Oxidation of stainless steel 2

[tripog]

I am troubleshooting an issue with some things I don't quite understand and would like some input.
set up:
Stainless Steel item is placed under vacuum and heated up to 300 degrees Celsius after multiple argon purge cycles. When part comes out there is oxidation on it, I am being told this can only happen if it is exposed to o2 during the bake. The vacuum is held at 1x10-1 mbar with no rise or drop in pressure during the procedure. Nearly identical setup on another machine performs the operation as intended with zero oxidation, the only difference that I can detect is a little damage to a couple of TiAlN coated parts inside the setup that is producing items with oxidation.

So my question is, can anything other that o2 produce oxidation on stainless steel at 300 degrees Celsius?
o2 and h2o is less than .1 ppm during argon purge.
I am unsure of the stainless steel grade.

 

[EdStainless]

What about H2O? That will do it also.
Every surface is covered with a thin (molecular) layer of water.
Your vacuum level is only a rough vacuum, not even medium vac levels.
I take it that you are getting a light yellow/gold tint?
You can try heating to 150C and then doing repeat Ar fill cycles.
This will help remove moisture.
And I don't believe that you can measure H2O and don't see it spike during heating.
Every vac oven or furnace that I have ever run would do that.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

Thank you for the quick response.

H2o is supposedly less than .1ppm
Yes light yellow/gold tint color is correct
Would the 150C ar purge cycles have a better chance at removing moisture than 600c 48hour bake? that was done over the weekend.
Also besides h2o and o2 is there anything else that could cause the discoloration?

 

[EdStainless]

Bake the oven out at 600C with a couple of Ar purge cycles.
Keep it closed and under vac until you are ready to process parts.
In fact it should be closed and under vac all of the time.
The parts should be cleaned with solvent (usually fresh alcohol) and stored in a desiccator if possible.
One mistake is using alcohol or acetone that has been open and has absorbed moisture from the air.
This leaves more water on the surfaces.
We used to do work at 0.1micron pressures (10e3 lower than you) and we measured oxygen and H2O using mass spec.
How you report gas concentrations is tricky. 0.1ppm of what?


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

I am not part of the process about how the parts are handled prior to entering the oven, I have inquired about that, They did clean the inside of the oven with iso alcohol though. If I have to head back down to the site I will be traveling with a mass spectrometer. The readings for o2 and h2o come from the box where the purge gas is located.

 

[EdStainless]

Those reading could be the purge gas.
I presume that it is from liquid Ar?
We never wiped out chambers with solvents, just clean dry rags.
Even if you leave just a little isopropanol it is much harder to pump out because it is higher molecular weight.
So you mass spec has its own vacuum pump (differential pumped), correct?
You should have a sample port just outside of the oven in the line to the primary vacuum pump.
Is there some big reason why this is a problem?
The light gold tint is very thin.
You could strip it chemically (or electrolytically) if you just don't like the color.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

the purge gas is a house supply, the second unit is connected to the same gas without issue.
They took it upon them selfs to clean the chamber with iso, I recommend clean dry rags as well.
I have not had a chance to use the mass spec yet.
Yes there is a port i can tie into
I can not answer that question
Do you believe the gold tint to be chromium oxide?

 

[EdStainless]

Yes, it is Cr oxide.
But it is still passing light through it so it is very thin.
And the Cr depletion under it is very minimal.
Do you have a corrosion test that you need to meet?
Why on earth are you baking parts at 300C?


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

I am not the one baking these items,
I asked the same question about 300c and was unable to get an answer, the ovens go from 60c up to 600c

 

[EdStainless]

I certainly serves no metallurgical purpose.
No stress relief will happen at that temperature and no metallurgical transformation.
Unless they are curing a coating or adhesive I don't know why they would do it.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

The batch ran last night came out with a blue tint more than anything. A helium leak test showed no signs of any leak.

 

[EdStainless]

Then they went in either wet or dirty.
Again, the vac levels are not very good.
Honestly I would expect more like <10microns and you are >75microns.
Your vac isn't helping pull impurities, only repeated backfill and pump cycles can help.
And doing this warm (150C) will help a lot.
You need to test the Ar at point of use.
It had better have dew point <80C and <0.5 ppm O2 (V/V corrected to atmospheric pressure).
I am dubious of the <0.1ppm H2O, this is <-90 DP

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

I will be visiting the site tuesday, hopefully will be able to run some tests then.
the blue tint is from moisture? how is the blue different from the golden color?

 

[TugboatEng]

It formed at a higher temperature.

Is this a vacuum sterilization process?

 

[tripog]

I am not familiar with what they are using the ovens for, I only know that the oxidation is a problem for them. I am just trying to get a little more familiar with the chemistry and process so, I can be better prepared for troubleshooting. Though I do not believe this to be an oven issue, I believe their process is part of the issue, I am just trying to cover all bases.

 

[EdStainless]

the color is related to the thickness of the oxide. If it is bright (regardless of color) then it is thin enough that you are seeing through it. It will start as a pale straw, then golden, then more of an amber, pale blue, darker blue, and then the oxide usually starts getting dull and goes from gray to brown and black.
If they leave the ovens open other than just to load or unload it would make things worse.
If it is one oven more than the other I would be concerned about the vacuum lines and the gas supply lines.
A tiny leak in a supply line under pressure will allow O2 and H2O into the line.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[3DDave]

If I understand the optics correctly the yellow/gold color is from the removal of the blue component of visible light, leaving the green and red to make yellow. The way this happens is that the light enters the surface of the metal, between the atoms and is absorbed. So either that thickness is between 200 and about 450 nanometers thick or the crystal structure is or it is about half that to make a self-interfering filter.

If one takes gold leaf and shines a light through it the light that is transmitted is blue as it blocks the red and green.

Things I would try -

swap the Ar feed at the oven to eliminate the potential there is a contaminated supply line.
create test coupons from a single piece of metal and try identical cycles on the two ovens.
ask if the same people are running the parts. (I've had problems where one technician decided to "make an improvement" that, at the time, ruined parts. Good news was it was so terrible it was evidence for doing the opposite as an actual improvement. Remember this: "No one is in trouble. I don't care what you did, I just need to know what it was. Then we can fix it and move along.")

Whenever everything is the same but the results are different, then not everything is the same.

 

[EdStainless]

There is an old test that slipped my mind
You run samples of Cu, steel, and 304SS.
All must be bright and clean.
By comparing the colors you can make judgements about the atmosphere.
Old article in HT magazine which I can't locate right now.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[tripog]

@3DDave

the ar feed has been swapped from the box atmosphere to right from house gas, i'll have to see if a tank is avaliable although the other oven is on the same house gas and even though it too is fed from the box it is producing parts fine, well was, they did something to cause the oven to no longer hold vacuum.

the cycles are identical according to them on both ovens, and the parts run have been from the same lot/batch, I am not sure how strict their tracability is.

the people swapping is a great suggestion, i dont know when they run their parts but it is typically a one batch a day thing.

the boxes the ovens are attached to are a little bit different, as the machines running inside the boxes are different, but properly purged and evacuated I dont see how much difference that could make. On a side note, the previous oven that was in place of the one producing golden and blue parts was apparently always breaking down and having issues.


At what temp does the golden oxide layer become blue? and at what temp / o2/h2o levels does they oxide layers begin to form?

 

[tripog]

@EdStainless

that would be a great article to read if youre able to track it down at somepoint.