Hot corrosion behavior of Y 3 NbO 7 as thermal barrier coating material

Abstract

Y₃NbO₇, a simple ternary oxide, has shown excellent high-temperature stability lately. The growing need for more durable thermal barrier coatings in extreme conditions underscores the importance of exploring materials that can resist challenges like hot corrosion. This study examines the thermal barrier properties and hot corrosion characteristics of Y₃NbO₇. Thermal diffusivity and the coefficient of thermal expansion were measured using laser flash technique and dilatometry respectively. The interaction with the thermally grown oxide (TGO) was evaluated by exposing the Y₃NbO₇ powder to Al₂O₃ at high temperatures. To demonstrate the viability of the as-synthesized powder for coating deposition, a plasma sprayed TBC architecture was developed using the Y₃NbO₇ powder. The behavior of as-deposited coating under thermal shock was studied using thermal cycling test. Hot corrosion study was performed on Y₃NbO₇ pellets in a species composed of 32 wt% Na₂SO₄ and 68 wt% V₂O₅ at 900°C. The corrosion products were examined with X-ray Diffraction and Scanning Electron Microscopy. The corrosion phenomenon was found to align with the Lewis acid-base theory, where Y₂O₃ was preferred over Nb₂O₅ for the acidic activity of V₂O₅.

Keywords

hot corrosion thermal barrier coating lanthanide niobates

Introduction

Thermal barrier coatings (TBCs) are critical in advanced gas turbines and aircraft engines, providing essential thermal protection for hot-section components while enhancing oxidation and corrosion resistance.¹ Traditionally, yttria-stabilized zirconia (YSZ) has been the preferred material, known for its low thermal conductivity and high thermal expansion coefficients (CTE). However, YSZ is limited to applications below 1200°C due to phase degradation, leading to potential failures.²

To address these limitations, alternative ceramics including lanthanide niobates (Ln₃NbO₇) are being increasingly explored. These materials, comprising rare earth elements like La, Nd, Sm, Eu, Gd, and Dy, show significant promise due to their unique properties. Unlike traditional TBC materials, lanthanide niobates exhibit low thermal conductivity, exceptional phase stability, and enhanced mechanical toughness.³ The hardness, fracture toughness, specific heat, diffusivity, and CTE of Ln₃NbO₇ ceramics are being actively investigated, revealing their potential as superior TBC materials.^4–6

The techniques generally employed to reduce thermal conductivity were introducing structural disorder by adding additional atoms^7,8 or increasing entropy through multi-component oxides,⁹ which result in enhanced phonon scattering. However, these multi-component oxides often present technological challenges, such as maintaining precise stoichiometry during coating deposition.

Recent studies highlighted the importance of inherent chemical inhomogeneity and charge disorder within lanthanide niobates, which contribute to their impressive thermophysical characteristics. Further, in comparison to other fluorite-structured ceramics, such as rare earth zirconates (RE₂Zr₂O₇) and lanthanum cerium oxides (La₂Ce₂O₇), lanthanide niobates stand out for their ability to maintain low thermal conductivity while providing high-temperature performance.² Techniques such as doping and structural optimization further enhance their properties, making them viable candidates for next-generation TBC applications. As the demand for high-efficiency thermal management systems grows, lanthanide niobates are emerging as key materials in the development of advanced TBCs with their unique combination of low thermal conductivity, good phase stability, and better mechanical strength.

While other rare earth niobates like La₃NbO₇, Nd₃NbO₇, and Sm₃NbO₇ also exhibit beneficial properties, Y₃NbO₇ often strikes a better balance between thermal conductivity, phase stability, and mechanical properties.^10,11 Y₃NbO₇ emerged as a promising candidate for TBCs due to its excellent thermal stability and low thermal conductivity. Studies have shown that Y₃NbO₇ exhibits superior phase stability at elevated temperatures, making it a potential alternative to traditional TBC materials like yttria-stabilized zirconia (YSZ).¹² Additionally, its high melting point make it a suitable material for protecting high-temperature components in gas turbines and aero-engines.¹³ The incorporation of Y₃NbO₇ into TBC systems has demonstrated improvements in thermal insulation and overall performance, which may address some of the limitations of conventional TBC materials, such as YSZ's sensitivity to thermal shock and sintering.¹² Although some studies have investigated the radiation thermal protection¹⁴ and sintering characteristics¹⁵ of Y₃NbO₇, there is a lack of comprehensive evaluation of the TBC characteristics of the material.

Moreover, the chemical interactions of Y₃NbO₇ with molten sulfate and vanadate salts have not been extensively studied. Hot corrosion is a major concern for TBC failure, especially in environments where low-quality fuels with contaminants like V, S, and Na are used.¹⁶ The salts based on these contaminants melt and tend to build up on TBCs at temperatures between 600 and 1000°C.¹⁷ This work aims to address these gaps by thoroughly analyzing the thermal barrier and hot corrosion characteristics of Y₃NbO₇.

Y₃NbO₇ was synthesized using a solid-state route, and atmospheric plasma spray coatings and pressureless-sintered pellets were fabricated using the powder. Inconel-600 was chosen as the substrate for plasma spray coating due to its excellent high-temperature strength, oxidation resistance, and ability to withstand extreme environments, making it ideal for applications in gas turbines and jet engines. With a service temperature range of 700–1100°C, it provides a stable base for TBC systems, enabling components to endure higher temperatures than the alloy's own limit. The pellets and coating were then examined for their thermal properties, thermal cycling behavior and hot corrosion resistance in Na₂SO₄+ V₂O₅ salt environment. The hot corrosion experiment focused on identifying the major corrosion products and modifications in morphology.

