CORE Metadata, citation and similar papers at core.ac.

u
Provided by University of Queensland eSpace

5th Australasian Congress on Applied Mechanics, ACAM 2007

10-12 December 2007, Brisbane, Australia

Experimental Studies on Slurry Based Thermal Barrier Coatings

Phuc Nguyen, Steven Harding and Sook-Ying Ho

School of Mechanical Engineering, University of Adelaide, Adelaide SA 5005

Abstract: Thermal barrier coatings were introduced to reduce operating temperatures and thermal
stresses in structures and machine components across a wide range of industries and applications.
The focus of the present study is to develop a new, relatively simple and low cost technique of
manufacturing thermal barrier coatings based upon the slurry spray method. The challenge of this
work is to make this technique comparable, in terms of the quality of the coating, to existing
manufacturing methods, which are often expensive and inapplicable to coat large or curved surfaces.
This paper describes the deve loped technique and selected results of thermo-mechanical testing of
fabricated coatings including functionally graded coatings.

Keywords: spray drying, thermal barrier coatings, thermal protection, thermal spray, slurry spray,
sintering, zirconia.

1 Introduction
Over the last 35 years Thermal barrier coatings (TBCs) have been developed for various applications
and structures working under super high temperatures and temperature gradients. A thermal barrier
coating represents a thin layer of a material typically 5 µm to 2 mm with high insulating properties,
such as ceramics, that is bonded to a substrate, which is usually metal, to protect the metal structure
during temperature excursions. The application of TBCs can significantly increase the operating
temperatures up to 1400-1500 0C, increase efficiency and improve the durability of the structural
components. TBCs are currently being developed or considered for hypersonic applications,
automotive engines, nuclear fusion reactors and heavy-duty utilities (i.e. diesel trucks, electric power
generators, etc [1]).

The durability requirements of TBCs for these applications are increasing rapidly [2]. In many of the
TBC applications, stresses due to the difference in thermal expansion of the TBC and substrate can
have a detrimental effect on the fatigue and hence service life of the high temperature component.
One important way to reduce the adverse effects of thermal stresses is to use Functionally Graded
(FG) thermal coatings, where thermal and mechanical properties vary gradually through the thickness
of the coating. FG TBCs can significantly reduce the thermal mismatch and, therefore, largely reduce
thermal stresses as well as the possibility of fractures caused by these thermal stresses and, as a
result, drastically improve the service life.

In the past decade, many methods have been developed and described in the literature for fabricating
FG TBCs [3]. The basic methods of applying a TBC to a metal substrate involve either mechanical or
chemical deposition. However of the two alternatives the mechanical deposition is more effective in
terms of cost, flexibility and manufacturing time. In particular the Electron Beam Physical Vapour
Deposition, Powder Flame Spraying, Plasma Thermal Spray and Cold Gas Dynamic Spray Coating
techniques are currently most popular in the industry. However, these techniques are still expensive
and impractical to protect large-surface areas. The high manufacturing cost is the motivation behind
the current study, which will be focused on the development of relatively simple and low-cost Slurry
Spray and Sintering methods for fabricating thermal barrier coatings.
First, the new developed technique will be briefly outlined. It will be demonstrated later in this work that
this technique is also suitable for producing graded coating. From numerous tests the optimal
manufacturing parameters to produce spallation free coatings will be described. Furthermore, the
preliminary experimental results on thermal cycling, investigation of microstructure and effect of
various manufacturing parameters on the quality and durability of the coating will be discussed. The
paper will be concluded with a summary of major outcomes of the current study and suggestions on
future work.
2 Slurry spray technique
The Slurry Spray technique for manufacturing thermal barrier coatings utilises traditional wet powder
spraying methods to deposit sinterable coating materials onto target substrates to be sintered to
produce a functional coating. The process involves suspending the coating material within a fluid to
form a slurry mixture that can be applied to a surface using common spray guns. Successive layers
are then sprayed onto the substrate and dried using varying slurry compositions. The optimal
thickness of the layers to deter surface cracking during the drying process is around 0.25 mm and the
drying time is approximately an hour, depending on ambient conditions. After the desirable number of
layers of the TBC is created the multi-layered coating is pressed in a compression chamber to create
a solid form before being sintered with an acetylene torch or furnace, as show in Figure 1 below. The
applied pressure varies depending on the number of coating layers, typically between 10 and 40 MPa.

