Effect of nano-Al2O3 addition on the microstructure and erosion wear of HVOF sprayed NiCrSiB coatings

Development of nanostructured high velocity oxy-fuel (HVOF) coatings with low porosity, high strength and increased wear resistance is still in its infancy. Combining nanoparticles with conventional microscale powders are increasingly being investigated to use with feedstock materials for thermal spray processes. Accordingly, this work investigates the addition of nano-Al2O3 particles on the microstructure and erosion wear of NiCrSiB HVOF coating in a stainless steel (AISI 304) substrate. Particle analysis of the NiCrSiB feedstock was conducted and the maximum allowable addition of Al2O3 nanoparticles have been identified using the ‘mass mixture ratio’ model considering both the particle size and density. Consequently, two cases are considered and their performance analysed: a maximum allowable case of 1.4 wt%, followed by a 0.17 wt% addition of nano-Al2O3 with NiCrSiB. Scanning Electron Microscope (SEM), Energy Dispersive Spectroscopy (EDS) and x-ray Diffraction (XRD) analysis were employed to inform the microstructure, material composition and phase spectrum of the resulting coatings. Subsequently, the nanostructured coating was exposed to both a pull-off adhesion strength test and hot air jet (450 °C) hard particle erosion to characterise its performance. It was found that the microhardness of the HVOF NiCrSiB coating improved from 576 HV0.3 to 748 HV0.3 with the addition of 1.4 wt% nano-Al2O3. Furthermore, the nanostructured coating also exhibited high erosion resistance at a 90° erodent impact angle. The increase in erosion wear resistance was due to the increase in the hardness as a result of the nano-Al2O3 addition.

they feature a low flame temperature (<3000 K) at a high particle velocity (∼500 m s −1 ). Furthermore, HVOF also has high hardness, low degree of oxidation, and higher bond strength despite being comparatively cost effectiveness [19][20][21].
Over the years, a wide variety of coating materials have been successfully developed for use with thermal spraying technique. In that context, NiCrSiB coatings have gained popularity for wear and hot corrosion resistance, along with high-temperature oxidation. More importantly, the NiCrSiB coatings feature a low environmental impact [22][23][24][25] in comparison to other coatings of similar properties. The excellent properties to a large extend can be attributed to the formation of hard phases influenced by chromium (Cr), boron (B) and silicon (Si) which is then dispersed in the nickel (Ni) matrix. The self-fluxing properties of B and Si also lowers the melting temperature allowing for a lower energy usage [26,27]. Despite the excllent wear and corrosion resistance of NiCrSiB coatings their lower hardness (in comparison to carbides and ceramics) makes them unsuitable for a range of industries. Consequently, the use of reinforcements are increasingly being sought to improve erosion-corrosion resistance of NiCrSiB coatings. In this regard, a range of micron-sized hard phases such as WC, CrC, TiC, SiC, Al 2 O 3 , Cr 3 C 2 , TiO 2 , CeO 2 and ZrO 2 have been investigated to reinforce NiCrSiB [28][29][30][31][32][33].
Recently, the effect of nano-structuring on coatings have been investigated and significant improvement has been noticed in comparison to their micron-sized variants. Recently, Fernandes et al [34] combined nanostructured ZrO 2 to Ni-based alloy coating using atmospheric plasma spraying technique on a low carbon steel substrate. The subsequent sliding wear tests at room temperature showed that both the hardness and wear of coatings improved despite the decrease in friction coefficient. Addition of nano-Al 2 O 3 particles to Ni-alloy coatings were carried out by Hou et al [35] using a plasma transferred arc overlay welding. It was found that a 0.8% (wt.) nanoscale Al 2 O 3 raised the sliding wear resistance by 1.57 times. Addition of nano-Al 2 O 3 to NiCrMoCu composite by Xu et al [36] through double glow plasma alloying on a stainless steel also showed excellent wear resistance. It was also reported that the wear weight loss of the composite alloying layer was 45% and 65% less than that of stainless steel and the NiCrMoCu respectively.
