Influence of Spray Angle on Particle Deposition and Thermal Shock Lifetime of Embedded Micro-Agglomerated Particle Coatings

Abstract

The development of gas turbine technology has led to an increase in the complexity of the geometric shape of the sprayed workpiece. Consequently, it has become more difficult to maintain the perpendicularity of the spraying angle during the spraying process, thereby impacting the structure and performance of the coating. This study uses the atmospheric plasma spraying method to simultaneously spray two types of powder for the preparation of embedded micro-agglomerated particle (EMAP) coatings. The spraying process is conducted at four different angles, ranging from 90° to 30°, in order to analyze the influence of the spray angle on the particle deposition and coating performance. The experimental results demonstrate that the relative deposition efficiency, hardness, and elastic modulus of the EMAP coatings decreased as the spray angle decreased. The porosity exhibited a reduction when the spraying angle dropped from 90° to 50°, followed by a significant rise at 30°. The greatest relative amount of second phase particles embedded in the coating appeared at a spraying angle of 90°, amounting to 10.8%. The smallest amount was found at a spraying angle of 30°, with a relative quantity of 2.2%. Furthermore, the molten droplets of the first phase matrix powder underwent extension and fragmentation along the angular direction at low angles. At an angle of 90°, the maximum average thermal shock life was 40.6 cycles, with the best stability of thermal shock life. The decrease in the spraying angle resulted in a deterioration in both the thermal shock life and its stability.

1. Introduction

Thermal barrier coatings (TBCs) are functional coatings widely used for thermal protection and for the insulation of hot end components in gas turbines and aviation engines [1]. The classic TBC is a double-layer structure composed of a metallic bond coat (BC) and a ceramic top coat (TC). The ceramic top coat takes advantage of the low thermal conductivity of ceramics to isolate them from high temperatures and protect the metal substrate [2]. The metal bond coat reduces the interface stress and reduces the mismatch of the thermal expansion coefficient between the ceramic layer and the metal matrix [1]. Atmospheric plasma spray (APS) is a common method for the preparation of TBCs. During the plasma spraying process, plasma is employed to rapidly heat ceramic powder to their molten state. The molten powder is subsequently propelled forward through a jet stream and deposited onto the desired surface. In this process, a large number of molten particles undergo deformation and rapid cooling after a high-speed impact on the surface of the workpiece. The solidified layer particles stack and deposit on the workpiece to form a coating [3,4,5]. Therefore, the coating of plasma spraying has a layered structure, and there are large numbers of pores and cracks between the coating layer particles formed by the incomplete combination of lamellar particles [6,7].

An optimal coating performance is typically achieved when the spraying angle is 90° [8,9]. However, the spraying process may not ensure a fully vertical spraying angle due to the intricate shape of the workpiece, unreasonable trajectory planning, and constraints imposed by the spraying machine’s structure. Variations in spraying angles can alter the direction of impact of sprayed particles on the substrate surface, resulting in a shielding effect and affecting the spreading morphology of the spraying particles, ultimately leading to changes in the structure and performance of the coatings [10]. Some studies have been carried out on the impact of the spray angle on the structure and performance of coatings. These studies mainly focus on high-velocity oxy-fuel (HVOF) spraying and APS technologies, which are commonly employed for TBC deposition. As the HVOF spraying angle decreases, the spreading of the spraying particles onto the substrate will generate additional elongation along the spraying angle, resulting in an increase in the defects and porosity of the coating and a decrease in the microhardness and elastic modulus [9,11]. The rise in horizontal velocity leads to a greater number of particles bouncing off and sliding away, which ultimately causes a reduction in the efficiency of the coating deposition [9,11]. The erosive effect of the particles themselves and the rebounding of high-speed, unmelted particles at low angles both reduce the roughness and friction coefficient of coatings [12]. Correspondingly, the coating exhibits good water lubrication wear performance [13]. The dry sliding wear performance of the coating is observed to decrease as a result of the increase in pore cracks [14]. APS provides a similar conclusion to HVOF spraying. As the angle decreases, the powder deposition morphology will change along the direction of the spraying angle, and the maximum thickness of the spray bead will also decrease [15]. In comparison to HVOF, APS has a smaller spray particle speed, and the shot blasting effect and erosion effect are not obvious, which increases the coating porosity and decreases the thermal conductivity at low angles [8]. For suspension plasma spraying (SPS), under the condition of a low spraying angle, the coating will produce columnar microstructures inclined along the spraying angle, resulting in a decrease in the erosion resistance and thermal shock life of the coating [16]. As observed from the previously described examples, investigating the alterations in the coating structure and performance when exposed to varying spraying angles holds considerable importance in directing the practical spraying procedure.

