YSZ thermal barrier coatings exposed to gas flame

YSZ thermal barrier coatings exposed to gas flame

Surface & Coatings Technology 205 (2011) 4291–4298 Contents lists available at ScienceDirect Surface & Coatings Technology j o u r n a l h o m e p a...

2MB Sizes 5 Downloads 72 Views

Surface & Coatings Technology 205 (2011) 4291–4298

Contents lists available at ScienceDirect

Surface & Coatings Technology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s u r f c o a t

Thermal cycling behavior and failure mechanism of LaTi2Al9O19/YSZ thermal barrier coatings exposed to gas flame Xiaoyun Xie, Hongbo Guo ⁎, Shengkai Gong, Huibin Xu Key Laboratory of Aerospace Materials & Performance (Ministry of Education), School of Materials Science and Engineering, Beihang University, No. 37 Xueyuan Road, Beijing 100191, China

a r t i c l e

i n f o

Article history: Received 28 January 2011 Accepted in revised form 15 March 2011 Available online 23 March 2011 Keywords: Thermal barrier coatings (TBCs) LaTi2Al9O19 (LTA) Thermal cycling Failure mechanism Sintering

a b s t r a c t As a potential thermal barrier coating (TBC) material, lanthanum titanium aluminum oxide (LaTi2Al9O19, LTA) possesses excellent phase stability between room temperature and 1600 °C and desirable thermo-physical properties, but its thermal cycling performance is rather poor partially due to the low fracture toughness. In this work, LTA/yttria stabilized zirconia (YSZ) double ceramic layer TBC was prepared and its thermal cycling lifetime was assessed by a gas burner facility at a coating surface temperature of 1300 ± 50 °C. Chipping of LTA layer, which was different from the failure mechanism of the traditional plasma-sprayed TBCs, was visible in the center part of the TBC after about 3000 cycles, equaling to a holding time of 500 h at 1300 ± 50 °C. The failure of the LTA layer was mainly due to the decomposition and sintering of LTA exposed to extremely hightemperature gas flame. Followed by another 1000 cycles, the YSZ layer was also cracked along its interface to the bond coat. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Thermally sprayed thermal barrier coatings (TBCs) have been used successfully in gas turbine components to protect the metal parts from hot burner gasses, leading to further increase in operating temperature and decrease of the amount of cooling air [1–4]. Yttria stabilized zirconia (YSZ) is the current industrial standard topcoat materials in TBC system, which can long-term operate at temperatures below 1200 °C. At higher temperature, phase transformations and porous coating sintering result in the formation of cracks in the coating and higher thermal conductivity, which would accelerate the spallation failure of TBCs [5–8]. Therefore, investigation of novel TBC materials with ultra-high temperature capability, low thermal conductivity and long-term life time is a key problem for next generation turbine engines. In general, ceramic materials for TBC applications must satisfy several items: good sintering resistance at high temperature, phase stability between room temperature and operating temperature, low thermal conductivity as to achieve desirable thermal barrier effect, and having the capability of forming a coating with stoichiometric composition and well-controlled structure using TBC manufacturing techniques such as plasma spraying (PS) or electron beam physical vapor deposition (EB-PVD) [9–11]. No single material satisfies all these criteria. During the last decade a number of ceramic materials, mostly oxides have been suggested as new TBC materials. The new

⁎ Corresponding author. Tel.: + 86 10 8231 7117; fax: + 86 10 8233 8200. E-mail address: [email protected] (H. Guo). 0257-8972/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2011.03.047

materials cover especially doped zirconia [12–14], pyrochlores [15,16], fluorite [17,18], perovskites [19,20], and aluminates [21–24]. Recently, attention has been paid on LaTi2Al9O19 (LTA) [25–27], which is featured with a huge unit cell, complex atoms arrangement and lower symmetry. LTA reveals excellent phase stability between room temperature and 1600 °C and desirable thermophysical properties, which can be attributed to its crystallographic feature [27]. However, LTA TBC has a very short life due to its relative low fracture toughness. The lower fracture toughness could be compensated by a doubleceramic layered design, and the LTA/YSZ TBC exhibited improved thermal cycling performance at 1100 °C as compared to the single LTA layer TBC [25]. As a new TBC material candidate, further experiments have to demonstrate whether this material is suitable for higher temperature applications. In this work, LTA/YSZ TBC was prepared and its thermal cycling performance was assessed by a gas burner facility at a coating surface temperature of 1300 ± 50 °C. In addition, the thermal stability and sintering resistance of the plasma sprayed LTA coatings above 1300 °C were also studied to understand the failure mechanism of the coatings. 2. Experimental 2.1. Preparation of the LTA/YSZ TBC LTA materials were synthesized by solid-state reaction of La2O3 (99.99%), TiO2 (≥ 99.7%) and Al2O3 (≥ 99.7%). The starting materials were ball-milled and calcined at 1500 °C for 24 h. This process was repeated for three times to obtain pure products. The powders for plasma-spraying were produced by spray-dried method and the

