Progress in the degradation of thermal and environmental barrier coating materials caused by calcium-magnesium-aluminum-silicate deposit

Abstract

Thermal and environmental barrier coatings play a crucial role in protecting high-temperature structural components in gas turbine engines. As turbine inlet temperatures continue to rise, corrosion challenges posed by dust, volcanic ash, and other particulate matter—collectively known as CMAS—have become increasingly severe. Understanding the reaction mechanisms between CMAS and these coatings, identifying the key factors influencing CMAS corrosion, and developing methods to inhibit CMAS infiltration are essential for advancing high-performance gas turbine engines. This review examines the origins of CMAS corrosion and summarizes recent research on CMAS corrosion mechanisms in thermal and environmental barrier coating materials. Additionally, the role of rare earth elements in CMAS corrosion and various strategies to mitigate CMAS effects are discussed. Finally, the review highlights potential directions for future research.

Graphical abstract

Keywords

CMAS corrosion / Thermal barrier coating / Environmental barrier coating / Ceramic matrix composites / High entropy

Highlight

1.	The historical overview of the CMAS corrosion in gas turbine engines is outlined.
2.	This review examines recent progress on the CMAS corrosion of thermal and environmental barrier coating materials.
3.	The role of rare earth elements in the CMAS corrosion and strategies to mitigate CMAS corrosion are summarized.

1. Introduction

Gas turbine engines are considered the “heart” of aircraft, continuously evolving toward higher thrust-to-weight ratios and greater efficiency to enhance maneuverability and economic performance [1,2]. Increasing the turbine inlet temperature is a key strategy for achieving these objectives [3,4]. The turbine inlet temperature of the aircraft engine, as shown in Fig. 1, has evolved over several decades from 1600~1700K in F100 to 2400K in AVENT, posing significant challenges for the longevity of hot-section components [5]. Fig. 2 illustrates the evolution of superalloy, thermal barrier coating, environmental barrier coating and ceramic matrix composite used in critical hot-section components of gas turbine engines [6]. Superalloys used as turbine blades have evolved from polycrystalline to single crystal to progressively improve the resistance to creep and high-temperature strength [7]. However, even high-performance nickel-based single-crystal superalloys are limited to a maximum temperature of approximately 1100 ℃, which remains insufficient to meet the temperature requirement of high-performance gas turbine engines [8,9]. To address this limitation, thermal barrier coatings (TBCs) have been developed to protect superalloys, which own low thermal conductivity and high melting points. TBCs help reduce the surface temperature of superalloys, thereby extending the service life of turbine blades [10]. With the advancement of TBC systems, the gas turbine engine of aircraft is continuously aiming for higher efficiency by pushing the turbine inlet temperature above 1400 ℃ [11]. However, as the demand for improved performance grows, the combination of thermal barrier coatings and superalloys struggles to deliver greater thrust.

Ceramic matrix composites (CMCs), which demonstrate exceptional high-temperature resistance, are anticipated to replace superalloys as preferred materials for hot-section components [12]. Although CMCs require less thermal insulation, their surfaces still necessitate an environmental barrier coating (EBC) to protect against corrosive water vapor [13]. Consequently, EBCs must possess compatible thermal expansion coefficients and excellent resistance to water vapor corrosion. Rare earth silicate ceramics are emerging as promising candidates due to their excellent resistance to water vapor corrosion and low thermal expansion coefficient [14].

Both superalloy and CMC blades depend on functional thermal or environmental barrier coatings to protect them from external damage [2,15]. In recent years, it has been discovered that foreign contaminants such as sand, dust, and volcanic ash (referred to as CMAS) can be ingested by gas turbine engines [16]. These contaminants originate from sources like deserts, sandstorms, volcanic eruptions, and airport runways [17]. Several notable CMAS corrosion events have occurred in aeroengines, such as engine failure in C17 military aircraft and NAVY V-22 rotorcraft, and volcano ash threats in Iceland as shown in Fig. 3 [18⇓-20]. At high temperatures, these deposits easily melt and react with thermal or environmental barrier coatings, leading to reduced strain tolerance, increased elastic modulus, and coating delamination, as shown in Figs. 4 and 5 [21]. Additionally, highly flowing CMAS molten salts can penetrate along the interface of column grains in the coating, ultimately approaching the substrate. As a result, CMAS corrosion has become a prominent research focus for developing protective coatings in gas turbine engines [22].

This article provides a historical overview of CMAS corrosion and summarizes recent research on CMAS corrosion mechanisms of thermal and environmental barrier coating materials. Given the challenges posed by CMAS corrosion, various mitigation strategies are also discussed. Finally, the article provides conclusions and future research perspectives in this evolving field.

2. Historical overview of CMAS corrosion and characteristics of CMAS

2.1. Historical overview of CMAS corrosion

Research into CMAS-induced corrosion has spanned over three decades. Fig. 6 illustrates the historical overview of this issue. Initial insights and research on coating corrosion due to environmental deposits were conducted by Toriz et al. [23], who identified that TBCs on turbine blades are exposed to extreme service conditions. Key failure mechanisms include thermal cycling, bond coat oxidation, and the deposition and corrosion caused by solid particles and molten debris in the gas stream. Subsequently, during the 1990 Gulf War, a decline in turbine performance was observed in Black Hawk helicopters [24]. Engine teardown and rehabilitation tests revealed that sand intrusion was the primary factor behind this performance degradation. In a sandy environment, fine particles (less than 10 μm) can bypass the particle separators and enter the cooling and combustion systems, not only depositing on the turbine blades but also blocking the cooling passages. Smialek et al. [24] found that the composition of sand varies by region but typically consists of calcium aluminum silicate glass, SiO₂, (Ca, Mg)CO₃, CaCO₃, and NaCl. In addition to sand, several commercial aircraft experienced engine failure after encountering volcanic ash clouds, including blockages of turbine blades and cooling passages. In response, Kim et al. [25] employed a Hot-End Test System to study the deposition behavior of volcanic ash on the hot-end components of aircraft engines. At elevated temperatures, volcanic ash can melt and adhere to blades. Thus, turbine inlet and blade surface temperatures are key factors influencing deposition behavior. Borom et al. [26] found that deposits from various environments and regions can melt at high temperatures, infiltrate TBCs, and cause their spallation. Electron probe microanalysis confirmed that the composition of the molten phase infiltrating the TBC included CaO, MgO, Al₂O₃, and SiO₂ (CMAS). The deposition and corrosive effects of CMAS can have potentially catastrophic consequences. For example, the eruption of the Eyjafjallajökull volcano in Iceland in 2010 generated vast ash clouds that caused widespread disruptions to air traffic, leading to significant economic losses for the aviation industry [27]. Over time, the complexity and severity of CMAS-induced corrosion have become increasingly evident. Therefore, comprehensive research into CMAS corrosion remains crucial.

2.2. Characteristics of CMAS

The composition of CMAS is highly complex and varies depending on operating conditions and geographic location. Moreover, the composition of the molten phase that infiltrates into the coating is not entirely consistent with that of the initial deposits, suggesting partial melting of the latter [30]. Smialek et al. [24] analyzed sands from Saudi Arabia and deposits from turbine engines, finding that SiO₂ was the dominant component, followed by smaller amounts of Al₂O₃, CaO, NiO, Fe₂O₃, MgO, and K₂O. Bansal et al. [31] examined dust samples from Middle Eastern countries, identifying the composition as 27.8% CaO, 4% MgO, 5% Al₂O₃, 61.6% SiO₂, 0.6% Fe₂O₃, 1% K₂O. Aygun et al. [32], using X-ray fluorescence spectroscopy, confirmed that the composition of sand was 50% SiO₂, 38% CaO, 5% MgO, and 4% Al₂O₃, with trace amounts of Na₂O, K₂O, and Fe₂O₃. Fig. 7 presents a summary of the CMAS composition from various sources, as compiled by Ericks et al. [33]. A comparison of these findings reveals that, despite the complexity and variability of these dust and deposit samples, their primary components—CaO, MgO, Al₂O₃, and SiO₂—are consistent, hence the acronym CMAS.

The diversity of CMAS compositions also leads to differences in their physical and chemical properties, particularly the melting point, which is one of the most crucial characteristics. Smialek et al. [24] found that the melting point of the deposits was 1137 ℃. After conducting differential thermal analysis on environmental deposits, Borom et al. [26] determined that the initial melting temperature was 1195 ℃. Naraparaju et al. [34] reported melting points of volcanic ash from Japan and Iceland at 1170 ℃ and 1150 ℃, respectively. Fig. 8 presents the differential scanning calorimetry (DSC) curve of CMAS with a composition of 33CaO-9MgO-13AlO_1.5-45SiO₂, as reported by Guo et al. [35] The CMAS fully melts at approximately 1230 ℃ and undergoes self-crystallization below its melting point. Poerschke et al. [22] conducted a systematic study on the effects of CMAS composition on its melting temperature and liquid constitution. Calculations and experimental measurements indicate that the melting temperature of CMAS with different compositions falls within the range of 1100~1250 ℃. Based on the ternary phase diagram of CaO-MgO-AlO_1.5, the influence of deposit composition on melting behavior and liquid constitution is clearly illustrated (Fig. 9) [36]. At 1300 ℃, some deposits are completely converted into liquid, while others are only partially molten. Variations in the Ca/Si ratio within the liquid phase occur at a relatively constant AlO_1.5 concentration, suggesting that, at this temperature, the Ca/Si ratio may be the most relevant parameter in determining the liquid behavior. Additionally, the inclusion of MgO significantly alters the stability range of the liquid composition (Fig. 9 (b) and (c)). Therefore, changes in CMAS composition have a profound impact on its melting behavior, thereby influencing its interaction with the coating.

At high temperatures, once CMAS melts on the surface of coatings, it undergoes a liquid flow stage and spreads out. Song et al. [37] collected volcanic ash from different regions and systematically studied its melting behavior. The melting process of volcanic ash can be divided into four stages based on geometric characteristics: shrinkage, deformation, hemisphere, and flow, as shown in as shown in Fig. 10 (a, c). Initially, the volcanic ash shrinks due to sintering, then adheres to the surface upon melting, followed by spreading and wetting on the surface, and finally, viscous flow. The transition temperature for each stage varies depending on the composition of the volcanic ash (Fig. 10 (b)). In-situ observations of the melting process of a typical model CMAS (33CaO-9MgO-13AlO_1.5-45SiO₂) were conducted as illustrated in Fig. 11. Similarly, as the temperature increases, CMAS first undergoes deformation (partial melting), then transitions to a molten phase rapidly, and finally completely melts, quickly wetting the substrate. As a result, the interaction between CMAS and the thermal or environmental barrier coating is predominantly influenced by its wettability and flowability on the coating surface.

The wettability and flowability of CMAS are closely related to its viscosity, making the investigation of its viscosity essential. Bansal et al. [31] calculated the viscosity of CMAS (7.8CaO-4MgO-5Al₂O₃-61.6SiO₂-0.6Fe₂O₃-1K₂O) using the Vogel-Fulcher-Tamman equation and Giordano’s model and found that viscosity decreases with increasing temperature. Webster et al. [38] tested the viscosity of CMAS with different compositions (Fig. 12). They observed that a reduction in the CaO/SiO₂ ratio corresponds to a significant increase in viscosity, an effect similarly observed with elevated AlO_1.5 levels or diminished MgO content. In contrast, the inclusion of FeO_1.5, YbO_1.5, and TiO₂ caused only minimal changes to the viscosity of CMAS. Meanwhile, the viscosity of CMAS calculated by the FactSage model aligns well with experimental results.