Previous work established a porous TBC deposition method by modifying the conventional APS technique, which simultaneously sprayed two types of 8 wt % yttria stabilized zirconia (YSZ) particle powders to prepare a hybrid structure consisting of a dense coating matrix and randomly distributed areas of porous embedded particle clusters (PEPCs). The controlled-structure TBC is referred to as an embedded micro-agglomerated particle TBC (EMAP TBC) [17], where the second phase of micro-agglomerated YSZ powder is the reason for forming the PEPC region and the main source of coating porosity. According to previous investigations, EMAP coatings exhibit significant advantages in strain resistance, insulation ability, and sintering resistance [18,19]. In addition, it is possible to regulate the distribution and morphology of PEPC within the coating through alterations in the spraying process [20].

Nonetheless, a lack of relevant studies exists regarding the influence of the spraying angle on the properties of EMAP coatings. Furthermore, when compared to conventional YSZ coatings, EMAP coatings need to take into account the influence of the angle on the two types of powders. Therefore, this paper prepared EMAP coatings under four different spraying angles and evaluated the changes in the microstructure and mechanical properties of the coatings under different spraying angles. Specifically, a quantitative analysis was conducted on the changes in the area and aspect ratio of special pores of the coatings at different angles. Then, to analyze the single particle impacts, two types of powders were also sprayed at different spraying angles to analyze the influence of the angles on the spreading of particles. Finally, the thermal shock test was conducted on EMAP coatings at different spraying angles to evaluate the influence of spraying angles on the thermal shock resistance of the coatings.

2. Experiments

2.1. Coating Deposition

In this study, the F4-MB APS system (Oerlikon Metco, Wohlen, Switzerland) was employed to prepare all samples. The substrate was a nickel-based superalloy disc (IN 738) with a diameter of 25.4 mm and a thickness of 3 mm, and the surface of the substrate was grit-blasted and then ultrasonically cleaned with alcohol. Subsequently, a NiCrAlY powder (45–106 μm, Beijing Sunspraying New Material Co., Ltd., Beijing, China) was sprayed via APS to develop a metal bonding layer with a thickness of 120 μm.

Two types of powders (Anhui Yingrui Youcai Technology Co., Ltd., Wuhu, China) with different microstructures were used for the preparation of the ceramic layers. Solid spherical 8YSZ powder was employed as the first phase matrix powder (named “Powder #1”), and micron agglomeration 8YSZ powder was selected as the second phase embedded particle powder (named “Powder #2”). They had similar particle size distribution ranges of approximately 20–80 μm, which had been measured in a previous work [19]. The specialized pore structure of Powder #2 is depicted in Figure 1, which was made through agglomeration using a significant quantity of 8YSZ primary feedstock particles ranging from 1 to 8 μm in size. The deposition process involved the simultaneous injection of two phase powders into a plasma jet at separate locations. Figure 2 illustrates that the injection location of Powder #1 into the plasma jet was fixed at the exit of the torch nozzle. Powder #1 would be entirely melted to form a dense coating matrix (the first phase matrix) when deposited onto the substrate. The injection position of Powder #2 was set at a distance of 35 mm from the outlet of Powder #1 to ensure that Powder #2 was unmelted. Finally, the EMAP coating was formed, consisting of a coating matrix provided by molten Powder #1 and unmelted Powder #2 uniformly embedded in the coating matrix (the second phase embedded particles). The spraying trajectory is depicted in Figure 3. Four sets of coatings were prepared using the conventional “Z”-shaped path at 90°, 70°, 50°, and 30° spraying angles. The spraying process parameters have been determined in a previous work [19]. The parameters exhibit consistency, and the specific parameters of the spraying process are presented in

Table 1. Plasma spraying parameters for TBC deposition.