(1) Mixing
(2) Multilayered Spraying

Oxy Torch

Pressure

Oven

(4) Pressing Stamping (3) Drying

(5) Sintering
Figure 1: Stages in producing thermal barrier coating

TBCs produced using the Slurry Spray technique have many advantages compared to other
fabrication methods in terms of increased simplicity, reduced fabrication costs, potential application to
a range of coating materials and the ability to coat complex surface geometries. The Slurry Spray
technique utilises unsophisticated equipment and techniques in simple progressive stages to create
the final coating. The process has the potential to be adopted in an automated environment but can
also be utilised for manual applications.
The Slurry Spray technique can be applied to a range of thermal barrier coating materials, provided
that the materials can be prepared as a powder capable of being sintered at temperatures below the
melting point of the substrate. As well as being an established thermal barrier coating material,
zirconia powders are readily suited for the Slurry Spray technique. Zirconia powders are required to be
either partially or fully stabilised using a metal oxide addition to prevent cracking within the thermal
barrier coating after sintering due to shrinkage associated with structural phase changes during
cooling [4]. The technique also addresses limitations associated with other thermal barrier coating
methods in terms of the complexity of the surface geometries that can be coated and also the total
coating area. The spray and sintering streams can be easily manipulated and controlled to coat
surfaces out of direct line of sight, such as internal bend sections of pipes, and is limited only to areas
that can be accessed by the spray and sintering streams. The process does not require isolated or
evacuated environments and can be used to coat large areas in a more cost effective manner in
comparison to other coating methods.

3 Fabricating Parameters

3.1 Mixing procedure

Initial slurry compositions are created using the mixing procedure used to create sintered beads of
Yttria Partially Stabilised Zirconia (YSZ) by Roy et el [5]. Initially a volume of distilled water solvent is
agitated using a magnetic mixing element in a beaker placed on a magnetic mixing pad. Tetra Sodium
Pyrophosphate dispersant is slowly added to the agitated solvent and allowed to mix [6]. Hydro
soluble polyvinyl alcohol binder crystals are added to the slurry mixture to complete the slurry [7].
Typically the slurry mixture composes of a ceramic and metal powder (Yttria partially stabilised
Zirconia and Nickel powder), binder (Tetra Sodium Pyrophosphate) and a dispersant, with the
remaining percentage being distilled water. The composition of the weight percentage of all the
components for the graded coatings can be seen in Table 1.
Table 1 – Composition of Powders
Powder (%) Binder (%) Dispersant (%)
Graded Composition % ZrO 2 + % Ni = 45% 4% 0.4 %

3.2 Spraying and Compression

To spray the TBC slurry, a gravity fed paint spray gun is connected to an air compressor and used to
spray the slurry mixture fed into the spray gun from a top mounted paint hopper. The gun is directed to
the surface to deposit the slurry particles in a uniform layer.
The spraying technique used to spray the TBC onto the substrate involves using several quick passes
of the spray jet over the substrate surface to form an even coating. The spray nozzle is kept a
moderate distance from the surface to allow a large cone shaped spray stream and prevent rapid
deposition of the slurry on the surface. The pressure of the spray stream is varied from 4-6 bars,
depending on the desired coating thickness. Thinner layer thicknesses have been shown to provide a
more stable TBC. A diagram of the graded composition can be seen in Figure 2.

100 % ZrO 2
66 % ZrO2 / 33 % Ni
33 % ZrO2 / 66 % Ni

Substrate: Inconel

Figure 2 – 3 Example of Layer Gra ded Composition

Pressure is applied to the dried multi-coating layers to promote contact between powder particles,
which in turn decreases the time and applied temperatures required for sintering as shown by Dahl et
el [8]. The lower applied temperatures and reduced times of sintering are very beneficial for the load-
caring substrate structure by reducing thermal damage to the substrate material during the coating
process.

