Magnesium Extraction From Epsom Salts

[Lord Malice]

Gentlemen I am relatively new to the Forum and I am asking this question here because frankly i dont know whether it classifies as one that warrants the boot. I know that the admins check here and if so tell me if it goes against rules or not.

Anyways, this is my question....i am going to try to extract magnesium from epsom salts (they contain small amounts of the element) and have already searched your database for any advice/techniques...and found nothing to help. Im doing this at home using bare essentials...hot plate, common chemicals, and obviously standard lab utensils.

If anyone has any advice it is appreciated.

Lord Malice

[Zeitgeist]

Epsom is hydrated MgSO4,

Hopefully you know enough, or are learing basic chemistry knowledge, so you know that you need to electrolyse the molten salt. An Aqeuous solution clearly won't work.

I'm not sure of the melting temperature off the top of my head, but i don't think it will be practical, unless you find something you can mix in to bring it down.

[Lord Malice]

Thanks for the reply brother,

I am in my fourth year of university chemistry with a minor in biology...so hopefully i do know something :confused:

Anyhow i know that i have to electrolyse the salt to separate the components but i was just wondering if anyone had any advice as to the best way to do that from home...perhaps using a car battery or 12V? and then just scrape the Mg off the cathode. secondly would this method gather the highest recovery of magnesium? Is there a better method?

Thanks for all your help.

----------------

Please use your shift key! And have a look at the screen you will see after you submit your posts. It tells you that your posts are subject to moderation and will show up after they have been approved.

Rhadon

[megalomania]

I have actually spent a lot of time researching this exact extraction, I am sure I have mentioned this before on other threads. Anyway, the subject interests me so I love to talk about it.

Theoretically this is possible on the lab scale. Industrially they do this all the time. The short of it is you can electrolyze molten magnesium chloride to produce magnesium metal.

Now some practical aspects to consider. First of all magnesium chloride melts at 712 degrees C, so no hotplate will do. You can mix this with some sodium or calcium chloride. I believe they use calcium chloride to lower the melt temperature because at these temperatures calcium reacts with the mix and is not electrolyticly reduced. This is something to consider for matters of purification.

Another situation to keep in mind is magnesium ignites in the presence of air at the melt temperature. Any electrochemical reduction will have to be done in a sealed container. The cyanide thread has an excellent description of an apparatus, used by Marvin I think, where a metal pipe is heated and an exhaust pipe dips under water.

If I was doing this, and I intend to some day, I would use about a 60-40 mix of magnesium to calcium chloride well mixed in a steel pipe, fat diameter, capped at the bottom. The top is sealed with cement, possibly the high temperature stuff. Also in this cement is a length of copper tubing that connects to a glass jar (like a pickle jar or mason jar). You can cut a hole in the jars lid and glue in the copper tubing, or cement it. Another hole in this jar leads some tubing into a bottle of water. As the chlorine is released it escapes through this copper tube, into the empty glass jar, and bubbles out through the bottle of water. In the mean time air is excluded from entering in because only water would be sucked back. The purpose of the empty glass jar is to prevent water from being sucked into the superheated pipe when it begins to cool as this could be disastrous.

Another hole in the cement of the pipe holds your electrode. Get a carbon welding electrode. You may want to use the steel pipe itself as the other electrode, although I have had mixed success getting an electrical contact to work in a more corroded pipe in other electrolysis experiments. If the pipe does not work then you have to stick in two electrodes.

Just make sure the electrodes do not touch and try to minimize the amount of time that chlorine gas remains in the system. Make it so the magnesium forms on the pipe electrode and the chlorine is released at the carbon electrode. The chlorine will react with magnesium and reform magnesium chloride at these temperatures. Stick the carbon electrode down about half way into your salt off to one side. The reason you locate the electrode nearer to the side of the pipe (your other electrode) is to minimize electrical resistance that can reduce yield.

The pipe system can be heated by building an improvised furnace. I suppose you could dig a hole in the ground, fill it with charcoal, and insert the pipe in the center. A steel pipe should also be placed into the hole to allow you to blow air into the bottom to get a high temperature. You can use a blow dryer or something. Anyway that’s too much to go into now, do a search for improvised furnaces for other ideas.

One thing I would like to build is a low voltage, high amperage power source that connects straight into the wall. Battery chargers are nice and all, but the higher the current, the faster your electrolysis will proceed, and you only have so much time for your charcoal to burn. You want to get the most magnesium you can in a single heating. I can’t vouch for the safety or effectiveness of this but you could rip the cord off a lamp, or extension cord or something, and separate the two wires. Place one wire in a glass pan (a casserole dish or pie plate) of salt water and connect the other wire to an electrode. Place another length of wire in the salt water and connect that to the other electrode. Tie the ends of the wire in the salt water to lead sinkers and separate these wires by several inches. Plug it in the wall and pray… Technically the salt water acts as an improvised rheostat to control the current so you don’t blow a fuse. By moving the wires closer together or farther apart you control the current. Never let them touch out there will be a short. The salt water will also heat up so it may need some ice if this runs too long. Never, ever touch the water, you will get electrocuted!

