Effect of SiCw Whiskers on the indentation thermal shock resistance of ZrB2-MoSi2 Composites

A comparative evaluation of the thermal shock resistance (TSR) of ZrB 2 -20 MoSi 2 20 SiC w (ZMS w 20) and ZrB 2 -20 MoSi 2 - 5SiC w (ZMS w 5) composites were studied using the indentation quench method. High-dense ZrB 2 based composites were prepared by multi-stage spark plasma sintering at 1700 °C. The results show that the ZMS w 20 composite expressed superior crack shielding and TSR under quenching circumstances. The critical temperature differential ΔT c of ZMS w 20 ceramic (ΔT c = 800°C) was higher than that of ZMS w 5 ceramic (ΔT c =600°C). The significant enhancement in TSR was imposed residual stresses that improved the resistance to crack progression during thermal shock. Furthermore, an increment in silicon carbide content reduces the crack growth, and increases the TSR of the composites.


Introduction
During re-entry at the earth atmosphere, leading edges and nose cones of the hypersonic vehicle are encountered in thermal and oxidation loads, leading to the evaporation of the sharp outer surface by aerodynamic heating [1,2]. Moreover, hypersonic velocity also generates shock waves in front of the vehicle [3]. Hence, it is demanded to develop thermal protection system (TPS) materials for overcoming those problems. The TPS materials should have a high melting point, adequate ablation, erosion, and oxidation resistance [4,5]. Therefore, the development of such material is a real challenge. Ultrahigh-temperature ceramics (UHTCs) are one of the most favorable high-temperature materials for overcoming these problems. Borides, carbides, and nitrides of early transition metals are considered UHTCs [6][7][8]. In the group of all ultra-high-temperature ceramics, zirconium diboride (ZrB2) is a suitable material due to it has a very superior melting temperature (3250 °C) and lower density (6.01 g/cm3) than other UHTCs [5].
However, zirconium diboride is one of the strong covalent bonds with a hexagonal crystal system [9]. Moreover, the boride ceramics show higher thermal and electrical conductivity at room temperature than nitride and carbide ceramics [8,[10][11][12][13][14]. Borides also have good chemical resistance [8]. So, those unique characteristics of mechanical, physical, and thermal properties make the ZrB2 ceramic suitable for thermal protection and propulsion systems of hypersonic vehicles, rockets, and other high-temperature industrial fields such as foundry electrodes for electric arc furnace and refractory fields [15,16]. However, the pure monolithic ZrB2 can not be used in actual application due to its low oxidation resistance and damage tolerance. It is also difficult to fabricate due to its poor self-diffusivity and strong atomic bond strength [17]. Also, a very high temperature (around 1900°C) is required for sintering during the pressure-assisted sintering technique to eliminate the porosity [8]. However, the densification, mechanical properties, oxidation resistance, creep, wear, and ablation resistance of diborides can be enhanced by the addition of the second phase additives like SiC, MoSi2, Si3N4, and tantalum compounds [18][19][20][21][22][23][24][25][26][27][28]. Moreover, SiC and MoSi2 are the most promising second phase element in ZrB2 ceramic to achieve a unique combination of mechanical, physical, thermal, and high-temperature oxidation properties [29,30]. Monteverde [29] reported superior sinterability, mechanical, and oxidation behaviors of ZrB2-SiC-MoSi2 composites than the monolithic ZrB2 and ZrB2-SiC composites.
In another work, Mashhadi et al. [18] observed an enhanced density, hardness, and fracture toughness of the ZrB2-SiC-MoSi2 ceramics as compared to ZrB2-SiC composites. Guo  Researchers have been trying to developing new thermal shock inhibitors of ZrB2-based UHTCs ceramics. Less number literature is obtainable on the ZrB2-MoSi2-SiCw ceramics densified by a multi-stage SPS sintering technique. However, the addition of the silicon carbide whisker (SiCw) on the thermal shock characteristic of the ZrB2-MoSi2 ceramic is unexplored. The thermal shock resistance characteristic investigation will director the applied application of SiCw added ZrB2-MoSi2 based composites. Therefore, the primary objective of this investigation is to identify the role of silicon carbide whisker content on the TSR of ZrB2-MoSi2 ceramics.

Composite preparation
High purity (≥99%) raw powders used in the investigation were ZrB2, SiCw whiskers, and MoSi2 powder collected from Alfa Aesar, MA, USA. Two ZrB2-20 vol.% MoSi2 based ceramics improved with 5/20 SiCw (ZMSw5 and ZMSw20) were fabricated by multi-stage spark plasma sintering. The procedure had been described in the previous article [36] in detail. The graphical representation of the multi-stage SPS sintering schedule is presented in Fig.1.

