PS-PVD gadolinium zirconate thermal barrier coatings with columnar microstructure sprayed from sintered powder feedstocks

Journal Pre-proof PS-PVD gadolinium zirconate thermal barrier coatings with columnar microstructure sprayed from sintered powder feedstocks

Shan Li, Wenting He, Jia Shi, Liangliang Wei, Jian He, Hongbo Guo PII:

S0257-8972(19)31233-2

DOI:

https://doi.org/10.1016/j.surfcoat.2019.125243

Reference:

SCT 125243

To appear in:

Surface & Coatings Technology

Received date:

7 August 2019

Revised date:

11 November 2019

Accepted date:

7 December 2019

Please cite this article as: S. Li, W. He, J. Shi, et al., PS-PVD gadolinium zirconate thermal barrier coatings with columnar microstructure sprayed from sintered powder feedstocks, Surface & Coatings Technology (2019), https://doi.org/10.1016/j.surfcoat.2019.125243

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier.

Journal Pre-proof

PS-PVD Gadolinium Zirconate Thermal Barrier Coatings with Columnar Microstructure Sprayed from Sintered Powder Feedstocks Shan Lia, Wenting Hea, Jia Shia, Liangliang Weia, Jian Hea,b,c, Hongbo Guoa,b,c* a

School of Materials Science and Engineering, bResearch Institute of Frontier Science, cKey

Laboratory of High-Temperature Structural Materials and Protective Coatings (Ministry of Industry and Information Technology), Beihang University, No. 37 Xueyuan Road, Beijing

ro

of

100191, China

-p

*To whom all correspondence should be addressed. E-mail: [email protected]

Abstract

lP

re

Tel & fax: 86 10 8231 7117

na

The characteristics of powder feedstock are very critical in the plasma spray physical vapor

Jo ur

deposition (PS-PVD) process to produce columnar structured ceramic coatings. In this study, gadolinium zirconate (GZO) powder feedstocks for PS-PVD process were prepared by solid-state reactions and spray drying. Initially, the GZO coating sprayed from the as-prepared powder revealed a low deposition rate of ~4 μm·min-1 because the powder couldn’t reach to the hottest zone of the plasma plume due to the weak binding between the primary particles. Therefore, heat treatment of the powder was performed at temperatures between 1000 °C and 1300 °C to enhance the binding strength and the powder after 1 h heat treatment at 1200 °C exhibited a moderate improved binding strength. The coating sprayed from the sintered powder showed a typical columnar structure and nearly stoichiometric composition as well as a deposition rate of 24 μm·min-1.

Journal Pre-proof

Keywords: Thermal barrier coatings (TBCs); Gadolinium zirconate (GZO); Powder feedstock;

Jo ur

na

lP

re

-p

ro

of

Plasma spray physical vapor deposition (PS-PVD).

Journal Pre-proof 1. Introduction In recent years, the demanding on higher thrust to weight ratio of aero-engines has promoted researches on advanced materials that can endure elevated temperatures. Thermal barrier coatings (TBCs) which apply on the surface of hot components are one of the key technologies in improving the durability and energy efficiency of gas-turbine engines [1]. Generally, TBCs consist of three layers,a metallic bond-coat on the superalloy substrate, a

-p

ro

formed between them under service environments [2].

of

ceramic top-coat providing main thermal insulation and a thermally grown oxide (TGO)

re

In the last decades, yttria-stabilized zirconia (7~8 wt. %Y2O3-ZrO2, YSZ) was widely utilized

lP

as the top-coat materials but limited to service temperatures below 1200 °C [3-5]. Therefore, many researchers have denoted to develop new ceramic materials for using at higher

na

temperatures in which gadolinium zirconate (GZO) is believed to be one of the promising

Jo ur

materials because of its phase stability at higher temperatures (from room temperature to 1550 °C), and lower thermal conductivities (1.33 W·m-1·K-1 at 1000 °C) [6, 7]. The coefficient of thermal expansion (CTE) of GZO (9-11.9×10-6 K-1) is very close to that of YSZ (10-11.9×10-6

K-1)

[8,

9].

Furthermore,

GZO

has

better

resistance

to

calcia-magnesia-alumino-silicate (CMAS) deposits since the infiltration of molten CMAS can be arrested by the crystalline reaction products of GZO with CMAS [10].

