Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Journal of the European Ceramic Society journal homepage: www.elsevier.com/locate/jeurceramsoc Original Article The effect of YB4 addition in ZrB2-SiC composites on the mechanical properties and oxidation performance tested up to 2000 °C Zuzana Kováčováa,b,*, Ľubomír Orovčíkc, Jaroslav Sedláčekd,e, Ľuboš Bačab, Edmund Dobročkaf, Michael Kitzmantela, Erich Neubauera a RHP-Technology GmbH, Forschungs- und Technologiezentrum, A-2444 Seibersdorf, Austria Department of Inorganic Materials, Institute of Inorganic Chemistry, Technology and Materials, Faculty of Chemical and Food Technology, Slovak University of Technology, Radlinského 9, 812 37 Bratislava, Slovak Republic c Institute of Materials and Machine Mechanics, Slovak Academy of Sciences, Dúbravská cesta 9, 845 13 Bratislava, Slovak Republic d Institute of Inorganic Chemistry, Slovak Academy of Sciences, Dúbravská cesta 9, 845 36 Bratislava, Slovak Republic e Centre of Excellence for Advanced Materials Application, Slovak Academy of Sciences, Dúbravská cesta 9, 84511, Bratislava, Slovakia f Institute of Electrical Engineering, Slovak Academy of Sciences, Dúbravská cesta 9, 845 04 Bratislava, Slovak Republic b A R T I C LE I N FO A B S T R A C T Keywords: ZrB2-SiC YB4 Oxidation Oxyacetylene torch ZrxY1-xO1.5+x/2 The influence of YB4 and Y2O3 on densification, mechanical properties and oxidation performance of ZrB2-SiC (ZS) composite was studied. The oxidation tests were performed in static air up to temperature of 1650 °C for 1 h as well as under dynamic conditions of oxyacetylene torch at 2000 °C. Static oxidation of ZS led to the formation of protective silica-based glass on the surface. However, ablation tests showed absence of silica in ablation centre. Only dense zirconia layer was left on the top of ZS. Composites with Y-containing additives exhibited significantly inferior oxidation performance in static conditions, since severe spallation and deeper degradation of the material were observed. On the contrary, the depth of material degradation after ablation was comparable with ZS. Samples were covered by solid solution of zirconia and yttria. Due to very low vapor pressure, yttriabased oxidation products are of interest considering even higher application temperatures exceeding 2000 °C. 1. Introduction Due to the growing interest in re-entry and hypersonic vehicles, there is an increasing need for materials able to withstand severe environmental conditions met during application. For the operation in an extreme environment, a proper material selection process must be emphasized. Ultra-high temperature ceramics (UHTCs) represent the most promising group of material candidates in view of of their very high melting temperatures (in excess of 3000 °C). Nevertheless, high melting temperature is only one of the key criteria. It is the oxidation performance which is nowadays limiting the application, and which is playing an important role in the selection process for materials [1]. Up to now, especially UHTCs such as HfB2 and ZrB2 composites with SiC reinforcement have been considered for such an application and have been subjected to extensive research efforts [2,3]. However, most of the studies are related to ZrB2-based UHTCs since ZrB2 has significantly lower price (approx. 10 x lower) and density values when compared to HfB2. Despite the superior combination of physical and chemical properties of high-temperature borides [4,5], these materials also have ⁎ their limitations. The oxidation resistance of ZrB2-based UHTCs depends on the formation of a protective layer on the surface and on oxidation products formed during exposure to an oxidizing environment. In fact, the formed oxide scale has a layered structure reported by several research studies. Typically, the formation of a protective borosilicate glass layer during oxidation of ZrB2-SiC composites at elevated temperatures is described by following equations [2,6–13]: ZrB2 (s) + 5/2 O2 (g) → ZrO2 (s) + B2 O3 (l) (1) B2 O3 (l) → B2 O3 (g) (2) SiC (s) + 3/2 O2 (g) → SiO2 (l) + CO (g) (3) The monolithic ZrB2 possesses a poor oxidation resistance. This is due to the emerging B2O3 formation which is liquid above 450 °C and wets the ZrO2 grains until it volatilizes above temperatures of 1100 °C due to the high vapor pressure. Above this temperature, the oxidation protection is provided by SiO2 (reaction (3)) which is significantly less volatile and more viscous compared to the boron oxide. This glassy layer reduces the diffusion into the bulk material and prevents the Corresponding author at: RHP-Technology GmbH, Forschungs- und Technologiezentrum, A-2444 Seibersdorf, Austria. E-mail address: z.ko@rhp.at (Z. Kováčová). https://doi.org/10.1016/j.jeurceramsoc.2020.03.060 Received 1 December 2019; Received in revised form 26 March 2020; Accepted 27 March 2020 0955-2219/ © 2020 Elsevier Ltd. All rights reserved. Please cite this article as: Zuzana Kováčová, et al., Journal of the European Ceramic Society, https://doi.org/10.1016/j.jeurceramsoc.2020.03.060 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. detailed studies of RE-B systems are rare and there has been very little research on the potential of rare earth metal borides for the use as materials for ultra-high temperature applications. Up to now, there is lack of fundamental studies on this material family. Some of them have been mentioned as possible candidates for application in extreme environments [23–25]. One of these candidates seems to be also yttrium tetraboride (YB4), which can act as a source of yttrium for the stabilization of an oxide scale formed due to oxidation of ZrB2 based UHTCs. In contrast to other borides of yttrium, YB4 possesses the highest melting point and chemical stability. Moreover, YB4 exhibits high hardness, strength, good electrical and thermal properties [25]. According to literature [25], an oxidation of YB4 is assumed as a two-step process, where the formation of Y2O3 and B2O3 by reaction (7) as a first step and their interaction to form YBO3 and B2O3 by reaction (8) as a second step is considered. oxidation progress. Unfortunately, SiC addition into the ZrB2 matrix is effective in terms of oxidation resistance only up to temperatures of 1700–1800 °C. On the one hand, SiC undergoes an active to passive transition of oxidation and a volatile SiO(g) is produced (reaction (4)). Active oxidation occurs at low partial pressures and very high temperatures. On the other hand, the initially formed silica based glassy phase is no more protective at ultra-high temperatures due to its evaporation and decomposition (reactions (5) and (6)). SiC (s) + O2 (g) → SiO (g) + CO (g) (4) SiO2 (l) → SiO2 (g) (5) SiO2 (l) → SiO (g) + 1/2 O2 (g) (6) ZrO2 as oxidation product is of interest because of its high melting point and relatively low vapor pressure [6]. However, the large volume changes due to the phase transformations (monoclinic to tetragonal at 1150 °C and tetragonal to cubic at 2370 °C) result in the destruction of material during operation. Thus, for practical application, ZrB2 based materials require a modification using appropriate additives [9]. Many researchers studied the influence of various additives in order to improve the performance, oxidation resistance and mechanical properties of this material system. Several studies have reported the beneficial effects of rare earth oxides (REO) on the densification, mechanical properties (hardness, fracture toughness and flexural strength), creep and wear resistance of REO-doped ceramics [14–19]. The addition of Y2O3 or more precisely yttrium-containing compounds offer several benefits. Yttrium oxide is generally used as a stabilization aid to zirconia. Regarding the volume changes connected to the polymorphic transformation of ZrO2 formed during oxidation, it could have an essential effect on the oxidation behaviour of ZrB2-SiC composites [2,14]. Typically, Y2O3 additions have been used to reduce the sintering temperature and to increase the densification of ZrB2-SiC materials [15,16]. Y2O3 reacts with SiO2 on the surface of SiC grains and forms an amorphous grain boundary phases. A stronger intergranular phase results in a higher flexural strength of the ceramics. Moreover, the Y2O3 addition was reported to supress the grain growth by reacting with oxides on the surface of the ZrB2 powder. Yttria is also considered to be a refractory phase with high melting point (2425 °C). Furthermore, oxidation above 1600 °C may lead to the formation of a refractory RE2Zr2O7 phase [20]. These zirconates have melting temperatures above 2300 °C and could provide oxidation protection when the borosilicate glass is vaporized from the exposed surface [2,15]. In fact, the compound where ZrO2 is associated with REO is known as pyrochlore phases. As reported by Opeka et al. [21], they exhibit significantly lower diffusion of oxygen compared to ZrO2. The in-situ formation of the pyrochlore phase could have an essential effect on oxidation performance of ZrB2 based materials. Moreover, due to the very