Microstructure and Wear Performance of CeO₂-Modified Micro-Nano Structured WC-CoCr Coatings Sprayed with HVOF

Abstract

Rare earth elements have been widely utilized in material manufacturing to enhance properties in various ways. In order to obtain the WC-10Co4Cr coating with uniform distribution of rare earths, CeO₂-modified powder was prepared by mixing 1 wt.% nano-sized CeO₂ during the initial ball-milling of the powder fabrication process. Bare and CeO₂-modified WC-10Co4Cr coatings were deposited via high velocity oxygen fuel spraying to investigate the impact of CeO₂ modification on the coating’s microstructure, mechanical properties and abrasive wear performance. The results show that the addition of CeO₂ increased the interface energy, inhibiting the formation of the Co₃W₃C phase during the powder sintering process, as well as the W₂C phase and CoCr alloy during the high-velocity oxy-fuel (HVOF) process. This led to a significantly decreased porosity and higher concentration of undissolved Cr-rich areas. The microhardness and fracture toughness of the CeO₂-modified coating were 1230 HV_0.3 and 5.77 MPam^1/2, respectively. The abrasive wear resistance of the CeO₂-modified coating was only 70.9% of that of the unmodified coating. Due to the weak cohesive strength between WC and Cr, Cr-rich areas were preferentially removed, resulting in an increased wear rate in the CeO₂-modified coating.

1. Introduction

Due to the high hardness and outstanding wear resistance, cermet coatings have emerged as a viable solution for a variety of wear protection applications under severe service conditions, including impeller shafts, pump housings, nuclear applications, etc. [1,2,3]. Among the coating preparation techniques such as thermal spraying and laser cladding [4,5], HVOF exhibits relatively lower particle temperature and higher particle velocity, and has been widely employed for the deposition of cermet coatings, especially WC-based cermet coatings to avoid serious formation of brittle carbides and achieve better coating properties [6,7]. The type and percentage of the metal binders in WC-based cermet also greatly affect the coating performance. For example, Liu et al. [8] investigated the influence of Co content on the microstructure, mechanical properties and cavitation erosion resistance of HVOF-sprayed coatings. It was found that WC-12Co coating showed improved cavitation erosion resistance compared to WC-17Co coating, although the increase in Co content led to lower porosity and higher fracture roughness. Among a variety of the most commonly utilized WC-based cermet coatings, such as WC-10Ni, WC-12Co and WC-10Co4Cr, the WC-10Co4Cr coatings generally show more excellent mechanical properties, corrosion and wear resistance than other coatings [9,10,11].

In addition to the chemical composition, the structure of feedstock powders also exerts a crucial influence on the performance of WC-Co(Cr) coatings sprayed with HVOF. According to previous studies, improved coating hardness and fracture toughness can be realized simultaneously by introducing nano WC particles [12,13,14]. However, serious decarburization of nano WC particles occurs during the HOVF process, resulting in the development of undesired phases such as W₂C or even W. As a consequence, the mechanical properties of nano WC-based cermet coatings can be negatively affected [15,16]. Bimodal WC-based cermet coatings, which consist of both nano and micron-sized WC particles, have been proposed and extensively investigated as a means of achieving a balance between high hardness and low decarburization [17]. In fact, it is not necessary for bimodal WC-based cermet coatings to contain nano-sized WC particles. For example, Yuan et al. [18] fabricated bimodal WC-12Co coatings by incorporating submicron WC particles into conventional HVOF-sprayed WC-Co coatings. The submicron WC particles were presented at splats’ interfaces, resulting in improved microhardness and a decrease in wear rate of approximately 15%. On the basis of bimodal coatings, Ding et al. [19] deposited a multimodal WC-10Co4Cr coating which consisted of micron, submicron and nano WC particles. The multimodal coating exhibited homogenous microstructures and a 36% increase in wet abrasive resistance compared to the nanostructured coating.