Additionally, Bansal et al. [31] conducted a systematic study of the thermomechanical properties of CMAS (7.8CaO-4MgO-5Al₂O₃-61.6SiO₂-0.6Fe₂O₃-1K₂O). They found that its density is 2.69 g/cm³, with a Young's modulus of 92 GPa and a shear modulus of 36 GPa. The Vickers hardness was reported as 6.3 ± 0.4 GPa, while the glass transition temperature and softening point were 706 ℃ and 764 ℃, respectively. CMAS remains amorphous even after being held at 850 ℃ for 10 hours; however, crystalline phases of CaSiO₃ and Ca₂Mg_0.5AlSi_1.5O₇ begin to precipitate at temperatures above 900 ℃. Zhang et al. [39] investigated the microstructural changes of CMAS (33CaO-9MgO-13AlO_1.5-45SiO₂) below its melting point, discovering that self-crystallization occurs at 1000 ℃. As temperature increases and holding time extends, both the number of crystalline phases and the thickness of crystal layers increase. At 1050 ℃, diopside and anorthite phases precipitate from CMAS, and at 1150 ℃, wollastonite also appears. Prolonged holding times cause diopside to transform into wollastonite and anorthite.

3. Interaction between CMAS and TBCs at high temperatures

As early as the 1950s, the National Advisory Committee for Aeronautics and the National Bureau of Standards in the US conducted a series of studies on frit enamel ceramic coatings aimed at aerospace applications, identifying alumina and zirconia-calcia as among the most successful [40]. In the 1960s, the National Aeronautics and Space Administration first applied CaO-ZrO₂/NiCr-type TBC to the exhaust nozzle of the X-15 aircraft, marking the debut of TBCs on manned spacecraft [41]. Since the 1970s, yttria-stabilized zirconia (YSZ), containing 7-8 wt.% Y₂O₃, emerged as the most widely used TBC material, owing to its low thermal conductivity and a thermal expansion coefficient compatible with the metallic substrate [42,43]. The TBC system comprises a superalloy substrate, bond coat, and top coat, as illustrated in Fig. 13 (a) [10]. However, ZrO₂ undergoes phase transformation at approximately 1200 ℃, inducing volumetric expansion and limiting its applicability at elevated temperatures. As turbine inlet temperatures continue to rise, the demand for advanced TBC materials to replace YSZ has intensified [44,45]. New TBC candidates, such as RE₂Zr₂O₇, RETaO₄, and REPO₄, are gaining significant attention [9,42,46]. Their thermal conductivity and thermal expansion coefficient have been thoroughly investigated (Fig. 13 (b)) [47] demonstrating significant potential for practical application. In service, TBCs are highly susceptible to CMAS corrosion, which can lead to cracking and spallation of the coating [48,49]. Currently, CMAS corrosion is a key factor limiting TBC development, and improving CMAS resistance remains a pivotal area of future research.

3.1. CMAS corrosion mechanism of ZrO₂-based TBC

Pure ZrO₂ undergoes a transformation from a tetragonal to a monoclinic phase at approximately 1200 ℃, accompanied by a volumetric expansion of 3% to 5% [50,51]. This transformation generates thermal stress between the layers, potentially resulting in interlayer failure [52,53]. As a result, ZrO₂-based coatings typically require the addition of stabilizing agents, such as CaO, MgO, or Y₂O₃, to maintain the stability of ZrO₂ at low temperatures. Under the influence of CMAS, ZrO₂-based coatings often experience damaged caused by both thermomechanical and thermochemical mechanisms [54]. In terms of the thermomechanical mechanism, Tseng et al. [55] found that the expansion strain caused by CMAS infiltration along pores and phase transformation in zirconates are the primary factors contributing to coating damage and spalling. Fang et al. [56] conducted experiments involving the coupling of thermal cycling and CMAS exposure on three types of coatings: YSZ, Yb₂O₃ and Y₂O₃ co-stabilized ZrO₂, and Gd₂Zr₂O₇. The results indicated that under thermal cycling, CMAS corrosion is mainly driven by channel cracks in the ceramic layer. As illustrated in Fig. 14, molten CMAS gradually infiltrates the pores of the top coat and solidifies upon cooling, leading to thermal mismatch and reducing the TBC’s strain tolerance. The channel cracks continue to expand, ultimately causing the progressive delamination of the TBC. The blockage of cooling holes by CMAS melt represents another thermomechanical damage mechanism, when CMAS melt adheres to the blade surface and blocks the cooling holes, the cooling efficiency of the blades decreases, and the mismatch in the coefficient of thermal expansion in the CMAS and the alloy readily induces cracks [57,58]. Qadeer et al. [59] characterized a nozzle guiding vane after CMAS corrosion. EDS analysis confirmed the coating to be YSZ, which was observed in SEM, along with cracks caused by CMAS infiltration. Notably, SEM also identified CMAS melt blocking the cooling holes and cracks resulting from cooling holes merging caused by the blockage (Fig. 15).

Regarding thermochemical mechanisms, Wu et al. [60] observed preferential grain boundary corrosion at the grain boundary in YSZ. As shown in Fig. 16, the composition within the grains remains close to that of 8YSZ, while EDS analysis at the grain boundaries reveals CMAS infiltration in these regions. This phenomenon is attributed to the tendency of the Y element to segregate at grain boundaries, increasing CMAS reactivity in these areas. Furthermore, the extent of corrosion becomes more severe with increasing Y₂O₃ content.

Li et al. [61] discovered the formation of a reaction layer between CMAS and YSZ during CMAS corrosion, as illustrated in Fig. 17 (a). The thickness of this reaction layer gradually increases with prolonged corrosion time (Fig. 17 (b)), and Y³⁺ ions are gradually depleted within the layer. This depletion weakens the suppression of ZrO₂ phase transformation, ultimately leading to the failure of the TBC [62,63]. Pujol et al. [64] further demonstrated that the dissolution-recrystallization reaction between YSZ and molten CMAS is a significant contributor of failure. Both ZrO₂ and Y₂O₃ dissolve in molten CMAS, and when ZrO₂ solubility in CMAS is exceeded, ZrO₂ grains re-precipitate through a recrystallization process (as shown in Fig. 17 (d)). In the absence of stabilizing effects from Y₂O₃ during this process, a transformation from the tetragonal phase to the monoclinic phase occurs, causing volume expansion. The stress generated by this phase expansion induces cracks propagate, ultimately leading to TBC delamination [62].

TBCs are predominantly fabricated using atmospheric plasma spraying (APS) or electron beam physical vapor deposition (EB-PVD), and the microstructures produced by these two methods are shown in Fig. 18 [65]. TBCs prepared by APS exhibit a distinct layered structure with minimal vertical cracks, which is effectively in mitigating CMAS infiltration. In contrast, TBCs fabricated by the EB-PVD method display a characteristic feather-like columnar crystal structure, though the vertical gaps within this structure facilitate CMAS infiltration. Shan et al. [66] studied the CMAS corrosion behavior of YSZ coatings produced by APS, as illustrated in Fig. 18 (c). Notably, some larger spherical pores in the YSZ remained unfilled by CMAS, suggesting that incorporating larger spherical pores could enhancing CMAS corrosion resistance by consuming CMAS. Wellman et al. [67] investigated the CMAS corrosion mechanisms in TBCs prepared by the EB-PVD method and observed that CMAS preferentially infiltrates the vertical gaps between the columnar grains (Fig. 18 (d)). This infiltration initiates corrosion at the column tips, propagating cracks into adjacent columns and ultimately leading to delamination of the ceramic layer. Although coatings fabricated by the EB-PVD method exhibit higher bond strength and damage tolerance, the vertical gaps between grains facilitate CMAS infiltration, thereby accelerating corrosion.

Overall, under high-temperature and CMAS exposure, YSZ experiences grain boundary corrosion due to the segregation of Y at the grain boundaries. Additionally, the dissolution and recrystallization process of ZrO₂ within the CMAS reaction layer induces phase transformations accompanied by volumetric expansion, compromising coating stability. Furthermore, the inherent vertical gaps between grains in YSZ provide channels for CMAS infiltration, exacerbating CMAS corrosion.

3.2. CMAS corrosion mechanism of RE₂Zr₂O₇

Although YSZ has been used for several decades, it undergoes phase transformation at temperatures above 1200 ℃. As turbine inlet temperatures continue to rise, YSZ’s applicability at temperatures above 1200 ℃ is considerably restricted. Rare earth zirconates (RE₂Zr₂O₇) exhibit lower thermal conductivity and similar thermal expansion coefficients compared to YSZ (as shown in Table 1 [68]) and lack phase transformations below their melting point, offering high phase stability, making them promising TBC candidates [69]. RE₂Zr₂O₇ can be categorized into two structural types: defective fluorite and pyrochlore, depending on the ionic radius ratio of RE³⁺ to Zr⁴⁺. Pyrochlore-structured RE₂Zr₂O₇ ceramics possess compatible thermal expansion coefficients with the superalloy, which making them extensively studied as promising TBC materials.

Stephan et al. [70] investigated the CMAS corrosion behavior of Gd₂Zr₂O₇ at 1300 ℃. Their results revealed that Gd₂Zr₂O₇ dissolved in the CMAS melt and formed an apatite phase predominantly composed of Gd₈Ca₂(SiO₄)₆O₂ and a fluorite phase rich in ZrO₂. This process involved substantial consumption of calcium and silicon from the melt, while magnesium and aluminum oxides precipitated as spinel within the melt (Fig 19 (a) and (b)). The combination of these three crystalline phases filled the crystal gaps, hindering further penetration of the remaining CMAS. The type of RE element significantly influences the CMAS corrosion behavior of RE₂Zr₂O₇. In Schulz et al.’s study [71], La₂Zr₂O₇ (which has a larger ionic radius of RE³⁺) exhibited superior resistance to CMAS corrosion compared to Gd₂Zr₂O₇, suggesting a correlation between the ionic radius of the RE element and CMAS corrosion resistance. Wu et al. [72] examined the CMAS corrosion resistance of Sm₂Zr₂O₇, Gd₂Zr₂O₇, and Yb₂Zr₂O₇. Their findings revealed that needle-like apatite and spherical cubic ZrO₂ were the main crystalline products formed in the CMAS melt. Additionally, the amount of apatite product increased with the cation radius of RE and decreased with a reduction in the CaO: SiO₂ ratio in the CMAS, as shown in Fig 19 (c). These observations provide guidelines for designing next-generation RE₂Zr₂O₇ ceramics with enhanced CMAS resistance.

Guo et al. [73] further investigated the corrosion resistance of Gd₂Zr₂O₇ under Fe-containing environments (CMAS-Fe) as shown in Fig. 20. The results demonstrated that Gd₂Zr₂O₇ continuously precipitated needle-like Gd-oxygen apatite, while calcspar precipitated in low iron content melts. The precipitation of calcspar reduced the Ca ratio in the melt, increasing the melt viscosity and slowing the penetration rate of CMAS. In the CMAS melt with the highest Fe content, garnet formed, which had a minimal effect on the viscosity on the melt viscosity and was ineffectively in resisting CMAS infiltration, leading to the weakest CMAS corrosion resistance. Moreover, during cyclic CMAS-Fe corrosion, Gd in the reaction layer continuously reacted with CMAS-Fe, and the re-deposition of molten CMAS-Fe altered the chemical composition of the reaction layer. This process converted the residual CMAS-Fe into a low-viscosity melt, which created penetration pathways for the re-deposition of CMAS-Fe, thereby diminishing the corrosion resistance of the ceramic layer. These findings revealed the shortcomings of Gd₂Zr₂O₇ in CMAS-Fe corrosive environments, providing direction for future modification strategies tailored to specific corrosive conditions for Gd₂Zr₂O₇.