low vapor pressure of yttria, it seems to be one of the most suitable candidates for this purpose. There are several beneficial effects of Y2O3 addition reported, especially on the densification and mechanical properties. However, the oxidation performance of Y2O3-doped material was inferior in comparison to a basic ZrB2-SiC composite. [14,22]. Besides oxides, rare earth elements form other compounds which could be considered for high-temperature applications. Borides of rare earth metals are interesting high-temperature materials with promising physical and structural properties. Despite this, 2 YB4 (s) + 7, 5 + O2 (g) = Y2O3 (s) + 4 B2 O3 (l,g) (7) Y2O3 (s) + 4 B2 O3 (l) = 2 YBO3 (s,g) + 3 B2 O3 (l,g) (8) Despite of relatively low liquidus temperature (1373 °C) of oxide mixtures from reaction (8) and a melting temperature of YBO3 (1650 °C), it is believed that YB4 has great application potential, especially in combination with other materials (mainly non-oxide ceramics). The main intention is to use YB4 as a new source of Y in ZrB2-SiC composites. This could lead to the modification of physical properties and enhancing the oxidation behaviour of the UHTCs [26]. YB4 has been the subject of only a few previous studies, mainly related to crystal structure [26,27] and investigation of physical and chemical properties [28–31]. However, recent data on physical properties are rare or still not available. Furthermore, YB4 powder is not commercially available, therefore various preparation methods have been investigated [23–25,28–30,32,33]. Up to now, only Zaykoski et al. [25] reported about oxidation behaviour and mechanical properties of YB4 material and its composites. The present study is focused on study of YB4 and Y2O3 addition in to a ZrB2-SiC conventional material. Densification, mechanical properties and oxidation performance of the reference and Y-modified ceramics are investigated. Oxidation behaviour was studied in static air as well as using an oxyacetylene torch setup up to 2000 °C. 2. Experimental procedure 2.1. Characterisation of starting materials and preparation of composites Commercially available ZrB2 (ABCR, Grade B, with a purity of 98.5 % min, Hf main contamination and a particle size of d90 of 5.2 μm), SiC (ABCR, UF25, with a purity of 99 % and a particle size of d90 of 0.76 μm) and Y2O3 (ABCR, with a purity of 99.95 % min and a particle size of d90 of 1.5 μm) were used as starting materials. The YB4 powder was synthesized by the boron carbide/carbothermal reduction method according to the following reaction: Y2O3 (s) + 2B4 C(s) + C(s) = 2YB4 (s) + 3CO(g) (5) The reaction was performed in a vacuum furnace at 1500 °C for 4 h as described in our previous study [34]. Three different composites have been prepared and analysed, as listed in Table 1. Theoretical density was calculated using the rule of mixture. The weight ratio of ZrB2 to SiC was maintained 8:2 for all Table 1 Starting composition, labelling and theoretical density of prepared composites. Material Labelling Theoretical density (g/cm3) ZrB2 (wt%) SiC (wt%) YB4 (wt%) ZrB2-SiC ZrB2-SiC-YB4 ZrB2-SiC-Y2O3 ZS ZSYB ZSYO 5.16 5.02 5.14 80 68.79 70.22 20 17.20 17.56 14.01 2 Y2O3 (wt%) 12.22 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Cu anode operating at 12 kW (40 kV/300 mA). All measurements were performed in parallel beam geometry with parabolic Goebel mirror in the primary beam producing the beam divergence of ∼ 0.03°. In order to achieve a good spatial resolution, the beam size was limited by two circular apertures with the diameter of 1 mm. To suppress the effect of defocusing, the width of the primary beam in the diffraction plane was further decreased by the slits of 0.6 mm and 0.2 mm (in some cases) wide, respectively. The XRD patterns were recorded in grazing incidence set-up with constant angle of incidence α = 12°. This rather large incidence angle (not typical for grazing incidence technique) was used in order to minimize the broadening of the irradiated area. At these measuring conditions the size of the irradiated area was estimated to be 1.7 mm × 2.9 mm and 1.7 mm × 1 mm, respectively. On of the advantages of the grazing incidence set-up is that the shape and the size of the measured area as well as the penetration depth of X-rays do not change during the measurement. The method is also insensitive to surface roughness and irregularities. All patterns were measured in the angular range 18° – 80° with a step size 0.05° and measuring a time of 1 s per step. The HighScore Plus analysis program as well as PDF-2 database were used for the qualitative analysis of the phase composition. The microstructure of the oxidized surfaces was observed by a 3D light microscope (VHX-5000, Keyence). The cross-sections of the hotpressed as well as oxidized samples were examined by a scanning electron microscope (SEM, JEOL 6061, FEG – JEOL 7600 F) equipped with an EDS analyser (OXFORD INSTRUMENTS, 50 mm2). Observation and measurements were performed in compo mode at 15 kV accelerating voltage using a backscattered electron detector. Cross-sections of selected samples were prepared by standard ceramographic processes (i.e., embedded in polymer matrix, grinded and polished with diamond pastes down to 1 μm finish). The average grain size was measured using ImageJ software. Mechanical properties of hotpressed materials, such as Vickers hardness and fracture toughness, were measured by an indentation technique using a hardness tester from Qness Q10 M equipped with a pyramidal indenter. 9.8 N (HV1) and 98 N (HV10) were applied and maintained during 5 s on the sample to measure the hardness and fracture toughness, respectively. Measurements were conducted 10 times to calculate average values. Young´s modulus and Poisson´s ratio were determined using ultrasound measurement system Olympus 38DL Plus. mixtures. The main goal was to prepare fully stabilized zirconia during oxidation of ZrB2. Therefore, the content of YB4 and Y2O3 was calculated in such way that after oxidation, 8 mol% of Y2O3 (either as starting material or as oxidation product of YB4 according to reaction 3) should be present in relation to the ZrO2 formed during oxidation of ZrB2 (reaction 1). The powder mixtures were homogenized in cyclohexane using yttria stabilized zirconia balls. The mixing was performed in a tumble mixer for 6 h. Subsequently the powders were dried in an evaporator and sieved. The prepared powder mixtures were filled into a graphite mould with an inner diameter of 65 mm. The consolidation of the prepared mixtures was conducted using a direct hot-pressing system (DSP518, Dr. Fritsch Sondermaschinen GmbH, Germany) at a temperature of 1900 °C in vacuum for 30 min with 30 MPa of uniaxial load applied from room temperature. The hot-pressed discs were subsequently planparallel grinded using a diamond wheel. After machining, bulk densities were measured using Archimedes´ method in distilled water at room temperature. The relative density (RD) was calculated as the ratio of Archimedes density to the theoretical density (calculated from the rule of mixtures) and expressed in percentage. Cylinders with a diameter of 10 mm and 20 mm and a height of 5 mm were cut out from the hot-pressed discs by electrical discharge machining (EDM). 2.2. Oxidation and ablation testing The oxidation behaviour of the composites was studied in static atmosphere of air as well as under dynamic conditions using an oxyacetylene torch in order to study the ablation behaviour. Static oxidation tests were performed on samples with a diameter of 10 mm in a furnace (super kanthal furnace, Classic 0518) at a temperature range of 1100–1650 °C for 60 min. Ablation tests were carried out in a flowing oxyacetylene torch environment on specimen with a diameter of 20 mm mounted into a graphite tube using vacuum under pressure. The used gas flows were adjusted experimentally to obtain a temperature of 2000 °C ( ± 20 °C) in the centre of the sample. The temperature was monitored by an optical two-colour pyrometer (Fluke Process Instruments). The gas flows were mixed in a ratio of 4.5:2.5 L/min for O2 and C2H2, respectively. The surface was vertically exposed to the flame for 60 s of ablation at final temperature. The distance between the nozzle tip of the oxyacetylene torch and the top of the specimen was approximately 10 mm. 3. Results and discussion 2.3. Characterization 3.1. Densification and microstructure of hot-pressed samples The performance with respect to the oxidation was evaluated according to the mass change and layer thickness after the oxidation test. The specific mass change of the samples tested in static conditions was calculated according to following equation: Fig. 1 shows the polished surfaces of hot-pressed ZS and ZSYB composites. It can be seen that the microstructure is characterized by coarser and bigger ZrB2 grains (bright phase) and dark SiC particles with near-equiaxed globular shape which are uniformly dispersed in the ZrB2 matrix for both ZS and ZSYB samples. The YB4 phase represented by grey colour was homogenously distributed within the matrix with size of approx. 