Advances in diffusion barrier coatings for high-temperature applications

doi:10.21203/rs.3.rs-3240347/v1

Download PDF

Research Article

Advances in diffusion barrier coatings for high-temperature applications

https://doi.org/10.21203/rs.3.rs-3240347/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 02 Nov, 2023

Read the published version in High Temperature Corrosion of Materials →

You are reading this latest preprint version

Diffusion barrier coating (DBC) systems on heat resistant alloys consist of a multi-layer structure: an outer Al-reservoir layer and an inner diffusion barrier layer. The outer Al-reservoir layer forms a protective Al₂O₃ scale and the inner layer acts as a barrier layer against alloy interdiffusion. Three kinds of diffusion barrier layers (DBL) were developed, containing Re on Ni-based super alloys, W on stainless steels, and Cr on Ni-Cr alloys. It was found that DBC systems have excellent mechanical properties (creep and fatigue), improving alloy substrate performance, and enhance the anti-exfoliation properties of YSZ in thermal barrier coatings (TBC) in addition to providing excellent oxidation resistance. At temperatures higher than 1,300℃, however, the DBC design based on kinetics (diffusion) is insufficient to form and maintain a protective Al₂O₃ scale. In this case a self-maintaining coating (SMC) system designed on the basis of thermodynamics (phase stability) is required. The SMC system formed on Nb-Hf (C-103) alloy consists of a multi-layer structure: an outer Re (Al, Si)_1.8 and inner NbSi₂ layers, plus a transient Nb₅Si₃ layer formed during oxidation. At temperatures higher than 1,300℃ the Al₂O₃ can be formed by changing the Al/Si ratio in the Re (Al, Si)_1.8 in which Si was supplied from the decomposition reaction of NbSi₂ to Si + Nb₅Si₃ during selective oxidation of Al.

It is proposed that coatings alloy should be designed for not only high temperature oxidation, but also alloy substrate mechanical properties and anti-exfoliation of oxide scales, based on both kinetic principles (DBC system) and thermodynamics (SMC system).

Figure 1 shows the micro-structure of three types of DBL formed by a combination of electro-plating or slurry coating, heat treatment, and Al diffusion treatment. The Re (Cr-Ni) based barrier layer can be precisely controlled in thickness, and is formed onto a Ni-based single crystal superalloy, and the W (Mo, WC)-based barrier layer can be formed relatively thickly, making it suitable for general-purpose substrates such as stainless steels. The Cr (Mo) barrier layer is a α-Cr phase that is naturally formed when Al diffusion treatment of Ni-Cr based alloys containing about 20 at% or more Cr.

Conventional coatings have an Al-containing alloy layer formed on the surface of the alloy substrates by making use of electroplating, thermal spraying, and EBPVD or the like. At high temperatures, as shown in Fig. 2, an aluminum depletion zone is formed under the Al₂O₃ film by selective oxidation of aluminum. Further, interdiffusion occurs between the coating and the alloy substrate, and the secondary reaction zone (SRZ) develops. The formation of the Al-depletion zone and the SRZ causes degradation of the oxidation resistance of the coating film. Furthermore, in a case of the Ni-based single crystal superalloy the formation of the SRZ causes reduction in the strength of the base material as will be described later.

As shown in Fig. 3, the diffusion barrier coating (DBC) system proposed by the author has a structure in which an alloy layer is inserted between the alloy substrate and the Al-containing layer in Fig. 2. This alloy layer inserted has the function of suppressing the mutual diffusion of elements between the base material and the Al-containing layer. From the viewpoint of diffusion suppression, refractory metals (W, Re, Mo, Nb, Ta, Cr) are candidate elements.

3 − 1 SRZ

Thermal barrier coating (TBC; bond coat of MCrAlY and top-coat of YSZ) was formed on Ni-based single crystal superalloy by using of thermal spraying (HVOF and Plasma), and then an oxidation test was performed at 1,050℃ for 1,000hr in air.

As shown in Fig. 4 − 1, in a case of conventional TBC poly-crystallization phenomena can be seen over about 80µm depth of the substrate on the bond coat side. At high temperatures and for long durations, interdiffusion between the alloy substrate and the bond coat MCrAlY proceeds significantly, resulting in poly-crystallization of the single crystal structure.

