Sciencemadness Discussion Board - Electrolyzing CaCl2 with DC welder

Electrolyzing CaCl2 with DC welder

metalresearcher - 17-5-2015 at 11:04

I tried to electrolyze some CaCl2 'drying agent' in a steel crucible connected to the (-) so it acts as the cathode and a carbon rod connected to the (+).
I used a 200 amp (set to 100 amp) DC welder as power supply. I started arcing to start the initial melt to get it conductive. Finally I saw a red hot liquid bubbling, sometimes it arced underwater ('water' which is actually liquid CaCl2) due to the louder bubbling with the sound of a submerged arc.
Sometimes I saw orange blinks which are actually Ca metal which then immediately burned again.
I smelled a swimming pool smell, but I don't think considerable amounts of Cl2 gas are generated.

Here the video (I held the camera in my hand while my other hand was holding the anode.

http://www.metallab.net/jwplayer/video.php?f=/clips/CaCl2-el...

Did somebody else try this ?

MeshPL - 17-5-2015 at 11:34

You cannot produce big amounts of chlorine by electrolysis of chloride solution. You will produce some chlorine and calcium hydroxide, but after some point all produced chlorine will react with calcium hydroxide (not super soluble, yet soluble enough) to form calcium hypochlorite and calcium chloride. For such a process to work you need to use:
-molten chloride
-diaphragm/membrane cell
-mercury cell (works with NaCl, not sure about CaCl2)

BromicAcid - 17-5-2015 at 11:40

MeshPL, obviously you did not read the original post carefully, this does appear to be molten and looking the video it is obvious.

@metalresearcher: Your setup might work for a proof of concept but it's not going to let you really make anything isolateable. You're using too much current and shorting out your cell constantly (or so it seems), also your cell is much too small (i.e, no separation of products) and it looks from the color that your setup is too hot, you're just burning off any calcium as you form it. You need to put some more care into this than tossing some CaCl2 into a crucible and hitting it with some amps.

blogfast25 - 17-5-2015 at 11:50

Quote: Originally posted by MeshPL

You cannot produce big amounts of chlorine by electrolysis of chloride solution.

Assuming the CaCl2 drying agent was relatively dry, he wasn't electrolysing a chloride solution. Small amount of water in the melt would be electrolysed first, because of the relative Standard Reduction Potentials of H+ and Ca2+.

Using some lower melting eutectic mixture could make things easier though.

[Edited on 17-5-2015 by blogfast25]

byko3y - 17-5-2015 at 11:56

Let me remind you of an old method of calcium production using aluminium The Production of Metallic Calcium by Thermal Reduction
Also, I'd like to say few words bout sodium metal production, for which not a regular NaCl molt is used, but a mixture of NaCl+CaCl2 or stuff like that (doesn't matter).
The starightforward method of CaCl2 electrolysis is complex and inefficient. There are many ways to improve the process, e.g. ELECTROLYTIC PRODUCTION OF CALCIUM METAL By J. J. Lukasko 1 and J. E. Murphl or US3226311
Also, you need to separate the chlorine from calcium metal, because those two react violently.

metalresearcher - 17-5-2015 at 12:12

BromicAcid : you are right, this is indeed a raw test, for really isolating Ca metal more is needed. But why is the current too much ? For the small size of the cell ?
Indeed a kind of a 'Downs cell' is needed and some fluxing to lower the mp. Pure CaCl2 melts at 772 C and I guessed by the color that the temperature is not far above it. I forgot my infrared pyrometer to check.

blogfast25: Using some lower melting eutectic mixture could make things easier though.

But adding NaCl will isolate Na as Na has priority over Ca in a mixture. So what salt should I add to lower the mp ? A potassium salt ?

[Edited on 2015-5-17 by metalresearcher]

blogfast25 - 17-5-2015 at 12:27

Aluminothermic calcium:

http://www.epa.gov/osw/nonhaz/industrial/special/mining/mine...

Quote: Originally posted by metalresearcher

But adding NaCl will isolate Na as Na has priority over Ca in a mixture. So what salt should I add to lower the mp ? A potassium salt ?

