Sciencemadness Discussion Board

unconventional sodium

 Pages:  1  2    4

ballzofsteel - 28-4-2004 at 20:47

Hey Bromic,

Did you take a look at the patent I posted
further on up the thread regarding reduction of the nitrate with Sodium carbonate??

It is fairly comprehensive,explaining in detail,construction materials ect.

Some real good info for you there man.

Pretty please,give it a read.

Fuse them salts,lower the temps needed
collect you nitric acid.:cool:

Correcting my mistakes

BromicAcid - 1-5-2004 at 11:34


Then after you get your fill you put it in, jeeze, I forget the exact name so you're going to have to settle for a phonic equivalent, linindale or something of that nature, regardless it is an inert substance, I think a paraffin of some sort with a low density around .56 and apply heat. The sodium will melt and float to the surface leaving hydroxide behind from there scoop it back up and store under your final storage solution.

I actually looked up the procedure in the library, it's actually for making lithium but the principle should be the same since we are working in terms of physical properties. You would take your Na with NaOH contaminates and put it in a paraffin bath. The sodium melts and floats to the surface and the NaOH stays on the bottom. The thing bout the ligroin is specifically related to the lithium procedure that this was written for. The density of this fraction of petroleum ether is only .56 g/ml which seems to be very slightly higher then the .53 g/ml for lithium. I guess at best it might make a suspension. It recommends storing Li in a container filled to the brim of this stuff regardless.

The_Davster - 1-5-2004 at 21:51

I used an alcohol not a bunsen burner to melt the sodium hydroxide, took a long time though. Also it seems to me that crystall drano has a lower melting point than pure sodium hydroxide. It must be mixed with some sort of oxidizer because if a popsicle stick is dipped into the molten draino it catches fire once removed.

Off Topic, but....

Saerynide - 1-5-2004 at 22:48

Neat!! I want to try the flaming popsicle stick thing now :D

BromicAcid - 2-5-2004 at 07:23

The commercial draino is somewhat hydrated, that is the main impurity that I know of that causes it to melt at a lower temperature. As for the flaming popsicle stick I think there was a thread posted here about it. How the hydration of NaOH can start a fire. Put a popsicle stick that has water throughout it into molten NaOH and watch it make excessive energy from hydration and set it on fire :cool:

Polverone - 2-5-2004 at 10:41

Drano contains some sodium nitrate (as mentioned upthread) as well as some sodium chloride, it appears:

BromicAcid - 2-5-2004 at 12:15

Wow, I guess I was mistaken about water of hydration being the main impurity.

My bad. :(

Regardless, I still stand by my belief that the inflamation of the wood in question is caused by the inheret moisture in the wood being acted upon by the hot molten hydroxide.

Saerynide - 3-5-2004 at 00:31

Maybe its like the "Gummybear goes to hell" thing with the molten KNO3 since Draino has NaNO3?

Strepta - 6-5-2004 at 16:59

Originally posted by chemoleo
Bromic, your method doesn't seem all that different, except maybe the water addition.
I dont understand why you'd get metalic Na globules - everyone who's tried it (inc. myself), one gets a dark grey mass, with the occasional yellow explosion from igniting H2/O2/Na.
It's not like your temperature is low (judging by the mark on the wooden bench :P), or that there is an oxygen deficiency in your local atmosphere (you wouldnt be writing this otherwise :P)
But seriously, I can't figure out why you'd get metalic sodium, as a nice little globule.
How about upscaling this?

patu - 9-5-2004 at 09:12

I purify the sodium by heating it in just plain mineral oil I got at the supermarket. The sodium has a very high suface tension. The very pure sodium rises to the top which creates a ball with a visible line between the shiny sodium and the dull sodium hydroxide. Even though the densities are different the sodium really doesn't want to detach from the sodium hydroxide. I guess you could take a razor blade and just cut away the sodium.

Tacho - 13-5-2004 at 08:02

I must confess, embarrassed, that I also had a hot NaOH burn in my tongue. So, statistically, tongue burning is a major hazard in sodium making. Wear googles and keep your mouth shut!

Saerynide - 13-5-2004 at 08:05

You guys with all your alkali burns are scaring me. And I want to get lithium from batteries soon (summer hoilday is coming in 2-3ish weeks!!) :P

Edit: typo

[Edited on 13-5-2004 by Saerynide]

The_Davster - 13-5-2004 at 13:28

Tacho: Ive tasted NaOH too:P. I was scraping sodium off the electrode with pliers and the electrode sprung up and flicked some sodium hydroxide onto my tongue. I still cant think of why my tongue was out at that moment though...

