Reactive ZrB 2 -induced oxidation resistance and kinetics in (Zr,Ti,W)C-based ceramics at 1000°C

Abstract

This study systematically investigates the high-temperature oxidation behaviour of (Zr,Ti,W)C-Me _x B _y multiphase ceramics fabricated by spark plasma sintering, focusing on their performance in static air at 1000°C–1100°C (operating temperature range for divertor components in nuclear fusion devices). The oxide layers consist predominantly of t-TiO₂, m-ZrO₂ and (Ti, Zr)O₂/(Zr,Ti)O₂ solid solutions. At 1000°C, oxidation follows parabolic kinetics, suggesting diffusion-controlled growth. At 1100°C, linear kinetics prevail as enhanced volatilisation of WO₃ and B₂O₃ leads to porous microstructures. The outer oxide layer develops voids due to WO₃ sublimation, while the inner layer remains dense owing to B₂O₃ filling the pores and cracks. The multiphase oxide structure, comprising (Zr,Ti)O₂ and (Ti,Zr)O₂ solid solutions along with dispersed ZrO₂ and TiO₂ particles, substantially enhances the oxidation resistance. The incorporation of ZrB₂ significantly increases the apparent activation energy of oxidation from 13.3641 kJ/mol (TW) to 47.4178 kJ/mol (TW60ZB), representing a 355% improvement in oxidation resistance. The key mechanisms include prolonged oxygen diffusion paths due to low-diffusivity ZrO₂ regions, grain boundary pinning by interphase boundaries and microstress fields at TiO₂/ZrO₂ interfaces that deflect microcracks. These results demonstrate the promising potential of (Zr,Ti,W)C-Me _x B _y ceramics for high-temperature applications such as nuclear fusion divertor components.

Keywords

(Zr,Ti,W)C oxidation behaviour oxidation kinetics

Introduction

Ultra-high-temperature ceramics (UHTCs) are in urgent demand for advanced equipment in aerospace, nuclear energy systems and hypersonic vehicles, yet their oxidation resistance in extreme environments still faces severe challenges.^1–3 Refractory metal carbides (e.g., ZrC, TiC, WC), with their high melting points, excellent high-temperature strength and good wear resistance, are considered promising candidates for UHTCs, with broad application prospects in aerospace thermal protection, nuclear fusion reactors, cutting tools and other fields.^4–8 However, these materials are prone to severe oxidation in high-temperature oxygen environments, leading to structural degradation and performance failure. Particularly, single-component carbides exhibit poor high-temperature oxidation resistance,^9–12 significantly limiting their engineering application prospects.^9,10,13–15

To enhance the oxidation resistance of refractory carbides, a synergistic strengthening strategy commonly employed involves multiple solid solutions combined with second-phase incorporation. Our previous study utilised a cation-anion dual solid solution (Zr,Ti)(C,N) and introduced SiC through an in-situ reaction, which significantly improved the oxidation resistance of the composite material at 850°C–950°C.¹⁶ Kang et al.¹⁷ enhanced the material's oxidation resistance by employing (Ti,W,Mo,Ta)(C,N) multi-component solid solution and second-phase composite effects.

Borides such as ZrB₂ exhibit certain oxidation resistance and thermal shock resistance at high temperatures.^18,19 ZrB₂ reacts with O₂ at temperatures above 650°C to form ZrO₂ and B₂O₃.²⁰ During oxidation, a ZrO₂ framework and B₂O₃ glass phase are formed, with B₂O₃ effectively filling cracks and pores within the oxide layer, thereby hindering further oxygen diffusion.^1,20–22 Consequently, ZrB₂ is considered an effective means to enhance the oxidation resistance of refractory metal carbides and is often used synergistically with SiC to improve the oxidation resistance of ZrC ceramics. Studies on the high-temperature oxidation behaviour of the ZrC-ZrB₂-SiC system^23,24 indicate that oxidation below 1200°C is reaction-controlled, while above 1300°C it becomes diffusion-controlled. At 1500°C, a ZrO₂ oxide layer with large pores forms, leading to a decline in oxidation resistance.

In our previous study,²⁵ (Zr,Ti,W)C-Me _x B _y multiphase ceramics were prepared by adding varying contents of ZrB₂ into a (Ti,W)C matrix, achieving synergistic enhancement in both strength and toughness of the material. On the one hand, the (Zr,Ti,W)C solid solution integrates the advantages of each component, improving material strength while maintaining a high melting point. On the other hand, ZrB₂ can react in-situ with the matrix to form platelet-like toughening phases such as TiB₂ and W₂B₅. However, there remains a lack of systematic understanding regarding the microstructural evolution, oxide layer formation mechanisms and kinetic behaviour of (Zr,Ti,W)C-Me _x B _y multiphase ceramics during oxidation at 1000°C–1100°C (the primary operating temperature range for divertor components in nuclear fusion devices). Particularly, the influence of ZrB₂ content on the phase composition, structural stability and protective performance of the oxide layer under coupled vapourisation-diffusion effects remains unclear, warranting further investigation. Notably, B₂O₃ exhibits significantly reduced protective capability due to its decreased viscosity and vapourisation at temperatures around 1000°C,²⁶ while WO₃ generated from the oxidation of W in the material also undergoes intense vapourisation within this temperature range,²⁷ collectively leading to a fundamental transformation in the oxidation mechanism. Therefore, studying the oxidation behaviour in this critical 1000°C–1100°C temperature range is decisive for elucidating the oxidation mechanisms of (Zr,Ti,W)C-Me _x B _y multiphase ceramics.

In this study, two characteristic temperature points of 1000°C and 1100°C were selected to systematically investigate the high-temperature oxidation behaviour of (Zr,Ti,W)C-Me _x B _y ceramics in static air. The phase evolution and microstructural characteristics of the oxide layer were revealed. Combined with thermogravimetric analysis and oxidation kinetic fitting, their oxidation mechanisms and variations in apparent activation energy were elucidated. This research aims to provide theoretical foundations and experimental support for the design and application of components such as nuclear fusion device divertors or oxidation-resistant structural parts.

Experimental procedure

(Zr,Ti,W)C-Me _x B _y multiphase ceramics were fabricated by Spark Plasma Sintering (SPS) according to our previous study.²⁵ The experiment used (Ti_0.602W_0.193)C powder (purity > 99.2 wt%, average particle size 2.05 μm) and ZrB₂ powder (purity > 99.9%, average particle size 1 μm) as raw materials. The powders were placed in a polyethylene ball milling jar according to the designed ratio, with silicon nitride (Si₃N₄) as the grinding medium and a ball-to-powder mass ratio of 5:1, and milled at 150 rpm for 24 h. After ball milling, the powder was sieved through a 120-mesh screen. Subsequently, an SPS system (SPS-10T-10-1 V, China) was used for sintering: the powder was first cold-pressed, then heated to 1800°C at a heating rate of 100°C/min under a vacuum environment of <0.1 Pa, held for 10 min and subjected to a uniaxial pressure of 50 MPa throughout the sintering process. After sintering, the samples were rapidly cooled to room temperature with the furnace. The relative densities of the sintered ceramics, as reported in our previous study,²⁵ were 88.8%, 96.8%, 98.7% and 98.3% for TW0ZB, TW40ZB, TW50ZB and TW60ZB, respectively. Table 1 presents the composition design, sintering parameters and relative density of the (Zr,Ti,W)C-Me _x B _y multiphase ceramics used in this study.