The utilization of rare earth elements in material manufacturing is widespread due to their ability to enhance properties in various ways [20]. One such example is the reinforcement of steel, where the addition of rare earths (REs) refines the grain structure and diminishes defects, ultimately improving the overall strength of steel [21]. In recent years, a number of studies have demonstrated that the addition of REs can also refine the microstructure of WC-based cermet coatings and enhance coatings’ properties [22,23]. Liu et al. [24] fabricated CeO₂-modified WC-12Co powder via mechanical mixing of the commercial WC-12Co powder with 1 wt.% nano CeO₂ and deposited the corresponding CeO₂-modified WC-12Co coating via HVOF. The CeO₂-modified coating showed reduced porosity, higher microhardness and improved sliding wear resistance. Reduced porosity, increased microhardness, lower coefficient of friction and smaller amount of wear can also be found in CeO₂-modified conventional WC-10Co4Cr coating sprayed via HVOF on 65Mn steel [25].

Although the addition of REs has a number of advantages, it can be challenging to assess the practice of incorporating REs into WC-based cermet coatings due to its complexity. The effectiveness of REs depends on the amount of REs added and the method employed for the addition. According to previous studies [23], the optimal amount of REs added to WC coatings is 1–2 wt.%. In most current research, the method of adding REs to WC-Co(Cr) coatings is by ball milling the REs and WC-Co(Cr) feedstock powders. In this case, REs can only gather at the splat boundaries of the coating. As a result, the influence of REs on the microstructure of the coating is insignificant, and the mechanism of RE modification is difficult to explain. In this work, a CeO₂-modified micro-nano WC-10Co4Cr coating was fabricated by adding CeO₂ during the initial powder agglomeration and sintering process. The effects of CeO₂ on the microstructure of WC-10Co4Cr powder and coating was thoroughly analyzed, as well as the further influence on the mechanical properties and abrasive wear resistance of the coating.

2. Materials and Methods

2.1. Materials

The 1 wt.% CeO₂-modified micro-nano WC-10Co4Cr powder (CeP) was fabricated via agglomeration and sintering process (Chongyi Zhangyuan Tungsten Co., Ltd., Ganzhou, China). To ensure the uniform distribution of CeO₂ in WC-10Co4Cr powder, the nano CeO₂ with particle size less than 50 nm were mixed with micro WC particles (1~2 μm), nano (60~180 nm) WC particles, Co metal and Cr metal during the initial ball-milling process before granulation. To obtain the feedstock powders, the ball mill product was dried and granulated using a centrifugal spray dryer, followed by continuous sintering in a molybdenum wire furnace at a temperature range of 1100–1250 °C under a hydrogen protective atmosphere. The mass frictions of micro and nano WC particles in the whole carbides are 70% and 30%, respectively. For comparison, bare micro-nano WC-10Co4Cr powder (BP) without CeO₂ was made from the same company and both feedstock powders have a size distribution of 15–45 μm after grading.

2.2. Coating Preparation

The bare WC-10Co4Cr coating (BC) and CeO₂-modified coating (CeC) were deposited on 316 stainless steel substrates via the JP8000 HVOF spraying system with a barrel length of 152.4 mm. Prior to spraying, the substrates were grit-blasted with 60 mesh Al₂O₃ and then cleaned with acetone to remove any contaminants from the surface. The coating deposition process was optimized by monitoring the temperature and velocity of in-flight particles using a SprayWatch 2i diagnostic system.

Table 1. Spraying parameters of the bare and CeO₂-modified WC-10Co4Cr coatings.

Kerosene (L/h)	Oxygen (m3/h)	Feed Rate (g/min)	Spraying Distance (mm)	Velocity (mm/s)
22.7	55.2	75	380	500

shows the spraying parameters that had been optimized for both powders. The spraying process was intermittent to avoid overheating of the substrates above 160 °C; meanwhile, the substrates were cooled via compressed air.