Tacho - 14-5-2004 at 03:05

Originally posted by rogue chemist
Tacho: Ive tasted NaOH too:P. I was scraping sodium off the electrode with pliers and the electrode sprung up and flicked some sodium hydroxide onto my tongue. I still cant think of why my tongue was out at that moment though...

See? See? Another one!

Hot NaOH is magnetically atracted to tongues!! Googles, gloves and MOUTHS SHUT!

High boiling organics with densities greater than 1

Strepta - 15-5-2004 at 16:24

You need an inert liquid with a density of 1 g/cm3 or above with a boiling point of greater then 100C. This is the technique used to purify lithium that I mentioned before, you melt the impure NaOH/Na mixture at the bottom and as it melts the pure-er Na will float to the top leaving the NaOH behind. You could use something like nitrobenzene possibly. I think most chlorinated hydrocarbons will react with Na at these temps and that is the massive drawback because most organic liquids that have a density of greater then 1 are halogenated hydrocarbons.

[Edited on 5/9/2004 by BromicAcid]

I thought I'd give a nitrated organic a try. I have toulene, so I prepared mono nitrotoluene (NT) according to Davis. It's a straightforward nitration with mixed acid. I placed a small ball of Na in the NT in a test tube. It appeared to darken a bit. I heated it carefully-- the Na began to react with the NT and soon all Na was consumed and the liquid, initially yellow, turned reddish brown. :( I suppose the Na is reducing the NT. Oh well, not much ventured..

[Edited on 16-5-2004 by Strepta]

[Edited on 16-5-2004 by Strepta]

Magnetic NaOH

Cyrus - 17-5-2004 at 17:09

I have gotten NaOH on my tongue too, but not from electrolysis. I was doing a simple NaOH +Al reaction, and stuck my head too close to the bubbles.

Is SS ok for molten NaOH, and why do you all use nickel? I thought SS was dissolved very slowly by NaOH. Could you not use carbon, or even make a carbon coating on your container with a candle?

Where can I get soda-lime glass to get
sodium that way?

BromicAcid - 17-5-2004 at 19:17


Is SS ok for molten NaOH, and why do you all use nickel? I thought SS was dissolved very slowly by NaOH. Could you not use carbon, or even make a carbon coating on your container with a candle?

I believe stainless reacts appreciably with NaOH(l) however I have no sources to back me up, some designs for castner cells call for cast iron which I come across even easier, it holds up fairly well. I use nickel whenever possible because for as far as I can tell it is nearly invulnerable to molten NaOH, I've used the same set of electrodes several times and no pitting and the molten NaOH does not turn colors from making nickel salts. Carbon will die in NaOH(l) there is no getting around it, I used graphite rods as electrodes and it quickly formed as semi-metalic solution that conducted electricity and thus accomplished no electrolysis.

As for where you can get soda lime glass, most everything made of glass that is not made for a lab is soda lime glass. But it needs to be thin, it should have a partial vacuum within it, and it does break after a period of use. NASICON ceramics are the way to go for a membrane that permits only Na though.

Edit: And Strepta, nice try, nice to see someone trying somthing an idea that I've put fourth, keep up the good work :)

[Edited on 5/18/2004 by BromicAcid]

Cyrus - 19-5-2004 at 15:18

I have tried electrolysis in Fe endcaps and Cu , and in food cans, and the food cans seem to withstand molten NaOH the best! In the other ones, it changes to brown quickly, and if I stick any iron or copper wires into the melt, it will turn brown too...:(

BromicAcid, does that mean I could use the glass in a lightbulb? It is pretty thin and will withstand a vacuum. If only the wires inside of the lightbulb touched the edge:( Then you could just plunk a lightbulb into the NaNO3 melt and connect it to your power source, as it already has a vacuum inside, run the cell, get the sodium, and crack the bulb open!

Saerynide - 20-5-2004 at 00:16

Inside a light bulb isnt a vacuum. Theres a noble gas in there, I cant remember which one though.

But I was also thinking about using lightbulbs to melt or heat stuff in, since I dont have pyrex, and if light bulbs can withstand constant heating, I hope they can withstand a gas stove :P

Edit: I was going to open a light bulb by removing the connector and taking out all the crap in it, to get a flask like container. But now come to think of it, Im kinda worried that H2 might build up in the bulb (since it has a narrow opening) and with the Na forming, it might turn the lightbulb into a glass pipebomb :(

[Edited on 20-5-2004 by Saerynide]

Esplosivo - 20-5-2004 at 05:49

I wouldn't be so worried about that. After all the opening is quite wide, wide enough for gasses to escape quickly.