Table 1.

The compositional design and sintering schedule and relative density for (Zr,Ti,W)C-Me _x B _y multiphase ceramics.

Identifications of samples	Compositions of raw materials (mol%)		Sintering parameters	Relative density (%)
Identifications of samples	(Ti, W)C	ZrB₂	Sintering parameters	Relative density (%)
TW0ZB	100	0	SPS/1800°C/10 min/50 MPa	88.8
TW40ZB	60	40	SPS/1800°C/10 min/50 MPa	96.8
TW50ZB	50	50	SPS/1800°C/10 min/50 MPa	98.7
TW60ZB	40	60	SPS/1800°C/10 min/50 MPa	98.3

The (Zr,Ti,W)C-based multiphase ceramic samples were cut into rectangular test blocks measuring 10 mm × 10 mm × 4 mm. The surfaces were sequentially ground and polished using SiC sandpaper with varying grit sizes. The pre-oxidation masses were precisely measured using an electronic balance (FA1004B, accuracy 0.1 mg). The samples were placed vertically. A large crucible with an inner diameter of 12 mm was used as the outer container, inside which a small crucible with an inner diameter of 7.5 mm was placed. The sample was then vertically placed into the large crucible. This configuration minimises the contact area between the sample and the crucible, thereby effectively reducing experimental errors. The mass measurements were performed as follows: an initial mass was recorded before oxidation; during the oxidation process, samples were removed after 1, 3, 5, 7, 9, 11, 13 and 15 h of exposure, cooled to room temperature and then weighed. To ensure data reliability, each sample was measured at least three times at each time point, and the most accurate value was taken as the final result. X-ray diffraction (XRD; X'Pert, Netherlands) was employed to analyze the phase composition of the sample surface layers and detached oxide powders, with the following test parameters: diffraction angle range 10°–90°, scanning speed 10°/min. Following oxidation, the samples were embedded in resin, and their cross-sections were sequentially ground and polished for microscopic examination using a scanning electron microscope (Hitachi SU5000, Japan) to observe the oxide layer microstructure, accompanied by energy-dispersive spectroscopy (EDS) for elemental composition analysis. Thermodynamic calculations to explore reaction energy changes were performed using HSC Chemistry 6.0 software.

Results and discussion

Oxidation kinetics

Figure 1 presents the thermogravimetric curves of the ceramics after oxidation at 1000°C and 1100°C. The oxidation weight gain of ceramic samples with varying ZrB₂ contents shows significant differences. At 1000°C, the order of oxidation weight gain is: TW > TW40ZB > TW50ZB > TW60ZB, while at 1100°C, the sequence changes to: TW40ZB > TW50ZB > TW > TW60ZB. The TW60ZB sample exhibits the smallest oxidation weight gain at both temperatures. Combined with the previous analysis of oxide layer depth, TW60ZB demonstrates the shallowest total oxide layer depth, the thickest dense inner oxide layer and the minimal mass change, collectively indicating its superior oxidation resistance.

Figure 1.

Evolution of weight gain per unit area (mg/mm²) over time during oxidation tests in static air for ceramics TW0ZB, TW40ZB, TW50ZB and TW60ZB at 1000°C and 1100°C.

The weight gain trend was successfully fitted using the kinetic equation (1):

\frac{Δ m}{S} = k_{n} \times t^{n} + c

(1)

where Δm represents the weight gain, S is the surface area of the ceramic, t denotes the oxidation time, k_n is the oxidation rate constant, n signifies the oxidation exponent and c corresponds to the constant oxidation before reaching isothermal conditions.

The value of n was determined by fitting experimental data using equation (1). As shown in Table 2, the exponent constants of the fitting function can be categorised into two types: the first involves n fluctuating around 1 (linear), while the second shows n fluctuating around 0.5 (parabolic). Linear and parabolic kinetic models are commonly used simplified models to describe the oxidation behaviour of pure ceramics. During pure metal oxidation, an oxide layer typically forms covering the surface; however, in ceramic oxidation processes, the presence of pores, cracks or volatile products within the oxide layer often leads to the formation of non-protective oxide layers.

Table 2.

Values of oxidation exponent (n) and coefficient of determination (R²) obtained by fitting curves using equation (1) for ceramics oxidised at 1000°C and 1100°C.

T (°C)	TW	TW40ZB	TW50ZB	TW60ZB
T (°C)	n (R²)	n (R²)	n (R²)	n (R²)
1000	0.89034 (0.99092)	0.48937 (0.98871)	0.49214 (0.99332)	0.49404 (0.99285)
1100	0.7171 (0.98003)	1.32752 (0.97267)	1.11795 (0.99185)	0.98428 (0.94885)

At both temperatures, the TW samples exhibit oxidation kinetic fitting exponents n close to 1, indicating their oxidation behaviour follows linear kinetics. This is attributed to the single-phase (Ti,W)O₂ oxidation product of TW, which forms a porous oxide layer structure under the combined action of CO, CO₂ and O₂, primarily due to WO₃ volatilisation. Although the TiO₂ therein provides some protective effect to the substrate, the overall oxidation resistance remains limited. In contrast, TW40ZB follows parabolic kinetics (n ≈ 0.5) at 1000°C, significantly differing from the linear behaviour of TW, demonstrating that ZrB₂ addition markedly alters the oxidation mechanism and oxidation resistance at moderate temperatures. Parabolic kinetics typically indicates the formation of protective oxide layers, where the growth rate is inversely proportional to the oxide layer thickness, gradually slowing with oxidation progress.

At 1000°C, TW40ZB, TW50ZB and TW60ZB all exhibit n values close to 0.5, confirming their oxidation weight gain follows parabolic law and the process is diffusion-controlled. However, deviations exist between the actual system and ideal models: oxygen diffusion through grain boundaries, presence of defects and CO/CO₂ gas generation from carbon oxidation in the material can all lead to porous oxide layers, accelerating oxidation and causing kinetic behaviour to deviate from the ideal parabola. Notably, at 1100°C, all ZrB₂-containing samples show n values close to 1, with oxidation behaviour transitioning to linear kinetics. This suggests the protective oxidation mechanism dominant at 1000°C weakens with rising temperature. The cause may be aggravated WO₃ volatilisation at high temperatures, forming numerous gas diffusion channels that render the oxide layer structure porous, degrading oxidation resistance.

The oxidation rate constant (K_p) for each ceramic sample at different temperatures was calculated from the relationship between the square of oxidation weight gain and oxidation time, with the corresponding values listed in Table 3. The results show that at the same oxidation temperature, the oxidation rate constant of the samples is significantly correlated with ZrB₂ content, confirming that ZrB₂ addition effectively enhances the oxidation resistance. For example, at 1000°C, the K_p value of TW ceramic is 16 times higher than that of TW60ZB. With increasing ZrB₂ content, the K_p value gradually decreases, further indicating enhanced oxidation resistance of the material.

Table 3.

Values of parabolic rate constant (K_p) obtained from slope of linear fitting and coefficient of determination (r) for ceramics oxidised at 1000°C and 1100°C.