In addition to materials, manufacturing methods also determine the overall performance of coatings. Plasma spray physical vapor deposition (PS-PVD) is a novel coating technology, which fills the gap between plasma spray (PS) and electron beam-physical vapor deposition

Journal Pre-proof (EB-PVD) and inherits their merits [11-15]. Owing to the lower working pressure (about 100~200 Pa) and higher input power (up to 180 kW), the size of the plasma jet can reach to 2 m in length and 400 mm in diameters exhibit in PS-PVD [16-18]. From the previous results of our group, five different microstructures of YSZ coatings can be obtained along the axial direction, named as “dense lamellar structure”, “closely packed columnar structure”, “quasi-columnar structure with more nanoparticles”, “EB-PVD-like columnar structure” and

of

“quasi-columnar structure with less nanoparticles” were achieved [19]. Among these

ro

structures, the “quasi-columnar structure with less nanoparticles” exhibited the best thermal

-p

cycling lifetime of ~2000 cycles in burner rig test comparable to EB-PVD coatings and

re

achieved a low thermal conductivity of 1.15 W·m-1·K-1 at 1200 °C very close to that of APS

lP

coatings [15, 19].

na

Up to now, GZO coatings which can withstand higher surface temperature (over 1300 °C) are

Jo ur

mainly produced by APS and EB-PVD techniques [20-22]. The aim of this paper is to fabricate quasi-columnar structured GZO coatings by PS-PVD. In this study, the pyrochlore GZO powder feedstocks used for PS-PVD were successfully prepared using solid-phase reaction followed by spray drying [23]. Afterward, heat treatments were conducted at temperatures between 1000 and 1300 °C to enhance the binding strength between the primary particles. The influence of heat treatment temperature on the feedstock and coating deposition are discussed, in particular, the effects of binding strength between primary particles. .

2. Material and Methods 2.1. Preparation of powder feedstocks

Journal Pre-proof Solid-phase synthesis: The raw materials, Gd2O3 (99.9 %) and ZrO2 (99.9 %) in a precise stoichiometric ratio (Gd: Zr molar ratio of 1:1), along with deionized water and zirconia grinding balls were mixed for ball milling (QM-QX2, Nanjing University Instrument Factory) to get a homogeneous mixing. The obtained slurry was dried, and then sintered at 1550 °C for 10h.

of

Spray granulation is divided into the following three steps. First, a slurry containing GZO

ro

powder, deionized water, dispersant (polyethylene glycol), and organic binder (PVA) was ball

-p

milled for 6 h. Afterward, the prepared slurry went through a spray dryer to obtain

re

micro-sized agglomerated spherical powder and the relevant process parameters are listed in

lP

Table 1. The inlet temperature of the spray dry tower was 280 °C and the outlet temperature was 120 °C measured with a thermocouple. During the spray drying, the feeding speed of the

na

slurry was about 100 ml/min. Finally, the agglomerated spherical powder was screened

Jo ur

through a sieve of 400 mesh and a sieve of 500 mesh to get a homogeneous particle size distribution (PSD). The PDS of the as-prepared feedstock (marked as No. 1) is given in Table 2. Evaluated from the cross-section in Fig. 1, most of the primary particles in the feedstock have a diameter below 5 μm. .

Heat treatments of the as-prepared feedstock were carried out in a furnace for 1 h at 1000 °C (marked as No. 2), 1100 °C (No. 3), 1200 °C (No. 4), and 1300 °C (No. 5), respectively.

2.2. Coating deposition Nickel-based superalloy (K403) with a NiCoCrAlY bond-coat deposited by multi arc ion

Journal Pre-proof plating was used as substrate, which had dimensions of 14 mm in diameter and 3 mm in thickness. The deposition of quasi-columnar-structured GZO coatings was performed in a PS-PVD system (Medicoat, AG, Switzerland) which can achieve a low working pressure down to 200 Pa and a gun power of 60 kW. The process parameters of spraying are given in Table 3, referring to PS-PVD parameters of depositing quasi-columnar YSZ coatings [19, 24]. Before powder injection, the substrates were preheated to 900 °C detected by an infrared

of

pyrometer (Pyroview 380 Compact). During coating deposition, there was no sample rotation

-p

ro

or plasma gun movement.

re

2.3. Characterizations

lP

The size distributions of the powder feedstocks were acquired by laser particle size analyzer (LPSA). Ultrasonic treatment was performed in the LPSA for semi-quantitatively evaluating

Jo ur

na

the binding strength between primary particles in the powder feedstocks.