Parameters	Bond Coat	Ceramic Top Coat
		First Phase	Second Phase
Current, A	600	550
Power, KW	40	36
Primary gas flow rate, Ar, slpm	50	40
Carrier gas flow rate, H2, slpm	7	8
Spray distance, mm	120	85
Traverse speed of gun, mm/s	1000	500
Powder feeding rate, g/min	10	10	20
Thickness, μm	120	300

.

2.2. Coating Characterization

Cross-sectional samples of the coatings were prepared using a metallographic polishing procedure. Specifically, this procedure involves grinding the sample from a low grit to high grit, and finally polishing it. Afterwards, an analysis of the microstructure of the coating cross section was conducted through the utilization of scanning electron microscopy (SEM, Hitachi S3400N, Tokyo, Japan). The Image J 1.53e software was utilized to conduct a quantitative analysis of the specific pore area in the SEM image of the coating. This analysis enabled the measurement of both the porosity and thickness of the coating. Specifically, binary images and the Otsu algorithm were used to partition pore areas through thresholds, and digital image processing techniques were used to analyze the area and aspect ratio of special pore areas. To reduce errors, a minimum of five backscattered electron images (at a magnification of 300×) were collected for every sample to calculate the mean porosity. The error bar was determined by the standard deviation of the data. Similarly, at least 15 points were measured for each sample to obtain the average thickness of the coating. Then, the thickness per pass of the coating was determined through a comparison between the mean thickness and the number of coating passes, and this ratio was used to characterize the relative deposition efficiency of the coating. The Vickers indentation hardness tester (HX-1000TM/LCD, Shanghai Taiming Optical Instrument Co., Ltd., Shanghai, China) was utilized to measure the hardness and elastic modulus of the coatings. A 300 gf indentation load was employed during the measurement process with a holding time of 15 s. To obtain reliable measurements, at least 15 points were measured on each sample. The elastic modulus (E) of the coating can be determined using the following equation [20]:

$$\frac{b}{a}-\frac{b^{\prime }}{a^{\prime }}=\alpha \frac{H}{E}$$

where α is a constant with a value of 0.45, H is the hardness, b′/a′ is the ratio of the short indention diagonal to the long indention diagonal after elastic recovery, and b/a is 1/7.11 for the Knoop indenter geometry.

2.3. Single Particle Spraying Experiment

The substrate of the nickel-based superalloy underwent a metallographic polishing process to achieve a mirror-like surface. The elimination of surface defects on the substrate was carried out to reduce any potential interferences that may arise from such defects during the observation of the single spraying samples using a scanning electron microscope. Subsequently, the two types of powders were subjected to a single spray pass using the APS system at four distinct spraying angles. The majority of the spraying parameters were the same as those used during the preparation of the EMAP coatings. To ensure the unobstructed observation of the powder morphology, the spraying speed was elevated to 1000 mm/s to decrease the risk of powder stacking at low spraying speeds. A SEM was used to observe the spreading morphology and embedding situation of two types of powders under four spraying angles. This article employs Image J 1.53e software to determine the area percentage of the second-phase embedded particles on a specific substrate surface using threshold analysis techniques. This approach enables the quantification of the relative embedding amount of Powder #2. To ensure measurement accuracy, a minimum of five images were counted per sample, with each plane having the same area.

