Enhanced stability of hafnia based coatings in a hot gas environment

Abstract

Yttria stabilized hafnia (YSH) coatings have been fabricated by magnetron sputtering, a physical vapor deposition (PVD) method. Coatings of a mixed composition of hafnia (HfO₂) and zirconia (ZrO₂) (YSHZ) stabilized by yttria (Y₂O₃) were also fabricated in order to compare and contrast the resulting properties. The composition of the material was varied by varying the ratio of HfO₂ and ZrO₂ (4 : 1, 2 : 1, 1 : 1, 1 : 2 and 1 : 4) while keeping the Y₂O₃ stabilizer content constant at 7.5 mol%. Thermal and chemical stability along with the durability of the YSH and YSHZ coatings were evaluated by exposing the coatings to hot gases in a combustor rig. A nanoindentation technique was used to evaluate the mechanical properties. A diamond tipped, sharp nanoindenter with known geometry and mechanical properties was forced into the sample while both the force and indentation depth were recorded. Hardness (H) and Young's modulus were estimated for the YSH and YSHZ coatings. Atomic force microscopy was utilized to provide imagery of the indentation area on the sample surface. Residual stress analysis was performed using X-ray diffraction (XRD). The results from nano-indentation indicate that the YSH sample possesses values of hardness and Young's modulus as high as 18 GPa and 220 GPa, respectively. The residual stress estimated using XRD indicates very high compressive stress within the coatings. The stability and durability tests demonstrate the enhanced stability of YSH coatings in a hot gas environment created by burning natural gas with oxygen.

---

1. Introduction

Thermal barrier coatings (TBCs) are being utilized as an advanced technology to protect metallic surfaces from high temperature exposure for extended durations. TBCs are extensively used in gas turbines and aero-engine parts. As TBCs protect the gas turbine components from high temperature exposure, this allows a further increase in engine temperature which subsequently increases the efficiency of the gas turbine power plant. The development of novel TBCs which can withstand higher temperatures can minimize fuel consumption by maximizing gas turbine efficiency, which will subsequently help protect the environment through the reduction of carbon emissions.¹ Additionally, different fuel compositions ranging from natural gas to a broad range of syngases with high hydrogen contents are being tested for next-generation gas turbine power plants.¹ Therefore, the TBCs for next-generation gas turbine systems should show reliability and durability in diverse chemical, thermal and mechanical environments.

Numerous works have been directed toward the development of TBCs for advanced turbine technology.^2–13 The current choice of material for TBC applications is yttria stabilized zirconia (YSZ). However, the maximum temperature tolerance of YSZ is up to 1200 °C for long term operation.⁴ YSZ undergoes phase transformation at temperatures higher than 1200 °C. This phase transformation is associated with a change in volume, leading to the initiation of cracks on the surface and at the interface between the TBC and substrate.⁴ The ultimate result is the failure of the TBC system. Thus the application of YSZ is limited to temperatures below 1200 °C.^5,7 Therefore, the design and development of new and better TBCs is desired to meet the challenging goals of syngas/hydrogen turbine technology. Manufacturing new materials or engineering coatings with novel chemistries must be taken into account to develop TBCs with unique structural, thermal, thermo-chemical, and mechanical properties. Motivated by these research needs in the topical area of TBCs for turbine applications in power generation systems, the present work is focused on strain-tolerant high-temperature ceramics based on hafnia. Recently, improved temperature tolerance up to 1400 °C has been reported for yttria stabilized hafnia.³ It has been shown that a HfO₂ and ZrO₂ co-system can result in significant reductions in thermal conductivity.⁷ Zr⁴⁺ replacement with Hf⁴⁺ also showed a reduction of thermal conductivity.^4,14 While these observations are promising, no information is available on the performance of hafnia-based coatings under hot gas exposure. Motivated by these facts and challenges, the present work is directed towards evaluating the structure and thermo-mechanical properties of nanostructured hafnia based TBCs under a hot gas environment. Interestingly, the results presented and discussed in this paper demonstrate the enhanced durability of the hafnia-based TBCs under hot gas exposure testing in a laboratory combustor.