4292

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

powders of 25–150 μm were selected for spraying LTA coatings, as shown in Fig. 1. A NiCoCrAlY bond coat with a chemical composition of Ni–21Co– 17Cr–12Al–1Y (in wt.%) was sprayed onto Ni-based superalloy K3 (Ф30 × 2.5 mm) by vacuum plasma spraying with an F4 gun. YSZ coating was then sprayed from 204 NS powders (Sulzer Metco) onto the bond coated specimen as the bottom ceramic layer in an atmosphere plasma spraying unit using a 7 M gun (Sulzer Metco). Finally, LTA coating was sprayed on the YSZ coating as the top ceramic layer. The processing parameters used for spraying YSZ and LTA coatings have been reported in the previous study [27]. Free-standing LTA coatings, which were used to investigate the thermal stability and sintering resistance ability were produced by spraying LTA powders onto the steel, followed by removing the substrate using chemical etching. 2.2. Thermal cycling test It is well known that the sprayed coatings usually contain amorphous phase due to rapid heating and cooling during spraying. If the coating is thermally cycled in an amorphous state, re-crystallization together with oxidation and thermal mismatch stresses could accelerate premature spallation failure of TBC. Considering this, the sprayed LTA/YSZ coatings were annealed in air at 1050 °C for 20 h for the recrystallization to improve thermal cycling lifetime of TBC. Thermal cycling test of the LTA/YSZ TBC was performed in a gas burner facility, as shown in Fig. 2a. The sample was heated for 20 s to the desired temperature and then held at this temperature for 10 min. During heating stage the backside of the sample was cooled by compressed air to maintain a controlled temperature gradient through the sample thickness and in this case, the coating surface temperature was 1300 ± 50 °C and the substrate temperature was 950 ± 50 °C. The surface temperature was measured by a two-color infrared pyrometer (Raytek, Model: MR1SBSF, USA, 700 °C–1800 °C) in the two-color mode with the spectral response of 0.75–1.1 μm and 0.95–1.1 μm. The ratio of the emissivity values in wavelength of 0.75– 1.1 μm and 0.95–1.1 μm is 1 during thermal cycling, which is calibrated by a single-color infrared pyrometer (Raytek3I) with long IR wavelengths of 8–14 μm and emissivity of 0.95. The substrate temperature was measured with a NiCr/NiSi thermocouple fixed to the center of substrate. During cooling stage the burner went out automatically to save the fuel and the sample was cooled by compressed air from both sides for 40 s to room temperature. The temperature curves of the surface and the substrate during thermal cycling acquired by computer are presented in Fig. 2b. The

Fig. 2. (a) Burner rig test equipment. (b) Temperature curves of the surface and the substrate during thermal cycling.

lowest and highest temperatures are 1257 °C and 1330 °C, respectively. Three samples were tested for LTA/YSZ TBC. The lifetime is determined according to the average cyclic life. 2.3. Thermal stability and sintering ability of LTA coating To value the thermal stability property, annealing was conducted at 1350 °C and 1500 °C in air using the free-standing LTA coating. The annealing temperature chosen in here was in accordance with the LTA surface temperature in the thermal cycling test. Test specimens were heated at a heating rate of about 240 °C h−1, held for a specified annealing time, and then furnace cooled. Phase stability of the LTA coating were also investigated by a differential scanning calorimeter (DSC, Netzsch STA 449C, Germany) in argon atmosphere with a heating rate of 20 °C min−1. The volume shrinkages of the LTA specimens at 1350 °C for 10 h were determined using a high temperature dilatometer (Netzsch DIL 402E, Germany) on specimens of 25 × 10 × 1 mm, respectively. 2.4. Microstructure characterization

Fig. 1. Morphology of spray-dried LTA powders used for plasma spraying.

