LMNts - 23-9-2018 at 04:50
This weekend I have been trying to make potassium metal using the procedures in the famous US4725311A patent.
The first attempt was with 30g magnesium shavings and 100g potassium hydroxide flakes (I know, way too much) in 250 mL mineral oil. The catalyst,
tert-butanol, was added at the start of the experiment. After about 10 minutes hydrogen started to evolve vigorously because of the water in the KOH
reacting with the Mg. After that tiny spheres (~1mm) of potassium became visible floating in the oil. After some more time the spheres disappeared and
never appeared again for the next 5 hours until the experiment was stopped.
The next attempt was with 15g Mg and 30g KOH. This time the tert-butanol was added after the solvent was hot. The mixture was heated for 8 hours and
no potassium was observed.
At this point I had ran out of mineral oil and switched to kerosene as solvent. The kerosene had a low aromatic content so should be usable. Both
previous described experiments were repeated with kerosene and again tiny spheres of potassium were observed if the catalyst was added at the start
and again they vanished after some more time.
Each time a vigorous evolution of gas was observed a few minutes after heating the mixture indicating the potassium hydroxide contains water. Could
this be a problem?
Each time the solvent remained completely clear after the experiment so no dark decomposition product were formed.
The whole reflux setup is sealed and the only gas escape is via a gas bubbler so it is not likely oxygen is getting in.
Does anyone have any idea what might be going wrong?
metalresearcher - 23-9-2018 at 07:35
Dry the KOH first, by heating it to 500 C in a (stainless) steel vessel for some time, e.g. half an hour.
Let it cool down and crush it and keep away from any moisture source.
SWIM - 23-9-2018 at 08:34
DanVizine has suggested elsewhere that just distilling off some of the solvent (was it like 10%?)after the Koh is added should dry the hydroxide well
enough.
This would be done before the catalyst is added of course.
LMNts - 23-9-2018 at 09:12
So it is probably the water content that is ruining the reaction?
Shouldn't the magnesium react with all the water during the hydrogen gas evolution phase? The patent states that they use wet potassium hydroxide with
success.
Drying the potassium hydroxide is not a problem for me as I can easily heat it to 1000 °C in a furnace but the resulting chunk may be hard to crush.
Any ideas how to do this? Just use a hammer?
I think another possibility for drying the potassium hydroxide is adding potassium metal to the mixture, just like NurdRage is doing with his sodium
synthesis. The problem is that I don't have potassium yet...
12thealchemist - 24-9-2018 at 04:58
Instead of adding potassium, you could add lithium - from ultimate energiser batteries. I've been following NurdRage's sodium production videos
closely, and using battery-lithium was his final solution for circumventing start-up problems.
LMNts - 24-9-2018 at 05:14
Thanks for the suggestion!
I don't have lithium available yet but I do have a small amount of sodium metal. I only want to use this as a last resort as it will contaminate the
product.
Currently I'm trying to make potassium oxide (K₂O) by heating potassium nitrate to 1000 °C. No detonation has occurred so far 
. Mixing this in excess with wet KOH should result in dry KOH, right?
MrHomeScientist - 24-9-2018 at 06:04
I never had a problem with "wet" hydroxide. KOH has about 15% water "built in" that is nearly impossible to remove, except, I guess, by the fusion
route suggested. But it will immediately start sucking up water as soon as it cools so you'll never get it completely dry. Distilling off some of the
solvent is a good idea - use a bit excess solvent, add the KOH, distill until the desired solvent level is reached, then add the rest of the
reactants. If you can do it easily, I guess drying is worth a try if only to eliminate that variable. But again, I never had any problems using
unprocessed KOH.
This reaction is extremely finicky and there are a whole host of issues that have come up for people. Seemed like the most common one was bad quality
magnesium (too coarse, too oxidized, etc.). I recommend reading the Make Potassium thread. Yes, I know it's long. But that just means it's packed full
of great experiments. If you want to skim past some of the theory, look for posts with pictures.
12thealchemist - 25-9-2018 at 02:18
I've summarised the theory on that thread, however, if you wish to read that but have no desire to trawl through the entire thread.