1 Introduction

For more advanced aircraft, there is insatiable demand for more powerful aero-engines, which can be accomplished by increasing the turbine gas-inlet temperature [1, 2]. Over several decades, the hot-section structural materials have developed from wrought, conventionally cast, directionally solidified to single-crystal alloys [3, 4]. This considerably elevates the gas-inlet temperature. However, this method faces “bottleneck” due to the high-temperature capability limit of superalloys. For further significant increase in the gas-inlet temperature, thermal barrier coating (TBC) in conjunction with advanced cooling technology have been employed for hot-section components of aero-engines [5,6,7]. A TBC system is composed of a ceramic top coat, a bond coat and the substrate. Usually, there is a thermally grown oxide (TGO) layer forming on the bond coat during TBCs operation. The ceramic top coat provides thermal insulation, which has large thermal expansion mismatch to the substrate. Hence, a metallic bond coat is prepared between the top coat and substrate to alleviate the thermal expansion mismatch, and also protect the substrate from oxidation.

The research on TBCs could trace back to the 1950s, and the first usage was reported in the 1960s [8]. After the identification of partially yttria-stabilized zirconia (YSZ) in the 1980s, TBCs development has made a major step forward [9, 10]. YSZ has many unique properties fit excellently to the requirements of a TBC system, such as low thermal conductivity, high thermal expansion coefficient (TEC), high toughness, good phase stability, good compatibility with the TGO layer, and low sintering rate [4, 7, 11,12,13]. However, with ever-increasing demands for higher gas-inlet temperature, YSZ TBCs face severe limitations. The as-fabricated YSZ coatings exhibit a non-transformable metastable tetragonal (t′) phase, which has high toughness resulting from a ferroelastic toughening effect. At temperatures higher than 1250 °C, t′ phase decomposes to tetragonal (t) and cubic (c) phases, and the former transforms to a monoclinic (m) phase during cooling accompanied with excessive volume expansion, which would cause cracks in the coating leading to premature failure of TBCs. Additionally, high-temperature accelerates the coating sintering and causes severe corrosion to TBCs. Therefore, search is underway for develo** TBC materials that have even better phase stability, higher sintering resistance, lower thermal conductivity and better corrosion resistance.

Although TBCs in combination with the cooling technology largely enhance the operating temperature of hot parts of aero-engines, superalloys still have their temperature-capability limit. It is thus not optimistic that the gas-inlet temperature achieves the goal of above 1700 °C [2]. In response to this, SiCf/SiC ceramic matrix composites (CMCs) have been proposed and designed to gradually replace nickel-based superalloys for aero-engine hot-end parts. However, one notable challenge hindering the practical implementation of CMCs is their insufficient environmental durability within combustion environments [14,15,16]. The presence of water vapor, a byproduct of the combustion reaction, triggers chemical reaction with the protective silica scale that forms on the CMCs, resulting in the formation of gaseous reaction products such as Si(OH)4. In high-pressure and high-velocity combustion environments, this reaction accelerates the degradation of CMCs materials. Additionally, CMCs are susceptible to other severe corrosion when exposed to combustion environments. Therefore, environmental barrier coatings (EBCs) have been developed, which play a critical role in protecting SiCf/SiC hot-section components from the harsh conditions encountered in aircraft engines. EBCs are applied to surfaces of components such as turbine shrouds, combustors, seal segments, vanes, and blades, in conjunction with CMCs, to protect against corrosion, oxidation and thermal cycling, and provide thermal insulation [17,18,19,20]. Thereby EBCs effectively improve the durability and reliability of CMCs components, ensuring optimal performance and extending their service life, which contributes to the safe and efficient operation of aircraft engines.