A Re-based DBC with about 20µm in thickness was formed on a single crystal superalloy, and then the same TBC as those shown in Fig. 4 − 1 was over-coated. The TBC/DBC coated superalloy was oxidized under the same conditions as those in Fig. 4 − 1. It was found that as shown in Fig. 4 − 2, an interdiffusion layer of about 10µm thickness is formed between the alloy substrate and the bond coat MCrAlY, while the SRZ is significantly suppressed.

3 − 2 Creep and Fatigue

The creep and fatigue characteristics of Ni-based single crystal superalloys coated with DBC/TBC or conventional TBC were investigated. When the stress of the creep test is plotted by the Larson Mirror Parameter (LMP), as shown in Fig. 5 − 1 two points (red arrows 920°C/450MPa and 1,040°C/140MPa) were on a straight line of the data such as the alloy substrate and decrease in creep strength with a DBC/TBC was scarcely observed. Fatigue tests at 920°C, (strain rate: 0.4%/s, full strain 0.7%, tri-angle wave, exposure at 1050°C for 100hr) showed that the number of rupture cycles of DBC/MCrAlY increased by about twice as high as that of the MCrAlY coating, as shown in Fig. 5 − 2. It is concluded that the DBC system is effective in improving fatigue properties.

3–3 Cyclic Oxidation

Cyclic oxidation tests (1093°C; 10hr⇔RT in air, DSS) of a Ni-based single crystal superalloy with DBC/TBC were performed. For comparison, the results of the thermal cycling test of the superalloy with conventional TBC are also given.

As shown in Fig. 6, it was revealed that the YSZ of the top coat layer of TBC was completely peeled off at 126 cycles, while the peeling rate of the YSZ layer of DBC/TBC was about 20% after 300 cycles. It can be seen that for Ni-based single crystal superalloys, the exfoliation resistance of the YSZ layer of TBC is greatly improved by DBC.

Figure 7 shows creep curves obtained for a Ni-based alloy (Hastelloy X) with or without a Re-based DBC/β-NiAl coating. Creep conditions are as follows; temperature 970 ° C; stresses 40, 27.5 and 22.5MPa. It was found at 27.5MPa that the creep rupture time of the DBC coated HX alloy is about twice as long as those of the bare alloy.

In the creep test at a stress of 22.5 MPa, the test was interrupted at 3% strain (creep time 180 hours), as shown by an arrow in Fig. 7. As shown in Fig. 8, the Re-based DBC layer is uniformly creep-deformed over the entire surface of the test specimen, and cracks and the like are not observed.

From these results, it is presumed that the σ phase (Re-Cr-Ni) constituting the DBC system can be creep-deformed due to the fact that the alloy elements and their concentrations of the DBC system differ from the σ phase of the single-crystal superalloy, which is well-known to be the harmful, mechanical phase. More detailed research will be conducted in the future.

Figure 9 shows the creep curves of Fe-based alloy (SUS310S) with or without a Re-based DBC / β-NiAl coating. A creep test was carried out under the following conditions: temperature 900℃; stress 22.5MPa in air. From this result, creep deformation of SUS310S proceeds rapidly from the initial stage due to grain boundary fracture, whereas in DBC / SUS310S, grain boundary fracture is suppressed and the creep deformation rate is small. While, little difference was observed in the fracture time.

A DBC/TBC coating was formed on a disk-shaped HX alloy, and a thermal cyclic oxidation test (1,100°C; 45min ⇔RT; 15min in air) was performed, and the surface morphologies during cooling process was observed with a camera after pre-determined cycles. For comparison, test pieces with conventional TBC were also tested under the same cyclic oxidation conditions. For the conventional TBC, as shown in Fig. 10 − 1, peeling of the YSZ layer proceeded from periphery into the substrate, and cracks formed within the YSZ layer, and then the entire YSZ peeled off after about 600 cycles.

Figure 10 − 1 Conventional TBC

Meanwhile, as shown in Fig. 10 − 2, the interface detachment of the YSZ layer of the DBC / TBC proceeds slowly from the periphery, but cracks in the YSZ layer and the like are not observed. It was completely exfoliated after about 1,850 cycles. The exfoliated YSZ layer retains the shape of the disk.