Surface morphologies and cross-sections of the feedstocks and coatings were observed by scanning electron microscope (SEM, FEI Quanta 200F). The thickness of the coatings was averaged out of multiple SEM images measured with the Adobe Photoshop CC 2019. X-ray diffraction (XRD, Regaku D/Max-2500, with the Cu Kα radiation) was used to identify the phase composition. Energy Dispersive Spectrometer (EDS, Inca, Oxford Instruments) was used to probe chemical composition.

3. Results and discussion 3.1. Effects of stand-off distance on the microstructure of GZO coatings

Journal Pre-proof The stand-off distance plays an important role in coatings deposition due to the distinct deposition mechanisms [19]. In order to study the influence of the feedstocks on the GZO coating deposition process, the stand-off distance was optimized first. Referring to the spraying parameters of YSZ coating by PS-PVD in our laboratory, the quasi-columnar structure is usually obtained by process parameters given in Table 3 at a stand-off distance from 1000 mm to 1200 mm. Initially, the No. 1 powder feedstock (without heat treatment)

of

was utilized for coating deposition. As shown in Fig. 2, two columnar structured GZO

ro

coatings were successfully deposited at stand-off distances of 1200 mm (coating-1-1200) and

-p

1000 mm (coating-1-1000), respectively. The coating-1-1200 (Figs. 2a and c) has typical

re

features of the “quasi-columnar structure with more nanoparticles” as described in ref. [19].

lP

The tops of the columns (Fig. 2b) are composed of some large spherical particles. From the high magnification SEM image in Fig. 2d, there are many nano-clusters and submicron-sized

na

particles adhering on the columns or exiting between the columns, which are smaller than the

Jo ur

primary particle in the feedstock. It was suggested that the feedstock would fragment into small primary particles immediately when it was injected into the plasma due to the weak agglomeration [25]. Therefore, these nano-clusters might be from partially evaporated primary particles or solidification of tiny liquid droplets [26]. As the stand-off distance decreased to 1000 mm, the coating-1-1000 exhibits different characteristics from the former one. The coating’s microstructure (Figs. 2e and g) is closer to the “quasi-columnar structure with less nanoparticles” YSZ coating described in ref. [19]. Besides, the microstructure of the column top (Figs. 2f) and the fracture surface (Figs. 2h) are faceted, which indicates that vapor deposition is dominant in this coating [27]. Referring to the calculated results in ref. [31], the supersaturated boundary layer where clusters can be formed thermodynamically is

Journal Pre-proof determined by the partial pressure and vapor pressure. As the stand-off distance increases, the supersaturated boundary layer becomes larger, and the formation of the clusters is easier. This could be the reason why there are more clusters in the coatings deposited at longer stand-off distance (1200 mm). Therefore, in the following spraying tests, coatings were all deposited at the stand-off distance of 1000 mm.

of

3.2. Effects of heat treatment temperature on feedstocks properties

ro

Although the quasi-columnar structured coating can be obtained with the No. 1 feedstock, the

-p

deposition rate (~3.9±0.4 μm·min-1) is rather low. We suppose that the feedstock probably

re

fragmented into smaller particles due to its low binding strength between the primary particles.

lP

Consequently, these particles are too small to have sufficient momentum to be injected into the core of the plasma inside the torch, which results in a low deposition rate. In order to

na

enhance the binding strength, heat treatment on the No. 1 feedstock was carried out at

Jo ur

1000 °C, 1100 °C, 1200 °C, and 1300 °C, respectively.

The microstructures of the No. 1 feedstock are displayed in Fig. 1. Most of the powders show apple-like morphology with porous surface, as shown in Figs. 1a and c. From the cross-section morphology in Figs. 1b and d, we can see that the internal structure of No. 1 feedstock is loose and porous. The micron-sized powder is agglomerated by small primary particles with binder, and there is no sintered neck between the primary particles. The microstructures of the feedstocks after heat treatment are displayed in Fig. 3. Apparent sintering is not observed in the No. 2 feedstock as the primary particles are still in angular shape as shown in Fig. 3a (I: un-sintered part). With elevated temperatures, sinter necks

Journal Pre-proof between the primary particles appeared. Seen from Fig. 3b (1100 °C), Fig. 3c (1200 °C) and Fig. 3d (1300 °C), the shape of particles became smoother, also necks between particles grew up (from II, III, to IV). It indicates that the binding strength between primary particles was enhanced by heat treatment at temperatures higher than 1100 °C.