Materials and methods

Y₃NbO₇ was synthesized using a solid-state technique with Y₂O₃ (99.0% purity) and Nb₂O₅ (99.99% purity) as precursors. The precursors were blended in the appropriate stoichiometry using a ball mill (Fritsch, Pulverisette 5). Afterwards, the blend was heated at 1400°C for 12 h with intermittent milling.

X-ray Diffraction (CuKα, wavelength 1.54060 Å) was performed on the synthesized powder for phase identification and crystal structure determination. The particle morphology and elemental proportions were examined using Scanning Electron Microscopy with Energy Dispersive Spectroscopy (SEM-EDS, Zeiss Sigma).

As-synthesized Y₃NbO₇ was deposited onto an Inconel-600 substrate using the plasma spraying (APS, Oerlikon Metco) to fabricate a TBC system. The surface of Inconel-600 was prepared through alumina grit blasting to achieve strong adhesion. The parameters for the coating deposition are provided in Table 1.

Table 1.

Parameters for APS coating fabrication.

Current (A)	Voltage (V)	Spray Distance (mm)	Argon		Hydrogen		Powder feeder (g/min)
500	50-70	75–100	Pressure (psi)	Flow rate (SCFH)	Pressure (psi)	Flow rate (SCFH)	50–60
500	50-70	75–100	100–120	100–115	50	15–18	50–60

Scanning Electron Microscopy (Zeiss Sigma) was performed on the coated sample to investigate its microstructure and splat structure. Additionally, XRD analysis was done to observe any probable decomposition of Y₃NbO₇ during the APS process.

For investigating thermal barrier and hot corrosion characteristics, cylindrical Y₃NbO₇ pellet samples were fabricated. These samples were fabricated using a hydraulic press and sintered at 1500°C for 6 h. The surface characteristics of the samples were analyzed using SEM imaging and X-ray diffraction was performed to identify any possible decomposition that might occur during the sintering process.

The thermal diffusivity of Y₃NbO₇ was determined by the Laser Flash Method (LFA, DLF-2, Waters) on a sample (φ10 mm×2 mm). Thermal conductivity (k) was then obtained using Equation 1:

k = Cp \times ρ \times α

(1)

where C_p is the speciﬁc heat capacity (J kg⁻¹ K⁻¹); ρ is the density (kg m⁻³); α is the thermal diffusivity (m² s⁻¹).

The Klemens relation was employed to calculate the thermal conductivity, with adjustments made to account for the porosity in the sintered pellet.¹⁸

\frac{k}{k^{'}} = 1 - \frac{4 δ}{3}

(2)

where δ is the fractional porosity; k is the thermal conductivity of the pellet (W m⁻¹ K⁻¹), and k’ is the thermal conductivity of 100% dense case.

The thermal expansion coefficient (CTE) was determined up to 1000°C with dilatometry technique. The measurement was performed on a cylindrical pellet of φ 10 mm × 30 mm dimensions. To examine the interaction of Y₃NbO₇ with thermally grown oxide (TGO), the as-prepared Y₃NbO₇ was blended with Aluminum Oxide (Al₂O₃) in equal molar amounts and was heated at 1200°C for 6 h. XRD was conducted to observe any probable reactions with Al₂O₃.

The ultrasonic velocities (both longitudinal and transverse) were obtained in a sample (φ10 mm×5 mm) with the transducers in pulse-echo configuration. These velocities were applied in Equations 3 and 4 to determine the elastic modulus.¹⁹ An ultrasonic probe (M110-RM, 5 MHz) along with a thickness gauge (Olympus, 45MG) was used to obtain the longitudinal velocity. The transverse velocity was found using a shear probe (5 MHz) in conjunction with an oscilloscope (Tektronix, MSO44)

μ = \frac{1 - 2 {(V_{T} / V_{L})}^{2}}{2 - 2 {(V_{T} / V_{L})}^{2}}

(3)

E = \frac{V_{L}^{2} ρ (1 + μ) (1 - 2 μ)}{(1 - μ)}

(4)

where μ is the Poisson's ratio, E is the modulus of elasticity (GPa), V_L is the longitudinal wave velocity (m s⁻¹) and V_T is the shear wave velocity (m s⁻¹), ρ is the density (kg m⁻³) of the sample.

Vickers microhardness testing was performed on a Y₃NbO₇ sample (φ10 mm × 5 mm) with a Digital Microhardness Equipment (Matsuzawa, MMT-X) based on ASTM E-384 with a load of 1 kg for 15 s (HV1).

To assess the behavior of the as-deposited coating under thermal shock, a thermal cycling test was conducted. The samples were held isothermally at 1000°C for 30 min and then rapidly cooled to room temperature within 5 min using compressed air. This procedure was repeated for 48 identical cycles to monitor any significant spallation or delamination of the coating.

To simulate sulfur and vanadium contaminants, a Na₂SO₄+ V₂O₅ salt blend was used for the hot corrosion experiment. The eutectic composition of the binary phase diagram (32 wt% Na₂SO₄+ 68 wt% V₂O₅),²⁰ was selected to ensure the early formation (at 630°C) of molten salt during exposure. A salt load of 1 wt% of the pellet was applied evenly over the surface. These designed conditions—including a highly acidic species, a thicker corrosive layer, and the earlier formation of corrosive melt—were intended to expedite the hot corrosion process and yield measurable results in a shorter period.²¹ The ratios of the salt mixture and the corrosion period used in the study do not precisely replicate the real conditions in TBC applications. Instead, they are meant to simulate high-temperature sulfate-vanadate corrosion conditions to evaluate the response of Y₃NbO₇.