2 μm. The Y2O3 phase on polished cross-section of the ZSYO samples was difficult to distinguish from ZrB2. Therefore, the fracture surface of ZSYO was observed and it is presented in Fig. 2. EDS mapping (not shown) revealed, that Y2O3 particles are homogeneously distributed in the matrix with a size of approx. 1.5–2.5 μm. Relative densities and average grain size of the composites hot pressed using the same conditions are listed in Table 2. The XRD analysis of hot-pressed samples confirmed only phases already present in the starting powders. The Y2O3 addition resulted in nearly full dense samples and supressed grain growth during hot-pressing, as expected and reported in literature [15,16,]. A very similar effect was observed also in the ZSYB samples. The YB4 addition positively affected the densification and the grain growth was also inhibited (compared to baseline ZS). w= m − mox A (9) where w is the specific mass change (mg cm ), m is the weight (g) of the sample before oxidation, mox is the weight (g) after oxidation and A is the surface area (cm2) of the sample exposed to oxidation. Specific mass change ablation rate was calculated as follows [35]: −2 Rm = m − mox Aox t (10) −2 -1 where Rm is the mass ablation rate (mg cm s ), Aox is the surface area (cm2) of sample exposed to ablation and t is the ablation time (s). The phase composition of samples oxidized in static air was identified by X-ray diffraction (XRD, Stoe Theta-Theta with Cokα radiation). XRD analysis of samples which have been tested using the oxyacetylene torch facility were carried out on selected areas using the diffractometer from Bruker D8 DISCOVER equipped with an X-ray tube with a rotating 3 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Fig. 1. Microstructures of ZS (a) and ZSYB (b) samples. investigated composites, c/a ratios were calculated and surfaces after indentation test were further polished and investigated using the 3D light microscope (shown for ZS in Figure S1). The depth of indents before polishing was approximately 8, 9 and 10 μm for ZS, ZSYB and ZSYO, respectively. Removing of material during polishing of the tested surface revealed presence of deep cracks extending under the indent tip and connected to the inverted pyramid. Moreover, the material has fallen out also between the cracks in some indents and the formation of cavitation was observed. Palmquist cracks are typically generated only under the apparent cracks and not under the indent [38]. For ZS sample, material was removed slightly under the indent. The cavity was clearly developed underneath the indent. The depth of cavity reached 10 μm (Fig. 4) and in some cases even 12 μm. Also, the cracks were still radiating outward from the indentation corners. This was observed for all samples. Moreover, cracks in ZSYO sample were still present even when a thicker layer of material was removed, and original indents were no longer observable. Considering c/a ratio, cracks remaining after polishing and depth of dropped out material, the present crack system was assessed to be median type. In our study, fracture toughness values were calculated according to following equation [37,39,40]: The densification process during hot-pressing was also evaluated by the shrinkage rate and derived from the displacement of punches due to the shrinkage (Fig. 3). Generally, a sharp increase of the shrinkage rate is associated with heating ramp. During an isothermal hold the shrinkage rate gradually slows down. The decrease in densification rate with time indicates achieving the final density. There is an obvious relationship between shrinkage rate and relative density – as the shrinkage rate increased, increased also RD (ZSYO > ZSYB > ZS). The onset temperature for ZS is around 1550 °C as observed also by Akin et al. [36]. The shrinkage rate as well as the displacement curve decreases smoothly. The shrinkage of ZS progressed continuously and was finished approx. 10 min before reaching the end of the holding time (see Fig. 3 (b)). The densification behaviour of composites with Y2O3 and YB4 was obviously different. The densification of ZSYB and ZSYO started earlier and the maximum peaks are slightly shifted to lower temperatures compared to the ZS. Moreover, shrinkage rates decreased more rapidly which can also be observed on displacement of the punch. Almost fully dense (99.9 %) ZSYO samples were obtained. No shrinkage appeared after approx. 15 min of isothermal heating. 3.2. Mechanical properties E 2/5 F ⎞ KIC = 0.0309 ⎛ HV ⎝ ⎠ c 3/2 The indentation technique was used to evaluate the fracture toughness of the studied composites. The present crack morphology should be known for selection of the appropriate mathematical model for the calculation of the fracture toughness values. The crack shape can be determined by investigation of the cross-section of the indent, further by successive polishing of the indented surface as well as indicated by the c/a ratio (the proportion of the crack distance from the centre of the indentation to the crack tip (c) to the half of the indentation diagonal (a)). The median (or half-peny) crack system remains connected to the indented pyramid and the ratio is > 2.5. Palmquist (or radial) cracks are detached from the indent also after polishing and the ratio is < 2.5 [37,38]. In order to determine the crack types in the (8) where E is Young´s modulus (GPa), HV is Vicker´s hardness (GPa), F is applied load during indentation (N) and C is the distance from the centre of the indentation to the crack tip. This equation was originally proposed by Niihara at al. [41,42] and used by several researches [37,39,43–45] for calculation of fracture toughness values considering median crack system. The obtained values of fracture toughness and c/a ratio show inversely proportional dependence, e.g. c/a ratio increases with decreasing KIC value. The mechanical property values are listed in Table 3. Young´s modulus values are comparable among all samples. The addition of YB4 Fig. 2. Fracture surface of ZSYO sample observed in compo mode (a) and secondary electron mode (b). 4 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Table 2 Relative densities and measured grain size in the prepared composites. Sample ZS ZSYB ZSYO RD (%) 98.8 ± 0.2 99.2 ± 0.3 99.9 ± 0.1 Average grain size (μm) ZrB2 SiC YB4 2.93 ± 0.76 2.36 ± 0.68 2.32 ± 0.65 1.00 ± 0.24 0.89 ± 0.28 0.76 ± 0.28 2.18 ± 0.51 Y2O3 1.94 ± 0.40 Fig. 3. The time dependence of shrinkage rate (a) and displacement of punch (b) during hot-pressing of ZS, ZSYB and ZSYO in relation to the temperature. Fig. 4. 3D inspection of polished indent in ZS sample - measurement of the indentation depth for ZS sample. 5 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. content to 8 vol% did not significantly improve the mechanical properties of ZrB2-SiC composite due to the excessive formation of a liquid phase on the grain boundaries. In present study, even a higher yttria content (approx. 16 vol%) was used with main focus on improvement of the oxidation performance. Although the grain growth suppression was observed in this study, the mechanical properties were deteriorated. Y2O3 as a monolithic ceramic exhibits inferior mechanical properties compared to the boride base ceramics. Relatively high content of yttria could also be a reason for degradation of mechanical properties. Decrease of fracture toughness of composites with Y2O3 can be interpreted as follows. As explained by Guo [53], fracture toughness is improved by large ZrB2 grains and/or small SiC grains. The large ZrB2 grains and intergranular cracks propagating along the grain boundaries enhance the fracture toughness. SiC contribution is attributed to the crack deflection that occurs near the SiC particles and/or at ZrB2/SiC interfaces. The small SiC grains increased the number of crack deflections and pull out of grains. The fracture toughness values for ZSYO is slightly inferior compared to the ZS, due to the lower content of ZrB2 and mainly SiC in matrix. Also, the ZrB2 grains were slightly smaller in ternary composites (Table 2). Regarding the ZSYB composite, there is no information about the effect of YB4 on mechanical or other properties of ZrB2-SiC composites up to now. In present study, measured values of Young´s modulus, hardness are fracture toughness are in the same range compared to the basic ZS material. This can be a consequence of similar RD. Table 3 Comparison of mechanical properties of ZS, ZSYB and ZSYO. Property ZS ZSYB ZSYO Young´s modulus (GPa) Poisson´s ratio Vicker´s hardness HV1 (GPa) Fracture toughness KIC (MPa m1/2) (c/a) 471 ± 2 0.137 ± 0.006 17.9 ± 1.0 474 ± 2 0.154 ± 0.004 18.3 ± 0.5 473 ± 4 0.145 ± 0.007 16.8 ± 0.2 4.48 ± 0.27 (3.64 ± 0.25) 4.22 ± 0.25 (3.78 ± 0.28) 3.91 ± 0.15 (3.89 ± 0.28) did not influenced the hardness and fracture toughness considerably. Although the addition of Y2O3 improved the RD, hardness and fracture toughness were noticeably inferior. The propagation of the fracture front involved a mixed transgranular and intergranular path. However, transgranular was significantly dominant, especially for ZrB2 grains. This was observed for all prepared composites (for ZSYO shown in Fig. 2). Obtained values of mechanical properties for ZS materials are comparable to that available in the literature taking into account materials with similar composition (e.g. ZrB2-30 vol% SiC; 20 wt% of SiC used in our study corresponds to approx. 