Community Research and Development Information Service - CORDIS

Final Report Summary - SAMBA (Self-Healing Thermal Barrier Coatings)

Executive Summary:
The new concept of self-healing thermal barrier coatings (TBCs), made of novel MoSi2 particles encapsulated with Al2O3 embedded in Yttria Stabilized Zirconia (YSZ), is demonstrated in the SAMBA project. Autonomous repair occurs of crack damage that evolves in the TBC during thermal cycles as experienced in service of turbine engines for power generation and airplanes.

The MoSi2 particles are alloyed with boron to promote the kinetics of the healing reaction and filling of the crack gap with amorphous silica that reacts to crystalline zircon. Three different methods have been developed to encapsulate the healing particles with alumina, which prevents premature oxidation and aids the triggering of the healing reaction when interacting with a crack. For manufacturing the self-healing TBC by atmospheric plasma spraying, encapsulation by selective oxidation of aluminium, added to the healing particles, proved to be most successful. The core of the encapsulated and embedded healing particles remained intact when exposed to high temperatures in air for long times. The size distribution of the alloyed MoSi2 particles was optimized by wind sifting. Both the manufacturing of the alloyed healing particles and the method of encapsulation are filed for patent. A new method to manufacture the self-healing TBC by plasma spraying with double injection has been developed, allowing simultaneous deposition of the YSZ matrix and the alloyed MoSi2 particles onto a turbine engine component. Also, this method is filed for patent.

Advanced characterization techniques have been employed to guide the design and development of the self-healing TBC and to determine their microstructure features and properties. The physics and chemistry of the healing mechanism have been established and the crack gap filling is determined. The effect of embedding the MoSi2 healing particles into the TBC has a small effect on its functional and mechanical properties.
The crack-particle interaction is studied and simulated by fracture mechanical modelling. The obtained knowledge is used to develop a simulation tool to detail the design of the self-healing TBC in terms of volume fraction and size of the healing particles as well as its configuration. This simulation tool is made publicly available and allows to predict damage and healing evolution during service and ultimately the lifetime of TBCs.

In furnace cycle tests (from 100 up to 1100 C), mimicking TBCs in applications, the crack damage evolution due to mismatch in thermal expansion is determined. Cracks run into the TBC where the self-healing mechanism can be triggered. This is even more pronounced when healing particles are embedded in the bottom part of the TBC. X-ray computed tomography and non-destructive thermographic analysis revealed the evolution of crack damage and the effect of the self-healing particles.

The lifetime of the self-healing TBC in the furnace cycle tests compared to a similar TBC but without healing particles, was prolonged and the scatter in the lifetime data reduced making the self-healing TBC more reliable. Further extension of the self-healing TBC lifetime can be expected when increasing the volume fraction of healing particles. This may be achieved when the composition of the alloyed MoSi2 healing particles is made more uniform and their shape more spherical.

To this end, the manufacturing of both the healing particles and the new process of plasma spraying must be made robust and hence reproducible.Overall, the lifetime of TBC with self-healing particles has increased and the first campaign of self-healing TBC showed a tremendous increase of lifetime by a factor of 2 to 3, which offers a very promising perspective to bring this concept to the market. Overall, the SAMBA project was successful.
Project Context and Objectives:
This Summary Description is divided into two sections:

1. Context and Project Set-up;
2. Objectives;

1. Context and Project Set-up

Thermal barrier coatings (TBC) are applied to metallic surfaces of combustion chambers, blades and vanes in gas turbine engines used for propulsion and power generation to allow for higher operation temperatures. This thermal barrier is crucial to enhance the engine’s fuel efficiency and consequently to reduce the emission of greenhouse gasses. The turbines and the TBCs experience severe thermal cycles spanning a range of over 1000 °C due to starts and stops of an engine or installation. During cooling from the operation temperature to room temperature, high stresses develop due to a mismatch between the coefficients of thermal expansion of the metal substrate and the ceramic TBC. In time, these stresses result in the development of crack patterns in the TBC that coalesce and ultimately lead to failure. Cracks that run through the TBC perpendicular to its surface are not detrimental per se, but in conjunction with cracks that develop parallel to the interface they lead to spallation (i.e. fragmentation) of the TBC, directly exposing the substrate to the high-temperature environment.

