An investigation on thermal shock resistance of ZrB2-SiC composites

In this research, the effect of particle size of the SiC powder was studied on the thermal resistance. In this regard, initial SiC powder was prepared in three sizes of 25 μm, 45 nm, and 80 nm. The sintering of the composites was performed at three temperatures of 1600 °C, 1700 °C, and 1800 °C, as well as at three times of 4 min, 8 min, and 12 min under the constant pressure of 40 MPa. The thermal shock resistance of the sample was investigated through fast cooling in water. Moreover, the flexural strength of the samples was measured through the three-point method. The results showed that the thermal shock resistance is improved by declining the particle size of initial SiC from micro to nano in such a way that the amount of the critical Δ T for the samples sintered at 1700 °C and 1800 °C reaches 510 °C and 590 °C. The flexural strength reaches the highest value (about 400 MPa) using nano scale SiC particles.


Introduction
Today, the need for novel materials that can endure temperatures higher than 2000°C for a long time in addition to the functionality in the oxidizing and corrosive atmosphere is necessary. Besides, sharp edges and surfaces can be employed in order to improve the aerodynamics level of ultrasonic carriers, re-entry aircraft, propulsions, and so on. The implementation of these types of designs requires high capabilities materials in various operational mediums like oxidizing atmosphere at high temperatures and corrosive gas at high speeds. Nowadays, there are a limited number of materials meeting these capabilities. To solve this problem, ultra-high temperature ceramics (UHTCs), which are known as an advanced group of materials, have been at the center of interest due to their attractive properties. Ultra-high temperature ceramics belong to one of the refractory material groups, including metallic borides, carbides, and nitrides [1][2][3][4][5][6][7][8][9][10].
Good oxidation resistance and high thermal and electrical conductivity of diboride refractory compositions compared to the other intermetallic compounds (carbides and nitrides) have been led to an extensive investigation on the transition metal diborides from the fourth to the fifth group of the periodic table (Ti, Zr, Hf, Nb, Ta) [7,11,12].
Borides and carbides such as TiB 2 , TiC, NbC, and NbB 2 also form several compounds like TiO 2 and Nb 2 O 5 with a temperature lower than 2000°C, which will be not usable [13]. Among diborides, ZrB 2 -based ceramics have been at the center of interest because of their exclusive compositions with relatively low density, good electrical and thermal conduction, high hardness, high melting temperature, good thermal shock resistance, and excellent chemical and mechanical stability at high temperatures [5][6][7].
However, extensive investigations have been implemented in order to improve the oxidation and shockability of this ceramic at high temperatures. Zimerman et al reported that the addition of 30 vol% SiC substantially improves thermal shock resistance [14]. The triple system of ZrB 2 -SiC-G (ZSG), as a suitable shock resistance system, has been studied by various researchers [15][16][17]. Additionally, the manufacture of ZrB 2 porous ceramics, layered composites, and using a secondary phase like ZrO 2 [18][19][20][21], which suffers a phase change during applying a thermal cycle, are of the research works conducted to improve the shock resistance of this ceramic. The addition of Al 3 B improves shockability up to 600°C in addition to improving sinterability and fracture toughness [22]. The manufacturing parameters like milling time and sintering conditions also significantly affect shockability so that the critical temperature difference increases from 580°C to 650°C by increasing time from 3 to 15 h [23]. The effect of additives like MoSi 2 , SiC, HfB 2 , and carbon fiber along with manufacturing parameters on the shock resistance has been examined using an experiment designed by the Taguchi method [24].
Different methods have been used in order to study the shock resistance behavior thus far; they are categorized into two groups of applying shock during heating and cooling, which are carried out through electrical resistance and cooling in water, respectively. The second method is the simplest one [17] that was utilized in this research.
According to the results obtained from the effect of milling time in previous research works, we tried to improve shock resistance using initial SiC nanoparticles. Table 1 presents the primary characteristics of SiC and ZrB 2 powders used in this research. For each composite sample, the powders were dispersed and mixed based on the values obtained from rule mixture composites based on ZrB 2 -30 Vol% SiC. In order to reduce the particle size and create better mixing with additives, the mixture of the powders was milled by a planetary mill with a speed of 200 rpm for 3 h after dispersion. Milling was performed in an ethanol medium because of the possibility of oxygen adsorption in the surface powder particles.