Sample	T (°C)	K_p (mg²/cm⁴)	r
TW0ZB	1000	11.67993	0.96511
TW0ZB	1100	29.292	0.95081
TW40ZB	1000	1.854	0.99129
TW40ZB	1100	51.5032	0.81883
TW50ZB	1000	1.69743	0.99495
TW50ZB	1100	55.08034	0.90658
TW60ZB	1000	0.7316	0.99562
TW60ZB	1100	19.10225	0.84007

The apparent activation energy (E_a) of the oxidation process for each system was determined using the Arrhenius equation (equation (2)) and the coefficient of determination (r) of ceramic samples, including TW, TW40ZB, TW50ZB and TW60ZB. The apparent activation energy is a key parameter reflecting the high-temperature oxidation resistance of ceramic materials. In this study, the multiphase oxide layer formed by (Zr,Ti)O₂ and (Ti,Zr)O₂ solid solutions doped with ZrO₂ and TiO₂ effectively inhibited the diffusion of oxygen ions and matrix elements in the ceramic, reduced the overall oxidation rate and thereby significantly altered the apparent activation energy of the oxidation process.

According to the Arrhenius equation, the oxidation rate constant K_p at thermodynamic temperature T can be expressed as:

K_{p} = A \times \exp (- \frac{E_{a}}{RT})

(2)

where A is the pre-exponential or frequency factor; E_a is the apparent activation energy; R is the universal gas constant; and T is the absolute temperature.

The apparent activation energy was calculated by plotting the relationship curve between lnK_p and 1/T (Figure 2) and determining its slope, with the results summarised in Table 4. The results demonstrate that ZrB₂ addition significantly enhances the apparent activation energy of oxidation for TW-based ceramics. Within the 1000°C–1100°C temperature range, the oxidation resistance of ceramics improves with increasing activation energy. Although ZrB₂-containing ceramics exhibit similar activation energy values, comprehensive analysis of total oxide layer depth, inner oxide layer thickness and oxidation weight gain confirms TW60ZB possesses the optimal overall oxidation resistance. Compared to unmodified TW, its oxidation resistance improves by 355%.

Figure 2.

Arrhenius plots showing the temperature dependence of oxidation rate constants for four ceramic materials at oxidation temperatures of 1000°C and 1100°C.

Table 4.

Apparent activation energy (E_a) of TW0ZB, TW40ZB, TW50ZB and TW60ZB ceramics oxidised at 1000°C–1100°C.

Sample	E_a (kJ/mol)
TW0ZB	13.3641
TW40ZB	48.3194
TW50ZB	50.5770
TW60ZB	47.4178

Oxidation substance analysis

Figure 3 shows the XRD patterns of (Zr,Ti,W)C-based multiphase ceramics with varying ZrB₂ contents before oxidation. The reaction between (Ti,W)C and ZrB₂ generated (Ti,Zr)B₂, (Ti,W,Zr)C and (Zr,Ti,W)C solid solutions. In the 40ZB sample, in addition to the diffraction peaks of these solid solutions, residual (Ti,W)C diffraction peaks were observed; in the 50ZB sample, (Ti,W)C reacted completely with ZrB₂, exhibiting diffraction peaks of WB, W₂B₅, W₂B and WB₄; while in the 60ZB sample, residual ZrB₂ diffraction peaks were detected.

Figure 3.

X-ray diffraction (XRD) pattern of (Zr,Ti,W)C-Me _x B _y ceramics before oxidation.

Figure 4 presents backscattered electron images of the as-sintered ceramics. As shown, the black phase is identified as (Ti, Zr)B₂, while the grey phase corresponds to (Zr,Ti,W)C. However, further analysis reveals that the grey phase actually consists of two distinct phases: (Zr,Ti,W)C formed via in-situ reaction, and (Ti,W,Zr)C resulting from the solid solution of Zr into the residual (Ti,W)C. These two phases are distributed among plate-like (Ti, Zr)B₂ grains.²⁵

Figure 4.

Backscattered electron (BSE) images of (a) TW40ZB, (b) TW50ZB and (c–d) TW60ZB ceramics before oxidation.

Figure 5 shows the macroscopic morphologies of (Zr,Ti,W)C-based multiphase ceramics after oxidation at 1000°C and 1100°C for durations of 1, 3, 5, 7, 9, 11, 13 and 15 h, respectively. During oxidation, with prolonged oxidation time, the sample without ZrB₂ addition (0ZB) exhibited cracking at the edges of its prismatic structure after oxidation, while ZrB₂-added samples maintained overall structural integrity. Additionally, the surface oxidation products gradually turned yellow with increasing oxidation time, due to the oxidation of TiC to rutile-type TiO₂ and the oxidation of ZrC to monoclinic ZrO₂ (m-ZrO₂).²⁸ This is confirmed by the strong diffraction peaks observed near 27° and 36° from TiO₂ in the temperature range of 800°C–1200°C,²⁹ as well as the oxidation of ZrC to ZrO₂ and CO₂ above 900°C.³⁰ With increasing ZrB₂ content, cracking and spalling of the oxide layer were significantly reduced, with the 60ZB sample showing the lowest degree of cracking. From the macroscopic morphology, it can be observed that even under the highest experimental temperature and duration conditions, the TW60ZB effectively resists severe oxidation damage. As shown in Figure 5(b), after 15-h oxidation at 1100°C, compared to other samples, the TW60ZB sample maintained the higher structural integrity, demonstrating its superior high-temperature oxidation resistance.

Figure 5.

The morphology of (Zr,Ti,W)C-Me _x B _y ceramics after oxidation at (a) 1000°C and (b) 1100°C.

To determine the phase composition of the oxide layer in (Zr,Ti,W)C-based multiphase ceramics, the detached oxide layer was ground into powder and subjected to XRD analysis, with the results shown in Figure 6. Figure 6(a) reveals that the primary phases of all detached oxide layers are tetragonal TiO₂ (t-TiO₂, PDF#99-000-3236) and monoclinic ZrO₂ (m-ZrO₂, PDF#04-005-7378). Further analysis of Figure 6(b) and (d) demonstrates that in ZrB₂-added ceramic samples, with Zr atoms dissolving into the TiO₂ unit cell, the lattice parameter of TiO₂ increases, causing its diffraction peaks to shift to lower angles; Simultaneously, Ti atoms dissolving into the ZrO₂ unit cell reduce the lattice parameter of ZrO₂, inducing its diffraction peaks to shift to higher angles. In contrast, for ZrB₂-free pure TW ceramic samples, W atoms dissolving into the TiO₂ unit cell decrease the lattice parameter of TiO₂, resulting in high-angle shifts of its diffraction peaks.

Figure 6.

(a) X-ray diffraction (XRD) patterns and (b) expanded patterns of detached oxide layers of (Zr,Ti,W)C-Me _x B _y ceramics after oxidation at 1000°C for 15 h; (c) XRD patterns and (d) expanded patterns of detached oxide layers of (Zr,Ti,W)C-Me _x B _y ceramics after oxidation at 1100°C for 15 h.