The SAMBA project created a unique self-healing thermal barrier coating (SH TBC) for turbines and other thermally loaded structures in order to realize a significant extension of the lifetime of critical high-temperature components. As currently available TBCs do not exhibit any self-repair, the new self-healing TBC will offer a reduction of the number of TBC replacements during an engine lifetime and enhance the reliability of the critical components. The concept is based on novel Al2O3 coated MoSi2 particles embedded in the TBC layer, typically consisting of yttria-stabilized zirconia (YSZ).

The research was focused not only on just creating the new particles, but also on the production and testing of complete TBC systems and a quantitative determination of the lifetime extension of such a new coating system. The research also focused on qualitative and quantitative understanding of the mechanisms of damage development and crack healing in TBCs. Modelling of these mechanisms, using input from state of microstructure and properties analysis techniques as well as microstructure and crack imaging techniques, enabled design and optimize the new self-healing TBC system.

2. Objectives

The primary goal of this project is to realize and optimize the self-healing capacity of thermal barrier coatings with MoSi2 dispersed particles for application in aero engines and industrial gas turbine engines to prolong the lifetime of their components. This is achieved through a combined theoretical, experimental and modelling approach of a new, innovative self-healing concept. Upon local fracture of the TBC, these particles fill the crack initially with a glassy phase that subsequently reacts with the matrix to form a load bearing crystalline ceramic phase. This self-healing concept can be exploited to other high temperature structural ceramics as well.

This project aimed to prolong the thermal barrier coating lifetime by 20-25 % through autonomous self-repair of occurring crack and delamination damage upon service (demonstrated in a furnace thermal cycle test).

Further, the project aims to improve European industry competiveness by designing a new generation of affordable self-healing materials and associated manufacturing technologies. A significant economic benefit can be obtained by reducing the number of TBC replacements in critical turbine engine components and by decreasing the effective downtime of the equipment.

Project Results:
This section describes the main Scientific and Technology results:

1. Concept and Design;
2. Manufacturing of self-healing TBC;
3. Properties of self-healing TBC;
4. Thermal cycle performance;
5. Outcome and Outlook.


1. Concept and Design

The self-healing TBC is applied to a nickel based superalloy (Hastalloy X) coated with a NiCoCrAlY bond coating. To provoke delamination cracks to run into the TBC, where the self-healing can be triggered, a rough interface with the bond coating is created. Such interface by itself already prolongs the lifetime of the TBC system.

The healing particles are made of MoSi2 alloyed with 1-2 wt% boron to enhance the formation of SiO2 as well as to promote its amorphous rather than crystalline structure. It has been demonstrated that both these aspects are beneficial for filling of the crack gap and zircon formation. For effective self-healing, the volume fraction of these particles in the bottom part of the TBC is designed to be about 10 %, when the average particle size is in the range of 20 to 30 micron.

This new concept involves the creation of an inert, oxygen impenetrable, shell of alumina (Al2O3) around the actual healing agent, which prevents premature triggering of the healing reaction. With this approach, the healing mechanism will become active only when a crack penetrates the alumina shell.

In order to prevent premature oxidation of the healing particles, different routes for encapsulation have been explored. Soft chemical methods (precipitation and sol-gel) to coat the particles with boehmite and subsequently calcination to form Al2O3 were successful and can be used when making self-healing TBC or other high temperature ceramics.
However, when applying self-healing TBC with plasma spraying techniques, as done in the SAMBA project, an alternative method is preferred. Then the encapsulation is realized by selective oxidation of aluminium, which besides boron is also alloyed with the MoSi2 particles. Tis newly developed in situ encapsulation method comprises a post high temperature annealing treatment at low oxygen partial pressure of the deposited self-healing TBC. This method resulted in fully encapsulation of the alloyed healing particles by a shell of alumina Al2O3.

As long as the self-healing mechanism is not triggered by a crack the healing particle should remain intact. Exposing the encapsulated healing particles to 1100 °C in air for 100 hours showed only little increase of mass. The mass change of encapsulated MoSi2 particles is compared with that of unprotected MoSi2 and MoSi2 alloyed with boron while exposing to 1100 °C in air. Also, high temperature stability tests of the encapsulated healing particles embedded into YSZ showed that the core of the healing particles is still composed of MoSi2 and the shell is somewhat thickened.