Wang et al [37] on the other hand studies a combination of Al 2 O 3 -SiC and nano-CeO 2 on the high temperature wear behaviour of Ni-alloy coatings produced by laser cladding. The results demonstrated the nano particles significantly increased the hardness of coatings (NiCoCrAlY: 350.5, NiCoCrAlY+Al 2 O 3 : 397.8, NiCoCrAlY+SiC: 484.9 and NiCoCrAlY+CeO 2 : 385.7). A further comparative study using nano-CeO 2 and Sm 2 O 3 to reinforce Ni through laser deposition was carried out by Zhang et al [38]; where an improvement in microhardness and wear-corrosion resistance was observed. Deposition of nano-Cr 2 O 3 in combination with Ni-alloys on steel using duplex surface treatment was reported by Liu and Xu [39]. It was found that the Ni-Cr 2 O 3 composite layer exhibited higher erosion-corrosion resistance. Furthermore, the slurry dynamic action was shown to not severely decrease the corrosion resistance and protective efficiency of the Ni-Cr 2 O 3 layer.
Dispersion of a 1% mass-fraction nano-Al 2 O 3 to create high-strength NiCoCrAlY superalloy coating by CO 2 laser was established by Wang et al [40]. This pioneering study aimed to quantify the 'nano-effect' of nanoscale Al 2 O 3 particles; i.e. the nanos-size plays a substantial role in improving the microstructure of the coating itself. Other notable works include the use of activated-combustion high-velocity air-fuel (AC-HVAF) by Liu et al [41] to add nano-WC-12Co to WC-10Co-4Cr coatings. The subsequent slurry erosion test showed enhanced microhardness, wear and erosion resistance of the coated samples. The best performance was reported at a 15% addition of nano-WC-12Co to WC-10Co-4Cr.
While these studies validates the 'nano-effect' potential when it comes to developing high performance coatings, the influence of nonscale particle addition on the wear resistance of NiCrSiB coatings are yet to be established. Of particular interest, in this regard, is the effect of nano-Al 2 O 3 addition on the microstructure, and erosion wear resistance of HVOF sprayed NiCrSiB coatings. The interest in adding nano-Al 2 O 3 ceramic material is primarily due to its high level of hardness, wear and oxidation resistance. Accordingly, this study attempts to use nano-Al 2 O 3 to reinforce Ni-based alloy coatings processed by HVOF thermal spraying technique. As opposed to plasma spraying, the HVOF technique is selected to demonstrate the suitability of the coating for high density applications. Accordingly, the nano-structured Al 2 O 3 -NiCrSiB coating will be deposited on an AISI304 stainless steel using HVOF spraying and the resulting performance in terms of microstructure, microhardness and erosion wear will be investigated.

Substrate and coating feedstock
The deposition of coating feedstock was carried out on AISI304 stainless steel 25×25×6 mm substrate. The selection of AISI304 as the substrate of choice was due to its wide application in erosion-corrosion environment at high (>600°C) service temperatures [8]. Particularly, the 304 variant is the most common boiler steel under both high temperature and oxidising environment [42,43]. The coating surface of the substrate was manually grit blasted at 320-500 μm resulting in a surface roughness of approximately 9±1 μm. According to Tillmann et al a surface roughness (∼9 μm) can improve coating adhesions by promoting material interlocking between the feedstock and the substrate material. A case also validated by Sen et al [44] and Wang et al [45] while preparing their respective samples for nano-particle reinforced coating. In order to remove any contaminants that may have been introduced during the grit blasting process, the substrate was subsequently cleaned in acetone following an ultrasonic bath. Optical emission spectroscopy (ASTME 1086-08) was employed to evaluate the chemical composition of the substrate material; the results of which are presented in table 1.
Regarding the coating feedstock, micron sized NiCrSiB along with nano sized Al 2 O 3 (hereafter referred to as nano-Al 2 O 3 ) powders were the choice of materials. Particle size characterization of the feedstock was carried out and found that the NiCrSiB and nano-Al 2 O 3 powder feature an average size D50 of 20 μm (micrometre) and 45 nm (nanometre) respectively. The nominal particle size distribution of the feedstock was characterised using the well-established laser diffraction technique using a Malvern Mastersizer 2000 (UK) following ASTM C1070 [46].

2.2.
Coating procedure HVOF has been regarded as one of the most efficient techniques for depositing high-performance coatings at moderate cost [47]. The procedure starts with the use of a carrier gas to transfer the feedstock powders into a combustion chamber. The feedstock is then heated to a high temperature as a result of the combustion of mixture gas. The molten and semi-molted particles are then accelerated by the combustible gases at supersonic velocity through a nozzle, subsequently being deposited on the substrate material resulting in a coating [48].