2. Experimental

Yttria stabilized hafnia (YSH) target (Plasmaterials Inc.) was employed to fabricate the coatings. The composition of yttria (Y₂O₃) in the YSH target was maintained at 7.5 mol%. 403-stainless steel (2.5 cm diameter and 3 mm thickness) was used as a substrate. Nanostructured YSH coatings were grown using sputter-deposition. The YSH target was placed on a 5 cm sputter gun, which was correspondingly placed at a distance of 8 cm from the substrate. A high vacuum was produced in order to create the plasma. When a pressure of 1 × 10⁻⁶ torr was achieved, a sputtering power of 30 W was initially applied to the target while introducing high purity argon (Ar) into the chamber, causing plasma ignition. Once ignited, the power was increased to 100 W to deposit the coatings. The flow of Ar was controlled using an MKS mass flow meter. Before each deposition, the YSH-target was pre-sputtered for 10 minutes using Ar with the shutter above the gun closed. A variety of operating parameters were used to optimize the deposition conditions. Starting from room temperature, various substrate temperatures up to 500 °C were maintained for all substrates to optimize the operating temperature for good uniform growth of the TBC. The substrates were heated by halogen lamps and the desired temperature was controlled by an Athena X25 controller. Various deposition durations were maintained to grow TBCs with different thicknesses. Various gas (Ar) velocities were used to optimize the plasma flow for forming a uniform thin film on the substrates. The deposition was carried out at a pressure of ∼4 × 10⁻³ torr.

Structural characterization of the fabricated TBCs was performed by using X-ray diffraction (XRD). A Bruker D8 Advance X-ray diffractometer was employed for crystal structure and phase analysis of the coatings. All the measurements were made ex situ as a function of the coatings' fabrication conditions. XRD patterns were recorded using CuKα radiation (λ = 1.54056 Å) at room temperature (RT). Scanning electron microscopy (SEM) was used to study the surface and interface morphology. Structural analysis of the surface and interface was performed using a Hitachi S-4800 scanning electron microscope. Secondary electron imaging was performed to probe the surface morphology and grain distribution characteristics. Energy dispersive X-ray spectroscopy (EDS) coupled with SEM was used to analyze the composition of the TBCs. In the EDS system, the elemental composition was investigated by analyzing the generated X-rays from the samples by using an incident electron beam of 20 keV. Grain size was calculated using the well-known Scherrer equation. It was also measured using the software provided with the scanning electron microscope.

Residual stress analysis was performed using an XRD method. The sin² ψ technique was followed to calculate the residual stress.^15,16 The variation of d-spacing was measured using the diffraction of Cu-Kα radiation by (133) planes. X-ray diffraction was carried out on the sample at various tilting angles.

The performance of the fabricated coating in a real gas turbine environment was evaluated by exposing the coating to hot gases in a combustor. A swirl stabilized laboratory scale combustor rig was used to perform the experiments. The schematic of the experimental setup for the hot gas exposure test is shown in Fig. 1. The YSH samples were exposed to hot gas produced by the combustion of methane. The detailed conditions of the hot gas experiment are given in Table 1.

Fig. 1 The schematic of the experimental setup for the hot gas exposure test.

Table 1 Details of the hot gas exposure experiments

Gases used for combustion	CH4 and air
Flow rate of CH4	20 L min^−1
Flow rate of air	198 L min^−1
Anticipated composition of the hot gases	CO, CO2, N2, O2 and unburned hydrocarbons
Temperature of the hot gas	800–1100 °C
Exposure time	15 min to 16 h
Angle of impingement	90°

---

The air/fuel ratio was maintained at an optimum value in order to keep the flame stabilized without any interruption from flashback or blowout. The sample was inserted from the top of the combustor rig using a laboratory made sample holder as shown in Fig. 1. The sample holder was made of stainless steel so that it could withstand the hostile environment inside the combustion chamber. The sample holder was designed in such a way that the impingement angle can be changed. Samples of different sizes can be accommodated by adjusting the threaded rod coupler. An R-type thermocouple was used to measure the temperature near samples inside the combustion chamber. After every specific interval, the samples were taken out and the crystal structure, surface morphology and residual stress were examined. Finally, the properties of the exposed samples were compared to those of the as grown sample to evaluate the performance of the coatings in a real hot gas environment.

