Ablation Resistance of ZrB₂-SiC-Si Composite Coatings Applied on Graphite Substrate by Spark Plasma Sintering

Abstract

In this research, a protective composite coating made of ultra-high-temperature ceramics (UHTCs) was applied on the graphite substrate to boost the ablation resistance of the graphite substrate. Spark plasma sintering (SPS) was used to create such coatings on the graphite substrates. Efforts were made to identify the appropriate chemical composition for the composite coating and the conditions for the SPS (temperature, pressure and holding time). The coatings with a base composition of ZrB₂-15 vol.% Si-15 vol.% SiC as well as MoSi₂ and WC additives in the same amounts of 0 vol.%, 2.5 vol.% and 5 vol.% of each were applied on the graphite substrates under sintering conditions of final temperature of 1875 ± 25°C, initial pressure of 10 MPa, final load of 25 MPa and holding time of 5 min. The results obtained from the ablation test by oxyacetylene flame verified that the presence of the protective composite coating significantly increases the ablation performance of the graphite substrate. The sample without WC and MoSi₂ additives had the lowest mass reduction in longer ablation times, which indicates the high ablation resistance of the base sample compared to the samples with those additives.

Introduction

Graphite as a member of the carbon materials family is an attractive material for high-temperature applications like electrical contacts, heaters, nozzles, high-temperature heat exchangers, supersonic spacecraft and space shuttle noses, and aircraft wing leading edges because of its high modulus, high strength, light weight and superior resistance to thermal shock.¹ However, the high oxidation rate at temperature > 400°C and poor ablation resistance of graphite significantly reduce its efficiency.^2,3 One of the most effective solutions to boost the oxidation and ablation resistances of this material is to use ultra-high-temperature ceramics (UHTCs) as a coating.^4,5,6,7,8,9 Recently, researchers have focused more on UHTCs for use in harsh environments, including carbides and borides of transitions metals, such as HfB₂, ZrB₂, HfC and ZrC,^{10,11,12,13,14,15} because of their prominent combination of properties, including good chemical stability and high melting points.^{16,17,18,19,20,21} Additionally, various methods were reported to apply UHTC coatings on the surfaces of carbonaceous substrates including pack cementation,^22,23 plasma spray,²⁴ slurry technique,^25,26 vapor silicon infiltration²⁷ and spark plasma sintering.^4,5,6,28

Akbarpour et al.⁴ applied ZrB₂-SiC-HfB₂-WC coating on the graphite by SPS method to improve the ablation behavior. In fact, the effects of the addition of WC and HfB₂ on ablation resistance of ZrB₂-SiC-based coating were investigated. The ablation resistance was significantly improved with 2.5 vol.% HfB₂ and 5 vol.% WC. The formation of glassy silicate phase in the coating surface helped in increasing the oxidation response. In addition, the main mechanisms of improving the ablation behavior of such coatings were oxidation of SiC and WC and formation of WO₃ and SiO as products of the oxidation process and, finally, evaporation of such gases simultaneous with heating. Another study by the same group⁵ found that a sample with 3.75 vol.% of each additive (HfB₂ and WC) showed a better performance than the other samples. Evaporation of gas byproducts over the ablation test led to the use of flame energy, which contributes to the longevity of the coating main structure. Moreover, high thermal conductivity of the coating composition due to existence of HfB₂ in the coating caused reduced flame energy, decreased temperature of the surface and enhanced ablation resistance.

Aliasgarian et al.⁶ applied double-layer coating on graphite substrate through pack cementation and shrouded plasma spray methods. In that study, SiC as the inner layer of coating was applied on graphite by pack cementation. Additionally, the ZrB₂-SiC composite as the outer layer was applied via SPS route. The Ablation properties of the coating at various times (15 s, 30 s, 120 s, and 360 s) and the ablation mechanisms were investigated. The weight of the samples after 15, 30 and 120 s exposure to oxyacetylene flame were enhanced because of synthesis of oxides. The zirconia layer as one of the oxidation products of ZrB₂-SiC coating acted like a thermal barrier, which prohibited graphite substrate from oxidative damage. Also, with increasing the ablation duration to 360 s, the inner layer of SiC oxidized slowly, and the gas product (SiO) of this phenomenon evaporated because of the creation of a gap between graphite substrate and ZrO₂ layer. Therefore, during the cooling from high to ambient temperature, phase transformation of ZrO₂ and its volume expansion caused the generation of cracks in the central region.

Jyoti et al.²⁹ manufactured ZrB₂-SiC composites through in situ reduction of ZrO₂ with B₄C, Si and graphite (c) by hot-pressing at 1500°C. Thermodynamic calculations confirmed the possibility of reactions. The in situ formation of homogeneous SiC particles prevented the growth of ZrB₂ grains and improved densification. Due to the formation of layers of SiO₂ and ZrO₂/SiO₂ on the surface of oxidized samples, SiC particles provided better dimensional and thermal stability during ablation at 2600°C. Dense layer of SiO₂ protected the composite because of its low oxygen permeability and the development of a SiC-depleted zone.