After heat treatment, the PSD of the feedstocks (given in Table 2) didn’t change. In order to

of

evaluate the binding strength between the primary particles, ultrasonic treatment was

ro

performed to intentionally destroy the binding between the particles. The corresponding PSD

-p

after ultrasonic treatment are depicted in Fig. 4. Without heat treatment, the binding between

re

the primary particles of the No. 1 feedstock was destroyed easily by ultrasonic treatment,

lP

which shows a PSD shift to smaller sizes d10=0.25 μm, d50=1.17 μm, and d90=4.66 μm in Fig. 4a. Similarly, the PSD of the No. 2 feedstock (1000 °C) reduced. But it should be noted in Fig.

na

4b that the d90 (14.90 μm) of the No. 2 feedstock is still larger than that of the No. 1 feedstock.

Jo ur

Although no obvious sinter necks were observed in the SEM image (Fig. 3a), this comparison reveals that heat treatment at 1000 °C already has enhanced slightly the binding strength between primary particles. When the feedstock was sintered at 1100 °C, as shown in Fig. 4c, the binding strength is significantly enhanced as d50 (19.72 μm) of the No. 3 feedstock is much larger compared with the No. 2 feedstock. When it came to 1200 °C, the binding strength between the primary particles cannot be destroyed easily by ultrasonic treatment as the PSD (Fig. 4d) of the No. 4 feedstock only shows a decrease of d10: 2.75 μm. Further increasing the sintering temperature to 1300 °C, the agglomerated feedstock can no longer be destroyed by the ultrasonic treatment as the PSD (Fig. 4e) of No. 5 feedstock almost has no change. Such an evaluation of the binding strength between the primary particles by

Journal Pre-proof ultrasonic treatment during PSD measurement seems to be effective.

Beside PSD, heat treatment also have influences on flowability and apparent density of the feedstocks. Thus, before deposition, these powder feedstocks were injected into feeding tube to determine their flowability. Except for the No. 2 feedstock, other feedstocks can be continuously injected into the plasma torch successfully. The poor flowability of the No.2

of

feedstock might be on account of its weak agglomeration as seen from its PSD after ultrasonic

ro

treatment. Heat treatment at 1000 °C burned off the organic binder in the feedstock resulting

re

-p

in even worse agglomeration than the No. 1 feedstock.

GZO coatings

lP

3.3. Effects of different powder feedstocks on microstructures and deposition rates of

na

According to the results in section 3.2, the No. 2 feedstock was not used in coating deposition.

Jo ur

Coatings deposited with feedstocks No. 1, No. 3, No. 4, and No. 5 are named as coating-1-1000, coating-2-1100, coating-2-1200, and coating-2-1300, respectively.

The morphologies of GZO coatings are shown in Fig. 5. The surface morphologies (Figs. 5a, c, e and g) of the four coatings are cauliflower like. From the top view, faceted structure is dominant in the coating-1-1000, coating-2-1100, and coating-2-1200, but is less pronounced in the coating-2-1300. Wherein, the column tops of the coating-2-1100 (Fig. 5c) and coating-2-1200 (Fig. 5e) exhibit pure faceted structures, which strongly indicate that they are deposited mainly from vapor phase. In contract, the coating-1-1000 is covered by some liquid droplets as seen in Fig. 5a, and the top of the coating-2-1300 (Fig. 5g) is mainly composed of

Journal Pre-proof many nano-particles. Such results imply that a large fraction of the No. 1 and No. 5 feedstocks was not evaporated. For the No. 1 feedstock, the particles might be not injected into the core of the plasma due to its weak binding strength as the results shown in Fig. 4a. As for the No. 5 feedstock (Fig. 4e), the binding strength of the sintered particle is too strong to fragment in the plasma, and thus the feedstock cannot be fully evaporated.

of

It is obvious that all the coatings are columnar structured as shown in Figs. 5b, d, f and h. The

ro

column widths range from 10-30 μm which is obviously larger than the 5-15 μm obtained by