2.4. Thermal Shock Assessment

This study uses the gas flame thermal shock test rig (Shaanxi Dewei Automation Co., Ltd., Xi’an, China) to simulate the thermal environment of the actual gas turbine in service with the aim of assessing the thermal shock resistance capabilities of the TBCs. In the burner rig test, the ceramic sides of the TBCs were subjected to heating via a propane/oxygen flame, while the opposite sides of the sample were cooled through the use of compressed air. The manipulation of the propane–oxygen ratio and compressed air flow rate allows for the regulation of temperature conditions. The surface temperature of the TBCs in this study was controlled at 1400 ± 50 °C, while the back temperature was maintained at 950 ± 50 °C. The surface and back temperatures mentioned above refer to the measured values when the sample reaches a stable state after heating. Specifically, in a thermal cycle test, a thermal cycle includes three stages. During the heating process, the sample surface is heated to a predetermined temperature of 1400 ± 50 °C (approximately 60 s). During the holding process, the surface temperature of the sample is maintained at a temperature of 1400 ± 50 °C (approximately 120 s). During the cooling process, the surface temperature of the sample is cooled using compressed air from 1400 ± 50 °C to approximately 100 °C (approximately 60 s). At least three samples were selected from each angle for evaluation to ensure experimental accuracy. When the surface peeling of the coating exceeded 20%, TBC failure was determined, and the thermal shock life is the number of thermal cycles carried out until coating failure. Afterwards, the failed sample was divided, and the alterations in fracture morphology and cross-sectional microstructure were analyzed using a SEM.

3. Results and Discussion

3.1. The Cross-Sectional Microstructure of the Coating

The cross-sectional microstructure of EMAP coatings at various spraying angles are depicted in Figure 4, and the porosity of the coating at different spraying angles is shown in Figure 5. It can be seen that when the spraying angle is 90°, the coating contains a large number of PEPC structures distributed in a dense matrix (Figure 4a). At this spraying angle, the coating has a relatively high porosity of 11.9%. The distinguishing characteristic of the PEPC structure is the large area of granular morphology (Figure 4b). This characteristic is ascertained by the micron-agglomerated powder structure of Powder #2 [17]. Furthermore, as a result of the influence of powder impact during the spraying process, the PEPC structure mostly presents a flat shuttle shape [21]. At an angle of 70°, the number of PEPC structures in the coating is rapidly diminished (Figure 4c), and the porosity of the coating decreases to 10.6%. However, according to the research by Chen et al. [8], the porosity in the conventional YSZ coating sprayed by APS increases as the spraying angle decreases. The opposite trend in porosity between the conventional coatings and the EMAP coatings may be attributed to the use of two types of powders during the spraying process for the EMAP coatings. Specifically, the first phase matrix utilized in this process is comparable to that of the conventional YSZ coating, which means that the porosity in the first phase matrix should be increased with the decrease in the spraying angle. Meanwhile, the PEPC structure is also an important source of pores in EMAP coatings, and a smaller number of PEPC structures can lead to a decrease in the coating porosity [21]. Therefore, it was speculated that the decrease in the number of PEPC structures at this angle surpassed the increase in the pores of the first phase matrix, resulting in a decline in the porosity of the coating. A similar phenomenon was also observed while spraying at an angle of 50°. The cross-sectional morphology of the coating resembled that at 70°, with a minor decrease in the PEPC structure (Figure 4e), leading to a further reduction in the porosity of the coating to 9.5%. At an angle of 30°, The porosity of the coating suddenly rose to 12.3%. In addition to holes and cracks, many pores with granular morphology are generated in the coating (Figure 4g). It can be seen that these granular pores are formed by the aggregation of a large number of small particles in a granular morphology, which is similar to the morphology of the PEPC structure, but with smaller areas and more irregular shapes (Figure 4h). These granular pores may originate from Powder #1 or Powder #2. Nevertheless, Tillmann et al. [9] and Houdková et al. [11] found that for HVOF-sprayed hard metal coatings, the rebounding and sliding of carbide particles at the surface of the coating is increased when smaller spraying angles are applied. This is analogous to the decrease in the number of PEPC structures in EMAP coatings at 70° and 50°, indicating that the number of second-phase embedded particles in the coating may be the smallest at 30°. In the study by Chen et al. [8], a small number of pores with granular morphology can be observed in the cross-sectional morphology of the coating when sprayed at a 60° angle. Thus, it can be reasonably speculated that these pores with granular morphology do not come from the second-phase embedded particle, but rather from the first phase matrix. That is, when the spraying angle is 30°, the increase in pores generated by the first phase matrix surpasses the decrease in the PEPC structure, resulting in an increase in pores in the coating. In summary, it can be speculated that as the spraying angle decreases, the number of PEPC structures in the coating decreases, leading to a decrease in the porosity of the coating. Simultaneously, the increase in defects in the first phase matrix, including granular pores, leads to an increase in the coating porosity. The combined effect of both factors results in a trend where the porosity of the coating initially decreases and then increases as the angle decreases. More research is required to confirm this assumption.