The general consensus is that HVOF is the most suitable methedology for NiCr [49][50][51] or WC-Co-based [52,53] coatings, to improve both wear and oxidation resistance at an industrial scale. The HVOF system adopted in this study heats the feedstock using a mixtures of oxygen and liquified petroleum gas (LPG). Although multitude of fuels such as hydrogen, methane, propane, propylene, acetylene, natural gas and kerosene maybe used in compatible systems. The use of hydrogen is known to result in enhanced coating quality. However, this comes at a higher price; consequently the use of hydrogen is hard to justify when the particular coating application is not targeted at demonstrating the highest quality. The scope of this research is limited to studying the influence of nano-Al 2 O 3 addition on the coating quality. Consequently, since all the coating cases being compared are conducted under consistent process parameters, the use of LPG is adequate. Table 2 summarises the HVOF process parameters used to generate the NiCrSiB+1.4 wt% nano-Al 2 O 3 coating at a constant thickness of 250±30 μm.

Coating characterisation
The key properties relating to the mechanical performance of the coated surface are the microhardness and the adhesion strength. Consequently, both microhardness and adhesion strength measurements were carried out for all the coated samples. The microhardness (HV) tests were carried out using a Wilson-402MVD (Buehler) at 300-g load (15 s.). The stud pull adhesion tests were performed using a 14.2 mm (dia.) aluminium stud glued to  the coating using an epoxy paste; the assembly was oven cured at 150°C (3 h.) following best practice reported by Elmoursi and Patel [54]. In order to ensure consistency, three samples were prepared and tested under identical conditions using an Elcometer-506 pull-off adhesion strength tester. Quantification of the coating porosity was carried out through optical microscopy investigation of the sample cross-section. A total of ten sliced sections were investigated and the cross-sectional data of each tile was analysed using ImageJ [55][56][57]. The average measurement is then reported as the single number porosity of the coating. Further to optical microscopy, x-ray Diffractometry (XRD) was carried out to the investigate the phases present both on the powder and the coating. The XRD equipment of choice was the Regaku Ultima featuring a 3 kW x-ray generator at a 0.4×12 mm focus size.
SEM was carried out to study the materialography [58][59][60]of both the feedstock powder, coated sample, erodent and eroded surface. Lastly, the material composition of both the powder and the coated surface were investigated using the Energy Dispersive x-ray Spectroscopy (EDS). While there are numerous techniques [61][62][63] to prepare the sample for EDS, a mirror finish grade sequence mechanically polishing and hot resin mounting were the choice of procedure adopted in this study.

Hot air jet erosion
Hot air jet erosion was chosen to characterize the erosion performance of the coating to simulate an erosion environment close to that experienced by powder generation power plants or boiler flue tubes. According to a critical review of powerplant economizer tube failures carried out by Pal et al [64], a temperature of 417°C is reported as the flue gas temperature. Consequently, hot air jet erosion of the NiCrSiB+1.4 wt% nano-Al 2 O 3 coated surface were carried out at a slightly higher temperature of 450°C.
Alumina particles (50 μm avg.) featuring a flaky and angular morphology was the erodent of choice. The erosion tests were carried out at three impact angles between 30°and 90°at 30°intervals. The coated surface were exposed to the hot air jet for 10 min at an erodent feed rate and velocity of 5 g min −1 and 40 m s −1 respectively. The exit nozzle diameter was 1.5 mm placed at a distance of 10 mm (standoff) from the sample. The erosion rate (g/g) is evaluated as the ratio of erosion wear mass loss in grams to the mass of the erodent particle in grams. The tests were carried out following ASTMG76-18 [65,66] using a Ducom air jet erosion tester. When it comes to addition of nano particles, the maximum amount (wt%) of nano-Al 2 O 3 powder that can be added NiCrSiB powder depends both on the particle diameter and the powder density. Consequently, this study adopts the nano (nm) to micron (μm) mass mixture ratio proposed by Wang et al [67] as shown in equation (1): where, wt e is the 'mixture mass ratio' of the respective micron and nano sizes feedstock powders. Accordingly, D is the average particle diameter of the micron powder and d is the diameter of the nano powder. Finally, the density of micron and nano powders are accounted as g m (NiCrSiB) and g n (nano-Al 2 O 3 ) respectively.