KCl is used.

[Edited on 17-5-2015 by blogfast25]

BromicAcid - 17-5-2015 at 18:05

Yes, I think the current is too much for such a small cell. I did electrolysis of LiCl/KCl eucetic in a similarly sized cell with a 10A power supply and even that was too much, as material was reduced at the cathode, the metal formed would help to short out the cell as it approached the side walls. For a cell that size it's just overkill. Of course I don't know what your current draw actually was, just what your supply was, if you only had the very tip of the electrode touching your current would be much lower, right? But really, your crucible was tiny, at 100A and 100% efficiency (of course neither happened in reality) you would be making 12.5g of calcium metal every 10 minutes, likely consuming all the electrolyte in that container.

j_sum1 - 17-5-2015 at 20:47

Quote: Originally posted by metalresearcher

Here the video (I held the camera in my hand while my other hand was holding the anode.

http://www.metallab.net/jwplayer/video.php?f=/clips/CaCl2-el...

Looks pretty in an evil, hell-is-hot-pink kind of a way.

blogfast25 - 18-5-2015 at 08:18

Quote: Originally posted by BromicAcid

Yes, I think the current is too much for such a small cell. I did electrolysis of LiCl/KCl eutectic in a similarly sized cell with a 10A power supply and even that was too much, as material was reduced at the cathode, the metal formed would help to short out the cell as it approached the side walls. For a cell that size it's just overkill.

Did you use len1's cell, as described in his book? Just curious...

[Edited on 18-5-2015 by blogfast25]

MeshPL - 18-5-2015 at 12:55

Sorry for wrong reply I posted a while ago.

blogfast25 - 18-5-2015 at 13:44

Quote: Originally posted by MeshPL

Sorry for wrong reply I posted a while ago.

Three 'Hail Science!' and we'll say no more about it.

BromicAcid - 18-5-2015 at 14:52

Quote: Originally posted by blogfast25

Did you use len1's cell, as described in his book? Just curious...

No, I don't think Len had his book out at that time. It was just a proof of concept (looking very similar to the opening post of this thread), you can see the glow in the picture from the lithium metal shorting out the cell which is one of my worries with high current in such a small cell.

blogfast25 - 18-5-2015 at 16:10

Quote: Originally posted by BromicAcid

What were the electrodes and crucible used?

BromicAcid - 18-5-2015 at 16:13

Cathode was nickel, anode was just the crucible (mild steel).

blogfast25 - 18-5-2015 at 16:17

Quote: Originally posted by BromicAcid

Cathode was nickel, anode was just the crucible (mild steel).

I see. Thanks.

420MLGnOhEADsCOPEpro - 18-5-2015 at 21:36

if we assume 100 amps at 100% efficiency then you were making 0.37 grams of chlorine per second = 22.2 grams per minute = 1,332 grams per hour
you definitely didn't get that much but a notable fraction of that is a reasonable guess
i would consider grams per minute of chlorine to be significant

metalresearcher - 19-5-2015 at 08:56

I doubt whether the current was really 100A. I set the welder to 100A, which is a maximum rating and will be achieved when holding a welding stick for drawing a welding arc. But when electrolyzing the resistance of the salt solution bath is leading which is much higher than steel.

Quote: Originally posted by 420MLGnOhEADsCOPEpro

if we assume 100 amps at 100% efficiency then you were making 0.37 grams of chlorine per second = 22.2 grams per minute = 1,332 grams per hour
you definitely didn't get that much but a notable fraction of that is a reasonable guess
i would consider grams per minute of chlorine to be significant

22.2kg / minute = 1.332kilograms per hour not grams.
22.2x60 =1332 not 1.332.

That is another hint that the 100 amps are not reached by far.

Marvin - 19-5-2015 at 10:14

I make 100 Amps as producing 0.037g/s Chlorine. 133g/h.

metalresearcher, he used a comma, and you have an errent k on the left hand side yourself ;P

blogfast25 - 19-5-2015 at 12:05

Faraday constant = 9.64853399 x 104 C per mol of electrons. 2 mol of electrons per mol of chlorine generated.

