Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-7591(Print)
ISSN : 2287-8173(Online)
Journal of Korean Powder Metallurgy Institute Vol.26 No.6 pp.455-462
DOI : https://doi.org/10.4150/KPMI.2019.26.6.455

Effect of SiC and WC additon on Oxidation Behavior of Spark-Plasma-Sintered ZrB2

Chang-Yeoul Kima*, Jae-Seok Choib, Sung-Churl Choia*
aNano- Material and Process Center, Korea Institute of Ceramic Engineering and Technology, 101 Soho-ro Jinju-si Gyeongsangnam-do, 52851, Republic of Korea
bDept. of Material Sci. & Eng. Hanyang University, Seoul 04763, Republic of Korea
-

김창열: 책임연구원, 최재석: 학생, 최성철: 교수


Corresponding Authors: Chang-Yeoul Kim, TEL: +82-55-792-2707, FAX: +82-55-792-2730, E-mail: cykim15@kicet.re.kr Sung-Churl Choi, TEL: +82-2-2220-0505, FAX: +82-2-2291-6767, E-mail: choi0505@hanyang.ac.kr
November 6, 2019 December 19, 2019 December 23, 2019

Abstract


ZrB2 ceramic and ZrB2 ceramic composites with the addition of SiC, WC, and SiC/WC are successfully synthesized by a spark plasma sintering method. During high-temperature oxidation, SiC additive form a SiO2 amorphous outer scale layer and SiC-deplete ZrO2 scale layer, which decrease the oxidation rate. WC addition forms WO3 during the oxidation process to result in a ZrO2/WO3 liquid sintering layer, which is known to improve the antioxidation effect. The addition of SiC and WC to ZrB2 reduces the oxygen effective diffusivity by one-fifth of that of ZrB2. The addition of both SiC and WC shows the formation of a SiO2 outer dense glass layer and ZrO2/WO3 layer so that the anti-oxidation effect is improved three times as much as that of ZrB2. Therefore, SiC- and WC-added ZrB2 has a lower two-order oxygen effective diffusivity than ZrB2; it improves the anti-oxidation performance 3 times as much as that of ZrB2.



초록


    1. Introduction

    Many research papers of the anti-oxidation of superhigh temperature ceramics have been published [1-14]. Thermal degradation of the material properties of super high temperature ceramics such as borides, carbides, nitrides, and oxides is very important because they have been used in high temperatures. Zirconium diboride (ZrB2) as a representative high temperature material has an excellent electrical conductive and mechanical strength. But has a weakness of oxidation from the surfaces exposed to high temperature. To inhibit the high temperature oxidation behavior, many research groups have conducted of the antioxidation of ZrB2 by the composition. The representative composite materials are ZrB2-silicon carbide (SiC) [3-10], and ZrB2-tungsten carbide (WC) [11-14]. For ZrB2-SiC, the formation of silicon dioxide (SiO2) layer from the oxidation of SiC is known to play a role to prohibit the diffusion of oxygen from the surface to the inner side. For ZrB2-WC, the volume expansion of tungsten oxide (WO3) layer formed from the oxidation of WC is known to decrease the thermal degradation of the composite property. In this paper, we focus on the antioxidation effects of the addition of SiC and WC to ZrB2. Until now, the addition of SiC and WC to ZrB2 ceramic was studied for their chemical reactions and strong mechanical properties [13, 14], but now the anti-oxidation properties are not studied. So, we think that the study of simultaneous SiC and WC addition effects to the anti-oxidation of ZrB2 ceramic is needed. Parthasarathy et al. [15, 16] suggested the model for the oxidation of ZrB2 in the temperature range; low temperature range below 1000°C, intermediate temperature range between 1000-1800°C, and high temperature range over 1800°C. Boric oxide (B2O3) does not evaporate and exists as a liquid phase in the low temperature range, but begins to evaporate in the intermediate temperature, and zirconium oxide (ZrO2) can evaporate in the high temperature range. In our study, we try to understand the effective oxygen diffusion in ZrB2 in the case of the addition of either SiC or WC.