-p

EB-PVD coatings [13]. The interior of a single quasi-column in the coatings is not fully dense

re

but dendritic, which is the same as quasi-columnar YSZ coatings deposited by PS-PVD.

lP

Between the columns, there are more or less some nano or sub-micro particles that might come from unmelted particles or solidification of liquid droplets [27-29]. But it seems that the

na

particles in gaps between the columns in the coating-1-1000 (Fig. 5b) and the coating-2-1100

Jo ur

and (Fig. 5d) are less than that of the coating-2-1200 and coating-2-1300 displayed in Figs. 5f and h. In addition, their deposition rates differ significantly. All coatings deposited with sintered feedstocks show higher deposition rates, and the coating-2-1200 has the highest deposition rate of approx. 24 μm·min-1. As mentioned above, heat treatment has improved the binding strength between the primary particles. In other words, such increased deposition rates confirm our assumption that the low deposition rate of the coating-1-1000 is because of poor binding strength between the primary particles in the No. 1 feedstock. However, excessive binding strength is also not good for PS-PVD as the relatively low deposition rate of the coating-2-1300 has illustrated this point.

Journal Pre-proof According to the results, a schematic diagram illustrating the possible deposition process with feedstocks sintered at different temperatures is proposed in Fig. 6. Without heat treatment, the No. 1 feedstock fragmented into primary particles immediately when they come into the plasma torch because of burning off of the binder. As a result, the primary particles are too small (light) and tend to fly along the edge of the plasma jet rather than being injected into the core region of the plasma jet as shown in Fig. 6a, which leads to partially evaporated particles

of

and a low deposition rate. In this case, many deposits might fly out of the plasma jet at

ro

already the exit of the plasma torch. When the No. 1 feedstock was sintered at 1200 °C and

-p

1300 °C, respectively, the binding strength between the primary particles is enhanced due to

re

the growth of the sinter necks. This improved binding strength contributed to increasing

lP

momentum at the same conditions and support the feedstock to be injected into the core region of the plasma jet. However, excessive binding strength is not good for PS-PVD. After

na

injection into the core region, the feedstock needs to be fragmented into smaller particles by

Jo ur

the plasma for a high evaporation rate and finally a high deposition rate, which is the case of the No. 4 feedstock as shown in Fig. 6b. If the feedstock cannot be fragmented easily in the plasma (the No. 5 feedstock, Fig. 6c), the evaporation rate and deposition rate will be low because many large unmelted or partially melted particles might not contribute to the column growth. Furthermore, as the binding strength becomes stronger with elevated sintering temperatures, major fraction of the feedstock could be injected into the core of the plasma, which could be the reason that relatively more particles are observed in the column gaps in the coating-2-1200 and the coating-2-1300 as described above.

3.4. Effects of heat treatment on elemental and phase compositions of the feedstocks and

Journal Pre-proof coatings XRD and EDS are performed to analyze phase composition and the molar ratio of Gd/Zr in the coatings. According to the phase diagram, in equilibrium conditions solidification starts with the fluorite phase, and then it transforms to pyrochlore phase when the temperature drops below 1530 °C. XRD patterns in Fig. 7 reveal that the feedstocks including the original and the calcined feedstocks are pyrochlore phase (P) while all the coatings are defect fluorite

of

phase (F). This is because fast quenching during the coating depositions process which leads

ro

to insufficient atomic diffusion, and therefore forming the metastable defect fluorite phase.

-p

Similar phenomena have been widely reported in producing rare earth zirconates by plasma

re

spraying as well as EB-PVD. EDS (Table 4) analyses indicate that Zr is a little bit excess in

lP

the feedstock while in the coatings Gd tends to be slightly higher than the stoichiometric ratio. This is because the vapor pressure of gadolinia is higher than that of zirconia, thus the Gd

na

excess in the coating deposited from vapor phase is expected. Such phenomenon has also

Jo ur

been reported in ref. [30]. Overall, sintering has no obvious effects on phase and chemical composition of the coatings.