32 vol% of SiC) [46–51]. Reported fracture toughness and Young´s modulus values ranged from 4.2 to 5.5 MPa m1/2 and from 484 to 520 GPa, respectively. These values are in good agreement with our study. However, the measured hardness (17.9 GPa) was slightly lower than reported (18–24 GPa), which is a consequence of lower relative density of ZS samples (98.8 %). As reported in several studies [51,52] - the hardness significantly increased as the porosity decreased. It has to be mentioned that UF-10 SiC quality (with a particle size of d90 of 2.1 μm) was used in most of the published studies. Only Zhu et al. [46] prepared ZrB2 composite using 30 vol% of SiC with the same quality as in our study (UF-25, d90 of 0.76 μm). They prepared almost fully dense samples with a relative density of 99.8 %. Higher RD is related to the prolongation of dwell time to 45 min and WC contamination (> 3.8 vol%) coming from mixing process. WC content could also have a beneficial effect on hardness (20.7 GPa) and Young´s modulus (520 GPa) of their samples. The same was observed also by Chamberlain et al. [47] – they prepared ZrB2 composite using 30 vol% of SiC with contamination of WC. The values of measured Young´s modulus, hardness as well as fracture toughness were higher compared to this study (484 GPa, 24 GPa and 5.3 MPa m1/2, respectively). Zhang et al. [16] observed that the addition of 3 vol% of Y2O3 positively affected the densification and the fracture toughness and supressed the grain growth of ZrB2-SiC. However, increasing of yttria 3.3. Oxidation in static air The graphs in Fig. 5 show the specific mass change and the thickness of the oxidized layer over temperature, given for three tested materials after oxidation performed in static air. To summarize all results regarding tested oxidation temperatures, the overall specific mass gain as well as thickness increase of the oxidized layer was observed to be lowest for ZS samples. Both evaluated values increase with the temperature of oxidation. However, the thickness of the oxidized layer for the ZS sample was still below 100 μm, even after oxidation at 1650 °C. Overview of all samples after static oxidation as well as table summarizing phase composition related to investigated oxidation temperature are presented in Figure S2 and Table S1, respectively. XRD analysis performed on oxidized ZS samples showed the formation of monoclinic ZrO2 (PDF 00-037-1484) as the predominant phase for all temperatures of oxidation. Moreover, a minor phase of tetragonal ZrO2 (PDF 01-079-1771) was found to be present after treatments above 1300 °C. Transformation from monoclinic to tetragonal modification occurs at around 1170 °C therefore only monoclinic modification was found in the samples after oxidation at 1100 °C. As the oxidation Fig. 5. Specific mass change (a) and thickness of oxidized layer (b) for ZS, ZSYB and ZSYO exposed for 1 h in static air at different temperatures. 6 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Y2O3 phase diagram. Local high-yttrium concentration regions present in the ZSYO samples could be the reason for the formation of stable cZrO2, even at this temperature. YBO3 was found as the main oxidation product of YB4 also by Zaykoski et al. [25]. The increase of the oxidation temperature led to the formation of mainly cubic solid solutions of zirconia and yttria as the main oxidation products. Generally, the binary ZrO2-Y2O3 system is characteristic for its wide range of solid solubility due to the structural similarity of the components. Solid solutions in this study correspond to the anion interstitial model represented by ZrxY1-xO1.5+x/2 formula [55]. Peak positions of this solid solutions are only slightly shifted; hence they appear as one single peak on the first sight. Detailed inspection of XRD patterns revealed the presence of cubic solid solution with different compositions: x = 0.76 (PDF 01-089-6687) and x = 0.85 (PDF 00-030-1468) at the oxidation temperature of 1400 °C and more. Moreover, in the ZSYB material most likely tetragonal solid solution is also present where x = 0.92 (PDF 00-048-0224). Formation of these solid solutions, also in the ZSYB materials, clearly indicate the presence of yttria during oxidation. Exposure of ZSYB as well as ZSYO to oxidation at 1400 °C resulted in the formation yttrium borate and yttrium disilicate phases. However, two different polymorphic modifications of Y2Si2O7 have been found. Monoclinic γ-Y2Si2O7 (PDF 01-074-1994) was identified in both oxidized samples. In ZSYO, also orthorhombic δ-Y2Si2O7 modification (PDF 01-082-0732) was found. The yttrium disilicate is well known for its polymorphism [56]. It exhibits up to seven different structural forms with different thermal stability, δ-type is reported being stable above 1650 °C [57]. It must be noted that Y2Si2O7 was found only after oxidation at 1400 °C. This phenomenon needs to be exploited in detail in future studies. Y2Si2O7 can be formed as an aimed oxidation product with beneficial influence on the oxidation performance of the base materials. Generally, there are 2 possible two reactions for the formation of yttrium disilicate. It can be formed due to the reaction of silica with yttria (reaction 11) or with yttrium borate (reaction 12). Fig. 6. ZrO2 grains size and SiO2 layer thickness of ZS as a function of oxidation temperature. temperature increased, thicker silica based glass formed (reaching approx. 23 μm after oxidation at 1650 °C) as shown in Fig. 6. Moreover, significant grain growth of ZrO2 was observed with increasing oxidation temperature. Fine ZrO2 grains below 0.5 μm were found after oxidation up to 1400 °C. ZrO2 grains reached approximately 2−3 μm when the sample was oxidized at a temperature of 1650 °C. Both ZrO2 grains as well as silica glass thickness followed a exponential growth with temperature. In Fig. 7 the image of a cross-section (a) as well as top surface (b) of a ZS sample oxidized at 1500 °C is presented. Oxidation of ZS led to the formation of a layered structure, typical for this material (Fig. 7 a). The oxidized layer of ZS consists of zirconia (reaction (1)) grains incorporated in silica glass (reaction (3)) and is shown in Fig. 7 b. Elemental mapping (not shown) clearly confirmed a compact silica layer on the top covering the ZrO2 grains. Moreover, a SiC depleted zone as well as oxygen diffusion into the material was observed. ZSYB and ZSYO samples did behave quite similar in terms of oxidation. XRD analysis of the oxidized surface showed very similar phase compositions after all temperatures of oxidation. Though some differences should be mentioned. The m-ZrO2 and h-YBO3 (PDF 01-0741929) have been identified as the main products of oxidation tests performed at 1100 °C and 1300 °C. The t-ZrO2 was found to be present as a minor phase in both materials oxidized at 1300 °C. Moreover, presence of t-ZrO2 and c-ZrO2 (PDF 00-049-1642) was confirmed in ZSYO samples already after oxidation at 1100 °C and 1300 °C, respectively. XRD patterns of ZSYB and ZSYO after oxidation at 1300 °C are shown in Fig. 8a. It is known from literature [54] that the existence of cubic zirconia modification is possible even at lower temperatures, mainly depending on the yttria content in the system according to ZrO2- Y2O3 + SiO2 = Y2Si2 O7 (11) 2SiO2 + 2YBO3 = Y2Si2 O7 + B2 O3 (12) With increasing oxidation temperature to 1500 °C formation of hYBO3 was found in addition to yttria-zirconia solid solutions. Besides this, minor phases of m-ZrO2 and crystalline h-SiO2 were identified in ZSYB and ZSYO, respectively. Incorporation of YB4 and Y2O3 into the ZS matrix led to the very similar microstructures after oxidation (as shown in Fig. 9), but quite different from the binary ZS composite. Elemental mapping measured on cross-sections of oxidized samples showed a chemical composition in good agreement with XRD results (Figure S3 and Figure S4). The creation of two layers due to the oxidation was observed for the Y-containing composites, where three segregated phases were observed in the outer layer. Bright ZrxY1xO1.5+x/2 grains of around 10 μm were agglomerating and surrounding Fig. 7. Cross-sectional view observed by SEM (a) and view of oxidized surface observed by light microscope (b) of ZS after oxidation at 1500 °C for 1 h in static air. 7 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Fig. 8. XRD patterns of ZSYB and ZSYO oxidized at 1300 °C (a) and 1650 °C (b) for 1 h in static air. The peaks are shifted vertically. The intensity scale for b is magnified twice. oxidation resistance of monolithic YB4 ceramics. Yttria is very stable in terms of oxidation and evaporation, therefore it is still present in the