The microstructure and composition of the coatings were investigated by a scanning electron microscopy (SEM, FEI Quanta 600, Netherlands) with an energy dispersive X-ray spectrometer (EDS) (Inca, Oxford Instruments, UK). Phase compositions were identified by X-ray diffraction (XRD, X' Pert Pro MPD, Cu Kα radiation, Netherlands). Fluorescence spectroscopy (14,200–14,900 cm−1,

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

4293

Fig. 3. The cross-section micrograph of plasma-sprayed LTA/YSZ TBC.

RM2000, Renishaw) was also used to detect the alumina phase on the LTA/YSZ TBC. 3. Results and discussion 3.1. Plasma-sprayed LTA/YSZ coating The cross-section micrograph of plasma-sprayed LTA/YSZ TBCs is shown in Fig. 3. A NiCoCrAlY bond coat of ~ 100 μm was deposited onto the K3 substrate. An YSZ layer of ~ 200 μm thickness and a LTA layer with the nearly same thickness were sprayed onto the bond coat as the bottom ceramic layer and the upper ceramic layer, respectively. Fig. 4 shows the XRD patterns of the LTA coatings as-sprayed and after 20 h heat-treatment at 1050 °C. Amorphous phase was formed in the sprayed coating due to rapid cooling of LTA coating during plasma spraying processing (pattern b). After annealing at 1050 °C for 20 h, crystallization of the coating was completed (pattern c). The chemical composition of LTA coating determined by electron probe microanalysis (EPMA) shows atom ratio of La:Ti:Al ≈ 1:2:9, indicating that the LTA coating with nearly stoichiometric composition was attained by plasma spraying. 3.2. Thermal cycling performance of LTA/YSZ coating

Intensity(a.u.)

Photographs of LTA/YSZ TBC during thermal cycling test are presented in Fig. 5. Apparent spot spallation was observed at the center of the upper LTA layer after 3000 cycles (equaling to a holding time of 500 h at 1300 ± 50 °C), as shown in Fig. 5d. The LTA layer spalled bit by bit gradually with the thermal cycling going on, as shown in Fig. 5e. After 4000 cycles, a few white areas were exposed

(e) (d) (c) (b) (a) 10

20

30

40

50

60

70

2theta(deg.) Fig. 4. XRD patterns of LTA/YSZ coatings: (a) LTA JCPDS card no. 37-1233; (b) as-sprayed; (c) heat-treated at 1050 °C for 20 h; (d) Al2O3 JCPDS card no. 71-1123; and (e) after 4000 thermal cycles (equal to a holding time of nearly 700 h at 1300 ± 50 °C).

Fig. 5. Photographs of LTA/YSZ TBC before (a) and after thermal cycling (b) 1000 cycles; (c) 2000 cycles; (d) 3000 cycles; (e) 3500 cycles; and (f) 4000 cycles (the dashed line shows the cutting direction of the coating for microstructure analysis by SEM).

which could be ZrO2 phase from the bottom ceramic layer (Fig. 5f), and thermal cycling was stopped. The surface morphologies of the LTA/YSZ coating with different thermal cycling cycles are shown in Fig. 6a–f, respectively. The micrographs are corresponding to the photographs that presented in Fig. 5. From the surface morphologies, the as-sprayed coating basically consisted of two areas: one is porous, the other smooth (Fig. 6a). The smooth area indicated a good molten state of the sprayed coating. After 1000 cycles, there were cracks formation on the coating surface (Fig. 6b). After 2000 cycles, the coating surface got gray especially in the porous part and the protrusion areas, as shown in Fig. 6c. After 3000 cycles, there were some spallations occurred and the coating included a dark area (labeled as A) and a bright area (labeled as B), as shown in Fig. 6d. According to the chemical composition analysis listed in Table 1, the composition of area A was nearly same as that of LaTi2Al9O19, whereas the composition of area B was depleted in Al. This suggests decomposition of area B occurred. After 3500 cycles, the decomposed LTA layer had spalled mostly from the part exposed to flame center (Fig. 5e) and the coating surface was mainly composed of dark areas except a few of residual white areas (Fig. 6e and f). Simultaneously, many black granules were presented in the area (Fig. 6g). According to the chemical composition analysis as shown in Table 1, the granules were basically Al2O3 phases. This also

4294

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

indicates that decomposition of LTA occurred when exposed to hightemperature flame. In addition, more cracks occurred in this area. After 4000 cycles, in the center part of the sample, where the whole LTA layer was already spalled off, a few white areas were exposed, which was determined to be ZrO2 phase of the bottom ceramic layer (Fig. 6h and i).