Wang et al.⁷ applied SiC/ZrB₂-SiC/ZrB₂/SiC multilayer coating on graphite through chemical vapor deposition and pack cementation to boost ablation resistance of graphite. They investigated the microstructural evolution and ablation behavior of multilayer coatings. The SiC/ZrB₂-SiC/ZrB₂/SiC coatings illustrated superior ablation resistance; hence, after exposure to oxyacetylene flame for 298 s, just the center of the SiC/ZrB₂-SiC/ZrB₂/SiC coated specimen was ablated. The mass and linear ablation rates under oxyacetylene exposure for 298 s were 0.27 mg/s and 0.57 μm/s, respectively. The significant increase in the ablation resistance of SiC/ZrB₂-SiC/ZrB₂/SiC coatings was attributed to the chemical vapor deposition process and proper ablation response of formed ZrO₂ phase during the oxidation of ZrB₂ layers.

Mu et al.³⁰ applied ZrB₂-SiC and ZrB₂-SiC-LaSi₂ coatings on the surface of carbon-carbon composites by atmospheric plasma spraying (APS) to improve the ablation resistance at temperatures > 1800°C. Addition of LaSi₂ increased the densification and resistance to ablation of ZrB₂-SiC coatings. In fact, the presence of LaSi₂ in the coating composition produced a more protective Si-O-La sealant phase and formed compounds with a high melting temperature that stabilized the sealant phase. Li et al.³¹ investigated the effect of Al₂O₃ as a catalyst on the ablation behavior of ZrB₂-SiC-based coatings. For this purpose, SiC/ZrB₂-SiC-Si (ZSS) and SiC/ZrB₂-SiC-Al₂O₃ (ZSA) coatings were prepared on the surface of carbon-carbon composites with a combination of pack cementation and APS methods. The SiC/ZSA coating had better ablation resistance because the phase transformations and grain growth of ZrO₂ were inhibited by Al₂O₃.

Wei et al.³² manufactured laminated graphite/ZrB₂-SiC ceramics through tape casting and hot-press sintering. The ablation behavior of the ceramics in two various directions was investigated. The results illustrated that the ablation behavior in two directions were different as the mass and linear ablation rates were 8.1 mg/s and 3.1 μm/s in parallel direction and 0.2 mg/s and 1.2 μm/s in perpendicular direction, respectively. The thermal conductivity of laminated graphite/ZrB₂-SiC composites in perpendicular direction (121 W/mK) was greater than that in parallel one (78 W/mK), which resulted in a less surface temperature compared to parallel direction at the same ablation condition. A higher thermal conductivity of perpendicular direction led to more heat transfer from the central part to the edges due to a less surface temperature of perpendicular one. Additionally, higher thermal conductivity boosted the thermal shock resistance and improved thermal stability at high temperatures. Therefore, in perpendicular direction, the ablation response of the laminated graphite/ZrB₂-SiC samples was better.

In this research, an appropriate composite coating of ultra-high temperature ceramics is applied on the graphite substrate using spark plasma sintering. The aim of this study is to investigate the impact of adding MoSi₂ and WC to the base composition of ZrB₂-Si-SiC coating to boost the resistance to ablation of the graphite substrate.

Experimental Procedure

Materials and Process

Starting materials of SiC (particle size < 10 µm, purity ~ 99%), ZrB₂ (particle size < 3 µm, purity ~ 99.9%) and MoSi₂ powders (particle size < 10 µm, purity ~ 99.5%) were supplied by Hongwu International Group. Moreover, WC (particle size < 15 µm, purity ~ 99%) and Si powders (particle size < 10 µm, purity ~ 99%) were supplied by Almaseh Saz Company. The raw material powders of coatings were weighed based on the compositions presented in the supplemental information (Tables S1, S2). To calculate the amounts of starting powders used, first, the density of the final coating applied to each sample was supposed to be 100% and the thickness of the final coating assumed 700 μm. It is necessary to mention that the volume ratios of ZrB₂/(Si + SiC) and WC/MoSi₂ are constant and approximately equal to 2.33 and 1, respectively.

