See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/316460250 Sintering behavior of ZrB2–SiC composites doped with Si3N4: A fractographical approach Article in Ceramics International · April 2017 DOI: 10.1016/j.ceramint.2017.04.144 CITATIONS READS 0 62 5 authors, including: Behzad Nayebi Mehdi Shahedi Asl Amirkabir University of Technology University of Mohaghegh Ardabili 37 PUBLICATIONS 158 CITATIONS 92 PUBLICATIONS 411 CITATIONS SEE PROFILE SEE PROFILE Mahdi Ghassemi Kakroudi Iman Farahbakhsh University of Tabriz Islamic Azad University, Quchan Branch 63 PUBLICATIONS 477 CITATIONS 36 PUBLICATIONS 61 CITATIONS SEE PROFILE SEE PROFILE Some of the authors of this publication are also working on these related projects: Spark plasma sintering of titanium matrix composites View project Taguchi design of experiments for optimization of processing parameters in ZrB2-based UHTCs View project All content following this page was uploaded by Mehdi Shahedi Asl on 04 June 2017. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. Ceramics International 43 (2017) 9699–9708 Contents lists available at ScienceDirect Ceramics International journal homepage: www.elsevier.com/locate/ceramint Sintering behavior of ZrB2–SiC composites doped with Si3N4: A fractographical approach ⁎ MARK ⁎ Zohre Ahmadia, Behzad Nayebib, , Mehdi Shahedi Aslc, Mahdi Ghassemi Kakroudid, , Iman Farahbakhshe a Young Researchers and Elite Club, Miyaneh Branch, Islamic Azad University, Miyaneh, Iran Young Researchers and Elite Club, Central Tehran Branch, Islamic Azad University, Tehran, Iran c Department of Mechanical Engineering, University of Mohaghegh Ardabili, Ardabil, Iran d Department of Materials Science and Engineering, University of Tabriz, Tabriz, Iran e Department of Mechanical Engineering, Quchan Branch, Islamic Azad University, Quchan, Iran b A R T I C L E I N F O A BS T RAC T Keywords: Silicon nitride Hot pressing Pressureless sintering Zirconium diboride Silicon carbide Fractograph ZrB2–SiC composite ceramics were doped with 0, 1, 3 and 5 wt% Si3N4 plus 1.6 wt% carbon (pyrolized phenolic resin) as sintering aids and fabricated by hot pressing process under a relatively low pressure of 10 MPa at 1900 °C for 2 h. For a comparative study, similar ceramic compositions were also prepared by pressureless sintering route in the same processing conditions, with no applied external pressure. The effect of silicon nitride dopant on the microstructural evolution and sintering process of such ceramic composites was investigated by a fractographical approach as well as a thermodynamical analysis. The relative density increased by the addition of Si3N4 in hot pressed samples as a fully dense composite was achieved by adding 5 wt% silicon nitride. A reverse trend was observed in pressureless sintered composites and the relative density values decreased by further addition of Si3N4, due to the formation of gaseous products which resulted in the entrapment of more porosities in the final structure. The formation of ZrC phases in pressureless sintered samples and layered BN structures in hot pressed ceramics was detected by HRXRD method and discussed by fractographical SEM-EDS as well as thermodynamical analyses. 1. Introduction Ultrahigh temperature ceramics (UHTCs) have been remarkably attended recently, due to their thermal stability which can be considered in thermal protection systems of hypersonic vehicles and next generation aerospace applications. For example, leading edges of a spacecraft are exposed to high temperatures and intense thermal shocks in both neutral and oxidizing environments. Such extreme conditions demand materials with high oxidation resistance, low creep resistance, and excellent thermal shock properties. Zirconium diboride (ZrB2) based UHTCs are potentially considered for the mentioned applications during last decades. However, there are four main limiting factors which challenge the applicability of monolithic ZrB2: relatively poor mechanical properties (such as strength and fracture toughness), low sinterability (derived by strong covalent bonds), low self-diffusion coefficient, and unfavorable oxidation resistance at elevated temperatures [1–8]. Several research works have been carried out in order to improve the properties of ZrB2-based materials. Most of these approaches are focused on using reinforcement phases such as silicon carbide (SiC ⁎ [9–11]), boron carbide (B4C [12–14]) and zirconium carbide (ZrC [4,6]) to compensate the weaknesses of ZrB2. It has been shown that SiC reinforcements not only enhance the oxidation resistance and mechanical properties of ZrB2-based materials, but also improve their sinterability and microstructural properties. Therefore, ZrB2–SiC composites have drawn the attention of researchers; hence, several fabrication methods have been developed to achieve favorable properties. Although most of the proposed techniques for consolidating ZrB2– SiC composites are based on hot pressing (HP), pressureless sintering (PS) is also considered in order to fabricate complex-shaped products. Despite the fact that HP process is the most applicable method to achieve fully-dense composites, fabrication of complex-shaped components requires consequent costly and time-consuming machining processes. In contrast to HP method, PS techniques offer advantages such as manufacturing near-net-shape components, which decreases the finishing costs. However, it is rather difficult to attain fully-dense ZrB2-based ceramics by PS process. Therefore, applying higher sintering temperatures and/or sintering aids are required to overcome the intrinsic low sinterability of ZrB2 [15–17]. Corresponding authors. E-mail addresses: behzad.nayebi@aut.ac.ir (B. Nayebi), mg_kakroudi@tabrizu.ac.ir (M. Ghassemi Kakroudi). http://dx.doi.org/10.1016/j.ceramint.2017.04.144 Received 26 March 2017; Received in revised form 23 April 2017; Accepted 25 April 2017 Available online 26 April 2017 0272-8842/ © 2017 Elsevier Ltd and Techna Group S.r.l. All rights reserved. Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. nitride additives in densification and sintering process of ZrB2–SiC composites. Briefly, the sintering process is reported to be accomplished at 1900 °C, by adding HfN or ZrN to ZrB2–SiC system [20,23,24]. Using BN as a sintering aid, greatly improved the densification of ZrB2–SiC composites through removing the surface oxide impurities of ZrB2 particles such as ZrO2 [25,26]. Recently published research project has indicated that AlN activated the liquid phase sintering mechanism as a result of formation of a glassy phase which fills the voids between particles and decreases the porosity content of the composite. It is also indicated that AlN dopant can decrease the sintering temperature and lead to a fine-grained microstructure [21,22]. The addition of β-SiAlON is reported to result in fully-dense ZrB2–SiC composites manufactured by hot pressing process [37]. Although there are few publications on the effects of Si3N4 sintering aid on the densification behavior of ZrB2–SiC composites [27,39–45], it can be concluded that silicon nitride significantly improves the densification of ZrB2. Most of the conducted research projects in this area are dealing with hot pressing of Si3N4-doped ZrB2-based materials. To the best of our knowledge, there is no comprehensive study which provides a comparative criteria between HP and PS processes. This work, as the 3rd part of a comprehensive research project about the effects of different nitride additives, deals with the improvements in sinterability, microstructure and densification behavior of hot pressed and pressureless sintered ZrB2–SiC composites. The 1st and 2nd parts of this project have been already published and allotted to SiAlON [37] and AlN [22] additives, respectively. 2. Experimental Fig. 1. (a) SEM micrograph and (b) XRD pattern of as-received Si3N4 powder. Commercially available ZrB2 powder (average particle size ~2 µm, purity > 99%, Leung Hi-tech Co., China), α-SiC powder (average particle size ~5 µm, purity > 99%, Carborundum Universal Limited, India) and Si3N4 powder (0.6 mm, α-phase, FCP 15 C, Sika Tech., Norway) were used as raw materials. Fig. 1 shows the scanning electron microscopy (SEM) image and X-ray diffraction pattern (XRD) of the silicon nitride powder. Such characteristic features of the starting ZrB2 and SiC powders were presented elsewhere [37]. Phenolic resin (Resol. 800, Iran Polymer and Petrochemical Institute, Iran) was added as a binder for the green parts. Upon heating, it pyrolyzed to 4 wt% carbon which may participate as another sintering aid. Certain amounts of ZrB2 powder plus 30 vol% SiC, and 4 wt% phenolic resin as well as 0, 1, 3 or 5 wt% Si3N4 were ball mixed at 90 rpm for 60 min in the Teflon cup using zirconia balls and ethanol. Samples were then labeled regarding the coding system presented in Table 1, which includes the Si3N4 content and the processing technique. The slurry was dried in a rotating evaporator (Tebazma HMS 14, Iran) at 90 °C to remove the ethanol and minimize the particles’ agglomeration. Before sintering, the dried mixtures were sieved through a 100-mesh metallic sieve to obtain uniform granules. High resolution X-ray diffraction pattern (HRXRD) of the powder mixture containing 5 wt% Si3N4 is presented in Fig. 2. It is well known that using extra-high sintering temperatures ( > 2000 °C) although facilitates the densification process, but leads to undesirable grain growth in both HP and PS processes. Fanatic grain growth during the sintering process has been introduced as the main cause of poor mechanical and microstructural characteristics of the final products [9,18]. Higher sintering temperatures also evoke more production costs. Therefore, it is highly desirable to perform the sintering process in lower temperatures [19]. Hence, using the appropriate sintering aids is attended as a key factor in achieving economical and industrial products. Several sintering additives have been reported to improve the densification behavior of ZrB2-based ceramics. Such additives can cooperate in the sintering process of the composites through activating several densification mechanisms including liquid phase sintering, reactive sintering, etc. Based on the literatures, the most common additives used in ZrB2–SiC system include: oxides (ZrO2 [4,6] and Re2O3: Re=La, Nd, Y, Yb [20]), nitrides (AlN [21,22], ZrN [23], HfN [24], BN [25,26] and Si3N4 [20,27]), disilicides (MoSi2, TaSi2, and ZrSi2 [20,28]) and carbides (B4C [12,13], WC and VC [29]). There are also several reports about using carbon sources (graphite [30,31], graphene [32], carbon fiber [33–35], carbon nanotube [3] and carbon black [36]) as well as other additives (SiAlON [37] and metals [38]), as sintering aids. Numerous research works have indicated the effective role of Table 1 Coding system of the samples. Code Processing technique Amount of Si3N4 (wt%) PS0 HP0 PS1 HP1 PS3 HP3 PS5 HP5 Pressureless sintering Hot pressing Pressureless sintering Hot pressing Pressureless sintering Hot pressing Pressureless sintering Hot pressing 0 0 1 1 3 3 5 5 Fig. 2. HRXRD pattern of the starting powder mixture containing 5 wt% silicon nitride. 