The elemental distribution on the Fig. 6f was analyzed by means of EDS, and the mappings of elements are shown in Fig. 7. The white area appeared on the coating surface is enriched in La and Ti but depleted in Al, while the dark area beneath the white area enriched in Al. It seems that a thin layer of oxides containing mainly La, Ti and O was formed on the coating surface during thermal cycling with a consequent

Fig. 6. SEM micrographs of surface morphology of LTA/YSZ TBC before (a) and after thermal cycling (b) 1000 cycles; (c) 2000 cycles; (d) 3000 cycles; (e) 3500 cycles; (f) high magnification of area C; (g) high magnification of area D; (h) 4000 cycles; and (i) high magnification of area E and related EDS analysis.

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

4295

Fig. 6 (continued).

precipitation of alumina beneath the layer. In addition, the chemical composition of the white area determined by EPMA shows the atom ratio of La:Ti:Al:O = 15.11:20.10:3.90:60.88. The only ternary compound reported in the Al2O3–La2O3–TiO2 system is LaTi2Al9O19 [28]. According to the ternary Al2O3–La2O3–TiO2 phase diagram [29] and the chemical composition ratio, the decomposed La- and Ti-rich oxides on the coating surface during thermal cycling test is possibly to be a solid solution of La2/3TiO3 and LaAlO3 in the formula of (1− x) La2/3TiO3 − x LaAlO3 with x ≈ 0.15. The XRD pattern of LTA/YSZ TBC after 4000 cycles comprises the main phase of LTA and some alumina (Fig. 4e). The decomposed LTA layer was very thin and spalled gradually during the thermal cycling process. The alumina phase was further determined by the fluorescence spectroscopy measurement conducted on the TBC sample after thermal cycles. The fluorescence spectroscopy result is shown in Fig. 8. After thermal cycling, the peaks at 14,402 cm−1 and 14,432 cm−1 are consistent well with the α-alumina characteristic peaks [30]. In considering the above results, it can be concluded that there was decomposition occurred on the top surface of LTA layer during thermal cycling process. The cross-section microstructure of LTA/YSZ TBC after 4000 thermal cycles is shown in Fig. 9. It is clear in Fig. 9a that failure of the coating occurred by chipping spallation. However, it is difficult to distinguish the decomposed La2/3TiO3–LaAlO3(ss) and Al2O3 in the cross-section micrograph for the gradual peeling off during thermal cycling process. Compared to the sprayed coating, there are a large number of cracks in the LTA layer, whereas the LTA and YSZ layers were still intact. These cracks would result in crashing and spallation of LTA under the erosion of high-temperature and high-speed gas flame. In the center part of sample (Fig. 9b), where the LTA layer was Table 1 The compositions of selected areas in Fig. 6. Area

La (at.%)

Ti (at.%)

Al (at.%)

O (at.%)

A B Black block

3.57 8.97 1.25

7.10 14.46 3.08

27.23 13.76 33.81

62.09 62.81 61.86

already spalled off, large delamination cracks of the bottom YSZ layer was observed closed to its interface to the bond coat. The bond coat was severely oxidized and numerous internal oxides were formed in the bond coat. It can be supposed that the failure of LTA/YSZ coating included two stages: firstly, the decomposition and peeling off of the LTA upper layer; and then delamination cracking of the YSZ bottom layer at the interface between the YSZ layer and bond coat, for the gas temperature far beyond the temperature limit of YSZ. It is admitted that the spallation of YSZ is due to phase transformation, sintering of YSZ and accelerated thickening of thermally grown oxide (TGO) formed on the bond coat. In the present work, spallation failure of the upper LTA layer occurred by chippings rather than by bulk spallation. For comparison, the thermal cycling lives of the traditional YSZ TBC and some new TBCs reported in literatures are also listed in Table 2. All of the coatings were prepared by atmospheric plasma spraying and the thermal cycling lives of the coatings were assessed by gas-burner test facility. Despite of the varieties in coating microstructures and failure criterions, the LTA/YSZ TBC showed very promising thermal cycling performance at 1300 ± 50 °C.

