Microstructure and wear resistance of laser cladding nano TiC/micro TiB2/C-276 composite coatings on 316L stainless steel

Four metal-ceramic composite coatings were prepared by laser cladding (LC) using Hastelloy C-276, nano-TiC and micro-TiB2 as raw materials to improve the surface properties and extend the service life of 316L stainless steel. The microstructure and mechanical properties were analyzed by metallographic microscopy, scanning electron microscopy (SEM), x-ray diffraction (XRD), Vickers hardness experiments, and friction and wear testing. The TiC and TiB2 phases were detected in the coating, exhibiting homogenous distribution as their content increased. As a result, the strengthening effect of the coating was enhanced, improving microhardness relative to that of the 316L substrate. The wear test revealed that the Hastelloy coating exhibited poor wear resistance, but the addition of TiC and TiB2 particles improved the wear resistance by reducing the wear rate. The lowest average coefficient and wear rate were attained at 50% and 70% ceramic contents, respectively. However, abundant hard particles on the surface easily fell off during the wear, which further impeded the surface degradation and increased the friction coefficient.


Introduction
The 316L stainless steel is widely used in aerospace vehicles, marine engineering, intelligent manufacturing, biomedicine, and other fields due to its excellent mechanical and good corrosion resistance characteristics [1][2][3][4]. However, in recent years, stringent requirements for various properties of metals with respect to mechanical processing, transportation and so on have been put forward [5]. In the face of increasingly complex working conditions, the service life of 316L stainless steel has been significantly reduced [6]. Therefore, it is essential to improve the surface properties of metals so as to prolong their service life.
Currently, the methods used for metal surface modification mainly include thermal spraying [7], hard chromium electroplating [8], ion implantation [9], chemical vapor deposition [10], physical vapor deposition [11], cold metal transfer [12], friction stir welding [13], microwave cladding [14], metal inert gas welding [15], tungsten inert gas welding [16], plasma spray [17] and laser cladding [18], to name a few. However, some of them may cause defects such as interface contamination, interface separation, and cracks that easily appear between the thermally sprayed coatings. For instance, the current efficiency of hard chromium electroplating against some toxic substances is still low. The ion implantation method is characterized by poor surface finishing and the formation of porous morphology. Chemical vapor deposition is a simple but slow deposition process also leading to the high surface roughness. In turn, physical vapor deposition results in fewer contaminants and higher coating density, but is not reproducible and expensive for the widespread use. Cold metal transfer can reduce spattering, but the low heat input is not conducive to the formation of brittle intermetallic compounds. On the contrary, friction stir welding has received consideration attention as an eco-friendly, cost-effective and efficient technology, but possesses shortcomings that may affect the quality of the final product. Microwave cladding allows the matter to be uniformly heated and mitigates the thermal gradient phenomenon; however, this method is not fully applicable to metallic materials. MIG welding is a simple and efficient approach, but may induce defects such as cracks and pores. Although TIG welding has a broad range of applications, it is mainly Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. used for welding thin sheets of various metals. Plasma spraying is an inexpensive and powerful method, but the surface quality of the sprayed coatings still needs to be substantially improved. Laser cladding consists in using a high-energy laser beam to melt the powder that is either pre-placed on the substrate or synchronously transported to the substrate surface so as to join both materials and solidify them after cooling. It has the advantages of low dilution rate, small heat-affected zone, and good combination with the substrate [19][20][21].
In recent years, there have been many studies on the laser cladding of cermet, and research has found that cermet coating can effectively improve surface performance [22][23][24]. Shengyu Xu et al [25] prepared TiC/Ni60 composite coatings on Ti6Al4V substrates and investigated the effect of three groups of scanning speeds on the microstructure and properties of the coatings. The hardness of the coating was found to significantly increase in all three groups. The highest hardness at a scanning speed of 8 mm s −1 was twice that of the substrate, and the corresponding wear resistance was four times greater than that of the substrate. Chonggui Li et al [26] investigated the influence of different vibration frequencies on the laser melting of WC-Ni composite coatings. It was shown that an appropriate increase in vibration frequency could refine the microstructure, and when the vibration frequency was 556 Hz, the microhardness was about three times that of the substrate (about 1021 HV). The wear of the coating was also reduced by about 25% compared to the substrate. Yongbin Peng et al [27] prepared FeCoCrNi-WC composite coatings in two different ways so as to analyze their properties. It was found that the microhardness of the laser cladding layer was higher than that of the plasma cladding layer, and the coefficient of friction of the former layer was lower by about 0.29, while that of the latter layer varied from 0.5 to about 0.9 due to adhesive wear. Hao Liu et al [28]added Ti to AlCoCrFeNi to generate TiC in situ and investigated the effect of Ti content on the composite coating. It was found that the addition of Ti could refine the grain size, and the microhardness continued to increase with the increase of Ti content. The AlCoCrFeNiTi1.0 composite coating had a microhardness value of up to 1027.5 HV, which was approximately 59% higher than that of the AlCoCrFeNiTi0 coating. Under wear conditions at 25°C, the wear scar width of the AlCoCrFeNiTi1.0 composite coating was 411.53 μm, being approximately 30% less than that of the AlCoCrFeNiTi0 coating. Chandrasegaran Raahgini et al [29]studied the effect of VC content on the properties of In625-VC composite coatings. Their results showed that the microhardness increased with the increasing VC content. When the VC mass percentage was 15%, the microhardness was approximately 1.65 times that of the In 625 coating. Shi Yan et al [30]prepared a Co50/Ni/WC gradient composite coating on 20CrMnTi alloy steel and carried out parameter optimisation and performance research. It was shown that the gradient composite coating increased the hardness of the substrate by 2 times and the wear resistance by 36.4 times. Currently, most researchers explore the performance of the cermet coatings by changing the laser power, scanning speed, and other factors. Meanwhile, there is relatively little research on the distribution of two kinds of ceramics with different particle sizes in the coating and their impact on the coating performance.
In this study, a continuous laser was used to prepare composite coatings with different ceramic contents on 316L stainless steel substrates by adding nano TiC, micro TiB2, and Hastelloy C-276 via a preset powder method and laser cladding technology. The microstructure and properties of the composite coatings were studied, in which nano TiC and micro TiB2 were half of the content.