Studies have shown that when exposed to elevated temperatures (∼1100°C), the evaporation of SO₃ from the Na₂SO₄+ V₂O₅ melt (Equation 5) decreases the activity of V₂O₅, leading to a reduction in the hot corrosion attack of the mixture.^22–24

{Na}_{2} {SO}_{4} (l) + V_{2} O_{5} (l) \to 2 NaV O_{3} (l) + {SO}_{3} (g)

(5)

k_{1} = \frac{α_{N a V O_{3}} {^{2} P}_{S O_{3}}}{α_{N a_{2} S O_{4}} α_{V_{2} O_{5}}}

(6)

Equation 6 indicates that the activity of V₂O₅ (

α_{V_{2} O_{5}}

) is affected by both the partial pressure of SO₃ (

P_{S O_{3}}

) and the square of the activity of sodium metavanadate (

α_{N a V O_{3}}

). Thus, it is essential to select an isothermal exposure temperature that minimizes SO₃ evaporation in the experiment. Consequently, a temperature of 900°C was selected for the experiment to ensure high V₂O₅ activity while preventing significant SO₃ evaporation.

The hot corrosion experiments were performed on Y₃NbO₇ pellets (φ10 mm × 5 mm) using salt melts of Na₂SO₄+ V₂O₅ mixture, pure V₂O₅ and pure Na₂SO₄ separately. The salt mixtures were prepared by thoroughly mixing the salts in deionized water. Each mixture was applied evenly (salt load of 30 mg/cm²) on the top surface of the samples using a glass rod, with careful attention to avoid from contacting the edges. The samples were then heated on a hot plate to evaporate any excess water and obtain adequate adhesion of the salt. The heating program for the corrosion experiment is illustrated in Figure 1. The heating and cooling rate for the heat treatment program were 5°C/min and 2.5°C/min respectively. The samples were subjected to five cycles of 6-h corrosion at 900°C, resulting in an aggregate heating period of 30 h. XRD analysis was executed to determine the corrosion products, and SEM-EDS was used to study the microstructural changes in the samples.

Figure 1.

Heating program of hot corrosion test.

Results and discussion

The XRD pattern of the synthesized Y₃NbO₇ powder was very similar to the standard ICSD pattern 98-009-9881 (Figure 2), which corresponds to a defect fluorite structure with a cubic unit cell (Fm-3m). The powder had a crystallographic density of 5.25 g/cm³, and the lattice parameters, as determined by Rietveld refinement, were a = b = c = 5.254 Å. The lattice parameter value closely matches with that of ICSD 98-009-9881 pattern, which is 5.2450 Å. In crystal structure of Y₃NbO₇, Y and Nb atoms are randomly distributed in cationic positions in a 3:1 ratio, with 1/8 of the oxygen sites unoccupied to ensure chemical balance. While this structure resembles that of conventional yttria-stabilized zirconia, Y₃NbO₇ features anion deficiency in 1/8 of the sites, compared to 1/50 in 7YSZ. This difference could significantly affect important TBC properties, such as thermal conductivity.

Figure 2.

XRD patterns: ICSD 98-009-9881, prepared powder, pellet and coating.

The XRD results from both the sintered pellet and coating sample (Figure 2) closely resembled that of the Y₃NbO₇ powder. There weren’t any discernible peaks in the pellet pattern other than those corresponding Y₃NbO₇ which suggested that Y₃NbO₇ did not undergo any kind of decomposition during sintering. However, the coating pattern exhibited broader peaks, likely due to the rapid cooling during the plasma spraying. The coating pattern included extra peaks not seen in the powder pattern, which have been identified as Y₂O₃ (ICSD 98-010-4015) and NbO (ICSD 98-006-9409). This indicates that a minor decomposition of Y₃NbO₇ occurred during deposition. This is likely due to the significant difference in vapor pressures of Y₂O₃ (10⁻⁶ to 10⁻⁵ Pa at 2000°C) and Nb₂O₅ (10⁻⁵ to 10⁻² Pa at 2000°C) and dissociation of Nb₂O₅ at elevated temperatures.^25,26 The accompanied variation in Y₃NbO₇ stoichiometry may affect the performance of the TBC, for which the composition of feedstock needs to be monitored to obtain stochiometric Y₃NbO₇ coating, particularly during coating deposition using methods such as EB-PVD.

The SEM image of the prepared Y₃NbO₇ powder, shown in Figure 3(a), reveals an irregular morphology with fairly consistent size distribution, likely resulting from ball milling. Figure 3(b) shows the EDS analysis of the prepared powder, indicating a composition that closely aligns with the theoretical stoichiometric values. The A : B cationic ratio obtained from EDS (3.13:1) aligns with the formula of the Y₃NbO₇ compound. Additionally, the DLS (Nano Plus, Micromeritics) method determined the average particle size of the Y₃NbO₇ sample to be 7.2 μm, with the particle size distribution provided in Table 2.

Figure 3.

(a) SEM micrograph and (b) EDS results of Y₃NbO₇ powder.

Table 2.

Particle size distribution of prepared Y₃NbO₇.

D (10%)	D (50%)	D (90%)
0.4809 (µm)	4.5183 (µm)	17.9681 (µm)

Figure 4(a) and (b) display surface micrographs of the pellet and the coating. The SEM image of the pellet shows grains with a consistent size distribution, averaging around 2 µm. The pellet also features several open micropores. On the other hand, the coating image reveals both open pores and cracks, which are typical for TBCs. The coating consists of fully molten and partially molten splats, along with some unmolten particles. The cross-sectional SEM of the coated sample (Figure 4(c)) depicts good adhesion of the bond coat (thickness ∼120μm) with base material. However, the Y₃NbO₇ topcoat is comparatively thin with a rough surface (thickness ∼60μm), since the powder lacks the free-flowing, spherical morphology that promotes easier melting and deposition.

Figure 4.