In order to conduct a more objective quantitative analysis of the changes in the PEPC structure and granular pores, digital image processing technology was used to exclude holes and cracks in the coating, and granular areas in the coating (including the PEPC structure and granular pores) were selected. Figure 6 shows the results of image processing on the coating at a spraying angle of 90°. Figure 6a depicts the original image of the coating’s cross-sectional structure, while Figure 6b represents the modified image. The distribution of two distinct categories of pore areas and aspect ratios at various angles, together with their respective average values for the area and aspect ratio, is shown in Figure 7. The peak coordinate of the pore area fitting curve is the greatest at a spraying angle of 90°, which corresponds to an average pore area of 297 μm². The peak coordinate shifts towards the left as the angle decreases, and the peak increases while the average area decreases (Figure 7a,c). The peak coordinate of the pore area fitting curve is the smallest at a spraying angle of 30°, which corresponds to an average pore area of 183 μm². This indicates that as the angle decreases, the pores in the coating change from large PEPC structures to small granular pores. Similarly, as the angle decreases, the peak coordinates of the pore aspect ratio fitting curve also shift to the left, and the peak increases while the average aspect ratio decreases (Figure 7b,d), indicating that as the angle decreases, the pore shape in the coating changes from a flat shuttle-shaped PEPC structure to irregular granular pores. The outcomes of the image analysis support the validity of the previous inference about different sources of PEPC structures and granular pores. However, the granular pores and PEPC structures are determined by Powder #1 and Powder #2. In order to further understand how the spray angle affects the coating structure, it is necessary to obtain the effect of the spray angle on the two powders. Therefore, this paper will separately conduct single particle spraying experiments on the two types of powders.

3.2. Single Particle Spray Results of Two Types of Powders

The morphology and relative quantities of the second phase powder sprayed onto a mirror substrate at various spraying angles are illustrated in Figure 8 and Figure 9, respectively. The results demonstrate that at a spraying angle of 90°, the substrate has the highest amount of second phase powder embedded (Figure 8a), with a maximum relative quantity of 10.8%. This phenomenon may be responsible for most of the PEPC structures and the greater porosity in the coating when sprayed at a spraying angle of 90°. Additionally, the powder exhibits a uniformly dispersed morphology towards the surrounding areas (Figure 8b), which aligns with the PEPC structure observed in the coating. When the angle reduces to 70° and 50°, the quantity of second phase powders embedded decreases, with relative amounts of 7.8% and 7.3%, respectively. This decrease corresponds to a reduction in the PEPC structure in the coating cross-section at both angles. Meanwhile, the difference in powder quantity between the two angles is minimal, which may explain the similar coating structures observed at 70° and 50°. In addition, the powder has a slight extension of powder spreading towards the direction of the spraying angle. The substrate exhibits marks of powder particles sliding and rebounding, with certain particles displaying reduced volumes and irregular morphologies (Figure 8c–f). There is a speculation that the observed phenomenon might be attributed to the heightened tangential velocity component of the second phase powder when spraying at lower angles, which leads to a significantly greater tendency towards powder fragmentation and rebound. The research conducted by Fefekos et al. [22] also demonstrated that ceramic powders exhibit a higher susceptibility to fracture and rebound when sprayed at lower angles. At an angle of 30°, there is a notable reduction in the amount of second phase powder present on the substrate (Figure 8g), with a relative quantity of only 2.2%. The embedded second phase powder is primarily composed of remnants from the rebounding process (Figure 8h), demonstrating that the rebound and sliding of the second phase powder are the most apparent at this angle. Despite the resemblance of these remnant powders to granular pores in terms of their morphology, the porosity of the coating cannot be enhanced due to the minuscule quantity of embedded powders. This indicates that these remnants cannot serve as the main sources of granular pores in the coating at 30°, which more directly supports the validity of the prior inference.