Overall, rare earth zirconates dissolved in CMAS melts primarily form an apatite phase composed of RE₈Ca₂(SiO₄)₆O₂ and a ZrO₂-enriched fluorite phase. The apatite phase fills crystal gaps, thereby hindering further infiltration of residual CMAS. The formation of the apatite phase is influenced by both the type of RE element and the composition of CMAS. Specifically, the amount of apatite increases with the growing ionic radius of the RE element, thereby improving the CMAS corrosion resistance of RE₂Zr₂O₇. Conversely, a decrease in the CaO: SiO₂ ratio in CMAS reduces apatite formation, which in turn diminishes the CMAS corrosion resistance. Moreover, in iron-containing CMAS melts, RE₂Zr₂O₇ form a garnet phase, which does not effectively resist CMAS attack.

3.3. CMAS corrosion mechanism of RETaO₄

Rare earth tantalates are considered promising candidates for next-generation high-temperature TBC due to their low thermal conductivity (1.4~2.1 W·m⁻¹K⁻¹ at 900 ℃), high melting points, and good phase stability [74,75]. Depending on the stoichiometric ratio between RE₂O₃ and Ta₂O₅, rare earth tantalates can be classified into three types of ceramic materials: RETaO₄, RE₃TaO₇, and RETa₃O₉, with their crystal structures illustrated in Fig. 21 [76⇓⇓-79]. Among these, RETaO₄ is distinguished by its ferroelastic toughening properties, as its grains are composed of layered ferroelastic domains [75,78]. These domains reduce the energy required for phase transformation and efficiently absorb fracture energy during crack propagation, thereby enhancing the material’s fracture toughness.

Yang et al. [80,81] evaluated the CMAS corrosion resistance of GdTaO₄, YbTaO₄ and YSZ at 1350 ℃. The interface microstructure of GdTaO₄ and CMAS is shown in Fig. 22 (a), with the main reaction product confirmed to be Ca₂Ta₂O₇. The surface state of the samples indicated that CMAS could not infiltrate the GdTaO₄ substrate, instead accumulating on the surface to form a CMAS glass. Furthermore, CMAS contact angle tests demonstrated that GdTaO₄ has poorer wettability than YSZ (Fig. 21 (b)), which contributes to reduce degradation at high temperatures. As a result, GdTaO₄ exhibits superior CMAS corrosion resistance compared to YSZ. Similarly, the main CMAS corrosion product of YbTaO₄ is also Ca₂Ta₂O₇. The dense layer formed on the surface of this product effectively hiders CMAS penetration. Under the same exposure duration, YbTaO₄ shows a thinner reaction layer thickness compared to GdTaO₄, as shown in Fig. 22 (d) which can be attributed to the lower cohesive energy (Fig. 22 (c)).

Tian et al. [82] studied the CMAS corrosion behavior of YTaO₄ (Fig. 23), and the primary corrosion product was identified as the solid solution (Ca_2-xY_x)(Ta_2-y-zMg_yAl_z)O₇. In addition, intergranular corrosion was observed in M-YTaO₄. This form of corrosion provides a rapid pathway for CMAS infiltration and dominates the entire corrosion process (Fig. 23 (a) and (b)). Typically, the interaction between CMAS and TBC reduces strain tolerance, leading to the formation of cracks. However, even after 80 hours of corrosion at 1300 ℃, no delamination cracks were observed in M-YTaO₄ (Fig. 23(a)). To investigate the mechanical properties of the interface, nanoindentation test was performed (Fig. 23 (c)). The modulus decreases smoothly from the reaction layer to M-YTaO₄ without significant modulus mismatch across the layers, demonstrating good reliability during CMAS corrosion.

RETaO₄ exhibits excellent resistance to CMAS infiltration due to its inert reaction and poor CMAS wettability. Additionally, its intrinsic ferroelastic toughening characteristic makes RETaO₄ a promising candidate for next-generation ultra-high-temperature TBCs. However, further research should focus on grain boundary CMAS corrosion to enhance its overall applicability.

3.4 CMAS corrosion mechanism of REPO₄

Rare earth phosphates are also promising candidate materials for TBCs due to their unique chemical and thermal stability [83]. Song et al. [84] studied the CMAS corrosion resistance of LaMgAl₁₁O₁₉/GdPO₄ (LMA/ GdPO₄). CMAS reacted with GdPO₄ to form the apatite-phase compound Ca_2+xGd_8-x(PO₄)_x(SiO₄)_6-xO₂. Observation of the cross-sectional view of the reaction layer (Fig. 24 (a-c)) revealed that a higher GdPO₄ content in the system led to a thinner reaction layer. This findings suggests that increasing the amount of GdPO₄ in the LMA/GdPO₄ system enhances CMAS corrosion resistance. Meng et al. [85] found that the reaction layer exhibited a dual-layer structure, consisting of a needle-like upper layer and a dense lower layer in the CMAS-attacked GdPO₄ sample. The discontinuous needle-like upper layer suppressed CMAS diffusion, while the lower layer effectively inhibited CMAS infiltration (Fig. 24 (d-g)). Surface energy calculation results indicated that GdPO₄ has the lowest surface energy, which is favorable for its CMAS wetting resistance.

Qin et al. [86] studied the CMAS corrosion resistance of LaPO₄ at 1250 ℃ and found that LaPO₄ reacted with CMAS early in the process, forming a protective layer of Ca₈LaMg(PO₄)₇ apatite. Additionally, an interlayer was formed between the reaction layer and LaPO₄ due to solution substitution and atomic rearrangement at the LaPO₄-CMAS interface (Fig. 25 (c)). To explain this reaction layer, which differs from the “dissolution-precipitation process”, a solution/diffusion-recrystallization theory was proposed (Fig. 25 (d)). This dual-layer structure prevents the formation of cracks or pores between LaPO₄ and the reaction layer, thereby ensuring the sealing effect of the reaction layer against CMAS. Due to its low surface energy and dual-layer reaction structure formed during CMAS corrosion, LaPO₄ demonstrates high resistance to CMAS wetting and infiltration, making it a promising candidate for next-generation TBCs.

A comparison of the CMAS corrosion resistance among the discussed TBCs is presented in Fig. 26. Under similar corrosion temperatures and exposure times, RETaO₄ and REPO₄ exhibit lower corrosion depths compared to ZrO₂-based and RE₂Zr₂O₇ TBCs, demonstrating their superior resistance to CMAS corrosion. However, grain boundary CMAS corrosion was found in RETaO₄, indicating the need for further improvement. Therefore, more extensive research is needed to explore the factors influencing CMAS corrosion, understand the underlying corrosion mechanisms, and develop methods for enhancing corrosion resistance, which will aid in the design and optimization of TBCs.

4. Degradation of EBCs caused by CMAS at high temperatures

Despite significant advancements in TBC technologies, nickel-based superalloys remain insufficient for meeting the extreme temperature requirements of next-generation gas turbine engines. SiC_f/SiC composites, with their exceptional high-temperature performance, offer a viable alternative for use in hot-section components of gas turbines. EBCs were developed to ensure material longevity and performance integrity [92⇓⇓⇓-96]. Researchers have made substantial progress in the development, fabrication techniques, and performance evaluation of EBC materials, which has led to the evolution of multiple generations of EBC systems [2]. The first- and second-generation EBCs were primarily based on mullite (3Al₂O₃·2SiO₂) and BSAS (1-xBaO-xSrO-Al₂O₃-2SiO₂, 0≤x≤1) as core materials. However, their stability at high-temperature is limited, rendering them prone to destabilization and spalling. As a result, these coatings are restricted to service temperatures below 1300 ℃ [97,98]. Currently, rare earth silicates (RE₂SiO₅ and RE₂Si₂O₇) are widely recognized as the preferred EBC systems due to their superior high-temperature stability and corrosion resistance [99].

4.1. CMAS corrosion behavior of rare earth monosilicate (RE₂SiO₅)

RE₂SiO₅ consists of [SiO₄] tetrahedra and [REO_x] polyhedral, with its crystal structure illustrated in Fig. 27. It typically forms the X1 phase with larger RE elements (La ~ Gd), while the X2 phase forms with smaller RE cations (Tb ~ Lu) [100,101]. The distinction between X1-RE₂SiO₅ and X2-RE₂SiO₅ lies in the coordination numbers of the RE atoms: in X1-RE₂SiO₅, the RE ions are coordinated in 9- and 7- coordination arrangements, whereas in X2-RE₂SiO₅, they are coordinated in 7- and 6- coordination arrangements, respectively.

Given the diversity of RE elements and their effects on crystal structure, the specific RE element species play a crucial role in influencing the CMAS corrosion behavior of RE₂SiO₅. Cao et al. [102] studied the CMAS corrosion behavior of Y₂SiO₅ and Yb₂SiO₅ at 1300 ℃. Y₂SiO₅ exhibited a much more vigorous reaction with CMAS than Yb₂SiO₅. After 32 hours, the reaction layer thickness of Y₂SiO₅ was four times that of Yb₂SiO₅. Additionally, Cao et al. [103] compared the CMAS corrosion resistance of Y₂SiO₅ and Yb₂SiO₅ at 1400 ℃, finding that after 12 h, the reaction layer thickness of Yb₂SiO₅ was less than half that of Y₂SiO₅. This difference is attributed to the greater optical basicity (OB) difference between Y₂SiO₅ and CMAS compared to that of Yb₂SiO₅, leading to a higher reactivity of Y₂SiO₅ with CMAS. Zhong et al. [104] compared the CMAS corrosion behaviors of X1-Gd₂SiO₅, X2-Y₂SiO₅, and X2-Er₂SiO₅ coatings. The CMAS corrosion resistance of the three was ranked as follows: X2-Er₂SiO₅ > X2-Y₂SiO₅ > X1-Gd₂SiO₅. The primary product of CMAS corrosion was Ca₂RE₈(SiO₄)₆O₂, and since the ionic radius of Gd is closer to that of Ca, X1-Gd₂SiO₅ is more prone to reacting with CMAS to form Ca₂RE₈(SiO₄)₆O₂, resulting in poorer CMAS corrosion resistance. Furthermore, the crystal structure suggests that X1- Gd₂SiO₅ has a weaker crystal field strength and weaker chemical bonds, making it more susceptible to corrosion. In contrast, X2-Er₂SiO₅ generates garnet-like substances during corrosion, which help form a dense apatite layer, thereby enhancing its CMAS corrosion resistance.

Tian et al. [100,105] systematically studied the CMAS corrosion behavior of X1-RE₂SiO₅ (RE = La, Nd, Sm, Eu, Gd) and X2-RE₂SiO₅ (RE = Tb, Dy, Ho, Er, Y, Tm, Yb, and Lu) ceramics at 1300 ℃. After 25 hours of interaction with CMAS, the thickness of the recession layer in X1-RE₂SiO₅ ceramics increased with decreasing RE ionic radius, as shown in Fig. 28 (a). This indicates that the CMAS corrosion resistance of X1-RE₂SiO₅ ceramics is related to the RE element species, with larger RE ionic radii correlating to better corrosion resistance [100]. In contrast, for X2-RE₂SiO₅ ceramics, the recession layer thickness after 50 and 100 hours of corrosion at 1300 ℃ exhibited a linear relationship with the RE ionic radius, as shown in Fig. 28 (b). Based on the CMAS corrosion behavior of X2-RE₂SiO₅, three distinct groups can be identified (Fig. 28 (c)). For RE elements with relatively larger ionic radii, such as Tb₂SiO₅, Dy₂SiO₅, and Ho₂SiO₅, the CMAS is almost entirely consumed, resulting in the thickest reaction layer. This indicates an intense interaction between the CMAS and RE₂SiO₅. In contrast, for RE elements with smaller ionic radii, such as Tm₂SiO₅, Yb₂SiO₅, and Lu₂SiO₅, a considerable amount of residual CMAS remains, the reaction zone is comparatively thin, and no cavities are observed, suggesting a mild CMAS corrosion process [105]. During the CMAS corrosion process of X2-RE₂SiO₅, the main reaction product was Ca₂RE₈(SiO₄)₆O₂. Research by Costa et al. [106] found that as the RE ionic radius increased, the formation enthalpy of Ca₂RE₈(SiO₄)₆O₂ became more exothermic, indicating improved stability. Thus, X2-RE₂SiO₅ with a larger RE ionic radius tends to react more readily with CMAS to form Ca₂RE₈(SiO₄)₆O₂, resulting in poorer corrosion resistance. On the other hand, the reactivity of X2-RE₂SiO₅ with CMAS can also be assessed by the difference in their OB [107]. A smaller OB difference indicates lower reactivity. Calculations reveal that as the RE ionic radius decreases, the OB difference between X2-RE₂SiO₅ and CMAS molten salt becomes smaller, leading to lower reactivity and thus better CMAS corrosion resistance.