After 15-h oxidation at 1000°C, XRD analysis revealed that the sample without ZrB₂ addition was oxidised to form (Ti,W)O _x solid solution. When the oxidation temperature was increased to 1100°C and maintained for 15 h, this sample underwent further oxidation, producing TiO₂ and WO₃. Due to the high-temperature volatility of WO₃, only TiO₂ was detected in the XRD pattern. For TW60ZB ceramics, the initial phases consisted of (Ti,Zr)B₂, (Ti,W,Zr)C, (Zr,Ti,W)C, as well as WB, W₂B₅, W₂B, WB₄ and unreacted residual ZrB₂ . After 15-h high-temperature oxidation, these phases ultimately transformed into (Ti,Zr)O₂, B₂O₃, WO₃, ZrO₂ and (Zr,Ti)O₂. Among these, WO₃ and B₂O₃ partially volatilised at high temperatures, while the remaining oxides collectively formed a multiphase oxide layer. Compared to a single-phase oxide layer, this multiphase oxide layer exhibited a denser structure, superior oxidation resistance and provided more effective protection to the underlying substrate.^31–33

To evaluate the oxidation tendencies of each phase, Figure 7 shows the standard Gibbs free energy curves of reactions between TiC, ZrC, WC, ZrB₂, TiB₂, WB, W₂B₅, W₂B and O₂ as a function of temperature. Thermodynamic analysis of the oxidation products indicates that the reaction forming CO₂ is more favourable in terms of Gibbs free energy compared to those forming CO or free carbon.

Figure 7.

Curves of Gibbs free energy change with temperature for different reactions.

The (Zr,W,Ti)C solid solution in this study has a crystal structure similar to ZrC, therefore its oxidation process can be described by referring to the oxidation model of ZrC, with the reaction equation expressed as equation (3):

(Zr, W, Ti) C + O_{2} \to W O_{3} + (Zr, Ti) O_{2} + C O_{2}

(3)

To simplify the thermodynamic expression, the influences of trace Ti and W elements can be neglected, and the competitive reaction pathways between Zr and the oxidation processes of TiC, WC and tungsten borides (W _x B _y ) can also be ignored. Similarly, the (Ti,W,Zr)C solid solution, which has a crystal structure analogous to TiC, can be treated using the oxidation model of TiC.

The Gibbs free energy variation trend clearly indicates that the oxidation reaction of W₂B₅ to form WO₃ and B₂O₃ proceeds most readily, with the reaction equation expressed as equation (4):

W_{2} B_{5} + 6 .75 O_{2} \to 2 W O_{3} + 2 .5 B_{2} O_{3}

(4)

Among all the phases considered, WC exhibits the highest thermodynamic stability,³⁴ with its oxidation reaction expressed as equation (5):

WC + 2 .5 O_{2} \to W O_{3} + C O_{2}

(5)

Therefore, based on the thermodynamic results, within the temperature range of 1000°C–1100°C, the ease of oxidation reactions for each phase in this ceramic system follows the order: W₂B₅> ZrB₂ > W₂B > TiB₂ > ZrC > TiC > WB > WC.

Oxidation microstructure characterisation

Figure 8 presents the cross-sectional microstructures and elemental distributions of the TW60ZB ceramic after oxidation at 1000°C and 1100°C for 15 h. At 1000°C (Figure 8(a) and (b)), irregularly distributed pores are observed within the oxide layer, likely originating from diffusion channels formed by the volatilisation of WO₃ and B₂O₃.³⁵ Energy-dispersive spectroscopy mapping reveals partial overlap of Zr and Ti between the inner and outer oxide layers, indicating the formation of Zr-rich (Zr,Ti)O₂ and Ti-rich (Ti,Zr)O₂ solid solutions. The W content in the outer layer is significantly lower than that in the inner layer, which is attributed to the volatilisation of WO₃ – formed by the reaction of W _x B _y with O₂ – in the temperature range of 800°C–900°C³⁶; the dense inner oxide layer suppresses WO₃ volatilisation, retaining more WO₃ within, whereas the vigorous reaction in the outer layer leads to substantial WO₃ loss. At 1100°C (Figure 8(c) and (d)), the overlap of Zr and Ti distributions persists, confirming the presence of both Zr-rich (Zr,Ti)O₂ and Ti-rich (Ti,Zr)O₂ phases in the inner and outer layers. Compared with the results at 1000°C, the W content further decreases due to the intensified oxidation reaction at elevated temperatures, which promotes more severe WO₃ volatilisation. During this process, a volatilisation-induced migration mechanism may transport some TiO₂ to the oxide surface via physical or transient chemical adsorption^37–40; as WO₃ continuously volatilises into the atmosphere, the transported TiO₂ remains on the surface, resulting in significant TiO₂ enrichment in the outermost oxide layer. A comparison of Zr content between the inner and outer layers shows that the outer layer is richer in Zr. This phenomenon may be related to the oxidation of ZrB₂ to form B₂O₃: B₂O₃ fills the voids in ZrO₂ and, during volatilisation at high temperatures, carries ZrO₂ toward the outer oxide layer, thereby enriching Zr in the outer region.^41,42

Figure 8.

Cross-sectional scanning electron microscope (SEM) and energy-dispersive spectroscopy (EDS) images of the oxidation outer layer (a) and oxidation inner layer (b) of TW60ZB multiphase ceramic after 15-h oxidation at 1000°C; the oxidation outer layer (c) and oxidation inner layer (d) after 15-h oxidation at 1100°C.

It should be clarified that B₂O₃ exists as an amorphous glassy phase in the temperature range of 1000°C–1100°C. According to the literature, B₂O₃ has a melting point of approximately 450°C,⁴³ above which it transforms into a low-viscosity glassy phase with good fluidity and filling ability, allowing it to effectively infiltrate pores and microcracks within the oxide layer to form a dense protective layer.^43,44 Although the volatilisation rate of B₂O₃ increases with temperature in this range, it predominantly remains in the glassy state within the oxide layer until complete volatilisation.⁴⁵ Therefore, the conclusion that B₂O₃ functions as a glassy phase for filling and densification within the experimental temperature range is widely accepted in the field of oxidation research on ZrB₂-based ultra-high temperature ceramics.^46,47

Figure 9 further identifies the phase composition of the oxide layer via point EDS. At 1000°C (Figure 9(a) and (b)), based on XRD and EDS analyses, the dark grey regions are identified as (Ti,Zr)O₂, the light grey regions as (Zr,Ti)O₂, the bright grey regions as WO₃ and the pale grey regions as ZrO₂. Additionally, a (Ti,W)O _x phase is detected at this temperature. As the temperature increases to 1100°C (Figure 9(c) and (d)), the (Ti,W)O _x phase disappears with the intensified volatilisation of WO₃, and TiO₂ is formed.

Figure 9.

Cross-sectional scanning electron microscope (SEM) images and energy-dispersive spectroscopy (EDS) analysis results of the oxidation inner layer (a) and outer layer (b) of (Zr,Ti,W)C-based multiphase ceramic after 15-h oxidation at 1000°C; the oxidation inner layer (c) and outer layer (d) after 15-h oxidation at 1100°C.

Collectively, Figure 8 and Figure 9 reveal a distinct temperature-dependent evolution of the oxide layer structure in the TW60ZB ceramic. At 1000°C, the oxide layer exhibits a bilayer structure comprising a porous outer layer and a relatively dense inner layer. The inner layer benefits from the filling of pores and microcracks by B₂O₃, which enhances the density and oxygen-barrier performance of the oxide layer. Meanwhile, the formation of (Ti,Zr)O₂ and (Zr,Ti)O₂ solid solutions further contributes to the structural integrity of the oxide layer. The Zr-rich phase is uniformly distributed as large blocks, whereas the Ti-rich phase exists as small agglomerated particles interspersed with ZrO₂ and (Ti,W)O _x ; such a multiphase composite structure effectively improves the density and oxidation resistance of the oxide layer.³¹ At 1100°C, the volatilisation of WO₃ and B₂O₃ intensifies, leading to increased porosity and cracking in the outer layer and a reduction in the protective efficiency of the oxide layer, although the inner layer retains a relatively dense structure.