Figure 10 depicts the spreading morphology of the first phase powder at various spraying angles. The first phase of molten droplet spreading shifts in the direction of the angle as the angle lowers, lengthening and fragmenting the particle spreading. The formation of the coating involves the stacking and spreading of numerous powders. Therefore, the morphology of powder spreading plays a crucial role in determining the structure and performance of the coating. In the first phase matrix (traditional YSZ coating), alterations in the spreading morphology of the first phase powder at low angles can cause more layer interfaces in the coating. This leads to an increase in the porosity of the coating, which, in turn, results in lower thermal conductivity and elastic modulus in the coating [8]. Simultaneously, the dispersion of droplets in the angular direction will cause a reduction in the thickness of spreading [15]. This, coupled with the rebound of particles at shallow angles, ultimately leads to a decline in the efficiency of coating deposition, as illustrated in Figure 11. The previous section demonstrated that the likelihood of particle-shaped pores originating from the second phase powder is extremely minimal at a spraying angle of 30°. In the meantime, the fractured edge of the molten first phase powder exhibits a degree of resemblance to the particle-shaped pores. Hence, it is speculated that the fractured edge of the molten droplet may be one of the factors for the augmentation of particle-shaped pores and coating porosity at low angles.

3.3. Coating Performance Characterization

Figure 12 displays the hardness and elastic modulus of coatings at various spraying angles. At a spraying angle of 90°, the coating achieves a maximum hardness of 833 HV_0.3 and a maximum elastic modulus of 103 GPa. As the angle decreases, the coating exhibits a decrease in both hardness and elastic modulus. At a spraying angle of 30°, the coating exhibits a minimum hardness of 689 HV_0.3 and a minimum elastic modulus of 53 GPa. Prior studies [21] have shown that coatings containing a higher number of PEPC structures demonstrate higher porosity and reduced hardness and elastic modulus. Based on this conclusion, the coating should show lower hardness and elastic modulus at a 90° angle. This is incongruous with the measurement results. Nevertheless, conventional thermal spray coatings demonstrate the same trend as the measured results at low angles [8,12]. This is because, when using the indentation method to measure the hardness and elastic modulus, the landing point is required to be a dense area far away from pore defects. These dense areas are mainly influenced by the first phase matrix, which is the same as conventional coatings. Additionally, coatings that possess lower hardness and lower elastic modulus generally demonstrate superior thermal shock resistance. However, in the case of EMAP coatings, the PEPC structure within the coating will have a greater influence on the thermal shock life. Therefore, it is essential to take into account the actual thermal shock life by considering both the first phase matrix and the second phase embedded particles.

3.4. Thermal Shock Resistance of Coatings

Figure 13 shows the thermal shock life of EMAP coatings under different spraying angles. Figure 13 demonstrates that the coating achieves its highest average thermal shock life of 40.6 cycles at a spraying angle of 90°. The difference between the best and worst samples is only one cycle, indicating excellent thermal shock life stability. At a spraying angle of 70°, the average thermal shock life decreased to 34 cycles, and the difference between the best and worst samples was 21 cycles. This led to a notable decrease in the stability of the thermal shock life. At a spraying angle of 50°, the average thermal shock life decreased to 32 cycles, with a notable difference of 26 cycles between the best and worst samples. This further diminished the stability of the thermal shock life. At a spraying angle of 30°, the average thermal shock life increased to 34.3 cycles, and there was a 15-cycle difference between the best and worst samples. The thermal shock life stability also experienced an increase. It can be seen that there exists a similar trend between alterations in the average thermal shock performance of coatings and variations in the porosity of the coatings.