Upon high-temperature exposure in an oxidising environment the embedded MoSi2 particles react to form a viscous silica (SiO2), which fills the cracks and re-establishes adherence in the TBC. Subsequently, the SiO2 reacts with the matrix forming zircon (ZrSiO4), which is a load bearing and solid crystalline ceramic phase.

The self-healing mechanism will be triggered when a crack opens the alumina shell and run into the healing particles. Since the MoSi2 particles are stiffer and stronger than the YSZ matrix the crack may be deflected by these particles. But numerous fracture mechanical simulations and experiments showed that a crack indeed interacts with the embedded healing particles.

When the healing reaction is triggered, silica (SiO2) is formed. This reaction is associated with a volume expansion of 2.2 (i.e. 120 %), which is necessary to fill the crack gap. The filling of the crack gap, can be estimated from a statistical model. A 3D diffusion model has been adopted to calculate the crack-filling behaviour of the self-healing mechanism. To sum it up, the filling of the crack gap depends, besides the volume expansion, on the volume fraction and size of healing particles, the width of the crack, and the oxidation kinetics, while taking into consideration both time and temperature.

Since the healing reaction is a thermally activated process, it obeys Arrhenius type equation. The parameters for the formation of SiO2 from MoSi2 particles with and without boron are identified. Clearly, the SiO2 formation from MoSi2 particles with boron is much faster (about a factor of 4) than from pure MoSi2. The effect of the added boron is evident. Moreover, when boron is added amorphous rather than crystalline SiO2 is formed according to X-ray diffraction analysis.

After filling the crack with amorphous silica, it reacts with the zirconia of the YSZ matrix, to form crystalline zircon. This reaction is associated with only a small volume decrease of about 18 %, which brings the fracture surfaces closer together. Also, the kinetics of this reaction is studied by preparing bi-layers of MoSi2 with boron and YSZ using spark plasma sintering. The formation zircon is slower than the formation of SiO2, but is also enhanced when boron is added.

The actual crack healing by MoSi2 particles embedded in YSZ TBC has been analysed. The healing particle becomes depleted from silicon and the silica formed by oxidation fills the crack gap. At the fracture surfaces zircon is formed resulting in a strong adhesion between the TBC matrix and the healing agent. In the smaller crack gap, all the silica is transformed into zircon. So, the self-healing mechanism is proven.

A Finite Element (FE) based simulation tool was developed to study the crack damage and healing in the TBC system and to further optimize the design of self-healing TBC. The governing equations for the thermal, mechanical and chemical processes, together with constitutive models, material parameters, loading conditions and geometrical data are included. Nucleation and growth of micro-cracks is modelled using a versatile numerical simulation technique where cohesive elements are inserted between all bulk elements. The fracture behaviour of the distinct phases and interfaces in the TBC system are accounted for using specialized cohesive relations in the corresponding cohesive elements.

Incorporation of the healing effect in the simulations entails:
(i) the triggering mechanism (cracks that expose the healing particles to oxidation) and
(ii) the re-cohesion of cracks due to oxidation.

The main modelling steps to predict the lifetime of the self-healing TBCs includes:
(i) the simulation of the fracture response of the representative volume elements under thermomechanical loading of a TBC system,
(ii) the coupling of the Thermally-Growth Oxide layer (TGO) growth model to the thermomechanical FE simulations and
(iii) the incorporation of a healing model to simulate the healing process.

The simulations include models to predict the onset, growth and coalescence of cracks that lead to final failure of a TBC system by spallation. A self-healing model was developed and implemented to account for the strength recovery associated to the healing particles. The simulation tool used in a thermal cycling analysis framework allows to estimate the lifetime of a self-healing TBC system and to further quantify the effect of healing on the lifetime extension of the TBC system.

Analysis of the results of many simulations with the current tool revealed that:
- A particle-based self-healing mechanism can be activated even under unfavourable conditions (i.e., even for relatively stiff particles and relatively weak interfaces). The main parameters governing the activation are the fracture strength and the fracture energy of the particles, which should be relatively low compared to the properties of the matrix.
- Healing particles should be placed in the top coat layer (TC) close to the TGO in order to delay coalesce of cracks in those layers that lead to final failure by spallation. The coefficient of thermal expansion of the healing particles should be slightly less than that of the hosting matrix to promote linking and subsequent healing between cracks in the TGO layer and cracks in the top coat layer (TC) where the healing particles are hosted.
- Strength recovery due to sealing of cracks with a healing product requires long distance propagation of the healing agent along the crack. An efficient mechanism can be achieved when the healing agent is transported along the crack in a fluid-like state at high temperature. Subsequent solidification upon cooling to a ceramic-like phase (that remains solid) provides the required strength recovery.