Since temperature significantly affects the rate of chemical reactions, Tian et al. also [108,109] investigated the CMAS corrosion behavior of RE₂SiO₅ (RE = Tb, Dy, Ho, Y, Er, Tm, Yb, and Lu) at 1500 ℃. The corrosion product was consistently apatite, Ca₂RE₈(SiO₄)₆O₂. In RE₂SiO₅ (RE = Tb, Dy, Ho, Y, Er, Tm, and Yb), apatite displayed a random orientation, and high-temperature in-situ observations indicated that numerous smaller, laterally arranged apatite crystals precipitated during the cooling process. Conversely, in Lu₂SiO₅, apatite was predominantly aligned longitudinally, as illustrated in Fig. 29. As the temperature increases, the viscosity of CMAS decreases, leading to accelerated reaction rates and more severe CMAS-induced corrosion of RE₂SiO₅ at 1500 °C relative to its behavior at 1300 ℃. Under consistent corrosion durations, the CMAS infiltration depths in RE₂SiO₅ (RE = Tb, Dy, Ho, Y, Er, Tm, and Yb) exhibit a slight reduction with increasing RE ionic radius (Fig. 30), suggesting that the CMAS corrosion behavior is minimally influenced by the RE element species. However, Lu₂SiO₅ demonstrates superior resistance to CMAS corrosion at 1300 ℃ but undergoes the most significant degradation at 1500 ℃ (Fig. 31 (a)). At 1300 ℃, a small amount of disorderedly arranged apatite forms at the reaction interface (Fig. 31 (b)). In contrast, at 1500 ℃, Lu₂SiO₅ undergoes decomposition into Lu₂Si₂O₇ and Lu₂O₃. The Lu₂Si₂O₇ preferentially reacts with CMAS, and the vertically arranged apatite further facilitates rapid CMAS penetration, allowing the molten CMAS to infiltrate along the grain boundaries of Lu₂SiO₅ (Fig. 31 (c)), exacerbating its CMAS corrosion resistance.

X2-RE₂SiO₅ demonstrates superior CMAS corrosion resistance compared to X1- RE₂SiO₅. Larger RE elements in RE₂SiO₅ facilitate the formation of Ca₂RE₈(SiO₄)₆O₂ apatite in corrosion products. At relatively low temperatures, the CMAS corrosion resistance of X2-RE₂SiO₅ shows a negative linear correlation with the RE ionic radius. However, at higher temperatures, X2-RE₂SiO₅ ceramics undergo more severe corrosion, and their CMAS resistance becomes less sensitive to the RE element species, instead being influenced by the distribution and structure of corrosion products.

4.2. CMAS corrosion behavior of rare earth disilicate (RE₂Si₂O₇)

RE₂Si₂O₇ has seven different polymorphs, namely β, α, γ, δ, F, G, and A phases, depending on the RE ionic radius and temperature (as shown in Fig. 32) [99]. Thus, the RE element species and temperature play a critical role in determining the CMAS corrosion behavior of RE₂Si₂O₇.

Kim et al. [110,111] studied the CMAS corrosion behavior of δ-Gd₂Si₂O₇ and β-Er₂Si₂O₇. The reaction product was determined to be Ca₂RE₈(SiO₄)₆O₂. Under the same conditions, the thickness of the reaction layer in β-Er₂Si₂O₇ was slightly lower than that in δ-Gd₂Si₂O₇, exhibiting better corrosion resistance. This improvement is attributed to the smaller ionic radius of Er, which results in a greater difference with Ca, thus hindering the formation of apatite. Wolf et al. [112] compared the CMAS corrosion behavior of Yb₂Si₂O₇ and Y₂Si₂O₇ ceramics. After 8 h at 1400 ℃, the reaction layer thicknesses of Yb₂Si₂O₇ and Y₂Si₂O₇ were 100 ± 6 μm and 255 ± 9 μm, respectively, with Yb₂Si₂O₇ showing significantly better CMAS corrosion resistance. This is because Yb₂Si₂O₇ has a smaller OB difference with CMAS, resulting in lower reactivity, and it requires more Ca to form stable apatite during the reaction process with CMAS.

Tian et al. [113] investigated the CMAS corrosion behavior and mechanism of RE₂Si₂O₇ (RE = Y, Yb, and Lu) ceramics at 1300 ℃ and 1500 ℃. After 50 h at 1300 ℃, an apatite-rich reaction layer developed within γ-Y₂Si₂O₇, as depicted in Fig. 33 (a). This reaction layer, positioned beneath the residual CMAS melts, exhibits a compact microstructure with minimal porosity, reaching an approximate thickness of 155 μm, effectively impeding further CMAS infiltration. Corrosion of β-Yb₂Si₂O₇ similarly resulted in the formation of an apatite reaction layer, although it was localized on the sample surface, with coarser apatite grains. The CMAS accumulated beneath this layer, characterized by the presence of numerous cracks and voids, and having a reaction zone thickness of approximately 125 μm (Fig. 33 (b)). For β-Lu₂Si₂O₇, no corrosion products were detected after CMAS corrosion, yet the CMAS extensively infiltrated along the grain boundaries, with near-complete CMAS consumption and abundant voids within the reaction zone, which exhibited a thickness of about 330 μm (Fig. 33 (c)). The comparative CMAS corrosion resistance ranks as follows: γ-Y₂Si₂O₇>β-Yb₂Si₂O₇>β-Lu₂Si₂O₇. At an elevated temperature of 1500 ℃, the CMAS is fully consumed, penetrating predominantly along the grain boundaries and permeating the entire sample, leading to “blister” cracks throughout the structure (Fig. 34). The temperature-dependent internal friction of γ-Y₂Si₂O₇, β-Yb₂Si₂O₇, and β-Lu₂Si₂O₇, shown in Fig. 34 (d), indicates low internal friction and a brittle nature below 1300 ℃. Under these conditions, the grain boundaries and interiors exhibit comparable strength, allowing for the gradual dissolution of RE₂Si₂O₇ into CMAS with subsequent apatite precipitation. Above 1300 ℃, however, internal friction increases significantly, signaling an elastic-plastic transition as grain boundary strength diminishes relative to grain interiors. Consequently, at 1500 ℃, CMAS preferentially infiltrates along grain boundaries, rapidly permeating the entire sample.

Padture et al. [114,115] investigated the CMAS corrosion behavior of γ-Y₂Si₂O₇, β-Yb₂Si₂O₇, and β-Sc₂Si₂O₇ at 1500 ℃, with cross-sectional morphologies illustrated in Fig. 35 (a-c). For γ-Y₂Si₂O₇, a substantial Ca₂Y₈(SiO₄)₆O₂ apatite reaction layer was formed, reaching a thickness of approximately 300 μm after 24 h. Conversely, β-Yb₂Si₂O₇ generated only a limited amount of apatite Ca₂Yb₈(SiO₄)₆O₂, while β-Sc₂Si₂O₇ showed no reaction products and instead underwent dissolution-reprecipitation. This contrast arises is attributed to the larger ionic radius of Y, which enhances its affinity for Ca and provides a strong driving force for Ca₂Y₈(SiO₄)₆O₂ formation. However, at 1500 ℃, CMAS infiltrated along the grain boundaries of both β-Yb₂Si₂O₇ and β-Sc₂Si₂O₇, leading to significant cracking after CMAS corrosion. This grain boundary infiltration is driven by the interfacial energy dynamics: the ceramic grain boundary energy is more than twice that of the ceramic/melt interface energy, so transforming a grain boundary into two ceramic/melt interfaces reduces the total energy of the system [116]. Additionally, CMAS infiltration along grain boundaries induces differential expansion, with more penetrated regions above expanding to a greater extent than the less-penetrated regions below. This differential expansion generates thermal stresses, leading to “blister” cracks, as shown in Fig. 35 (d). Further studies by Padture et al. [117,118] examined the microstructural and interfacial changes in β-Yb₂Si₂O₇ before and after CMAS corrosion. They observed persistent anisotropy in the alignment of grain boundaries, facet planes, and disconnections, consistently oriented with the {110} planes of β-Yb₂Si₂O₇ both pre- and post-corrosion. This anisotropy enhances the high-temperature stability of β-Yb₂Si₂O₇ and actively impedes CMAS infiltration.

5. Strategies to mitigate CMAS corrosion

Despite the widespread application of TBCs and EBCs to enhance the stability and durability of hot-section components in gas turbine engines, protection against CMAS corrosion remains critical. Research into CMAS corrosion protection methods is vital for improving gas turbine engine performance and enhancing aircraft reliability and safety. While EBCs and TBCs offer protection, the development of new materials and structures resistant to CMAS corrosion is essential. The design of EBCs faces challenges that require multi-layer, multi-component, or multi-scale coating designs, along with advanced coating preparation technologies. Proposed protection strategies include the design of the surface protection layer, optimization of the coating structure, modification of coating components, grain boundary engineering, and the development of new corrosion-resistant materials, as shown in Fig. 37.

5.1. Design the non-permeable protective coating

A dense, crack-free, unreactive, non-permeable protective coating can effectively block CMAS penetration. These protective layers can include oxide, non-oxide, or metal coatings, such as Pt, SiC, Al₂O₃, Si₃N₄ and their mixtures [125,126]. These materials can effectively slow down the penetration and chemical reaction of CMAS, as well as reduce CMAS adhesion to the coating surface. Yang et al. [127] investigated the CMAS corrosion resistance of Al₂O₃ and YSZ-Al₂O₃ coatings on 8YSZ (yttria-stabilized zirconia) by APS. They found that dense YSZ-Al₂O₃ coatings exhibited superior anti-penetration properties initially. However, over time, interactions with CMAS compromised the structure and phase stability of YSZ-Al₂O₃, reducing its thickness. In contrast, Al₂O₃ coatings demonstrated better long-term stability and protection (Fig. 38). In addition, Guo et al. [128] studied the effectiveness of Pt thin films and Gd₂Zr₂O₇ layers in preventing CMAS corrosion. They found that Pt thin films electroplated onto YSZ coatings provided effective resistance to CMAS, while EB-PVD Gd₂Zr₂O₇ coatings were ineffective due to large inter-columnar gaps.

Fig. 38 Diagram of CMAS interaction mechanism with YSZ, Al₂O₃ modified YSZ, and YSZ-Al₂O₃ modified YSZ. Reproduced with permission from Ref. [127], © Elsevier 2024.

Bakkar et al. [129] applied a Si₃N₄ impermeable layer using plasma-enhanced chemical vapor deposition. The Si₃N₄ seals the YSZ surface to mitigate CMAS attack. The interaction between CMAS and Si₃N₄ causes no cracks or delamination. The defect-free dense layer of Si₃N₄ presents excellent capability of protecting the YSZ by blocking CMAS infiltration.