2.3. Characterization

The crystalline structure of the bare and CeO₂-modified WC-10Co4Cr powders and coatings were measured via X-ray diffraction (XRD; BRUKER, Bremen, Germany). The measurement was conducted in a θ–2θ scan mode using Cu Kα radiation with a wavelength of 1.5406 Å. The surface and cross-section morphologies of the coatings were analyzed via a VEGA3 scanning electron microscope (SEM; TESCAN, Brno, Czech Republic) at an accelerating voltage of 20 kV, while the energy dispersive spectrometry (EDS) system embedded in the SEM was used to detect the element composition. The hardness was measured on the coating cross-sections by using Vickers microhardness tester, with an applied load of 300 g and a dwell time of 10 s. The fracture toughness was measured similarly via the Vickers hardness tester, with an applied load of 5 kg for 15 s, and then calculated according to Evans and Wilshaw Equation (1) [26].

$$K_{IC}=0.079\frac{P}{a^{3/2}}\log \left(\frac{4.5a}{c}\right)$$

(1)

where P, a and c represent the load, half of the length of indentation diagonal and half of the crack length, respectively. The average and standard deviation of microhardness and fracture toughness are based on the results of ten measurements.

2.4. Abrasion Wear Tests

The abrasive wear tests were carried out by using an MLG-130B wet sand/rubber wheel abrasion tester (Zhangjiakou Chengxin Testing Equipment Manufacturing Co., Ltd., Zhangjiakou, China), which was designed in accordance with ASTM G105. Figure 1 displays the schematic of the abrasion tester. The rubber wheel had a hardness of 70 HA and a diameter of ~177.5 mm. The samples were wire-cut to the size of ~57 × 25 × 5 mm³ to fit the device. A load of 100 N was applied over the samples. Throughout the testing process, the coatings were submerged in a slurry consisting of 1.5 kg of 20~40 mesh size SiO₂ abrasives and 1 kg of water. Following the initial 600 revolutions carried out to remove the loose surface layer, the tests persisted for 5 rounds, each comprising 3000 revolutions. The material loss was measured using a balance with 0.1 mg accuracy and the average weight loss was divided by the theoretical density of the bare and CeO₂-modified WC-10Co4Cr coatings (13.92 and 13.79 mg/mm³, respectively). Samples were cleaned with acetone and then dried before weighing.

3. Experimental Results

3.1. The Microstructure of the Powders and Coatings

The surface morphologies of the bare and CeO₂-modified WC-10Co4Cr powders are presented in Figure 2. Both powder particles exhibit a spherical or nearly spherical shape and possess a size distribution that is close to normal, which is consistent with previous studies [27]. Further examination of EDS shows that in addition to W, Co metal binders are almost uniformly distributed in both powers. The distribution of Cr metal binders is much worse as a number of remaining micro particles can be seen in Figure 2d,g, which is associated with the higher hardness and lower ductility of Cr metal compared to Co metal. Uneven distribution is also occasionally found in CeO₂. Although the typical particle size of CeO₂ addition is less than 50 nm, the particles may agglomerate into micro-clusters due to the high activity. From EDS, there is no obvious correlation between the locations of the Cr micro particles and CeO₂ micro-clusters.

The XRD patterns of the powders and coatings are shown in Figure 3. The powders are mainly composed of WC and also a small amount of Co₃W₃C and Co phases. Under the sintering temperature of ~1200 °C and a hydrogen protective atmosphere, Co₃W₃C is formed because of the loss of carbon to Cr metal and the metallurgical reactions of W and Co. If chromium is added not by metal but by Cr₃C₂, according to previous studies by the authors, chromium is already carbon-rich and Co₃W₃C cannot be formed [28]. It is noteworthy that by examining the main Co₃W₃C diffraction peaks located at ~42.8°, the diffraction intensity of Co₃W₃C in the bare powder exceeds that in the CeO₂-modified powder by four times. This indicates that CeO₂ can inhibit the metallurgical reactions that occur even during the sintering process.