Btw, if you intend to use a light bulb as a container remove the metal part completely. When I used to leave it the bulb would crack after heating.

t_Pyro - 20-5-2004 at 06:04

Light bulbs contain an inert gas at low pressure. Since the inert gas (at low temperature) is a bad conductor of heat, the glass capsule does not get hot too fast. After prolonged use, it does get hot, but by then it has already expanded more or less uniformly to prevent cracking.

Using the glass bulb as lab glassware could be dangerous as the temperature changes involved would be far greater than the glass is designed to withstand. Drop a drop or two of water on a glowing light bulb, and you'll see...

[Edited on 20-5-2004 by t_Pyro]

Cyrus - 10-6-2004 at 20:14

Maybe I'm missing something but...

Does Na+ migrate through soda lime glass at all temperatures- if so, why can't the electrolysis be done by placing the bulb in a soln. of NaCl- does H2O also migrate through the glass? Why not use Nafion or Nasicon membranes at room temperature in a soln.?

Sure, Bromic Acid, NASICON is the way to go, but where are you going to get some of that stuff- make your own? The materials aren't too bad but the process looks pretty involved! Also, you speak as if you had actually done the soda lime glass electrolysis- have you? I would love to hear more details, as the sodium produced this way sounds like it would be very clean.

I am dissolving the metal end off of a small bulb currenty, any interesting results definitely are going to be posted.

Marvin - 11-6-2004 at 04:51

Sodium ions arnt mobile at room temperature in glass, or it wouldnt be an insulator. At 300C or so its noticable, but extremly poor by other conductors standards.

As far as I am aware, all the sodium ion conductors have to be used at elevated temperature. Many of them were intended for use in applications like the sodium sulphur battery.

Cyrus - 15-6-2004 at 21:42

Originally posted by Polverone
At one point xoo1246 posted a fascinating patent about producing alkali metals by heating their hydroxides with magnesium in a high-BP hydrocarbon.

Sounds very interesting.
I bet polyethylene would work. I should test that.
I just need to go get some Mg (tomorrow)
at $0.25 a lb from a scrap dealer. I think its a powder too. Amazing. I love the Market.

Then I can get around to all those reactions I've been saying I would do. ;)

All of my previous experiments with sodium have failed. :(

I did 2 early experiments with NaOH a while back, but I forget what, I just remember that they were hopeless.

Electrolysis 3 - 9v battery, molten NaOH in food can, copper wires. nothing happened- I think the battery was nearly dead by then .:mad:

#4- larger 12 v 0.5 a (iirc) power supply,
molten NaOH in food can, Fe wires. Some bubbles, pops, fizzes. The current helped heat the melt, not enough to keep it molten though. Melt turned green at one wire, brown at the other.

More coming in this sector.

I cleaned out the soda-lime glass lightbulb, and filled the bottom with copper wires. Then I found another smaller lightbulb which had a Cu wire actually TOUCHING the glass for some reason. Yes!!! Now I don't have to supply vaccuum, heat, and electricity, just heat and electricity! :)

More coming here.

Sodium thermites

I heated about 1 g NaOH, 1 g Na2CO3, 1 g NaCl, 2.3 g Al in a copper test tube - (got that one for .90$ on one of my recent excursions from Persia to Home Depot) The Al was not a fine powder, more like finely crumpled shreds of Al - it was from a blender, I have almost no fine Al left.

the apparatus looked like this

/ ______________####_- more pipe
/ / __ 90 degree fitting, teflon tape seal
| |
| |- copper test tube
| |
|* | * = thermite
| *|
|* |
|* | # = steel wool
* /

After heating for about 0.5 minutes with my torch, there was a loud thump, I lit the H2 gas escaping with the torch, after about 1.5 minutes, the bottom half of the test tube started glowing very brightly for about 0.5 min, more than the torch could get it alone.

I let it cool, and then looked at the steel wool - should be gobs of Na, right? NOTHING!!! :mad: The inside of the copper pipe seemed to have a light white haze on it. The leftover reactants could not be chipped out, so I added water. Then the fun started. The tube got hot, the mixture fizzed, and lots of H2 came off- enough to light in medium sized fireballs. I did the reaction again, and the tube gets too hot to hold when water is added.

