Sciencemadness Discussion Board

Electrolysis of MgCl2 (a realistic objective?)

Swinfi2 - 13-10-2018 at 21:14

As a result of nurdrage's hard work with the sodium production reaction I've been looking into how magnesium is made. The common industrial routes afaik are the silicothermic pidgin process (at 1000c) and electrolytic process (at 700+c).

Just going out on a limb here I'll say those are out of reach for all of us so I've been looking for a workaround.

So far I only see one way to bring this down to reality.. by exploiting a eutectic mixture. It would seem the most convenient method would be anhydrous electrolysis of 2AlCl3.MgCl2 which melts around 275c under an inert atmosphere to grow a magnesium electrode with concurrent production of useful chlorine.

Attachment: einarsrud1987.pdf (1.1MB)
This file has been downloaded 611 times

Now I don't currently have the means or the minerals to try this (yet) but does it seem possible or too good to be true?

[Edited on 14-10-2018 by Swinfi2]

symboom - 13-10-2018 at 22:15




Sounds like the FFC Cambridge process only oxides of metals are used instead. But your process sounds interesting that's going to be some hot chlorine gas make sure the metal can handle hot chlorine gas at that temperature. Coppernickel is some chemical resistant metal and tin is stubborn to dissolve in HCl.

Magnesium sulfate and calcium chloride
Magnesium chloride HCl and aluminum metal.

https://en.m.wikipedia.org/wiki/FFC_Cambridge_process

[Edited on 14-10-2018 by symboom]

[Edited on 14-10-2018 by symboom]

Whenyou - 13-10-2018 at 22:18

Hmm. Having a look at the table of standard reduction potentials, I'm pretty sure that using a mixture like that will produce aluminium on the cathode rather than magnesium. The potential for Al reduction (-1.662v) being a fair bit higher than that for Mg reduction (-2.372v). At best, I'd assume that you'd possibly get a mixture of aluminium and magnesium as a product rather than pure magnesium. Though it's much more likely that you'd get only aluminium, that 0.7 volt gap would be pretty difficult to overcome.

Aside from the AlCl3-MgCl2 mixture, the only other eutectic mixtures I can find that would work melt somewhere above 400C, so possibly Mg(OH)2 (melting at 350C) is the best you could do.

Do you have anything stating that this mixture has been used for electrolytic magnesium production?

[Edited on 14-10-2018 by Whenyou]

Swinfi2 - 14-10-2018 at 00:12

@symboom the FFC camebridge process was interesting as I'd not seen it before, I thought generally metal oxides were insulators (as mostly covalent structures) like PbO2 is iirc conductive because of deficiencies but Al2O3 is an insulator as I'd expect. How do you predict that?

@whenyou I did find this however as it's in relation to organic solvent systems not molten salts so I'm not sure how transferable it is. They claim up to 100% current efficiency :o

Attachment: acsenergylett.7b00269.pdf (1MB)
This file has been downloaded 578 times

Ubya - 14-10-2018 at 01:34

Quote: Originally posted by Whenyou  
Hmm. Having a look at the table of standard reduction potentials, I'm pretty sure that using a mixture like that will produce aluminium on the cathode rather than magnesium. The potential for Al reduction (-1.662v) being a fair bit higher than that for Mg reduction (-2.372v). At best, I'd assume that you'd possibly get a mixture of aluminium and magnesium as a product rather than pure magnesium. Though it's much more likely that you'd get only aluminium, that 0.7 volt gap would be pretty difficult to overcome.

Aside from the AlCl3-MgCl2 mixture, the only other eutectic mixtures I can find that would work melt somewhere above 400C, so possibly Mg(OH)2 (melting at 350C) is the best you could do.

Do you have anything stating that this mixture has been used for electrolytic magnesium production?

[Edited on 14-10-2018 by Whenyou]


standard reduction potentials are exactly that, reduction potentials at STANDARD conditions (25°C 1atm 1mol/L), so in conditions as described by swinfi2 magnesium could have a higher reduction potential


[Edited on 14-10-2018 by Ubya]

Whenyou - 14-10-2018 at 02:51

Quote: Originally posted by Ubya  

standard reduction potentials are exactly that, reduction potentials at STANDARD conditions (25°C 1atm 1mol/L), so in conditions as described by swinfi2 magnesium could have a higher reduction potential


Sorry, I should have been a bit more specific. Magnesium could certainly have a higher reduction potential in the molten mixture. I simply assumed that seeing as the post had no evidence for the mixture's use in magnesium production, it would be more likely that the 0.7 volt gap between the potentials wouldn't change drastically to the point where pure magnesium is reduced. If it were easy enough to set up, it would be a worthwhile test, though it probably wouldn't be worth the effort to make the setup for high temperature electrolysis in an inert atmosphere.