    2. Experiment

    2.1 Raw materials and the sample powder preparations

    The raw materials used were ZrB2 (99.5%, 1~2 μm, APS powder, Alfa Aesar), SiC (99.8%, 1 μm, S.A. 11.5 m2/g beta-phase, Alfa Aesar), WC (99%, 2 μm, Sigma- Aldrich). Four ZrB2-based composites with different composition were used for this study. Table 1 summarizes the four batches prepared in this study, which are referred to as Z (ZrB2), ZS (ZrB2-SiC), ZW (ZrB2-WC), ZSW (ZrB2-SiC-WC), respectively. In Table 1, we presents the weight and volume increase of the samples, supposing that 100% of the samples are oxidized to form ZrO2, SiO2 and WO3 and calculate the weight and volume of the oxidized samples. The reference samples of Z, ZS, and ZW were prepared for the exact analysis of SiC-WC addition effects to the thermal degradation. The compositions of samples were shown in Table 2. It is reported that ZrB2-SiC shows the best antioxidation properties of dense and optimal viscosity after the thermal oxidation at high temperature for the volume fraction of 76-84%, and so we determined the SiC volume fraction as 20% for the excellent sinterability and antioxidation. The addition of WC at 6 mole% is reported that the volume expansion of WO3 generated from the oxidation of WC through the phase transformation shows the crackfree and excellent anti-oxidative sintered ZrB2 composite. So, the 6 mol% of WC addition was determined for the preparation for the ZrB2-WC composite. Small amount (about 0.5%, respectively) of B4C and C (graphite powders) were added to remove the oxidized impurities of ZrB2 surfaces and improve the sinterability. Raw materials of ZrB2, SiC, WC, B4C and C were ball-milled in the methanol for 24 h and rotary-evaporated. After that, the mixed powders were sieved using #100 mesh screen and dried in a vacuum oven at 80°C for 24 h.

    2.2 SPS sintering

    The prepared powders were packed in the graphite mold and then sintered by using the spark plasma sintering equipment (SPS). The sintering conditions are like these; the temperature rise of 100°C/min and the sintering temperature and pressure at 1800°C and 30 MPa for Z and ZS and at 1900°C and 60 MPa for ZW and ZSW, and the sintering holding time of 5 min. The densities of the samples (the bulk densities) were measured by Archimedes’ method and theoretical densities were calculated for the composition of the samples. The relative densities mean the ratio of measured bulk densities to theoretical densities. The full relative densities of the samples were obtained except for ZS with 99.7% of relative density.

    2.3 Anti-oxidation property analysis

    The anti-oxidation analyses were conducted through two different methods: torch thermal oxidation and ther- mogravimetric (TGA) method. The torch oxidation was conducted by the exposure of the samples to propane and oxygen (20%) flame at 1600°C measured by optical pyrometer for 15 min (Fig. 1). The TGA method were conducted for the oxidation of the samples at 1500°C for 15 min at the temperature rise of 10°C/min at ambient atmosphere to measure the increase of sample weights (TGA-DTA, DTG-60H and TA-60WS, Shimadzu). The thicknesses of the oxidation layers were observed by using field emission scanning electron microscopy (FESEM) and energy dispersive X-ray spectroscopy (EDS) mapping images (JSM-6700F, JEOL, Japan).

    3. Results

    3.1 Oxide scale formations

    1) ZrO2 columnar scale formation from B2O3 evaporation

    FE-SEM images before and after torch oxidation test were observed in Fig. 2. After the torch oxidation test, we analyzed the oxidation behaviors by SEM and EDS mapping image observations (Fig. 2). Z sample without the addition of SiC and WC showed 49 μm of oxidation layer thickness. The oxidation layer thicknesses are 29.4, 33.4, and 16.9 μm for ZS, ZW and ZSW, respectively. For Z, the surface layer is composed of ZrO2 scale layer, which has the columnar structure. The channels between the columns are considered to be paths for oxygen diffusion and boron evaporation.