Surface micrograph of the (a) Y₃NbO₇ pellet (b) Y₃NbO₇ coated sample; (c) cross-sectional micrograph of the coating.

Thermal diffusivity of Y₃NbO₇ was measured from 30°C to 1200°C at 200°C intervals which decreased from 0.5089 mm² s⁻¹ at 31.7°C to 0.4593 mm² s⁻¹ at 1200.5°C (Figure 5(a)). Heat capacities at these temperatures were calculated using the Neumann-Kopp rule (Figure 5(b)).²⁷ The density of the sample used for the measurement was 5.03 g cm⁻³ (95.8% of the theoretical value). Thermal conductivity was obtained using Equation 1, and the value for the 100% dense material was obtained using Equation 2 by correcting the error due to porosity. The values are plotted in Figure 5(a), demonstrating an initial decrease from 2.28 W m⁻¹ K⁻¹ (30°C) to 1.77 W m⁻¹ K⁻¹ (600°C) further increase to 2.68 W m⁻¹ K⁻¹ (1200°C)

Figure 5.

(a) Thermal diffusivity and conductivity (b) calculated specific heat of Y₃NbO₇.²⁷

The thermal conductivity values of Y₃NbO₇ were comparable to those of YSZ (2.2–2.9 W m⁻¹ K⁻¹). Unlike common binary oxides used in TBCs, which typically show a decrease in thermal conductivity with temperature (following a 1/T dependence), the thermal conductivity of Y₃NbO₇ showed slight increase with temperature after initial decrease from room temperature value. This unusual behavior is amorphous-like and characteristic of lanthanide niobates. The lower thermal conductivity of lanthanide niobates, despite being simple binary oxides, is often attributed to substantial chemical variability caused by disparity in cationic charge and bond length.⁴ The difference in atomic radii and ionic charges between Y³⁺ (1.019 Å) and Nb⁵⁺ (0.74 Å) leads to the disparity in bond lengths and strengths of Y-O and Nb-O, aligning with this interpretation.

An important factor influencing the performance of TBCs at high temperatures is the coefficient of thermal expansion (CTE) of the topcoat. Ideally, the CTE of the topcoat should align with that of the metallic superalloy to reduce stresses caused by differing thermal expansion. The density of the sample used for the measurement was 5.055 g cm⁻³ (96.2% of theoretical density). Y₃NbO₇ exhibited a linear thermal expansion, as illustrated in Figure 6(a). The coefficient of thermal expansion (TEC) for Y₃NbO₇ rose from 2.2 × 10⁻⁶ K⁻¹ at room temperature (40°C) to 10.84 × 10⁻⁶ K⁻¹ at 670°C, before slightly decreasing to 10.25 × 10⁻⁶ K⁻¹ at 1000°C, as depicted in Figure 6(b). These values for TEC were comparable to that of YSZ, remaining fairly constant throughout most of the measured temperature range (ranging from 10 × 10⁻⁶ K⁻¹ at 310°C to 10.25 × 10⁻⁶ K⁻¹ at 1000°C). Y₃NbO₇ possesses a defect fluorite structure with unoccupied anionic positions that allows for greater atomic displacement under thermal stress than more rigid structures. Furthermore, the bonding in Y₃NbO₇, characterized by both ionic and covalent components, promotes increased atomic mobility, leading to a higher thermal expansion coefficient.

Figure 6.

(a) Linear expansion and (b) CTE of Y₃NbO₇.

The elastic modulus of Y₃NbO₇ was obtained by finding the ultrasonic wave velocities using the pulse-echo method. The ultrasonic wave velocities, along with the calculated elastic modulus (Equations 3 and 4), are presented in Table 3.

Table 3.

Ultrasonic velocities and elastic constant of Y₃NbO₇.

Longitudinal wave velocity, V_L (m s⁻¹)	Shear wave velocity, V_T (m s⁻¹)	Young's modulus, E (GPa)
5817 ± 90	3934 ± 60	169.9 ± 3

The average microhardness of the Y₃NbO₇, measured using Vickers micro indentation (ASTM E-384), was found to be 7.79 GPa (794 HV1).

The mechanical properties of Y₃NbO₇ were found to be less than that of conventional YSZ.²⁸ Table 4 provides a comparison of the mechanical properties prepared Y₃NbO₇ with that of YSZ. The lower hardness value is suggestive of reduced coating performance in erosive conditions. However, it is promising while compared to the alternative materials presently considered for TBC application.^29–31 The lower elastic modulus indicates improved strain compliance, implying enhanced performance during thermal cycling. This indicates that a TBC topcoat made of Y₃NbO₇ may exhibit better performance under thermal cycling conditions.

Table 4.

Mechanical properties of Y₃NbO₇ compared with YSZ.

Material	Modulus of Elasticity, E, (GPa)	Hardness (GPa)
YSZ^32,33	210	13
Y₃NbO₇	169.9 ± 3	7.79 ± 0.3

Chemical compatibility with the TGO layer is crucial when evaluating the suitability of a material as a topcoat. To maintain the coating's integrity at high temperatures, it is essential that it remains stable and non-reactive with the thermally grown oxide (TGO). The TGO, mainly consisting of aluminium oxide (Al₂O₃), develops from the oxidation of aluminium in the bond coat as oxygen diffuses through the topcoat.^34–37 The XRD pattern of Y₃NbO₇ - Al₂O₃ mixture after heating (Figure 7) revealed no additional peaks except those corresponding to Y₃NbO₇ (ICSD 98-009-9881) and Al₂O₃ (ICSD 98-005-3283). Thus, it can be inferred that the Y₃NbO₇ topcoat is expected to remain unreactive with TGO during high-temperature exposure.

Figure 7.