The simulation tool consisting of several numerical modules together with a comprehensive manual can be downloaded from the website www.SAMBAproject.eu.

2. Manufacturing of self-healing TBC

Since atmospheric plasma spraying (APS) is the most widely used method to apply thermal barrier coatings, also this method is used to deposit the self-healing TBC. The plasma spray process is basically the spraying of molten or heat softened material onto a surface to provide a coating. Material in the form of powder is injected into a very high temperature plasma flame, where it is rapidly heated and accelerated to a high velocity. The hot material impacts on the substrate surface and rapidly cools forming a coating. The plasma spray gun is made up with water cooled copper anode and tungsten cathode. The plasma gas (a mixture of argon and helium) flows around the cathode and through the anode, which is shaped as a constricting nozzle. The plasma is initiated by a high voltage discharge which causes localized ionization and a conductive path for a direct current (DC) arc to form between cathode and anode.

In the first pass of spraying both yttria stabilized zirconia and the alloyed MoSi2 particles are deposited resulting in a layer of about 100 microns. The next layers are composed of only YSZ making the total thickness of the self-healing TBC about 500 microns. The parameters for plasma spraying (like: electric current, spraying distance, gas composition and flow rate, etc.) are adjusted to the powder properties.
The certified YSZ powder used is optimized for plasma spraying and the deposition parameters are well established. The alloyed MoSi2 healing particles made for manufacturing the self-healing TBC requires different plasma spray parameters. For example, the melting point of MoSi2 (2030 °C) is much lower than the melting point of YSZ (about 2700 °C). Most critical is the particle size; too small particles will evaporate in the plasma plume, while too large particles will not melt. Hence, a narrow size distribution of the healing particles is required.
To this end, the batch of alloyed MoSi2 particles were classified by wind sifting prior to plasma spraying. After wind sifting the fines are removed and a sharp size distribution is obtained.

Moreover, the original composition of the particles must be preserved after plasma spraying. The silicon in MoSi2 has the tendency to evaporate during plasma spraying. Also, some loss of aluminium may occur. To deal with these aspects a double injection method was developed to avoid these risks.

This newly adopted method allows optimization of the plasma spray conditions for simultaneously depositing YSZ and MoSi2 by varying the injection distance of MoSi2. A design of experiment (DoE) was used to find the optimal spray conditions for a self-healing TBC. An analysis of variables with the determined values for porosity and content of MoSi2 shows a dependency of the current (I), spraying distance (SD) and their combination (I*SD) for the porosity. The content of MoSi2 depends on the current (I), injection distance (ID) and their combination (I*ID).

Series of self-healing TBC were applied onto super alloy buttons with a diameter of one inch provided with a bond coating for determining the lifetime in furnace cycle tests. For comparison of the results also series of the same TBC system were prepared, but without healing particles.

3. Properties of self-healing TBC

Embedding the alloyed MoSi2 particles into a YSZ matrix may affect the mechanical properties as well as the thermal conductivity. The mechanical properties are relevant for fracture and ultimately delamination of the TBC, while the thermal conductivity is vital for its functionality. Therefore, the SAMBA project studied the characteristics of the SH TBC.
Dense composites of YSZ with different volume fraction MoSi2 particles up to 20 % were prepared by spark plasma sintering (SPS) to determine the effect on the mechanical properties and thermal conductivity.

The Young’s modules (260±20 GPa), hardness (16±2 GPa) and fracture toughness (5.2±0.5 MPa.m1/2), all determined by Vickers indentation, were within the statistical error independent of the volume fraction MoSi2.

The thermal conductivity of the YSZ composites, measured in argon from room temperature up to 1000 C with a laser flash system, increases with the volume fraction of MoSi2 particles. The measured thermal conductivity for the pure YSZ samples (1.2 ~ 1.3 Wm-1K-1) are lower than its bulk value (2 ~ 2.5 Wm-1K-1), presumably due to the pores within the sample. It also reveals the effect of MoSi2 addition on the effective thermal conductivity. It can be seen that the effective thermal conductivity becomes more temperature dependent as more MoSi2 are added. Moreover, the thermal conductivity at room temperature increased from 1.3 Wm-1K-1 to 2 Wm-1K-1 when 20 vol.% of MoSi2 is added. Still, this value is well below the thermal conductivity of pure MoSi2 with a similar porosity (~ 38 Wm-1K-1).