    2) SiO2 glass outer layer and ZrO2 oxide inner layer

    The oxidation layer of ZS is composed of two layers; the outer layer is a SiO2 layer with about 5 μm thickness and the inner layer is a porous ZrO2 layer with about 25 μm of thickness. The anti-oxidation effect of SiC is known to result from the formation of SiO2 layer by the oxidation of SiC on the surface to inhibit the oxidation of ZrB2. SiO2 generated from the oxidation of SiC diffuses out to form amorphous silica layer on the surface of the samples, which was identified by XRD data in Fig. 2. After oxidations of ZS and ZSW, the XRD patterns of the samples showed the broad band at around 20°, which is known to be amorphous silica phase. We also confirmed the formation of amorphous silica phase on the surface of the samples, ZS and ZSW in the FE-SEM microstructures in Fig. 2. We guess that SiO2 amorphous phase migrates onto the surface although ZrO2 formation from ZrB2 occurs at the same place so that B2O3 gas evolves out and the oxidized ZrB2 layer is porous. A. Rezaie et al. [3] suggested that B2O3 rich borosilicate scale layer formed on the surface of ZrB2-SiC ceramics below 1200°C and SiO2 rich scale layer is generated from the evaporation of B2O3 gas above 1200°C. Below the surface scale layer, ZrO2-SiO2 layer exist and the below are ZrB2-SiC bulk layer. It indicates that borosilicate glass layer forms below 1200°C but turns to be SiO2 amorphous layer from the evaporation of B2O3 above 1200°C. Fahrenholtz [9] also investigated a thermodynamic behavior of oxidation of ZrB2-SiC. He insists that the initial formation of ZrO2 and B2O3 layer on the surface at 1200°C due to the rapid oxidation rate of ZrB2 rather than SiC. At 1500°C, SiC oxidation rate increases to form SiO2 layer and SiC depletion layer between the surface SiO2 layer and SiO2 and ZrO2 layer. In our study, the surface layer is thought be SiO2 amorphous phase, and SiC depletion layer exists below the SiO2 layer which was thought to be consisted of ZrO2 or ZrB2.

    3) ZrO2/WO3 oxide layer formation

    ZW has an oxidation layer with 33.4 μm of thickness, which is thought be composed of ZrO2 and WO3 mixtures. It is known that WC addition to ZrB2 improves the anti-oxidation properties of ZrB2 ceramics due to its densification of oxidized ZrO2 scale by the volume increase of WO3 formed from the oxidation of WC [11, 12]. During the oxidation, WO3 turns to liquid phase within ZrO2 scales so that the liquid phase sintering densifies the oxidized ZrO2/WO3 scale to result in the inhibition of oxidation. In our study, WC addition formed WB after sintering, which indicates that WC reacts with ZrB2. We guess that WB is oxidized to form WO3 liquid phase at 1500°C and a dense oxidized scale of ZrO2/WO3 phases. It increases the anti-oxidation effects of ZrB2 so that there are no cracks on the surface of the samples after the oxidation test, although Z shows the cracks due to the formation of ZrO2 scale and its phase transformation. In our case, the oxidized thickness of ZW was 33.4μm, 32% decrease against the oxidized thickness with 49 μm for Z.

    4) SiO2 glass phase outer layer and ZrO2/WO3 oxide layer formation

    The oxidation layer of ZSW is composed of a 5-μmthick dense surface layer and a 12-μm-thick porous inner layer. From the EDS analysis mapping image, the location of ZrO2 and WO3 crystal phase and SiO2 amorphous phase is not so clear. From the EDS mapping images for ZS and ZW and XRD data for ZSW, we guess that the dense surface layer is thought to be a SiO2 amorphous layer and the inner porous layer is a ZrO2 and WO3 crystal phase mixtures.