5.2. Design the sacrificial protective coating

In contrast to the design of a non-permeable protective layer, a sacrificial protective layer can be employed to consume the CMAS and mitigate the corrosion. This layer actively reacts with CMAS to form a denser, impermeable substance, thereby protecting the underlying coating from erosion. Acting as a “sacrificial” component, it consumes harmful substances, ensuring the integrity and durability of the coating. Wang et al. [130] investigated a double ceramic layer plasma-sprayed YSZ coating that incorporates a sacrificial mullite-YSZ protective layer to suppress CMAS attacks. The double ceramic layer coating exhibits enhanced CMAS resistance due to the formation of the CaAl₂Si₂O₈ layer. No CMAS penetration was observed in the double ceramic layer coating even after exposure for up to 12 hours. Cao et al. [131] studied the deposition of an aluminum film on YSZ TBCs by magnetron sputtering technology, and the formation of a sacrificial alumina protective layer through in-situ oxidation to improve CMAS corrosion resistance. The research shows that the alumina layer, as a sacrificial protective layer, can effectively reduce the penetration of corrosive media on YSZ coating and extend the service life of the YSZ coating (Fig. 39).

5.3. Optimizing coating surface structure

In addition to blocking the CMAS corrosion through a chemical reaction, another approach is to optimize the coating structure to mitigate the interaction of CMAS with TBCs or EBCs. The coating surface can be treated with laser processing to mimic the structure of a lotus leaf, creating a superhydrophobic surface that prevents molten CMAS from wetting the surface and delays penetration. Fan et al. [132] used femtosecond laser texturing and aluminum film deposition to develop modified TBCs with excellent superhydrophobic properties at 1230 ℃. Compared to untreated TBCs, the contact angle between TBCs and molten CMAS was significantly increased, effectively inhibiting the wetting and diffusion of CMAS (Fig. 40).

Except for femtosecond laser technology, two other research groups introduced innovative techniques for surface modification. Similarly inspired by the natural superhydrophobic properties of the lotus leaf, Song et al. [133] employed ultrafast laser direct writing technology to create a biomimetic-structured surface on (Gd_0.9Yb_0.1)₂Zr₂O₇ TBCs, which demonstrated "silicate" phobicity at high temperatures. The molten-volcanic-ash-repelling (CMAS-phobic) characteristics of the surface were attributed to the lotus-leaf-like dual-scale microstructure, particularly the presence of nanoparticles. This surface modification approach was considered a significant step toward developing next-generation aviation engines with reduced vulnerability to environmental siliceous debris (Fig. 41) [133].

Some researchers combined laser technology with finite element analysis methods. They utilized finite element analysis to investigate the temperature field and stress distribution within the laser glazing process and subsequent exposure to CMAS within the TBCs. The results showed that a tensile stress state occurs along the laser scanning direction, and the maximum principal stress occurs at the interface between the glaze layer and the unmodified coating, which may be the key area for crack initiation and failure of the coating. In addition, the study found that when attacked by CMAS, modified TBCs exhibit high radial stress and maximum principal stress (about 1094 MPa) in the radial range of 0-6 mm, making these areas prone to cracking. The method not only improves the mechanical integrity of the TBC, but also provides insights into stress distribution that can guide future design improvements to enhance CMAS resistance. In addition to laser technology, spraying technology was applied to modify the surface morphologies [134]. Liu et al. [135] designed a unique surface on Yb-doped Y₂O₃ stabilized ZrO₂ (YbYSZ) coatings through solution precursor plasma spraying, which leverages micro-nano structures to provide superior anti-wetting properties. These properties are crucial for preventing CMAS penetration at high temperatures. Their study investigated the structural and compositional changes in YbYSZ-SPPS coatings after prolonged thermal exposure at 1300 ℃, demonstrating the potential of the coatings for long-term use in high-temperature environments. These properties are crucial for preventing CMAS penetration at high temperatures.

5.4. Enhancing CMAS corrosion resistance via grain boundary engineering

In gas turbine engines, CMAS molten salt can infiltrate along grain boundary of TBCs or EBCs, leading to significant material performance degradation. Grain boundary engineering, a field focuses on the manipulation of grain boundaries to improve material properties, has shown potential in mitigating the effects of CMAS corrosion [136]. Grain boundary phase can react with CMAS and block its infiltration pathways. For instance, Tian et al. [137] coated Yb₂SiO₅ on the surface of YTaO₄ powder using the sol-gel method. During high-temperature exposure, Yb₂SiO₅ forms at the grain boundary and act as corrosion inhibitor, effectively preventing CMAS from corroding the grain boundaries of YTaO₄.

In addition to specially designed corrosion inhibitor, some materials can actively react with CMAS as sacrificial phases to inhibit the CMAS corrosion. Wang et al. pre-mixed mullite and Al₂O₃ with YSZ. The introduction of mullite and Al₂O₃ altered the grain boundary characteristics of ceramic pellets. Due to its high chemical reactivity, mullite quickly reacts with CMAS to form a sealing layer, reducing the penetration of molten CMAS (Fig. 42).

Beyond the use of corrosion inhibitors and sacrificial agents, mixing the CMAS glass phase in the ceramic powder can effectively prevent the rapid penetration of CMAS glass. N. P. Padture et al. [114] observed that two EBC ceramic materials, β-Yb₂Si₂O₇ and β-Sc₂Si₂O₇ suffered from “blister” crack damage caused by the penetration of CMAS glass into their grain boundaries. To address this issue, they introduced 1 volume % of CMAS glass into β-Yb₂Si₂O₇ powder before sintering. By incorporating the CMAS glass phase at the grain boundary, the rapid penetration of CMAS was effectively prevented avoiding the expansion gradient and reducing the interface energy. (Fig. 43).

In summary, grain boundary engineering provides a promising strategy for the development materials with improved resistance to CMAS corrosion. By understanding and controlling the grain boundary phase, it is possible to design TBCs and EBCs that can withstand the attack of CMAS, thereby extending the service life and enhancing the performance of high-temperature components in gas turbine engines.

5.5. Development of high entropy TBCs and EBCs

The composition of TBC and EBC materials significantly impacts their resistance to CMAS. Most TBCs and EBCs contain rare earth elements, and the large family of rare earth elements offers ample opportunity for composition modification through the solid solution method. Previous studies have shown that the presence of rare earth elements is strongly correlated with resistance to CMAS. As a result, high-entropy TBCs and EBCs are particularly valuable for CMAS corrosion protection due to their flexible and customizable composition design. High-entropy materials, which form single-phase solid solutions with a variety of elements, exhibit enhanced thermal, mechanical, and corrosion-resistant properties compared to traditional single-phase materials, overcoming the limitations of conventional coatings.

Wang et al. investigated the excellent resistance to CMAS of multi-component rare earth disilicates, specifically (Er_0.25Tm_0.25Yb_0.25Lu_0.25)₂Si₂O₇, as an EBC material at 1500 ℃. From the perspective of high-entropy materials, the focus is on the synergistic effect of multiple rare earth elements. Y and Lu as inert elements maintain phase stability, while Er and Tm as active elements promote the formation of precipitated phase formation. This rapid phase formation reduces CMAS corrosion activity and significantly enhances the material’s corrosion resistance to CMAS (Fig. 44).

By combining the stability of apatite with the synergistic effect of high entropy, Chu et al. [140] designed (Ho_1/4Er_1/4Dy_1/4Gd_1/4)₂Si₂O₇ as a promising T/EBC material. This material exhibits superior CMAS corrosion resistance and thermal properties, including a low corrosion rate of 8.2 ± 0.14 µm/h at 1673 K. The exceptional performance is attributed to the synergistic effects of high chemical reactivity and the stability of the formed apatite in the CMAS melt. Liu et al. [141] focused on a novel high-entropy ceramic material with low thermal conductivity, (Nd_0.2Gd_0.2Y_0.2Er_0.2Yb_0.2)₂Zr₂O₇, as a potential material for TBCs. The results showed that the high-entropy ceramic formed a dense reaction layer composed of apatite and RE/Ca-ZrO₂, significantly reducing the infiltration rate of CMAS at high temperatures. The study concluded that the high-entropy ceramic demonstrates exceptional resistance to CMAS corrosion, making it a promising candidate for TBCs (Fig. 45).

Liu et al. [142] investigated the hot corrosion behavior of (La_0.2Nd_0.2Sm_0.2Eu_0.2Gd_0.2)₂Zr₂O₇ (HEZ) ceramics when exposed to molten CMAS at 1300 ℃. The high-entropy zirconate ceramics demonstrated a graceful behavior during CMAS corrosion, with no horizontal or vertical cracks forming in either the ceramic substrate or the reaction layer. This was attributed to the lower Young's modulus and higher thermal expansion coefficient of (La_0.2Nd_0.2Sm_0.2Eu_0.2Gd_0.2)₂Zr₂O₇ ceramics, which resulted in a fine-grained reaction layer without the formation of MgAl₂O₄ phase. The study concluded that high-entropy rare-earth zirconate ceramics show superior resistance to CMAS corrosion at high temperatures compared to lanthanum zirconate ceramics (Fig. 46 (a) and (b)). Luo et al. [143] investigated the CMAS corrosion behavior of two high-entropy ceramics, (La_0.2Nd_0.2Sm_0.2Eu_0.2Gd_0.2)₂Zr₂O₇ (LaHZ) and (Nd_0.2Sm_0.2Eu_0.2Gd_0.2Yb_0.2)₂Zr₂O₇ (YbHZ). YbHZ demonstrated significantly reduced infiltration depths compared to La₂Zr₂O₇. After CMAS corrosion, both LaHZ and La₂Zr₂O₇ exhibited similar microstructures, with apatite and fluorite embedded in CMAS residue, while YbHZ formed a dense nanoscale dual-phase zirconia layer without visible CMAS residue. The study concluded that the formation of a dense and continuous fluorite Ca&RE-ZrO₂ layer could provide excellent CMAS resistance, offering a promising strategy for designing CMAS-resistant TBCs (Fig. 46 (c)).

6. Concluding remarks and future directions

The increase in inlet temperatures, driven by the high thrust-to-weight ratio of modern gas turbine engines, has significantly accelerated the corrosion and infiltration of CMAS, presenting a major challenge to the safe operation of gas turbine engines. This review summarizes the behavior and mechanisms of CMAS-induced corrosion in TBCs and EBCs. CMAS corrosion can result in reduced strain tolerance, increased elastic modulus, sintering, thermal expansion mismatch, crack formation, and “blister” cracks in protective coatings. Additionally, CMAS can penetrate and corrode along grain boundaries, pores, and interfaces, causing damage to the turbine blade.

Current research has highlighted the variation in corrosion behavior across different coating materials. For example, the Y element in YSZ ceramics tends to segregate at grain boundaries, promoting grain boundary corrosion. Similarly, rare earth tantalate ceramics exhibit grain boundary corrosion, although without the segregation of rare earth elements. CMAS corrosion resistance is strongly correlated with RE elements in rare earth zirconates. Among these, rare earth monosilicate ceramics show better resistance to CMAS corrosion compared to rare earth disilicates. Furthermore, studies have shown that factors such as the presence of rare earth elements and the corrosion temperature significantly influence the corrosion resistance of TBCs and EBCs. At 1300 ℃, the CMAS corrosion resistance of X2-type rare earth monosilicate ceramics improves as the rare earth ion radius decreases, though this trend diminishes at higher temperatures (1500 ℃). Rare earth disilicate ceramics exhibit milder CMAS corrosion at 1300 ℃ but develop “blister” cracks at temperatures above 1500 ℃.