Similar measurements were performed on the plasma sprayed TBCs with and without healing particles. The elastic modules increase on average from 105 to 123 GPa with MoSi2 healing particles. Due to the porous nature of the APS coatings, both data sets show considerable scattering. The slightly higher value for the self-healing TBCs can be explained as the effect of MoSi2 whose elastic modulus (~ 460 GPa) is much higher than YSZ (~ 210 GPa).

The effective thermal conductivity of the plasma sprayed TBCs with and without healing particles at room temperature were compared. Although the self-healing TBC still show a higher thermal conductivity than the benchmark TBC, the increase by adding MoSi2 particles has been limited to only about 22%. More importantly, the effective thermal conductivity for the APS samples are only about half of their counterparts manufactured via SPS with a similar porosity level.
4. Thermal cycle performance

Furnace Thermal Cycle Tests (FCTs) were executed to observe the performance of TBC system and to determine its End-of-Life (EoL). A strict protocol was devised and followed to perform the FCTs as well as the characterization and post mortem analysis of the TBC system.

A first series of self-healing TBC was produced and subjected to a furnace cycle test. A similar TBC, but without healing particles, last about 150 cycles in this FCT at 1100 °C with a hot dwell time of 2 hours and a heating and cooling times of 7 and 15 minutes, respectively. When increasing the concentration of aluminium in the MoSi2 healing particles and reducing their volume fraction, while increasing the injection distance, led to a large increase in lifetime. These self-healing TBCs lasted 400 to 500 cycles in the FCTs, thus nearly tripled the lifetime. These very promising results also indicate that the composition and therewith the encapsulation of the healing particles is critical.

Then, second series of of self-healing TBC with seemingly the same optimized parameters were produced. One batch tested at 1060 and 1100 °C showed a prolonged lifetime. The lifetime data are compared with those of similar TBC, but without healing particles. One so-called benchmark TBC was produced early on in the project, and another batch at the same time as the self-healing TBCs, denoted as benchmark A and B, respectively. Inherent to the brittle and porous nature of the ceramic TBC a large scatter is observed in the lifetime data. Moreover, there is a significant difference in the lifetime of both benchmark TBCs, although these were prepared from the same materials and with the same plasma spray parameters.

However, in both FCTs the self-healing TBCs last the longest! It is interesting to note that also the scatter of the lifetime data of the self-healing TBCs, are smaller when comparing with benchmark A TBCs. For manufactures, designers and operators of gas turbine engines provided with TBCs, this aspect is as important and attractive as prolonging the lifetime, because the lower limit or threshold lifetime is increased.

The FCTs were accompanied with X-ray computed tomography (XCT) and thermography analysis to monitor the evolution of crack damage and healing of TBCs non-destructively. Two thermographic techniques have been explored, one based on thermal diffusivity and another based on thermal effusivity. Small crack damage that is apparent, even after a few thermal cycles, can be observed. A quantitative evaluation of crack damage in terms of fraction delaminated during FCT can be determined after preliminary calibration on some samples of the same TBC set.
The normalised thermal diffusivity approach gives evidence of a constant growth of diffuse micro-cracking in the central sample area up to a certain level close to the end of life, when an acceleration can be observed. A significant number of diffuse cracks nucleates and propagate before the occurrence of the final failure of the self-healing TBC and a remarkable increase of the effective cracking thickness can be observed as well. This provides an indirect evidence of some effectiveness of the self-healing particles in promoting cracks and their subsequent healing when exposed to high temperatures.

XCT and apparent effusivity maps clearly show for almost all the tested TBC samples that macro–cracking occurs even before the first 1/3 of life all around the sample edges forming a circular ring. This edge cracking is typical for the disc shape sample geometry. After the initial growth of the circular ring, small diffuse not connected delamination start to appear in the central area of samples.

Later, when the delaminated fraction reaches a value in the range 40- 50%, a significant acceleration of macroscopic damage involving even the central region of the sample occurs. This phase corresponds to the coalescence of micro-cracks towards macro-delamination up to values in the range 70-80%. Then there is a final phase where macro-delamination rapidly coalesces up to the complete spall-off of the TBC.