    The anti-oxidation effect by the addition of SiC and WC to ZrB2 is prominent for our study. The oxidized scale thickness is about 17 μm, which was about one third of that for Z. From XRD and SEM-EDS results, we observed that the oxidized scale is composed of two layers, the outer surface layer is SiO2 dense layer and the inner surface layer is composed of ZrO2 and WO3 layer. The two layers are considered to inhibit the diffusion of oxygen toward the bulk ZSW. After sintering, the major crystal phase of ZSW is ZrB2 and the secondary phases are SiC and WB. WB crystal phase is known to form from the following chemical reactions [13, 14].

    5ZrB 2  +4WC = 5ZrC +2W 2 B 5 ZrB 2  + 2WC = ZrC + 2WB

    During the oxidation, ZrB2 and SiC are oxidized to form ZrO2 (s) and B2O3 (g) and SiO2 and CO2 (g) phases. SiO2 amorphous phase is diffused out to the outer surface area and dense amorphous scale layer, which acts as an oxidation inhibition layer [3-10]. We guess that WB is also oxidized to form WO3 inner scale layer, which are composed of WO3 and ZrO2 inner scale layer [11-14]. This layer is reported to be composed of ZrO2/WO3 liquid sintered phase to form a dense layer, which improves the anti-oxidation effect. So, in our study, the simultaneous addition of SiC and WC improves the anti-oxidation effect three times as much as that of ZrB2 by the formation of SiO2 amorphous outer scale (1) and ZrO2/WO3 inner scale (2).

    SiC(s) + 2 O 2 (g) = SiO 2 (s) + CO 2 (g); SiO 2  amorphous inner scale formation
    (1)

    ZrB 2 (s) + 2 .5 O 2 (g) = ZrO 2 (s) +B 2 O 3 (g) WB(s) + 2 .25 O 2 (g) = WO 3 (s) + 0 .5 B 2 O 3  (g) ZrO 2 /WO 3  inner scale formation }
    (2)

    3.2 Weight changes analysis of the samples by torch and TGA

    The weight changes of the samples by TGA are shown in Fig. 4. The samples were placed in alumina sample holder and heated to 1500°C holding for 15 min. The weight increases of the samples are 106.6% for Z, 101.9% for ZS, 105.2 for ZW and 101.0% for ZSW. When the samples were maintained at 1500°C for 30 min, the weight increase of the samples are 112.0% for Z, 110.8% for ZS, 110.9 for ZW and 102.1% for ZSW. The weight increase of the samples before and after the torch oxidation test at 1600°C for 15 min in an ambient atmosphere were 27.4, 22.6, 21.9, and 13.9 μg/mm2 for Z, ZS, ZW and ZSW, respectively.

    4. Discussion

    4.1 The oxidation reactions and behaviors

    The oxidation behavior of ZrB2 is suggested that ZrB2 is oxidized to produce ZrO2 structure and B2O3 liquid phase between ZrO2 columns and from the interface between B2O3 interphase and ambient atmosphere is evaporated B2O3 gas phase (Fig. 3(a)). Oxygen gas phase permeates through the outer ZrO2 porous layer and diffuses into the B2O3 liquid phase, and then reacts with ZrB2 in the ZrB2 and ZrO2/B2O3 interface. For ZS, the outer scale layer is found as SiO2 amorphous layer (Fig. 3(b)). First, ZrO2 crystals and B2O3 liquid phase are formed from the oxidation of ZrB2 and B2O3 gas phase evaporates onto the surface and SiC is also oxidized to from SiO2 phase. Borosilicate glass (88% of SiO2 and 10% of B2O3 and less than 2% of ZrO2) is known to form from the oxidation of ZrB2 containing 15% of SiC. In our study, amorphous phase of SiO2 was confirmed from XRD data and EDS data (Fig. 2). The outer SiO2 dense amorphous layer is considered to inhibit the diffusion of oxygen gas into ZrB2. The addition of WC, which is converted as WB from the reaction with ZrB2, plays a role of liquid phase sintering, and so the ZrO2/WO3 oxide layer is known to be formed. Compared to ZrB2, WCadded ZrB2 shows a dense oxide scale layer due to the volume expansion of WO3 (Fig. 3(c)). For ZSW, the oxidation of SiC and WC (converted to WB) produces SiO2 outer layer and ZrO2/WO3 liquid phase sintered inner oxide layer. The dense diffusion inhibiting SiO2 glass phase and the expansion of WO3 cause the synergistic anti-oxidation effects of ZrB2. The chemical oxidation reactions are expressed as followed (Fig. 3).

