CTE Measurement - How To

[EmilAV]

Hi,

We are currently trying to evaluate CTE for different ceramic top coats with various spraying parameter.

How is it normal to measure CTE for coatings?

Best Regards,
EmilAV

[loriolo]

Hi EmilAV
Behavior of metal oxides as sprayed are different about full ceramics pieces. In which is possible to have CTE, but the Coefficient of Thermal Expansion on metallic oxides sprayed, changes with systems, parameters, gun in use, this because porosity of the coating has different porosities, so it isn’t standard, you could do a valuation piece by piece, measuring it directly while change the temperature

[EmilAV]

Thank for the reply Luigi,

I have some problems understanding the numbers supplied by powder manufacturers. For myself, I have done some powder pellets and done dilatometry analysis (DIL). Intuitively, I am perfectly clear that these values probably doesn't represent the actual coatings. CTE in a coatings should also be an anisotropic in thermal spray coatings. Normally and parallel to the surface should be different due to the nature of the build up. Therefore, I seek methods of measuring the relevant behaviour. It would also be interesting to understand numbers presented by the powder suppliers.

[Vadim Verlotski]

Thank for the reply Luigi,

I have some problems understanding the numbers supplied by powder manufacturers. For myself, I have done some powder pellets and done dilatometry analysis (DIL). Intuitively, I am perfectly clear that these values probably doesn't represent the actual coatings. CTE in a coatings should also be an anisotropic in thermal spray coatings. Normally and parallel to the surface should be different due to the nature of the build up. Therefore, I seek methods of measuring the relevant behaviour. It would also be interesting to understand numbers presented by the powder suppliers.

Hi Emil,
Dense ceramic made from YSZ has a coefficient of thermal expansion of approx. 10.8x10-6K-1. Theoretically, the plasma-sprayed coating has a similar value, but because of the crack network, the term "thermal expansion coefficient" loses its meaning. Since the coating is connected to a certain solid substrate, it moves together with the substrate and shows a coefficient of thermal expansion of the substrate to the outside.
In short: ceramic layers with a network of cracks adapt to the deformation of the substrate, provided that the substrate has a higher theoretical coefficient of thermal expansion.

[EmilAV]

Thank you Dr. Verlotski for your reply.

I understand your point regarding networks of cracks. The substrate being key to the key driver for the global deformation/expansion is for me quite clear. As you mention, as long as it is not the other way around. However, my interest in these numbers is based on the urge to understand and quantify the mismatch between different substrate and dense ceramic HVOF coatings where cracking would be a limitation to the actual application. The best approach I can come up with today would be spraying extremely thick coatings and isolate them from the substrate to perfrom DIL on small pieces of pure coating.

[Vadim Verlotski]

Thank you Dr. Verlotski for your reply.

I understand your point regarding networks of cracks. The substrate being key to the key driver for the global deformation/expansion is for me quite clear. As you mention, as long as it is not the other way around. However, my interest in these numbers is based on the urge to understand and quantify the mismatch between different substrate and dense ceramic HVOF coatings where cracking would be a limitation to the actual application. The best approach I can come up with today would be spraying extremely thick coatings and isolate them from the substrate to perfrom DIL on small pieces of pure coating.

Your reasoning is absolutely correct. To determine the CTE of the coating, it must be sprayed with a thick layer (more than 2 mm) on a soft metal substrate, and then saw off the substrate from the coating. A standard beam must be cut out of the sawn-off coating, which is suitable for measuring with a dilatometer. I myself made such polished beams, 30x2x2 mm in size, cut from thick coatings sprayed onto an aluminum substrate.
By the way, by cutting out beams from the coating, you can determine not only the CTE, but also the modulus of elasticity of the coating, as well as its bending strength.