discussed region. For ZSYO composites, it can be assessed as a Si depleted region. In Fig. 9, severe cracking and defects in depleted layers can be seen. However, defect layers did not provide effective oxidation protection which can also be seen in the thickness as well as mass gain (Fig. 5). Oxidized layers of both samples are cracked and spalled from the surface. This unfavourable interaction between oxide scale and bulk material may be caused by several factors such as CTE mismatch, polymorphous transformations connected with volume changes or successive formation of volatile products escaping from the material. Further increasing of oxidation temperature to 1650 °C led to the formation of mentioned solid solutions and monoclinic modification of zirconia (Fig. 8 b). It has to be emphasized that the intensity scale for XRD patterns of samples oxidized at 1650 °C is magnified twice. The intensity peaks were noticeably lower due to the formation of significant amount of amorphous phase as indicated by hump observable in XRD pattern. Specific mass change as well as the thickness of the oxidized layer increased with oxidation temperature and both were found to be higher the YBO3 phase (indicated in Fig. 10 for both ZSYB and ZSYO). These agglomerates were dispersed in a high contrast dark phase consisting of Si and O, clearly identified as SiO2, formed due to the oxidation of SiC (reaction 3). Moreover, silica was also penetrating into the yttrium borate. The observed microstructure suggests the spinodal decomposition of formerly formed borosilicate glass and thus the formation of two separated phases of borate and silica. Elemental mapping of the ZSYO sample (Figure S5) also revealed the zirconia grains incorporated in the YBO3 area. Yttria was probably consumed for the formation of YBO3 which allowed the local formation of zirconia particles and not the solid solution mentioned above. YBO3 phase in ZSYB sample was more compact and tighter compared to the appearance found in ZSYO. It is also known from literature [21] that the introduction of transition metal oxides results in phase immiscibility of borate and silica glass increasing its viscosity. An increase in viscosity slows down the oxygen diffusion and suppresses the evaporation of boria from the glass. Elemental mapping also revealed depleted regions located between the outer oxidized layer and the unreacted bulk material. In case of the ZSYB composite, the region can be described as Si and also Y depleted zone. Low yttrium content in depleted zone can be assigned to the low Fig. 9. Comparison of cross-section of oxidized ZSYB (a) and ZSYO (b) at 1500 °C for 1 h in static air. 8 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Fig. 10. Detailed view of the cross-section of oxidized ZSYB (a) and ZSYO (b) at 1500 °C for 1 h in static air. compared to the undoped ZS composite, especially at elevated oxidation temperatures (see Fig. 5). Generally, the specific mass change of ZSYO is markedly higher compared to the one of ZSYB. This is due to volatile boria which is formed in much larger quantity during oxidation of ZS composite containing YB4. Although the XRD analysis (not shown for all temperatures of oxidation) did not show qualitative differences among phase compositions formed during oxidation, a variation of the peak intensity proportions corresponding to individual phases was clearly observed. This is assumed to be the reason for the discrepancies between specific mass change and thickness of oxidized layers of ZSYB and ZSYO. For example, despite of lower mass gain measured after oxidation at temperatures of 1100 °C and 1300 °C, the oxidized layer of ZSYB is thinner compared to the one of ZSYO. However, when analysing the peak intensities, it can be assumed that a higher amount of the YBO3 phase was formed for ZSYB. Yttrium borate has lower theoretical density compared to zirconia, 4.46 g cm−3 and 5.68 g cm−3, respectively. The increased thickness for the layer of the ZSYB samples after oxidation at temperature of 1650 °C could be explained otherwise. The top surfaces of ZS, ZSYB and ZSYO after oxidation at 1650 °C are shown in Fig. 11 a, b and c, respectively. ZS was covered with continuous thick glassy layer. The topography of oxidized surface of ZSYB was quite inhomogeneous in terms of roughness. The layer formed during oxidation was damaged and contained extensive and deep cavities and became interconnected. This is assumed to be the consequence of evaporation of boria during the oxidation. When comparing with the previous results for ZSYB, the oxidation layer of ZSYO appeared more compact and tighter. Fig. 12. ZS sample during oxyacetylene torch testing. qualitative way. The oxidation behaviour under the oxyacetylene torch was quite different compared to the testing of the samples in static air. Generally, all the materials tested in this severe environment were covered by a dense and adherent oxide layer. On the contrary, ZS composites with YB4 and Y2O3 oxidized in static conditions showed severe spallation of the oxidized layer, especially after oxidation at elevated temperatures. Morphology and macrostructure of these oxidized samples appeared different when comparing throughout the distances from the torch impact centre and the circumference. Therefore, XRD measurements were taken in labelled positions in order to investigate phase formations depending on the distance from the tip of 3.4. Ablation tests The illustrative image of a ZS sample during oxyacetylene torch testing is shown in Fig. 12. After the testing, the torch was removed, and the samples were cooled down to room temperature for posttreatment investigations. It has to be mentioned, that none of the tested samples broke or cracked during or after the test. This observation strongly suggests good thermal shock resistance of tested materials in a Fig. 11. Images taken by optical microscope on the top surface of ZS (a), ZSYB (b) and ZSYO (c) oxidized at 1650 °C for 1 h in static air. 9 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. ZrO2 (PDF 00-037-1484) was identified as the only phase in regions 1 and 2 and as dominant phase in region 3. A minor phase of tetragonal ZrO2 (PDF 01-081-1545) was found to be present in the outer area of the surface (position 3). However, it has to be emphasized that XRD patterns were measured using grazing incidence set-up. The penetration depth was below 10 microns at an incidence angle of 12 degrees. That means that only a thin surface layer was measured. It is possible that tetragonal modification is also present in positions 1 and 2 but it was not detected in the thin outer layer measured. Probably emerging tetragonal zirconia was transformed to the monoclinic during rapid cooling. The experiment was performed at 2000 °C which is below t→c transformation temperature, therefore it was not expected to find any cubic modification. The microstructure on the surface obviously changed depending on the distance measured from the tip of torch. The dense ZrO2 layer with diameter of approx. 10 mm was formed in the sample center, directly under the tip of the torch in the ablation centre (region 1). The presence of cracks could be the consequence of the internal stresses during cooling, CTE mismatch between oxide layer and bulk ZS ceramic and t-m transformation is accompanied by volume changes (increase 5%). Towards the edge of the sample, 2 additional regions can be observed. Region 2 is an intermediate region represented by an increasing proportion of the glassy silica phase at the expense of the crystalline ZrO2 phase. The outer ablation region (region 3) was characterized by dominating silica glass with incorporated zirconia crystals. Examination of cross-section of the ablation centre revealed a layered structure as shown in Fig. 14. Light microscopy as well as elemental mapping showed 4 distinctive layers as indicated in pictures. The outer layer was identified as a dense ZrO2 layer (1) with thickness of approx. 10 μm. There was an absence of Si as a consequence of several factors. Firstly, active oxidation of SiC (reaction 4) occurred at this temperature. Moreover, evaporation and decomposition of silica Table 4 Comparison of specific mass change ablation rate and thickness of oxidized layers for ZS, ZSYB and ZSYO samples oxidized under oxyacetylene torch at 2000 °C for 60 s. Specific mass change ablation rate Rm (mg cm−2 s-1) Thickness of oxidized layer in centre (μm) ZS ZSYB ZSYO 0.049 0.116 0.209 142 ± 2 144 ± 53 152 ± 15 torch flame. Unfortunately, the used experimental setup did not allow to monitor the actual temperature gradient on the sample surface during the experiment. It can be also noticed that the adjusted ratio of oxygen to acetylene flux was set to 1.8. This is following Miller-Oana et al. [58], who determined the volumetric flow rate (VFR) ratio as the ratio of oxygen to acetylene gas flow. They found out that increasing of VFR up to 1.7 resulted in a more oxygen rich flame. This led to the increase local pO2 as well as an increase of the ablation rate of graphite. In our study, even higher VFR values of 1.8 were used indicating oxygen rich conditions and high ablation rate present during testing. The specific mass change ablation rates and the thickness of the oxidized layers in the centre of the samples were measured for quantitative assessment of the oxidation performance. The results are summarized in Table 4 and will be discussed later in the text. In general, ZS´s mass ablation rate was less than a half of ZSYB´s and less than a quarter of ZSYB´s. The oxidized layers measured in the ablation centre were comparable for all composites. Although considering slight deviations found, the most inhomogeneous thickness was measured for ZSYB samples. Fig. 13 shows the surface morphology evolution of ZS samples after exposure to the oxyacetylene torch test. A detailed view is given on labelled positions where XRD measurements were taken. Monoclinic Fig. 13. Top surface morphology of ZS after exposure to oxyacetylene torch at 2000 °C for 60 s with enlarged view of regions 1, 2 and 3. 