XRD pattern of Y₃NbO₇ – Al₂O₃ mixture after 6 h heating at 1200°C.

To assess the durability of this TBC system, a thermal cycling test was performed on the as-deposited Y₃NbO₇ coated sample for 48 cycles. The coating remained intact without any notable indication of cracking or spallation on the surface. In the thermal cycling experiment, the failure of the TBC often begins at the beveled edge, where the coating is subjected to tension. As illustrated in Figure 8(a) and (b), there was no evidence of significant spallation at any of the edges, and no color change was observed in the sample during exposure. A comparison of the EDS results from the sample before and after thermal cycling (Table 5) indicates that the stoichiometry of the coating remained consistent throughout the experiment. This finding suggests that the Y₃NbO₇ coating can maintain its integrity over multiple heating and cooling cycles compared to other alternative TBC materials currently being investigated.³⁸ To examine the coating microstructure before and after thermal cycling, SEM-EDS analysis was conducted. The coating appears slightly denser (Figure 9(a)) compared to its state before thermal cycling (Figure 4(b)), and the porosity level has significantly decreased, as evidenced by the micrographs in Figures 4(b) and 9(a). This may be attributed to the sintering of the coating during isothermal exposure, leading to more uniform grain sizes after thermal cycling. However, the cross-sectional micrographs depicting the splat morphology before (Figure 9(c)) and after thermal cycling (Figure 9(b)) show a greater number of microcracks. One fundamental reason for these cracks is the thermal stresses that arise during thermal cycling.

Figure 8.

Y₃NbO₇ coated sample (a) before and (b) after thermal cycling test.

Figure 9.

SEM image of Y₃NbO₇ coating: (a) surface and (b) cross-sectional splat morphology after thermal cycling; (c) cross-sectional splat morphology before thermal cycling.

Table 5.

EDS results of Y₃NbO₇ coating before and after thermal cycling.

Element	Before thermal cycling		After thermal cycling
Element	Weight %	Atomic %	Weight %	Atomic %
O K	24.08	64.04	25.74	64.62
Y L	57.84	27.68	56.53	27.41
Nb L	18.08	8.28	17.73	7.97
Total	100	100	100	100

The surfaces of the samples exposed to in a Na₂SO₄+ V₂O₅ corrosive mixture showed notable color changes, turning predominantly grey. These color shifts indicate corrosive reactions. SEM micrographs of the pellet surfaces after the experiment are shown in Figure 10(a) and (b). The micrographs display altered particle morphologies, suggesting significant corrosive activity. The ceramic pellet prior to exposure (Figure 4(a)) exhibited a relatively dense microstructure with a uniform grain size distribution. However, high-temperature exposure to corrosive salt led to significant disintegration of the sample surface, as shown in Figure 10. The corroded samples revealed the presence of rod/plate-like particles with varying thicknesses, as seen in Figure 10(b).

Figure 10.

SEM micrograph of Na₂SO₄+ V₂O₅ exposed Y₃NbO₇ at (a) 10KX (b) 50KX.

SEM-EDS analysis indicated that the particles primarily consist of Y, V, and O elements (Table 6). XRD results (Figure 11(c)) confirmed that the rod/plate-like particles are mainly YVO₄ (ICSD 98-003-5198). The XRD patterns also showed minor peaks corresponding to Y₂V₂O₇ (ICSD 98-011-3329), YNbO₄ (ICSD 98-007-2062), and NaNbO₃ (ICSD 98-007-1460). The Y:V atomic ratio in some EDS spectra (spectrum 2) matched that of Y₂V₂O₇, supporting its presence. However, the presence of YNbO₄ and NaNbO₃ particles could not be confirmed by EDS, likely due to their low concentrations as suggested by the XRD data. Additionally, the detection of small amounts of Na and Nb within YVO₄ and Y₂V₂O₇ particles in the EDS results implies that the hot corrosion process likely involved dissolution and reprecipitation. Thus, the main corrosion product of Y₃NbO₇ in a 32 wt% Na₂SO₄+ 68 wt% V₂O₅ mixture at 900°C is inferred to be YVO₄.

Figure 11.

XRD pattern of the Y₃NbO₇ pellets after 30 h of high-temperature exposure to (a) Na₂SO₄ (b) V₂O₅ (c) Na₂SO₄+ V₂O₅ salt mixture.

Table 6.

EDS spectra at different rod/plate-like particles on the Y₃NbO₇ sample exposed to Na₂SO₄+ V₂O₅ mixture.

Element	Spectrum 1		Spectrum 2		Spectrum 3		Spectrum 4
Element	Weight %	Atomic %	Weight %	Atomic %	Weight %	Atomic %	Weight %	Atomic %
O K	24.49	58	21.11	52.6	33.65	68.05	22.21	53.57
Na K	2.18	3.59	2.71	4.7	2.4	3.38	3.86	6.48
V K	22.53	16.76	25.75	20.15	19.69	12.51	24.47	18.54
Y L	50.8	21.65	46.92	21.04	41.61	15.14	45.72	19.85
Nb L	0	0	3.5	1.5	2.65	0.92	3.75	1.56
Total	100	100	100	100	100	100	100	100

Figures 11(a) and (b) illustrate the XRD results of Y₃NbO₇ samples after 30 h exposure to pure Na₂SO₄ at 900°C and pure V₂O₅ salt melts at 900°C, respectively. For the Na₂SO₄ corrosion test, the XRD pattern primarily displayed sharp peaks corresponding to Y₃NbO₇ (ICSD 98-009-9881) and minor peaks of Na₂SO₄ (ICSD 98-001-2491). This indicates that Y₃NbO₇ remained practically unreactive in pure Na₂SO₄ environment. In contrast, the XRD pattern for the pellet corroded in V₂O₅ exhibited sharp peaks of YVO₄ (ICSD 98-000-7818) and Y₃NbO₇, along with minor peaks of YNbO₄ (ICSD 98-007-2062) and NbO₂ (ICSD 98-001-0013).