Substituting the respective data to equation (1) reveleated that the maximum content (wt%) of nano-Al 2 O 3 that can be added to micron-NiCrSiB is 1.4 wt%. Consequently, two proportion of mixing are considered in this study: the maximum limit and half, at of 1.4 wt% and 0.7 wt% of nano-Al 2 O 3 . The two cases (1.4 and 0.7) were considered in order to identify the influence of the nanoparticle quantity on the microstructure and surface properties of the resulting coating (micron-NiCrSiB+nano-Al 2 O 3 ). The mechanical alloying of the feedstock powder mixture was carried out using a ball mill [68][69][70] featuring a planetary arrangement of 250 ml hardened steel vial and 10:1 powder to ball weight ratio at 250 rpm. In order to prevent excessive cold welding during the mechanical alloying process ethanol was used as the process control agent (PCA). Typically, ethanol is absorbed at the surfaces of the powdered particles and aids in reducing excessive cold welding between the powders along with the grinding medium and the milling container. This subsequently reduces agglomeration of particles which affects both flowability and the packing density. The mechanical alloying of all powder mixtures considered in this study were carried out in normal atmospheric conditions. Figure 2 shows the influence of milling time on the particle size distribution (PSD) of the mechanically alloyed NiCrSiB+1.4 wt% nano-Al 2 O 3 powder. It can be seen that a 4-hrs. milling time yields a near normal PSD that is suitable for thermal spraying. Investigating the associated morphology through SEM shows that the particles are spherical and composed of a range of sizes which is beneficial both for powder flowability and splat density.
Excessive milling (6-hrs.) was found to causes fracturing of the Ni-based powder along with significant reduction in particle size (<12 μm). Further increments in grinding time will result in particle and grain sizes decreasing progressively with milling time, reaching nanometre levels. Consequently, mechanically alloyed powder post 4 h. were used as the feedstock for the HVOF process. Figure 3 shows the characteristics of the mechanically alloyed feedstock (NiCrSiB+1.4 wt% nano-Al 2 O 3 ). From the electron diffraction spectroscopy (EDS) analysis (figure 3), it was found that the mechanically alloyed composition is formed of nickel, aluminium, and chromium.

3.2.
Characteristics of the 'NiCrSiB+nano-Al 2 O 3 ' coating XRD was conducted to assess the structure of NiCrSiB+1.4 wt% nano-Al 2 O 3 multilayer coating deposited on steel (AISI304) substrate ( figure 4). The diffraction peaks were indexed by γ-Ni (JCPDS 04-0850) with traces of Cr 23 C 6 , Cr 7 C 3, Ni 3 B and Al 2 O 3 . However, the NiO component reveals that oxidation of the Ni particles has taken place as a result of the HVOF thermal spraying process. Furthermore, amorphisation can also be validated by the dimensional difference of the intensity peaks. The low intensity of Al 2 O 3 peaks in the XRD pattern of NiCrSiB+0.7 wt% nano-Al 2 O 3 , indicates either a low concentration of nanoparticles (that is below the XRD detection threshold) or that the nanoparticle have been fully melted as a result of the high-temperature facilitated by the HVOF flame decomposing to constituent elements.
To facilitate understanding of both the morphology and the composition of the coatings, the cross-sections were analysed and profiled using EDS. Three coating cases were considered as shown in figure 5, NiCrSiB, NiCrSiB+0.7 wt% nano-Al 2 O 3 and NiCrSiB+1.4 wt% nano-Al 2 O 3 . The resulting elemental composition summarized in table 3. Figure 5(a) shows that the HVOF sprayed NiCrSiB coating is uniform and feature fewer number of pores. In the case of nano-Al 2 O 3 added coatings (figures 5(b) and (c)), a lamellar structure of deformed dark Al 2 O 3 reinforcing phase with grey Ni phase is formed.