Increasing operation temperature of aero-engines leads to aircraft more easily be detected in modern information technology warfare. The stealth of aero-engines and aircraft has become crucial to their capabilities to survive and penetrate defenses [21]. Therefore, stealth technology, also known as low detection technology, has become a prominent feature of next-generation fighter and a key focus of military powers [22,23,24]. Due to the high application ratio of radar wave and infrared detection, accounting for 60% and 30% respectively, radar and infrared stealth is particularly important. For the aeroengine of fighters, the radar scattering signals generated by the engine aft compartments, and the infrared radiation signals generated by hot-end components of the aft compartment and the exhaust jet account for over characteristic 95% of signals at the rear of the aircraft. Therefore, the aft compartments of engines must be made of high-temperature stealth materials to achieve low detection and also a high thrust-weight ratio [36, 37]. EB-PVD technology utilizes high-energy electron beams to bombard evaporative materials under vacuum conditions, rapidly heating them to convert into atomic and molecular forms, which then condense onto the preheated substrate surface to form a coating. EB-PVD coatings have a unique columnar crystal structure, which enables the coating to have high strain tolerance [38,62, 63]. For example, Guo et al. [64] selected Gd2O3 and Yb2O3 to co-dope YSZ, and confirmed that it exhibits better phase stability and lower thermal conductivity than the YSZ counterpart.

In order to meet the requirements of next-generation aeroengines, within the last decades much work has been done to explore novel promising TBCs candidates, which is generally restricted by some basic principle: 1) low thermal conductivity, 2) high fracture toughness, 3) no phase transformation from room temperature to the operating temperature, 4) TEC matching the metallic substrate, 5) high corrosion resistance against CMAS and molten salt, and 6) good sintering resistance. Several ceramic materials suitable for the topcoat of TBCs are summarized as follows.

2.1.1 Hexaaluminates

Lanthanate hexaaluminates (LnMAl11O19, M = Mg, Mn to Zn, Cr, Sm) with a magnetoplumbite-type structure are regarded as potential TBC materials due to the low thermal conductivity, high phase stability, excellent sintering resistance and high melting point [65, 66]. In particular, LaMgAl11O19 (LaMA) is the most interesting one [67]. LaMA/YSZ double-ceramic-layer (DCL) coatings produced via APS have long thermal cycling lifetime of ~ 11,749 cycles at ~ 1370 °C [68]. Note that during the spraying process, LaMA partially decomposes and forms much amorphous phase upon rapid cooling. In service, recrystallization of the amorphous phase at high temperatures probably affects the reliability of the coating [69].

2.1.2 Rare earth phosphates

Rare earth phosphates (REPO4) exhibit low thermal conductivity, good chemical compatibility with TGO and high phase stability, which have been considered as promising candidates for TBC applications [70, 71]. Specially, it has been reported that REPO4 (RE = Nd, Sm, Gd) is highly resistant to CMAS attack [72]. In addition, GdPO4 has outstanding anti-corrosion behavior in the presence of V2O5 and V2O5 + Na2SO4 at 900 °C [73].

2.1.3 Rare earth zirconates

Rare earth zirconates (RE2Zr2O7), first reported by Vaßen [74], possess lower thermal conductivity, higher melting point and better phase stability than traditional YSZ for application at high temperatures (> 1300 °C) [75]. RE2Zr2O7 commonly has pyrochlore structure or defect fluorite structure (Fig. 1), which is determined by the r(RE3+):r(Zr4+) ratio, i.e., the former is formed when the value falls into the range between 1.46 and 1.78, otherwise the latter is formed [76]. The pyrochlore structure is generally written as A2B2O6O´, in which A, B, O and O´ ions is located in 16c, 16d, 48f and 8b sites, respectively. Note that there is an unoccupied 8b site and the neighboring O2– deviates from the common site. In comparison, the cations in defective fluorite structure are completely disordered and there is only one 8c site for O2–.

Fig. 1
figure 1

Schematic illustrations demonstrating the pyrochlore structure (a) and defective fluorite structure (b)

Among these zirconates, Gd2Zr2O7 has been reported to be the most promising one. The thermal conductivity of APS Gd2Zr2O7 coating varies from 0.6 to 1.4 W m–1 K–1 depending on the temperature [77]. The TEC ranges between 10.4 × 10–6 and 10.6 × 10–6 K–1, slightly lower than that of the YSZ coating [65, 78]. The distinct disadvantages of Gd2Zr2O7 are its relatively low fracture toughness and chemical incompatibility with TGO. In order to extend the service life of Gd2Zr2O7 coatings, Doleker et al. designed a Gd2Zr2O7/YSZ DCL coating, and confirmed that its thermal cycles are nearly twice that of simple Gd2Zr2O7 counterpart [79]. Furthermore, Guo et al. selected Yb2O3 to modify Gd2Zr2O7, aiming to improve its thermophysical properties [80]. The optimized (Gd0.9Yb0.1)2Zr2O7 exhibits very low thermal conductivity (0.8–1.1 W m–1 K–1, 20–1600 °C), and has been regarded as one of the most promising RE2Zr2O7 materials (Fig. 2).