The macro-crack damage evolution of both the self-healing and benchmark B TBCs are compared. Clearly, when the crack damage starts accelerate and micro cracks coalescence in the self-healing TBC final failure is delayed due to the presents of the encapsulated MoSi2 healing particles.
The data of the thermographic analysis at intermediate stages of the furnace cycle tests can also be used to predict the EoL of the TBC system. A comparison was made between these estimated and the experimentally determined EoL of TBC systems.

The results of XCT and thermographic analysis are in agreement with microscopic analysis. Cross sections of the samples were prepared after different stages of their lifetime in furnace cycle tests to observe the crack damage using electron microscopy. These analysis show that after about 1/3 to 1/4 of lifetime, cracks occurred at specimen edges, but not in the specimen centre due to the disc shape geometry. The edge crack propagates within the ceramic topcoat along topcoat porosity aligned parallel to the bond coat surface (typical feature of this particular system) and stops approximately a few millimetres away from the edge.

After the initial growth of this circular ring, small diffuse not connected cracking some hundreds of microns long and few microns wide can be observed also in the central area of samples. These cracks stay both within the TBC top layer and the interface between the bond coating and TBC top layer.

Comparing sections of samples aged up to different fractions of their life, there are indications supporting the idea that a constant growth occurs of diffuse micro-cracking in the central sample area up to a certain level close to the end-of-life. Thus, the final failure of the system occurs in the centre of the specimen by crack propagation at or in near vicinity to the bond coating. This two-stage failure mode consists the so-called “white failure” along sample edge and “grey failure” in the specimen centre.

It is worth noting that most of the benchmark TBCs failure occurred near the interface between the bond coating and the YSZ top layer, while the self-healing TBCs exhibit a statistically significant tendency to fail more within the YSZ top layer. Thus, favourable for triggering the self-healing mechanism. This experimental finding is in agreement with our fracture mechanical FE simulations.

5. Outcome and Outlook

An affordable self-healing TBC has been developed and prepared. The lifetime compared with a similar but non-self-healing TBC is prolonged. Moreover, the scatter in the lifetime data of the self-healing TBC is reduced, which raises the lower bound or threshold lifetime and thus more reliable.

The development of the SH TBC is close to TRL4, which is an excellent achievement. In order to fully reach TRL 4 and work towards TRL5-9 further research and development is needed, especially focusing on the reliability and reproducibility of the particles and composition methods. Once these processes are stabilized further testing can be initiated. Project partners have already deployed follow-up activities and aim to achieve real-time application within 2 man-years.

To bring the concept of self-healing TBC to the market, the manufacturing of both the healing particles and the process of plasma spraying with double injection must be made more robust and hence reproducible.
Yet, the composition size as well as shape of the healing particles vary. Significant improvement has been realized by wind sifting the initial powder resulting in a sharp size distribution. Still the particles have an angulated rather than spherical shape. Moreover, the variation in composition from particle to particle is also of concern in view of proper in situ encapsulation via selective oxidation of aluminium.
As the volume fraction of healing particles in the current self-healing TBC is only 3 %, significant further extension of the lifetime can be realized when this volume fraction is brought to its design level, i.e. up to 10 %. Then, the composition of the healing particles should be better controlled, i.e. the variation of its composition and size should be small.

These aspects are yet subject of further research at Delft University of Technology. Preparation of the alloyed MoSi2 particles by spark plasma sintering to obtain a uniform composition is in progress. Next, methods of powdering and spherodization are under development. The knowledge with regards to the manufacturing of the alloyed self-healing particles is protected by a patent that is about to being filed.

The plasma spray process for the self-healing TBC can be made more robust by detailed monitoring of this spray process and by investigating the sensitivity of the coating system microstructure and properties to the process parameters as: plasma current and gas flow, spray distance, powder feed (both particles and ceramic), self-healing particle injection distance, angles (relative to plume and rotational angle relative to gun), etc. Next, the reproducibility of the plasma spray process should be demonstrated, which is under consideration at Forschungszentrum Jülich (FZJ) together with our industrial partners GKN Aerospace, GE power and Flame Spray Technology (FST). The invention of manufacturing the self-heling TBC by double injection of YSZ and the healing particles, respectively, is filed for patenting.

