Notes on home-made tcea crucible fabrication:

-McMaster.com P/N 9121K71: Very fine grain graphite rod, 1/4" dia. x 6", enough for four crucibles. Current cost is $7.47 per rod, so ~$2 per crucible (Thermo wanted $48.33 each in 2016). Max claimed T is 800F (427C) in air, so helium brings 1450C. Or that's the idea we hope. This very fine grained graphite is surprisingly tough and machineable.

SUMMARY: drill a 3.5 mm dia. hole 29 mm into the rod end. Cut the rod to 31 mm length. Repeat twice and trim the stub for 4 pieces per rod. Mill the drilled cylindrical rod pieces into perfect 4.5x4.5x31 mm square prisms, centered perfectly upon their respective holes. Bob's your uncle.

TRIGGER WARNING: To a proper machinist, the following may extend beyond cringeworthy.

-Wrap a 9/64" drill bit at 29mm with a bit of tape and drill the rod with a lathe to that depth. The small lathe in Watkins' lab works well.

-With the rod still in the lathe chuck, use a 1/4" or larger drill bit (hand held to truly expedite matters) to lightly bevel the interior end of the hole just drilled. A funnel shaped opening, of sorts.

-Fabricate a 31 mm length exterior gauge (e.g. from a paper clip :o) and with the rod still in the lathe chuck, cut the rod with a hand held hacksaw blade to 31 mm length

-Repeat twice and trim the stub for 4 pieces per rod.

-Plane flat sides by hand with a flat bastard file. Find a small machine vise - as at the drill press in Watkins' lab. Make a shim from sheet metal slightly thinner than the finished crucible, erring toward too thin rather than too thick; here we used 8 ga. aluminum @ 0.129 in. This is for adjusting the height of the crucible in the vise. It should have two heights available, one with the crucible lower for filing down sides one and two, and one a bit higher for sides three and four. **Clamp crucible securely enough to not spin while filing, but DO NOT OVERCLAMP (because if crunching sound is heard, may thence crumble.)**  Shim height is arrived at with a fine flat file, by trial and error.

** Once the first side has been filed flat, it and subsequent flat sides are then indexed on the face of either vise jaw - so for the remaining three sides it will not spin even with the very light pressure from the vise. We predict that we will never again crunch a crucible. Or that's the idea we hope. We would have not thought of this on our own - again credit to tsa.uoregon.edu.

These inexpensive units appear to perform fabulously.

Update: While the factory units have a square cross-section, it's not necessary to remove that much material for the home made units. The idea is to allow the carrier and sample gases to flow easily past the crucible. File away enough material to make eight sides, with four each alternating flat and curved surfaces; an "octagon" of sorts. The slightly larger crucible cross-section also will provide less chance of sample solid residue falling outside and past the crucible.