10 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Fig. 14. Cross-sectional view observed by light microscopy (a) and SEM image (b) with elemental mapping (c) of Zr, Y, Si and O of ZS centre after oxidation in oxyacetylene torch at 2000 °C for 60°seconds. Surface morphology of as received ZSYB and ZSYO samples after ablation test is shown in Fig. 15. The surface phase compositions were determined by XRD in relation to indicated positions and are summarized in Table S2. Generally, doping of ZS with YB4 and Y2O3 resulted in a formation of cubic ZrxY1-xO1.5+x/2 as the main phase independently on the distance from the tip of torch. The c-ZrxY1-xO1.5+x/2 was identified as the dominant phase in both materials in ablation centre (Fig. 15 - regions 1 and 5 for ZSYB and ZSYO, respectively). Detailed inspection of XRD patterns revealed the presence of yttria and zirconia solid solutions with various compositions, similar to what formed upon oxidation in static air. Most likely also rhombohedral solid solution (x = 0.82, PDF 00037-1307 was present in postions 1 and 2. In addition, trace amount of t-ZrO2 was found in ZSYO sample. Ablation centre of both samples appeared to be very dense with inhomogeneous topography, more significant for ZSYB (this corresponds to relatively high deviation of the layer thickness). Although the main phase identified by XRD was the same for both composites, they differ in minor phases found to be present toward the outer ablation area (shown for ZSYB in Fig. 16). In case of the ZSYB sample, 2 intermediate regions were apparent. XRD analysis showed the presence of different secondary phases. The t-ZrO2 and small amount of m-Y2Si2O7 were observed in region 2. The yttrium disilicate could be formed due to the reaction of silica with yttria or yttrium borate. The temperature here was obviously lower – silica and yttrium borate were not evaporated which allowed the formation of disilicate phase. In region 3, monoclinic and orthorhombic crystal systems of Y2Si2O7 were confirmed by XRD analysis. In addition, also glass (reaction 5 and 6) took place as well. ZrO2 grains reached size of 4.33 ± 1.29 μm. They possessed elongated shape indicating the direction of oxidation. The second layer was around 70 μm thick and can be described as ZrO2-based interlayer (2) infiltrated with silica glass. The grain size of zirconia grains decreased within this layer from 2.60 ± 0.72 μm to 1.84 ± 0.63 μm. Moreover, this layer was cracked due to the reasons mentioned above. Si-depleted layer (3) was clearly indicated by elemental mapping and observable also by light microscopy. This layer can be also described as ZrB2-based region where no diffused oxygen was observed. Below this there was the unreacted bulk material (4). In fact, compact ZrO2 outer layer acted like diffusion barrier and became protective against the proceeding oxidation as reported by Han et al. [59]. In this study, dense ZrO2 outer layer was formed at 2000 °C. SiC is not responsible for the improvement of oxidation resistance since volatile oxidation products are formed at present conditions. The equilibrium vapor pressure is good indication of a evaporation rate. Both B2O3 and SiO2 have relatively high vapor pressures, which means that they are volatile at elevated temperatures. ZrO2 has lower vapor pressure, it did not evaporate at present conditions and remained on the surface [59,60]. Generally, layered structure formed during oxidation of ZrB2-SiC composite under the oxyacetylene torch depend on testing conditions. The layer becomes porous when performed at higher temperatures. Volatile gaseous products degrade the oxide scale which leads to its rupture. [59,61,62]. The addition of YB4 and Y2O3 to ZS material resulted to a very similar appearance after the ablation test, thus quite different from ZS. 11 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Fig. 15. Comparison of top surface morphology of ZSYB and ZSYO after exposure to oxyacetylene torch at 2000 °C for 60°sec. all YBO3 reacted to the disilicate phase or evaporated. For ZSYO sample, the secondary phase composition was quite similar for the intermediate as well as outer ablation region (position is 6 and 7 in Fig. 15). Mainly Y2Si2O7 and ZrO2 were identified as minor phases, trace amount of m-ZrO2 and h-SiO2 were identified. Crystalline SiO2 was formed during rapid cooling of the sample. In addition to Y2Si2O7, also YBO3 was found in outer ablation region. This finding indicates that the temperature here was even lower (compared to region 3) – not Fig. 16. XRD patterns of ZSYB (left) and detail in the 20-40 2-Theta range (right). The peaks are shifted vertically. XRD measurements were taken in areas indicated in Fig. 15. 12 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Fig. 17. Cross-sectional view of centre of ZSYB observed by light microscopy (a) and detailed SEM image (b) after oxidation in oxyacetylene torch at 2000 °C for 60°seconds. higher operation temperatures. Tan et al. [65] studied the ablation resistance of ZrB2-SiC coatings modified with 10 mol% of Sm2O3 and Tm2O3. They observed quite similar surface morphologies as well as cross-sections. Zr0.8RE0.2O1.9 (RE = Sm or Tm) was identified as the primary phase after ablative testing, in analogy with this study. Inspection of cross-section revealed, that the ZSYB and ZSYO composites differ mostly in second layer, underlaying the protective ZrxY1xO1.5+x/2 layer. For ZSYB, the second layer was formed by above mentioned solid solution infiltrated with silica (Figure S8). In case of ZSYO, zirconia, silica and yttria were clearly observed in the mapping (Figure S9). Despite of high stability of yttria, the CTE mismatch, volatile species and zirconia phase transformations led to the detachment of layers as observed in Fig. 18. The formed oxide scale appears to be more damaged for ZSYO, it is obviously not coherent, and it is separating from the bulk material. though in different crystal systems, indicating the phase transformations related to the temperature gradient occured. Significantly higher specific mass ablation rate of ZSYO well agrees with the observed microstructure of samples, especially beyond the ablation centre. In fact, B2O3 and YBO3 are the primary gaseous product of oxidation and evaporated from the system readily due to the high vapor pressure. YB4 can be assessed as boron rich compound and its oxidation results in significantly higher quantity of volatile products compared to the Y2O3 (which has very low vapor pressure). This can be observed also in the microstructure shown in Figure 15 – 3 and 4 where extensive pores are present. Many small voids appearing on the surface provided more paths for the inward flow of oxygen and testify outward flow of gaseous ablation products oxidized beneath the outermost ablation layer which explains deeper degradation as well as lower specific mass ablation rate. On the contrary, ZSYO seemed to be more compact on the surface and less pores could be observed (see Figure 13 – 6 and 7). Analysis of the cross-section in the ablation centre clearly shows the layered structure of ZSYB and ZSYO, as shown in Figs. 17 and 18, respectively. Elemental mapping (see Figures S6 and S7) was performed in order to understand chemical differences between formed layers. Outer layer of both composites consists of Zr, Y and O, which is consistent with XRD analysis. The formed solid solution acted like the protective layer and endured on the surface when SiO2 already evaporated. This was confirmed also by elemental mapping since no Si was found in the outer layer in the ablation centre, similar as for ZS sample. Moreover, Y2O3 is reported to have even lower vapor pressure than ZrO2 [63]. Yttria is more stable and should remain on the surface even when zirconia already evaporates. Moreover, thermodynamic data published in this study for Y2O3.2(ZrO2) shows also very low vapor pressure of this system. It has also been reported as the material with the highest maximum use temperature of ∼ 2300 °C [64]. Formation of identified solid solution could be very beneficial, especially at even 4. Conclusion The influence of YB4 and Y2O3 on sintering, mechanical properties and high temperature oxidation up to 2000 °C of of ultra-high temperature ZrB2– 20 wt% SiC (ZS) composite prepared by hot-pressing was investigated. Y-containing additives were selected to improve the oxidation performance of ZrB2-SiC composites and stabilize the forming oxide scale preventing volume changes upon cooling. The content of YB4 and Y2O3 was 14.01 wt% and 12.22 wt°, respectively, which correspond to 8 mol% of Y2O3 present in system after oxidation of composites. The YB4 showed beneficial impact on densification, supressed the grain growth and slightly improved the hardness and Young´s modulus. The Y2O3 addition resulted to full densification and also suppressed the grain growth, as expected form the literature. However, the mechanical properties were inferior to ZS material, most probably due to the relatively high content used. Fig. 18. Cross-sectional image of centre of ZSYO after oxidation in oxyacetylene torch at 2000 °C for 60°seconds. 