Based on the above microstructural observations, the multiphase mixed oxide layer enhances oxidation resistance primarily through the following mechanisms. First, the oxygen ion detouring mechanism: oxygen ions must bypass ZrO₂ regions with low diffusion rates during transport, substantially lengthening the diffusion path and markedly reducing the overall oxygen diffusion rate, thereby inhibiting the oxidation reaction. This principle is analogous to the barrier network formed by second-phase particles (graphene) in the coatings studied by Wei et al.,⁴⁸ which forces corrosive media to detour, prolonging the diffusion path and reducing the overall diffusion rate. Second, the grain boundary pinning mechanism: phase boundaries between different phases can effectively pin grain boundaries, suppressing grain growth of the oxide layer at high temperatures. Fine grains and tortuous grain boundaries collectively hinder the short-circuit diffusion of oxygen ions and metal ions, maintaining the stability and density of the oxide layer even under prolonged high-temperature exposure. This observation aligns with the conclusion of Mao et al.,⁴⁹ who reported that phase boundaries (TiB₂-ZrO₂) in multilayer structures pin grain boundaries and improve oxidation resistance. Additionally, the mismatch in thermal expansion coefficients (CTE) between TiO₂ and ZrO₂ generates a micro-stress field near the phase interfaces, which helps deflect or terminate microcracks generated during oxidation, effectively inhibiting crack propagation and oxide layer spallation, thereby preserving the structural integrity of the oxide layer.⁵⁰

Other studies on the high-temperature oxidation behaviour of refractory carbides also indicate that mixed oxides often exhibit superior protective performance. For instance, during the high-temperature oxidation of ZrC-20TaSi₂, mixed oxides such as TaZr_2.75O₈ formed a thick, dense oxide layer with isolated pores, effectively hindering oxygen diffusion into the substrate.¹¹ Similarly, in the present study, the (Zr,Ti)O₂ solid solution forms a dense and continuous outer layer that protects the underlying porous ZrO₂ and TiO₂ layers, reducing the oxygen diffusion rate and progressively depleting the oxygen partial pressure within the composite, thereby effectively suppressing further oxidation of the substrate.⁵¹ Under identical oxidation conditions, the multiphase oxide layer composed of (Ti,Zr)O₂ and (Zr,Ti)O₂ solid solutions interspersed with TiO₂ and ZrO₂ exhibits superior oxidation resistance compared to a single-phase solid solution oxide layer. This conclusion is further supported by extensive literature: single-phase WC develops an oxide layer depth of approximately 1.35 mm after oxidation in air at 800°C for 2 h,⁵² whereas under the same conditions, the TWT sample investigated by Zheng et al. exhibits a reduced oxide layer depth of 812 μm,⁵³ indicating that the incorporation of MeC (Me = Ti, Ta) partially suppresses oxygen penetration through the porous WO₃ layer into the substrate, indirectly confirming the beneficial role of multiphase structures in enhancing oxide layer protectiveness⁵⁴; Song et al. introduced reinforcing phases such as YAG (yttrium aluminium garnet) and Al₂O₃ to form a multiphase structure, significantly improving the oxidation resistance of ZrB₂ ceramics – compared with pure ZrB₂ ceramics, the resulting multiphase ceramics exhibited markedly thinner oxide layers and higher oxide layer densification⁵⁵; furthermore, Xia et al. reported that multiphase oxide layers typically contain glass phases (SiO₂) or dense oxide phases (Al₁₈B₄O₃₃), which can infiltrate and fill pores and cracks within the oxide layer at elevated temperatures, forming a dense protective film that substantially reduces oxygen diffusion rates and effectively blocks oxygen penetration into the substrate.⁵⁶ Collectively, these findings demonstrate that the superior oxidation resistance of multiphase oxide layers compared to single-phase counterparts is well established in high-temperature ceramics such as ZrB₂ and SiC, as well as in metal matrix composites.

Based on the above microstructural analysis and literature corroboration, the TW60ZB ceramic exhibits outstanding oxidation resistance. In the TW60ZB composition, residual ZrB₂ reacts with oxygen to form ZrO₂ and B₂O₃. ZrO₂, which possesses excellent high-temperature chemical stability, serves as an oxygen barrier layer that reduces oxygen diffusion into the substrate, further enhancing the overall oxidation resistance of the material.⁵⁷ Considering the oxide layer thickness and activation energy analysis, TW60ZB demonstrates the best oxidation resistance among all tested samples.

Figure 10 shows the cross-sectional morphologies of the oxide layer in (Zr,Ti,W)C-based multiphase ceramics after 15-h oxidation at 1000°C and 1100°C. The oxidised interface exhibits a distinct layered structure, sequentially comprising the outer oxide layer, inner oxide layer and substrate, with a continuous oxidation gradient formed between them. The outer oxide layer contains numerous voids and cracks, while the inner oxide layer is relatively dense. The higher-density inner layer effectively hinders inward oxygen diffusion, thereby providing excellent protection to the substrate.

Figure 10.

Scanning electron microscope (SEM) images of oxidised cross-section of TW40ZB, TW50ZB and TW60ZB after oxidation at 1000°C and 1100°C for 15 h.

Quantitative analysis of the oxide layer thickness reveals that after oxidation at 1000°C, the TW60ZB sample exhibits an inner oxide layer thickness of 112.59 μm and a total oxide layer depth of 280.9 μm. When the temperature increases to 1100°C, the inner oxide layer thickness rises to 159.47 μm, with the total oxide layer depth reaching 440.61 μm. For comparison, existing studies show that pure ZrC can achieve an oxide layer depth of 550 μm after 1-h oxidation in air at 900°C,⁹ while ZrC-TiC composites reduce this to 200 μm under identical conditions.³¹ Remarkably, our study obtains shallower oxide layer depths under more severe conditions (higher temperature at 1100°C and longer duration of 15 h), a phenomenon analogous to that observed in ZrC + SiC + ZrB₂ composites after oxidation.⁵⁸ The experimental results demonstrate that ZrB₂ addition and the formation of (Zr,Ti)O₂ solid solution significantly enhance the material's oxidation resistance.

Numerous pores and microcracks are observed in the outer oxide layers of TW50ZB and TW60ZB samples. Their formation may be attributed to the high-temperature oxidation of volatile WO₃ and B₂O₃ generated from W _x B _y compounds, with these gaseous products escaping outward from the oxide layer interior, creating diffusion channels. Particularly in TW60ZB samples after 15-h oxidation at 1100°C, cracks between the oxide layer and substrate are visible, likely caused by CTE mismatch between them. During cooling, if the average linear CTE of the oxide layer (e.g., α_ZrC = 6.7 × 10⁻⁶ K⁻¹,¹¹ α_ZrO₂=10.5 × 10⁻⁶ K⁻¹,^59,60), thermal mismatch stress will induce tensile stress on the outer surface while the substrate experiences compressive stress,⁵⁹ leading to interfacial separation.

To visually compare the oxidation resistance of different samples, Figure 11 presents a line chart of oxide layer depth versus ZrB₂ content. The results show that with increasing ZrB₂ content, the inner oxide layer thickness gradually increases while the total oxide layer thickness progressively decreases. Among them, TW60ZB samples exhibit the thickest inner oxide layer and thinnest total oxide layer, demonstrating superior oxidation resistance.