Experimental process
A 25 mm × 25 mm × 10 mm 316l stainless steel (Taizhou Qiuxin Metal Products Co., Ltd.) was selected as the substrate. Before the experiment, the surface of the substrate was cleaned with ethanol, and its chemical composition is presented in table 1. The cladding materials were Hastelloy C-276 powder (Xindun Metal Materials Co., Ltd.), TiC powder (with the average particle size of 40 nm and 99.99% purity, Xindun Metal Materials Co., Ltd.) and TiB2 powder (with the particle size of 1 μm and 99.9% purity, Xindun Metal Materials Co., Ltd.) shown in figure 1. The powders were mixed using a planetary mill (Model-F-P400; Make-FOCUCY, China) for 2 h to prepare four compositions with different contents before coating. The coating numbering method is summaried in table 2. This experiment adopted the preset powder method, and fiber laser(Model-A2000D fiber laser; Make-Raycus RFL, China) was used for laser cladding. The laser power was 1000 W, and the scanning speed was 10 mm s −1 . The motion was controlled by the ROKAE XB16 six-axis robot(Model-XB16 six-axis robot; Make-ROKAE, China). Argon was used as the protective gas during the cladding process. The

Test method
To better study the performance of the coating, the samples were cut by the wire EDM instrument into 10 mm×10 mm×10 mm pieces. The pieces were embedded in a metallographic inlaying machine and then polished with 360 #, 600 #, 1000 #, 1500 #, and 2000 # sandpapers. Manual polishing was afterward performed until there was no scratch on the sample surface. After polishing, the samples were etched with metallographic etching solution for 65-90s and then cleaned with ethanol. The microstructure of the strengthened layer was observed in an optical microscope (Model-M3LY630T; Make-Shunyu, Holland). The scanning electron microscope (Model-Nova Nano SEM 450; Make-FEI, America), equipped with an energy dispersive spectrometer (Model-VANTAGE-DS1; Make-NORAN, America), was applied to examine the microstructure of the strengthened layer and the element distribution in the specific areas. The x-ray diffractometer (Model-X PERT3 POWDER; Make-PANalytical B.V., Japan) was used to analyze the phase composition of the strengthened layer. In addition, an X'pert HighScore Plus software was chosen for peak matching [31]. This software is widely used in related studies because of its high accuracy. The microhardness of the coating was measured at the same distance by means of a microhardness tester