To address the challenges posed by CMAS corrosion, researchers have proposed several mitigation strategies. Some approaches focus on optimizing the coating structure and composition, such as employing non-permeable layers to delay the penetration and reaction of CMAS or designing high-entropy ceramics that incorporate components with excellent CMAS resistance while maintaining desirable thermal and mechanical properties. Another approach emphasizes the structural design of the protective coatings. Given the strong flow and corrosiveness of CMAS at high temperatures, biomimetic designs that create surfaces to hinder CMAS spread can enhance corrosion resistance.

Based on the above summary, several future directions can be outlined:

(1) Current stage of research on CMAS corrosion in extreme environments typically employs high-temperature CMAS corrosion and characterization at room temperature. However, the high-temperature corrosion process is often considered a “black box”, making it difficult to observe the events occurring during this period. As a result, characterizations at room temperature need to deduce the reaction mechanism. The nucleation, growth, precipitation, movement, and solidification of CMAS corrosion products are crucial for the understanding of CMAS corrosion mechanisms and the degradation of TBCs and EBCs. Despite their importance, relevant in-situ studies remain limited. Therefore, conducting in-situ characterization of CMAS corrosion will provide deeper insights into the corrosion behavior and mechanisms, which will significantly assist in the design of effective anti-corrosion protective coatings.

(2) Most current research on CMAS corrosion in TBCs and EBCs is conducted under corrosion only by CMAS melts. However, in actual service conditions, the presence of water vapor in the combustion gases also poses a significant corrosion threat. Water vapor exhibits high corrosive capability and oxygen diffusivity at elevated temperatures, which can adversely affect both the coatings and the underlying substrates. Under the influence of high-temperature water vapor, the wetting behavior of CMAS may change, potentially enhancing its reactivity. Thus, investigating the synergistic corrosion effects of water vapor and CMAS will be crucial for accurately assessing the performance of TBCs and EBCs in the real-world combustion environment of gas turbine engines.

(3) Most TBCs and EBCs contain rare earth elements, and the extensive family of rare earth elements offers considerable design flexibility for high-entropy ceramics. However, there is currently a lack of theoretical guidance for the compositional design of high-entropy TBCs and EBCs with strong resistance to CMAS corrosion. Moreover, the synergistic effects of multiple components and the selective corrosion of rare earth elements remain poorly understood. Therefore, it is essential to investigate the role of rare earth elements in enhancing CMAS corrosion resistance and to establish design principles for constructing high-entropy TBCs and EBCs.

References

[1] Z.Y. Wei, G.H. Meng, L. Chen, G.R. Li, M.J. Liu, W.X. Zhang, L.N. Zhao, Q. Zhang, X.D. Zhang, C.L. Wan, Z.X. Qu, L. Chen, J. Feng, L. Liu, H. Dong, Z.B. Bao, X.F. Zhao, X.F. Zhang, L. Guo, L. Wang, B. Cheng, W.W. Zhang, P.Y. Xu, G.J. Yang, H.N. Cai, H. Cui, Y. Wang, F.X. Ye, Z. Ma, W. Pan, M. Liu, K.S. Zhou, C.J. Li, J. Adv. Ceram. 11 ( 2022) 985-106. http://doi.org/10.1007/s40145-022-0581-7.

[2] D. Tejero-Martin, C. Bennett, T. Hussain, J. Eur. Ceram. Soc. 41 ( 2021) 1747-1768. http://doi.org/10.1016/j.jeurceramsoc.2020.10.057.

[3] L. Liu, S.Z. Wang, B.Q. Zhang, G.Y. Jiang, H. Liu, J.Q. Yang, J.L. Wang, W. Liu, J. Manuf. Process. 97 ( 2023) 12-34. http://doi.org/10.1016/j.jmapro.2023.04.052.

[4] R. Darolia, Int. Mater. Rev. 58 ( 2013) 315-348. http://doi.org/10.1179/1743280413Y.0000000019.

[5] Q.H. Zeng, X.W. Chen, Prog. Aerosp. Sci. 140 ( 2023) 100927. http://doi.org/10.1016/j.paerosci.2023.100927.

[6] N.P. Padture, Nat. mater. 15 ( 2016) 804-809. http://doi.org/10.1038/nmat4687.

[7] B. Liu, Y.C. Liu, C.H. Zhu, H.M. Xiang, H.F. Chen, L.C. Sun, Y.F. Gao, Y.C. Zhou, J. Mater. Sci. Technol. 35 ( 2019) 833-851. http://doi.org/10.1016/j.jmst.2018.11.016

[8] S. Zhang, X.H. Wang, C. Zhang, H.M. Xiang, Y.W. Li, C. Fang, M.L. Li, H.L. Wang, Y.C. Zhou, J. Adv. Ceram. 13 ( 2024) 373-387. http://doi.org/10.26599/JAC.2024.9220862.

[9] D.R. Clarke, S.R. Phillpot, Mater. today 8 ( 2005) 22-29. https://doi.org/10.1016/S1369-7021(05)70934-2.

[10] N.P. Padture, M. Gell, E.H. Jordan, Science 296 ( 2002) 280-284. http://doi.org/doi:10.1126/science.1068609.

[11] J. Wang, Q.Q. Jin, J.B. Song, D. Zhang, B. Xu, Z.Y. Ren, M. Wang, S.X. Yan, X.L. Sun, C. Liu, X.Y. Chong, J. Feng, J. Adv. Ceram. 12 ( 2023) 2087-2100. http://doi.org/10.26599/JAC.2023.9220811.

[12] C.K. Song, F. Ye, L.F. Cheng, Y.S. Liu, Q. Zhang, J. Adv. Ceram. 11 ( 2022) 1343-1374. http://doi.org/10.1007/s40145-022-0611-5.

[13] R.X. Liu, W.P. Liang, Q. Miao, H. Zhao, X.F. Zhang, M. Zhang, R.X. Yan, B. Ramasubramanian, S. Ramakrishna, J. Adv. Ceram. 13 ( 2024) 1677-1696. http://doi.org/10.26599/JAC.2024.9220969.

[14] K.N. Lee, D.S. Fox, N.P. Bansal, J. Eur. Ceram. Soc. 25 ( 2005) 1705-1715. https://doi.org/10.1016/j.jeurceramsoc.2004.12.013.

[15] X. Shan, Z.H. Zou, L.J. Gu, L.X. Yang, F.W. Guo, X.F. Zhao, P. Xiao, Scr. Mater. 113 ( 2016) 71-74. https://doi.org/10.1016/j.scriptamat.2015.09.029.

[16] R.B. Nair, D. Brabazon, npj Mater. Degrad. 8 ( 2024) 44. http://doi.org/10.1038/s41529-024-00462-w.

[17] Y.F. Wang, Q.G. Fu, H. Dong, X.S. Zhuo, X.H. Liang, G. Liu, L.X. Huang, X.F. Zhang, Mater. Today Commun. 41 ( 2024) 110756. http://doi.org/10.1016/j.mtcomm.2024.110756.

[18] A. Nieto, R. Agrawal, L. Bravo, C. Hofmeister-Mock, M. Pepi, A. Ghoshal, Int. Mater. Rev. 66 ( 2021) 451-492. http://doi.org/10.1080/09506608.2020.1824414.

[19] J.B. Roche, Unveiling the Greatest Threats: Top Hazards of Sand and Volcanic Ash for Aircraft.

[20] CBC, Volcanic ash: Hazardous to planes, environment.

[21] G. Costa, B.J. Harder, V.L. Wiesner, D.M. Zhu, N. Bansal, K.N. Lee, N.S. Jacobson, D. Kapush, S.V. Ushakov, A. Navrotsky, J. Am. Ceram. Soc. 102 ( 2019) 2948-2964. http://doi.org/10.1111/jace.16113.

[22] D.L. Poerschke, R.W. Jackson, C.G. Levi, Annu. Rev. Mater. Res. 47 ( 2017) 297-330. http://doi.org/10.1146/annurev-matsci-010917-105000.

[23] F.C. Toriz, A.B. Thakker, S.K. Gupta, Turbo Expo: Power Land, Sea, Air 79221 ( 1988) V005T13A012. https://doi.org/10.1115/88-GT-279.

[24] J.L. Smialek, The chemistry of Saudi Arabian sand: A deposition problem on helicopter turbine airfoils. New London, United States, 1991.

[25] J. Kim, M.G. Dunn, A.J. Baran, D.P. Wade, E.L. Tremba, J. Eng. Gas Turbines Power 115 ( 1993) V003T05A001. https://doi.org/10.1115/1.2906754.

[26] M.P. Borom, C.A. Johnson, L.A. Peluso, Surf. Coat. Technol. 86 ( 1996) 116-126. http://doi.org/10.1016/S0257-8972(96)02994-5.

[27] R. Gertisser, Geol. Today 26 ( 2010) 94-95. http://doi.org/10.1111/j.1365-2451.2010.00757.x.

[28] S. Krämer, J. Yang, C.G. Levi, C.A. Johnson, J. Am. Ceram. Soc. 89 ( 2006) 3167-3175. http://doi.org/10.1111/j.1551-2916.2006.01209.x.

[29] W. Braue, J. Mater. Sci. 44 ( 2009) 1664-1675. http://doi.org/10.1007/s10853-008-3215-8.

[30] C.G. Levi, J.W. Hutchinson, M.H. Vidal-Sétif, C.A. Johnson, MRS Bull. 37 ( 2012) 932-941. http://doi.org/10.1557/mrs.2012.230.

[31] N.P. Bansal, S.R. Choi, Ceram. Int. 41 ( 2015) 3901-3909. http://doi.org/10.1016/j.ceramint.2014.11.072.

[32] A. Aygun, A.L. Vasiliev, N.P. Padture, X.Q. Ma, Acta Mater. 55 ( 2007) 6734-6745. http://doi.org/10.1016/j.actamat.2007.08.028.

[33] A.R. Ericks, F.W. Zok, D.L. Poerschke, C.G. Levi, J. Am. Ceram. Soc. 105 ( 2022) 3665-3688. http://doi.org/10.1111/jace.18413.

[34] R. Naraparaju, J.J.G. Chavez, U. Schulz, C.V. Ramana, Acta Mater. 136 ( 2017) 164-180. http://doi.org/10.1016/j.actamat.2017.06.055.

[35] L. Guo, H. Xin, Y.Y. Li, Y. Yu, Z. Yan, C.W. Hu, F.X. Ye, J. Eur. Ceram. Soc. 40 ( 2020) 5683-5691. http://doi.org/10.1016/j.jeurceramsoc.2020.07.025.

[36] D.L. Poerschke, T.L. Barth, C.G. Levi, Acta Mater. 120 ( 2016) 302-314. http://doi.org/10.1016/j.actamat.2016.08.077.

[37] W.J. Song, Y. Lavallee, K.U. Hess, U. Kueppers, C. Cimarelli, D.B. Dingwell, Nat. Commun. 7 ( 2016) 10795. http://doi.org/10.1038/ncomms107958.

[38] R.I. Webster, E.J. Opila, J. Non-Cryst. Solids 584 ( 2022) 121508. http://doi.org/10.1016/j.jnoncrysol.2022.121508.

[39] B. Zhang, Y. Yu, L. Guo, S. Kang, Y.X. Cheng, Z. Yan, L. Wang, Ceram. Int. 48 ( 2022) 32877-32885. http://doi.org/10.1016/j.ceramint.2022.07.215.

[40] R.A. Miller, J. Therm. Spary Technol. 6 ( 1997) 35-42. http://doi.org/10.1007/BF02646310.

[41] H. Davies, Aeronaut. J. 67 ( 1963) 79-91. http://doi.org/10.1017/S0368393100090222.