13 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. Static oxidation (tested up to temperature of 1650 °C) conditions led to the formation of a typical layer structure of ZS composite with protective silica layer on the surface. However, exposure to ablation test at 2000 °C caused evaporation and elimination of silica in ablation centre. The protective function was undertaken by an outer zirconia layer. Generally, the oxidation resistance of materials mixed with YB4 as well as Y2O3 tested in static air was inferior compared to basic ZS composite, especially at oxidation temperatures above 1300 °C. This is attributed mainly to the formation of porous and not compact oxidation layer; a high content of boria evaporating from the surface, forming voids allowing deeper diffusion of oxygen into bulk and CTE mismatch between present compounds. Although the YB4 as well as Y2O3 did not improved the oxidation resistance during oxidation in static air, the performance during oxidation in dynamic conditions indicates different results. During ablation at 2000 °C, a dense cubic solid solution of ZrxY1-xO1.5+x/2 was formed as the main oxidation product. This compound is very promising to protect the material at even higher temperatures when zirconia already shears. Moreover, the oxidation layer of YB4 containing composites appeared to be less damaged than for materials prepared with Y2O3. Considering the oxidation in static air, Y-based additives should be decreased to have sufficient content of protective silica glass formed during oxidation. However, for oxidation at temperature exceeding 2000 °C, silica is no more present in outer area and therefore does not function as protective layer. In this case, yttria based oxidation products are more stable under severe ablation conditions and are promising candidates for ultra-high temperature applications. Moreover, yttria has even lower vapor pressure compared to zirconia. It could protect the material at higher temperatures when zirconia is not effective anymore. [3] J.F. Justin, A. Jankowiak, Ultra high temperature ceramics: densification, properties and thermal stability, AerospaceLab J. 3 (2011) 1–11. [4] W.G. Fahrenholtz, G.E. Hilmas, Refractory diborides of zirconium and hafnium, J. Am. Ceram. Soc. 90 (5) (2007) 1347–1364, https://doi.org/10.1111/j.1551-2916. 2007.01583.x. [5] A. Bellosi, S. Guicciardi, V. Medri, F. Monteverde, D. Sciti, L. Silvestroni, Processing and properties of ultra-refractory composites based on Zr- and Hf-borides: State of the art and perspectives, in: N. Orlovskaya, M. Lugovy (Eds.), Boron Rich Solids: Sensors, Ultra High Temperature Ceramics, Thermoelectrics, Armor, Chapter in NATO Science for Peace and Security Series B: Physics and Biophysics, Springer Science + Business Media B.V., 2011, pp. 147–160, , https://doi.org/10.1007/97890-481-9818-4_10. [6] J. Han, P. Hu, X. Zhang, S. Meng, W. Han, Oxidation-resistant ZrB2-SiC composites at 2200°C, Compos. Sci. Technol. 68 (2008) 799–806, https://doi.org/10.1016/j. compscitech.2007.08.017. [7] W.G. Fahrenholtz, G.E. Hilmas, Oxidation of ultra-high temperature transition metal diboride ceramics, Int. Mater. Rev. 57 (1) (2012) 61–72, https://doi.org/10. 1179/1743280411Y.0000000012. [8] W.-M. Guo, G.-J. Zhang, Oxidation resistance and strength retention of ZrB2-SiC ceramics, J. Eur. Ceram. Soc. 30 (11) (2010) 2387–2395, https://doi.org/10.1016/ j.jeurceramsoc.2010.01.028. [9] R. Loehman, E. Corral, H.P. Dumm, P. Kotula, R. Tandon, Ultra High Temperature Ceramics for Hypersonic Vehicle Application, Sandia report, 2006, pp. 1–46. [10] R. Inoue, Y. Arai, Y. Kubota, Oxidation behaviors of ZrB2-SiC composites above 2000°C, Ceram. Int. 43 (2017) 8087–8088, https://doi.org/10.1016/j.ceramint. 2017.03.129. [11] M. Mallik, K.K. Ray, R. Mitra, Oxidation behaviour of hot-pressed ZrB2-SiC and HfB2-SiC composites, J. Eur. Ceram. Soc. 31 (2011) 199–215, https://doi.org/10. 1016/j.jeurceramsoc.2010.08.018. [12] W.G. Fahrenholtz, Thermodynamic analysis of ZrB2-SiC oxidation: formation of a SiC-depleted region, J. Am. Ceram. Soc. 90 (1) (2007) 143–148, https://doi.org/10. 1111/j.1551-2916.2006.01329.x. [13] J. Li, T.J. Lenosky, C. Först, S. Yip, Thermomechanical and mechanical stabilities of the oxide scale of ZrB2+SiC and oxygen transport mechenisms, J. Am. Ceram. Soc. 91 (5) (2008) 1475–1480, https://doi.org/10.1111/j.1551-2916.2008.02319.x. [14] Z. Kováčová, Ľ. Bača, E. Neubauer, M. Kitzmantel, Influence of sintering temperature, SiC particle size and Y2O3 addition on the densification, microstructure and oxidation resistance of ZrB2-SiC ceramics, J. Eur. Ceram. Soc. 36 (2016) 3041–3049, https://doi.org/10.1016/j.jeurceramsoc.2015.12.028. [15] W.M. Guo, J. Vleugels, G.-J. Zhang, P.-L. Wang, O. Van Der Biest, Effect of Re2O3 (Re = La, Nd, Y and Yb) addition in hot-pressed ZrB2-SiC ceramics, J. Eur. Ceram. Soc. 29 (2009) 3063–3068, https://doi.org/10.1016/j.jeurceramsoc.2009.04.021. [16] X. Zhang, X. Li, J. Han, W. Han, Ch Hong, Effects of Y2O3 on microstructure and mechanical properties of ZrB2-SiC ceramics, J. Alloys. Compd. 465 (2008) 506–511, https://doi.org/10.1016/j.jallcom.2007.10.137. [17] W. Li, X. Zhang, Ch Hong, J. Han, W. HAN, Hot-pressed ZrB2-SiC-YSZ composites with various yttria content: microstructure and mechanical properties, Mater. Sci. Eng. 494 (2008) 147–152, https://doi.org/10.1016/j.msea.2008.04.010. [18] P. Tatarko, M. Kašiarová, J. Dusza, J. Morgiel, P. Šajgalík, P. HVIZDOŠ, Wear resistance of hot-pressed Si3N4/SiC Micro/Nanocomposites sintered with rare-earth oxide additives, Wear 269 (2010) 867–874, https://doi.org/10.1016/j.wear.2010. 08.020. [19] S. Lojanová, P. Tatarko, Z. Cchlup, M. Hhatko, J. Dusza, Z. Lenčeš, P. Šajgalík, Rareearth element doped Si3N4/SiC Micro/Nano-composites – RT and HT mechanical properties, J. Eur. Ceram. Soc. 30 (2010) 1931–1944, https://doi.org/10.1016/j. jeurceramsoc.2010.03.007. [20] D.D. Jayaseelan, E. Zapata-Solvas, R.J. Chater, W.E. Lee, Structural and compositional analyses of oxidized layers of ZrB2-based UHTCs, J. Eur. Ceram. Soc. 35 (2015) 4059–4071, https://doi.org/10.1016/j.jeurceramsoc.2015.07.026. [21] M.M. Opeka, I.G. Talmy, J.A. Zaykoski, Oxidation-based materials selection for 2000°C+ hypersonic aerosurfaces: theoretical considerations and historical experience, J. Mater. Sci. 39 (2004) 5887–5904. [22] F. Monteverde, A. Bellosi, Oxidation of ZrB2-based ceramics in dry air, J. Electrochem. Soc. 150 (11) (2009) B552–B559, https://doi.org/10.1149/1. 1618226. [23] Q. Guo, H. Xiang, X. Sun, X. Wang, Y. Zhou, Preparation of porous YB4 ceramics using a combination of in-situ borothermal reaction and high temperature partial sintering, J. Eur. Ceram. Soc. 35 (2015) 3411–3418, https://doi.org/10.1016/j. jeurceramsoc.2015.05.023. [24] J. Li, A. Peng, Y. He, H. Yuan, Q. Guo, Q. She, L. Zhang, Synthesis of pure YB4 powder via the reaction of Y2O3 with B4C, J. Am. Ceram. Soc. (2012) 2127–2129, https://doi.org/10.1111/j.1551-2916.2012.05272.x. [25] J.A. Zaykoski, M.M. Opeka, L.H. Smith, I.G. Talmy, Synthesis and characterization of YB4 ceramics, J. Am. Ceram. Soc. 94 (11) (2011) 4059–4065, https://doi.org/10. 1111/j.1551-2916.2011.04676.x. [26] R.F. Giese Jr., V.I. Matkovich, J. Economy, The crystal structure of YB4, Z. Fã¼r Krist. 122 (1965) 423–432. [27] J. Günster, T. Tanaka, R. Souda, Surface reorganization observed on YB4(001), Phys. Rev. B 56 (24) (1997) 15962–15966. [28] G.A. Kudintseva, G.M. Kuznetsova, V.P. Bondarenko, N.F. Selivanova, V.Ya. Shlyuko, Preparation and emissive properties of some Yttrium and Gadolinium Borides, PoroshkovayaMetallurgiya 2 (1968) 45–53. [29] B. Jäger, S. Paluch, W. Wolf, Characterization of the electronic properties of YB4 and YB6 using 11B NMR and first-principle calculations, J. Alloys. Compd. 