9700 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Iran) at 400 MPa for 3 min. The green samples were then pressurelessly sintered at the same temperature and a dwelling time of the HP samples (1900 °C, 120 min). The PS samples were heated up in a resistance-heated graphite element furnace (Shenyang Weitai Science & Technology Development Co., Ltd., China). Finally, the sintered samples were cooled down naturally with an average cooling rate of ~5 °C/min to the room temperature. 2.3. Characterization Fig. 3. Relative densities of HP and PS samples. The starting materials and the powder mixtures were characterized using X-ray diffractometer (XRD: Philips PW1800) as the X-ray generated by a Cu lamp (λ=1.54 Å) was operated at 40 kV and 30 mA. Sintered samples were also analyzed using high-resolution XRD apparatus (HRXRD: Bruker Discover D8 X-Ray Diffraction analysis system, Cu lamp, λ=1.54056 Å, maximum detecting depth: 300 nm). The bulk densities of specimens were measured by the Archimedes method. The measured bulk density was divided by the theoretical density to obtain the relative density. The theoretical density was estimated using the rule of mixtures. The microstructural and fractographical investigations of the samples were carried out using a field emission scanning electron microscope (FESEM: Mira3 Tescan, Czech Republic). Chemical analysis was carried out simultaneously by an energy dispersive spectroscope (EDS: DXP–X10 P Digital X–Ray Processor). Thermodynamical calculations were carried out via HSC Chemistry software (ver. 5.11, Outokumpu Research Oy, Pori, Finland). 2.1. Sintering process of HP samples A set of milled powders was fabricated by hot pressing in a graphite die (pellet shape; diameter: 20 mm; thickness: 5 mm) lined with boron nitride to protect the die from reacting with the powder mixture. The samples were initially heated up to 900 °C at 20 °C/min, in order the phenolic resin to get pyrolized. Then, the samples were hot pressed (Shenyang Weitai Science & Technology Development Co., Ltd., China, vacuum: 5×10−2 Pa) at 1900 °C under a pressure of 10 MPa for 120 min. Finally, the pressure was released and the samples were cooled down to the room temperature (cooling rate: ~5 °C/min). 2.2. Sintering process of PS samples Another set of milled powders was initially compressed using a biaxial cold press machine (CP-B35T, Tajhiz Ceram Engineering Co., Fig. 4. (a) Secondary electron SEM micrograph of the fracture surface of sample PS1, (b) Back scattered electron micrograph of a formed phase at interface of ZrB2/SiC particles in sample PS1 and (c) EDS spectrum of the mentioned phase. 9701 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Fig. 5. (a) Secondary electron SEM micrograph of the fracture surface of sample PS3, (b) Back scattered electron micrograph of a formed phase at interface of ZrB2/SiC particles in sample PS3 and (c) EDS spectrum of the mentioned phase. PS1 sample are presented in Fig. 4. Although the microstructural features of the fracture surface in PS1 sample reveal the progression of sintering process (e.g. neck formations indicated by thick marker in Fig. 4a), it is clear that several porosities are still present in the composite. The fractograph of PS1 sample shows a mixed inter/intragranular mode (thin markers in Fig. 4a), which endorses a limited sintering phenomenon, particularly in SiC/ZrB2 interfaces. It seems that most intra-granular fractures occurred in SiC particles and SiC/ ZrB2 interfaces; whereas, most of the neighbor ZrB2 particles did not form strong interfaces and were fractured inter-granularly. Fig. 4b shows the presence of an interfacial particle which seems to be formed during the sintering process. Related EDS spectrum of the mentioned particle can be found in Fig. 4c. According to the chemical composition from EDS analysis, the mentioned phase can be potentially characterized as zirconium or boron carbides (ZrC or B4C). Low and high magnification SEM micrographs of the fracture surface of PS3 sample are presented in Fig. 5a and b, respectively. Compared to PS1 sample, no remarkable progression in porosity removal is observed in PS3 sample. Although it appears that the fracture mode of PS3 sample is similar to PS1 sample, some regions with intragranular fracture mode are also seen in ZrB2 particles (marked area in Fig. 5a). As it can be concluded from the EDS spectrum of the interfacial phase in Fig. 5b (Fig. 5c), the chemical composition of the mentioned phases is approximately similar to that of PS1 sample. Such phases contained noticeable amounts of carbon, 3. Results and discussion Fig. 3 shows the relative densities of both HP and PS samples. As it can be clearly seen, hot pressing process led to the fabrication of fully dense composites with 5 wt% silicon nitride (HP5 sample), whereas maximum relative density in PS samples was achieved in additive free (PS0) sample. It seems that although more Si3N4 additive resulted in improved densification in HP samples, it had an inverse effect on densification process in PS samples. Moreover, the gap between the values of relative densities in PS and HP samples which were sintered at same processing conditions (time, temperature and composition), increases at higher Si3N4 contents. As it was expected, pressureless sintered samples contain remarkable amounts of porosity, which