Figure 11.

Total oxidation layer depth and oxidation inner layer thickness of (Zr,Ti,W)C-based multiphase ceramics with varying ZrB₂ content after 15-h oxidation at 1000°C and 1100°C.

The effect of ZrB₂ content on the oxidation behaviour of the composites exhibits a distinct temperature dependence, which can be attributed to the varying roles of oxidation products in densification and volatilisation at different temperatures.

At 1000°C, the B₂O₃ glassy phase derived from ZrB₂ oxidation possesses relatively high viscosity and low vapour pressure, enabling it to effectively fill pores and microcracks within the oxide layer. Consequently, the inner layer undergoes significant densification, which substantially impedes inward oxygen diffusion. As a result, the total oxide layer thickness decreases markedly with increasing ZrB₂ content.⁶¹ In this regime, the increase in inner layer thickness is negatively correlated with the reduction in total layer thickness,^61,62 indicating that inner layer densification serves as the dominant mechanism governing oxidation resistance.

At 1100°C, the oxidation mechanism shifts fundamentally due to enhanced volatilisation of key oxidation products. On one hand, WO₃ generated from the oxidation of tungsten begins to volatilise above 800°C–900°C⁶³ and undergoes severe volatilisation at 1100°C, leaving numerous gas escape channels in the outer layer. On the other hand, the viscosity of B₂O₃ decreases with increasing temperature, accompanied by an accelerated volatilisation rate, which compromises its ability to continuously fill voids, particularly in the outer layer. These combined effects result in a loose and porous outer layer structure, allowing rapid oxygen diffusion toward the inner layer. Although the inner layer thickens with increasing ZrB₂ content, it fails to suppress the growth of the total oxide layer thickness.⁶⁴

Collectively, the densification-dominated mechanism that governs oxidation resistance at 1000°C is rendered ineffective at 1100°C due to the pronounced volatilisation of WO₃ and B₂O₃. This transition from ‘inner-layer densification dominance’ to ‘outer-layer volatilisation dominance’ accounts for the temperature-dependent influence of ZrB₂ content on the oxidation behaviour of the composites.

Oxidation mechanism

Based on the above analysis, the oxidation mechanism schematic of TW60ZB after 15-h oxidation at 1000°C and 1100°C is illustrated in Figure 12. The diagram distinguishes different oxide layer regions using yellow and dark grey colours, visually presenting the oxidation process and product distribution.

Figure 12.

Schematic diagram showing oxidation mechanism of TW60ZB oxidised at 1000°C and 1100°C. (a) and (d) show the unoxidised state of the samples at 1000°C and 1100°C; (b) and (e) The samples during oxidation at 1000°C and 1100°C; (c) and (f) The samples after oxidation at 1000°C and 1100°C.

Figure 12 presents schematic diagrams illustrating the oxidation mechanism of TW60ZB at 1000°C and 1100°C. Figure 12(a) shows the initial state of TW60ZB before oxidation at 1000°C (unoxidised). Figure 12(b) illustrates the oxidation process of TW60ZB during 15 h of oxidation at 1000°C, depicting the gradual formation of the oxide layer, the filling of pores by the B₂O₃ glass phase and the formation of diffusion channels resulting from WO₃ volatilisation. Figure 12(c) presents the final oxidation mechanism of TW60ZB after 15 h of oxidation at 1000°C, revealing a complete multi-layered oxide structure: a porous volatile outer layer, a dense inner layer formed by B₂O₃ consolidation and the formation of (Zr,Ti)O₂ and (Ti,Zr)O₂ solid solution layers at the interface. Figure 12(d) shows the initial state of TW60ZB before oxidation at 1100°C (unoxidised). Figure 12(e) illustrates the oxidation process of TW60ZB during 15 h of oxidation at 1100°C, highlighting significantly enhanced volatilisation of WO₃ and B₂O₃ increased porosity and cracking in the outer layer, and an increased thickness of the dense inner layer accompanied by weakened protective capability. Figure 12(f) depicts the final oxidation mechanism of TW60ZB after 15 h of oxidation at 1100°C, showing the resulting oxide layer structure characterised by severe porosity in the outer layer and a relatively dense inner layer still containing volatile escape channels.

Under 1000°C oxidation conditions, the outer oxide layer primarily consists of monoclinic m-(Zr,Ti)O₂, (Ti,Zr)O₂ solid solution, (Ti,W)O_x and incompletely volatilised WO₃. Within the m-(Zr,Ti)O₂ and (Ti,Zr)O₂ composite oxide layer, ZrO₂ and TiO₂ particles are dispersed. These particles likely originate from WO₃ and B₂O₃ being transported to the surface through physical adsorption or micro-area chemical reactions during high-temperature outward volatilisation. The inner oxide layer contains significant amounts of unvolatilised WO₃ and B₂O₃ in addition to the aforementioned oxides. B₂O₃ fills pores in ZrO₂ and oxidation microcracks, thereby enhancing the density of the inner layer. The dense inner oxide layer effectively inhibits outward carbon diffusion and inward oxygen penetration, significantly improving the material's overall oxidation resistance.⁶⁵

When the oxidation temperature increases to 1100°C, substantial thickening of the entire oxide layer is observed due to intensified oxidation reactions. At this temperature, WO₃ becomes undetectable in the outer layer owing to aggravated high-temperature volatilisation, while TiO₂ and ZrO₂ contents significantly increase, attributable to migration effects induced by WO₃ and B₂O₃ volatilisation. Furthermore, the (Ti,W)O _x phase is no longer observed in either the inner or outer oxide layers, indicating complete oxidation of W elements into WO₃ through vigorous volatilisation at elevated temperatures. Under these conditions, rapid evaporation of B₂O₃ reduces the effectiveness of the oxidation barrier.^20,66

Conclusions

This study systematically investigated the high-temperature oxidation behaviour of (Zr,Ti,W)C-Me _x B _y multiphase ceramics prepared by SPS with varying ZrB₂ additions, in static air at 1000°C–1100°C (the primary operating temperature range for divertor components in nuclear fusion devices). The oxide layer primarily consists of rutile-type t-TiO₂, monoclinic m-ZrO₂ and (Ti,Zr)O₂ and (Zr,Ti)O₂ solid solutions. Volatilisation of W and B elements in the forms of WO₃ and B₂O₃, respectively, leads to porous structure formation in the outer oxide layer, while B₂O₃ effectively fills pores and microcracks in the inner layer, significantly enhancing density and thereby inhibiting inward oxygen diffusion and outward carbon migration. This multiphase oxide structure substantially reduces oxygen diffusion rate by impeding oxygen ion migration and pinning grain boundaries, thereby strengthening protective effects. At 1000°C, ZrB₂-containing ceramics follow parabolic oxidation kinetics, with the rate constant K_p decreasing markedly with increasing ZrB₂ addition. At 1100°C, vigorous WO₃ and B₂O₃ volatilisation render the oxide layer porous, transitioning oxidation behaviour to linear kinetics. ZrB₂ addition elevates oxidation activation energy from 13.3641 kJ/mol for (Ti,W)C to 47.4178 kJ/mol. When ZrB₂ content reaches 60 mol% (TW60ZB), the material exhibits optimal oxidation resistance, with 355% improvement over (Ti,W)C. This research provides theoretical foundations and experimental support of (Zr,Ti,W)C-Me _x B _y multiphase ceramics for the design and application in components such as nuclear fusion device divertors or oxidation-resistant structural parts.