Composite characterization
X-ray diffractometer (XRD) analyses using Cu Kα radiation were performed to identify the existent phases in the sintered composites [37]. The densities of the sintered samples were measured on the basis of the Archimedes principle. Sintered samples were partitioned for microstructure characterization. The microstructure and phase composition of the multi-stage sintered composites were examined by FESEM (field emission scanning electron microscopy) and energy-dispersive spectroscopy (EDS), respectively.
Flexural specimen bars were subjected to a 3-point bend method in a universal testing instrument (Instron Model 3365, USA). The crosshead velocity was 0.1 mm.min -1 . The test was conducted according to ASTM C 1161-02 (ASTM, 2006) [38]. The microhardness and fractural specimen surface were polished up to a 1µm surface finish using diamond abrasive.
Then, the hardness and indentation fracture toughness (IFT) were determined by a microhardness machine (MMT-X7B, Hibiki Co., 2012) operated with an applied load of 500 gf and 5 kgf, respectively. Loads were applied for 15s for both cases. Minimum ten indents were generated on each specimen to estimate the average magnitude of IFT, which was determined using the following expression [39,40]: In which, E denotes the elastic moduli, HB denotes bulk hardness, L denotes the indentation load, a half diagonal length of indentation, and C = x+a, and x is the length of crack.

Thermal shock tests
The thermal shock resistance of the investigated composites was assessed by the indentation quenching method following the procedure established by Anderson and Rowcliffe [41]. Small samples with 10 mm × 6 mm × 3 mm were cut from the sintered specimens and polished metallographically. Four indentations were made on the surface of the polished sample by using a Vickers indenter at a load of 49.05 N in order to develop small initial cracks on the surface. The dwell time of each indentation was 15s, and a 653µm gap was maintained between indentations. Each indentation generated 4 cracks, and a total of 16 cracks were formed on the surface of the sample. After that, the specimens were kept inside a furnace atmosphere for 30 min at temperatures of 225°C, 275°C, 325°C, 375°C, 425°C, 625°C, 825°C, 1025°C, and 1225°C. Afterward, the specimens were taken out from the furnace and immediately quenched in room temperature (25 °C) water. The quenched samples were characterized by XRD analysis using Cu Kα radiation. The indentation crack length was evaluated before and after soaking using FESEM. The primary indentation crack length was restricted as the distance from the center of the indentation to the crack tip. After the thermal shock, the final crack length was equal to the initial crack length plus crack length propagated due to thermal shock.

Microstructure and mechanical properties
The relative densities of the sintered ZMSw5 and ZMSw20 composites were 97.50% and 99.50 %, respectively. Typical XRD patterns obtained from the multistage sintered   Cracks were generated by the Vickers indentation on the surface of the sample. Figure 5 shows indentation cracks present before thermal shock in ZMSw5 composite. The mechanical and thermo-physical characteristics of the ZMSw composites are tabulated in Table 1. Especially, ZMSw20 ceramic composite provides better fracture toughness due to the presence of the elongated and huge amount of whiskers that absorb a large quantity of fracture energy. The interface between matrix and whiskers deflects cracks that cause a tortuous fracture path (Fig. 5b). In addition, the introduction of Whisker SiC can reform the fractured nature of the composite by reducing the ZrB2 grain size.
Therefore, fracture toughness was improved with the addition of Whisker SiC [40].
In which, denotes the whole number of radial cracks generated from the indentation, and ∆ denotes the number of total available radial cracks propagated afterward water quenching.
In which, and ∆ denote the original indentation crack length prior water quenching and the extended crack length afterward water quenching.