The XRD patterns of the two coatings consist primarily of the WC peaks, accompanied by some minor peaks corresponding to W₂C, while the Co₃W₃C peaks are absent. Although the metallurgical reactions become much more violent during the HVOF process as decarburized carbides dissolve into the molten metallic matrix to form a number of Co-W-C compounds, the Co-W-C compounds further form nanocrystalline or even amorphous phase due to high cooling rates and disappear from XRD patterns [29]. The presence of brittle W₂C phases, as a result of high-temperature oxidation during the HVOF process, is much sharper in the bare WC-10Co4Cr coating than in the CeO₂-modified coating, indicating that CeO₂ can effectively inhibit the oxidation of carbides during the spraying process.

The cross-sectional SEM images of the bare and CeO₂-modified coatings are shown in Figure 4, suggesting that both WC-10Co4Cr coatings possess a dense microstructure. According to EDS analysis, the mass percentage of the Ce element is slightly larger than 1%, indicating that CeO₂ has been effectively incorporated into the WC-10Co4Cr coating. In addition to the normal WC-10Co4Cr structure which shows light color, a small amount of dark color areas is also observed in the coatings. Based on the analysis of the images (250× magnification) by using ImageJ software, the proportion of dark areas in the bare and CeO₂-modified coatings is 1.04 ± 0.17% and 2.06 ± 0.37%, respectively, as shown in

Table 2. Porosity and defect density of the bare and CeO₂-modified WC-10Co4Cr coatings.

Coating	Porosity (%)	Defect Density (%)
BC	0.85 ± 0.11	1.04 ± 0.17
CeC	0.23 ± 0.04	2.06 ± 0.37

. These values are usually considered as the porosity of the coatings. However, through direct observation of pores in Figure 4b,d, not only is the number of pores in the CeO₂-modified coating decreased by one order of magnitude compared with the bare coating, but the size of pores is also reduced. The porosity was evaluated by using Photoshop software to manually extract all pores in 2000× SEM images via the quick selection tool and calculate the size of selected areas. Based on this method, the porosity of the bare and CeO₂-modified coatings is 0.85 ± 0.11% and 0.23 ± 0.04%, respectively, as shown in Table 2. One of the main reasons for the formation of pores in HVOF-sprayed coatings is the unescaped gas bubbles within the coatings produced by oxidation of carbides during coating deposition. This is in line with prior research indicating that plenty of phase interfaces between nano-oxides and the matrix can effectively capture point defects and suppresses the growth of bubbles [21]. The significantly reduced porosity in the CeO₂-modified coating confirms that CeO₂ can effectively inhibit the oxidation during the spraying process.

Since the porosity of the CeO₂-modified coating is significantly decreased, the dark areas in Figure 4a,c must contain other abnormal structures. A closer examination at the dark areas in the figures reveals microstructures that are completely different from the pores, as shown in Figure 5. Combined with the EDS results, the areas are enriched with Cr, implying that some of the Cr micro particles in the feedstock powders (as shown in Figure 2) have persisted after deposition. By considering the size difference between the overall dark regions and the porosity, the area occupied by Cr particles is estimated to be 0.19% and 1.83% for the bare and CeO₂-modified coatings. Although there can be some discrepancies in the results due to variations in size calculation methods, it is evident that the concentration of Cr particles in the CeO₂-modified coating is significantly greater. Cr metal can form CoCr alloy with Co at high temperatures. However, the core part of spraying powders has lower temperatures compared to the periphery because of the uneven heating during spraying; as a result, some Cr metal remains due to insufficient reaction temperature and time. The Cr particles in Figure 5 show no sign of forming alloy as the areas are Co-free. More residual Cr particles in the CeO₂-modified coating suggests CeO₂ can also hinder the alloying process of Cr and Co.