You can make magnesium chloride from Epsom salts (magnesium sulfate) in a few easy steps. First, prepare a concentrated solution of magnesium sulfate and a concentrated solution of sodium bicarbonate (baking soda). Bring the magnesium sulfate solution to a boil and add the baking soda solution. I’m not going to calculate the molar ratios of how much to add because it’s late and your problem anyway, suffice it to say you can calculate exactly how much sodium bicarb solution to add based on solubility in a saturated solution. If you have a scale even better, one mole of sodium bicarb reacts with one mole of magnesium sulfate. What you end up with is a solid precipitate of magnesium carbonate. Collect the crystals on a filter or something and wash them well with water to remove any traces of sodium sulfate. SLowly add hydrochloric acid to the magnesium carbonate and it will form magnesium chloride. Boil the solution down to get the crystals.

If you try this I sure would like to hear your results.

[MightyQuinn®]

One thing I would like to build is a low voltage, high amperage power source that connects straight into the wall........the higher the current, the faster your electrolysis will proceed, ....... I can’t vouch for the safety or effectiveness of this but you could rip the cord off a lamp, or extension cord or something, and separate the two wires......


As I have done "time" at a plating plant I can assure you that you will want to invest in (or construct) a DC power supply. Your theory of using AC current would not be prudent, nor would it be effective.

As a side note……..as a rookie at the plant I ruined a batch of valve stem covers while nickle plating by “goosing” the rectifier to a higher than spec amperage….the parts were referred to as “burned” and had a mottled, discolored and crystalline quality…..NOT desirable for that application, but possibly what you require.

[Boomer]

Why not get a cheap welding transformer? They go to 100-200 Amps! Could even rent one or ask a friend.

All you need then is a rectifier bridge (e.g. B80C50000) for 50A screwed to a metal heat sink (thick alu plate placed half in water) per element, working with 2-4 pipes. These rectifiers cost 3 bucks each (in radio shops more like 10 bucks).

Or put them in parallel, but then they MUST be on the same heat sink, or one will overheat and take all the current. Small resistors (<0.1 Ohm, or just a metre of wire) in series to each can prevent that.

P.S. Older PC power supplies give 5V at 20-30A. You could start with one of them.

[Anthony]

I bought a Switch Mode Power Supply on ebay. Originally intended to power racks of hard disks, I bought it for the auxillary outputs of +12v and -12v at 15A. But the main output is 5v (DC) at 65A.

IIRC, 5VDC is good for electrolysis, and the current should be enough. The PSU is only 1-1/2" tall and runs cool.

Cost me £20 and there's always something on there, there will be more on ebay.com too than the UK one.

[Zeitgeist]

IIRC, 5VDC is good for electrolysis,

Of course, it depends on the elements involved.

A list of standard redox potentials is your friend.

[Marvin]

Megalomania,
Ive described a similar apperatus used by someone else, but I think you are probably thinking of Polverones phosphorous method?

Aside from the annoying filtering of calcium sulphate, is there a problem with simply mixing epsom salts and calcium chloride to get magnesium chloride?

There is some useful information in 'Elements of electrochemistry' 1897, it describes a demonstration method that could potentially be scaled up.

20 grams of crystalised magnesium chloride,
7.5 grams of potassium chloride,
3 grams of ammonium chloride,

Fused in a crucible and then poured into an unglazed clay pipe (as in for smoking) previosly strongly heated, kept hot with a blowpipe flame and clamped in place. A knitting needle is introduced to the stem of the pipe that serves as the cathode and a carbon rod is dipped into the bowl as the anode and approx 18v is applied (the book predates power supplies this so this actually 10 lead acid batteries in series) next a thick layer of finely ground wood charcol is put in the bowl covering the molten salt. This prevents the magneium burning in the air, and also stops it being distributed as a powder through the molten salt. At the end, the contents are broken open and globules of magnesium up to 5mm are found inside.

It seems a little wood charcol can work wonders, its a shame this trick cant be used for making sodium from sodium hydroxide.

I think an electric welder would be a very good idea, some are available with DC output and the 50v odd open circuit voltage would probably drop to less than 20v or so under moderate load. Aditionally they have a degree of control over the current. For an AC welder boomers method of stacking diodes in parallel to avoid thermal avalanche is absolutly the way to go, though I think 10 to 20 mohms per 50 to 10A rated diode would be neerer the optimum values.

Switched mode power supplies might be workable with ordinary water based electrolysis, where cell resistance is known, fairly high, and and not variable but I suspect they wont take a lot of abuse before they die, unlike welders. Using them with molten salt electrolysis might be unwise, they are designed to power electrics with a regulated constant voltage after all.