In-situ Synthesis and Characterization of Ceramic Reinforced Inconel 718 Coatings Using B4C-Inconel 718 Powders by Laser Directed Laser Energy Deposition

Inconel 718 has been widely used in aerospace, nuclear and marine industries due to excellent high-temperature mechanical properties and corrosion resistance. In recent years, laser directed energy deposition (DED) become a competitive method in the fabrication of Inconel 718 coatings. Compared with other surface coating processes, laser DED has the advantage of extremely ne-grained structures, strong metallurgical bonding, and high density. However, the hardness and wear resistance of Inconel 718 coatings still need to be improved. To further improve these properties, ceramic reinforced Inconel 718 coatings have been investigated. Compared with ex-situ ceramic reinforcements, the in-situ synthesized reinforcements have the advantage of rened ceramic particle size, uniform distribution, and low thermal stress. B 4 C was a preferable additive material to fabricate in-situ synthesized multi-component ceramic reinforced Inconel 718 coatings. The addition of B 4 C could form a large number of borides and carbides as ceramic reinforcements. In addition, the in-situ reactions between Inconel 718 and B 4 C could release heat during the fabrication, thereby promoting the melting of material powders. However, there are currently no investigations on the in-situ synthesis mechanisms, microstructure, and mechanical properties of laser DED fabricated B 4 C-Inconel 718 coatings. In this study, the effects of B 4 C on the properties of Inconel 718 coatings were investigated. Results show that Ni 3 B, NbB, and Cr 3 C 2 phases were formed. With the addition of B 4 C, the size of Laves phase was rened and the porosity was decreased. The hardness and wear resistance of B 4 C reinforced coatings were improved by about 34% and 28%, respectively.


Introduction
Inconel 718, as a kind of nickel-based superalloy, has remarkably high-temperature mechanical properties and corrosion resistance.It is a good choice for strengthening or repairing the structural parts that are prone to be damaged under conditions with complicated stress and elevated temperature [1,2].In recent years, laser-directed energy deposition (DED) becomes a competitive technology for repairing and coating structural parts, due to its advantages of excellent bonding quality, easy controllability, excellent stability, and the capability of functionally gradient materials fabrication [3,4].Due to these advantages, the fabrication of Inconel 718 coatings on stainless steel and nickel-based alloys by laser DED process have been received extensive research attention [5,6].Results show that the laser DED fabricated coatings are free of micropores and micro-cracks and have good adherence to the substrates.However, the microhardness and wear resistance of laser DED fabricated Inconel 718 coatings is relatively low, which limited their further applications in many areas.
In order to improve the service life and mechanical properties of laser DED fabricated Inconel 718 coatings, ceramic materials such as BN, SiC, and Al 2 O 3 have been directly added to Inconel 718 coatings as reinforcement phases [7][8][9][10].However, the addition of ex-situ ceramic reinforcements also has a serious negative impact on the quality of coatings.Due to the high melting point of ex-situ ceramic reinforcements, the feedstock powders are hard to be fully melted.The un-melted particles could decrease the uidity of the molten pool, leading to the generation of fabrication defects and the uneven distribution of the ceramic reinforcements [11].In addition, the thermal expansion coe cients of ex-situ ceramic reinforcements and Inconel 718 matrix are different.During the solidi cation process, the matrix cracking in composites could be generated.Different from the ex-situ process, in the in-situ process, the reinforcements are synthesized in the matrix itself by chemical reactions.The ultra-ne ceramic reinforcement particles could be generated and uniformly distributed in the fabricated coatings, which could further improve the consistency and the mechanical properties of the fabricated coatings.In addition, the large thermal expansion coe cient differences between ceramic particles and matrix are reduced, which suppresses the crack generation [12,13].
Over the past decade, researchers have demonstrated interest in the fabrication of in-situ synthesized ceramic reinforced Inconel 718 coatings.TiC has been the most commonly used as the reactant of in-situ synthesized ceramic reinforced Inconel 718 coatings [14].Results show that a large number of sub-grain boundaries appear in the TiC/Inconel 718 nanocomposite coatings, which signi cantly increases the hardness and wear resistance.Compared with TiC, B 4 C has more advantages to serve as the reactant of in-situ synthesized ceramic reinforced Inconel 718 coatings.Firstly, due to the low molar mass of B 4 C, a small amount of B 4 C could form a large number of borides and carbides as hard phases that can signi cantly improve the hardness and the erosion resistance of the coating.Secondly, the borides generate from the reactions between B 4 C and Inconel 718 could provide heterogeneous nucleation sites for the Laves phase during the solidi cation, which could re ne the size of Laves phase and improve the microhardness of the fabricated coatings.Finally, the in-situ reactions between B 4 C and metallic materials always generate a large amount of heat [15].The generated heat has the potential to be utilized to promote the melting of feedstock powders and reduce the laser energy input.As far as the authors know, there are no investigations on the fabrication of in-situ synthesized ceramic reinforced Inconel 718 coatings using B 4 C-Inconel 718 powders.The in-situ synthesis and characterization of the fabricated coatings are still unknown.In this study, the multi-component ceramic reinforced Inconel 718 coatings were successfully fabricated by laser DED process.The effects of B 4 C on molten pool size and temperature, element and phase compositions, microstructure, porosity, and mechanical properties of the fabricated coatings (including microhardness and wear resistance) were investigated.