Table 7 presents the EDS spectra from the V₂O₅ hot corrosion sample. The EDS results confirm the XRD findings, showing the presence of YVO₄ (spectrum 1), Y₃NbO₇ (spectrum 2), and YNbO₄ (spectrum 3). The Y:V and Y:Nb molar ratios observed in the respective particles match closely with theoretical values. However, NbO₂ particles were not detected by EDS, likely due to its low concentration. Additionally, the presence of minor quantities of Nb in YVO₄ and V in both Y₃NbO₇ and YNbO₄ suggests that the hot corrosion process involved dissolution and reprecipitation mechanisms.

Table 7.

EDS spectra from Y₃NbO₇ sample exposed to pure V₂O₅.

Element	Spectrum 1		Spectrum 2		Spectrum 3
Element	Weight %	Atomic %	Weight %	Atomic %	Weight %	Atomic %
O K	36.96	72.88	15.91	51.02	17.93	55.06
V K	18.39	11.39	2.27	2.28	0.78	0.75
Y L	37.19	13.2	60.3	34.8	50.69	28.01
Nb L	7.47	2.54	21.52	11.89	30.6	16.18
Total	100	100	100	100	100	100

The corrosive reactions during the tests can be understood through Lewis acid-base theory. Y₃NbO₇ is a complex oxide composed of Y₂O₃ and Nb₂O₅ in 3:1 ratio. Under corrosive conditions, it is expected to decompose into component oxides because of the acidic leaching of Y³⁺ ions from Y₃NbO₇ (Equation 7).

2 Y_{3} {NbO}_{7} (s) \to {Nb}_{2} O_{5} (s) + 3 Y_{2} O_{3} (s)

(7)

As the temperature is increased, V₂O₅ melts at 690°C, followed by Na₂SO₄ (884°C). Nonetheless, the corrosive species is formulated as a eutectic blend of Na₂SO₄ and V₂O₅, with the eutectic compound NaVO₃ forming earlier (630°C). This allows NaVO₃ to react with Y₃NbO₇ earlier in the process. Additionally, NaVO₃ has been reported to act as a good solvent for oxides,³⁹ enhancing their atomic mobility.⁴⁰

Given the lack of studies on the Y₂O₃-NaVO₃ system compared to the extensively reported Y₂O₃ -V₂O₅ system, it is plausible to forecast the reactions between Y₃NbO₇ and NaVO₃ based on the Y₂O₃-V₂O₅ phase diagram.⁴¹ In the Y₂O₃ -V₂O₅ system, two major compounds, YVO₄ and Y₂V₂O₇, are formed depending on the Y₂O₃:V₂O₅ ratio. Therefore, it can be concluded that the reaction between Y₂O₃ and NaVO₃ starts at 630°C (Equation 8), leading to the formation of YVO₄, as indicated by the XRD results (Figure 11).

Y_{2} O_{3} (s) + 2 NaV O_{3} (l) \to 2 YV O_{4} (s) + {Na}_{2} O (s)

(8)

The freed Nb₂O₅ (Equation 7) can further react with NaVO₃. However, since Nb₂O₅ has lower basicity, this reaction happens only when Y₂O₃ is unavailable. This explains the lack of evident peaks for any Nb-V-O compounds. However, Nb₂O₅ could react with Y₃NbO₇, leading to the following reaction:

Y_{3} {NbO}_{7} (s) + {Nb}_{2} O_{5} (s) \to 3 YNb O_{4} (s)

(9)

By integrating these findings, the corrosive action of the salt mixture with 32 wt% Na₂SO₄ and 68 wt% V₂O₅ on Y₃NbO₇ can be summarized as follows:

Y_{3} {NbO}_{7} (s) + 2 NaV O_{3} (l) \to 2 YV O_{4} (s) + {YNbO}_{4} (s) + {Na}_{2} O (s)

(10)

Some of the YNbO₄ that formed could further react with the excess NaVO₃ during extended exposure, leading to the following reaction:

4 YNb O_{4} (s) + 4 NaV O_{3} (l) \to 2 Y_{2} V_{2} O_{7} (s) + 4 NaNb O_{3} (s) + O_{2} (g)

(11)

Equations 10 and 11 combinedly explains the formation of corrosion products on Y₃NbO₇ samples and presence of YVO₄, Y₂V₂O₇ and NaNbO₃ in the XRD pattern (Figure 11(c)). Also, the Na₂O released can undergo further reactions with V₂O₅ to yield NaVO₃.³⁹ This recurring reaction raises the amount of NaVO₃ solvent, promoting the corrosion phenomena.

A comparison of the hot corrosion tests conducted with pure Na₂SO₄ and pure V₂O₅ versus the 32 wt% Na₂SO₄+ 68 wt% V₂O₅ mixture shows notable differences in the activity of V₂O₅. Figure 11(a) illustrates that exposure to Na₂SO₄ produced prominent peaks for Y₃NbO₇ and minor peaks for Na₂SO₄, suggesting that Na₂SO₄ by itself did not cause degradation of Y₃NbO₇. In contrast, the V₂O₅-exposed sample displayed sharp peaks for both YVO₄ and Y₃NbO₇ (Figure 11(b)), suggesting that the acidic action of V₂O₅ was less severe compared to the 32 wt% Na₂SO₄+ 68 wt% V₂O₅ mixture. The corrosion pattern for the mixture sample, which showed no peaks for Y₃NbO₇ (Figure 11(c)), indicates complete disintegration of Y₃NbO₇ at the exposed surface. This demonstrates that the combined effect of V₂O₅ and Na₂SO₄ was much more corrosive than V₂O₅ alone. Additionally, the presence of NaVO₃ resulted in enhanced atomic mobility, which accelerated the corrosion phenomena.