is increased by silicon nitride addition. This effect is shown in Fig. 3. The fracture surface of additive-free pressureless sintered sample (PS0) has been presented by our group elsewhere. Related discussions about the densification behavior of such a sample can be found in references [37] and [22]. Briefly, densification behavior of PS0 sample was not favorable, as the processing parameters were not able to fulfill the minimum levels of sintering phenomenon. Such observation is the main cause of seeking suitable sintering additives for pressureless sintering method. Therefore, it was expected that using sintering aids leads to improved densification, but the obtained results (Fig. 3) showed an inverse trend. The fracture surface and EDS spectrum of an interfacial phase in 9702 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Fig. 6. (a) Back scattered electron micrograph micrograph of the fracture surface of sample PS5, (b) Secondary electron SEM micrograph of a formed phase at interface of ZrB2/SiC particles in sample PS5 and (c) HRXRD pattern of the mentioned phase. zirconium carbide (ZrC). Some weak peaks in this pattern can normally be attributed to the background phases including ZrB2 and SiC. Microstructure analysis and related fractographical discussions about HP0 sample were reported elsewhere [22,37]. In brief, the presence of uniformly distributed pores (about 5%, based on the results of density measurements) was identified in the mentioned sample. The relatively large remained porosities were attributed to the insufficient hot pressing parameters [49–51]. This observation was in agreement with literature which reports that a fully dense ZrB2–SiC composite is achievable by hot pressing at ~2000 ℃, due to the activation of diffusion-based mechanisms. Such mechanisms, derived by higher diffusion rates at elevated temperatures, promote the complete removal of porosities from the as-sintered samples. In Fig. 7, SEM micrograph of the fracture surface, high-resolution SEM nanograph of an interfacial phase and EDS spectrum of the mentioned phase in HP1 sample are presented. The fracture surface of HP1 composite indicates a favorable sintering process which has led to intragranular fracture mode. It is due to the activation of densification mechanisms (e.g. particles’ rearrangement and plastic deformation) derived by applied pressure. In addition, the size and distribution of porosities have obviously changed in comparison with PS samples. Presence of interfacial flaky phases is also noticeable in HP1 sample. Detailed morphological characteristics of such phases can be paraphrased from Fig. 7b. In comparison with PS samples, the laminar morphology of the interfacial phase in HP1 sample promotes the hypothetical formation of graphite-like structures. EDS analysis of the mentioned phase (Fig. 7c) potentially addresses the formation of boron-based compounds (BN or B4C). Several published reports [52] endorsed the graphite-like growth of BN which is more compatible with the observations of this study. boron and zirconium elements which promotes the hypothetical formation of B4C or ZrC compounds. Fig. 6 shows the SEM fractogeraphs and related point HRXRD pattern of an interfacial particle (marked in Fig. 6b) of PS5 sample. As it can be clearly seen, a progressive sintering phenomenon has been occurred in the mentioned sample, as most of the particles have wide interfacial joinings (marked in Fig. 6a). although the relative density of this sample is the worst among pressureless sintered samples, the fracture mode dominantly consisted of intragranular fractures (according to the morphology of sintered particles). Such observation verifies a better sinterability along with unfavorable densification, in comparison with other PS samples. It may be due to the formation of gaseous phases during the heating step, which can result in more porosities. Anyway, as it has been observed in several other additives, densification is majorly controlled by in-situ formation of secondary phases (e.g. liquid, gaseous or solid glassy/crystalline phases), whereas interfacial phenomena are the most effective parameters in sintering behavior [12,13,46–48]. In the present case, it seems that more Si3N4 additions, on one hand caused higher contents of output gaseous phases (more porosities), but on the other hand, resulted in a better joining between the neighibor particles. Comparing PS1, PS3 and PS5 samples, it seems that the size of the interfacially formed particles increases as a function of Si3N4 content. It may be due to higher interfacial reactivity of the composite which is promoted by higher amounts of additive. It was already discussed that the EDS spectra of interfacial phases in PS1 and PS3 samples proposed the formation of carbide phases (ZrC or B4C). In the case of PS5 sample, the interfacial phase (Fig. 6b) is large enough to be characterized by point HRXRD phase analysis. According to this pattern (Fig. 6c), it can be concluded that the interfacial phases are majorly consisted of 9703 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Fig. 7. (a) Back scattered electron micrograph micrograph of the fracture surface of sample HP1, (b) Secondary electron SEM micrograph of a formed phase at interface of ZrB2/SiC particles in sample HP1 and (c) EDS spectrum of the mentioned phase. graphene-like structure is formed due to the 2-D hexagonal growth of boron nitride phase [53]. Comparing the sintering behavior of PS and HP series, it is worthy to discuss the