Then I tried doing the experiment just like Tacho did- lower bowl, upper bowl with water and ice- I heated the lower bowl red hot in places, H2 came off- enough to make the upper bowl filled with water jump a cm or to into the air :) , but no violent reaction happened. The lower bowl did not glow bright red except where the torch was. No residues on the upper bowl at all. Argg!:mad:

The only thing I can think of is the Al could be bad- it came from Al foil and the blender. What do you think of the copper condenser apparatus- it's cheaper and easier to work with than steel pipes.

Sorry for the longevity of the post.:P

Tacho - 16-6-2004 at 04:41


There is no way the sodium will deposit nicely in the steel wool. The high temperature makes it vaporize. Even the condensed sodium is quite dirty. I couldn’t even melt it into a single blob.

The only practical uses for this reaction I could find until now are: 1)impress kids igniting the condensed powder with water; 2) to make methoxide/ethoxide. Even the leftovers react vigorously with alcohols. But be careful, those who try. The leftovers still react violently with water after they reacted with alcohol.

Cyrus - 16-6-2004 at 09:42

Originally posted by Tacho
There is no way the sodium will deposit nicely in the steel wool. The high temperature makes it vaporize. Even the condensed sodium is quite dirty. I couldn’t even melt it into a single blob.

Ok, I was exaggerating about my expectations for nice shiny globs of sodium- dirty stuff is fine for now. And yes, the Na will be a vapor when it is produced- but some should deposit on the steel wool if it is cold enough, clean or not.

Edit, why won't the sodium be pure- if its done in an inert gas, assume, and the only products are MgO, H2, H2O and Na, what is the impurity? NaOH? MgO? NaH? NaOH can be removed, but I don't know about the others-could the sodium be distilled to purify, even if it won't melt?

Also, I tried magnesium turnings and sodium hydroxide under molten candle wax today, I heated the candle wax till boiling, there was vigorous bubbling of hydrogen, the Mg turned black, the wax caught fire- when the torch was placed straight at it, the flames billowed about 1m high.:D:D:D. When placed under water, nothing happened. (the reactants were covered in solidified wax) If a decent way of extracting the Na from the wax and the MgO/ Mg(OH)2 is found , this might be a simple and promising method.

Halfway throught the reaction, I discovered that a prill of NaOH had located themselves on my arm, and were busy making Cyrus-soap... :mad: I was most annoyed because I had to stop watching the reaction and wash myself.

[Edited on 18-6-2004 by Cyrus]

[Edited on 18-6-2004 by Cyrus]

A hair raising experience

Cyrus - 22-6-2004 at 14:31

Today, I tried to make sodium. Once again.

I used the copper test tube setup previously described, but I used lots of teflon tape to keep the joints secure, and the outlet of the setup went diagonaly into a food-can with about 2 cm of mineral spirits in the bottom.

I used 1g of NaOH and 0.6 g of Mg, this mixture was placed into the bottom of the test tube. I was doing this quick and dirty, so I didn't even bother using a clamp to hold the setup in place, I just held the food can with my left hand, and held the torch in my right. Bad move.

Upon heating, some bubbles peacefully drifted out of the copper setup for a minute or two. I began thinking, either this reaction is very slow, or it takes a while to get started. Then


I remember
-1- a very bright flash and a loud noise
-2-a large wave of heat
-3-a hot and sticky sensation
-4-the apparatus decomposing into several flaming objects,
-5-my goggles dislocating themselves from my head somehow

At this point I decided it would be best to leave the area, and then I noticed that the wave of heat was COMING FROM MY HEAD! :mad::mad::mad:

I ran into my garage looking for a bucket filled with ice-cold water (there are no buckets of water in my garage, ever) while beating down the flames on my head with my hands.

Finally, I put the flames out.

-the still flaming apparati were burning holes in the grass
-the propane burner was still on, only heating concrete, though
-mineral spirits were all over me, the concrete, and my goggles
-the goggles were on fire
-All hope of sodium was lost (the most important) :D
-Oh yes, and the last 0.5 cm of almost every hair on my head had turned white and crinkled, including my eyebrows and lashes, which were behind the safety goggles!:o One of my ears was burnt.
My hair sticking up in fused masses at every angle, I truly looked the mad scientist.:D

I smelled like burnt hair and mineral spirits. Pleasant.

Things I have learned
-BromicAcid has good reason to use those beefy iron pipes-which I will use from now on.

-Do NOT stand within 3m of such reaction.