In summary, the addition of CeO₂ significantly alters the microstructure of both the WC-10Co4Cr powder and coating, including reducing the porosity and hindering the formation of Co₃W₃C, W₂C and CoCr alloy. Since CeO₂ cannot be dissolved into either WC, Co or Cr, it can only be distributed in the interface. It is evident that the existence of CeO₂ in the interface increases the interface energy, resulting in decreased chemical reaction rates between WC, Co, Cr and oxygen. On one hand, the coating microstructure is improved by reducing the porosity and brittle W₂C phases. On the other hand, many more undissolved Cr particles remain in the coating, which may lead to deterioration of mechanical properties.

3.2. The Mechanical Properties of the Coatings

The mechanical properties of the bare and CeO₂-modified coatings are presented in

Table 3. Properties of the bare and CeO₂-modified WC-10Co4Cr coatings.

Coating	Hardness (HV0.3)	Fracture Toughness (MPam1/2)
BC	1223 ± 27	6.19 ± 0.14
CeC	1230 ± 13	5.77 ± 0.05

, which suggests the two coatings have almost identical average hardness. The hardness of a coating is influenced by a variety of factors, including the porosity, the particle size and the original hardness of the coating materials. The CeO₂-modified coating has denser microstructure than the bare coating, but the materials are softened overall due to the addition of a relatively softer CeO₂ phase and the decrease in the harder W₂C phase compared to the main WC phase. Under the combined effect of these factors, the CeO₂-modified coating has a slightly improved hardness of 1230 ± 13 HV_0.3. It is noteworthy that the standard deviation of the hardness of the CeO₂-modified coating is only half of that of the bare coating (1223 ± 27 HV_0.3), indicating that the coating microstructure is more homogeneous after modification except for the Cr particle areas. The standard deviation of the fracture toughness of the two coatings shows the same trend, but the average fracture toughness of the CeO₂-modified coating is reduced by 7%. The fracture toughness is determined via the indentation method and negatively correlated with the length of the crack, which preferentially propagates along defects such as splat boundaries. CeO₂ in the splats can be regarded as a defect on the interface, which may weaken the interface strength and lead to a decrease in the fracture toughness.

3.3. Abrasive Wear Results

Figure 6 displays the specific wear rates of the bare and CeO₂-modified WC-10Co4Cr coatings in abrasive wear tests. Both coatings have a much higher specific wear rate in the initial 600 r, which is caused by the surface states of HVOF-sprayed coatings. During the HVOF process, the deposited particles are strengthened by the impact of subsequent spraying particles, but this effect is dramatically weakened in the surface. As shown in Figure 7, the surfaces of the two as-sprayed coatings have loose microstructures with a few holes. After removing the surface layer, specific wear rates of bare and CeO₂-modified coatings decrease and become stable after three and one rounds, respectively, suggesting the CeO₂-modified coating has more homogeneous microstructure near the surface. The typical abrasive worn surface morphologies of the two coatings are shown in Figure 7b,d. Although all the worn surfaces show similar smooth morphologies after five round abrasive tests, from Figure 6, the average specific wear rate of the last three rounds of the CeO₂-modified coating is 1.41 × 10⁻⁵ mm³/Nm, which is 38% higher than that of the bare coating (1.02 × 10⁻⁵ mm³/Nm). This suggests the wear resistance of micro-nano WC-10Co4Cr coating is obviously weakened after CeO₂ modification.