possible mechanisms. To clarify the sintering mechanisms in Si3N4-doped ZrB2–SiC composites, some possible chemical reactions were investigated by a thermodynamical approach in the following. As it has been previously reported by several researchers, the surface of ZrB2 and SiC particles is commonly covered by surface oxide layers such as ZrO2, B2O3 and SiO2. Therefore, the powder mixture should be consisted of starting materials, the mentioned oxides, Si3N4 (additive), and carbon (pyrolized phenolic resin). Hence, several potential chemical reactions which can take place among the present components, are introduced in Fig. 10, regarding their Gibbs free energies as a function of temperature. It should be noted that the thermodynamical calculations were assumed based on the standard state (pressure of 1 atm). Therefore, the reactions can occur at temperatures lower than those of discussed here. As it can be paraphrased from Fig. 10, the chemical reaction between Si3N4 and B2O3 (Eq. (1)) has the most negative value of ΔG through the heating process. The graphite-like structure of the interfacial phase is also clearly seen in SEM fractograph of HP3 sample, presented in Fig. 8a and b. Fig. 8c shows the EDS spectrum of interfacial phase (marked in Fig. 8b), which is mainly consisted of boron and nitrogen, and can be considered as boron nitride (BN). It should be noted that as HP3 sample has a high relative density of ~99%, no considerable pores must be remained in the as-sintered structure. Such density measurement, based on the results showed in Fig. 3, is in agreement with the microstructural observations, because only a submicron pore can be seen in Fig. 8a (marked by arrow). Fractographical and analytical results of HP5 sample are presented in Fig. 9. Based on Fig. 9a, no porosity is visible in the fracture surface of the composite. This observations are in agreement with the density measurements presented in Fig. 3, which verifies achieving a fully dense composite by hot pressing process at 1900 °C under 10 MPa for 120 min and addition of 5 wt% Si3N4. It appears that the observed increase in the relative density of the composite is accompanied by formation of interfacial boron nitride phase. HRXRD pattern of HP5 sample presented in Fig. 9b, completely proves the formation and presence of BN at the interfaces of the composite. It is also given no sign of formation of boron carbide (B4C), which was previously imagined not to exist through the morphological discussions. SEM nanograph of interfacial BN phase in HP5 sample is presented in Fig. 9c. the multilayer boron nitride phase seems to be consisted of around 15 platelets. Hence; it can be categorized as graphene-like structures, as previously reported by Zhang et. al [52]. such a Si3N4+2B2O3 = 4BN+3SiO2 (1) The mentioned reaction results in the formation of BN and SiO2 phases at the interface of ZrB2 particles. The occurrence of Eq. (1) is 9704 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Fig. 8. Secondary electron SEM micrograph of (a)the fracture surface and (b) the formed phase at the interface of ZrB2/SiC particles in sample HP3 and (c) EDS spectrum of the mentioned phase. composite). It also should be noted that the occurrence of Eq. (3) can be affected by Eqs. 1–2, as it is highly depended on the amounts of unconsumed Si3N4 and C and heating temperature. Remained carbon can independently play a reductive role in oxide removal process, based on the following reactions. dominantly possible through the whole heating temperature range, based on its lowest Gibbs free energy. Therefore, it only depends on the availability of the reactants. Si3N4 can also react with remained carbon and ZrB2 particles throught the following reaction: 4C+Si3N4+ZrB2 = ZrC+2BN+3SiC+N2(g) (2) Eq. (2) leads to the formation of carbides (ZrC and SiC) together with BN, and N2 gaseous phase. the value of Gibbs free energy for the mentioned reaction turns to negative values at temperature about 975 °C. therefore, it can be concluded that Eq. (2) will be the second dominant reaction of the Si3N4-doped ZrB2–SiC system during heating process, if adequate amount of carbon is available. The reaction also depends on the amount of Si3N4 and thus, it is more favorable in samples with higher Si3N4 contents. Si3N4 additive can also react with the remained carbon and form SiC and N2 through the following reaction. Si3N4+3C = 3SiC+2N2(g) ZrO2+B2O3(l)+5C = ZrB2+5CO(g) (4) SiO2+3C = SiC+2CO(g) (5) Both Eqs. (4) and (5), which are favorable at temperatures > 1500 °C, lead to the formation of fine particles (ZrB2 and SiC) together with CO, and endorse the reductive role of carbon in oxide removal and sinterability improvement of ZrB2–SiC composites. Such reactions have also been reported in literature [31–33], but it should be considered that the amount of remained carbon in the present case, may be decreased due to the occurrence of several aforementioned reactions. Thus, the progression of such reactions may be affected by carbon shortage. At temperatures above 1675 °C, carbon can also independently react with ZrO2 and form ZrC, through the following reaction. (3) As it can be derived from Fig. 10, Eq. (3) is favorable at temperatures > 1450 °C, according to the Gibbs free energy values. Although the formation of fine SiC particles due to the occurrence of Eq. (3) initially seems to result in better sintering behavior, but more favorable results can be achieved if the reactants act as oxide removal agents (as the presence of oxide impurities on the surface of starting particles is known as a harmful factor in grain coarsening of the ZrO2+3C = ZrC+2CO(g) (6) It has been reported that the formation of fine ZrC interfacial phase can improve the sinterability of ZrB2-based materials [22,31– 33,37,54,55], although it seems that the resulted CO may negatively influence the densification process. 