Finally, GKN Aerospace and GE power, as major actors in the aero and gas turbine industry, have the ability to introduce new technologies in gas turbine engines when the technology and improvements are proven with sufficient data. The participation of FST as powder supplier will increase the chance of market introduction in aero and gas turbines and other areas of applications. The approach to get the self-healing TBC into an engine is to do an engine test where a low risk component (low risk component is defined as TBC spallation with no risk of significant damage in engine) will be identified and used as a candidate for the self-healing TBC. After successful tests, these consortium partners could proceed towards market introduction.

Potential Impact:
The following sections will be described under Impact:

1. Context;
2. Profiling of current and near-future impact;
3. Strengths and Weaknesses;
4. Opportunities;
5. Economic impact;
6. Societal impact;
7. The main dissemination activities;
8. Conclusion.



1. Context

Ceramic thermal barrier coatings (TBCs) themselves already contribute tremendously to
sustainable technology, because it enhances the engine efficiency by allowing higher operation
temperatures, which saves fuel and thus reduces CO2 emissions. Furthermore, it protects the high tech structural components (made of single crystal super alloys) against severe high-temperature corrosion, thereby contributing to durable use of resources. The materials and their amounts used to produce the current as well as the prospective self-healing TBCs are abundant, relatively cheap and not environmentally hazardous. The added value grows exponentially when starting from the raw materials to application by plasma spraying onto components integrating these into complete parts for gas turbine engines, which finally are applied in complete systems such as aeroplanes or power stations.

The innovative concept of healing crack damage proposed is generic and can be adopted
naturally to develop a wide range of new self-healing ceramics. Clearly, this will offer substantial opportunities to prolong the lifetime and enhance the reliability of other ceramic components used in high-temperature applications. Overall, the new applications may ultimately outperform other materials (e.g. scarce metals) due to reliability, longer lifetime, cost-efficiency and capabilities. Such developments will benefit from the knowledge and experience gained in this project.

The impact of SAMBA must be seen in this larger context. The proven concept of the self-healing TBC has clearly taken a step forward. The modelling tool for design is available to the world, so all actors in the field of TBC can make use of this in their research and development activities. This will lead towards further development of the application of SH TBC. However, one must take into account that the stability and reproducibility of the mechanism needs initial enhancement. Nevertheless, the experiments have shown that the SH TBC was triggered and prolonged the lifetime sufficiently to further investigate its potential. Direct application and bringing it to market could follow within approximately 2 years FTE effort. With the reached objectives of this high-risk project it can be stated that is has created impact through presenting and making available its results to a wide-spread community.

2. Profiling of current and near-future impact

The currently achieved impact is of course less wide and not very pronounced yet. It is mainly focused on the market of the participating partners. However, it is foreseen that in two year from now the technology could already been improved so it can be brought to the market. Follow-up projects are considered to bring the results to TRL 5-9.

The application fields are in SH TBC, where first the reproducibility and stability of the production of the self-healing particles and stability of the double-injection plasma spraying need to increase, for which the consortium partners continue to collaborate. Follow-up projects are taken into consideration.

3. Strengths and Weaknesses

The experiments showed an increased lifetime, the costs of the materials needed for the particle production are low. These two aspects can be seen as a strength.

However, there is still need for improvement in the fields of stabilizing and up-scaling the production. Also, the reproducibility needs to be addressed. To bring the concept of self-healing TBC to the market, the manufacturing of both the healing particles and the process of plasma spraying with double injection must be made more robust and hence reproducible. These weaknesses need to be taken into account.

The technique is currently not yet able to be applied directly and/or brought to the market.
In order to bring the new self-healing TBC into industrial production and thus towards the market, in a follow-up of this project many more steps will have to be taken and more research will be required. Assuming the concept to work as foreseen and shown during the experiments, the project has delivered a roadmap to take the new TBC system through the TRL 5-9.

4. Opportunities

The cost-effectiveness and shown pro-longed lifetime (which could increase even further) are main cost drivers to investigate further development of this technique towards a more sustainable and reliable end-product. Comparing the current test results to equal TBC without SH particles show already an improved reliability.

The current technique has reached TRL 4, the industry partners will take up the technique internally, in close collaboration with the academic partners in order to optimize the plasma spray technique with double-injection and the controlling (and optimization) of the particle size and distribution.

GKN Aerospace and GE power, as major actors in the aero and gas turbine industry, have the ability to introduce new technologies in gas turbine engines when the technology and improvements are proven with sufficient data. The participation of FST as powder supplier will increase the chance of market introduction in aero and gas turbines and other areas of applications. It is estimated that within two man-year the application could be brought to the market.