Fig. 2
figure 2

XRD patterns (a) and thermal conductivity (b) of (Gd1–xYbx)2Zr2O7 (x = 0, 0.1, 0.3, 0.5, 0.7) ceramics [80]

2.2 Thermal barrier coating fabrication technologies

Nowadays, EB-PVD and APS are the two main methods used for TBCs preparation [81,82,83]. During EB-PVD processing, the working pressure is low and can vary from 5 to 0.1 Pa, and sometimes even lower [84]. The high energy electron beam heats ingots, producing a steam that subsequently deposits on the surface of the substrate. This enables to produce columnar structured coatings with both intra- and inter-columnar porosities. EB-PVD TBCs exhibit high-strain tolerance and good erosion resistance but high thermal conductivity as compared to APS TBCs. Figure 3 shows a typical EB-PVD YSZ TBC microstructure in cross section [85]. EB-PVD has also been used to produce new ceramic compositions such as alternative stabilizers doped zirconia, fluorites, and pyrochlores [80, 86,87,88]. EB-PVD is a line-of-sight process. Only surfaces which are in direct line-of-sight to the coating source can be coated. Therefore, components with complex geometries and shadowed areas are very difficult to be coated homogeneously.

Fig. 3
figure 3

Typical EB-PVD YSZ TBC microstructure in fractured cross section in Ref. [85]

During APS processes, ceramic powder particles injected into the plasma jet are heated into molten or semi-molten splat, and accelerated simultaneously followed by deposition onto the substrate. A lamellar microstructure is obtained by the stacking of splats. It allows rapid coating manufacturing, and the produced coatings are thick (300 ~ 3000 μm) and porous (10–25%) which exhibits good thermal insulation [89, 90]. As reported, the thermal conductivity of APS YSZ can reach 0.8–1.2 W/(mK) [91]. However, APS coatings are unavoidable to have drawbacks, i.e., low thermal shock resistance especially for thick TBCs. In recent years, with the introduction of segmentation cracks, which run perpendicular to the coating surface, the thermal shock resistance of thick TBCs has been improved, however compromising the thermal insulation performance. Figure 4 shows the SEM micrographs of cross-section of APS YSZ coating with segmentation cracks from Ref. [89]. The segmentation cracks were accompanied by horizontally oriented branching cracks (as marked by arrows in Fig. 4). These cracks improve the strain tolerance of the coating as an opening of the cracks during tensile loading, similar to the behavior of inter-column gaps of EB-PVD coatings. Besides, APS is also a line-of-sight process.

Fig. 4
figure 4

SEM micrographs of APS YSZ coatings with segmentation cracks in Ref. [89]

PS-PVD is a newly emerging processing technology, which combines the advantages of APS (high deposition rates and cost-efficiency) and EB-PVD technologies [92]. It is developed based on the low-pressure plasma spray process (LPPS, also known as vacuum plasma spraying, VPS). However, the plasma jet of PS-PVD is greatly different from that of LPPS due to the lower working pressure (ranging from 1 ~ 2 mbar) and higher power plasma gun [93]. The plasma plume of PS-PVD can expand to more than 2 m long and 200–400 mm in diameter [41, 94]. Moreover, the temperature of supersonic plume can exceed 6000 K due to the high power level (> 100 kW). With this technology, the injected powder is not only molten into liquid droplets but evaporated to build up a columnar coating. By adjusting the spray parameters, different microstructures of coatings could be obtained [95,96,97], as shown in Fig. 5. Among the coatings investigated, the quasi-columnar coating exhibits the lowest thermal conductivity of around 1.1 W/mK at 1200 °C, and is highly strain tolerant during thermal shock tests [44]. Its erosion resistance is lower than that of EB-PVD coatings but at a level of standard APS TBCs with 15% porosity [93]. Besides, due to its high velocity and large dimension, the plasma gas stream is able to flow around complex geometries and “forced” to go through the shadowed areas, thereby making the PS-PVD a non-line-of-sight coating process [93]. This allows the complete coverage of parts having complex geometries and shadowed areas, such as multiple turbine air foils or double vanes. So far, with the use of suitable feedstock materials, many types of TBCs can be produced by PS-PVD such as La2Ce2O7 and modified Gd2Zr2O7 coatings [98,99,100].