Mol Cl2 generated per second = 100 C/s divided by 2 x 9.64853399 x 104 C/mol = 0.00052 mol Cl2/s = 0.037 g Cl2/s = 133 g/h.

metalresearcher - 20-5-2015 at 11:33

Well, I did another try (proof of concept) and set the welder to 25A. I did not notich much difference but less shortings, more current through the salt bath instead of shorting arcs. On the edge lots of yellow sparks appeard but the color was more sodium like yellow, which means that some NaCl is in the solution, which does not surprise me as it is just commercial drying agent.
According to the color I guessed the temperature about 800C slightly above the mp of CaCl2 of 772C.

BTW I did this outdoors due to the Cl2 stench.

Now I try to find ways to isolate the Ca metal which is not easy as it floats on top of the solution. The only way is not to use the crucible as cathode but a steel cylinder closed on top as cathode, completely immersed in the solution. After cooling and when the salt is frozen I have to dig out the cylinder and to break out the Ca metal under the salt, as removing the cathode when the salt is still liquid, the still hot Ca metal will be exposed to the air which burns immediately.

The clip again (refreshed tonight): http://www.metallab.net/jwplayer/video.php?f=/clips/CaCl2-el...

[Edited on 2015-5-20 by metalresearcher]

Marvin - 20-5-2015 at 13:25

My only try at calcium was an utter failure. I think the reason I couldn't get my CaCl₂ molten was down to a large amount of oxide or carbonate in the drying chips, but it might just be that a Bunsen and a metal container couldn't reach the temperature. I've since read about adding hydrochloric acid to the CaCl₂ and heating to fusion to get a good melt. You'd need a good draught for the fumes but then you do anyway.

One key to this seems to be getting the cell absolutely right for the current available and cooling the cathode enough to use a window between the melting point of the chloride and the metal to pull a stick of metal (covered with a little chloride) out of the molten chloride as it forms. This probably needs a lot of planning and experimentation as normally the cathode will be the hottest part of the cell.

blogfast25 - 20-5-2015 at 16:51

Quote: Originally posted by metalresearcher

Now I try to find ways to isolate the Ca metal which is not easy as it floats on top of the solution. The only way is not to use the crucible as cathode but a steel cylinder closed on top as cathode, completely immersed in the solution. After cooling and when the salt is frozen I have to dig out the cylinder and to break out the Ca metal under the salt, as removing the cathode when the salt is still liquid, the still hot Ca metal will be exposed to the air which burns immediately.

The clip again (refreshed tonight): http://www.metallab.net/jwplayer/video.php?f=/clips/CaCl2-el...

Much better video, thanks.

In order for the closed cylinder cathode idea to work, the outside of the cathode would also have to be covered in some insulating material, as current seeks the path of least resistance.

I've played with that idea in the past by covering the outside of such a 'cupola cathode' in annealed alumina, by using KClO3 + 2 Al === > Al2O3 + KCl. Or KNO3 + 2 Al === > Al2O3 + 1/2 K + 1/2 N2.

A mixture of the ingredients could be made into a wet (water) paste, applied onto the cupola, dried and then lit with the resulting alumina covering the cupola.

[Edited on 21-5-2015 by blogfast25]

metalresearcher - 22-5-2015 at 08:27

Quote: Originally posted by blogfast25

Or KNO3 + 2 Al === > Al2O3 + 1/2 K + 1/2 N2.

This will NEVER happen, particularly at those temperatures. Al powder will NEVER release elemental K from its salts.

But indeed, the exterior of a steel cylinder cathode should also be covered.

blogfast25 - 22-5-2015 at 09:07

Quote: Originally posted by metalresearcher

Quote: Originally posted by blogfast25

Or KNO3 + 2 Al === > Al2O3 + 1/2 K + 1/2 N2.

This will NEVER happen, particularly at those temperatures. Al powder will NEVER release elemental K from its salts.

Oh but it does because Al reduces K2O easily, it also reduces KOH in a thermite-style reaction even. Of course the K re-oxidises immediately in air.

I've used the KNO3 + 2 Al reaction many times to heat boost titanium (dioxide) thermites.