The acidic leaching Y₂O₃ by pure V₂O₅ could be described as follows:

2 Y_{3} {NbO}_{7} (s) + 3 V_{2} O_{5} (l) \to 6 YV O_{4} (s) + {Nb}_{2} O_{5} (s)

(12)

The liberated Nb₂O₅ may react separately with Y₃NbO₇ over extended exposure, leading to the formation of YNbO₄.

Y_{3} {NbO}_{7} (s) + {Nb}_{2} O_{5} (s) \to 3 YNb O_{4} (s)

(13)

Combinedly the complete corrosive reactions of pure V₂O₅ on Y₃NbO₇ is given by:

Y_{3} {NbO}_{7} (s) + V_{2} O_{5} (l) \to 2 YV O_{4} (s) + {YNbO}_{4} (s)

(14)

None of the XRD results obtained from the hot corrosion samples of Y₃NbO₇, displayed any signs of a Nb-V-O compound. This supports the notion that the corrosion proceeded via the Lewis acid-base principle, where V₂O₅ preferentially reacted with the more basic Y₂O₃ rather than Nb₂O₅. The acidic reaction of V₂O₅ with Y³⁺ ions from the mixed oxide and further chemical reactions resulted in the formation of Y-V-O compounds identified in the experiments.

Conclusions

Single-phase Y₃NbO₇ powder was prepared through a solid-state method, utilizing Y₂O₃ and Nb₂O₅ as precursors. Pellet samples, prepared through pressureless sintering were used for hot corrosion experiments and thermal property measurements. A plasma sprayed TBC architecture was fabricated using the prepared powder to illustrate the viability of Y₃NbO₇ for coating deposition.

The thermal properties of Y₃NbO₇ indicate its potential for use as a TBC material. Its thermal conductivity is lower than that of YSZ, and the thermal expansion coefficient values are appropriate for TBC applications. Y₃NbO₇ was unreactive with Al₂O₃, indicating good TGO compatibility during service. The thermal cycling test demonstrated that the as-deposited Y₃NbO₇ retained its original stoichiometry and coating integrity for a minimum of 48 cycles.

Out of the measured mechanical properties lower elastic modulus was found to be beneficial for TBC applications, suggestive of improved thermal cycling performance. However, the lower hardness compared to YSZ indicates potentially poor coating response under erosive conditions.

Y₃NbO₇ experienced aggressive corrosion in 32 wt% Na₂SO₄+ 68 wt% V₂O₅ melt, with Y³⁺ ions leaching out and reacting with NaVO₃ to form YVO₄. In comparison, corrosion in pure V₂O₅ was less severe, and pure Na₂SO₄ did not significantly destabilize Y₃NbO₇. Combining Na₂SO₄ with V₂O₅ increased the activity of V₂O₅ by producing NaVO₃ and enhanced corrosion process.

None of the corrosion tests provided evidence of significant amounts of Nb-V-O compounds, confirming that the corrosion process followed the Lewis acid-base principle.

Footnotes

Declaration of conflicting interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Guo

Chen

, et al. Progress on high-temperature protective coatings for aero-engines. United Kingdom: Springer Nature Singapore. 2023. Epub ahead of print 2023. DOI: https://doi.org/10.1007/s44251-023-00005-6.

Chen

Song

, et al. Potential thermal barrier coating materials: rE3NbO7 (RE=La, Nd, Sm, Eu, Gd, Dy) ceramics. J Am Ceram Soc 2018; 101: 4503–4508.

Chen

Guo

Zhu

, et al. Features of crystal structures and thermo-mechanical properties of weberites RE3NbO7 (RE=La, Nd, Sm, Eu, Gd) ceramics. J Am Ceram Soc 2021; 104: 404–412.

Yang

Qian

Pan

, et al. Diffused lattice vibration and ultralow thermal conductivity in the binary Ln–Nb–O oxide system. Adv Mater 2019; 31: 1–9.

Chen

, et al. Thermo-mechanical properties of fluorite Yb3TaO7 and Yb3NbO7 ceramics with glass-like thermal conductivity. J Alloys Compd 2019; 788: 1231–1239.

Yang

Pan

Han

, et al. Mechanical properties, oxygen barrier property, and chemical stability of RE3NbO7 for thermal barrier coating. J Am Ceram Soc 2020; 103: 2302–2308.

Zhao

Pan

Wan

, et al. Defect engineering in development of low thermal conductivity materials: a review. J Eur Ceram Soc 2017; 37: 1–13.

Tian

Lin

Zheng

, et al. Defect-mediated multiple-enhancement of phonon scattering and decrement of thermal conductivity in (YxYb1-x)2SiO5 solid solution. Acta Mater 2018; 144: 292–304.

Liu

Chen

Zhao

, et al. Entropy as a gene-like performance indicator promoting thermoelectric materials. Adv Mater 2017; 29: 1–7.

10.

Kumar

Pandey

. Tailoring mechanical, microstructural and toughening characteristics of plasma-sprayed graphene-reinforced samarium niobate coatings for extreme environments. Proc Inst Mech Eng Part E J Process Mech Eng. Epub ahead of print 2024. DOI: https://doi.org/10.1177/09544089241279724.

11.

Yang

, et al. Phase transformation inhibition and improved thermo-mechanical properties of rare earth niobates for thermal barrier coatings. Ceram Int 2024; 50: 29799–29805.

12.

