Sciencemadness Discussion Board

Electrolysis of alkali metal halides in nonaqueous solvents

Lotilko - 4-8-2016 at 11:54

Greetings to every fellow member here. I watched NurdRage's video about making potassium and reagents cost here a fortune so I thought about using an alternative method. The idea is to use an organic solvent which dissolves the salt and does not react with the metal which collects at the cathode. I thought about using glycerol as a solvent and potassium iodide as the salt, however I have a few questions about this:

-Since glycerol is a trietary alcohol, does it react with potassium to form an alkoxide?

-Would graphite electrodes suit my needs?

-I know iodine is partially soluble in glycerol, so could it cause any problems during electolysis, like reacting with glycerol?

-What current should be used?

Thanks for any suggestions.


Metacelsus - 4-8-2016 at 14:22

Glycerol will not work as an electrolyte, as it will react with any alkali metal that forms. Also, it's not very conductive.

Propylene carbonate has been suggested as a good solvent for this, but as far as I know, nobody here has managed to get any good results with it.

WGTR - 4-8-2016 at 14:27

I don't know the answers to all of your questions. However, a solvent can work, provided that the salt ionizes in solution, that potassium forms a passive layer that is ionically conductive, and that the passivation layer isn't soluble in the solvent. I'm not sure you'll meet with success with potassium, but with the right solvent it's possible with lithium.

Cryolite. - 4-8-2016 at 16:56

I remember reading somewhere (I think BromicAcid's attempt at a sciencemadness home chemistry book) that nitrobenzene could work as a solvent for the electrolysis of sodium. I don't know of a reference, though.

DraconicAcid - 4-8-2016 at 17:50

No alcohol is going to work, and I'm pretty sure ketones and esters are out, too.

BromicAcid - 4-8-2016 at 18:52

Quote: Originally posted by Cryolite.  
I remember reading somewhere (I think BromicAcid's attempt at a sciencemadness home chemistry book) that nitrobenzene could work as a solvent for the electrolysis of sodium. I don't know of a reference, though.


You will find various references to this across the internet. The first time you read this the thought is 'sounds easy enough'. Then you dig, you find out that it's not sodium chloride that is the electrolyte but a melt of sodium chloride and anhydrous aluminum chloride. Not impossible but certainly not as easy. Then you find out that the mess that deposits at the anode is prone to exploding. At that point you write the whole thing off. I can dig up part of the reference, I think mine was from Gmelin.

Lotilko - 4-8-2016 at 23:57

Alright, thanks for everyone. I'll try a different method then. Have a nice evening. ;)

DraconicAcid - 4-8-2016 at 23:59

I don't think I'd want nitro-anything mixed with alkali metals.

Cezium - 5-8-2016 at 12:35

maybe some kind of solution of alkali metal chloride in one of these liquid chlorides as TiCl4, SnCl4, GeCl4, PbCl4, VCl4...

DraconicAcid - 5-8-2016 at 15:12

Quote: Originally posted by Cezium  
maybe some kind of solution of alkali metal chloride in one of these liquid chlorides as TiCl4, SnCl4, GeCl4, PbCl4, VCl4...


That won't work, because the titanium/tin/lead/vanadium is more easily reduced than the alkali metal. And that's *if* the alkali metal is even soluble in the liquid chloride to any great degree.

Cezium - 6-8-2016 at 10:09

Probably won't, there are oxidizing chlorides and will be reduced to lower ox. state like II. by oxidizing alkali metal back to chloride maybe even polychloride. This was just an idea.
Better will be mixture of various chlorides like LiCl+CsCl+KCl for lowering m.p.

clearly_not_atara - 7-8-2016 at 02:33

Maybe K2S in CS2 with K2CS3 as the electrolyte? It's really hard to think of a solvent compatible with potassium.

j_sum1 - 7-8-2016 at 03:22

Quote: Originally posted by clearly_not_atara  
Maybe K2S in CS2 with K2CS3 as the electrolyte? It's really hard to think of a solvent compatible with potassium.


I think that would smell bad.
And it might turn out that potassium trithiocarbonate is as difficult to obtain as potassium itself.

Interesting idea though.

Dan Vizine - 9-8-2016 at 10:45

Where do you guys live?

If you're making these because you simply can't buy them where you live, I guess these threads make sense. Or, if you really think you'll devise a useful new preparatory route worthy of note.

But getting lithium, sodium, potassium or cesium is as easy as going to GalliumSource's website in the US.

Judging from the endless stream of users seeking to make active metals by new, easy, and largely mystical routes, I don't expect this topic to go away anytime soon, but I wouldn't spend 15 minutes on it.

It all boils down to one useful maxim...If you are an actual chemist and you can buy something more cheaply than you can make it, don't make it. Find a project where you can make things that you can't easily purchase.

elementcollector1 - 9-8-2016 at 10:49

Quote: Originally posted by Dan Vizine  
Where do you guys live?