A nanoindentation technique using a Hysitron TI 750 Tribo Indenter was used to evaluate the mechanical properties. Hardness (H), modulus of elasticity (E_s), reduced modulus of elasticity (E_r), stiffness (S) and strain rate sensitivity were estimated using appropriate mathematical expressions for YSH and YSHZ coatings. A diamond Berkovich tipped sharp nanoindenter with a radius of curvature of 396 nm was forced into the sample. The diamond tip, whose geometry and mechanical properties are known, was pressed into the sample while both the force and indentation depth were recorded continuously. The load placed on the indenter tip was increased as the tip penetrated further into the sample until it reached the defined maximum depth of 85 nm. At the maximum depth the load was held constant for a set time and then removed. The area of the indentation in the sample was measured using the known geometry of the indentation tip. Atomic force microscopy was utilized to image the indentation area on the sample surface. A record of these values was plotted to create a load–displacement curve and the mechanical properties were calculated using the corresponding relationships.

3. Results and discussion

The XRD patterns of the YSH coatings before and after exposure to hot gases are shown in Fig. 2. The XRD patterns of the YSH coatings exhibit peaks corresponding to the cubic structure of HfO₂. The peak at 2θ ∼30.28° corresponds to the diffraction from (111) planes. It can be seen from the XRD patterns that the (111) peak is sharper and more intense, indicating the preferred orientation of the film along the (111) plane. The monoclinic and tetragonal phases consist of several shorter Hf–O bond lengths (2.0–2.1 Å) compared to cubic HfO₂ (2.37 Å).¹⁷ As a consequence, the strain energy because of the size mismatch becomes significant in the monoclinic and tetragonal phases. On the other hand, the distortion is less in the cubic phase with oversized Y³⁺ doping, which makes cubic HfO₂ stable even at low temperature. Furthermore, structural relaxation plays another role in the stabilization of the cubic phase of HfO₂ by Y₂O₃ dopants.

Fig. 2 XRD patterns of the YSH coatings before and after exposure to hot gases.

The effect of hot gas exposure on the YSH coatings can be understood as follows. It is important to recognize that no significant structural change can be seen from the XRD patterns. However, the strongest peaks show a slight positive shift in the peak position towards higher 2θ values with increasing exposure time to hot gases. The peak shift is indicated by drawing a line through the (111) peaks (Fig. 2). This peak shift indicates a decrease in inter-planar spacing with increasing exposure time. Little change in orientation is evident after 9 h of exposure to 1100 °C. However, the material does not show any change from the overall cubic crystal structure. Since no significant change in crystallographic structure is evident after exposure to hot gases, it is expected that there would be no change in volume due to phase change. This would ensure there is no mismatch in thermal expansion between the topcoat and the thermally grown oxide (TGO)/bond coat. The ultimate outcome of this performance is the enhanced stability of YSH coatings in a hot gas environment.

The surface morphology of the YSH coatings before and after exposure to hot gas is compared in Fig. 3 and 4. No significant change other than slightly more compaction of the grains is visible from the SEM images when the samples are subjected to hot gases for 9 h at 1100 °C (Fig. 4a). The morphology after exposure for a duration of 16 h at 1100 °C is slightly different. It is evident from Fig. 4b that the grains are more defined and densely packed together compared to those shown in Fig. 3a, 3b and 4a.

Fig. 3 SEM images of the YSH samples (a) before and (b) after exposure to hot gases for 1 h at 800 °C, in order of increasing magnification from top to bottom.

Fig. 4 SEM images of the YSH samples after exposure to hot gases for an extended time: (a) 9 h at 1100 °C and (b) 16 h at 1100 °C. Shown in order of increasing magnification from top to bottom.

The thermal-induced changes in the chemical composition were probed using EDS measurements before and after exposure to hot gases (Fig. 5a and b). The enhanced chemical stability of the coatings was found through composition analysis by EDS. No compositional change was observed in the YSH coatings exposed to hot gases. EDS indicates that the composition of the coatings is the same before and after hot gas exposure. Thermo-chemical analysis based on the EDS measurement indicates the coatings' chemical stability at high temperatures and in hot gases.