Powder treatment
The powder materials used in this study were B 4 C powder (99.7% purity) (Atlantic Equipment Engineers Inc., Upper Saddle River, NJ, USA) and Inconel 718 powders (99.7% purity) ( Carpenter Powder Products Inc., Bridgeville, PA, USA).A stainless steel 304 of 6.65 mm thickness was used as the substrate.The feedstock material powders (Inconel 718 and Inconel 718 + 2 wt.%B 4 C) were adopted to fabricate coatings.As shown in Figure 1, before the LDED process, B 4 C powder and Inconel 718 powder were mixed by a planetary ball milling machine (ND2L, Torrey Hills Technologies LLC., San Diego, CA, USA) for four hours with the rotation speed of 200 rpm.The weight ratio of milling balls to powders was 1:1.B 4 C powders were mixed with Inconel 718 powders after the ball milling process.

Experiment setup
As shown in Figure 2, experiments were conducted on an LDED machine (LENS 450, Optomec Inc., Albuquerque, NM, USA) equipped with a 400 W ber laser source (YLM-1070, IPG Photonics, USA).Before the fabrication, to avoid the oxidization of B4C and Ti at a high temperature, argon gas was utilized to purge the chamber system.During the fabrication, the laser beam was generated and transformed to the substrate to generate a molten pool, which caught and melted the material powders.When the laser beam moved away, the molten pool was solidi ed to fabricate the rst layer.After the fabrication of the rst layer, the cladding head moved up the distance of a layer thickness to fabricate the second layer on the top surface of the rst layer.By repeating this procedure, the coatings were fabricated layer by layer.The values or ranges of input fabrication variables are listed in Table 1.