The presence of pores and un-melted Al 2 O 3 can be observed for both the cases (0.7 and 1.4 wt%) were nano-Al 2 O 3 was added. The melting point of Al 2 O 3 (2100°C) is higher than that of Ni-base alloy (1040°C-1200°C), consequently, the Ni-base alloy powder and nano-Al 2 O 3 particles can absorb sufficient heat for complete melting, extending and depositing to form a lamellar structure in the thermal spray process [32]. The number of lamellae was observed to be more in the 1.4 wt% nano-Al 2 O 3 coating in comparison to the 0.7 wt%.  Consequently, this effect can be attributed to the amount of nano-Al 2 O 3 particle present in the feedstock mixture. table 4 shows the results of porosity, surface roughness, microhardness, and adhesion strength. There is no significant variation observed in the porosity, surface roughness and adhesion strength with the addition of nano-Al 2 O 3 particles. The adhesion strength from the stud pull off test was found to be in the range of [35][36][37][38] MPa for all the coating compositions tested. Investigating the macroscopic images of the failed samples as shown in figure 6, localised failing of the coating can be observed. Nevertheless, the major failure can be characterised as adhesion failure.  When Wang et al [71] and Yin et al [72] added Al 2 O 3 to 75Ni and Al using HVOF and plasma spray respectively; an adhesion strength of 22.9 MPa and 28.8 MPa was observed. In comparison, it is clear that an improvement in adhesion strength (35)(36)(37)(38) MPa) is obtained. The 20% improvement in adhesion strength can be attributed to increase in contact points facilitated by the well-flattened splats. This is a result of the high impact pressure of the melted/un-melted particles owing to the high impact velocity during HVOF spraying.
When it comes to the microhardness performance of the coating as shown in figure 7, the addition Al 2 O 3 nanoparticles was found to have a significant effect (table 4). It was found that the microhardness of NiCrSiB coating increased with the addition of nano-Al 2 O 3 as a result of the high hardness and compressive strength of the Al 2 O 3 particles acting as a reinforcing phase. Furthermore, the Al 2 O 3 addition also resulted in a good cohesive strength and lower porosity HVOF coating.
As listed in table 4, the highest microhardness (748±80 HV 0.3 ) was observed for coating featuring the highest quantity of nano-Al 2 O 3 (1.4 wt%), followed by 0.7% nano-Al 2 O 3 (620±70); a different of approximately 20%. Compared to the coating that featured NiCrSiB alone, this is an improvement of 30% at 1.4% addition of nano-Al 2 O 3 . Comparing the microhardness of the best performing coating (1.4 wt% nano-Al 2 O 3 ) with that of literature as listed in table 5, an improvement in performance can be observed. It can be also seen that HVOF offers superior microhardness performance when adding Al 2 O 3 in comparison to cold spray.

Hot air jet erosion wear
Analysing the eroded sample post hot air jet erosion as shown in figure 8, a gradual reduction in wear scar can be observed as angle of impact increase. This shows that the mechanism of erosion is dependent on the process parameters involved. In the case of hot air jet erosion, it is primarily the impact angle when the erodent material is kept constant. This is because the impact angle can change the erosion behaviour of ductile and brittle materials [77]. Consequently, when the angle is lower, in this case 30°, the divergence of the alumina stream is higher, resulting in an elliptical shape as shown in figure 8; a case also reported by Laguna-Camacho et al [65]. This effect reduces as the erodent impact angle increases resulting in a comparatively smaller circular case at 90°. Nevertheless, in all the cases irrespective of the erodent angle, zonal erosion can be observed: these can be divided into areas of primary, secondary and tertiary erosion. The primary zone is where the erodent makes the first contact with the coating resulting in severe removal of the material, this was followed by an annular section where the erosion is medium followed by the and outside area where the erodent influence is minimum. Classical analysis primarily refers to two classification of material removal from a coating under erosion ductile erosion and brittle erosion. However, according to Aquaro and Fontani [78], classification of ductile versus brittle erosion can be interchangeable based on the shift in peak erosion with respect to the impact angles and associated conditions. A 'ductile erosion' classification is appropriate when a higher wear rate occurs at low impact angles which then goes on to decrease as the impact angle increases. A brittle erosion on the contrary tend to peak at high angle of erodent impact subsequently being reduced as the impact angle decreases.
Comparing erosion performance of the coating as shown in figure 9, the erosion wear was found to be the highest at 30°, subsequently reaching the best performance at 90°. This clearly classifies that the erosion wear mechanism for all the coatings considered in this study are of a ductile nature. This also means that the shear strength of the coating is lower, and they have the capability to absorb a comparatively larger energy resulting from erodent at an impact angle of 90°as opposed to lower. The reason for this can be due to the presence of more Ni phase in the coatings. Overall, the results (figure 9) indicate that an addition of 1.4 wt% nano-Al 2 O 3 improves the erosion resistance of NiCrSiB coatings up to 1.5 times at a 90°impact angle. Although a slight improvement in erosion resistance was observed at 0.7 wt% nano-Al 2 O 3 , the overall effect was found to be insignificant comparison to the performance of NiCrSiB alone.