Figure 14 shows the typical failure process of EMAP coatings from different angles. When the spraying angle is 90°, layered peeling occurs at the center of the coating as the number of thermal cycles increases. Then, as the number of thermal cycles increases, the peeling area continues to expand. Overall peeling happens as the coating becomes close to the thermal shock life. After the onset of overall peeling, there is a notable acceleration in the rate of coating peeling, leading to a rapid attainment of the failure state. This phenomenon occurs in the thermal shock process of all 90° specimens. The specific reason for coating delamination is attributed to the formation and propagation of cracks along the interface between the coating matrix and the embedded particle area, as a result of thermal stress induced during the thermal shock cycle [23]. Similar phenomena have also occurred in our previous studies [19], which belong to the normal failure process of EMAP coatings. In fact, the same process occurs at spraying angles of 70° and 50°, and when the layered peeling failure process takes place, the thermal shock life of the coating can essentially reach about 40 cycles, which is comparable to the thermal shock life of the coating at spraying angles of 90°. In contrast to the coating observed at a spraying angle of 90°, the coating at spraying angles of 70° and 50° displays a distinct failure process. Specifically, the overall peeling of these coatings occurred at a lower number of cycles without layered peeling and rapid reach failure. Currently, the thermal shock life of the coating is typically reduced to approximately 20 cycles, representing only 50% of its lifetime at 90°. The outcome of this phenomenon is a reduced average thermal shock life and suboptimal stability in the coatings at 70° and 50° in comparison to those at 90°. It is notable that this effect is more pronounced at 50° than at 70°. At an angle of 30°, the coating will simultaneously exhibit layered peeling at the centre and overall peeling at the edge. The thermal shock life of a coating is limited to approximately 26 cycles when the failure manifests as edge peeling. The thermal shock life of a coating can increase significantly, up to approximately 36 cycles, when the failure of the coating occurs in the form of layered peeling. Compared to the 90° angle, the thermal shock life of the layered peeling failure process of the 30° coating is relatively diminished.

Figure 15 shows the fracture morphology and cross-sectional microstructure of failed coatings at different spraying angles. At a spraying angle of 90°, the coating experiences a failure process characterized by layered peeling, a stepped fracture edge, and cracks located at the top part of the coating. These phenomena can be attributed to the layered peeling of the coating. The PEPC structure in the coating experiences partial sintering while still preserving some of its initial structure, accompanied by the presence of a small number of transverse cracks within the coating. At spraying angles of 70° and 50°, the coatings exhibit the overall peeling failure process in the center. The fracture surface of the coating with an angle of 70° exhibits a perpendicular form, while the fracture surface of the coating with an angle of 50° displays a sloping form. The phenomenon of layered peeling does not result in the formation of any cracks at the top part of the coating. Furthermore, the slope of the 50° coating is solely attributable to the remaining portion subsequent to overall peeling. In contrast to the 90° coating section, the PEPC structure within the coating section is relatively small, resulting in a higher coating density. Consequently, the coating is more prone to the formation of transverse cracks, which propagate along the holes. At a spraying angle of 30°, the coating experiences a failure process characterized by the overall peeling of its edges. The fracture surface exhibits a perpendicular form at the edge of the coating, and there exist interface cracks that traverse the coating between the bonding layer and the ceramic layer. The coating exhibits a significant quantity of voids, with the highest number of transverse cracks. The propagation of transverse cracks leads to their connection across a multitude of voids, resulting in the formation of larger transverse cracks.

The interface morphology of the coatings reveals the absence of a discernible thermally grown oxide (TGO) layer in all four TBCs. This result suggests that TGO is not a contributing factor to the failure of the coatings. Park et al. [24] obtained comparable outcomes, primarily attributed to controlling the temperature of the sample substrate below 1050 °C subsequent to cooling it with compressed air during the experiment. The observed amount of severe oxidation of the bonding layer was prolonged at the given temperature. During the thermal shock evaluation, it was seen that if the surface temperature of the sample of TBC exceeds 1300 °C, there is a possibility of phase transformation occurring in the YSZ coating. Va ßen et al. [25] discovered that, when subjected to more rigorous evaluation conditions, the YSZ TBC did not exhibit a phase transition subsequent to the flame impact assessment. The primary cause of this phenomenon is attributed to the quick cooling rate experienced by the coating during the thermal shock process, which effectively inhibited the monoclinic phase transition. Therefore, it can be posited that the coating did not undergo any phase transition while undergoing the thermal shock process. Hence, it can be inferred that the failure life of coatings is significantly influenced by the microstructure of the coating from different angles.