Measurement procedures
During the fabrication, An infrared thermal camera (PYROVIEW 768 N, DIAS INC, Dresden, Germany) was used to obtain the thermal images of the molten pool.The infrared thermal camera was xed in the chamber at a 20 cm distance to the molten pool.The professional software (PYROSOFT 3.22, DIAS INC, Dresden, Germany) was used to record the collected thermal data with a sample rate of 25 Hz.The phases of the fabricated coatings were analyzed by an X-ray diffraction (XRD) machine (Ultima III, Rigaku Corp., The Woodlands, TX, USA).The samples were scanned from 20 to 80 degrees (2θ) with a scanning step of 0.02 degrees (2θ), and the parameter in XRD tests are wavelength 0.154 nm, voltage 40 kV, and current 44 mA.Each peak was tted by the MDI/JADE software (Version 2020, Materials Data, Livermore, CA, USA).A scanning electron microscope (SEM) (Crossbeam 540, Carl Zeiss AG, Oberkochen, Germany) was used to observe the microstructure morphologies of the fabricated coatings.An energy dispersive Xray spectroscopy (EDS) system equipped was used to analyze the element compositions.Before the observation, a thin Iridium layer was coated on the samples to create excellent conductivity and the signal to noise ratio dramatically, resulting in crisp and clear SEM images.To further quantify the size and volume content of the reinforcements, the Image J software was utilized to process the SEM images for both Inconel 718 coatings and B 4 C-Inconel 718 coatings under the mode of black and white.The volume fraction of Laves phase is quanti ed by the area ratio of Laves phase in the cross-sectional surface.To obtain an accurate result, for both Inconel 718 coatings and B 4 C-Inconel 718 coatings, ve cross-sectional surfaces were analyzed to get the average volume fractions of Laves phases.
The microhardness of the deposited coating layers was tested by a Vickers microhardness tester (Phase II, Upper Saddle River, NJ, USA) with a 10 N normal load.Two samples fabricated by the LDED process were tested to measure the microhardness.For each sample, the microhardness test was performed ten times on random positions of the polished surface.The wear rate was tested and measured by dry sliding tests with a silicon carbide ball at room temperature using a mechanical testing system (PB1000, Nanovea, Manufacturer in Irvine, CA, USA).During the dry sliding test, the SiC ball was sliding on the surface of the coating for 15 minutes with a load of 0.2 N, a constant sliding speed of 3mm/s, and a sliding distance of 3 mm.An OLYMPUS DSX-510 optical microscope (OM) (Tokyo, Japan) was used to observe the worn surface and obtain the scratching width after the dry sliding tests.The wear rate W r was calculated by Eq.1.
where, V was the wear volume lost, mm 3 , F was the normal load, N; v was the sliding speed, mm/s; T was the duration time, s.

Phase composition
Figure 3 shows the XRD results of the phase composition analysis on Inconel 718 coatings and B 4 C-Inconel 718 coatings.In the XRD results of Inconel 718 coatings, the peaks matched up reasonably well for a γ Ni-Cr-Fe phase (PDF-# 65-0380) but were shifted to the left slightly.Most likely there were a small amount of other Nb-rich and Mo-rich precipitations doped into the structure.Similar results were also reported in the laser DED fabricated Inconel 718 parts [16,17].With the addition of B 4 C, some new phases

Molten pool size and temperature
The thermal image of the molten pool for the 3 layers single track coatings (t = 25s) and the maximum temperature during the fabrication are shown in Figure 4.As shown in Figure 4(a), the shape of both Inconel 718 coatings and B 4 C-Inconel 718 coatings was an irregular circle, when taking the melting point of the Inconel 718 (1450 °C) as the molten pool boundaries.With the same laser deposition input parameters of laser power, hatch distance, scanning speed, and powder feed rate, the molten pool size of B 4 C-Inconel 718 coatings (~ 2.25 mm) was much larger than that of Inconel 718 coatings(~ 1.35 mm).
As shown in Figure 4(b), the maximum temperature of Inconel 718 coatings was slightly increased from 1700 °C to 1800 °C during the fabrication due to the heat accumulation effects.As a comparison, the maximum temperature of B 4 C-Inconel 718 coatings was much higher than Inconel 718 coatings, which were signi cantly increased from 1900 °C to 2200 °C during the fabrication.The higher temperature increasing rate and the higher maximum temperature of B 4 C-Inconel 718 coatings indicated that the insitu reactions between B 4 C and Inconel 718 released a large amount of heat [15].A larger molten pool size and higher molten pool temperature had the potential to increase the uidity of the liquid materials in the molten pool, which could improve the bonding quality, coating density, and reinforcement distribution.