Wear surface characterisation
The SEM worn surfaces of the coatings reveal the wear mechanism of NiCrSiB and NiCrSiB+1.4 wt% nano-Al 2 O 3 are shown in figures 10 and 11 respectively. Given that a sharp-edged and flaky alumina erodent is used, four major wear features can be observed: ploughing, microcutting, indentation and type I cutting. Comparatively, for the 1.4 wt% nano-Al 2 O 3 addition, no long cracks could be observed, and the debris size is relatively small with significant fracturing of silica. This reflects the good adhesion between nanoparticles and  the NiCrSiB. There are numerous thoughts reported on the mechanism of erosion wear in coatings [79]. It is generally considered that at oblique erodent impact, wear happens primarily as a result of cutting action of the particle irrespective of its shape and size. Consequently, material removal by oblique cutting action can be observed primarily at 30°and 60°impact angles. The material is removed by a ploughing action, displacing materials to the front and side of the particle. Further impacts on the neighbouring site cause removal of highly strained materials from the rim or terminal lip of the crater. When the alumina particle rolls forwards post-contact with the surface, material is removed by repeated impacts on a prominent lip formed by the indenting angular particle and this is termed type I cutting which can be observed at higher angles. Material removal 90°shows indentions and craters which is a result of localised strain. The extruded rim of material that an bee seen around the craters further validates that a ductile wear is taking place.  In addition to the microcutting and ploughing, the material loss in NiCrSiB+1.4 wt% nano-Al 2 O 3 was due to the fracturing and pull of the Al 2 O 3 splat. At 90°impact angle, the impact energy reaches a maximum and causes extensive indentation deformation accompanied with the formation of dislocations. The accumulation of dislocations due to the further impact of the erodent particle, accelerates the Al 2 O 3 splat pull out at the weak bonding boundaries. The fracturing of the Al 2 O 3 splat was attributed to the intersection of lateral cracks formed due to the indentation of irregularly shaped alumina erodent particles on the brittle Al 2 O 3 splat on the coating. Consequently, the addition of 1.4 wt% nano-Al 2 O 3 to NiCrSiB (figure 10) show better erosion resistance as compared to NiCrSiB coating (figure 11), primarily attributable to the 'pinning effect' of the hard Al 2 O 3 particles with higher hardness in the coating in spite of fracturing and pull off of the Al 2 O 3 splat. The large concentration of the fractured alumina observed in figure 11 also shows the improved hardness of the surface.
In summary, the erosion of NiCrSiB coating post hot air jet alumina erodent involves wear by cutting mechanism observed in ductile mode of erosion mechanism (ploughing, microcutting, indentation, type 1). For the case of NiCrSiB+1.4 wt% nano-Al 2 O 3 coating, the fracturing and pull out of the alumina splat observed shows that, even though a ductile mechanism dominates the wear mechanism partial brittle modes are also prevalent. Nevertheless, the best performance for all coatings were observed at perfectly axial impact of erodent (90°) in comparison to oblique angles (30°and 60°) primarily due to the high volumes of Ni-metallic phase.

Conclusions
The study demonstrates the development of NiCrSiB/nano-Al 2 O 3 HVOF coating deposited on a steel (AISI304) substrate. In addition to characterising the erosion wear mechanism, the influence of nano-Al 2 O 3 on the microstructure and surface properties are presented. A 4 h. mechanically alloying was found to be most suitable to obtain a near normal particle size distribution (PSD) when incorporate 1.4 wt% nano-Al 2 O 3 with NiCrSiB for HVOF thermal spraying. The resulting microstructure of the coating consists primarily of Ni, Cr 23 C 6 , and Cr 5 B 3 phases with minor phases of alumina and NiO. Overall, no significant variation in porosity, surface roughness and adhesion strength of NiCrSiB coatings was observed as a result of adding either 0.7 or 1.4 wt% nano-Al 2 O 3 . However, the microhardness of the NiCrSiB coating increased significantly from 576 HV 0.3 to 748 HV 0.3 when 1.4 wt% addition of nano-Al 2 O 3 to the NiCrSiB feedstock. Furthermore, the erosions resistance of NiCrSiB coating improved up to 1.5 times at 90°erodent impact as a result of 1.4 wt% addition of nano-Al 2 O 3 . Even though, the ductile mechanism (ploughing, microcutting, indentation, type-I cutting) was the most prevalent erosion mechanism, the addition of nano-Al 2 O 3 was found to show certain brittle characteristics such as mild cracking and pull out of the alumina hard phase.