3.3. Phase stability and sintering ability of LTA coating To investigate the failure mechanism of the top coat LTA layer, the phase stability analyses of LTA coating were carried out. The previous study showed that LTA showed phase stability up to 1600 °C [27]. The differential scanning calorimeter analyses (DSC) were carried out with LTA free-standing coatings and the results are shown in Fig. 10. There are obviously two exothermic peaks at 860 °C and 1130 °C for the sprayed LTA coating, which correspond to the re-crystallization of LTA. For the LTA coatings heat-treated at 1050 °C for 20 h, there are neither endothermic nor exothermal peaks from room temperature to 1350 °C in the DSC curve, indicating the re-crystallization was completed. The XRD patterns of free-standing LTA coating annealed at 1350 °C and 1500 °C are compared in Fig. 11. There is no decomposition or phase destabilization of LTA coating after 500 h annealing at 1350 °C or 100 h annealing at 1500 °C, respectively.

4296

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

Fig. 7. The elemental distribution on the surface of LTA/YSZ TBC after 3500 thermal cycles, as shown in Fig. 5f.

According to the above results, there was no decomposition or phase destabilization of LTA coating from room temperature to 1500 °C, whereas the decomposition occurred during burner flame testing. It is 14402 cm

-1

-1

Intensity (a.u.)

14432 cm

14200 14300 14400 14500 14600 14700 14800 14900

Wave number (cm-1) Fig. 8. The fluorescence spectroscopy obtained from the surfaces of LTA/YSZ coatings after 4000 thermal cycling.

worth noting that, the actual temperature of burner flame could be as high as 2000 °C, although the coating surface was measured to be lower than 1400 °C. The transient surface temperature could be higher than the melting point of LTA. Moreover, the excellent insulating properties of ceramic layer would lead to overheating and even melting of the top surface. On the other hand, the combustion product of the burner that includes significant amount of water vapor may also have an impact on the phase stability of LTA. At temperatures around 1300 °C, the vapor may react with aluminum oxide producing aluminum hydroxide as a vapor phase and, thus, selectively removing alumina from the LTA. In order to understand the effect of sintering on the coating microstructure, the dilatometric measurements of the LTA coating was conducted from room temperature to 1350 °C and holding at 1350 °C for 10 h. The sprayed LTA coating has two serious contractions in the temperature ranges of 731–845 °C and 1060–1350 °C caused by the recrystallization, as shown in Fig. 12, which are in accordance with the DSC results. After 10 h sintering at 1350 °C, the coating revealed shrinkage of ~ 0.2%. For the LTA coating after 20 h heat-treatment at 1050 °C for recrystallization, the thermal expansion coefficient between room temperature and 1350 °C was in the range of 8.5–11.2 × 10− 6 K− 1, which is comparable to that of YSZ (10–11× 10− 6 K− 1). After 10 h heattreatment at 1350 °C, the coating yielded a contract of ~ 0.12% that is

Intensity(a.u.)

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

4297

(b) (a)

10

20

30

40

50

60

70

2theta(deg.) Fig. 11. XRD patterns of free-standing LTA coating heat-treated at high temperatures: (a) 1350 °C, 500 h and (b) 1500 °C, 100 h.

Fig. 9. SEM micrographs of the cross-section of LTA/YSZ coating after thermal cycling (a) rim and (b) center part.

Table 2 Lifetimes of TBC specimens for burner-rig test. TBC

YSZ [31] LZ [15] LZ/YSZ [32] LC [17]

TSurf (°C) 1238 Lifetime (h) 147

~ 1200 83

1445 270

LC/YSZ [33] LTA/YSZ

1230 ± 5 1250 ± 50 265 175

1300 ± 50 500

close to that of 8YSZ coating (0.137%) [23]. However, the actual temperature of gas flame could be far higher than 1350 °C, and the time exposed to the extremely high-temperature gas flame is as long as 700 h by the end of thermal cycling test. Thus, it could infer that LTA coating would sinter severely during thermal cycling process. The sintering of the LTA layer would lead to the fragmentation of the coating, for instance, formation of loosely attached fragments (such as shown in Fig. 9). This deteriorates mechanical integrity of the coating, and individual fragments chip off easily under the erosion of burner gas. Fracture toughness is an indication of the amount of stress required to propagate a preexisting flaw. The fracture toughness of both LTA bulk and coating was determined by an indentation technique and the results are given in Table 3. Details of this procedure can be found elsewhere [27]. The values of 1.9–2.5 MPa m1/2 for the LTA bulk are