9705 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Si3N4 not only can react with the surface oxide layer of SiC (SiO2) based on Eq. (7), but also directly with SiC itself via following reaction: Si3N4+SiC = 4Si(l)+C+2N2(g) (8) The product similarly contains N2 and liquid Si phases plus carbon. Anyway, it is noticeable that Eq. (8) can occur at temperature about 1900 °C (the sintering temperature of the samples). Reverse state of such a reaction is also reported by Ohtsuka et al. [56] at about 1800 °C, which is compatible with thermodynamical assessment of this research. If Si3N4 directly reacts with ZrB2 particles, following reaction (Eq. (9)) should progress. Anyway, the progression of Eq. (9) seems to be unfavorable at the maximum applied temperature of this study, as the Gibbs free energy values of this reaction are positive, based on Fig. 10. 2Si3N4+3ZrB2 = 6BN+3ZrSi2+N2(g) (9) As it was noticed at the beginning of the thermodynamical discussions, the occurrence and consequence of the mentioned reactions may change due to the processing parameters. For example, although the formation of BN is favorable (Eqs. (1) and (2)), no sign of boron nitride is found in PS samples. It may be due to the fact that in vacuum conditions, liquid B2O3 can easily evaporate and be exhausted from the microstructure of the pressureless sintered samples [18]. The B2O3 shortage then can postpone/avoid the occurrence of Eq. (1), which is one of the favorable BN-forming reactions. Also, as no effective contact area between ZrB2 and Si3N4 can be provided in pressureless sintering process, the progression of Eq. (2) is not favorable. Inversely, in hot pressing process, the applied pressure can promote plastic deformation of the adjacent particles and lead to a better contacts between the reactant particles. Therefore, Eq. (2) can be activated and result in the formation of BN. Such hypotheses were previously verified via fractographical and HRXRD analyses (Figs. 6c and 9b). As the surface B2O3 can be easily evaporated in PS samples, the remained surface ZrO2 can independently react with carbon and form ZrC phase. But, in HP samples, applied pressure not only provides adequate contact area, but also leads to B2O3 entrapment and promotes Eq. (1) and consequent formation of BN. such a hypothesis can be more clarified according to the temperature priority of the Eq. (2) compared to Eq. (6). As the gaseous phases can be easily escaped from the composite in PS samples, the occurrence of Eqs. 7–8 is more favored in pressureless sintering process. Therefore, some liquid phases can be formed in such conditions. Referring to Figs. 5b and 6b, it seems that the interfacial phases are consisted of nano-particles agglomerated/distributed in a solidified bed (liquid). Anyway, no sign of liquid phase sintering mechanism was detected in fractographs of the HP samples which denies the occurrence of liquid phase-forming reactions. 4. Conclusions Fig. 9. (a) Secondary electron SEM micrograph of the fracture surface, (b) HRXRD pattern and (c) SEM nanograph of the formed phase at the interface of ZrB2/SiC which shows the approximate thickness of BN layers in sample HP5. ZrB2–SiC composite ceramics doped with different amounts of Si3N4 as additive were fabricated via pressureless sintering and hot pressing routes, in order to compare the influences of Si3N4 on microstructure and densification behavior of such ultrahigh temperature ceramics. Hot pressing at 1900 °C for 2 h under 10 MPa and addition of 5 wt% Si3N4 resulted in fully dense ceramic. The reactive sintering mechanism led to the formation of graphene-like BN platelets at the interfaces as the dominant sintering phenomenon in the hot pressed composites. Several chemical reactions which may result in the formation of other interfacial glassy or crystalline phases were thermodynamically studied; some of them may lead to the formation of gaseous products. As opposed to hot pressed samples, increasing Si3N4 Regarding Fig. 10, Si3N4 can also react with SiO2 at temperatures above 1850 °C through Eq. (7) and form non-solid products. SiO2 reactant in Eq. (7) can be supplied by the surface oxide layer of SiC particles and/or the progression of Eq. (1). The produced liquid Si can activate the liquid phase sintering and improve the densification of the composite, if the gaseous products (SiO and N2) can successfully escape from the samples. Si3N4+SiO2 = 2SiO(g)+2Si(l)+2N2(g) (7) 9706 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. Fig. 10. Gibbs free energy as a function of temperature for potential chemical reactions in Si3N4/C-doped ZrB2–SiC composites (Diagrams obtained by HSC Chemistry). porous ZrB2–40 vol% B4C UHTC, Ceram. Int. 42 (2016) 17009–17015. [13] M. Shahedi Asl, M. Ghassemi Kakroudi, B. Nayebi, A fractographical approach to the sintering process in porous ZrB2–B4C binary composites, Ceram. Int. 41 (2015) 379–387. [14] H. Zhang, Y. Yan, Z. Huang, X. Liu, D. Jiang, Pressureless sintering of ZrB2–SiC ceramics: the effect of B4C content, Scr. Mater. 60 (2009) 559–562. [15] R. Zhang, R. He, X. Zhang, D. Fang, Microstructure and mechanical properties of ZrB2–SiC composites prepared by gelcasting and pressureless sintering, Int. J. Refract. Met. Hard Mater. 