure induced by PHT, as seen in Figure 12, and its influence on chemical composition distribution, seen in Figures 9 and  11, and indicate that the HAZ region is less prone to cracking.

Conclusions
The deposition of Inconel 625 claddings onto a 42CrMo4 steel substrate was performed using direct laser deposition (DLD) and varying processing conditions: laser power, scan-ning speed, feed rate, and preheating. Macro-and microstructural analysis, in addition to the hardness measurements, led to the following main conclusions: • A DLD process window map considering processing variables shows that several combinations can be used. However, the cladding produced with 2 kW of laser power, a scanning speed of 6 mm/s, and a 20 g/min feed rate presented adequate dilution and wettability.

•
The deposited layers were produced without significant structural defects such as cracks, pores, or other types of discontinuities. • Substrate preheating to 300 • C influences the microstructure of the cladding/substrate interface, reducing the formation of the deleterious Laves phase. • PHT also alters the hardness profile, mainly in the heat-affected zone, due to modification of the martensite microstructure and increased residual austenite. Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.