Footnotes

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Nos. 52002098, 52002003 and 52402074), the Natural Science Foundation of Heilongjiang Province (LH2023E080) and the Natural Science Foundation of Hubei Province (Nos. 2025AFD102 and 2025AFD040).

ORCID iD

Boxin Wei

Funding

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was supported by the Natural Science Foundation of Hubei Province, National Natural Science Foundation of China, Natural Science Foundation of Heilongjiang Province (grant numbers 2025AFD040, 2025AFD102, 52002003, 52002098, 52402074, LH2023E080).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

References

Eakins

Jayaseelan

Lee

. Toward oxidation-resistant ZrB2-SiC ultra high temperature ceramics. Metall Mater Trans A 2010; 42: 878–887.

Zhang

Deng

, et al. A novel ultra-high temperature ceramic composite coating prepared by high-speed laser cladding and pack cementation on Ta–W alloys for higher plasma ablation resistance above 2300 °C. J Adv Ceram 2025; 14: 9221009.

Feltrin

Hedman

Akhtar

. Thermal properties and high-temperature ablation of high-entropy (Ti0.25V0.25Zr0.25Hf0.25)B2 coating on graphite substrate. J Adv Ceram 2024; 13: 1268–1281.

Yao

Wei

Hou

, et al. Core-shell structure-designed TiC-based ceramics with synergistic strength-toughness via low-temperature in-situ reaction sintering. Int J Refract Met Hard Mater 2025; 132: 107276.

Cheng

Xie

Liu

. Spark plasma sintering of TiC-based composites toughened by submicron SiC particles. Ceram Int 2013; 39: 5077–5082.

Cheng

Xie

Liu

. Spark plasma sintering of TiC ceramic with tungsten carbide as a sintering additive. J Eur Ceram Soc 2013; 33: 2971–2977.

Yung

D-L

Antonov

Hussainova

. Spark plasma sintered ZrC-mo cermets: influence of temperature and compaction pressure. Ceram Int 2016; 42: 12907–12913.

Wei

Yang

, et al. In-situ architecting of reinforcing/toughening units in (Ti,W)C composites via two-step reactive SPS for enhanced strength and toughness. J Eur Ceram Soc 2026; 46: 118338.

Gasparrini

Chater

Horlait

, et al. Zirconium carbide oxidation: kinetics and oxygen diffusion through the intermediate layer. J Am Ceram Soc 2018; 101: 2638–2652.

10.

Gherrab

Garnier

Gavarini

, et al. Oxidation behavior of nano-scaled and micron-scaled TiC powders under air. Int J Refract Met Hard Mater 2013; 41: 590–596.

11.

Grossman

. High-Temperature thermophysical properties of zirconium carbide. J Am Ceram Soc 1965; 48: 236–242.

12.

Lavrenko

Glebov

Pomitkin

, et al. High-temperature oxidation of titanium carbide in oxygen. Oxid Met 1975; 9: 171–179.

13.

Lou

Zou

, et al. In-situ fabrication and characterization of TiC matrix composite reinforced by SiC and Ti3SiC2. Ceram Int 2023; 49: 20849–20859.

14.

Shimada

Ishil

. Oxidation kinetics of zirconium carbide at relatively low temperatures. J Am Ceram Soc 1990; 73: 2804–2808.

15.

Wei

Zhuang

Yang

, et al. Achieving superior strength-toughness synergy in ZrC-based ceramics: an in-situ multiscale construction strategy via two-step reactive SPS process. J Adv Ceram 2026; 15: 9221263.

16.

Wang

Wei

Zhang

, et al. Oxidation behavior and kinetics of (Zr,Ti)(C, N)–SiC ceramics with enhanced oxidation resistance at 850 °C–950 °C. J Mater Res Technol 2023; 27: 6744–6755.

17.

Kang

Xie

Yuan

, et al. Enhanced mechanical properties and oxidation resistance of (Ti,W,Mo,Ta)(C,N)-doped binderless Ti(C,N)-based ceramics. J Eur Ceram Soc 2025; 45: 117543.

18.

Medri

Pinasco

Sanson

, et al. ZrB2-based laminates produced by tape casting. Int J Appl Ceram Technol 2012; 9: 349–357.

19.

Sonber

Murthy

TSRC

Subramanian

, et al. Investigations on synthesis of ZrB2 and development of new composites with HfB2 and TiSi2. Int J Refract Met Hard Mater 2011; 29: 21–30.

20.

Guo

W-M

Zhang

G-J

Kan

Y-M

, et al. Oxidation of ZrB2 powder in the temperature range of 650–800°C. J Alloys Compd 2009; 471: 502–506.

21.

Cheng

, et al. Fabrication and characterization of B4C–ZrB2–SiC ceramics with simultaneously improved high temperature strength and oxidation resistance up to 1600 °C. Ceram Int 2016; 42: 8000–8004.

22.

Igawa

Nagasaki

Ishii

, et al. Phase-transformation study of metastable tetragonal zirconia powder. J Mater Sci 1998; 33: 4747–4758.

23.

Zhang

Gao

, et al. Oxidation behavior of hot pressed ZrB2-ZrC-SiC ceramic composites. Int J Appl Ceram Technol 2014; 11: 178–185.

24.

Guo

W-M

Zhou

X-J

Zhang

G-J

, et al. Effect of si and zr additions on oxidation resistance of hot-pressed ZrB2–SiC composites with polycarbosilane as a precursor at 1500°C. J Alloys Compd 2009; 471: 153–156.

25.

Wei

Jiang

Wang

, et al. Dual plate-like grains of (Ti, Zr)B2 and W2B5 toughened solid solution composites fabricated via in-situ reaction spark plasma sintering of (Ti, W)C and ZrB2 powders. J Eur Ceram Soc 2024; 44: 6206–6222.

26.

Guo

Yang

, et al. B2O3-reinforced ablative materials with superior and comprehensive ablation resistance used in aerospace propulsion thermal protection systems. Polym Degrad Stab 2024; 223: 110740.

27.

Wang

Yuan

, et al. Effect of (ti,W)C/TaC addition on the early oxidation behavior of surface layer of WC-co cemented carbides. Corros Sci 2020; 174: 108857.

28.

Calatayud

Alarcón

. V-containing ZrO 2 inorganic yellow nano-pigments prepared by hydrothermal approach. Dyes Pigm 2017; 146: 178–188.

29.

Thamaphat

Limsuwan

Ngotawornchai

. Phase characterization of TiO2 powder by XRD and TEM. Nat Sci. 2008; 42: 357–361.

30.

Parthasarathy

Rapp

Opeka

, et al. Modeling oxidation kinetics of SiC-containing refractory diborides. J Am Ceram Soc 2012; 95: 338–349.

31.

Yung

D-L

Maaten

Antonov

, et al. Oxidation of spark plasma sintered ZrC-mo and ZrC-TiC composites. Int J Refract Met Hard Mater 2017; 66: 244–251.

32.

, et al. Superior synergistic oxidation resistance of medium-entropy carbide ceramic powders rather than multi-phase carbide ceramic powders. J Adv Ceram 2024; 13: 1223–1233.

33.