Results and discussion
3.1. Phase composition and microstructure Figure 2 shows the XRD patterns of the substrate and four coatings. The results showed that the substrate consisted mainly of the Cr-Ni-Fe-C austenitic phase. Coating A was a single-phase face-centered cubic Hastelloy C-276 alloy. A Ni-Cr-Co-Mo rich phase was detected, which accounted for the substrate phase of solid solution. At the same time, elements such as Cr, Mo, Co, and other were dissolved into their structures, playing a positive role in their solution strengthening [32,33]. The main phases in coatings B, C, and D were TiC, TiB2, Fe3Ni2, and (Fe, Ni) phases. With the increase of TiC/TiB2 content, the peak values corresponding to TiC and TiB2 also increased. During laser cladding, TiC and TiB2 particles are not completely decomposed, which played a role in strengthening hard particles in the cermet coating. Figure 3 depicts the microstructure of coatings A, B, C, and D. The main component of coating A was Hastelloy C-276, and its melting point was 1325 ℃ ∼ 1370 ℃. During laser cladding, the alloy powder could be fully melted. After cooling and solidification, the coating exhibited a well-formed and even microstructure free of cracks, pores, and other defects. Figures 3(b)-(d) show the microstructures of coatings B, C, and D, in which the ceramic mass proportions were 30 wt%, 50 wt%, and 70 wt%, respectively. The EDS surface scanning of the coatings was carried out, and the point scanning was performed at the position A. The energy spectrum is shown in figure 4, and the percentage of elements are given in table 3. According to the results, the granular material in the coating was TiC, and the lath material was TiB2. Under the Marangoni effect caused by the surface tension gradient, some small light particles flowed and formed agglomerates during cladding [34,35]. As shown in figure 3(b), the ceramic content in the coating was low (about 30%). The granular nano-TiC and lath micro-TiB2 inclusions were unevenly distributed. Although the both phases exhibited no obvious aggregations, the distribution of TiC particles was relatively concentrated, which strengthened the coating area. When the ceramic content increased gradually, its distribution in the coating became more uniform. As shown in figure 3(c), TiC granules were almost uniformly distributed around TiB2, forming agglomerates, and their size increased. At the same time, no agglomerations were formed by micro-TiB2 inclusions, and the interior of the coating was strengthened. When the ceramic content reached 70%, the ceramic was distributed over the entire coating. As the TiC particles become larger, the TiB2 phases also appeared to gather, strengthening the entire coating.  hardness of the substrate was relatively low (about 180 ± 15 HV), followed by those of the heat-affected and cladding regions. As shown in figure 5(b), the average microhardness of the four coating groups significantly increased compared to the substrate. The hardness value of coating A was 265 ± 25HV, being 47.2% greater than that of the substrate. However, the average hardness of the coating was significantly improved after adding ceramic particles. At 30 wt%, 50 wt%, and 70 wt% ceramic contents, the average hardness values were 563 ± 40   HV, 861 ± 55 HV, and 1169 ± 70 HV, respectively. The microhardness of the cladding layer was greatly increased due to the presence of the hard TiC and TiB2 phases. During laser cladding, the convection mass transfer could have affected the uniform distribution of the phase within the cladding, so that the microhardness value of the top of the cladding might be lower than that of the inner region [36]. In addition, as the TiC and TiB2 contents increased, the microhardness of the cladding layer was refined, which had a positive effect on the microhardness and strengthening of the ceramic phase. At the same time, the uniform distribution of TiC and TiB2 phases hindered the movement of dislocations, which promoted the uniform dispersion strengthening [37,38].