The ZrB₂ and SiC particles were separately dispersed for 1 h in ethanol employing an ultrasonic stirrer. Then, the other coating powders (Si, MoSi₂ and WC) were introduced to the mixtures with preferable composition (Table S1), and the slurries were mixed ultrasonically for an additional 30 min. Notably, the reason for using WC in the composition is that this material has an anti-ablation feature and somehow sacrifices itself so that the main composite material does not suffer from ablation. The cause for adding MoSi₂ in the mixture is because this substance forms a layer of SiO₂ when exposed to oxygen at high temperature, which protects the lower parts and prevents oxygen diffusion. The main purpose of Si addition is to improve the bond between the graphite substrate and the composite coating. The detailed explanations are reported elsewhere.³³ In the next step, the slurries were dried on a rotating drier at 130°C for 2 h and in an oven for 12 h in 110°C. Subsequently, the dried powder mixtures were milled and loaded on HT-600 graphite pellets (diameter: 29 mm; thickness: 6 mm), which were placed in the graphite dies. The characteristics of the graphite substrate used in this research are given in the supplemental information (Table S3).

The sintering step was performed in a spark plasma sintering furnace (SPS-20T-10) according to these conditions: vacuum in the range of 30–100 Pa, current rate of 0.15–0.2 KA/min, final temperature of 1875 ± 25°C, holding time at final temperature of 5 min, initial pressure of 10 MPa, and final pressure of 25 MPa. The temperature of graphite dies during SPS process was checked by an infrared thermometer. Finally, after the SPS chamber had cooled down, the samples were extracted from the dies.

Characterization

After removing the samples from die, samples were ground and then polished for elimination of the graphite layers from the surface. The thickness of the coating after sintering and polishing in ZSS and ZSS-2.5WM samples was almost the same (676–674 µm). Notably, in ZSS-5WM sample, the thickness was greater than the desired value. The microstructural investigations of the coatings were conducted by a Tescan Mira3 field emission scanning electron microscopy (FESEM). The chemical compositions of coatings were examined using a DXP-X10P energy-dispersive x-ray spectroscopy (EDS). Phase investigations of the coatings after ablation tests were carried out via a Philips PW1730 x-ray diffractometer (XRD). For evaluating the ablation properties of the coatings, an oxyacetylene flame including acetylene (purity ~ 98%) and oxygen (purity ~ 99.2) with pressures of 8 and 6 bar, respectively, was used. Nozzle diameter, nozzle distance to the sample surface, flame temperature, surface temperature and test times were 2 mm, 10 mm, 2000°C, 1800°C, and 30 s, 60 s, 90 s, 150 s and 210 s, respectively. Two laser pyrometers were used to measure the temperatures of the flame and the coating surface. The flame temperature was adjusted to the desired value (2000°C) by changing the gas level (acetylene and oxygen). When the temperature of the central part of the coating surface reached 1800°C, the time was measured using a chronometer. To keep the flame temperature constant at 2000°C, the gas pressure was adjusted. Considering that the flame hits the central part of the samples in a concentrated manner, the temperature of the center was higher (about 300°C) than the sides at the beginning of the test, but after a while, the temperature difference between the center and the side parts dropped < 50°C. Ablation weight loss (ΔW%) of the coatings was calculated using the following formula:

$$ \Delta W = \frac{{m_{0} - m_{1} }}{{m_{0} }} $$

(1)

where $m_{0}$ and $m_{1}$ are the sample weights before and after the ablation test, respectively.

Result and Discussion

To study the ablation resistance of the composite coatings applied on the graphite, the surfaces of the graphite and the samples with composite coatings were subjected to oxyacetylene flame at a temperature of 2000°C for different dwell times. The mass reduction percentage of each sample is calculated according to Eq. 1, and the results are reported in Table S4 and graphed in Fig. 1. The amount of mass reduction in samples with protective composite coating (graphite substrates with protective coating) is much less than the sample without coating. Therefore, the resistance to ablation of graphite substrate with protective coating is higher than graphite without protective coating.

Fig. 1

Mass loss in samples with composite coatings.

Figure 2 shows the macroscopic images of the coating surfaces of ZSS, ZSS-2.5WM and ZSS-5WM specimens before and after the ablation test. The variations caused by the oxyacetylene flame on the surfaces of the composite coatings of the samples after the ablation test for 90 s and 210 s are shown in this figure. In the ZSS and ZSS-5WM samples, as a result of the oxyacetylene flame hitting the composite coating surface, the flame spread on the composite coating surface, but in the ZSS-2.5WM sample, it was concentrated in one part. Also, melting and detachment can be observed in the central part of the composite coating of the ZSS-2.5WM specimen. As the ablation time increases, the amount of flame destruction intensifies, and this causes the traces of melting, detachment, cracking and fracture on the surface of the composite coating to increase.

Fig. 2

Macroscopic images of the surfaces of composite coatings applied on HT-600 graphite substrate before and after flame tests: (a) ZSS, (b) ZSS-2.5WM and (c) ZSS-5WM.

