I have been warned that generally the anodic oxide layer is a porous,
hygroscopic material that is a poor coating for high vacuum service,
due to its high outgassing rate. Yet, anodized aluminum does seem to
be necessary for certain vacuum applications such as systems with
corrosive gases/plasmas. Is there a type of anodization that would be
a practical (but perhaps imperfect) coating for high vacuum service?
For example, would a thin (0.0001"/2.5 micron), sulfuric acid clear
(Class 1) anodized coating be o.k.? Which would be best, low-temp.
hard-coat (Type III) or room-temp standard (Type II)? Should the
coating be sealed and if so, by what method, hot water? Or does the
sealing process just trap water that will later outgas?
Altenatively, has anyone tried the Cerafuse (Microplasmic) aluminum
anodization process on parts for high vacuum service? Any problems
with outgassing? Other problems?
Thanks,
Grant
Unless specified otherwise by your client, I'd go for an oxalic/sulfuric
anodize coat, A-8625, type 3, ~ 1.5-2 mils, followed with a NiAc2/Cr2O7
sealing step, then dehydration by baking for 12 hrs at 375 +/- 25°F.
Any possible inclusions, defects, outgassing will show up as spalling
in the coating.
Take care,
hanson
"Grant Kiehne" <grant_...@hotmail.com> wrote in message
news:817cdbd2.0310...@posting.google.com...
You suggested:
> If you are not happy with anodic oxidation, you should give ANOF a
> try. Thats a german abbreviation for a method best translated as
> "ANodic Oxidation applying Spark Discharges". These sparks with their
> high temperature cause a very hard alpha-alumina layer (nearly
> ceramic) having no pores.
I've done some preliminary investigation into this method. A start-up U.S.
company called Microplasmic (www.microplasmic.com) has a patent on such a
process (I found it on www.uspto.gov). It is commercially available under
the trade name CeraFuse from Ceramic Coatings Technologies
(www.cerafuse.com). The process yields a hard, dense alpha-alumina layer
covered by a gamma-alumina layer that is buffed off. With buffing only, the
surface of the alpha-alumina is quite rough (~120 microinch), but it can be
diamond grit polished like a piece of sapphire down to the desired
roughness. I am interested in ~32 microinch or better. Reportedly, there
are some pores due to the micrplasmic discharge (sparks).
Reportedly, the CeraFuse process has been applied to aluminum vacuum chamber
surfaces, although it is not clear if one can consider it suitable for high
vacuum or uhv application.
Any academic or commercial contacts you can provide would be helpful. Might
there be non-U.S. vendors who can apply a similar process? I wonder if the
Microplasmic process is patent-protected outside the U.S.?
Cheers,
Grant
"Ansgar Kursawe" <A.Ku...@nurfuerspam.de> wrote in message
news:la9npvc8gd8b4nfr6...@4ax.com...
> On 25 Oct 2003 06:30:32 -0700, in sci.chem.coatings you wrote:
>
> Hi Grant,
>
> >I have been warned that generally the anodic oxide layer is a porous,
> >hygroscopic material that is a poor coating for high vacuum service,
> >due to its high outgassing rate.
>
> Correct.
>
> >Yet, anodized aluminum does seem to
> >be necessary for certain vacuum applications such as systems with
> >corrosive gases/plasmas. Is there a type of anodization that would be
> >a practical (but perhaps imperfect) coating for high vacuum service?
>
> You should perform the anodization with your required layer thickless,
> then followed by a sealing of the pores with superheated steam. To
> prevent water being trapped, I would heat the whole part up to 380K
> while treating it with steam.
>
> >Should the coating be sealed and if so, by what method, hot water?
>
> I would seal it, definitely. And I would make a small test with an
> electrically heated surface while steaming it. If the sealing is still
> successful, the amount of water being trapped should be much lower
>
> Unsealed anodized alumina is a very good catalyst
> support material because of its high inner surface area. Its surface
> is strongly acidic and it will definitely soak with water when exposed
> to air. You should pay good attention to the electolyte used for
> anodization: H2SO4 as electrolyte will cause very narrow pores of
> roughly 5-10 nm, oxalic acid will have approximately 20-40 nm pores
> and phosporic acid will have approximately 100. The former will have
> the highest inner surface area, the latter the lowest and the oxalic
> is somewhere between both. Therefore, I'd suggest (for your
> application) you give phosphoric acid a try.