7.1 Self-maintaining coating (SMC) system

Refractory metals and their alloys have the highest specific strength at elevated temperatures, while they are used to be suffered from severe degradation due to high temperature oxidation. As an example, a Nb-Hf based alloy (C103) has the NbSi₂ coating to form protective SiO₂. Although the NbSi₂ coating has much attention, there are several subjects to be solved. One of them is a formation of non-protective Nb₅Si₃ on the NbSi₂ layer due to selective oxidation of Si.

The present author proposed a concept of the self-maintaining coating (SMC) system, based on the phase diagrams for the Re-Nb-Si and Re-Nb-Si-Al systems. The SMC system with multi-layer structures as Re (Al, Si)_1.8/NbSi₂ on Nb-Hf alloy was prepared by making use of electroplating, slurry, and heat-treatment followed by Si and/or Al pack cementations.

The SMC coating structure on Nb-Hf alloy with the distribution of each element are shown in Fig. 11. It was determined experimentally that the Re (Al, Si)_1.8 in the outermost layer varies the Re / Al / Si concentration ratio.

After oxidation at 1,400℃ for 50hrs in air, the cross-sectional structure and the concentration of each element are shown in Fig. 12. From this result, a protective Al₂O₃ film is formed, and Re (Al, Si)_1.8 maintains the same crystal structure, while the Si concentration increases up to ReSi_1.8. At the same time, NbSi₂ changes to Nb₅Si₃.

Oxidation of Re(Al,Si)_1.8 to form 2Al+(3/2) O₂ = Al₂O₃

NbSi₂ decomposes into Nb₅Si₃ + Si

Si will enter into Re(Al,Si)_1.8 to form ReSi_1.8

7.2 Composition Paths

When the composition paths are plotted on the phase diagram in Fig. 13, at the stage of forming Re (Al, Si)_1.8 the concentration of each element changes along the solid line on the upper side (Si side) of the phase diagram, and in the process of high-temperature oxidation they move along dotted lines on the lower side (Re and Nb).

As shown in section 6, it was clarified that the exfoliation resistance of the YSZ (top layer) of the TBC is greatly improved by DBC / TBC. There are also differences in the peeling behavior (delamination and cracking) of the YSZ layer. The exfoliation behavior of the YSZ layer of DBC/TBC and TBC may be controlled by several factors. One of them is the so-called thermal grown oxide (TGO) formed between the YSZ layer and the bond layer (MCrAlY or β-NiAl). The other one seems to be as follows. The author notes the change in the coefficient of linear expansion due to inserting DBC between the alloy and TBC layer, as shown in Fig. 14. In other words, during the cooling process of thermal cycling, a large tensile stress is generated in the periphery YSZ layer in the substrate/TBC, which promotes cracking of YSZ, while the tensile stress generated in the YSZ layer can be greatly reduced by inserting a DBC layer with low linear expansion coefficients.

The similar arrangement of thermal expansion coefficients has been adopted for the alloy/Si₃N₄ ceramic joints, where in a case of the turbo charger rotors an intermediate (Ni/W/Ni) layer was inserted between alloy and ceramics.

Although the above-mentioned explanation is still at the estimation stage, in the design of the oxidation-resistant coating, it is desirable to establish a suitable construction method that takes into account not only oxidation resistance but also structural stability, stress, and mechanical characteristics of the alloy base material.

This report, focusing on the experimental results of the DBC systems developed, has shown that appropriately designed DBC systems improve not only oxidation resistance but also the mechanical properties (creep and fatigue) of alloy substrates and the peel resistance of YSZ. Furthermore, at extreme high temperatures the thermodynamic stability of the alloy plus coating assemblages should be taken into account in addition to kinetic factors.

In much of the earlier work, mechanical properties and high-temperature corrosion/protection have been investigated independently. Although elements such as Al, Cr, Si, etc are desirable for the oxidation resistance they provide, their adverse effects on alloy mechanical properties limits their use. The well-known solution is to concentrate these elements in coating, but the coatings must be formulated to be both kinetically and thermodynamically stable during service.

When considering changes in the service environment of materials, it is desirable to develop methods that solve simultaneously the problems of mechanical properties and high-temperature corrosion and protection.

