thermal radiation shields

Hi folks,
I'd appreciate any guidance I can get on the suitability of a couple of different choices of metals in a high temp inert atmosphere furnace application.
We are building a new series of small 1750 C furnaces for our instruments, to be used in measuring glass and metal properties in the molten state. The furnace, which is roughly 2.5" wide and 3 inches working length, is based around a graphite element, and that is surrounded by radiation shields, to insulate the core. We are using molybdenum sheet at the moment for the shields, but as you know this is very sensitive to oxygen, creating large volumes of MoO3 (?) "snow", and several other oxides, like dark violet Mo2O5. We have an inert gas purge of argon or nitrogen feeding the furnace, and we don't even switch on until we have well under 0.5% O2 present.
Still though, we get a trace of what can only be described as a soot formation. The soot seems to be evolved from the moly, and while iits is more an irritant than a serious problem, we wondered whether there was any good reason NOT to use Tungsten instead ? Are there "snowy" or "sooty" reaction products ?
We DID wonder whether we could actually use graphite as shields too. Does anyone have any comments on that idea ?
Thanks for any pointers.
Steve

steve wrote:
  Quoted Text Here  
Platinum just might take the temperature you cite. I remember use of platinum in oxygen containing furnaces as heating elements, but am uncertain of the max temp rating.
Graphite burns off, and helps to knock down the oxygen, but it will need replacement. Some grades are "dirty" with oxidizable inclusions.
An easily replaced "cheap" inner graphite radiation shield could be backed up by various metal shields which would be operating at lower temps, and would have fewer oxidation problems.
SiC type refractories may also be of some use.
My informal screening of refractory metals as heat shields 30 years ago, didn't turn up anything other than expensive platinum family alloys that would be good heat shields against oxygen at your temps. I am now uncertain as to how thorough that old search really was, however. Oxides tended to be poorly bound to the metal substrates, often powdery, and that led to contamination risks.
There aren't all that many oxides that cling tightly to their metallic substrates, or the use of metals at high temperatures in oxidizing atmospheres would be more common.

jbuch wrote:
  Quoted Text Here  
The oxygen we have is only "left over" from the purge - less than 0.5% during operation, so I rather hoped Tungsten might be acceptable.
Steve