[42] X.Q. Cao, R. Vassen, D. Stoever, J. Eur. Ceram. Soc. 24 ( 2004) 1-10. https://doi.org/10.1016/S0955-2219(03)00129-8.

[43] P. Li, X.C Jin, P. Lu, D.L Liu, R. Mou, X.L Fan, J. Adv. Ceram. 12 ( 2023) 1805-1820. http://doi.org/10.26599/JAC.2023.9220789.

[44] L. Guo, Y.Y Li, G. Li, J. Adv. Ceram. 12 ( 2023) 1712-1730. http://doi.org/10.26599/JAC.2023.9220781.

[45] D. Gao, Y.D Liu, A.H Huang, Adv. Ceram. 45(5) ( 2024) 434-445. http://doi.org/10.16253/j.cnki.37-1226/tq.2024.05.004.

[46] G. Mauer, M.O. Jarligo, D.E. Mack, R. Vaßen, J. Therm. Spray Technol. 22 ( 2013) 646-658. http://doi.org/10.1007/s11666-013-9889-8.

[47] R. Vaßen, E. Bakan, D.E. Mack, O. Guillon, J. Therm. Spray Technol. 31 ( 2022) 685-698. http://doi.org/10.1007/s11666-022-01330-2.

[48] J. K. Wang, L. Chen, B.H Li, H. Xu, X. Jiang, J.F. Li, J. Feng, J. Adv. Ceram. 13 ( 2024) 1771-1785. http://doi.org/10.26599/JAC.2024.9220973.

[49] G.R. Li, T. Liu, X.T. Luo, G.J. Yang, C.J. Li, J. Adv. Ceram. 12 ( 2023) 1317-1330. http://doi.org/10.26599/JAC.2023.9220750.

[50] J. Chevalier, L. Gremillard, A.V. Virkar, D.R. Clarke, J. Am. Ceram. Soc. 92 ( 2009) 1901-1920. http://doi.org/10.1111/j.1551-2916.2009.03278.x.

[51] V.P. Gorelov, S.A. Belyakov, R.K. Abdurakhimova, Phys. Solid State 65 ( 2023). http://doi.org/10.21883/PSS.2023.03.55589.541.

[52] W.W QU, Z.H CHEN, Y.L PEI, S.S LI, F.H WANG, J. Chin. Soc. Corros. Prot. 43 ( 2023) 1407-1412. http://doi.org/10.11902/1005.4537.2022.362.

[53] X.M Zhang, H. Xin, L. Guo, Corros. Commun. 10 ( 2023) 1-9. http://doi.org/https://doi.org/10.1016/j.corcom.2022.08.003.

[54] K. Wang, L.X Zhou, L. Guo, K. Yan, F.X Ye, H.L Liu, H.B Guo, J. Chin. Soc. Corros. Prot. ( 2024) 1-27. http://doi.org/10.11902/1005.4537.2024.227.

[55] S.C. Tseng, C.K. Chao, D.X. Li, X.L Fan, Coatings 11 ( 2021) 1148. http://doi.org/10.3390/coatings11101148

[56] H.J. Fang, W.W. Wang, C.C. Zhang, Y.X. Wang, T. Yang, Z.N. Yang, Y.G. Liu, D. D.Ye, Ceram. Int. 49 ( 2023) 12390-12407. https://doi.org/10.1016/j.ceramint.2022.12.099.

[57] D.E. Mack, T. Wobst, M.O.D. Jarligo, D. Sebold, R. Vaßen, Surf. Coat. Technol. 324 ( 2017) 36-47. https://doi.org/10.11902/1005.4537.2024.227.

[58] J. Wu, H.B. Guo, Y.Z. Gao, S.K. Gong, J. Eur. Ceram. Soc. 31 ( 2011) 1881-1888. https://doi.org/10.1016/j.jeurceramsoc.2011.04.006.

[59] Q. Ahmed, I.N. Qureshi, I.U. Salam, Key Engineering Materials 778 ( 2018) 245-250. http://doi.org/10.4028/www.scientific.net/KEM.778.245.

[60] D. Wu, X. Shan, H.Y. Cai, F. Yang, X.F. Zhao, Corros. Sci. 209 ( 2022) 110803. http://doi.org/10.1016/j.corsci.2022.110803.

[61] D.X. Li, P. Jiang, R.H. Gao, F. Sun, X.C. Jin, X.L. Fan, J. Adv. Ceram. 10 ( 2021) 551-564. http://doi.org/10.1007/s40145-021-0457-2.

[62] H.F. Garces, B.S. Senturk, N.P. Padture, Scr. Mater. 76 ( 2014) 29-32. https://doi.org/10.1016/j.scriptamat.2013.12.008.

[63] G.N. Xu, L. Yang, Y.C. Zhou, Z.P. Pi, W. Zhu, J. Mech. Phys. Solids 133 ( 2019) 103710. https://doi.org/10.1016/j.jmps.2019.103710.

[64] G. Pujol, F. Ansart, J.P. Bonino, A. Malié, S. Hamadi, Surf. Coat. Technol. 237 ( 2013) 71-78. https://doi.org/10.1016/j.surfcoat.2013.08.055.

[65] A. Feuerstein, J. Knapp, T. Taylor, A. Ashary, A. Bolcavage, N. Hitchman, J. Therm. Spray Technol. 17 ( 2008) 199-213. http://doi.org/10.1007/s11666-007-9148-y.

[66] X. Shan, W.F. Chen, L.X. Yang, F.W. Guo, X.F. Zhao, P. Xiao, Corros. Sci. 167 ( 2020) 108478. https://doi.org/10.1016/j.corsci.2020.108478.

[67] R. Wellman, G. Whitman, J.R. Nicholls, Int. J. Refract. Met. Hard Mater. 28 ( 2010) 124-132. https://doi.org/10.1016/j.ijrmhm.2009.07.005.

[68] W. Pan, X.R. Ren, M. Zhao, Thermal barrier coating ceramic YSZ material properties and high temperature aging, Science Press, Beijing, 2018.

[69] R.B. Zhu, J.P. Zou, J. Mao, X.F. Zhang, C.M. Deng, M. Liu, W.L. Chen, Res. Appl. Mater. Sci. 1 ( 2019) 27-32. http://doi.org/10.33142/msra.v1i1.670.

[70] S. Krämer, J. Yang, C.G. Levi, J. Am. Ceram. Soc. 91 ( 2008) 576-583. https://doi.org/10.1111/j.1551-2916.2007.02175.x.

[71] U. Schulz, W. Braue, Surf. Coat. Technol. 235 ( 2013) 165-173. http://doi.org/10.1016/j.surfcoat.2013.07.029.

[72] Y. Wu, W.B. Zhi, Y. Li, X.Y. Guo, Z. Zhou, W. Shao, D.Y. He, Corros. Sci. 224 ( 2023) 111526. http://doi.org/10.1016/j.corsci.2023.111526.

[73] L. Guo, Y.Y. Li, K. Yan, J. Adv. Ceram. 13 ( 2024) 447-462. http://doi.org/10.26599/JAC.2024.9220867.

[74] Q. Flamant, M. Gurak, D.R. Clarke, J. Eur. Ceram. Soc. 38 ( 2018) 3925-3931. http://doi.org/10.1016/j.jeurceramsoc.2018.04.002.

[75] L. Chen, M.Y. Hu, P. Wu, J. Feng, J. Am. Ceram. Soc. 102 ( 2019) 4809-4821. https://doi.org/10.1111/jace.16328.

[76] L. Chen, P. Wu, J. Feng, Int. J. Appl. Ceram. Technol. 16 ( 2019) 230-242. https://doi.org/10.1111/ijac.13079.

[77] L. Chen, Y.H Jiang, X.Y Chong, J. Feng, J. Am. Ceram. Soc. 101 ( 2018) 1266-1278. http://doi.org/https://doi.org/10.1111/jace.15268.

[78] L. Chen, M.Y Hu, J. Guo, X.Y Chong, J. Feng, J. Mater. Sci. Technol. ( 2020) 20-28, https://doi.org/10.1016/j.jmst.2020.02.051.

[79] L. C, J. F, J. Exp. Fluid Mech. 36 ( 2022) 56-76. http://doi.org/ 10.11729/syltlx20220020.

[80] W.Q Yang, F.X Ye, Ceram. Int. 47 ( 2021) 28685-28697. https://doi.org/10.1016/j.ceramint.2021.07.028.

[81] W.Q Yang, F.X Ye, H. Qi, Ceram. Int. 48 ( 2022) 22179-22186. https://doi.org/10.1016/j.ceramint.2022.04.219.

[82] Z.L Chen, Z.L Tian, L.Y Zheng, B. Li, J. Eur. Ceram. Soc. 44 ( 2024) 7277-7284. https://doi.org/10.1016/j.jeurceramsoc.2024.04.073.

[83] L. Guo, M.Z. Li, S.X. He, C.L. Zhang, Q. Wang, F.X. Ye, J. Alloys Compd. 698 ( 2017) 13-19. http://doi.org/10.1016/j.jallcom.2016.12.241.

[84] C.X. Song, X. Qi, L. Huang, J.B. Sun, L.Z. Li, C.G. Li, W.H. Lu, Ceram. Int. 49 ( 2023) 26578-26588. http://doi.org/10.1016/j.ceramint.2023.05.193.

[85] S.J. Meng, L. Guo, H.H. Guo, Y.P. Wang, H.L Liu, J. Adv. Ceram. 13 ( 2024) 1254-1267. http://doi.org/10.26599/JAC.2024.9220934.

[86] S.Y. Qin, H.T. Cao, G. Brewster, Z.H. Gao, X.L. Zhong, B. Han, Y. Chen, P. Xiao, Scr. Mater. 239 ( 2024) 115819. http://doi.org/10.1016/j.scriptamat.2023.115819.

[87] B.W. Li, J.Y. Sun, L. Guo, Corros. Sci. 193 ( 2021) 109899. https://doi.org/10.1016/j.corsci.2021.109899.

[88] W.Q. Yang, F.X. Ye, Surf. Coat. Technol. 423 ( 2021) 127584. https://doi.org/10.1016/j.surfcoat.2021.127584.

[89] S.S. Gong, M. Chen, X.P. Huang, F. Zhang, Trans. Nonferrous Met. Soc. China 32 ( 2022) 4014-4029. https://doi.org/10.1016/S1003-6326(22)66074-6.

[90] L. Guo, J.Y. Feng, S.J. Meng, Corros. Sci. 208 ( 2022) 110628. https://doi.org/10.1016/j.corsci.2022.110628.

[91] J.S. Han, Y. Zou, D.T. Wu, J. Chen, Y.G. Zhang, Surf. Coat. Technol. 446 ( 2022) 128799. https://doi.org/10.1016/j.surfcoat.2022.128799.

[92] K. Bryce, B.P. Majee, L.P Huang, J. Lian, J. Adv. Ceram. 13 ( 2024) 1807-1822. http://doi.org/10.26599/JAC.2024.9220978.

[93] L. Dong, W.Q. Yang, L. Chen, G.J. Yang, C.J. Li, J. Adv. Ceram. 13 ( 2024) 976-986. http://doi.org/10.26599/JAC.2024.9220911.

[94] G.H Zhang, J.Y Shi, J. Zhang, J.Y Wang, J. Adv. Ceram. 13 ( 2024) 789-799. http://doi.org/10.26599/JAC.2024.9220898.

[95] X. Wang, M.Y Meng, F.H Xu, L. Liu, L.H Gao, S.Z Zhu, Z. Ma, J. Adv. Ceram 13 ( 2024) 549-560. http://doi.org/10.26599/JAC.2024.9220877.

[96] M.Z Ren, D. Lin, G.J Y, J. Chin. Soc. Corros. Prot. ( 2024) 1-17. http://doi.org/10.11902/1005.4537.2024.283.