Analysis of element composition
Figure 5 shows the element compositions of Inconel 718 coatings and B4C-Inconel 718 coatings.In Inconel 718 coatings, there were white regions distributed in a grey matrix.Point 1 and point 2 were probed in the white regions and grey matrix regions, respectively.The element compositions of the white regions and grey matrix regions were shown in Table 2(a).Both white regions and grey matrix regions were rich in Ni, Cr, and Fe.Compared with grey matrix regions, white regions had a higher content of Nb and Mo and represented the Laves phase, which was generated during the solidi cation process through the segregation of Nb and Mo from the matrix [18,19].The grey matrix regions were Ni-Cr-Fe phase, in which the contents of Nb and Mo were lower than that in feedstock Inconel 718 powders.The reason was that the formation of Laves phase consumed the solid solution elements of Nb and Mo in the matrix.
Different from the Inconel 718 coatings, there were three kinds of regions in B 4 C-Inconel 718 coatings.
Besides the white precipitated regions and the grey matrix regions, there were also irregular black regions distributed near the white regions.The size and morphology of black regions were different from the feedstock B 4 C powders, which could be considered as the in-situ synthesized ceramic reinforcement.As discussed in Section 3.1, during the fabrication, there were in-situ reactions between B 4 C and Inconel 718 to form Ni 3 B, NbB, and Cr 3 C 2 .To further investigate the composition of the in-situ synthesized ceramic reinforcement, point 1 and point 2 were probed in the black regions and grey matrix, respectively.The detailed element compositions of different regions were shown in Table 2(b).In the black regions, the major elements were Ni, B, Cr, and C. Compared with the grey matrix regions, the black regions had higher element content of B and C, indicating the black regions were mainly borides and carbides that were formed through the reactions between B 4 C and Inconel 718.Combined with the XRD results, the black regions could be con rmed as the in-situ synthesized ceramic reinforcements of Ni 3 B, NbB, and Cr 2 C 3 .

Analysis of microstructural morphology
Figure 6(a) shows the comparisons on microstructural morphologies of Inconel 718 coatings and B 4 C-Inconel 718 coatings fabricated by laser DED process.In Inconel 718 coatings, it could be seen that the growth of long strip-shaped precipitated Laves phase was in the building direction (vertical direction).This phenomenon was caused by the unidirectional solidi cation direction.During the fabrication of Inconel 718 coatings and B 4 C-Inconel 718 coatings by the laser DED process, the localized heat ux within the molten pool was the main factor in uencing the grain growth, which was usually vertical downward due to the good thermal conductivity of Ti substrates [2].The gradient of temperature in the vertical direction was higher than that in the horizontal direction, leading to the bottom-up generation of Laves phase.
With the addition of B 4 C, the in-situ formed ceramic reinforcements were generated along with the precipitated Laves phase.The ceramic reinforcements and Laves phase were uniformly distributed in laser DED fabricated B 4 C-Inconel 718 coatings.The formation mechanism could be in three stages: Secondly, when the laser beam moved away, the liquid materials in the molten pool cooled down and started to be solidi ed.In the beginning, due to the high melting point, NbB and Ni 3 B precipitated and grew into long strip particles, which provided nucleation sites for the Laves phase.Then, the Laves phase nucleated around the ceramics in the liquid Ni-Cr-Fe.
Finally, the liquid Ni-Cr-Fe solidi ed to generate the matrix.The in-situ formed ceramics and precipitated Laves phase were uniformly distributed in the Ni-Cr-Fe matrix.
To further investigate the morphology, size, and volume fraction of precipitated Laves phase, the SEM images under the mode of black (matrix and ceramic reinforcement) and white (precipitated Laves phase) were shown in Figure 6 (b).With the addition of B 4 C, the size of precipitated Laves phase in B 4 C-Inconel 718 coatings was decreased.The major reason was that the element B from the decomposition of B 4 C reacted with Inconel 718 and then formed boride particles, which could act as the active nuclei in the molten pool.A large number of active nuclei increased the number of precipitated Laves phase grains in a certain melting volume, which suppressed the growth of Laves phase into a coarse stripshaped structure during the solidi cation.It could be also seen that with the addition of B 4 C, the volume fraction of Laves phase was signi cantly decreased from 17.742% to 6.396%.The major reason was that the Nb, an important constituent element of the laves phase, reacted with B 4 C to form NbB during the fabrication, resulting in the lower Nb content in the molten pool.As a result, the formation of Laves phase in B 4 C-Inconel 718 coatings was suppressed.Similar results had been reported in the Inconel 625 with 0.4 wt.% boron coatings fabricated by the gas tungsten arc deposition process [20].