The results of the phase examination of the central part of the composite coatings of ZSS, ZSS-2.5WM and ZSS-5WM specimens after the ablation test with oxyacetylene flame for 210 s are presented in Fig. 3. Notably, the x-ray diffraction spectra of the specimens after applying the coating by SPS method and before the ablation test were already reported in Ref. 33. As can be seen in the x-ray diffraction spectrum (Fig. 3), in the ZSS sample, which contained primary phases of ZrB₂ and SiC before the flame test,³³ all ZrB₂ has been converted into ZrO₂ because of oxidation, but still some primary SiC phase is present in the structure. Considering that due to the high heat of the flame, the probability of oxidation of SiC and its conversion to SiO₂ is high, the SiO₂ is formed as a layer on the surface, and its non-identification by XRD indicates its evaporation and removal during analysis. Therefore, it can be concluded that the presence of SiC and its detection by XRD analysis is related to the lower parts of the composite coating, not the upper part that has been subjected to direct flame and has undergone erosion.

Fig. 3

X-ray diffraction spectra from the center of the specimens after ablation for 210 s.

The phases of ZrO₂, ZrB₂ and SiC were identified in the ZSS-2.5WM sample, which had WC and MoSi₂ additives (2.5 vol.% of each) in addition to the raw materials of the main composition (ZrB₂, SiC and Si). Notably, here some initial ZrB₂ phase remains, which may be due to the presence of sacrificial phases of WB and MoB.³³ These phases are formed during the SPS process due to the presence of raw materials such as MoSi₂ and WC, which during the ablation test use the flame energy for their oxidation and, by increasing the heat transfer rate, spread the heat caused by the flame throughout the coating. In this way, they caused a loss in the heat of the flame. As observed in the x-ray diffraction spectrum of the ZSS-5WM sample (Fig. 3), due to the heat of the flame, all the primary phases in the composition of the composite coating (after applying the composite coating and before the ablation test³³) have undergone oxidation and ZrO₂ phases and small amounts of SiC have been identified. The disappearance of the primary phases in the composition of the composite coating applied by SPS and the formation of the ZrO₂ oxide phase after exposure to the oxyacetylene flame (Fig. 3) is due to a series of chemical oxidation reactions.^34,35

To investigate the effects caused by the heat of oxyacetylene flame during the ablation test on the microstructure of composite coatings, a scanning electron microscope was used. Figure 4 displays the SEM images of the ZSS composite coating surface exposed to oxyacetylene flame for 210 s. As the figures show, the amount of destruction caused by the heat of the oxyacetylene flame in the central part is more than other parts and this part has undergone the most change in the initial microstructure (before the flame test). In addition to melting and changing the primary microstructure, the central part has many holes on its surface, which is caused by the release of gaseous products resulting from the oxidation of the raw materials of the composite coating. Based on the SEM images, the destruction rate of the middle part (between the center and the edge) is less than the central part; however, it is clear that an amorphous (glass-like) phase layer is synthesized on the surface of the composite coating due to melting. Also, many holes have been created because of the release of gases from oxidation. Notably, the formation of this glass-like phase acts as a barrier against further erosion and destruction of the lower parts.

Fig. 4

SEM images of the ZSS composite coating surfaces after the ablation test: (a) central part and (b) middle part (between the center and the edge).

EDS elemental map analysis was used to more closely examine the degree of destruction and the changes made in the distribution of elements in the central part of the ZSS composite coating surface. Figure 5a displays the SEM image of the central part of the ZSS composite coating that was exposed to oxyacetylene flame for 210 s. From examining the results of elemental map analysis (Fig. 5b, c, d, e, and f) and the results obtained from x-ray diffraction test (Fig. 3), the presence of ZrO₂ phase in the central part can be confirmed. The light-colored parts in Fig. 5a are related to this phase. On the other hand, the dark-colored parts are seen slightly lower than the other parts. Based on the results of the elemental map analysis, they are related to the phases containing elements Si, C and O (SiC and SiO₂). According to the distribution of Si and C in the images obtained from elemental map analysis in Fig. 5, as well as the results achieved from XRD, there are still some amounts of the initial phase of SiC in the central part. The reason for this observation is that over the ablation process, the phases containing Si in the upper parts of the surface have gradually undergone surface oxidation because of the high temperature of the flame and have turned into oxide phases (such as SiO₂); due to the low melting temperature and evaporation, these oxide phases have melted and finally evaporated and thus been removed from the upper parts. The identified SiC phase in XRD was related to the lower parts of the surface. Of course, according to the distribution of Si and O elements in Fig. 5, there are phases containing Si-O in the central part of the ZSS composite coating surface.

Fig. 5

Elemental EDS maps of the central part of the ZSS sample after the flame test: (a) SEM image, (b) Zr, (c) Si, (d) B, (e) O and (f) C.