Indentation-quench thermal shock investigations
The percentages of crack extension after water quench at room temperature vs.
quenching temperature differential for ZMSw5 and ZMSw20 ceramics are displayed in Fig.7. Figure 8 shows crack propagation behavior after water quench at room temperature as a function of the quenching temperature differential for ZMSw5 and ZMSw20 composites sintered via multi-stage sintering route. It is observed that the fractions of crack extension and crack propagation are significantly related to temperature difference (∆T). The higher thermal-shock temperature difference shows a higher percentage of crack extension and cracks propagation. The ZMSw5 ceramics shows weak crack-growth resistance behavior after quenching. Moreover, %E and %P are greater than for the ZMSw20 ceramics. Figures 7 and 8 show that the ZMSw5 and ZMSw20 ceramic composites have an identical outline of crack progression. Hence, the crack progression can be parted into three sections. There is no significant crack growth at low ∆T (Section I). Cracks grow stably at medium ∆T (Section II). Cracks grow unstably over a certain high ∆T (Section III). The critical ∆TC for unstable crack growth of the ZMSw-20 ceramic is 800 °C, which is higher than that of the ZMSw5 composite (∆TC value for the ZMSw5 is 600 C). This is confirmed that multi-stage sintered ZMSw20 ceramics express superior thermal shock resistance behavior than multi-stage sintered ZMSw5 ceramic. Thermal shock resistance strongly depends on thermal and mechanical characteristics such as thermal conductivity, fracture toughness, elastic modulus, strength, and thermal expansion coefficient. Besides, all these properties are fundamentally related to the microstructure, density, chemical composition, and fabrication procedure of composites. Both ZMSw5 and ZMSw20 ceramics were prepared by spark plasma sintering via multi-stage sintering route. Even both ceramics have a similar chemical composition and almost the same microstructure. Consequently, one important question is why ZMSw20 has better thermal shock resistance than ZMSw5.
Ultra-high temperature ceramics (UHTCs) are candidates for applications under extreme environments such as thermal shielding elements and sharp leading edges for hypersonic re-entry vehicles, advanced nuclear fuels, rocket engines, nuclear reactors, furnaces elements, plasma -arc-electrodes, and refractory crucibles. So, understanding the temperature distribution and corresponding thermal stresses in UHTC materials under high temperatures is essential in order to assess their integrity. The TSR of ceramics is found to be very sensorial to their temperature-related material properties.
For ZMSw20, ∆TC for unstable crack propagation is 800 °C, superior to the ZMSw5. It could be investigated from Fig. 8, the crack progressions in both inspected samples introduced a stepwise model, which is directed that the temperature for crack extension is not steady because of the existence of various stresses at the crack tip. During multistage sintering, the formation of (Mo, W)Si2 is a product of the sintering reaction. In this viewpoint, the secondary phase of (Mo, W)Si2, which has a more significant CTE (8.3-6.7 × 10 -6 /K) [42] than the matrix of ZrB2 (6.91 × 10 -6 K-1) [43], results in the development of compressive residual stress during cooling from the sintering temperature. Such high compressive residual stress resists crack extension during thermal shock. Therefore, cracks in the ZMSw5 composite have grown unstably at ∆T ~ 600 °C, whereas cracks in the ZMSw20 composite have grown unstably until ∆T > 800 °C. Fig. 7. The percentage of crack extension after water quench at room temperature vs.
quenching temperature differential for ZMSw5 and ZMSw20 composites. Fig. 8. The percentage of crack propagation (% P) after water quench at room temperature vs. quenching temperature differential for ZMSw5 and ZMSw20 composites.

Oxide Phase identification upon thermal shock test
Typical XRD spectrums of the quenched surface of ZMSw5 and ZMSw20 composites are displayed in Fig. 9. Figure 9 reveals that the surface phases of ZMSw5 and ZMSw20 composites after water quenched at a series of temperature differences are consistent with primary materials. The XRD results also specify that the manifestation of oxides of ZrO2 and other oxide phases of Mo5Si3 and ZrSiO4. ZrB2 is partly oxidized to ZrO2. The relatively more peaks of the ZrO2 phases are detected in ZMSw5 ( Fig. 9(a)). The transformation of ZrB2 to ZrO2 can be progressed by raising the temperature differences. The amount of ZrO2 expresses an increasing trend from ΔT=800 °C.
However, poor oxidation resistance of ZMSw5 at 1200 °C exhibits the presence of more amount of ZrO2 phases. Overall, thermal shock for ∆T = 600 °C is less damaging than that for ∆T = 800 °C due to the presence of a high amount of oxide phase. However, this oxide phase is not favorable to resist crack growth during thermal shock.