Fabrication Defects
The effects of B 4 C on the fabrication defects of Inconel 718 coatings were shown in Figure 7.It can be seen that the irregular-shaped defects were distributed in the In Inconel 718 coatings.As a comparison, there was no irregular-shaped defect in the B 4 C-Inconel 718 coatings.In the LDED process, the irregularshaped fabrication defects on the cross sectional surface were usually caused by the lack of fusion, as demonstrated by Zhang et al. [21].Since the molten pool size and temperature of B 4 C-Inconel 718 coatings were much larger than that of Inconel 718 coatings (as discussed in Section 3.1), more feedstock powders could be caught and fully melted by the molten pool.The formation of su cient overlaps was promoted, which could reduce the lack of fusion at the boundaries of the molten pool.
It can be also seen that there were also some micropores distributed in both Inconel 718 coatings and B 4 C-Inconel 718 coatings.In the laser DED process, the cooling rate was extremely high.Entrapped gas bubbles were di cult to be e ciently expelled before the molten pool solidi ed [22].Compared with Inconel 718 coatings, the number of micropores in B 4 C-Inconel 718 coatings was much smaller than that in Inconel 718 coatings.There were two major reasons.Firstly, due to the heat generated from the in-situ reactions between B 4 C and Inconel 718, the molten pool of B 4 C-Inconel 718 coatings had the thermodynamic conditions with higher uidity.Gas bubbles could oat upward faster and escape from the molten pool before the molten pool solidi cation.Secondly, the solidi cation process of B 4 C-Inconel 718 coatings during the fabrication was extended by the exothermic reactions and the melting point depressant (B element), which could also promote the escape of gas bubbles [20].

Microhardness
Figure 8 shows the comparison of the microhardness of Inconel 718 coatings and B 4 C-Inconel 718 coatings fabricated by laser DED process.The microhardness of B 4 C-Inconel 718 coatings (348 HV) was signi cantly higher than that of Inconel 718 coatings (264 HV).As discussed in Section 3.3, ceramic reinforcement particles were uniformly distributed in the Ni-Cr-Fe matrix.These reinforcements had higher hardness than Inconel 718 alloy, which could bear the load during the microhardness tests.In addition, it was well known that a large amount of Laves phase in the form of long-strip morphology was detrimental to the strength and hardness of Inconel 718 since it depletes the elements needed for precipitation strengthening.With the addition of B 4 C, the coarse and long strip-shaped Laves phase was changed to needle-shaped Laves phase with re ned size, uniformed distribution, and lower volume fraction.The negative impacts of Laves phase on mechanical properties of Inconel 718 coatings were greatly reduced.In addition, the microhardness was positively correlated to the density.As discussed in Section 3.4, with the addition of B 4 C, the fabrication defects and micropores were signi cantly decreased, which increased the ability to support the load and improve the microhardness.

Wear resistance
Figure 9 shows the differences in the morphology of worn surfaces between the Inconel 718 coatings and B 4 C-Inconel 718 coatings.In the dry sliding tests of Inconel 718 coatings, scratches and a large area of coating delamination can be observed on the worn surface of the Inconel 718 laser cladding layer.These features indicated that local cold welds between the coating surfaces and silicon balls occurred under a load.Such phenomenon always occurred in adhesive wear, which had been reported by Hurricks et al. [23].In addition, some grooves could also be observed on the worn surface of Inconel 718 coatings, which occurred when a hard surface (SiC ball) pass over a soft surface (Inconel 718 coatings).The wear mechanism of Inconel 718 coatings in the dry sliding tests was the combination of adhesive wear and abrasive wear.Compared with Inconel 718 coatings, the worn surface of laser DED fabricated B 4 C-Inconel 718 coatings showed less delamination and smoother grooves.The major reason was that the high hardness of the B 4 C-Inconel 718 coatings made it di cult for the silicon ball to penetrate into the surface.
The morphologies of the worn surfaces indicated that the abrasive wear played a dominant role in the B 4 C-Inconel 718 coatings.
Figure 10 shows the wear rate of Inconel 718 coatings and B 4 C-Inconel 718 coatings.The wear rate of B 4 C-Inconel 718 coatings was much smaller than that of Inconel 718 coatings, indicating a higher wear resistance.As discussed in Section 3.5.1, the microhardness of B 4 C-Inconel 718 coatings was much higher than that of Inconel 718 coatings.According to Archard's wear equation, the volume of wear is inversely proportional to the hardness of the material.The higher hardness of B 4 C-Inconel 718 coatings also had a great in uence on the improvement of wear resistance.In addition, the metal matrix composite coatings with higher density always showed excellent wear resistance.Compared with Inconel 718 coatings, there were fewer fabrication defects in B 4 C-Inconel 718 coatings, which was attributed to suppressing fracture and the spalling of hard particles and hard protuberances during the dry sliding test.
Besides the effects of hardness and density on wear resistance, the wear mechanism also had great impacts on the wear resistance.The B 4 C-Inconel 718 coatings with abrasive wear type had a smoother worn surface, which could reduce the coe cient of friction during the dry sliding test.Under the same test condition, the friction force was reduced, resulting in a smaller material removal volume and higher wear resistance.