The cross section of the central part of the ZSS sample, which had the highest amount of destruction, was subjected to EDS to examine the distribution of elements and phases after the ablation test. Figure 6 presents the outcomes of elemental map analysis of the cross section of the central part of the ZSS sample. According to the distribution of Zr, Si and O elements in some areas, there is a possibility of the formation of ZrO₂ and SiO₂ phases in those areas. Notably, the lower density of areas containing Zr and Si indicates severe destruction caused by oxidation and the formation of oxide phases. There is also the possibility of intact ZrB₂ and SiC phases under the oxidized layers.

Fig. 6

Elemental EDS maps of the central part of the cross section of ZSS sample: (a) SEM image, (b) Zr, (c) Si, (d) B, (e) O and (f) C.

From the results of the calculation of the reduction in the mass of the samples due to exposure to the oxyacetylene flame (Table S4) and SEM images, it appears that at the beginning of the oxidation/erosion process (up to 90 s), high temperature in the center of the specimen causes the transformation of ZrB₂ and SiC materials into ZrO₂ and SiO₂, respectively. The results of XRD analysis (Fig. 3) also show that the ZrO₂ phase is synthesized in the center of the eroded part. But in the XRD pattern, there is no indication of the presence of SiO₂, which is due to the rapid evaporation of this material at elevated temperatures. The presence of a significant number of holes on the surface of the composite coating is caused by the escape of B₂O₃, CO, CO₂ and SiO gases resulting from the oxidation of the starting materials of the composite coating during the ablation test. This event has reduced the density of the starting materials and created a porous microstructure up to a certain distance from the surface of the composite coating (Fig. 7). In this example, the composite coating has almost preserved its original structure and by creating a porous microstructure in the upper parts near the surface; this acted as an insulation and thermal barrier, which reduces the temperature of the lower parts of the composite coating and the substrate and increases resistance to ablation. The results of the mass reduction of the samples due to flame (Table S4) also prove the same thing, as the mass reduction in this sample is less than other ones.

Fig. 7

SEM images of composite coating of ZSS-2.5WM sample after ablation for 210 s: (a) surface and (b, c) destroyed central part (EDS results are taken from area C).

In Fig. 7, the SEM micrographs of the eroded surface of the ZSS-2.5WM specimen are presented. Figure 7b is related to the detached region in the central part of the sample coating. For this part, in addition to being detached, its lower part has also melted because of the high heat of the flame. After the completion of the ablation test and during high-speed cooling, the glassy oxide phase has formed. Figure 7c shows the white layer created in this part with a higher magnification. This part has many cracks formed during the cooling after the ablation test because of volume changes caused by phase transformations. Also, in this area, many holes are observed, which were created because of the release of gaseous products of oxidation/erosion reactions from the coating. To check the destruction in detail, EDS analysis was taken from the white layer created in the central part of this sample, based on which probably the white layer in Fig. 7c is related to the phases consisting of Zr elements such as ZrB₂, ZrO₂ or ZrC. On the other hand, according to the results obtained from the XRD analysis of the central part of this sample (Fig. 3), the presence of ZrO₂ phase in this part was confirmed.

The SEM micrographs of the middle part of the coating surface of the ZSS-2.5WM sample exposed to oxyacetylene flame for 210 s are shown in Fig. 8. The interesting thing is that a series of island phases have been created on the surface of the coating during cooling. To further examine these parts, images with higher magnification were captured, and EDS analysis was also taken to determine the chemical composition of different parts in the middle part of the coating surface. According to the obtained results, the bright island parts are related to phases rich in Zr, W, B, C and O elements. Also, the dark colored parts (area E) are attributed to phases rich in Si, Zr, O, C and W elements. As a result of the elevated oxygen concentration in this environment, there is a significant probability of the formation of oxide phases. Ouyang et al.³⁶ also observed the formation of dendritic structure including high-temperature oxides such as ZrO₂.

Fig. 8

SEM images of the middle part of the ZSS-2.5WM specimen coating surface after the ablation test for 210 s along with the EDS results from the specified places.

Figure 9 shows the SEM micrographs of the cross section of the ZSS-2.5WM specimen after the flame test. As can be seen, moving from the middle part to the center of the sample, the thickness of the coating decreases. The decrease in thickness in the middle part was not so severe and occurred almost uniformly (Fig. 9a), but the decrease in the thickness of the protective composite coating intensified in the central part, so that a strong thickness gradient was created in the center (Fig. 9b). The ZrB₂-SiC-AlN system has demonstrated significant potential when employed as a protective coating. In this case, ZrO₂ emerges as the outer thermal barrier layer, supported by the aluminum silicate scale, and viscosity can be adjusted by changing the Al:Si ratio.³⁷

Fig. 9

SEM images of the cross section of the ZSS-2.5WM specimen after the ablation test for 210 s: (a) middle and (b) the central parts.