Theoretical study of indentation thermal shock behaviour
The indentation of thermal shock mechanism should be analyzed by measuring stress intensity, which is generated at the indentation crack tips. For the ZMSw ceramic composites, the total stress intensity factor at the indentation crack tip is exposed after water quenching (KI) = residual stress intensity (Kr) is prompted by indentation load + thermal stress intensity by thermal shock experiment (Kth) Kr can be defined by: Where, ℵ is a constant value which is estimated as 0.016 ( ⁄ ) 0.5 , Pv and c represent the Vickers indentation load and Vickers indentation flaw of length, respectively.
E denotes the elastic moduli; H denotes hardness of the specimens.
In the case of thermal shock, the thermal stress intensity (Kth) of the indentation crack method is only measured at the surface of the specimen and is expressed by: Where, represents a geometry constant factor for a crack of semi-elliptical surface( = 4 2 ⁄ ), which is more or less the shape of median-radial indentation cracks.
The thermal stress is described as follows: Where, is the coefficient of thermal expansion, ∆ denotes the temperature difference and ν denotes the Poisson's ratio of the specimen.
( ) is a geometry factor and it is a function of heat transfer condition as well as the Biot modulus of and it is represented by: Where, denotes the characteristic length of heat transfer (i.e., the thickness of the specimen 2 ⁄ ), h represents the coefficient of heat transfer between specimens and the water cooling medium, and k represents the thermal conductivity of the specimen. In the case of cylindrical indentation specimens, the constant values of M, N, Z, and W are 3.15, 1.33,-0.266, and 5.14, respectively. [44].
Hence, after combining all equations, the total stress intensity ( ) of an indentation crack for ZMSw composite can be defined as follows [45]: The assumptions of mechanical equilibrium are considered for this equation (equation 10). will be determined to find out whether the indentation crack will propagate or not, even it also finds that the indentation crack propagates either a stable or unstable form. The indentation crack will expand in unstable form during thermal shock loading if ≥ and ≤ Where, is the indentation fracture toughness of the specimen.    The total stress intensity (KI) at the indentation crack tip vs. crack length (c) for the ZMSw20 composite is shown in Fig.10 (a), which are quenched under several thermal shock temperatures. The value of KI can be measured according to Eq. (10). The value of KI rises with increasing ∆T (Figure 10 (a)). Every KI has a minimum value at all ∆T.
Each minimum value of KI rises with the rise in temperature difference ΔT. When ∆T > 800 °C, the indentation cracks grow unstably because the KI value higher than the fracture toughness (KIC). Fig.10  Thermal shock temperature variance ∆T is prolonged, which confirms that cracks can propagate continuously. While ∆T≥800 °C, KI will be continuously increased than KIC in every period. Hence, indentation cracks will propagate in unstable while ∆T≥800 °C.
The critical thermal shock temperature difference is calculated from the theoretical analysis at 800 °C, which is also the best agreement with the experimentally obtained outcomes. Figure 11 shows that ZMSw5 also follows the same patterns as ZMSw20. For ZMSw5, the cracks will propagate unstably when ∆T≥600 °C. It also shows the best agreement with the experimentally obtained outcomes. Finally, it can be noted that ZMSw20 ceramics shows better thermal shock resistance than ZMSw5 ceramics because of the presence of high residual stress in ZMSw20 than ZMSw5 samples. Thus, crack propagation in the ZMSw20 composite needs an upper thermal shock temperature than that for the ZMSw5 composite. For this reason, the critical thermal shock temperature (∆TC) of 800 °C in ZMSw20 is higher than 600 °C in ZMSw5.

Microhardness
The changes in microhardness of both the ZMSw ceramics due to thermal shock are displayed in Fig. 12. It is explicit from Fig. 12 that (i) hardness of both ceramic losses with a rising thermal shock temperature (∆T); (ii) net drop in hardness detected for the ZMSw20 ceramic is lesser than that ZMSw5 composite. Figure 9 reveals that the generation of an intermittent scale of ZrO2 is more for the ZMSw5 composite. The CTE of ZrO2 (10.3 × 10 -6 K -1 ) [5] is considerably greater than that of matrix ZrB2, while its thermal conductivity is also lower (2.0 W/mK). This ZrO2 scale contributed to thermal stress due to the huge dissimilarity of thermal properties of the composites leads to spallation of the surface scale. Moreover, the mismatch of CTE of the W-rich interfacial phase and matrix phase or oxidation of W succeeded by evaporation of WO3 can play an essential function in damage through thermal shock, mainly on soaking at temperatures ∆T≥1000 °C.

Conclusions
The TSR of ZMSw5 and ZMSw20 composites are investigated through the indentationquench approach. The critical thermal shock temperatures of ZMSw5 and ZMSw20 are 600°C and 800°C, respectively, which indicate that the ZMSw20 composite exhibits better TSR under quenching conditions, compared to that of ZMSw5 composites. The theoretical analysis confirms that the results have a good agreement with the experimentally obtained outcomes. High residual stress is observed in multi-stage SPS sintered ZMSw20 composite. Such high compressive residual stress is a barrier or obstacle to the stress field intensity at the crack tip. It contributes to the improvement in crack progression resistance during thermal shock. However, poor oxidation resistance of ZMSw5 at 1200 °C exhibits more ZrO2 phases. This oxide phase is not favorable to resist crack growth during thermal shock. For this reason, an increment of SiCw content reduces the crack growth, and that successively enhances the thermal shock of the composites. The drop in hardness due to thermal shock for the ZMSw20 composite is less than that of the ZMSw5 composite.

Figure 1
Graphical representation of the multistage SPS sintering heating cycle.       The percentage of crack extension after water quench at room temperature vs. quenching temperature differential for ZMSw5 and ZMSw20 composites.

Figure 8
The percentage of crack propagation (% P) after water quench at room temperature vs. quenching temperature differential for ZMSw5 and ZMSw20 composites. XRD spectrums of sample surface afterward thermal shock at different temperature differences: (a) ZMSw5 and (b) ZMSw20.

Figure 10
Total stress intensity with various ΔT vs. of crack length and stress intensity with certain ΔT of Vickers indentation cracks at the surface vs. indentation crack length after water quenching for ZMSw20 ceramics (a,b).   Bar chart displaying changes in hardness of the ZMSw composites due to thermal shock through at soaking various temperatures.