Experimental procedure
After milling, the powders were dried and the resulted powder mixture coated into the cylindrical graphite mold by a graphite layer was charged and was then located into the SPS container. The SPS process was carried out using an SPS device (model: SPS-20T-10) at different temperatures, times, and pressures (table 2). After the SPS process, a polishing process was accomplished with a diamond stone to remove the graphite layer on the surface of the samples.
The samples were cut to a thickness of 0.1 mm by wire-cut to perform mechanical properties, shockability, and microstructure studies. To determine the density, three slices were cutted from each sample and their relative density and open porosity percent were measured by the Archimedes method. To evaluate the shockability, the samples were heated into a furnace up to the temperatures of 400°C, 450°C, 500°C, 550°C, 600°C, 650°C, and 700°C and held for 20 min 25°C. For each samples, five bars were quenched into water held at the temperature of 25°C and an average retained strength was calculated and reported. The flexural strength of the samples was determined through the 3-point bending test (five bars for each sample) and the results compared with those of un-heated (original) samples. Scanning electron microscopy (SEM) was employed to investigate the microstructure of the samples. 3. Results and discussion

Mechanical properties
The grain size of the composite, the relative density, and the open porosity percent of the sintered samples are presented in table 3. The microstructure analysis was given in the previous work [22,23]. Figure 1 shows the flexural strength of the samples together with their SEM images. The studies indicate that the flexural strength depends on grain size and density.
For instance, the addition of SiC to the HfB 2 ceramic (in comparison with pure HfB 2 ) leads to depletion of HfB 2 surfaces from oxide impurities (i.e. HfO 2 and B 2 O 3 ) under the influence of chemical reaction with SiC.
It was revealed that the presence of oxide impurities on the surface of diborides (e.g. ZrB 2 , TiB 2 and HfB 2 ) [24] not only acts as negative factor on the densification process, but also accelerates the grain growth. On one hand, by the addition of SiC to the TiB 2 , the oxide impurities were reacted and eliminated from the surface of TiB 2 particles and consequently resulted to inhabit the sharp grain growth. On the other hand, the unreacted SiC grains as well as the in situ formed interfacial TiC phases acted as the colonized grain growth inhibitors by pinning the grain boundaries.
Hardness and flexural strength improve due to removing the porosities on one hand and controlling the grain size of SiC on the other. Wuchina and co-workers [25] reported an ambient temperature the flexural strength of 440 MPa for the modified monolithic HfB 2 , which is hot pressed at 2150°C. This value is lower than that reported by Monteverde (590 MPa) [24].
According to the data presented in table 1, the amount of porosities substantially reduces from 20% to about 7% while due to the short time of the sintering process, it is not observed substantial grain growth in spite of increasing sintering temperature. Hence, reaching a low porosity and suitable grain size microstructure results in increasing flexural strength at 1800°C. Similar to sintering temperature, the holding time parameter has a direct effect on the microstructure (porosities and grain size). The best flexural strength was obtained for 8 min sintering time. Although increasing the sintering time to 12 min results in a relatively dense sample (about 3% porosity), a decreasing trend of flexural strength is expected. On the other hand, the sensible growth of the grain size (1.1 μm) is as a result of high sintering temperature (1800°C) for 12 min, which in turn leads to decreasing flexural strength. In addition to sintering parameters, the effect of the initial powder size of SiC was studied on flexural strength. Using microscale nanoparticles brings about improving sinterability and reducing the size of the composite grains. Liu et al [26] studied the effect of SiC nanoparticles on the microstructure and mechanical properties of ZrB 2 . The particle size of ZrB 2 and SiC initial powders used in this research were 2 μm and 30 nm, respectively. The powder mixture was sintered at 1900°C for 30 min under a pressure of 30 MPa by the hot press (HP) method. They reported that the SiC intergranular particles with a size of lower than 1 μm play a key role in locking boundaries, so it hinders the grain growth. The reduction of the grain size along with sinterability improvement lead to increasing flexural strength in the samples having nano-scale SiC particles compares to micro-scale SiC ones. Moreover, there are high residual stresses in the interface of the transgranular nanoparticles and matrix grains (sub-interfaces) due to the presence of transgranular nanostructures, which can lead to the formation of the subgrain boundary. Due to these sub-grain boundaries, a huge amount of fracture energy is consumed during the flexural strength tests and consequently leads to improving flexural strength [26].

