[lmenwe]

Naphthalene is stirred constantly with thermometer so that temperature will not fluctuate during the cooling of naphthalene. The experiment is heat the naphthalene until it is melted and wait for it to cool down to become solid. I need to determine the freezing point. Why I need to stir the naphthalene constantly with thermometer? Why if not stir constantly the temperature will fluctuate?

[P]

I guess this is to avoid 'locallised' areas of cooling -   e.g cooling faster at the surface where evapouration occurs - or faster cooling at the sides of the flask where you have contact to the ice bath - i.e. the liquid in the middle of the flask may end up slightly warmer (convection currents will average this out eventually, but not imeadiatley).   If your measuring the exact freezing/melting point, then you want to make sure your reading is accurate and representitive of the whole body.

This might be an interesting (or at least informative) experiment:

do your cooling in a flask that is large enough.  Clamp in 3 or 4 thermometers - one just below the surface, one in the middle of the sample, one right near the edge of the vessel and one right at the bottem.  Don't stir and observe the readings on each thermometer as the sample cools to freezing from warm.   You won't see that much of a difference, but there will be enough to confuse a freezing point expt. I would bet.

[Yggdrasil]

The stirring also prevents for the formation of a supercooled solution, that is a solution that is liquid despite being below its freezing point.  The strirring disturbs the solution so that nucleation (http://en.wikipedia.org/wiki/Nucleation) can occur and it will freeze when it reaches its freezing point.  Without stirring, nucleation may not occur until the solution is well below its freezing point.