3.3 Debinding and Sintering

To investigate the effects of different debinding and sintering methods the dried and pressed thermal
barrier coating samples are subjected to different combinations of debinding and sintering methods.
The three methods for debinding and sintering involve either using an oven to both debind the coating
and continue to sinter the coating, debind the coating within the oven where it is then removed to be
sintered using an oxyacetylene torch or using an oxy-acetylene torch to simultaneously debind and
sinter the coating.
4 Testing

4.1 Surface Structure Studies

To promote sintering of the coating surfaces and decrease the total sintering time pressure is applied
to the unsintered coatings to reduce the porosity of the coating by forcing the powder particles
together into a denser arrangement with increased contact areas between particles as depicted in
Figure 3. The increased contact areas promote the growth of grain boundaries between the powder
particles and significantly decrease both the time and temperatures required for sintering of the
coatings [7].

Pressure

High Porosity coating powder Powder particles with reduced porosity

particles with little contact area and increased contact area

Figure 3: Reduction of porosity via pressure application

Scanning Electron Microscope (SEM) images of the top surfaces of the coatings (a) prior to and (b)
after pressure application depicted in Figure 4(a) show the increased surface contact between coating
powder particles as a result of the application of pressure to the coating surface prior to being
sintered. The effect of the application of pressure can be seen by the flattened and distorted powder
particles observable in Figure 4(b) and the increased contact boundaries.

(a) Before (b) After

Figure 4: SEM images of thermal coatings, before and after application of a 30 MPa pressure.
It is important that prior to subjecting the thermal barrier coatings to thermal loads that the entire layer
has been fully sintered to ensure durability and resilience to spalling. Therefore the time taken to reach
a completely sintered state is important in characterising the cost effectiveness and possible
applications of the Slurry Spray technique.
The mechanics behind the sintering effect are driven by the reduction in surface energies of the
individual particles by coalescing to form a continuous medium with increased grain boundary areas
and reduced porosity [9] as depicted in Figure 5.
Images were taken via SEM of the thermal coating before and after sintering. From Figure 6(a) it can
be noted that the grain boundaries are still evident, and have yet to start forming into singular bodies.
In comparison to Figure 6(b), it can be seen that the grains have started to coalesce to form a
continuous medium, which show a good agreement with the theory and mechanics of sintering.
 Grain
Boundary