4. Conclusions

In this study, the GZO powder feedstock used for PS-PVD was successfully fabricated. The influence of sintering on the feedstock, the coating deposition, and the microstructure and composition of coatings were investigated. Some main conclusions can be drawn as follows: 1) The as-sprayed GZO coatings by PS-PVD are defect fluorite phase, with nearly stoichiometric composition. 2) The quasi-columnar structured GZO coatings were fabricated by PS-PVD at spraying

Journal Pre-proof distances between 1000 mm and 1200 mm. Wherein, the coating deposited at 1000 mm revealed fewer nanoparticles but a low deposition rate (about 3.9 μm·min-1) due to the weak binding strength between the primary particles in the powder. 3) The powder feedstock sintered at temperatures above 1100 °C revealed enhanced binding strength. The GZO coating sprayed from the powder after 1 h heat treatment at 1200 °C

of

yielded a deposition rate of around 24 μm·min-1.

ro

Acknowledges

-p

This research is sponsored by National Science and Technology Major Project under No.

re

2017-VI-0010-0081, Nature Science Foundations of China (NSFC) under grant No. 51590894

Jo ur

na

grant No. 2018M640060.

lP

and 51425102, the 111Project (B17002) and Postdoctoral Science Foundation of China under

Journal Pre-proof

References [1] N.P. Padture, G. Maurice, E.H. Jordan, Thermal barrier coatings for gas-turbine engine applications, Science, 296 (2002) pp. 280-284. [2] D.R. Clarke, M. Oechsner, N.P. Padture, Thermal barrier coatings for more efficient gas-turbine engines, MRS Bull., 37 (2012) pp. 891-898. [3] X. Ren, W. Pan, Mechanical properties of high-temperature-degraded yttria-stabilized zirconia, Acta Mater., 69 (2014) pp. 397-406.

ro

of

[4] E.P. Busso, L. Wright, H.E. Evans, L.N. Mccartney, S.R.J. Saunders, S. Osgerby, J. Nunn, A physics-based life prediction methodology for thermal barrier coating systems, Acta Mater., 55 (2007) pp. 1491-1503.

re

-p

[5] Jaeyun, CHOI, KIM, Hyungjun, LEE, Changhee, The effects of heat treatment on the phase transformation behavior of plasma-sprayed stabilized ZrO2 coatings, Surf. Coat. Technol., 155 (2002) pp. 1-10.

lP

[6] S. Lakiza, O. Fabrichnaya, C. Wang, M. Zinkevich, F. Aldinger, Phase diagram of the ZrO2-Gd2O3-Al2O3 system, J. Eur. Ceram. Soc., 26 (2006) pp. 233-246.

na

[7] Z.-G. Liu, J.-H. Ouyang, Y. Zhou, X.-L. Xia, Structure and thermal conductivity of Gd2(TixZr1−x)2O7 ceramics, Mater. Lett., 62 (2008) pp. 4455-4457.

Jo ur

[8] D. Stöver, G. Pracht, H. Lehmann, M. Dietrich, J.E. Döring, R. Vaßen, New material concepts for the next generation of plasma-sprayed thermal barrier coatings, J. Therm. Spray Technol., 13 (2004) pp. 76-83. [9] X.Q. Cao, R. Vassen, D. Stoever, Ceramic materials for thermal barrier coatings, J. Eur. Ceram. Soc., 24 (2004) pp. 1-10. [10] S. Krämer, J. Yang, C.G. Levi, Infiltration‐ inhibiting reaction of gadolinium zirconate thermal barrier coatings with CMAS melts, J. Am. Ceram. Soc., 91 (2010) pp. 576-583. [11] H.B. Guo, R. Vaßen, D. Stöver, Thermophysical properties and thermal cycling behavior of plasma sprayed thick thermal barrier coatings, Surf. Coat. Technol., 192 (2005) pp. 48-56. [12] J. Wu, H.-b. Guo, L. Zhou, L. Wang, S.-k. Gong, Microstructure and thermal properties of plasma sprayed thermal barrier coatings from nanostructured YSZ, J. Therm. Spray Technol., 19 (2010) pp. 1186-1194.

Journal Pre-proof [13] J.R. Nicholls, K.J. Lawson, A. Johnstone, D.S. Rickerby, Methods to reduce the thermal conductivity of EB-PVD TBCs, Surf. Coat. Technol., 151 (2002) pp. 383-391. [14] H.D. Steffens, Z. Babiak, M. Gramlich, Some aspects of thick thermal barrier coating lifetime prolongation, J. Therm. Spray Technol., 8 (1999) pp. 517-522. [15] L. Gao, H. Guo, L. Wei, C. Li, H. Xu, Microstructure, thermal conductivity and thermal cycling behavior of thermal barrier coatings prepared by plasma spray physical vapor deposition, Surf. Coat. Technol., 276 (2015) pp. 424-430.

of

[16] K.V. Niessen, M. Gindrat, Plasma Spray-PVD: a new thermal spray process to deposit out of the vapor phase, J. Therm. Spray Technol., 20 (2011) pp. 736-743.