As seen in the surface SEM images (Fig. 7a), severe detachment has occurred in the central part. The cause of this phenomenon can be expressed as that, simultaneously with the first stage of erosion, a liquid phase oxide layer is synthesized on the surface, which protects the lower parts from the flame and prevents their oxidation and erosion. With the increase of ablation duration and the expansion of erosion and destruction, the lower parts are gradually oxidized, and the gases resulting from this process find their way to the surface through the creation of holes and leave the system. With the expansion of erosion in the lower parts, the bottom of the oxide layer is gradually emptied and creates a gap between the oxide layer on the surface and the lower parts. After the completion of the ablation test and during cooling due to the temperature gradient and the stresses caused by the differences in the coefficients of thermal expansion of the surface oxide layer and the lower parts of the surface layer, flaking and detachment have occurred.

To industrially use ZrB₂/SiC plasma sprayed carbon-carbon substrates, Gao et al.³⁸ deposited a dense continuous SiC intermediate layer with a thickness of 150 μm on the substrate. That action led to the improvement of the wettability of the ZrB₂/SiC coating and created a strong interface between the SiC intermediate layer and the ZrB₂/SiC coating. The intermediate SiC layer significantly (up to 60% compared to the case without an intermediate layer) reduced the thermal stress between the substrate and the ZrB₂/SiC coating. After ablation for 1800 s, the SiO₂ layer was completely evaporated and the C/C substrate was exposed, which effectively lost its effective coverage. In other words, surface erosion of ZrB₂/SiC coating occurred first. With the passage of time, oxygen continued to diffuse and the intermediate layer of SiC participated in the reaction and gradually thinned. Finally, the SiC layer was nearly depleted. The intermediate SiC layer improved the erosion resistance because of the formation of a continuous SiO₂ layer during ablation, which had anti-permeability and self-sealing properties.

To accurately check the extent of ablation and the resulting destruction, EDS analysis was taken from the central part of the composite coating of the ZSS-2.5WM specimen, which has experienced detachment so that a proper interpretation can be made based on the intensity of the scattering of elements in different areas. Based on the results of elemental map analysis (Fig. 10), it was found that the intensity and type of distribution of W and Mo elements in different areas of the central part (the part with detachment and the higher part without detachment) is almost uniform. In addition, the intensity of Zr element in the region without detachment is higher than in the region with detachment, and the intensity of Si element in the region with detachment is higher than its upper part which is free of detachment. From these findings, it can be concluded that, first, an oxide layer of ZrO₂ and SiO₂ is formed on the entire surface, and as the erosion time increases and the resulting destruction spreads, part of the central part becomes detached because of the stresses applied during cooling. The Si element detected in the detached region is related to the formation of the low temperature oxide layer of SiO₂, which protects its lower regions. The reason why the phases containing W and Mo were not detected in the x-ray analysis (Fig. 3) is that the XRD test was taken from the center of the samples, where the phases containing W and Mo were eroded, eventually evaporated and left the system. Notably, in the EDS analysis (Fig. 10), a wider area of the surface has been analyzed; hence, W and Mo elements were revealed in the elemental maps.

Fig. 10

SEM/EDS analyses of the destroyed coating surface of ZSS-2.5WM sample after exposure to oxyacetylene flame for 210 s.

The SEM images of the central and middle parts of the ZSS-5WM composite coating surface after the flame test for 210 s are shown in Fig. 11. The upper region is attributed to the central part and the lower region belongs to the middle part. From the examination of these images, it appears that the central part of the surface of the composite coating has been severely damaged and changed in structure because of ablation. The middle part has undergone more melting than fundamental structural changes.

Fig. 11

SEM/EDS analyses of the central and middle part of ZSS-5WM sample after the ablation test for 210 s.

To further study the trend and severity of the occurred destruction, EDS analysis was taken from the central part and the interface of the central part/middle part to get a better understanding of the erosion by examining the changes in the distribution of elements in different parts of the surface. The results of elemental map analysis (Fig. 11) show that W and Mo elements are almost uniformly present in the entire analyzed surface. It is also observed that in the central part of the surface, the accumulation of Zr element is more intense, which indicates the formation of a layer of ZrO₂ in this part. According to the findings obtained from the x-ray diffraction test (Fig. 3), in the central part of the sample, the ZrO₂ phase is formed, which is a confirmation of the results obtained from the analysis of the elemental map taken from this part. Due to the heat of the oxyacetylene flame, in the central part of the surface of the composite coating, the phases resulting from the oxidation of Si (SiO₂ and SiO) are completely removed from this part because of their low melting and evaporation points and the high heat of the flame. For this reason, the amount of Si element in this part has dropped drastically. Despite this, in contrast to the oxide phase obtained from the element Zr (ZrO₂), it still resists the heat of the flame because of its higher refractoriness and is formed as a layer on the surface of this part as a result of surface melting, which protects its lower parts from extra erosion. Therefore, the concentration of Zr element is higher in this part.