Fig. 5 EDS spectra showing the composition of the YSH sample before (a) and after exposure (b) to hot gases.

Analysis of the residual stress in the coatings was performed after hot gas exposure for a specific duration. In the sin² ψ method for residual stress measurement, the sample is tilted within a certain range of tilting angles (ψ) and then scanned with an X-ray beam. The inter-planar spacing (d_φψ) for the corresponding tilting angle is computed using a higher angle peak (133 plane in this case). A linear relation is obtained between the inter-planar spacing and sin² ψ values. Then the residual stress is calculated using the following well-known expression.^15,16

where,

d₀ = unstressed lattice spacing

d_φψ = stressed lattice spacing

σ_φ = stress component along the direction φ defined in the plane of coating

Ψ = tilt angle

E = Young's modulus

ν = Poisson ratio

σ₁₁ and σ₂₂ are in-plane principal stress components.

A representative plot for (d − d₀)/d₀ is shown in Fig. 6, which shows the d-spacing as a linear function of sin² ψ. The residual stress of the YSH samples before and after exposure to hot gases is shown in Fig. 7.

Fig. 6 d-spacing as a function of sin² ψ for the YSH samples.

Fig. 7 Residual stress in the YSH coatings before and after exposure to hot gases.

It is evident that the as grown YSH coating holds high residual stress within the coating. It is interesting to note that the coating shows a significant relaxation of residual stress through exposure to hot gas for a short period of time, such as 2 hours. However, the coating gains the stress back after exposure to hot gas at a higher temperature for a longer duration. The as grown coating, the coating under hot gas for 9 h and the coating under hot gas for 16 h exhibit similar compressive stress, which is as high as 2 GPa.

A stress versus displacement plot for nano-indentation on the YSH coating is shown in Fig. 8. After gathering the nano-indentation data, the mechanical properties were calculated using the corresponding relations as described by Oliver and Pharr.¹⁸ Fig. 9 shows the hardness (H), modulus of elasticity (E_s), reduced modulus of elasticity (E_r) and stiffness (S) for YSH and various compositions of YSHZ coatings. It is evident (Fig. 9) that YSH shows the highest mechanical strength compared to all other compositions. Optical images showed the damage on the sample surface after nano-indentation (not shown). One of the most important microstructural parameters that influences the mechanical properties is grain size.¹⁹ There is a critical grain size at which the mechanical properties, e.g., Young’s modulus, start decreasing drastically.¹⁹ This critical grain size in pure hafnia is ∼2 μm. The grain size of the YSH and YSHZ coatings employed for the evaluation of the mechanical properties are in the range of ∼5–20 nm, which is quite low compared to the critical grain size where the mechanical properties degrade in pure hafnia. The lack of drastic changes in the mechanical properties with the variation of composition of YSHZ may actually be due to the fact that the grain size is extremely low, so is unlikely to induce changes in the mechanical characteristics. Interpretation of the mechanical properties could be based on the residual strain energy stemming from the anisotropy of thermal expansion. If the grains are big enough, micro-cracks are initiated which are responsible for the reduction in mechanical strength.¹⁹ This is not the case in this work as no phase change was evident in the operating range of temperature which could be responsible for significant thermal expansion anisotropy. Since there is no transformation toughening in either the YSH or YSHZ coatings, the hardness, Young's modulus and stiffness values are a reflection of the combination of the individual material properties of hafnia, zirconia and yttria. These properties in fully or partially stabilized hafnia or hafnia–zirconia mixed compositions are strongly influenced by microstructure, such as porosity and grain size. As the porosity was not investigated in this work, the quantitative influence of porosity on the mechanical properties of YSH and YSHZ is not discussed here. The high values of hardness and Young's modulus are, perhaps, due to the dense columnar structure which is clearly seen in both YSH and YSHZ. As was previously reported, the hardness value of fully stabilized hafnia decreases with the content of alloying oxides.^19,20 This observation is also visible here when zirconia is incorporated in the YSH structure; YSHZ exhibits lower hardness values than YSH. Although the hardness and Young's modulus are very high and close to YSH, there is a slight randomness among all the compositions of YSHZ. The randomness is due to the experimental uncertainty resulting from the columnar structure of both YSH and YSHZ. The mechanical properties were measured using nano-indentation on the surface of the samples. It is really hard to identify from the top of the surface exactly where the columns are located. If the nano-indenter is placed on a column it will show a higher hardness value compared to that obtained if the indenter is placed in between the columns.