[97] K.N. Lee, R.A. Miller, Surf. Coat. Technol. 86 ( 1996) 142-148. http://doi.org/10.1016/S0257-8972(96)03074-5.

[98] Y. Xu, X.X. Hu, F.F. Xu, K.W. Li, Ceram. Int. 43 ( 2017) 5847-5855. http://doi.org/10.1016/j.ceramint.2017.01.153.

[99] L.R. Turcer, N.P. Padture, Scr. Mater. 154 ( 2018) 111-117. http://doi.org/10.1016/j.scriptamat.2018.05.032.

[100] Z.L. Tian, J. Zhang, T.Y. Zhang, X.M. Ren, W.P. Hu, L.Y. Zheng, J.Y. Wang, J. Eur. Ceram. Soc. 39 ( 2019) 1463-1476. http://doi.org/10.1016/j.jeurceramsoc.2018.12.015.

[101] Z.L. Tian, L.Y. Zheng, J.M. Wang, P. Wan, J.L. Li, J.Y. Wang, J. Eur. Ceram. Soc. 36 ( 2016) 189-202. http://doi.org/10.1016/j.jeurceramsoc.2015.09.013.

[102] G. Cao, Y.H. Wang, Z.Y. Ding, H.L. Yang, Z.G. Liu, J.H. Ouyang, Y.M. Wang, Y.J. Wang, Corros. Sci. 202 ( 2022) 110319. http://doi.org/10.1016/j.corsci.2022.110319.

[103] G. Cao, S.Q. Wang, Y.H. Wang, Z.Y. Ding, Z.G. Liu, J.H. Ouyang, Y.M. Wang, Y.J. Wang, J. Eur. Ceram. Soc. 43 ( 2023) 600-611. http://doi.org/10.1016/j.jeurceramsoc.2022.10.003.

[104] X. Zhong, Y.W. Wang, P.P. Liu, L.P. Huang, Y.R. Niu, Q.L. Li, L. Zhang, X.B. Zheng, Corros. Sci. 199 ( 2022) 110174. http://doi.org/10.1016/j.corsci.2022.110174.

[105] Z.L. Tian, J. Zhang, L.Y. Zheng, W.P. Hu, X.M. Ren, Y.M. Lei, J.Y. Wang, Corros. Sci. 148 ( 2019) 281-292. http://doi.org/10.1016/j.corsci.2018.12.032.

[106] G. Costa, B.J. Harder, N.P. Bansal, B.A. Kowalski, J.L. Stokes, J. Am. Ceram. Soc. 103 ( 2019) 1446-1453. http://doi.org/10.1111/jace.16816.

[107] A.R. Krause, B.S. Senturk, H.F. Garces, G. Dwivedi, A.L. Ortiz, S. Sampath, N.P. Padture, J. Am. Ceram. Soc. 97 ( 2014) 3943-3949. http://doi.org/10.1111/jace.13210.

[108] Z.L. Tian, K.Y. Ming, L.Y. Zheng, Z.L. Chen, Y.P. Zhang, P. Liu, B. Li, Surf. Coat. Technol. 492 ( 2024) 131196. http://doi.org/10.1016/j.surfcoat.2024.131196.

[109] Z.L. Tian, K.Y. Ming, L.Y. Zheng, Z.L. Chen, F. Zhou, P. Liu, Z.H. Qiu, D.H. Wei, B. Li, J.Y. Wang, J. Adv. Ceram. 12 ( 2023) 2315-2330. http://doi.org/10.26599/JAC.2023.9220822.

[110] S.H. Kim, C.A.J. Fisher, N. Nagashima, Y. Matsushita, B.K. Jang, Ceram. Int. 48 ( 2022) 17369-17375. http://doi.org/10.1016/j.ceramint.2022.03.001.

[111] S.H. Kim, N. Nagashima, Y. Matsushita, B.K. Jang, J. Eur. Ceram. Soc. 43 ( 2023) 593-599. http://doi.org/10.1016/j.jeurceramsoc.2022.10.040.

[112] M. Wolf, D.E. Mack, O. Guillon, R. Vaßen, J. Am. Ceram. Soc. 103 ( 2020) 7056-7071. http://doi.org/10.1111/jace.17328.

[113] Z.L. Tian, X.M. Ren, Y.M. Lei, L.Y. Zheng, W.R. Geng, J. Zhang, J.Y. Wang, J. Eur. Ceram. Soc. 39 ( 2019) 4245-4254. http://doi.org/10.1016/j.jeurceramsoc.2019.05.036.

[114] L.R. Turcer, A.R. Krause, H.F. Garces, L. Zhang, N.P. Padture, J. Eur. Ceram. Soc. 38 ( 2018) 3914-3924. http://doi.org/10.1016/j.jeurceramsoc.2018.03.010.

[115] L.R. Turcer, A.R. Krause, H.F. Garces, L. Zhang, N.P. Padture, J. Eur. Ceram. Soc. 38 ( 2018) 3905-3913. http://doi.org/10.1016/j.jeurceramsoc.2018.03.021.

[116] T.M. Shaw, P.R. Duncombe, J. Am. Ceram. Soc. 74 ( 1991) 2495-2505. http://doi.org/10.1111/j.1151-2916.1991.tb06791.x.

[117] H. Sternlicht, D.W. McComb, N.P. Padture, Acta Mater. 241 ( 2022) 118360. http://doi.org/10.1016/j.actamat.2022.118360.

[118] H. Sternlicht, D.W. McComb, N.P. Padture, Acta Mater. 241 ( 2022) 118359. http://doi.org/10.1016/j.actamat.2022.118359.

[119] S.H. Kim, N. Nagashima, Y. Matsushita, B.N. Kim, B.K. Jang, J. Am. Ceram. Soc. 104 ( 2021) 3119-3129. http://doi.org/10.1111/jace.17703.

[120] V.L. Wiesner, B.J. Harder, N.P. Bansal, Ceram. Int. 44 ( 2018) 22738-22743. http://doi.org/10.1016/j.ceramint.2018.09.058.

[121] V.L. Wiesner, D. Scales, N.S. Johnson, B.J. Harder, A. Garg, N.P. Bansal, Ceram. Int. 46 ( 2020) 16733-16742. http://doi.org/10.1016/j.ceramint.2020.03.249.

[122] H. Zhang, J. Lu, X. Shan, D. Wu, X.F. Zhao, F.W. Guo, P. Xiao, J. Eur. Ceram. Soc. 40 ( 2020) 4101-4110. http://doi.org/10.1016/j.jeurceramsoc.2020.04.021.

[123] H.B. Zhao, B.T. Richards, C.G. Levi, H.N.G. Wadley, Surf. Coat. Technol. 288 ( 2016) 151-162. http://doi.org/10.1016/j.surfcoat.2015.12.053.

[124] X. Zhong, Y.W. Wang, Y.R. Niu, L.P. Huang, Q.L. Li, X.B. Zheng, Corros. Sci. 191 ( 2021) 109718. http://doi.org/10.1016/j.corsci.2021.109718.

[125] W.C. Hasz, M.P. Borom, C.A. Johnson, Protection of thermal barrier coating with an impermeable barrier coating, U.S Patents, 1999.

[126] W.C. Hasz, M.P. Borom, C.A. Johnson, Protected thermal barrier coating composite with multiple coatings, U.S Patents, 2001.

[127] P.W. Ru, Y. Hu, Y. Yang, H.J. Zhao, Y.X. Shao, Y.W. Wang, W. Li, X.Y. Wang, W.W. Sun, Y.D. Ma, Ceram. Int. 50 ( 2024) 44401-44416. http://doi.org/10.1016/j.ceramint.2024.08.288.

[128] L. Wang, L. Guo, Z.M. Li, H. Peng, Y. Ma, S.K. Gong, H.B. Guo, Ceram. Int. 41 ( 2015) 11662-11669. http://doi.org/10.1016/j.ceramint.2015.05.128.

[129] S. Bakkar, E. Zucha, J. Beam, W.H. Flanagan, S. Dixit, T.Z. Hossain, J. Am. Ceram. Soc. 106 ( 2023) 6221-6229. https://doi.org/10.1111/jace.19205.

[130] H.J. Fang, W.Z. Wang, J.B. Huang, D.D. Ye, Corros. Sci. 173 ( 2020) 108764. http://doi.org/10.1016/j.corsci.2020.108764.

[131] C.X. Shu, J.S. Wang, X.J. Lu, T.Y. Chen, M.Y. Xu, G. Du, Y.C. Rao, Z.W. Luo, Z.X. Duan, Y.F. Xie, J. Eur. Ceram. Soc. 44 ( 2024) 2537-2549. http://doi.org/10.1016/j.jeurceramsoc.2023.11.030.

[132] X.S. Zhuo, X.M. Sun, J. Wu, H. Dong, P. Shen, X.F. Zhang, X.S. Mei, J.L. Cui, Z.J. Fan, Npj. Mater. Degrad. 8 ( 2024) 85. http://doi.org/10.1038/s41529-024-00505-2.

[133] Y.Q. Guo, W.J. Song, L. Guo, X.X. Li, W.T. He, X.D. Yan, D.B. Dingwell, H. Guo, Adv. Sci. 10 ( 2023) 2205156. https://doi.org/10.1002/advs.202205156.

[134] J.Y. Feng, J. Wu, L. Guo, H.X. Guo, Corros. Commun. 6 ( 2022) 29-39. https://doi.org/10.1016/j.corcom.2021.12.002.

[135] Y.G. Liu, Y.H. Wang, W.Z. Wang, W.K. Zhang, J.H. Wang, K.B. Li, H.C. Li, P.P. Liu, S.L. Yang, C.C. Zhang, Coatings 14 ( 2024) 1123. https://doi.org/10.3390/coatings14091123.

[136] C. Gao, H.Y. Liu, J. Zhang, Z.C. Guo, H.H. Huang, J. Power Sources. 582 ( 2023) 233434. https://doi.org/10.1016/j.jpowsour.2023.233434.

[137] Z.L. Tian, B. Li, L.Y. Zheng, Z.L. Chen, CN116462507A, granted at 2024-05-07.

[138] H.J. Fang, W.Z. Wang, J.B. Huang, D.D. Ye, Ceramics International 45(16) ( 2019) 19710-19719, http://doi.org/https://doi.org/10.1016/j.ceramint.2019.06.223.

[139] L.C. Sun, X.M. Ren, Y.X. Luo, X.R. Lv, J.M. Wang, Y. Oh, J.Y. Wang, Corros. Sci. 203 ( 2022) 110343. http://doi.org/10.1016/j.corsci.2022.110343.

[140] Y. Liu, S.X. Zhao, L. Zhuang, H.L. Yu, Y.H. Chu, J. Am. Ceram. Soc. ( 2024) e20194, http://doi.org/10.1111/jace.20194.

[141] G.Q. Lin, Y.L. Wang, L.X. Yang, R.F. Sun, L.K. Wu, X.F. Zhang, H.J. Liu, C.L. Zeng, Corros. Sci. 224 ( 2023) 111529. https://doi.org/10.1016/j.corsci.2023.111529.

[142] T.Z. Tu, J.X. Liu, L. Zhou, Y.C. Liang, G.J. Zhang, J. Eur. Ceram. Soc. 42 ( 2022) 649-657. https://doi.org/10.1016/j.jeurceramsoc.2021.10.022.

[143] M. Zhou, H. Zhang, G.J. Yang, Y. Chen, X. Shan, H.T. Li, L.R. Luo, X.F. Zhao, J. Eur. Ceram. Soc. 44 ( 2024) 4055-4063. https://doi.org/10.1016/j.jeurceramsoc.2024.01.014