383 (2004) 232–238, https://doi.org/10.1016/j.jallcom.2004.04.067. [30] A. Waskowska, L. Gerward, J. Staun Olsen, K. Ramesh Babu, G. Vaitheeswaran, Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements This work was carried out within the dissertation project investigating ceramic materials: Boride-based composites for high-temperature applications. This work was supported by the Slovak Grant Agency VEGA Project No. 2/0152/18, MVTS – ULTRACOM project (in the framework of FLAG ERA II), and APVV-15-0540. This study was performed during the implementation of the project Building-up Centre for advanced materials application of the Slovak Academy of Sciences, ITMS project code 313021T081 supported by Research & Innovation Operational Programme funded by the ERDF. The authors wish to thank Mr. Emanuel Feuerstein and his colleagues at RHP for technical support. We thank the university staff for support in measurements and analysis. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jeurceramsoc.2020.03. 060. References [1] E. Wuchina, E. Opila, M. Opeka, W. Fahrenholtz, I. Talmy, UHTCs: Ultra-High Temperature Ceramic Materials for Extreme Environment Applications, The Electrochemical Society Interface, Winter, 2007, pp. 30–36. [2] Eakins, E. Jayaseelan, D. D, W.E.L. Lee, Toward oxidation-resistant ZrB2-SiC ultrahigh temperature ceramics, Minerals Met. Mater. Soc. ASM Int. 42A (2011) 878–887, https://doi.org/10.1007/s11661-010-0540-8. 14 Journal of the European Ceramic Society xxx (xxxx) xxx–xxx Z. Kováčová, et al. [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] V. Kanchana, A. Svane, V.B. Filipov, G. Levchenko, A. Lyaschenko, Thermoelastic properties of ScB2, TiB2, YB4 and HoB4: exerimental and theoretical studies, Acta Mater. 59 (2001) 4886–4894, https://doi.org/10.1016/j.actamat.2011.04.030 2011. Y.-Y. Fu, Y.-W. Li, H.-M. Huang, Elastic and Dynamical Properties of YB4: FirstPrinciples Study, Chinese Phys. Lett. 31 (11) (2014), https://doi.org/10.1088/ 0256-307X/31/11/116201 p. 116201 – 1-4. Gmelin Handbook of Inorganic and Organometallic Chemistry, 8th ed., Springer – Verlag, 2020 ISBN 0-387-93604-1. R.M. Menelis, G.A. Meerson, N.N. Zhuravlev, T.M. Telyukova, A.A. Stepanova, N.V. Gramm, Vacuum.-Thermal preparation of yttrium and gadolinium borides, and some of their properties, Poroshkovaya Metallurgiya 11 (1966) 77–84. Z. Kováčová, Ľ. Bača, E. Neubauer, Ľ. Orovčík, M. Kitzmantel, M. Vozárova, Synthesis and reaction sintering of YB4 ceramics, Ceram. Int. 45 (2019) 18795–18802, https://doi.org/10.1016/j.ceramint.2019.06.108. X. Zhang, Z. Chen, X. Xiong, R. Liu, Y. Zhu, Morphology and microstructure of ZrB2SiC ceramics after ablation at 3000 °C by oxy-acetylene torch, Ceram. Int. 42 (2016) 2798–2805, https://doi.org/10.1016/j.ceramint.2015.11.012. I. Akin, M. Hotta, F.C. Sahin, O. Yucel, G. Goller, T. Goto, Microstructure and densification of ZrB2-SiC composites prepared by spark plasma sintering, J. Eur. Ceram. Soc. 29 (2009) 2379–2385, https://doi.org/10.1016/j.jeurceramsoc.2009. 01.011. D. Ćorić, M.M. Renjo, L. Ćurković, Vickers indentation fracture toughness of Y.TZP dental ceramics, Int. J. Refractory Met. Hard Mater. 64 (2017) 14–19, https://doi. org/10.1016/j.ijrmhm.2016.12.016. D. Chicot, A. Pertuz, F. Roudet, M.H. Staia, J. Lesage, New developments for fracture toughness determination by Vickers indentation, Mater. Sci. Technol. 20 (1985) 1–8, https://doi.org/10.1179/026708304225017427. D. Ćorić, M.M. Renjo, I. Žmak, Critical evaluation of indentation fracture toughness measurements with Vickers indenter on yttria-stabilized zirconia dental ceramics, Mater. Sci. Eng.Technol. 48 (2017) 767–772, https://doi.org/10.1002/mawe. 201700026. F. Sergejev, M. Antonov, Comparative study on indentation fracture toughness measurements of cemented carbides, Proceed. Estonian Acad. Sci. Eng. 12 (4) (2006) 388–398. K. Niihara, R. Morena, D.P.H. Hassekman, Evaluation of KIc of brittle solids by the indentation method with low crack-to-indent ratios, J. Mater. Sci. Lett. 1 (1982) 13–16. K. Niihara, A fracture mechanics analysis of indentation-induced Palmqvist crack in ceramics, J. Mater. Sci. Lett. 2 (1983) 221–223. E. Rudnayová, J. Dusza, M. Kupková, Comparison of fracture toughness measuring methods applied on silicon nitride ceramics, J. de Physique III 3 (1993) 1273–1276. M. Mallik, S. Pan, H. Roy, Fracture toughness measurement of hot pressed ZrB2MoSi2 composite, Int. J. Curr. Eng. Technol. 3 (5) (2013) 1647–1652, https://doi. org/10.1016/j.jeurceramsoc.2008.01.003. C.B. Ponton, R.D. Rawlings, Vickers indentation fracture toughness test, Part 2, Application and critical evaluation of standardised indentation toughness equations, Mater. Sci. Technol. 5 (1989) 961–976. S. Zhu, W.G. Fahrenholtz, G.E. Hilmas, Influence of silicon carbide particle size on the microstructure and mechanical properties of zirconium diboride-silicon carbide ceramics, J. Eur. Ceram. Soc. 27 (2007) 2077–2083, https://doi.org/10.1016/j. jeurceramsoc.2006.07.003. A.L. Chamberlain, W.G. Fahrenholtz, G.E. Hilmas, High-strength zirconium diboride-based ceramics, J. Am. Ceram. Soc. 87 (6) (2004) 1170–1172. A. Rezaie, W.G. Fahrenholtz, G.E. Hilmas, Effect of hot pressing time and temperature on the microstructure and mechanical properties of ZrB2-SiC, J. Mater. Sci. 42 (8) (2007) 2735–2744, https://doi.org/10.1007/s10853-006-1274-2. M. Lugovy, V. Slyunyayev, N. Orlovskaya, E. Mitrentsis, C.G. Aneziris, T. Graule, [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] 15 J. Kuebler, Temperature dependence of elastic properties of ZrB2-SiC composites, Ceram. Int. 42 (2016) 2439–2445, https://doi.org/10.1016/j.ceramint.2015.10. 044. R. Stadelmann, M. Lugovy, N. Orlovskaya, P. Mchaffey, M. Radovic, V.M. Sglavo, S. Grasso, M.J. Reece, Mechanical properties and residual stresses in ZrB2-SiC spark plasma sintered ceramic composites, J. Eur. Ceram. Soc. 36 (2016) 1527–1537, https://doi.org/10.1016/j.jeurceramsoc.2016.01.009. B. Nayebi, M.S. Asl, M.G. Kakroudi, M. Shokouhimehr, Temperature dependence of microstructure evolution during hot pressing of ZrB2-30 vol.% SiC composites, Int. J. Refract. Metals Hard Mater. 54 (2016) 7–13, https://doi.org/10.1016/j.ijrmhm. 2015.06.017. Y. Zhao, L.-J. Wang, G.-J. Zhang, W. Jiang, L.-D.E. Chen, Effect of holding time and pressure on properties of ZrB2-SiC composite fabricated by the spark plasma sintering reactive synthesis method, Int. J. Refract. Metals Hard Mater. 27 (2009) 177–180, https://doi.org/10.1016/j.ijrmhm.2008.02.003. S.-Q. Guo, Densification of ZrB2-based composites and their mechanical and physical properties: A review, J. Eur. Ceram. Soc. 29 (2009) 995–1011, https://doi.org/ 10.1016/j.jeurceramsoc.2008.11.008. G. Dhanaraj, K. Byrappa, V. Prasad, M. Dudley, Handbook of Crystal Growth, Springer – Verlag, 2020 ISBN: 978-3-540-74182-4. R.J. Bratton, Defect structure of Y2O3-ZrO2 solid solutions, J. Am. Ceram. Soc. (1969) 213. A.I. Becerro, A. Escudero, Revision of the crystallographic data of polymorphic Y2Si2O7 and Y2SiO5 compounds, Phase Transit. 77 (12) (2004) 1093–1102, https:// doi.org/10.1080/01411590412331282814. J. Parmentier, P.R. Bodart, L. Audoin, G. Massouras, D. Thompson, R.K. Harris, Goursat, J.-L. Besson, Phase transformations in gel-derived and mixed-powder-derived yttrium disilicate, Y2Si2O7, by X-ray diffraction and 29Si MAS NMR, J. Solid State Chem. 149 (2000) 16–20. M. Miller-Oana, P. Neff, M. Valdez, A. Powell, M. Packard, L.S. Walker, E.L. Corral, Oxidation behaviour of aerospace materials in high enthalpy flows using an oxyacetylene torch facility, J. Am. Ceram. Soc. 98 (4) (2015) 1300–1307, https://doi. org/10.1111/jace.13462. J. Han, P. Hu, X. Zhanh, S. Meng, W. Han, Oxidation-resistant ZrB2-SiC composites at 2200°C, Compos. Sci. Technol. 68 (2008) 799–806, https://doi.org/10.1016/j. compscitech.2007.08.017. X. Zhang, P. Hu, J. Han, S. Meng, Ablation behaviour of ZrB2-SiC ultra high temperature ceramics under simulated atmospheric re-entry conditions, Compos. Sci. Technol. 68 (2008) 1718–1726, https://doi.org/10.1016/j.compscitech.2008.02. 009. X. Zhang, Z. Chen, X. Xiong, R. Liu, Y. Zhu, Morphology and microstructure of ZrB2SiC ceramics after ablation at 3000°C by oxy-acetylene torch, Ceramic Int. 42 (2016) 2798–2805, https://doi.org/10.1016/j.ceramint.2015.11.012. M. Mallik, A.J. Kailath, K.K. Ray, R. Mitra, Effect of SiC content on electrical, thermal and ablative properties of pressureless sintered ZrB2-based ultrahigh temperature ceramic composites, J. Eur. Ceram. Soc. 37 (2017) 559–572, https://doi. org/10.1016/j.jeurceramsoc.2016.09.024. N.S. Jacobson, Thermodynamic Properties of Metal Oxide-zirconia Systems, Report No. NASA TM-102351 NASA Technical Memorandum, 1989, p. 102351 https:// ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19900004350.pdf. N.S. Jacobson, High-temperature Durability Considerations for HSCT Combustor, Report No. NASA TP-3162 NASA Technical Paper, 1992, p. 3162 https://books. google.sk/books?id=iZgucIZ3RfIC&printsec=frontcover&hl=sk&source=gbs_ge_ summary_r&cad=0#v=onepage&q&f=false. W. Tan, M. Adduci, R. Trice, Evaluation of rare-earth modified ZrB2-SiC ablation resistance using an oxyacetylene torch, J. Am. Ceram. Soc. 97 (8) (2014) 2639–2645, https://doi.org/10.1111/jace.12991.