If swinfi2 is intent on magnesium production by electrolysis, then testing the eutectic mixture would be the first step anyway, and if it doesn't work then they could switch the molten salt to that of Mg(OH)2, which is guaranteed to work (with a proper setup at least), though it will require a higher temperature.

streety - 14-10-2018 at 13:09

I've also been following Nurdrage's sodium production efforts with interest. Back in the beginning when glassware was being destroyed and the reaction was taking a shocking amount of time I began to wonder whether it wasn't all more trouble than the electrolytic approach.

The temperatures are high but they are achievable in an amateur setting.

Is your interest in making magnesium for its own sake or as a reagent to make sodium?

Fulmen - 14-10-2018 at 13:42

Quote: Originally posted by Whenyou  
Mg(OH)2 (melting at 350C)

Nope. Magnesium hydroxide does not have a melting point, it decomposes to magnesium oxide at 350°C

Swinfi2 - 14-10-2018 at 14:16

My interest in (home/amateur) chemistry is kinda from a doomsday prep perspective, having the knowledge and skills to do this stuff after "The Apocalypse".

Magnesium is too useful as a metal to loose in this scenario so if it can be made under mild conditions with high efficiency it would make it accessible to anyone who can get Epsom salts. So both really, direct NaCl electrolysis is simpler but would it be as efficient?

That and something just doesn't sit right with me and being anywhere near molten flammable metals.

streety - 14-10-2018 at 16:04

Doesn't NurdRage's protocol also use molten sodium? Or, potentially worse molten sodium in flammable liquids?

Given your interest, how are you planning to source your precursors?

Whenyou - 14-10-2018 at 16:45

Quote: Originally posted by Fulmen  

Nope. Magnesium hydroxide does not have a melting point, it decomposes to magnesium oxide at 350°C


Ah, damn. That'll teach me to trust a simple google search. Possibly a mixture of MgCl2-CaCl2-NaCl might be possible, melting at somewhere around 420C, though I'd imagine >400C is getting very difficult to set up. I don't know of any other ionic magnesium compounds/mixtures that will melt at a lower temperature than that though, possibly someone else might.



Attachment: MgCl2-CaCl2-NaCl.pdf (827kB)
This file has been downloaded 689 times

symboom - 14-10-2018 at 18:55

Calcium would be a better alternative

Henri Moissan first isolated pure calcium in 1898 by reducing calcium iodide with pure sodium metal

CaI2 + 2 Na → 2 NaI + Ca
Calcium iodide can be formed by treating calcium carbonate, calcium oxide, or calcium hydroxide with hydroiodic acid.


Bonus sodium and calcium iodide is used
Calcium crystals are also formed in this reaction
So nurdrage demonstrated sodium is possible next would be calcium first by fire then by slow chemical reaction.

Calcium iodide is soluble in acetone and alcohols
Substitution of sodium hydroxide with calcium iodide
Could work.

[Edited on 15-10-2018 by symboom]

Sulaiman - 15-10-2018 at 01:34

Magnesium is such a useful metal that it is likely to be manufactured for as long as we live,
mainly from magnesium chloride.

Due to economies of scale, magnesium metal is not particularly expensive
https://www.ebay.co.uk/itm/1200g-Magnesium-metal-ingot-block...
https://www.ebay.co.uk/itm/5Pcs-Magnesium-Rod-High-Purity-99...
Magnesium metal can also be recycled from many consumer products.
Almost all commercial magnesium chloride is 'dead sea salt' or similar
https://en.wikipedia.org/wiki/Dead_Sea_salt
so refining/isolating/separating will have to be performed at some stage(s).

If anyone survives Armageddon there will be relatively unlimited resources for re-use, refining and recycling available above ground.
Just try to avoid the zombies.

Melgar - 19-11-2018 at 07:59

I agree that magnesium metal is easy enough to get that making it seems like your effort could be spent elsewhere.

I think it's kind of interesting how they produce commercial-grade MgCl2 and KCl though. They pump brine into a flat salt basin, and the water gradually evaporates. The different salts precipitate out based on their solubilities, and they're collected and separated based on what concentration they crash out at. This sometimes happens naturally in inland seas that later become salt mines, and different layers of a salt mine can be different salts.