-p

ro

[17] J.L. Dorier, M. Gindrat, C. Hollenstein, M. Loch, A. Refke, A. Sailto, G. Barbezat, in: Proceedings of the International Thermal Spray Conference, 2001, pp. 759-764.

lP

re

[18] A. Refke, D. Hawley, J. Doesburg, R. Schmid, LPPS thin film technology for the application of TBC systems, Proceedings of the International Thermal Spray Conference, (2005) pp. 438-443.

na

[19] C. Li, H. Guo, L. Gao, L. Wei, S. Gong, H. Xu, Microstructures of yttria-stabilized zirconia coatings by plasma spray-physical vapor deposition, J. Therm. Spray Technol., 24 (2015) pp. 534-541.

Jo ur

[20] R. Vaßen, F. Traeger, D. Stöver, New thermal barrier coatings based on pyrochlore/ysz double‐layer systems, Int. J. Appl. Ceram. Tec., 1 (2010) pp. 351-361. [21] M.P. Schmitt, A.K. Rai, R. Bhattacharya, D. Zhu, D.E. Wolfe, Multilayer thermal barrier coating (TBC) architectures utilizing rare earth doped YSZ and rare earth pyrochlores, Surf. Coat. Technol., 251 (2014) pp. 56-63. [22] E. Bakan, D.E. Mack, G. Mauer, R. Vaßen, T. Troczynski, Gadolinium zirconate/YSZ thermal barrier coatings: plasma spraying, microstructure, and thermal cycling behavior, J. Am. Ceram. Soc., 97 (2014) pp. 4045-4051. [23] R. Vassen, S. Schwartz, W. Jungen, Spray-drying of ceramics for plasma-spray coating, J. Eur. Ceram. Soc., 20 (2000) pp. 2433-2439. [24] C. Li, H. Guo, L. Gao, L. Wei, S. Gong, H. Xu, Microstructures of yttria-stabilized zirconia coatings by plasma spray-physical vapor deposition, J. Therm. Spray Technol., 24 (2014) pp. 534-541.

Journal Pre-proof [25] W. He, G. Mauer, M. Gindrat, R. Wäger, R. Vaßen, Investigations on the nature of ceramic deposits in plasma spray–physical vapor deposition, J. Therm. Spray Technol., 26 (2016) pp. 83-92. [26] G. Mauer, R. Vaßen, D. Stöver, Plasma and particle temperature measurements in thermal spray: approaches and applications, J. Therm. Spray Technol., 20 (2011) pp. 391-406. [27] W. He, G. Mauer, M. Gindrat, R. Wäger, R. Vaßen, Investigations on the nature of ceramic deposits in plasma spray–physical vapor deposition, J. Therm. Spray Technol., 26 (2016) pp. 1-10.

ro

of

[28] B.J. Harder, D. Zhu, M.P. Schmitt, D.E. Wolfe, Microstructural effects and properties of non-line-of-sight coating processing via plasma spray-physical vapor deposition, J. Therm. Spray Technol., 26 (2017) pp. 1-10.

re

-p

[29] B. Zhang, W. Song, L. Wei, Y. Xiu, H. Xu, D.B. Dingwell, H. Guo, Novel thermal barrier coatings repel and resist molten silicate deposits, Scripta Mater., 163 (2019) pp. 71-76.

lP

[30] G. Mauer, N. Schlegel, A. Guignard, M.O. Jarligo, S. Rezanka, A. Hospach, R. Vaßen, Plasma spraying of ceramics with particular difficulties in processing, J. Therm. Spray Technol., 24(1-2) (2015) pp. 30-37.