Conclusions
In this study, the in-situ synthesized semicontinuous network microstructural ceramic reinforced Inconel Layer thickness (µm) 432 Powder feeding rate (g/min) 2

include
Ni 3 B (PDF-# 48-1223), NbB (PDF-# 32-0709), and Cr 2 C 3 (PDF-# 65-0897) were identi ed by the XRD analysis.In addition, there was no peak representing B 4 C in the XRD results.It could be con rmed that the feedstock B 4 C powders were completely reacted with Inconel 718 to form multi-component ceramic reinforcements during the fabrication.Inconel 718 as a kind of nickel-chromium alloy also contained signi cant amounts of niobium, iron, and molybdenum along with lesser amounts of aluminum and titanium.Because of the instability of B 4 C in liquid metallic materials, there were in-situ reactions to form Ni 3 B, NbB, and Cr 3 C 2 phases: It could also be seen that compared with Inconel 718 coatings, the intensity of the Ni-Cr-Fe peaks in B 4 C-Inconel 718 coatings was slightly decreased.The major reason was that Ni and Cr reacted with B 4 C to form ceramics during the fabrication, reducing the content of Ni-Cr-Fe.The positions of Ni-Cr-Fe peaks in B 4 C-Inconel 718 coatings were further shifted to the left, which was farther to the positions of the standard Ni-Cr-Fe phase.The peak shift indicated that the content of Nb-rich and Mo-rich precipitations (Laves phase) in B 4 C-Inconel 718 coatings was higher than that in Inconel 718 coatings.

Firstly, the laser
beam melted the substrates and generated a molten pool.As discussed in Section 3.1, the temperature of the molten pool reached the melting point of B 4 C during the fabrication.Both B 4 C and Inconel 718 powders were fully melted to form the liquid-state mixture.The added B element would react with part of liquid Nb and Ni constituent to form liquid NbB and Ni 3 B.
718 coatings were fabricated by laser DED process.The in-situ synthesis of the fabricated B 4 C-Inconel 718 coatings and the effects of B 4 C on the phase composition, microstructure, microhardness, and wear resistance were investigated.The major conclusions are drawn as follows: 1.During the laser DED process, there were reactions between B 4 C and Inconel 718 to form Ni 3 B, NbB, and Cr 2 C 3 .During the solidi cation, the needle-shaped NbB and Ni 3 B whiskers were uniformly distributed in Inconel 718 matrix and provided nucleation sites for the Laves phase.With the addition of B 4 C, the size and volume fraction of Laves phase were signi cantly reduced.2. The in-situ reactions generated a large amount of heat.The molten pool temperature was signi cantly increased, which could suppress the formation of the lack of fusion defects and micropores in B 4 C-Inconel 718 coatings.3.With the addition of B 4 C, the microhardness of B 4 C-Inconel 718 coatings was signi cantly increased due to the generation of ceramic reinforcements and the re nement of Laves phase.The wear mechanism of B 4 C-Inconel 718 coatings was abrasive wear, which was different from the Inconel 718 coatings (adhesive and abrasive wear).The wear resistance of B 4 C-Inconel 718 coatings was higher than Inconel 718 coatings because of higher hardness, lower porosity, and smoother worn surface.