Consortium members have already been invited to different conferences to present the project results. By doing so, the wider research and development community in the field of material science and engineering can be reached and used in other activities.

5. Economic impact

Although the current technique is not yet at the desired TRL to thoroughly investigate its cost-effectiveness, the test results have shown an increase of life-time at low cost production methods. Even though, this production process needs further optimization, it shows it potential economic impact in a further developing market. The market is moving more and more to smart and sustainable solutions. The application of SH TBCs could enhance this as it would also result in fewer maintenance which increases the potential return on investment (RoI) rate of its application and use; e.g. airplanes could fly longer without the need for maintenance, fewer down-time per engine, fewer repair rounds per engine lifetime.
The growing market for engineering ceramics can still be considered to make use of the SAMBA results which welcomes the new functionality of self-healing for engineering ceramics.

6. Societal impact

Once the application can be brought to the market, which can be done within an estimated 2 year-period and directly by our consortium partners, we can see the societal impact in the following fields:

• Electric power generation and critical infrastructures;
• Safety of turbine engines for aeroplanes;
• High thermos-mechanicals and engineering.

Society would benefit from the direct application in critical infrastructures like hospitals, ensuring a reliable and solid electric power supply. Equipment could run on higher temperature, making it more efficient and reliability increases due to the self-healing mechanism. All together it would lead to fewer down time and lower the need for repairs. The operating systems become more reliable.
In addition, the application in turbine engines in the aviation sector would lead to an increased safety, as fewer damage would occur and cracks would be self-healed instead of further propagating.
The three examples above all aim for more sustainable and smart solution and cost-effectiveness form which society can gain.


7. The main dissemination activities

The results of this project have been published and are widely spread throughout the scientific community through presentations at conferences.
In addition to that, two patents have been filed by the consortium.

Forschungzentrum Jülich applied for a patent on the manufacture of this self-healing TBCs. The patent has been filed and is under embargo. After this sealed-off period, it is foreseen that the patent will be made publicly available.

TU Delft applied for a patent on the novel self-healing particles including the method of in situ encapsulation. It is foreseen that the patent will be publicly available in due time.

8. Conclusion

The call text requested the following impact: “i) Improved materials with prolonged lifetime and reliability leading to enhanced safety in applications such as for example vehicles, roads and bridges; and/or ii) Societal and economic benefits deriving from the reduction of accidents, injuries, casualties, and permanent damages; and/or iii) Improved competitiveness of European industry via more favourable cost/benefit ratios.”

Looking at the results of the SAMBA project, it can be stated that is has succeeded as the project did produce a self-healing TBC with prolonged lifetime and reliability applicable in turbine engines in the aero industry. Furthermore, the direct application, which could occur within 2 years from now, would result in a reduction of permanent damages and fewer accidents involving aeroplanes caused by engine failure. More significantly shown impact at this stage is the overall cost-effectiveness and the already identified optimization of its production methods that would lead to an improved competitiveness of European industry based on a higher RoI rate due to longer lifetime.

All project results have been made available to the scientific and industrial communities to be taken up in further research and development activities. Follow-up projects are considered to continue the promising work of SAMBA.


List of Websites:
Below you will find the relevant contact details.

Public website: www.SAMBAproject.eu

TU Delft:
Wim G. Sloof
Mekelweg 2
2628 CD Delft
The Netherlands

Forschunszentrum Jülich:
Robert Vassen
Wilhelm-Johnen-Strasse
52425 Jülich
Germany

Manchester University:
Ping Xiao
Oxford Road
Manchester M13 9PL
UK

INPT - CIRIMAT:
Daniel Monceau
4 allée Emile Monso
31030 Toulouse cedex 4
France

RSE SPA:
Federico Cernuschi
Via Rubattino 54
20134 Milan
Italy

General Electric:
Bettina Bordenet
Brown Boveri Strasse 7
5401 Baden
Switzerland

GKN Aerospace:
Björn Kjellman
Flygmotorvägen 1
46181 Trollhättan
Sweden

Flame Spray Technologies:
Menno Zwetsloot
Dijkgraaf 40
6921RL Duiven
The Netherlands

Related information

Reported by

TECHNISCHE UNIVERSITEIT DELFT
Netherlands
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top