    ZrB 2  (s) +5/2O 2  (g) = ZrO 2  (s) + B 2 O 3  (g) WC(s) + 2O 2 (g) = WO 3 (s) + CO(g) SiC(s) + 3/2O 2 (g) = SiO 2 (s) + CO(g)

    We propose that WC reacts with oxygen to form WO3 for the simple model calculation, although WC reacts with ZrB2 to form WB for the samples. From the oxidation reactions, we calculated the weight increase of Z, ZS, ZW, and ZSW, supposing that 100% of ZrB2, SiC and WC are oxidized to form ZrO2, SiO2, and WO3 and no evaporation of the oxides occur. The weight gains of the samples are 109.19, 113.92, 110.10, and 113.60% for Z, ZS, ZW, and ZSW, respectively. The correspondent volume gains are 116.98, 129.86, 122.80, and 133.39%, respectively (Table 1). We measured the oxidation thickness after torch test by FE-SEM and EDS analysis (Fig. 2) and the weight increase (Table 2). We also converted the thickness date to the weight gain and vice versa as shown in Table 2. The weight increase and thickness changes according to the samples show consistent results.

    4.2 Diffusion behavior from TGA

    The weight increase of Z, ZS, ZW and ZSW are 8.2, 6.8, 5.3 and 1.2% for the oxidation in TGA test at 1500°C for 30 min in ambient atmosphere (Fig. 4(a)). The oxidation behavior shows parabolic increase behavior with oxidation time. It can be converted to oxidation ratio from the weight increase data of 100% oxidation from Table 1. Then, as is shown in Fig. 4 (b), we can know that 89% of Z, 49% of ZS, 52% of ZW and 8.8% of ZSW are oxidized (Table 3). It is considered that the difference of oxidation rate between torch test and TGA analysis result from the sample dimension differences. For example, the sample dimension is about 6% of torch test sample (φ: 10 mm, t: 1 mm). We also converted the oxidation rate data to the oxide scale thickness data supposing that the sample thickness is 500 μm (Fig. 4(c)). We calculated the effective diffusivity of oxygen gas using Fick’s second law. From that law, we can calculate the effective diffusivity, using the equation of L = D t , where L is diffusion length (that is, oxide scale thickness), D is oxygen effective diffusivity, and t is diffusion time. From this, we can calculate the effective oxygen diffusivity, D=L2/t. The effective diffusivity is given by the following equation [15, 16].

    D =  ( D k 1 + D 12 1 ) 1

    Here, Dk is the Knudsen diffusivity, and D12 is The diffusivity of species 1 (oxygen) in 2 (metal oxide layer). It indicates that the effective diffusion is the parallel sum of Knudsen diffusion through porous scale and oxygen gas diffusion through the oxide scale layer. The effective diffusivity is saturated at 1×10−10 m2/s for Z, 2×10−11 m2/s for ZS and ZW and 1×10−12 m2/s for ZSW.

    5. Conclusions

    ZrB2 ceramics and ZrB2 ceramics with addition of SiC, WC and SiC/WC were successfully synthesized by spark plasma sintering method. SiC additive forms SiO2 amorphous outer scale layer and SiC-deplete ZrO2 scale layer. The amorphous SiO2 layer is considered to decrease the oxidation rate. WC addition forms WO3 during oxidation process to produce ZrO2/WO3 liquid sintering layer, which is more dense scale layer due to volume expansion effect. The addition of SiC and WC to ZrB2 has an effect to reduce the oxygen effective diffusivity by one fifth of that for ZrB2. The addition of both SiC and WC shows the formation of SiO2 outer dense glass layer and ZrO2/WO3 layer so that the anti-oxidation effect is improved three times as much as that of ZrB2. Therefore, ZSW has two order lower oxygen effective diffusivity than Z.