Thermal stress, which results from a mismatch in the coefficient of thermal expansion between the ceramic coating and bonding coating, controls the majority of crack initiation and evolution in TBCs during the thermal shock cycle [26,27,28]. Previous research has indicated that the second phase particles embedded in the EMAP coating can improve the strain tolerance and reduce thermal stress, and it has been observed that the thermal shock life of the coating is positively correlated with the quantity of unmelted embedded particles [17,18,19]. In terms of the coating structure, it has been observed that the presence of embedded particles in the coating is the most pronounced when the spraying angle is 90°. This results in the formation of a significant number of PEPC structures, which effectively mitigate thermal stress and impede the formation and propagation of cracks within the coating. Hence, the optimal average thermal shock life and stability of thermal shock life within the coating at an angle of 90° are observed. A reduction in the second phase embedded particles within the coating can be observed when the spraying angles are 70° and 50°. This leads to a denser coating, decreased strain tolerance, heightened thermal stress, and ultimately, overall peeling failure of the coating. Consequently, a reduction in the mean thermal shock life and stability of coatings is observed. At a spraying angle of 30°, the coating exhibited a slight increase in the average thermal shock life and improved its stability, which can be attributed to the abundant small pores generated by the first phase matrix. These pores improve the mechanical properties of the coating to a certain extent, reducing thermal stress. In contrast to the PEPC structures, the granular pores produced by the first phase of the matrix exhibit an inclination to sinter, leading to the formation of voids and facilitating the emergence and spread of transverse fractures. The propagation and interconnection of transverse cracks through the holes lead to the formation of larger cracks, thereby reducing the thermal shock life and stability of the coating in comparison to the coating sprayed at an angle of 90°.

4. Summary

The present study investigated the impact of various spraying angles on the microstructure of EMAP coatings prepared by APS. The investigation focused on analysing the alterations in relative deposition efficiency, porosity, hardness, and elastic modulus of coatings from different angles. Then, the effects of varying angles on the spreading and embedding of two types of powder were examined. On the other hand, the present study also investigated the alterations in thermal shock resistance exhibited by coatings from varying angles. The primary results indicate the following:

(1)

With a decrease in the angle, there is a corresponding reduction in the PEPC structure within the coating, leading to an increase in the number of granular pores in the first phase matrix. In comparison to the PEPC structure, the granular pores in the first phase matrix exhibit reduced area and aspect ratio.

(2)

The reduction in the spraying angle is associated with decreases in the deposition efficiency, hardness, and elastic modulus of the coating, which are consistent with the findings of conventional YSZ coatings. The porosity of the coating is observed to decrease as the spraying angle ranges from 90° to 50°. Conversely, an increase in porosity is noted when the spraying angle is 30°. This phenomenon can be attributed to a decrease in the angle, resulting in a reduction in the number of embedded particles within the coating and a related rise in the number of pores of the first phase matrix. The former resulted in a reduction in porosity, whereas the latter resulted in an increase in porosity. The coupling between the two phase powders results in a porosity trend in the coating that initially decreases and subsequently increases.

(3)

As the angle decreases, the spread of the first phase molten droplets shifts in the direction of the angle, leading to elongation and fragmentation in the particle spread. This phenomenon results in an elevation in granular pores within the coating. The second phase particles exhibit a tendency to rebound and slip when sprayed onto the substrate at low spraying angles due to their unmelted states, leading to a reduction in the quantity of embedded particles and a decline in the PEPC structure within the coating.

(4)

A decrease in the spraying angle from 90° to 50° results in a reduction in both the average thermal shock resistance and the stability of the thermal shock life of the coating. At a spraying angle of 30°, the coating exhibits a slight increase in the thermal shock life and improved stability. The reduction in the spraying angle from 90° to 50° leads to a decrease in the quantity of particles embedded in the coating, resulting in two types of failure: layered peeling and overall peeling. The thermal shock life disparity between the two forms is approximately twofold, resulting in coating instability. At a spraying angle of 30°, the first phase of the matrix of the coating generates additional pores that mitigate the thermal stress between the ceramic layer and bonding layer. This leads to a marginal improvement in the thermal shock life and stability of the coating. Nevertheless, these pores are susceptible to sintering, which facilitates the formation and propagation of cracks, resulting in a coating lifetime below 90°.