Wear resistance
The coefficients of friction (COF) trend and the average COFs of substrate and coatings are shown in figure 6. As can be seen from figure 6(a), the wear evaluation took two stages. At the run of wear, COF also increases sharply with the increase of wear time, which is mainly dependent on the properties of the substrate and coating. The COF of substrate and coating tends to be stable over time. However, coating B still exhibited a significant change in the stable stage, mainly because the brittle material phase during the wear process peeled off from the surface and even participated in the wear, thus accelerating the structure degradation. Figure 6(b) shows the average COFs of the substrate and the coatings. The average COF of the substrate was 0.401 ± 0.005. The COFs of coatings A and D were higher than that of the substrate, being 0.489 ± 0.01 and 0.403 ± 0.006, respectively. In turn, the COFs of coatings B and C were lower than that of the substrate, being 0.343 ± 0.005 and 0.314 ± 0.005, respectively. Coating A exhibited the uniform structure and element distribution, resulting in its relatively stable change in the friction coefficient but relatively poor wear resistance to the substrate.
At 30%∼50% ceramic mass fractions, the average COF of the coatings decreased, and the wear resistance increased; When the ceramic content reached 70%, the average COF of the coating was similar to the substrate, with no significant improvement in the wear resistance, which was due to the presence of abundant ceramic hard phases. During the wear process, the hard phases fell off and became the part of the worn pair, the coating, and the wear debris, thus making the friction coefficient relatively larger. The change in the trend of the COF curve of coatings B, C, and D was considerable in comparison to that of the substrate. This was owing to the different melting temperatures of TiC and TiB2, which standed at 3150°C [39] and 2980°C [40]. Therefore, the ceramic particles could not be completely melted during laser cladding. Their height distributions in the horizontal direction were inconsistent, resulting in uneven wear surfaces, which had a great impact on the COF of the coating [41]. Figure 7 shows the low-magnification, three-dimensional, and high-magnification wear morphology images of the substrate surface. The wear width of the substrate was 1410 μm. As seen from figures 7(b) and (c), many furrows and pits emerged in the wear process of the substrate, indicating that the substrate was undergoing adhesive wear during the wear. Figure 8 displays the surface wear morphologies of the four groups of coatings. The wear width of the coating reflected the degree of the loading caused by the counterface ball to the surface. According to figure 8, coating A had the largest wear width of 1469 μm, indicating its poor wear resistance compared to the substrate. However, the wear width of the coating decreased with the addition of ceramics. The smallest wear width of 645 μm was observed at 50% ceramic addition. It meant that the wear resistance of the coating could be improved by adding a certain amount of ceramic. Figures 9 and 10 depict the high-magnification and three-dimensional wear surface morphologies of the coating, respectively. As shown in figures 9(a) and 10(a), the furrow in coating A was deeper than that of the substrate, and there were also many pits formed by metal peeling. The wear marks were obvious, and the coating was relatively soft, indicating that the wear resistance of coating A was worse compared to that of the substrate, and its wear mechanism was adhesive wear. When ceramic particles were added, the coating structure was refined, and the hardness of the coating was greatly improved relative to the substrate. Thus, the hardness contributed significantly to the wear resistance [42]. As seen from figures 9(b) and 10(b), the wear of the 30 wt% coating was mainly characterized by a series of furrow-like features and drop pits. The wear degree was shallower than that of coating A. The wear mechanism was mainly adhesive wear. Figures 9(c) and 10(c) show that the wear degree at 50 wt%, becomes lighter, and there were a number of shallow furrows and drop pits. The wear mechanism corresponded to adhesive wear. Figures 9(d) and 10(d) revealed that at 70 wt%, the surface was characterized by small and shallow drop pits, and the wear mechanism was mainly adhesive wear. The above finding indicated that TiC and TiB2 could alleviate the adhesive wear, reduce the coating wear, and improve the wear resistance. During the wear process, Hastelloy C-276 was first cut off due to its low hardness, and the  ceramics gradually protruded, effectively preventing further wear degradation. Hence, the wear degree of the coating was also small. Figure 11(a) reflects the influence of nano-TiC and Micro-TiB2 on the strengthening of cermet coatings. When a few ceramic particles were added, the lighter particles followed the Marangoni effect, resulting in the concentrated distribution. However, when two kinds of ceramics of different particle sizes were added, their  distribution gradually became uniform as the ceramic content increased. The nano-TiC inclusions first formed aggregates and then enlarged. The aggregation of micro-TiB2 particles was observed at a 70% ceramic content. Figure 11(b) depicts the influence of ceramics on the wear process of the coating. Initially, the counterface ball scratched the surface, damaging the coating and wearing off the Hastelloy from the surface, which gradually exposed ceramic particles. Their presence hindered the wear process and reduced the wear degree. The hardness is a key factor in the wear performance of the coating. The more ceramic particles are present in the coating, the higher the coating hardness is and the flatter the wear trace is. However, when there are abundant hard phases, ceramic particles fall off easily due to their brittleness in the wear process and contribute to the degradation, resulting in a larger friction coefficient.

Conclusions
In this study, four cermet strengthening layers were prepared on 316L stainless steel substrates via laser cladding technology using nano-TiC, Micro-TiB2, and Hastelloy C-276 powders as raw materials, and their microstructures and properties were studied.
(1) When the ceramic content was 30%, TiC and TiB2 phases were relatively dispersed, and only local areas were strengthened. The uniform distribution of TiC and TiB2 was observed at a 50% ceramic content, at which TiC inclusions tended to aggregate, and a strengthened coating was obtained. At a 70% ceramic content, the ceramic particles were evenly distributed in the coating, with agglomerated TiC and TiB2 particles, and the coating interior was strengthened. This also shows that ceramics with different particle sizes can be evenly distributed in the coating.
(2) The microhardness of the coating was effectively increased due to the presence of TiC and TiB2 phases, and the microhardness value was proportional to the ceramic content. The average microhardness of the ceramic-free Coating A was 265 ± 25 HV. At 30 wt%, 50 wt%, and 70 wt% ceramic additions, the average microhardness values of the coating were 563 ± 40HV, 861 ± 55 HV, and 1169 ± 70 HV, respectively.
(3) The Coating A exhibited the lower wear resistance than the 316L substrate. In turn, TiB2 and TiC improved the wear resistance of the coating and alleviated its adhesive wear. At a 30% ceramic mass fraction, the wear surface of the coating was mostly characterized by pits. In comparison, at 50%, the wear surface exhibited furrows, and the average COF was the minimum. The wear rate was the lowest at a 70% ceramic content due to the higher hardness, and many hard phases easily peeled off in the wear process. The average COF became larger when the peeled hard phases were added to the wear process. Overall, the wear mechanisms of the four groups of coatings could be mainly classified as adhesive wear.