relatively lower than the values for the YSZ bulk (6–9 MPa m1/2) [34]. The values of 0.9–1.7 MPa m1/2 for the LTA coating are nearly half of those for the YSZ coating. Lower fracture toughness is not desirable for TBCs because TBCs tend to be cracking more easily when suffering from large stresses. The re-crystallization would make volume contraction for LTA coating, and this may be the main reason for the cracks formation in the early stage of thermal cycling. With the thermal cycling going on, the stress due to CTE difference and temperature gradient accompanied the lower fracture toughness values would result in the preexisting cracks development in the LTA layer. In general, the failure mechanisms that cause spallation of LTA/YSZ TBCs are different from those of the traditional plasma-sprayed TBCs. The failure of the LTA/YSZ TBCs was mainly due to the decomposition and the sintering of LTA exposed to extremely high-temperature gas flame. The top surface LTA layer decomposed into La2/3TiO3– LaAlO3(ss) and Al2O3 and peeled off gradually during thermal cycling process. Thermal stress accompanied the lower fracture toughness values would accelerate the cracks growth in the LTA layer. Gradual spallation of LTA and the reduction in thermal insulation would result in the increase in the temperature on the YSZ layer. Then the YSZ spalled rapidly from the bond coat for the phase transformations, sintering and accelerated oxidation of metal bond layer. It is expected that the thermal cycling life of the LTA/YSZ coating can be improved by further optimization of the process parameters and coating microstructure.

10

1.4

5

1.2

(a) (b)

0

1200

1.0

1130 C

-10

dl/l0 (%)

1000

-5

-15

0.12%

0.8

800

0.6

600

1060 C

0.4

exo

400

Temperature(°C)

DSC (mW/mg)

1400

(b)

731 C

0.2 -20

845 C

(a)

0.20%

600

800

200

860 C

0.0 200

400

600

800

1000

1200

Temperature (°C) Fig. 10. DSC curves of LTA coatings (a) as-sprayed and (b) heat-treated at 1050 °C for 20 h.

0

200

400

0 1000

Time(min.) Fig. 12. Dilatometric measurements of LTA coatings (a) as-sprayed and (b) heat-treated at 1050 °C for 20 h for re-crystallization.

4298

X. Xie et al. / Surface & Coatings Technology 205 (2011) 4291–4298

Table 3 Fracture toughness of LTA samples determined by indentation technique. Sample

LTA-bulk (density: ~ 95%)

Fracture 1.9–2.5 toughness [MPa m1/2] a

LTA-coating YSZ-bulk (porosity: ~ 8%) (full dense)

YSZ-coating (porosity: ~ 12%)

0.9–1.7

1–3 [15,35,36] 1.6–3.6a

6–9 [34]

Determined in this work.

4. Conclusions The LTA/YSZ TBC prepared by plasma spraying was tested with a gas burner facility. The TBC survived more than 2000 cycles when exposed to gas flame, equaling to a holding time of more than 300 h at 1300 ± 50 °C. However, slight decomposition of LTA into La2/3TiO3– LaAlO3 (ss) and Al2O3 occurred on the surface layer of the coating after about 2000 cycles due to the ultra-high temperature of gas flame. Apparent spot spallation of LTA layer was observed in the center part of the TBC after about 3000 cycles due to decomposition and sintering of LTA. Followed by another 1000 cycles, the YSZ layer was also cracked along its interface to the bond coat.

[3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]

The research is sponsored by the National Natural Science Foundation of China (NSFC, no. 50771009 and no. 50731001) and the National Basic Research Program (973 Program) of China under grant no. 2010CB631200.

[25] [26] [27] [28] [29] [30] [31] [32] [33] [34]

References

[35] [36]

Acknowledgments

[1] R.A. Miller, Surf. Coat. Technol. 30 (1987) 1. [2] N.P. Padture, M. Gell, E.H. Jordan, Science 296 (2002) 280.