In the interface and the middle part of the surface of the composite coating, because less heat has reached these parts and the temperature is lower than the central part, it only causes surface melting, molten oxide phases including Si and Zr are formed, and no evaporation takes place. Therefore, the presence of phases containing Si and Zr elements is evident in this part. Notably, by moving further away from the interface and getting closer to the middle part, the concentration of the Zr element decreases and the intensity of the accumulation of the Si element increases, which indicates the presence of the SiO₂ layer created by the heat of the oxyacetylene flame is in this part.

Figure 12 shows the SEM micrographs of the cross section of the ZSS-5WM specimen after the flame test for 210 s. Figure 12a is related to the cross section of the middle part, and Fig. 12b is attributed to the cross section of the central part. As observed in these micrographs, the thickness of the coating has been reduced on the entire surface of the sample because of exposure to the oxyacetylene flame. Notably, the thickness of the coating in the ZSS-5WM specimen before the flame test was around 836 µm, and due to the erosion caused by the heat of the oxyacetylene flame, the thickness of the composite coating in the central part reached 612 µm and in the middle part it was 640 µm. It appears that the heat of the flame is distributed almost evenly in the central and middle parts of this sample, and the amount of detachment from the middle and central parts is almost the same. Therefore, it can be concluded that the addition of WC and MoSi₂ in the initial composition of the coating causes heat loss from the flame through the distribution of this heat over the entire surface. Moreover, the consumption of flame heat for the oxidation of phases including W and Mo is one of the positive effects of using these additive ceramics in the composition of composite coating.

Fig. 12

SEM images of the cross section of ZSS-5WM specimen: (a) middle and (b) central parts.

In this part, the possible oxidation reactions in the investigated system are discussed. As a result of the impact of oxyacetylene flame with a temperature of ~ 2000°C on the surface of the samples with protective coating, the structure of the composite coating applied on the graphite substrate has undergone phase changes due to chemical reactions, which will be discussed further.

In the flame test and during oxidation, ZrB₂ converts to stable ZrO₂ and B₂O₃; in contrast, boron oxide evaporates at temperatures > 900°C. Consequently, the development of the reaction products is controlled by the rate of ZrO₂ formation and B₂O₃ evaporation. The oxidation of ZrB₂ can be described as follows:^39,40,41

$$ {\text{ZrB}}_{{2}} \left( {\text{s}} \right) + {5}/{\text{2O}}_{{2}} \left( {\text{g}} \right) = {\text{ZrO}}_{{2}} \left( {\text{s}} \right) + {\text{B}}_{{2}} {\text{O}}_{{3}} \left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 740.67{-}0.43T\;\left( {{\text{kJ}}} \right). $$

(2)

SiC shows two types of oxidation behavior at high temperatures, depending on the oxygen potential of the environment, as mentioned in Ref. 42. At high oxygen potentials, passive oxidation occurs so that a protective SiO₂ film is formed on the surface by the reaction in Eq. 3,^40,41 which we are facing in this research:

$$ {\text{2SiC}}\left( {\text{s}} \right) + {\text{3O}}_{{2}} \left( {\text{g}} \right) = {\text{2SiO}}_{{2}} \left( {\text{l}} \right) + {\text{2CO}}\left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 2834.3 + 0.919T\left( {{\text{kJ}}} \right). $$

(3)

Due to the presence of the SiC phase in the composition of the composite coating and its oxidation during the ablation test, the SiO₂ phase is formed (Eq. 3). This phase together with other oxides (resulting from the oxidation of other primary phases) forms a glassy phase with high viscosity on the surface of the composite coating, which prevents the diffusion of oxygen and the oxidation of the lower parts at the beginning of oxidation/ablation.

At low oxygen potentials, due to the formation of gaseous products, active oxidation occurs according to the reactions in Eqs. 4 and 5:^40,41

$$ {\text{SiC}}\left( {\text{s}} \right) + {\text{2SiO}}_{{2}} \left( {\text{s}} \right) = {\text{3SiO}}\left( {\text{g}} \right) + {\text{CO}}\left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 490.56 + 0.683T\left( {{\text{kJ}}} \right) $$

(4)

$$ {\text{SiC}}\left( {\text{s}} \right) + {\text{O}}_{{2}} \left( {\text{g}} \right) = {\text{SiO}}\left( {\text{g}} \right) + {\text{CO}}\left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 692.56 + 0.212T\left( {{\text{kJ}}} \right). $$

(5)