Jo ur

na

[31] M. Liu, K. Zhang, Q. Zhang, M. Zhang, G. Yang, C. Li, C. Li, Thermodynamic conditions for cluster formation in supersaturated boundary layer during plasma spray-physical vapor deposition. Appl. Surf. Sci., 471 (2019) pp. 950-959

Journal Pre-proof Table captions Table 1 Parameters of spray drying of GZO powders Table 2 Properties of the GZO feedstocks Table 3 Processing parameters of PS-PVD GZO coatings

Jo ur

na

lP

re

-p

ro

of

Table 4 The Gd/Zr molar ratio of the GZO powder feedstock and the coatings

Journal Pre-proof

Figure captions Fig. 1. SEM micrographs of the as-prepared GZO powders: (a, c) surface morphologies, (b, d) cross-sectional morphologies. Fig. 2. Surface and fracture morphologies of the coatings sprayed with the as-prepared powder: (a, b) surface and (c, d) fracture morphologies of the coating-1-1200, (e, f) surface and (g, h) fracture morphologies of the coating-1-1000. Fig. 3. Surface morphologies of the powder feedstocks after heat treatment at different temperatures: (a) 1000 °C, (b) 1100 °C, (c) 1200 °C, (d) 1300 °C.

ro

of

Fig. 4. Particle size distributions of the feedstocks after ultrasonic treatment: (a) the as-prepared powder, (b)-(e) powders heat-treated at 1000 °C, 1100 °C, 1200 °C, and 1300 °C, respectively.

re

-p

Fig. 5. Surface morphologies (a), (c), (e) and (g) and cross-sectional morphologies (b), (d), (f) and (h) of the GZO coatings sprayed from the as-prepared powder, and the powders heat-treated at 1000 °C, 1200 °C and 1300 °C, respectively.

lP

Fig. 6. Schematic diagrams of deposition process with feedstocks sintered at different temperatures. (a) without heat treatment (low binding strength), (b) 1200 °C (appropriate binding strength), (c) 1300 °C (excessive binding strength).

Jo ur

na

Fig. 7. XRD patterns of (a) the GZO powder feedstock, and the as-sprayed coatings: (b) coating-1-1000, (c) coating-2-1100, (d) coating-2-1200 and (e) coating-2-1300.

Journal Pre-proof Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Jo ur

na

lP

re

-p

ro

may be considered as potential competing interests:

of

☒The authors declare the following financial interests/personal relationships which

Journal Pre-proof Author Contribution Section Hongbo Guo designed the experiment. Shan Li did the heat treatment part of the experiment and wrote the paper. Wenting He analyzed the data and modified the paper.

Jo ur

na

lP

re

-p

ro

of

Jia Shi and Liangliang Wei helped spray the coatings. Jian He did XRD and EDS test.

Journal Pre-proof Table 1 Parameters of spray drying of GZO powders Inlet

Feeding

Gas

Outlet temperature

Feeding pressure

temperature

speed (°C)

(r·min-1)

(°C)

40

1.0

na

lP

re

-p

ro

of

120

Jo ur

280

(1.013×105 Pa)

flow (m3·h-1) 5

Journal Pre-proof Table 2 Properties of the GZO feedstocks Particle size distribution (μm)

Heat-treatment temperature (°C)

D10

D50

D90

No. 1

----

19.2

27.7

40.3

No. 2

1000

18.8

27.0

39.2

No. 3

1100

18.8

27.2

39.5

No. 4

1200

18.7

26.9

39.1

No. 5

1300

18.8

of

Number

39.5

Jo ur

na

lP

re

-p

ro

27.1

Journal Pre-proof Table 3 Processing parameters of PS-PVD GZO coatings Gun

Carrier Current

Ar

He

power (A)

(slpm) (slpm)

Feed rate

Distance

(g·min-1)

(mm)

gas

(kW)

(slpm)

Coating-1 60

2000

30

60

~4

1200

11

60

2000

30

60

~4

1000

11

-1200 Coating-1

Jo ur

na

lP

re

-p

ro

of

-1000

Journal Pre-proof Table 4 The Gd/Zr molar ratio of the GZO powder feedstock and the coatings Coating-2-1100

Coating-2-1200

Coating-2-1300

0.98

0.99

1.04

1.03

1.05

na

lP

re

-p

ro

of

Coating-1-1000

Jo ur

Gd/Zr

Feedstocks

Journal Pre-proof

Highlights 

The GZO feedstocks for PS-PVD were prepared.



Effects of heat treatment on GZO feedstocks and coating deposition were studied.



Sintering of feedstocks at 1200 °C helps to improve deposition rate of the

Jo ur

na

lP

re

-p

ro

of

coating.