Fig. 8 Stress versus displacement plot for nano-indentation of the YSH coating.

Fig. 9 Mechanical properties of YSH and YSHZ coatings measured using nano-indentation.

4. Conclusion

Yttria stabilized hafnia (YSH) and YSHZ coatings crystallize in a cubic structure with characteristic columnar growth and dense structures as revealed by the interface analysis. These coatings exhibit structural stability without any noticeable changes in their crystal structure and morphology after exposure to hot gases. YSH coatings show a significant relaxation of residual stress in the earlier stage of exposure to hot gases, however, the high residual stress comes back after exposure to hot gases at a higher temperature for a longer duration. YSH and YSHZ coatings show very high mechanical strength. The results demonstrate the enhanced stability of the YSH and YSZH coatings in hot gases.

Acknowledgements

This material is based upon work supported by the Department of Energy under Award Number DE-FE0000765.

References

1. W. P. Parks, E. E. Hoffman, W. Y. Lee and I. G. Wright, J. Therm. Spray Technol., 1997, 6(2), 187–192.
2. N. P. Padture, M. Gell and E. H. Jordan, Science, 2002, 296, 280–284.
3. K. Matsumoto, Y. Itoh and T. Kameda, Sci. Technol. Adv. Mater., 2003, 4, 153–158.
4. D. R. Clarke and S. R. Phillpot, Mater. Today, 2005, 8, 22–29.
5. R. A. Miller, J. L. Smialek and R. G. Garlick, in Science and Technology of Zirconia, ed. A. H. Heuer and L. W. Hobbs, Am. Ceram. Soc. Inc., Columbus, Ohio, 1981, vol. 3, pp. 241–251.
6. P. K. Schelling and S. R. Phillpot, J. Am. Ceram. Soc., 2001, 84(12), 2997–3007.
7. D. Zhu and R. A. Miller. Proceedings of the 26th Annual International Conference on Advanced Ceramics, Mater. Struct B, Florida, 2002.
8. S. Raghavan, H. Wang, R. B. Dinwiddie, W. D. Porter and M. J. Mayo, Scr. Mater., 1998, 39(8), 1119–1125.
9. H. Yang, G. Bai, L. J. Thompson and J. A. Eastman, Acta Mater., 2002, 50, 2309–2317.
10. W. A. Nelson and R. M. Orenstein, J. Therm. Spray Technol., 1997, 6(2), 176–180.
11. L. L. Shaw, D. Goberman, R. Ren, M. Gell, S. Jiang, Y. Wang, T. D. Xiao and P. R. Strutt, Surf. Coat. Technol., 2000, 130(1), 1–8.
12. M. Gell, E. H Jordan, Y. H. Sohn, D. Goberman, L. Shaw and T. D. Xiao, Surf. Coat. Technol., 2001, 146–147, 48–54.
13. R. S. Lima, A. Kucuk and C. C. Berndt, Surf. Coat. Technol., 2001, 135, 166–172.
14. K. S. Murphy and N. Shores, US Pat. 0118873 A1, 2003.
15. V. Teixeiraa, M. Andritschky, W. Fischer, H. P. Buchkremer and D. StoÈver, J. Mater. Process. Technol., 1999, 92–93, 209–216.
16. C. H. Ma, J. H. Huang and H. Chen, Thin Solid Films, 2002, 418, 73–78.
17. C. K. Lee, E. Cho, H. S. Lee, C. S. Hwang and S. Han, Phys. Rev. B: Condens. Matter, 2008, 78, 0121021–0121024.
18. W. C. Oliver and G. M. Pharr, J. Mater. Res., 2004, 19(1), 3–20.
19. J. Wang, H. P. Li and R. Stevens, J. Mater. Sci., 1992, 27, 5397–5430.
20. C. D. Wirkus and M. F. Berard, J. Mater. Sci., 1982, 17, 109–114.

---