Fig. 5
figure 5

SEM morphologies of different microstructures in Refs. [96, 97] deposited by PS-PVD: (a) dense splat-like microstructure, (b) mixed microstructure coating, (c) EB-PVD-like columnar coatings and (d) quasi-columnar coatings

Suspension Plasma Spraying (SPS) and solution precursor plasma spraying techniques (SPPS) are also very attractive for producing TBCs. Here by using liquid feedstock instead of solid particles, delivering finer particles was realized. This allows the generation of unique microstructures such as nano-sized architectures, high density of segmentation cracks, and columnar structures [101,102,103]. The unique structure imparts SPS/SPPS coatings superior properties as compared to conventional coatings, such as lower thermal conductivity, higher fracture toughness and higher strain tolerance. Since the last two decades, increasing work has been devoted to understand the SPS/SPPS process and properties of the obtained coatings. So far, SPS has also successfully been used to deposit novel TBCs like perovskites and pyrochlores coatings [104,105,106].

2.3 Thermal barrier coating corrosion and protection

High-temperature corrosion is a key factor causing TBCs premature failure, involving CMAS corrosion, molten salt corrosion, and their coupling corrosion. CMAS corrosion behavior of YSZ TBCs was first reported by Stott et al. [107], who collected natural CMAS from the Middle East and deposited it on APS TBCs, followed by heat treatment at 1300–1600 °C for 120 h. They found that the stabilizer Y2O3 is prior to dissolve into molten CMAS as compared to ZrO2, causing the coating deficient in Y. As a result, the coatings undergo a phase transformation to m phase accompanied with excessive volume expansion. Then, Kramer et al. systematically investigated the CMAS corrosion mechanisms of TBCs, and proposed a dissolution–reprecipitation model [61], which has been accepted by many researchers in an aspect of thermo-chemical interaction. Besides, CMAS destroys TBCs in a thermal–mechanical interaction aspect. Molten CMAS penetrates the coating through its pores and micro-cracks, and solidifies during cooling. This causes the coating become compact, degrading its strain tolerance and producing large stress in the coating. When the stress is accumulated to a certain extent, the coating would spall off.

In view of the molten salt corrosion to TBCs, it occurs when the low-grad fuels are used. Na, S, V and other impurities in fuels are extremely corrosive to TBCs, especially at 600–1050 °C. Exposure to molten salt, rare earth elements in YSZ and other rare earth contained novel TBC candidates have high tendency to leach out [108,109,110,111,112]. This induces phase destabilization and microstructure destruction of TBCs, and also degrades thermo-physical and mechanical properties of the coating. As a result, the coating faces performance degradation and premature failure.

CMAS and molten salt coexist further aggravates the corrosion to TBCs. Naraparaju et al. found that in the presence of CMAS, CaSO4 decomposes at much lower temperature, and the penetration ability of CMAS containing CaSO4 in EB-PVD 7YSZ is stronger than its calcium sulfate-free counterpart [113]. Before, it was considered that CMAS attack only occurs in aero-engines, there was no CMAS issue in ship engine components because the probable operating temperature of marine engines is 200 ~ 400 °C below CMAS melting temperature (≥ 1100 °C). However, Shifler from Office of Naval Research (USA) and Choi from Naval Air Systems Command (USA) observed CMAS attack in ship engine components [114], and they first indicated that in marine environment, CMAS melts at much lower temperature due to the coupling of sea salt. When operated in inshore environment, the air with high humidity and high salt spray inhaled by gas engines contains much salt. Accordingly, CMAS and sea salt can be coupled attacking the TBCs. Guo et al. found that CMAS + salt (NaVO3, Na2SO4 or NaCl) has lower melting point and viscosity than CMAS, e.g., the melting temperature of NaVO3 + CMAS mixture is lower than that of CMAS by ~ 50 °C [123,124,125,126]. Some newly developed TBC materials, such as rare earth phosphate and Sc2O3 doped Gd2Zr2O7, were reported to have some resistance to molten salt corrosion [127, 130]