The small amount of ZrC phase, formed during the coating process by the SPS³³ as shown in the phase analysis of the samples after the sintering, undergoes oxidation during the ablation test and assists in the formation of ZrO₂ protective layer (Eq. 6):⁴⁰

$$ {\text{ZrC}}\left( {\text{s}} \right) + {\text{2O}}_{{2}} \left( {\text{g}} \right) = {\text{ZrO}}_{{2}} \left( {\text{s}} \right) + {\text{CO}}_{{2}} \left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 719.94 - 0.235T\left( {{\text{kJ}}} \right). $$

(6)

By increasing the amount of WC and MoSi₂ additives (in ZSS-5WM sample), the heat distribution on the entire surface of the coating has become more uniform. Based on the observation of the phase analysis³³ and after the flame test, it seems that the phase containing W (WB) formed during the SPS process is oxidized to WO₃ (Eq. 7)⁴⁰ and its vaporization absorbs the thermal energy of the flame. This phenomenon prevents the oxidation and destruction of the main structure of the composite coating. On the other hand, the thermal conductivity of the phase containing W (WB) is higher than that of ZrB₂.⁴³ Therefore, by increasing the thermal conductivity, the composite coating causes the heat of the flame to spread throughout the composite coating and the graphite substrate, which leads to a decrease in temperature. This phenomenon can help in increasing the ablation resistance of the composite coating. The oxidation of MoB, which was formed during the SPS process,³³ through Eq. 8 and its conversion to MoO₃⁴⁰ and finally its evaporation, absorbs a lot of heat from the flame and causes loss in the energy of the oxyacetylene flame. This causes a delay in the oxidation process and destruction of the main structure of the composite coating and can help to increase the resistance against oxidation/ablation.

$$ {\text{2WB}}\left( {\text{s}} \right) + {9}/{\text{2O}}_{{2}} \left( {\text{g}} \right) = {\text{2WO}}_{{3}} \left( {\text{g}} \right) + {\text{B}}_{{2}} {\text{O}}_{{3}} \left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 1057.9 - 0.379T\left( {{\text{kJ}}} \right) $$

(7)

$$ {\text{2MoB}}\left( {\text{s}} \right) + {9}/{\text{2O}}_{{2}} \left( {\text{g}} \right) = {\text{2MoO}}_{{3}} \left( {\text{g}} \right) + {\text{B}}_{{2}} {\text{O}}_{{3}} \left( {\text{g}} \right)\quad \Delta G^{ \circ } = - 1281.1 + 0.142T\left( {{\text{kJ}}} \right) $$

(8)

In general, because of the chemical reaction between the raw materials such as WC and MoSi₂ additives in the coating material and the synthesis of new phases during the SPS process, flame energy is consumed due to the oxidation reactions mentioned above. The phases created during the ablation test, through absorption of flame heat, oxidation and finally evaporation, as well as by distribution of flame heat, due to their high thermal conductivity, cause heat loss on the surface of the composite coating and protection of the main structure of the coating.

Notably, the amounts of WC and MoSi₂ additives were selected based on the data obtained from the previous works of our group. For example, higher amounts of WC lead to more refractoriness, requiring a higher sintering temperature, while the optimum content selected in this research has led to improved properties, especially resistance to ablation. On the other hand, MoSi₂ acts as a sintering aid and improves sinterability, thereby reducing the temperature required for sintering. Alternatively, using a high amount of it in the composition leads to a decrease in oxidation resistance (due to the formation of a low-temperature SiO₂ phase in a large amount, which has poor properties at high temperature) and partial melting on the surface of the composite coating.

Conclusion

In this research, the protective composite coatings based on ZrB₂-Si-SiC along with MoSi₂ and WC additives in amounts of 0 vol.%, 2.5 vol.% and 5 vol.% were applied by the SPS method on the graphite substrate to boost their ablation response. Based on the results obtained from the ablation test, the samples with a protective composite coating have a much lower mass reduction compared to the graphite substrate without a protective coating and therefore have a very high resistance to ablation. The presence of additives such as WC and MoSi₂ in the composition of the composite coating led to the synthesis of new phases such as WB and MoB. These new phases formed during ablation cause flame heat loss by absorbing flame heat, oxidizing and finally evaporating as well as by spreading flame heat on whole surfaces of the composite coating due to high thermal conductivity and protect the main structure of the composite coating. The ZrB₂-SiC-Si coating had better ablation resistance than other samples against oxyacetylene flame. In the first stage of oxidation/erosion of this sample, the parts close to the surface that are exposed to the flame and withstand more heat, due to the oxidation process and the formation of new phases such as ZrO₂ and SiO₂ and the evaporation and removal of SiO₂, get a porous microstructure that acts like a barrier against ablation and protects the underlying parts.