[SorryDnoodle (OP)]

Hi again, I just got my tertiary alcohol from poland, finally and I tried a fairly small scale attempt at conjuring up some elemental potassium by mixing KOH with Magnesium turnings / Powder in separate attempts under mineral oil.

My results were not particularly satisfying..

First off was the powder, I put both reagents under oil at their correct ratios ( 2.4 Mg / 5g KOH)
Right after the flux I added the tert alcohol ( 0.4 ml), but nothing seemed to happend, I kept adding 0.1 ml every 10 minutes for about 45 minutes but nothing still was happening so this try was a complete bust, I cleaned up and got rid of the rests in a overly cautious way and pushed on.

This time turnings, same deal, added everything and at reflux the alcohol, but this is where I was surprised: all of the KOH had turned this dark red(purple when dried(?) almost bromine colored, I immediately stopped the experiment and did a flame test and everything else I could think of.

the flame test was very weakly purple indicating Potassium which was proven when I took up the big torched and burned with a higher temperature, at which point it started glowing and eventually kind of "popping" sort of when you would throw K in water.

The red/purple stuff melted rather quickly under a flame and had the look of molten sulfur if I had to guess, i had very small amounts to look at.

So, my question is: what's the purple stuff?

[chiralSPO]

How dry is your alcohol (and your KOH)? If there is enough moisture it will eat up the magnesium in unproductive side reaction. You can dry the alcohol with various desiccants, CaH2 or molecular sieves are probably the best options here, but anhydrous MgSO4 or NaSO4 could help--if you have enough Mg to spare, you might be able reflux the source of alcohol of the Mg turnings for an hour, and then distill it before use.) The flame test would show the presence of potassium in anything coming from your KOH, so I don't think that is the best tool here. How did you dry the dark material? I have seen potassium form a dark blue/purple oxide layer before, but that was over the course of months in a <5 ppm O2 glovebox... Your purple could also have been an organic sludge formed from impurities in the oil or alcohol reacting with the KOH... I say try it again!

[SorryDnoodle (OP)]

The KOH is dry technical grade, but my alcohol is from some company in Poland that I haven't bought from before so it might well be the problem.

But, could the oil be the problem?
I dried the purple stuff by just removing it from the oil, running it through some papers to absorb most of the water I had washed it with and letting it sit in a test tube, I was going to deal with it the next day but it started drying up on itself so I left it be to observe it.
Also an update on that, it seems to have reverted back to a white substance again, I assume KOH.


But thank you for the reply, I will try refluxing it and dry the alcohol and see if it helps, if nothing works I'll try buy some new reagents.
cheers

[alancalverd]

Not sure the KOH would be reduced by metallic magnesium anyway, and reactions between solids are depressingly slow at any temperature where alcohol is a liquid.

Better to dissolve the KOH in water, pass plenty of carbon dioxide through it, and dry off the resulting K2CO3 (or just buy potassium carbonate). Then mix that with finely ground carbon and apply plenty of heat. Somewhere around red heat the carbonate will melt and 2K2CO3 + C -> 4K + 2CO2 will do the business for you.   

The trick with metal extraction is to get rid of the bit you don't want, so you can drive the equilibrium in the desired direction. It's a lot easier to blow away some carbon dioxide than to persuade Mg(OH)2 to dissolve in alcohol.

[SorryDnoodle (OP)]

It will definently reduce KOH to K, but you are right, it is a very slow process.
But, the reaction you describe is done at very high temp, how till the K react at that temp?
In this one it floats to the top of the oil and coalecess into one single ball which is quite handy.

But anyhoo, I am mostly doing it because I think it's a cool reaction and I wanna explore it.

[alancalverd]

I've just watched a YouTube presentation and as I expected, the reaction takes place at a high temperature with molten KOH. You will have to wait for thousands of years to get a measurable yield under alcohol, and I suspect even that may not happen as there is nowhere for the MgO to go.

You seem to use "tertiary alcohol" and "oil" interchangeably. IIRC potassium is kept under a layer of hydrocarbon oil (no -OH group) and will react with alcohols.

[chiralSPO]

alancalverd, I believe the method being discussed is the one described in this link: http://www.strategicdefaultagent.com/produce-the-note/make-potassium-metal-catalyzed-magnesium-reduction-method.

[alancalverd]

Ah! Cool!

[SorryDnoodle (OP)]

I first found it here: http://www.sciencemadness.org/talk/viewthread.php?tid=14970

Then have a look at the patent, and Nurdrage video also helped, it's since been taken down but it can be found here: http://v.youku.com/v_show/id_XNTAxODAzMTc2.html