>
> If you are not happy with anodic oxidation, you should give ANOF a
> try. Thats a german abbreviation for a method best translated as
> "ANodic Oxidation applying Spark Discharges". These sparks with their
> high temperature cause a very hard alpha-alumina layer (nearly
> ceramic) having no pores. I've once got a layer at my former
> university for catalytic tests. The layer was inert, so I wanted to
> get the catalyst by cooking / dissolving it with diluted HNO3. It was
> resistant to diluted (20%) and hot HNO3 for 48h. If you need more
> information about this method, I can give you a very competent contact
> person at my former university. He has forgotten more about aluminum
> and coating methods than I have learned in my life ;-)
>
> Greets from Germany and kind regards,
>
> Ansgar
> --
> The two most common things in universe are hydrogen and bureaucracy.
>
>> I have been warned that generally the anodic oxide layer is a porous,
>> hygroscopic material that is a poor coating for high vacuum service,
>> due to its high outgassing rate.
>
> Correct.
>
>> Yet, anodized aluminum does seem to
>> be necessary for certain vacuum applications such as systems with
>> corrosive gases/plasmas. Is there a type of anodization that would be
>> a practical (but perhaps imperfect) coating for high vacuum service?
>
> You should perform the anodization with your required layer thickless,
> then followed by a sealing of the pores with superheated steam. To
> prevent water being trapped, I would heat the whole part up to 380K
> while treating it with steam.
Except for thickness, what are the important differences between the surface
of aluminum oxidized in air with respect to anodized aluminum? From what I
understand, clean aluminum surfaces in air are immediately oxidized.
Bill
http://www.ahc-oberflaechentechnik.de/ahce/home-neu.htm
http://www.keronite.com/license.html
Grant
"Grant Kiehne" <grant_...@hotmail.com> wrote in message
news:CdPmb.20098$Ye3....@newssvr31.news.prodigy.com...
Grant Kiehne wrote:
>Does anyone have experience using anodized aluminum parts in a high
>vacuum or ultra-high vacuumm (UHV) system? I am investigating using
>some relatively large-area anodized aluminum parts in a system that
>must achieve ~5x10^-8 Torr (cool) after baking at 135 deg. C for ~12
>hrs. The alloys are 6061-T4 and 6063-T5. I am familiar with the
>standard preparation of aluminum for vacuum service of machining and
>hot basic solution etch, followed by hot air bake. This yields a
>fresh native oxide on the aluminum surface. My application requires a
>thicker oxide layer (microns).
>
Generally one uses stainless steel for high vacuum. Al might be ok for 10-8
josh halpern
Could you help me understand this?
TM
--
Toadmonkey: "Now now. Brain popping and world crashing may be hazardous to ones perception of reality.
Very dangerous business that can lead to madness or something worse for some, truth."
Remove "3+4" from addy before replying
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----
"toadmonkey" <html5...@7yahoo.com> wrote in message
news:3fa3989b...@news.vcn.com...
>Atmospheric pressure on the ground near sea level is ~760 Torr. Pressure
>decreases with altitude. --Grant
>
Thank you Grant!
A hard vacuum system is not about survival, it is about outgassing.
Aluminum is notorious for outgassing hydrogen. Forever. Large scale
electrolytic aluminum production typically sparged elemental chlorine
through the molten product to purge hydrogen by reaction. Castings
would otherwise solidify with serious voids.
One encounters a similar problem with copper and silver. The molten
metals are incredible sponges for oxygen. Thomas Edison is said to
have been bothered by a copper king at a party. He asked for a nice
trophy - a cast cubic foot of copper weighing its proper theoretical
density. If you can get cast copper to 90% of its theoretical
density, you are having an exceptionally fine day (or do vacuum
casting). Quality copper objects are usually forged.
--
Uncle Al
http://www.mazepath.com/uncleal/
(Toxic URL! Unsafe for children and most mammals)
"Quis custodiet ipsos custodes?" The Net!