Chen

Wang

Chen

, et al. Nb- and Ta-doped (Hf,Zr,Ti)C multicomponent carbides with enhanced oxidation resistance at 2500 °C. J Adv Ceram 2024; 13: 332–344.

34.

Zheng

Zhang

, et al. High-temperature wear and cyclic oxidation behavior of (Ti, W)C reinforced stainless steel coating deposited by PTA on a plain carbon steel. Surf Coat Technol 2021; 425: 127736.

35.

Cao

Wang

, et al. Oxidation behavior of melt-infiltrated SiC–TiB2 ceramic composites at 500–1300 °C in air. Ceram Int 2021; 47: 9881–9887.

36.

Millner

Neugebauer

. Volatility of the oxides of tungsten and molybdenum in the presence of water vapour. Nature 1949; 163: 601–602.

37.

Kalha

Bichelmaier

Fernando

, et al. Thermal and oxidation stability of TixW1−x diffusion barriers investigated by soft and hard x-ray photoelectron spectroscopy. J Appl Phys 2021; 129: 195302.

38.

Xiao

Zhao

, et al. Tailored TiO2/WO3 composites for enhanced electrocatalytic and photocatalytic applications. Ceram Int 2025; 51: 13586–13596.

39.

Hao

Dou

Z-h

Wan

X-y

, et al. Interphase migration and enrichment of lead and zinc during copper slag depletion. Trans Nonferrous Met Soc China 2024; 34: 3029–3041.

40.

Palmas

Ampudia Castresana

Mais

, et al. TiO2-WO3 nanostructured systems for photoelectrochemical applications. RSC Adv 2016; 6: 101671–101682.

41.

Vinci

Reimer

Zoli

, et al. Influence of pressure on the oxidation resistance of carbon fiber reinforced ZrB2/SiC composites at 2000 and 2200°C. Corros Sci 2021; 184: 109377.

42.

Yan

Liu

Zhang

, et al. Synthesis of ZrB2 powders from ZrO2, BN, and C. J Am Ceram Soc 2016; 99: 16–19.

43.

Bongiorno

Först

Kalia

, et al. A perspective on modeling materials in extreme environments: oxidation of ultrahigh-temperature ceramics. MRS Bull 2006; 31: 410–418.

44.

Servadei

Zoli

Galizia

, et al. Development of UHTCMCs via water based ZrB2 powder slurry infiltration and polymer infiltration and pyrolysis. J Eur Ceram Soc 2020; 40: 5076–5084.

45.

Jarvand

Balak

. Oxidation response of ZrB2–SiC–ZrC composites prepared by spark plasma sintering. Synth Sinter 2022; 2: 191–197.

46.

Zhou

S-G

Guo

Y-J

Liu

. Simulation of ZrB2 oxidation behavior at constant temperature ambient. J Inorg Mater 2019; 34: 60.

47.

Saha

Pramanik

Banik

, et al. Cyclic oxidation behaviour of ZrB2-20 vol.%MoSi2 based ultra-high temperature ceramic matrix composite between 1100°C and 1300°C, (2022), https://doi.org/10.21203/rs.3.rs-1353928/v1.

48.

Wei

Gao

. Recent research advances on corrosion mechanism and protection, and novel coating materials of magnesium alloys: a review. RSC Adv 2023; 13: 8427–8463.

49.

Mao

Dong

, et al. Optimization of the oxidation behavior and mechanical properties by designing the TiB2/ZrO2 multilayers. Coatings 2019; 9: 00.

50.

Khan

Swain

Patra

, et al. Effect of Y2O3, TiO2, ZrO2 dispersion on oxidation resistance of W–Ni–Nb–Mo alloys. Int J Mater Res 2024; 115: 960–969.

51.

Keneshea

Douglass

. The diffusion of oxygen in zirconia as a function of oxygen pressure. Oxid Met 1971; 3: 1–14.

52.

Zhao

Utton

Tsakiropoulos

. On the microstructure and properties of Nb-12Ti-18Si-6Ta-5Al-5Cr-2.5W-1Hf (at.%) silicide-based alloys with Ge and Sn additions. Materials (Basel) 2020; 13: 3719.

53.

Zheng

Wei

Yang

, et al. Entropy-stabilization meets ZrB2: dual pathways to oxidation resistance of (Ti,Zr)B2-reinforced (Zr,Ti,W,Ta)C medium-entropy carbide ceramics at 1000 °C–1100 °C. J Mater Res Technol 2026; 40: 3409–3421.

54.

Sengupta

Manna

. Role of TiC and WC addition on the mechanism and kinetics of isothermal oxidation and high-temperature stability of ZrB2–SiC composites. High Temp Corros Mater 2024; 101: 57–83.

55.

Song

Wang

, et al. Influence of YAG content on the properties of YAG-ZrB2 multi-phase ceramics. Powder Metall Technol 2013; 31: 334–339,348.

56.

Wang

Zhang

, et al. Dual-feedback healing mechanism redefining anti-oxidation coatings in fiber-reinforced composites. J Adv Ceram 2024; 13: 1643–1654.

57.

Shi

Wang

Sun

, et al. Effect of the surface oxidation on the flexural strength of the ZrB2–SiC–ZrC ceramic. Mater Sci Eng A 2012; 546: 162–168.

58.

Han

Guo

. Oxidation behavior of ZrC-based composites in static laboratory air up to 1300°C. Int J Refract Met Hard Mater 2014; 46: 159–167.

59.

Lipke

Ushakov

Navrotsky

, et al. Ultra-high temperature oxidation of a hafnium carbide-based solid solution ceramic composite. Corros Sci 2014; 80: 402–407.

60.

Hannink

Kelly

Muddle

. Transformation toughening in zirconia-containing ceramics. J Am Ceram Soc 2000; 83: 461–487.

61.

Saha

. New frontiers in characterising ZrB2-MoSi2 ultra-high temperature ceramics, (2022), https://doi.org/10.48550/arXiv.2202.11162.

62.

Cheng

Zhang

, et al. Advances in ultra-high temperature ceramics, composites, and coatings. J Adv Ceram 2022; 11: 1–56.

63.

Zakaryan

Nazaretyan

Aydinyan

, et al. Joint reduction of NiO/WO3 pair and NiWO4 by Mg + C combined reducer at high heating rates. Metals (Basel) 2021; 11: 1351.

64.

Pogozhev

Potanin

Bashkirov

, et al. Self-propagating high-temperature synthesis of the heterophase materials in the Zr–Mo–Si–B System: kinetics and mechanisms of combustion and structure formation. Russ J Non-Ferr Met 2022; 63: 649–658.

65.

Rezaie

Fahrenholtz

Hilmas

. Evolution of structure during the oxidation of zirconium diboride–silicon carbide in air up to 1500°C. J Eur Ceram Soc 2007; 27: 2495–2501.

66.

Monteverde

. The thermal stability in air of hot-pressed diboride matrix composites for uses at ultra-high temperatures. Corros Sci 2005; 47: 2020–2033.

Reactive ZrB 2 -induced oxidation resistance and kinetics in (Zr,Ti,W)C-based ceramics at 1000°C–1100°C

Abstract

Keywords

Introduction

Experimental procedure

Results and discussion

Oxidation kinetics

Oxidation substance analysis

Oxidation microstructure characterisation

Oxidation mechanism

Conclusions

Footnotes

Acknowledgements

ORCID iD

Funding

Declaration of conflicting interests

References