Shockability
Thermal shock means the resistance to failure resulted from the temperature changes applied to the material, or in other words, cracking in the material owing to the thermal stress initiated from severe temperature changes; ceramics and glass segments are prone to this type of fracture owing to low toughness, low thermal conduction, and high thermal expansion coefficient [27][28][29]. Temperature difference and stresses created lead to making a crack on the surface of ceramic material, resulting in the fracture of the sample with its growth and propagation [27][28][29].
The fracture of the sample, which is initiated from thermal shock, can be controlled by the following factors: (1) Reduction of material thermal gradient (through slower applying temperature changes), (2) Reduction of material thermal expansion coefficient, (3) Increase of material strength, (4) Heat treatment and stress relief, (5) Reduction of young modulus, and (6) Increase of toughness (control of crack) [27][28][29]. In order to investigate the effect of temperature, sintering time, and SiC particle size on thermal shock resistance, at first, the residual flexural strength graphs in terms of DT were drawn separately for sample separately (see figure 2(a)) and were analyzed in the next sections.
It is clear, applying higher temperature difference values lead to declining flexural strength, which is initiated from residual stresses created. Obtaining the critical temperature of the shockability is expressed for sample No. 1. For this purpose, the point which the flexural strength reaches to 70% of its initial value is discernible and from there a line is drawn on the X-axis; the intersection location with the X-axis displays the critical temperature difference of the shockability, which is equal to 480°C.

Effect of sintering temperature on thermal shock resistance
To investigate the effect of sintering temperature on the thermal shock resistance, the samples No. 7, 3, and 9 (sintering temperature: 1600°C, 1700°C, and 1800°C; sintering time: 4 min; SiC particle size: 45 nm; sintering pressure: 40 MPa) were compared. As shown in figure 3, the thermal shock resistance increases by increasing the sintering temperature.
The size of the SiC grains increases from 2.1 to 4 μm by increasing the sintering temperature from 1600°C to 1800°C. The studies reveal that the finer grain microstructures indicate better shock resistance from themselves [20]. However, it is observed that increasing the sintering temperature results in a dramatic improvement of sinterability and reduction of porosities from 8% to 2%. Porosities act as channels to heat transfer and help to improve shockability. Furthermore, the presence of open porosities leads to consuming crack energy and preventing its propagation in the microstructure, and finally improving fracture toughness through activating the toughening mechanisms (entrapping cracks in porosities). According to that fracture toughness has a direct relation with shockability [27,28] (see figure 4), and also the information presented in sections 3-4, it can be concluded that the presence of porosities is a positive factor for improving shockability. It should be paid attention to the amount of porosities available in the microstructure. Due to the high amount of porosities in the samples sintered at 1600°C and 1700°C and the lower initial flexural strength of these samples compared to the sample sintered at 1800°C, the thermal shock resistance of these samples are lower.

Effect of sintering time on thermal shock resistance
To investigate the effect of sintering time on the thermal shock resistance, samples No. 4, 5, and 6 (sintering temperature: 1800°C; SiC particle size: 45 nm; sintering pressure: 40 MPa) were compared. As can be seen in figure 5, thermal shock resistance decreases by increasing sintering time. An investigation on the graphs of grain 3.2.3. Effect of SiC grain size on thermal shock resistance The effect of SiC particle size on the thermal shock resistance at two temperatures of 1700°C and 1800°C is presented in figure 6. As seen, thermal shock resistance has an upward trend by decreasing particle size in both temperatures; this increase is more evident at 1800°C. Thermal sock resistance decreases by reason of the constant percent of the porosity in this range on one hand and the increase of the grain size on the other. By examining the trend of microstructural and density changes of these samples using the data presented in table 3, it is observed that the grain size of the resulted composite decreases about 1 μm at both temperatures by decreasing initial particle size (from 4.2 to 3.2 for 1700°C and from 4.4 to 3.8 for 1800°C) while the percent of the porosities in the samples sintered at 1800°C and 1700°C reaches 7% and 17%, respectively. On the other hand, the high amount of porosities leads to decreasing the initial flexural strength, fracture toughness, and finally shock resistance.

Conclusion
ZrB 2 -30 vol% SiC were fabricated by SPS using different initial SiC particle size successfully. It was cleared that the shockability improved by sintering temperature up to 1800°C. In fact in this temperature, desirable microstructure with good combination of grain size and relative density was achieved. Raising the holding time of SPS at the temperature of 1800°C, the shockability decreased continuously. It means both coarsening the grain size and reaching to nearly full relative density are detrimental for shockability. Finally, it was found that using finer initial SiC powder in nano scale (from 25 μm to 40 nm) significantly improves the shockability especially at the upper temperature (1800°C in comparison with 1700°C).

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).