4. Discussions

According to previous research, the abrasive wear behaviors of the coating primarily rely on the relative size and hardness of the abrasive and hard phases within the coating [30]. In this study, the hardness of the SiO₂ abrasive is lower than that of the main WC phase and the brittle W₂C phase, but higher than the CoCr metal binders. Thus, the metal matrix is preferentially removed by SiO₂; then, under continuous impact by the abrasives, the carbides are broken and finally pulled out from the metal binders when the support is insufficient. From Figure 7b,d, most observed microstructures are micro and nano WC particles stuck out from the metal matrix, which is consistent with the literature [10]. However, if the sizes of WC particles are smaller than the cutting depth of the abrasives, the wear mechanism varies in which the abrasives can cut the carbides and metal binders together at relatively high speed, leaving a number of grooves on the coating surface. This can occur since the average size of the SiO₂ abrasive used in this work is two orders of magnitude higher than that of micro WC particles. The typical groove microstructure can be observed in Figure 7b (marked with a rectangle), in which there is no obvious height difference between the WC particles and metal binders. Therefore, the abrasive wear of the two coatings is the combination of the two mechanisms and the wear rate of the second mechanism is higher than that of the first one. However, this cannot explain the 38% higher specific wear rate of the CeO₂-doped coating since the main influential factor of both mechanisms is hardness and the two coatings have almost identical microhardness.

It has been found that the CeO₂ modification results in a considerably greater quantity of undissolved Cr particles that remain within the coating. On closer examination, the Cr-rich areas are not merely composed of Cr but also a number of WC particles inside, as shown in Figure 8, suggesting WC particles are mixed into Cr particles during ball milling to form Cr-WC composite. From Figure 8a,c, only nano and a part of submicron WC particles can still remain on the cross-sectional surface of Cr-rich areas, while the fact that the pits match micro WC particles in terms of shape and size suggests micro WC particles have been pulled out from Cr during the process of grinding the samples with sandpaper. This indicates the cohesive strength between WC and Cr is rather weak, as confirmed by the morphology of the Cr-rich areas on the worn surface in which the pull out of micro WC particles is also observed. The shape of the Cr-rich areas is flat as the spraying particles have undergone a flattening process during HVOF; thus, the size of the areas on the surface is much larger and the impact on the wear resistance cannot be ignored. Compared to the surrounding area, the Cr-rich areas are rapidly worn out due to weak cohesive strength, forming a large crater with a diameter of tens of microns and a depth of several microns on the coating surface. When a SiO₂ abrasive passes through the crater, it can easily penetrate the surface with a depth larger than the size of micro WC particles, which can lead to the formation of groove and acceleration of the wear rate by launching a higher-speed wear mechanism. It is widely acknowledged that the binder phases of WC-based cermet coatings are easy to cut to form grooves due to the lower hardness [9]. The Cr-rich region in the coating can be regarded as a large area of low hardness binder. Thus, the reduced wear resistance of the CeO₂-modified coating is caused by the much higher concentration of undissolved Cr. Figure 9 depicts a schematic of the wear mechanism, showing that the preferred removal of the Cr-rich area leads to the acceleration of abrasive wear.

5. Conclusions

In this work, the microstructure, mechanical properties and wear performance of HVOF-sprayed WC-10Co4Cr coatings containing uniformly dispersed CeO₂ were systematically investigated. The following key findings can be drawn from the study regarding the impact of CeO₂ on the microstructure and properties of the coating:

• The addition of CeO₂ caused significant changes in the microstructure of both the WC-10Co4Cr powder and coating, such as impeding the formation of Co₃W₃C, W₂C and CoCr alloy. The presence of CeO₂ at the interface is expected to elevate the interface energy, leading to a reduction in chemical reaction rates between WC, Co, Cr and oxygen. There was a notable decrease in the porosity of the coating modified with CeO₂, which also contained a significantly increased density of Cr-rich regions.
• The CeO₂-modified coating showed comparable microhardness (1230 HV_0.3) and fracture toughness (5.77 MPam^1/2) to the bare coating, but with significantly reduced standard deviations, suggesting that the coating microstructure became more homogeneous as a result of CeO₂ modification.
• Due to the weak cohesive strength between WC and Cr, the CeO₂-modified coating that contained more Cr-rich areas suffered a faster wear rate, resulting in reduced abrasive wear resistance compared to the bare coating.