43 (2014) 83–88. [16] D. Sciti, L. Silvestroni, V. Medri, S. Guicciardi, Pressureless sintered in situ toughened ZrB2–SiC platelets ceramics, J. Eur. Ceram. Soc. 31 (2011) 2145–2153. [17] X.G. Wang, W.M. Guo, G.J. Zhang, Pressureless sintering mechanism and microstructure of ZrB2–SiC ceramics doped with boron, Scr. Mater. 61 (2009) 177–180. [18] M. Shahedi Asl, M. Ghassemi Kakroudi, Fractographical assessment of densification mechanisms in hot pressed ZrB2-SiC composites, Ceram. Int. 40 (2014) 15273–15281. [19] M. Zhu, Y. Wang, Pressureless sintering ZrB2–SiC ceramics at low temperatures, Mater. Lett. 63 (2009) 2035–2037. [20] S.Q. Guo, Densification of ZrB2-based composites and their mechanical and physical properties: a review, J. Eur. Ceram. Soc. 29 (2009) 995–1011. [21] W. Han, G. Li, X. Zhang, J. Han, Effect of AlN as sintering aid on hot-pressed ZrB2– SiC ceramic composite, J. Alloy. Compd. 471 (2009) 488–491. [22] Z. Ahmadi, B. Nayebi, M. Shahedi Asl, M. Ghassemi Kakroudi, Fractographical characterization of hot pressed and pressureless sintered AlN-doped ZrB2–SiC composites, Mater. Charact. 110 (2015) 77–85. [23] F. Monteverde, A. Bellosi, Development and characterization of metal-diboridebased composites toughened with ultra-fine SiC particulates, Solid State Sci. 7 (2005) 622–630. [24] F. Monteverde, The thermal stability in air of hot-pressed diboride matrix composites for uses at ultra-high temperatures, Corros. Sci. 47 (2005) 2020–2033. [25] H. Wu, W. Zhanga, Fabrication and properties of ZrB2–SiC–BN machinable ceramics, J. Eur. Ceram. Soc. 30 (2010) 1035–1042. [26] G. Li, W. Han, B. Wang, Effect of BN grain size on microstructure and mechanical properties of the ZrB2–SiC–BN composites, Mater. Des. 32 (2011) 401–405. [27] F. Monteverde, A. Bellosi, Effect of the addition of silicon nitride on sintering behaviour and microstructure of zirconium diboride, Scr. Mater. 46 (2002) 223–228. [28] S.Q. Guo, Y. Kagawa, T. Nishimura, H. Tanaka, Pressureless sintering and physical properties of ZrB2-based composites with ZrSi2 additive, Scr. Mater. 58 (2008) 579–582. [29] J. Zou, G.J. Zhang, Y.M. Kan, P.L. Wang, Pressureless densification of ZrB2–SiC composites with vanadium carbide, Scr. Mater. 59 (2008) 309–312. [30] Z. Wang, S. Wang, X. Zhang, P. Hu, W. Han, C. Hong, Effect of graphite flake on microstructure as well as mechanical properties and thermal shock resistance of ZrB2–SiC matrix ultrahigh temperature ceramics, J. Alloy. Compd. 484 (2009) 390–394. [31] M. Shahedi Asl, M. Ghassemi Kakroudi, R. Abedi Kondolaji, H. Nasiri, Influence of graphite nano-flakes on densification and mechanical properties of hot-pressed ZrB2–SiC composite, Ceram. Int. 41 (2015) 5843–5851. [32] M. Shahedi Asl, M. Ghassemi Kakroudi, Characterization of hot-pressed graphene reinforced ZrB2–SiC composite, Mater. Sci. Eng. A 625 (2015) 385–392. [33] M. Shahedi Asl, F. Golmohammadi, M. Ghassemi Kakroudi, M. Shokouhimehr, content elevated the formation of gaseous products in the pressureless sintered ones; hence, the relative density values significantly decreased. However, Si3N4 dopant boosted the sinterability via the formation of interfacial ZrC phase between the particles. The formation of secondary phases was characterized by HRXRD, investigated by SEM fractographs, EDS analysis and by the assistance of a thermodynamical approach. References [1] B. Nayebi, M. Shahedi Asl, M. Ghassemi Kakroudi, M. Shokouhimehr, Temperature dependence of microstructure evolution during hot pressing of ZrB2–30 vol% SiC composites, Int. J. Refract. Met. Hard Mater. 54 (2016) 7–13. [2] M. Jaberi Zamharir, M. Shahedi Asl, M. Ghassemi Kakroudi, N. Pourmohammadie Vafa, M. Jaberi Zamharir, Significance of hot pressing parameters and reinforcement size on sinterability and mechanical properties of ZrB2–25 vol% SiC UHTCs, Ceram. Int. 41 (2015) 9628–9636. [3] M. Shahedi Asl, I. Farahbakhsh, B. Nayebi, Characteristics of multi-walled carbon nanotube toughened ZrB2–SiC ceramic composite prepared by hot pressing, Ceram. Int. 42 (2016) 1950–1958. [4] N. Pourmohammadie Vafa, B. Nayebi, M. Shahedi Asl, M. Jaberi Zamharir, M. Ghassemi Kakroudi, Reactive hot pressing of ZrB2-based composites with changes in ZrO2/SiC ratio and sintering conditions. Part II: mechanical behavior, Ceram. Int. 42 (2016) 2724–2733. [5] M. Jaberi Zamharir, M. Shahedi Asl, N. Pourmohammadie Vafa, M. Ghassemi Kakroudi, Significance of hot pressing parameters and reinforcement size on densification behavior of ZrB2–25 vol% SiC UHTCs, Ceram. Int. 41 (2015) 6439–6447. [6] N. Pourmohammadie Vafa, M. Shahedi Asl, M. Jaberi Zamharir, M. Ghassemi Kakroudi, Reactive hot pressing of ZrB2-based composites with changes in ZrO2/ SiC ratio and sintering conditions. Part I: densification behavior, Ceram. Int. 41 (2015) 8388–8396. [7] M. Shahedi Asl, M. Ghassemi Kakroudi, S. Noori, Hardness and toughness of hot pressed ZrB2–SiC composites consolidated under relatively low pressure, J. Alloy. Compd. 