Exploring Zirconium Use in Reactors: Hydrogen & Oxidation

[Jack_O]

Few things I'm curious about regarding zirconium use in reactors. I know at high temperatures it can release hydrogen from water by forming zirconium oxide. But:

Zirconium alloys readily react with oxygen, forming a nanometer-thin passivation layer.
...
A sub-micrometer thin layer of zirconium dioxide is rapidly formed in the surface and stops the further diffusion of oxygen to the bulk and the subsequent oxidation.

http://en.wikipedia.org/wiki/Zircaloy"

this suggests that once a surface layer is formed no more oxidation can take place, judging by the events at Fukishima there is a mechanism for oxidation to continue, otherwise there would not have been so much hydrogen released. Can anyone elaborate?

It seems oxidation becomes a serious issue around 800-1000C in a water environment, does anyone know what temperature would lead to problems for zirconium/zircaloy in a CO₂ rich environment, for example in an AGR?

 

[Astronuc]

Zr is a reactive metal and it readily forms an oxide. Similarly, Ti and Hf, in the same group as Zr, form thin layers of oxides - nm to microns. Aluminum and Al alloys also form a protective oxide layer. And stainless steels are protected by a passive layer of chromia, chromium oxide.

When Zr operates in high temperature water, > 280°C in a BWR, it slowly reacts with the water or steam and oxidizes. The reaction is relatively slow, but can accelerate according the water chemistry, as well as compostion and microstructure of the Zr alloy, or Zircaloy-2 in a BWR.

With Zircaloy-2, one has to control the microstructure, particulary the second phase particles in order to avoid nodular corrosion (a localized corrosion) or enhanced uniform corrosion.

As temperature increases, the corrosion rate increases. It can accelerate as temperatures increase above 300°C, and particularly above 400°C. At temperatures approaching 600°C, it would be a matter of days before corrosion through the all occurs. As the temperatures goes higher up to 800 to 1200°C, the corrosion rate accelerates such that through-wall corrosion occurs in hours, minutes, seconds.

What is significant in modern BWR cladding is the internal surface of the cladding, which typical is layer of Zr about 3 mils or 75 microns thick. Since the mid 90's the Zr layer has been alloy with Sn or Fe depending on the manufacture. The cladding at Fukushima is likely a liner cladding with a low relatively Fe content in the liner, and that would like oxidize fairly rapidly if exposed to steam at high temperature. Oxidation of the liner would produce a fair amount of hydrogen, as well as putting a significant stress near the cladding breach. It would be possible then for cracks to extend axially from the breach. Locallized hydriding of the cladding would cause embrittlement, and the cladding may break into pieces.

We'll have to wait until TEPCO extracts some fuel assemblies to see what the damage looks like. On the other hand, if the tie-rods have failed, the it many not be possible to lift the fuel assemblies intact. Specially tooling and procedures will be necessary.

AGRs do not use Zircaloy, but rather a special stainless steel.

For looking at stability of metals in CO2, one should become familiar with Ellingham diagrams/plots.
http://www.doitpoms.ac.uk/tlplib/ellingham_diagrams/ellingham.php

 

[Dmytry]

this suggests that once a surface layer is formed no more oxidation can take place

when zirconium is reacting with water, you get not only the oxide but also hydrogen. When you get gas and solid as a reaction product, typically, the solid will be porous (don't know about specific case of zirconium, just the general case). Furthermore, water may react with zirconium oxide, producing zirconium hydroxide, which won't have same volume as original oxide, meaning it will end up breaking apart.

Consider regular (non stainless) steel or iron. It also forms layer of oxide. But in presence of water, iron oxide hydrates, forms oxide-hydroxide, etc etc. which makes the rust porous, allowing oxidation to continue.
http://en.wikipedia.org/wiki/Rust

 

[Jack_O]

when zirconium is reacting with water, you get not only the oxide but also hydrogen. When you get gas and solid as a reaction product, typically, the solid will be porous (don't know about specific case of zirconium, just the general case). Furthermore, water may react with zirconium oxide, producing zirconium hydroxide, which won't have same volume as original oxide, meaning it will end up breaking apart.

Thanks, that clears up my confusion over how the oxidation progresses.

 

[Jack_O]

AGRs do not use Zircaloy, but rather a special stainless steel.

For looking at stability of metals in CO2, one should become familiar with Ellingham diagrams/plots.
http://www.doitpoms.ac.uk/tlplib/ellingham_diagrams/ellingham.php

I have heard from people connected to industry that some magnox/AGRs used zirconium pins while layering graphite blocks to help them stay in place. I guess they have probably completely oxidised and carried away in the gas stream.

Apparently the mass loss of some graphite blocks in magnox reactors is as high as 30%, do you think this is mainly due to standard oxidation due to the high temperature CO₂ rich environment or due to radiolytic oxidation?

 

[Astronuc]

when zirconium is reacting with water, you get not only the oxide but also hydrogen. When you get gas and solid as a reaction product, typically, the solid will be porous (don't know about specific case of zirconium, just the general case). Furthermore, water may react with zirconium oxide, producing zirconium hydroxide, which won't have same volume as original oxide, meaning it will end up breaking apart.

Consider regular (non stainless) steel or iron. It also forms layer of oxide. But in presence of water, iron oxide hydrates, forms oxide-hydroxide, etc etc. which makes the rust porous, allowing oxidation to continue.
http://en.wikipedia.org/wiki/Rust

Water doesn't generally react with ZrO₂, which is a very stable oxide. One finds it on filters in nuclear plants. Hyrdogen has been found in Zr oxides taken from spent fuel rods in hotcell.

Similarly, stainless steels don't really rust - unless somehow they contact some strong oxidance. Normally a layer of Cr₂O₃ prevents the formation of FeO and higher oxides or hydroxides.

 

[Astronuc]

I have heard from people connected to industry that some magnox/AGRs used zirconium pins while layering graphite blocks to help them stay in place. I guess they have probably completely oxidised and carried away in the gas stream.

Apparently the mass loss of some graphite blocks in magnox reactors is as high as 30%, do you think this is mainly due to standard oxidation due to the high temperature CO₂ rich environment or due to radiolytic oxidation?

As far as I know, the Magox fuel cladding is a magnesium alloy. AGRs have used stainless steel.

I believe the fuel in the steam generating heavy water reactor (SGHWR) used Zircaloy cladding.

 

[Freddy_uk]

I have heard from people connected to industry that some magnox/AGRs used zirconium pins while layering graphite blocks to help them stay in place. I guess they have probably completely oxidised and carried away in the gas stream.

Apparently the mass loss of some graphite blocks in magnox reactors is as high as 30%, do you think this is mainly due to standard oxidation due to the high temperature CO₂ rich environment or due to radiolytic oxidation?

All AGR's in the UK use st/st clad pins. http://www.nuclearsites.co.uk/resources/upload/Brochure%20-%20fuel%20manufacturing.pdf"

The graphite loss in the core structure (I'm no expert) i believe is down to its irradiation and the chemistry of the gas. Its not a pure CO2 atmosphere, other gases are introduced to prevent boiler tubes being coated with graphite, some of these gases have an adverse effect on the core. So its balancing core life with boiler efficiency.