From Matt Heizler
Hi all,
I'm sure my list is not complete, but this is a start at getting
information to those of you who have a furnace from modifications
limited. The price and availability of Ta is becoming quite a
problem for Ken Heine and thus he is trying to break into the moly
crucible market. In order for him to do so, he requires some volume
and standardization. Eventhough we as a group will probably never
agree on an age of a standard, perhaps we could come close on a
crucible design which could work for most of us.
There are lots of designs out there, but for the most part they are
quite similar - a tube of metal gets hot and degasses a rock or
mineral.
We here in New Mexico have been using the attached design for about 6
years. Thus far I have been working with Schwarzkopf to get the moly
tube machined and then Applied Fusion to get the 2 3/4" conflat
attached.
During the past 17 years I have used lots of Ta and Moly crucibles
and based on this I am quite confident that moly makes for a much
better crucible. Unlike reactive Ta, moly crucibles never fail from
corrosion, however liners do get stuck sometimes. We went to the
larger 3/4" crucible to make it easy to remove the liner when the
ring of crude (mostly copper) that precipitates high up in the
crucible kept preventing liner removal from our 5/8" crucibles. In
order to do this the Modifications Lmt. heat shields required minor
changes.
Here's what I know about our crucibles.
1. The argon blanks are very good. I have attached a blank curve
which is typical for our lab.
2. Durability is very good. Primary failure mode is a stuck liner.
3. Thermocouple temperature versus sample temperature is generally
off-set by about 50°C. I do not understand this, but I guess there
must be a significant vertical temperature gradient between the tip
of the thermocouple and the sample. We calibrate our temperature via
copper melting experiments and kinetic data on K-feldspars suggests
that the temperature off-set is linear.
4. Time for temperature equilibration of thermocouple and sample is
significant. Our Eurotherm parameters allow the thermocouple to
reach set point in about 2 minutes, however the sample requires
approximately 2 more minutes. When doing diffusion studies I try to
do pretty long heating steps at low temperature to minimize
uncertainty for calculating diffusion coefficients associated with
uncertainty related to heating and cooling of the sample relative to
set point temperature.
5. Cost. Depending upon volume, a complete crucible costs about
$500 to $600. If Ken Heine could not be too out of line on this cost
I'd much rather call him to get a crucible than go thru the ordering
myself. Also, I'd rather see Ken hold the inventory cost rather than
me.
Drawbacks? Empirically I haven't found any, however a couple of you
have mentioned to me in the past that Moly has a high vapor pressure.
I haven't found this to be a problem that I know of for argon
analysis, but I could make some specific measurements to quantify
total pressure versus temperature the next time we change liners.
Standardization? We are not married to any specific design with
respect to crucible wall thickness, liner length, or overall length.
We originally kept the liner short in thinking it could help our
temperature control, but in light of what I discussed in #3 above it
probably doesn't matter a great deal. Overall length fits with our
current power feedthrough - furnace element - thermocouple geometry,
but relatively small changes to the inside of the furnace could
accommodate a length change. We will stay with the 3/4" crucible as
this is a major advantage relative to the 5/8" no step-down inside
diameter design. It would be simple to design both a 3/4" and 5/8"
crucible which accommodates the same liner diameter.
If you see missing addresses for users who might be interested in
this please forward my message to them and let me know who they are.
If Ken Heine can't get into the moly crucible business he may go out
of business all together. Regardless, Ta is going to become very
expensive and simply based on this you may consider using moly if you
aren't already.
Hope all is well out there. Cheers,
Matt
|