R. Taylor, J.R. Brandon, Surf. Coat. Technol. 50 (1992) 141. J.T. DeMasi-Marcin, D.K. Gupta, Surf. Coat. Technol. 68/69 (1994) 1. H.B. Xu, S.K. Gong, Y. Zhang, C.X. Zhang, Intermetallics 13 (2005) 315. H.B. Guo, S.K. Gong, K.A. Khor, H.B. Xu, Surf. Coat. Technol. 168 (2003) 23. D. Zhu, R.A. Miller, Surf. Coat. Technol. 108–109 (1998) 114. A.F. Renteria, B. Saruhan, J. Eur. Ceram. Soc. 26 (12) (2006) 2249. R.A. Milller, J. Therm. Spray Technol. 6 (1) (1997) 35. X.Q. Cao, R. Vassen, J. Eur. Ceram. Soc. 24 (1) (2004) 1. R. Vaβen, M.O. Jarligo, T. Seinke, D.E. Mack, D. Stöver, Surf. Coat. Technol. doi:10.1016/ j.surfcoat.2010.08.151. M. Matsumoto, N. Yamaguchi, H. Matsubara, Scr. Mater. 50 (2004) 867. M.N. Rahaman, J.R. Gross, R.E. Dutton, H. Wang, Acta Mater. 54 (2006) 1615. Q.L. Wei, H.B. Guo, S.K. Gong, H.B. Xu, Thin Solid Films 516 (2008) 5736. R. Vaβen, X. Cao, F. Tietz, D. Basu, D. Stöver, J. Am. Ceram. Soc. 83 (2000) 2023. H. Lehmann, D. Pitzer, G. Pracht, R. Vaβen, D. Stöver, J. Am. Ceram. Soc. 86 (2003) 1338. X.Q. Cao, R. Vaβen, W. Fischer, F. Tietz, W. Jungen, D. Stöver, Adv. Mater. 17 (2003) 1438. W. Ma, S.K. Gong, H.B. Xu, X.Q. Cao, Scr. Mater. 54 (2006) 1505. W. Ma, D. Mack, J. Malzbender, R. Vaβen, D. Stöver, J. Eur. Ceram. Soc. 28 (2008) 3071. H.B. Guo, H.J. Zhang, G.H. Ma, S.K. Gong, Surf. Coat. Technol. 204 (2009) 691. C. Friedrich, R. Gadow, T. Schirmer, J. Therm. Spray Technol. 10 (2001) 592. N.P. Bansal, D.M. Zhu, Surf. Coat. Technol. 202 (12) (2008) 2698. X.Q. Cao, Y.F. Zhang, J.F. Zhang, X.H. Zhong, Y. Wang, H.M. Ma, J. Eur. Ceram. Soc. 28 (2008) 1979. X.L. Chen, Y.F. Zhang, X.H. Zhong, J.F. Zhang, Y.L. Cheng, Y. Zhao, J. Eur. Ceram. Soc. 30 (2010) 1649. X.Y. Xie, H.B. Guo, S.K. Gong, J. Therm. Spray Technol. 19 (2010) 1179. H.B. Guo, X.Y. Xie, S.K. Gong, H.B. Xu, China Patent ZL 200710118236.5. X.Y. Xie, H.B. Guo, S.K. Gong, H.B. Xu, J. Eur. Ceram. Soc. (in press). P.E. Morgan, Mater. Res. Bull. 19 (1984) 369. S. Skapin, D. Kolar, D. Suvorov, J. Am. Ceram. Soc. 76 (1993) 2359. Q. Ma, D.R. Clarke, J. Am. Ceram. Soc. 76 (1993) 1433. H.B. Guo, R. Vaβen, D. Stöver, Surf. Coat. Technol. 186 (2004) 353. R. Vaβen, D. Stöver, Int. J. App. Ceram. Technol. 1 (2005) 351. H.B. Guo, Y. Wang, L. Wang, S.K. Gong, J. Therm, Spray Technol. 18 (2009) 665. T.K. Gupta, in: R.C. Bradt, D.P.H. Hasselman, F.F. Lange (Eds.), Fracture Mechanics of Ceramics, 4, Plenum Press, New York–London, 1978. R. Taylor, J.R. Brandon, P. Morrell, Surf. Coat. Technol. 50 (1992) 141. G.K. Beshish, C.W. Florey, F.J. Worzala, W.J. Lenling, J. Therm, Spray Technol. 2 (1993) 35.