    Figure

    KPMI-26-6-455_F1.gif
    Torch oxidation experiment set-up scheme (a), the photographs of samples before (b) and after (c) torch oxidation experiment.
    KPMI-26-6-455_F2.gif
    FE-SEM images and their correspondent EDX elementary images after torch oxidation. (red; Zr, green : B, blue: oxygen, purple : Si, cyan : W) of the oxidized samples, (a) Z, (b) ZS, (c) ZW, and (d) ZSW. (The scale bar is 40 μm.)
    KPMI-26-6-455_F3.gif
    Schematic diagram of oxidation of Z (a), ZS (b), ZW(c), and ZSW (d).
    KPMI-26-6-455_F4.gif
    Weight increases (a), oxidation rate (b), scale thickness variations (c) and oxygen diffusivity variations (d) of the samples with reaction time at 1500°C in an ambient atmosphere.

    Table

    The compositions of the samples, Z, ZS, ZW and ZSW
    Oxidation layer thickness, measured and calculated weight increases after torch test
    Weight increases and oxidation rates from oxide thickness after torch test (sample size : φ= 10 mm, t = 1 mm) and from TGA test (sample size : about 6% of torch test sample)

    Reference

    1. F. Monteverde and A. Bellosi: J. Electrochem. Soc., 150 (2003) B552.
    2. W. C. Tripp and H. C. Graham: J. Electrochem. Soc., 118 (1971) 1195.
    3. A. Rezaie, W. G. Fahrenholtz and G. E. Hilmas: J. Eur. Ceram. Soc., 27 (2007) 2495.
    4. G.-J. Zhang, Z.-Y. Deng, N. Kondo, J.-F. Yang and T. Ohji: J. Am. Ceram. Soc., 83 (2000) 2330.
    5. S. S. Hwang, A. L. Vasiliev and N. P. Padture: Mater. Sci. Eng. A, 464 (2007) 216.
    6. J. Han, P. Hu, X. Zhang and S. Meng: Scr. Mater., 57 (2007) 825.
    7. S. N. Karlsdottir and J. W. Halloran: J. Am. Ceram. Soc., 92 (2009) 481.
    8. W.-M. Guo and G.-J. Zhang: J. Eur. Ceram. Soc., 30 (2010) 2387.
    9. W. G. Fahrenholtz: J. Am. Ceram. Soc., 90 (2007) 143.
    10. M. Mallik, K. K. Ray and R. Mitra: J. Eur. Ceram. Soc., 31 (2011) 199.
    11. S. C. Zhang, G. E. Hilmas and W. G. Fahrenholtz: J. Am. Ceram. Soc., 91 (2008) 3530.
    12. S. C. Zhang, G. E. Hilmas and W. G. Fahrenholtz: J. Am. Ceram. Soc., 94 (2011) 1198.
    13. J. Zou, S.-K. Sun, G.-J. Zhang, Y.-M. Kan, P.-L. Wang and T. Ohji: J. Am. Ceram. Soc., 94 (2011) 1575.
    14. J. Zou, G.-J. Zhang, C.-F. Hu, T. Nishimura, Y. Sakka, J. Vleugels and O. Van der Biest: J. Am. Ceram. Soc., 95 (2012) 874.
    15. T. A. Parthasarathy, R. A. Rapp, M. Opeka and R. J. Kerans: J. Am. Ceram. Soc., 92 (2009) 1079.
    16. T. A. Parthasarathy, R. A. Rapp, M. Opeka and R. J. Kerans: Acta Mater., 55 (2007) 5999.