619 (2015) 481–487. [8] Z. Balak, M. Shahedi Asl, M. Azizieh, H. Kafashan, R. Hayati, Effect of different additives and open porosity on fracture toughness of ZrB2–SiC-based composites prepared by SPS, Ceram. Int. 43 (2017) 2209–2220. [9] M. Shahedi Asl, M. Ghassemi Kakroudi, A processing–microstructure correlation in ZrB2–SiC composites hot-pressed under a load of 10 MPa, Univers. J. Mater. Sci. 3 (2015) 14–21. [10] D. Sciti, L. Silvestroni, G. Saccone, D. Alfano, Effect of different sintering aids on thermoemechanical properties and oxidation of SiC fibers reinforced ZrB2 composites, Mater. Chem. Phys. 137 (2013) 834–842. [11] M. Shahedi Asl, M. Ghassemi Kakroudi, A. Farzaneh, B. Nayebi, Influence of nanoSiC participation on densification and mechanical properties of ZrB2, in: Proceedings of the 10th Nanoscience and Nanotechnology Conference of Turkey (NanoTR10), Istanbul, 2014. [12] B. Nayebi, M. Shahedi Asl, M. Ghassemi Kakroudi, I. Farahbakhsh, M. Shokouhimehr, Interfacial phenomena and formation of nano-particles in 9707 Ceramics International 43 (2017) 9699–9708 Z. Ahmadi et al. [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] Synergetic effects of SiC and Csf in ZrB2-based ceramic composites. Part I: densification behavior, Ceram. Int. 42 (2016) 4498–4506. H. Salimkhani, P. Palmeh, A.B. Khiabani, E. Hashemi, S. Matinpour, H. Salimkhani, M. Shahedi Asl, Electrophoretic deposition of spherical carbonyl iron particles on carbon fibers as a microwave absorbent composite, Surf. Interfaces 5 (2016) 1–7. M. Shahedi Asl, M. Ghassemi Kakroudi, I. Farahbakhsh, B. Mazinani, Z. Balak, Synergetic effects of SiC and Csf in ZrB2-based ceramic composites. Part II: grain growth, Ceram. Int. 42 (2016) 18612–18619. M. Shahedi Asl, M. Ghassemi Kakroudi, B. Mazinani, B. Nayebi, The morphologydependent properties for nano-carbon-reinforced ZrB2–SiC composites, in: Proceedings of the 12th International Nanoscience and Nanotechnology Conference (NanoTR12), Darica-Kocaeli, Turkey, 2016. M. Shahedi Asl, B. Nayebi, Z. Ahmadi, P. Pirmohammadi, M. Ghassemi Kakroudi, Fractographical characterization of hot pressed and pressureless sintered SiAlONdoped ZrB2–SiC composites, Mater. Charact. 102 (2015) 137–145. F. Monteverde, A. Bellosi, S. Guicciardi, Processing and properties of zirconium diboride-based composites, J. Eur. Ceram. Soc. 22 (2002) 279–288. F. Monteverde, S. Guicciardi, A. Bellosi, Advances in microstructure and mechanical properties of zirconium diboride based ceramics, Mater. Sci. Eng. A 346 (2003) 310–319. I.G. Talmy, J.A. Zaykoski, M.M. Opeka, High-temperature chemistry and oxidation of ZrB2 ceramics containing SiC, Si3N4, Ta5Si3, and TaSi2, J. Am. Ceram. Soc. 91 (2008) 2250–2257. M. Mallik, K.K. Ray, R. Mitra, Effect of Si3N4 addition on compressive creep behavior of hot-pressed ZrB2–SiC composites, J. Am. Ceram. Soc. 97 (2014) 2957–2964. F. Luo, D. Zhu, X. Su, W. Zhou, Properties of hot-pressed of SiC/Si3N4 nanocomposites, Mater. Sci. Eng. A 458 (2007) 7–10. W.M. Guo, L.X. Wu, T. Ma, S.X. Gu, Y. You, H.T. Lin, S.H. Wu, G.J. Zhang, Chemical reactivity of hot-pressed Si3N4–ZrB2 ceramics at 1500–1700°C, J. Eur. Ceram. Soc. 35 (2015) 2973–2979. M. Panneerselvam, K.J. Rao, Preparation of Si3N4–SiC composites by microwave route, Bull. Mater. Sci. 25 (2002) 593–598. H. Ghanem, H. Gerhard, N. Popovska, Paper derived SiC–Si3N4 ceramics for high temperature applications, Ceram. Int. 35 (2009) 1021–1026. K. Farhadi, A. Sabahi Namini, M. Shahedi Asl, A. Mohammadzadeh, M. Ghassemi [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] 9708 View publication stats Kakroudi, Characterization of hot pressed SiC whisker reinforced TiB2 based composites, Int. J. Refract. Met. Hard Mater. 61 (2016) 84–90. M. Shahedi Asl, A. Sabahi Namini, M. Ghassemi Kakroudi, Influence of silicon carbide reinforcement on the microstructural development of hot pressed zirconium and titanium diborides, Ceram. Int. 42 (2016) 5375–5381. A. Sabahi Namini, S.N. Seyed Gogani, M. Shahedi Asl, K. Farhadi, M. Ghassemi Kakroudi, A. Mohammadzadeh, Microstructural development and mechanical properties of hot pressed SiC reinforced TiB2 based composite, Int. J. Refract. Met. Hard Mater. 51 (2015) 169–179. M. Shahedi Asl, M. Ghassemi Kakroudi, F. Golestani-Fard, H. Nasiri, A Taguchi approach to the influence of hot pressing parameters and SiC content on the sinterability of ZrB2-based composites, Int. J. Refract. Met. Hard Mater. 51 (2015) 81–90. M. Shahedi Asl, M. Ghassemi Kakroudi, B. Nayebi, H. Nasiri, Taguchi analysis on the effect of hot pressing parameters on density and hardness of zirconium diboride, Int. J. Refract. Met. Hard Mater. 50 (2015) 313–320. M. Shahedi Asl, M. Ghassemi Kakroudi, M. Rezvani, F. Golestani-Fard, Significance of hot pressing parameters on the microstructure and densification behavior of zirconium diboride, Int. J. Refract. Met. Hard Mater. 50 (2015) 140–145. Z. Zhang, Y. Liu, Y. Yang, B.I. Yakobson, Growth mechanism and morphology of hexagonal boron nitride, Nano Lett. 16 (2016) 1398–1403. M.S. Bresnehan, M.J. Hollander, M. Wetherington, K. Wang, T. Miyagi, G. Pastir, D.W. Snyder, J.J. Gengler, A.A. Voevodin, W.C. Mitchel, J.A. Robinson, Prospects of direct growth boron nitride films as substrates for graphene electronics, J. Mater. Res. 29 (2014) 459–471. M. Shahedi Asl, Z. Ahmadi, B. Nayebi, M. Ghassemi Kakroudi, Effect of nano-AlN on densification behavior of hot-pressed ZrB2-based ceramics, in: Proceedings of the 12th International Nanoscience and Nanotechnology Conference (NanoTR12), Darica-Kocaeli, Turkey, 2016. I. Farahbakhsh, Z. Ahmadi, M. Shahedi Asl, Densification, microstructure and mechanical properties of hot pressed ZrB2–SiC ceramic doped with nano-sized carbon black, Ceram. Int. (2017). http://dx.doi.org/10.1016/j.ceramint.2017.03.188. T. Ohtsuka, H. Mabuchi, H. Tsuda, K. Morii, Fabrication of Si3N4/SiC composite ceramics by reactive hot-pressing from elemental powders, J. Jpn. Soc. Powder Metall. 45 (1998) 321–325.