[Xuanny]

Say an extremely exothermic reaction takes place in a test tube (KClO₃ and glucose). To determine the energy released by the reaction, the test tube should be placed in water, right? (like in a typical calorimetry experiment) Is there a way to do this without breaking the test tube? Heating the water doesn't fix the problem; the tube still shatters.

[Borek]

The only reasonable advice - don't use glass.

[gluedudeguru]

what about a tempered test tube, something that costs a bit more but is harder to break and allows for the large swing in temp.  Perhaps Pyrex

[Borek]

Better approach will be something made of Hastelloy or something similar, at least from what I have read that's what is used for such purposes.

[Xuanny]

I've never heard of Hastelloy, but from their website, it seems they only make larger piping. Would a plastic test tube work? (It's not a good conductor, but neither is glass... right?) Could the heat lost to the tube be determined the same way heat lost to a calorimeter is determined?

[Borek]

http://www.helgroup.com/home/reactor-systems/safety.html?subpage=2

http://www.thermalhazardtechnology.com/spares/arc.htm

[KeithM]

I would have thought Pyrex test tubes would work but I'd assume you are already using them. Its the expansion coefficient of the glass that is the problem as much as the heat conduction.

So, use a test tube made out of clear fused silica, otherwise called fused quartz. As far as I can see they are not regularly available but here is a tube that would probably do:

http://www.hellma-worldwide.com/kuevetten/137/en/pg_id,42$g_id,18$item_id,209/cells-and-optical-elements-for-special-applications.html#209

Note the letters QS in the code, that means the Quartz option.

Otherwise find a local scientific glass worker and have him knock one up out of a bit of clear fused silica tubing. Shouldn't cost too much.
NOTE
Expansion coefficient of Pyrex is 3.2
Expansion coefficient of Pyrex is 0.5 !!