Guené-Girard

Jubera

Viers

, et al. Liquid phase synthesis and sintering of Y3NbO7. Ceram Int 2020; 46: 26361–26367.

13.

Song

Pyeon

Ryu

, et al. Effect of Ta5+ doping on the thermal physical properties of defective fluorite Y3NbO7 ceramics. J Am Ceram Soc 2022; 105: 1358–1366.

14.

Chen

Zou

, et al. A promising radiation thermal protection coating based on lamellar porous Ca-Cr co-doped Y3NbO7 ceramic. Adv Funct Mater 2023; 33: 2305650.

15.

Guené-Girard

Jubera

Weill

, et al. Thermodynamic and experimental approach of the effect of Si on the sintering of Y3NbO7. J Eur Ceram Soc 2021; 41: 5316–5323.

16.

Falak

Khan

Grzegorz

. A review on the degradation of coatings under hot corrosion process. Contemp Probl Power Eng Environ Prot 2022; 1: 27–34.

17.

She

Guo

Tan

, et al. Na2SO4 + V2O5 corrosion behavior of BaNd2Ti3O10 for thermal barrier coating applications. Coatings 2020; 10: 901–922.

18.

Schlichting

Padture

Klemens

. Thermal conductivity of dense and porous yttria-stabilized zirconia. J Mater Sci 2001; 36: 3003–3010.

19.

Wang

. The elastic constants measurement of metal alloy by using ultrasonic nondestructive method at different temperature. Math Probl Eng 2016. Epub ahead of print 2016. DOI: https://doi.org/10.1155/2016/6762076.

20.

Yugeswaran

Kobayashi

Ananthapadmanabhan

. Hot corrosion behaviors of gas tunnel type plasma sprayed La 2Zr 2O 7 thermal barrier coatings. J Eur Ceram Soc 2012; 32: 823–834.

21.

De la Roche

Alvarado-Orozco

Toro

. Hot corrosion mechanism of yttria-stabilized zirconia powder in the presence of molten Na2SO4 + V2O5 salts. Rare Met 2021; 40: 1307–1316.

22.

Reddy

Gandhi

. Molten salt attack on t’ yttria-stabilised zirconia by dissolution and precipitation. J Eur Ceram Soc 2013; 33: 1867–1874.

23.

Rapp

. Chemistry and electrochemistry of hot corrosion of metals. Mater Sci Eng 1987; 87: 319–327.

24.

Nagelberg

. Destabilization of Yttria-stabilized zirconia induced by molten sodium vanadate-sodium sulfate melts. J Electrochem Soc 1985; 132: 2502–2507.

25.

Schulz

Saruhan

Fritscher

, et al. Review on advanced EB-PVD ceramic topcoats for TBC applications. Appl Ceram Technol 2004; 1: 302–315.

26.

. Thermodynamic study on vapourization of niobium oxides from slag melts . 2010.

27.

Leitner

Chuchvalec

Sedmidubský

, et al. Estimation of heat capacities of solid mixed oxides. Thermochim Acta 2002; 395: 27–46.

28.

Xie

Guo

Gong

, et al. Lanthanum-titanium-aluminum oxide: a novel thermal barrier coating material for applications at 1300°C. J Eur Ceram Soc 2011; 31: 1677–1683.

29.

Gao

Zhang

, et al. Influence of mechanical properties on thermal shock resistance of TBCs. Surf Eng 2021; 37: 572–580.

30.

Zeng

Huang

Zhang

, et al. Effects on the crystallite growth behavior of LaPO4 nanopowders and its mechanical properties. Ceram Int 2021; 47: 33223–33231.

31.

Wei

Zhu

, et al. Influence of size disorder parameter on the thermophysical properties of rare-earth-zirconate medium-entropy ceramics. J Am Ceram Soc 2023; 106: 2037–2048.

32.

Vassen

Cao

Tietz

, et al. Zirconates as new materials for thermal barrier coatings. J Am Ceram Soc 2000; 83: 2023–2028.

33.

Gupta

Lange

Bechtold

. Effect of stress-induced phase transformation on the properties of polycrystalline zirconia containing metastable tetragonal phase. J Mater Sci 1978; 13: 1464–1470.

34.

Leyens

Schulz

Pint

, et al. Influence of electron beam physical vapor deposited thermal barrier coating microstructure on thermal barrier coating system performance under cyclic oxidation conditions. Surf Coatings Technol 1999; 120–121: 68–76.

35.

Levi

. Emerging materials and processes for thermal barrier systems. Curr Opin Solid State Mater Sci 2004; 8: 77–91.

36.

Pint

Wright

Lee

, et al. Substrate and bond coat compositions: factors affecting alumina scale adhesion. Mater Sci Eng A, Struct Mater Prop Microstruct Process 1998; 245: 201–211.

37.

Sun

Tong

Feng

, et al. Microstructural evolution and deformation features in gas turbine blades operated in-service. J Alloys Compd 2015; 618: 728–733.

38.

Yuan

Sun

Wang

, et al. Srceo3 as a novel thermal barrier coating candidate for high–temperature applications. J Alloys Compd 2018; 740: 519–528.

39.

Krishna Prasad

Rao

Reddy

, et al. Hot corrosion of SrTiO3 perovskite in Na2SO4 + 50wt.% V2O5 and Na2SO4 + 10wt.% NaCl environments at 900°C. Int J Corros 2018. Epub ahead of print 2018. DOI: https://doi.org/10.1155/2018/4763085.

40.

Habibi

Wang

Liang

, et al. An investigation on hot corrosion behavior of YSZ-Ta2O5 in Na2SO4+V2O5 salt at 1100°C. Corros Sci 2013; 75: 409–414.

41.

Levin

. The system Y2O3-V2O5. J Am Ceram Soc 1967; 50: 381–382.