Central
Pore

Un-sintered Particles Beginning Of Sintering Fully Sintered

Figure 5: Stages of grain boundary development during sintering

(a) Before (b) After

Figure 6: SEM images of thermal coatings, before and after sintering

4.2 Preliminary Testing

Very little details were available in the open literature on the processing and fabrication procedures for
the Wet Powder and Sintering method. The initial experiments were conducted in a trial and error
fashion to produce a working coating sample.
It was found in our study that TZ-3Y-E, a partially stabilised Zirconia powder, was the most suitable
ceramic powder for these preliminary experiments. The percentage of ceramic powder, which yielded
smooth and consistent slurry compositions, was found to be 45%, while a percentage of 50%
produced flocculated slurry, producing surface irregularities during spraying onto substrate. This
caused problems with pressure stamping, and in turn generated cracking and spalling during and after
the sintering process. The optimum percentage of binder was found to be 4%; a higher percentage
would produce slurry that is too vicious to spray.
To ensure surface integrity, each layer was left to dry for 1 hour between application of coats, and at
least 3 hours after the final coat. The application of the surface pressure greatly reduced the sintering
time. Pressures of 10 to 60 MPa were applied to the samples, where the optimum amount of applied
pressure was found to be 25 MPa.
After experimenting with different sintering durations, all coatings were observed to be adequately
sintered after 30 minutes. The oxy-acetylene torch flame was set a distance from the substrate such
that the torch flame was touching the surface of the substrate. The ambient environment was also
found to be influential on the coating integrity during cooling.
To ensure the integrity of the thermal coatings, tests were developed to determine whether fabricated
coatings could withstand thermal cycling and show that the Slurry Spray method is a viable alternative
to other thermal spray methods. A laboratory oven was used to thermal cycle the specimens. The
specimens were cycled twenty times between 800oC and ambient temperature. Scanning electron
microscope (SEM) images were taken every five cycles. No noticeable cracking of the coating
occurred from 0 to 10 cycles, as shown in figure 7(a). After 20 cycles cracks developed along the left
hand section of the image Figure 7(b). While cracking does occur throughout the cycling, degradation
of the coating is minimal over the cycle range.
 (a) 10 Cycles (b) 20 Cycles
Figure 7 – Thermal Cycling
5. Conclusion
A feasibility study of a relatively simple and economical method for producing functionally graded
thermal barrier coatings using the Slurry (Wet Powder) Spray and Sintering technique showed
promising results. It utilises a pressurised spray gun to spray a slurry mixture of a powdered coating
material suspended in a liquid solution directly onto a substrate surface followed by sintering using an
oxyacetylene torch or oven.
The effects of slurry composition, type of ceramic powder, compatibility with substrates, spraying and
sintering conditions were studied. The optimum time and amount of applied pressure for sintering
were determined and described above. The manufactured FG TBCs were shown to survive 15 to 20
thermal cycles from ambient temperature to 800 Co, which complies with the requirement on TBCs
survivability in aerospace applications. Further study will include a theoretical study involving
micromechanical modelling of macro and micro-stresses during the temperature excursions. From this
study we expect to find optimal number of layers, their thickness and composition to further increase
durability of the thermal barrier coating [10].

6 Acknowledgements
The authors would like to thank Dr. Andrei Kotousov for their advice and guidance in the development
of this work, and the financial support from the US Airforce.

References
[1] Koizumi, M, 1997, FGM activities in Japan, Composites Part B; Engineering, 28 (1-2), 1 -4.
[2] Martena M., Botto D., Fino P., Sabbadini S., Gola M.M. and Badini C. (2006) Modelling of TBC system
failure: Stress distribution as a function of TGO thickness and thermal expansion mismatch, Engineering
Failure Analysis, 13, 3, 409-426.
[3] B. Kieback, A. Neubrand and H. Riedel, 2003, Processing techniques for functionally graded materials,
Materials Science and Engineering A, 362 (1 –2), pp. 81–105
[4] Fevre, M, Finel, A & Caudron R 2004, ‘Local order and thermal conductivity in yttria stabilized zirconia’,
Laboratoire d’Etude des Microstructures, Onera, Chatillion, France.
[5] Roy, P, Bertrand, G & Coddet, C 2005, ‘Spray drying and sintering of zirconia based hollow powders’,
Powder Technology Journal, vol. 157, pp. 20-26.
[6] Greenwood, R & Kendall, K, 1999, ‘Selection of suitable dispersants for aqueous suspensions of zirconia
and titania powders using acoustophoresis’, Journal of the European Ceramic Society, vol 19, pp. 479-488.
[7] Baklouti, S, Chartier, T & Baumard, JF 1998, ‘Binder distribution in spray dried alumina agglomerates’,
Journal of the European Ceramic Society , vol. 18, pp. 2117 -2121.
[8] Dahl, P, Kaus, I, Zhao, Z, Johnsson, M, Nygren, M, Wiik, K, Grande, T & Einarsrud, M-A 2006,
‘Densification and properties of zirconia prepared by three different sintering techniques’, Ceramics
International, online paper.
[9] Guillaume B-G, Nathalie M and Christian G 2007, Sintering of ceramic powders: Determination of the
densification and grain growth mechanisms from the “grain size/relative density” trajectory, Scripta
Materialia, vo l. 75, pp 137 -140.
[10] Tsukamoto, H. and Kotousov, A. 2007, Micromechanical study on high-temperature behaviour of ZrO2
particle-dispersed Ni composites, Key Engineering Materials, Vols. 340-341, pp. 95-100.