　Acknowledgements 　

The author expressed his deepest appreciation for the proposed DBC system, which was the result of a large number of collaborators. Theoretical understanding and explanation on DBC and SMC and their combination are beyond the scope of the author, and then I look forward to the efforts of those who are interested in this report. It is a great pleasure to contribute to the 80th anniversary special issue of my esteemed Professor David Young.

T. Narita, M. Shoji, Y. Hisamatsu, D. Yoshida, M. Fukumoto, and S. Hayashi, Rhenium coating as a diffusion barrier on a nickel-based superalloy in high temperature oxidation, HIGH TEMPERATURE CORROSION AND PROTECTION 2000 Science Reviews, pp.351~357
Y. Matsumura, M. Fukumoto, S. Hayashi, A. Kasama, I. Iwanaga, R. Tanaka, and T. Narita, Oxidation Behavior of Re-Base Diffusion Barrier /β-NiAl Coating on Nb-5Mo-15W at High Temperature, Oxidation of Metals, Vol.61(2004), Nos.1/2, pp. 105~125.
T. Izumi, T. Yoshioka, and T. Narita: Effect of Re-W-Cr-Ni Diffusion Barrier on High-Temperature Oxidation of Pt-Modified β-NiAl Coated Superalloy, J. Japan Inst. Metals, Vol.72, No.9(2008), pp.728~732.
S. Saito, K. Kurokawa, S. Hayashi, T. Takashima, and T. Narita: Phase Equilibria and Tie-Lined Compositions in a Ternary Ni-Al-Re System at 1423K, J. Japan Inst. Metals, Vol.71 No.9(2007), pp.793~800. Experimental Determination of Isothermal Phase Diagram at 1423K in the Ternary Re-Cr-Ni Systems; J. Japan Inst. Metals, Vol.72 No.2(2008), pp.132~137.
Y. Wang, S. Ohnuki, S. Hayashi, T. Yoshioka, M. Hara, and T. Narita; Thermal Stability of Rhenium-Based Diffusion Barrier Coating Layer on a Ni-Based Superalloy, Materials Transaction JIM Vol.48 No.2(2007), pp.127~132.
T. Narita, S. Hayashi, L. Fengquin and K. Z. Toshin; The Tole of Bond Coat in Advanced Thermal Barrier Coatings; Materials Science Forum Vol.502, (2005), 99~104.
I. R. Sohn and T. Narita; Influence of Pre-oxidation on the Corrosion and Mechanical Strength of Fe-25Cr and Fe-25Cr-20Ni Alloy at 973, Oxidation of Metals, Vol.65(2006), No3/4, 181~194.
M. Fukumoto, S. Kudo, M. Hara and T. Narita; Cyclic Oxidation Resistance of Electroplating Ni-Al-La Film, J. Japan Inst. Metals, Vo.70, No.8(2006), pp.607~610.
M. Fukumoto, T. Saruta, M. Hara and T. Narita; Formation of Ni-Aluminide Surface layer Containing Zr by Molten salt Electrodeposition and Cyclic oxidation Resistance, J. Japan Inst. Metals, Vo.71, No.1(2007), pp.001~006.
Japanese patent: #5295474, #3904197, #4271399, #4323816

No competing interests reported.

Download PDF

Journal Publication

published 02 Nov, 2023

Read the published version in High Temperature Corrosion of Materials →

Editorial decision: Major revision
28 Aug, 2023
Reviews received at journal
28 Aug, 2023
Reviewers agreed at journal
17 Aug, 2023
Reviewers invited by journal
17 Aug, 2023
Editor assigned by journal
17 Aug, 2023
Submission checks completed at journal
17 Aug, 2023
First submitted to journal
06 Aug, 2023

You are reading this latest preprint version

Advances in diffusion barrier coatings for high-temperature applications

Status:

Journal Publication

Version 1

Abstract

Figures

1. Formation of DBC system

2. Conventional Coatings and DBC Systems.

3 Ni-based Single Crystal Superalloys

3 − 1 SRZ

3 − 2 Creep and Fatigue

3–3 Cyclic Oxidation

4. Ni-based Alloy (Hastelloy-X)

5. Fe-based alloy (SUS310S)

6. Peel Resistance of YSZ layer

7 Coatings on Refractory Metals

7.1 Self-maintaining coating (SMC) system

7.2 Composition Paths

8. Effect of DBC on Peel Resistance of YSZ layer

9. Conclusion

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1