If you're making these because you simply can't buy them where you live, I guess these threads make sense. Or, if you really think you'll devise a useful new preparatory route worthy of note.

But getting lithium, sodium, potassium or cesium is as easy as going to GalliumSource's website in the US.

Judging from the endless stream of users seeking to make active metals by new, easy, and largely mystical routes, I don't expect this topic to go away anytime soon, but I wouldn't spend 15 minutes on it.

It all boils down to one useful maxim...If you are an actual chemist and you can buy something more cheaply than you can make it, don't make it. Find a project where you can make things that you can't easily purchase.


Personally, I think it depends on preference. I'd much rather have fun figuring out endless ways not to make sodium than just buy the stuff. Amateur chemistry, and all that.

DraconicAcid - 9-8-2016 at 10:59

I don't know if an ester would work, but if you made ethylene glycol diacetate, that might work better than other esters.

Dan Vizine - 9-8-2016 at 11:08

You may have missed my larger point, elementcollector. My point wasn't to buy things rather than make them, it's why make something so pedestrian? Put that same effort into trying to make a novel product.

Admittedly, the mental stretch of selecting that novel product is often the hardest part.

[Edited on 9-8-2016 by Dan Vizine]

Lotilko - 9-8-2016 at 15:14

Quote: Originally posted by Dan Vizine  
Where do you guys live?

If you're making these because you simply can't buy them where you live, I guess these threads make sense. Or, if you really think you'll devise a useful new preparatory route worthy of note.

But getting lithium, sodium, potassium or cesium is as easy as going to GalliumSource's website in the US.

Judging from the endless stream of users seeking to make active metals by new, easy, and largely mystical routes, I don't expect this topic to go away anytime soon, but I wouldn't spend 15 minutes on it.

It all boils down to one useful maxim...If you are an actual chemist and you can buy something more cheaply than you can make it, don't make it. Find a project where you can make things that you can't easily purchase.


If you live in the US, then good for you. There is a difference between an "actual chemist" and a "backyard scientist". Yeah, sure you can buy a shit ton of chemichals and exotic metals in the glorius west, but unfortunately I don't have acces to these kinds of materials because I live in eastern europe. Why bother to make a chlorate cell when chlorates are commercially available. The thing is, it's either restricted, too expensive, or just out of reach, or a combination of these. So why not make potassium? You say you wouldn't spend 15 minutes on a project like this, but I am willing to spend coutless hours researching and conducting the final experiment, because I find it interesting. Don't tell me you didn't try anything new before.

Dan Vizine - 9-8-2016 at 16:42

You sound like you think I'm deriding where some of you live. I wasn't. Many, perhaps most, users don't seem to give their countries of origin (just look at this thread) and the question was serious.

I'm often surprised when I find that the US has easier access to "stuff" than other countries, since our laws are oppressive, and getting worse, as everyone wets their collective pants about the 0.001 % chance that a big bad terrorist might jump up and bite 'em.

If you can't buy something in your country, try China. They'll sell you anything....white phosphorus, thallium by the kilo. Only reasonable thing they don't seem to sell is radioactive stuff AFAIK.

But you also missed the point...."IF" you can easily buy something cheaper, why make it? If you can't, and apparently you can't, then you have little choice.

And yes, I've been known to try a few new things, but that's part of the job description if you work as a research chemist.

Good luck, I hope you succeed!

Oh...enjoyed your joke about the "Glorious" West.

[Edited on 10-8-2016 by Dan Vizine]

Lotilko - 10-8-2016 at 03:46

I didn't think that, I was just pointing out that I don't have access to metallic potassium. I want to try NaK, but I can only get sodium at the moment. Good luck on obtaining black phosphorus and thanks.

clearly_not_atara - 14-8-2016 at 19:37

Diglyme might work as a solvent, actually, come to think of it; it chelates alkali metal cations so they tend to be soluble in it. At the same time it's extremely unreactive with bases.

You do need an anion that doesn't generate reactive byproducts when it's oxidized, though. Sulfate and chloride are right out. Consider electrolyzing sodium formate in diglyme, maybe? Assuming formate decomposes like this:

HCO2- >> HCO2* + e-

2 HCO2* >> (HCO2)2 "formyl peroxide"

(HCO2)2 >> CO2 + HCO2H

[Edited on 15-8-2016 by clearly_not_atara]

Dan Vizine - 23-8-2016 at 14:50

My guess would be that diglyme, being an ether, is unlikely to dissolve enough anhydrous sodium formate to allow electrolysis to occur, despite the tendency toward chelation.

Your idea is interesting. The chemistry of formates is truly a fascinating area. A number of the simplest members of different homologous series are special, but CO2 makes the carboxylic acid series fairly unique.

[Edited on 23-8-2016 by Dan Vizine]

NEMO-Chemistry - 24-8-2016 at 09:48

Wasnt NAK once used as a coolant for nuclear reactors? It had a kind of non moving parts pump??

I am sure i have seen a video about it somewhere, i need to file things in my brain in a more orderly fashion!