-Teflon tape will not hold very well when using magnesium- it works with Al.

-Water is very good to have at hand.

All in all, a very exiting day. I think I'll take a little siesta from reactions for a day or two though. :)


BromicAcid - 22-6-2004 at 19:01

Jeeze, atleast you were using small amounts.

Have you ever tried this before with magnesium as your reducing agent with out a coating like you had in your previous post? It was mentioned in this thread that it can be explosive
By chemoleo:
instead of Al I used 200 mesh magnesium (so no grains/filings like BromicAcid once did). As soon as I put the propane torch onto the mix (from below, so the mix never contacted the open flame),it ignited, making a massive flash of light, and spewing molten white glowing bits everywhere!!

And in one of my quotes:
"The hydroxides or carbonates of the alkalimetals -- excepting cesium -- are reduced by heating a mixture of one mol. of the carbonate with three gram -- atoms of magnesium.... The reacton with lithium proceeds with explosive violence.... with potassium and rubidium the reaction proceeds quickly.."

So where you expecting this? Magnesium oxide can be added to help moderate the reaction.

Best of luck with future experiments, and those 'big beefy pipes' really do make you feel safer ;)

[Edited on 6/23/2004 by BromicAcid]

Cyrus - 22-6-2004 at 20:12

No, I was not being cautious enough.

But I was using large Mg turnings, not fine powder, (as Tacho advised IIRC) and about 2 g total materials, so I just expected the reaction to go "nice and quickly",
not explode like a small bomb.:( I was not expecting anything like what happened, of course.

Now that I have done it, it is not a reaction I would say proceeded "nice and quickly"

Don't worry about my health, the ear is doing ok, and all of the crisped hair seems to have fallen off, thankfully.

Bromic acid, with your large steel pipes, to prevent suckback, instead of using a one way valve, why not make the level of water such that whenever it gets sucked back, the water level lowers enough to let air through. This has probably been said somewhere before, but oh well.

Be warned, I have just caught the phosphorus bug.

[Edited on 23-6-2004 by Cyrus]

[Edited on 23-6-2004 by Cyrus]

Tacho - 23-6-2004 at 04:39

I’m very sorry about what happened to you Cyrus! I hope you are OK. You were very wise to test the reaction with little reactants!

Thank you for describing what happened, thus helping other people NOT to do the same mistakes.

I’ll start a list of things that should be remembered:

Stay many meters away from any thermite reaction;

Never do thermites in closed tubes or pipes, even it it has a small vent. Any unexpected explosion should have space to release its energy without shards or shrapnel;

Test the reaction first with little reactants;

Magnesium in thermites is not a good idea;


chemoleo - 23-6-2004 at 11:36

Let me add to this:

Heat your reaction REMOTELY, not with the burner in one hand, and the setup in the other!!!!

NEVER EVER use flammable liquids together with an open flame!!! Particularly in the combination burner/flame/thermite/mineral spirits!!! (unbelievable really - I got hurt badly with much less stupidity!)

By the sound of it, you really have to get your act together, or next time you will get injured badly. Trust me I am speaking from experience. Recklessness punishes you faster and harder than you can blink your eye.
It's not a cool story anymore once you lie in hospital with 2nd/3rd degree burns, and a few shards in your abdomen.

Cyrus - 23-6-2004 at 15:39

Thanks for the good advice - the act is now together, for all such future reactions I'll use some clamps, thick iron pipes, a "blast shield" made out of spare sheetrock, and more common sense. And I'll be farther away.

I'm glad this happened because it was a good warning.

Tacho, the apparatus was not a closed tube, it had an opening, but my description may have been confusing.

Chemoleo, what did you do to yourself:(, and also, if I want to make/distill sodium, I think flammable liquids must be used to bubble the vapors through, any suggestions?

chemoleo - 23-6-2004 at 16:19

To be honest, I don't think you have much chance of condensing Na vapours in those oils. I played around a little with potassium in inert oils, and the metal tends to separate into thousands of tiny globules rather than condensing into a coherent one. With K/Na - vapours.... I'd rather think you will get an emulsion... if that. Sodium gas... sounds good in theory... but again, I doubt very much it occurs for real, with the thermite reaction. Remember I tried it, several ways. Distillation in an inert atmosphere is the only way I can see this to work (with massive heat, well above the boiling point of Na), if that. Even then, I do think most of the thermite products are aluminates/aluminium hydrides, despite stoichiometric equations. As I said above once, I tried heating the thermite product in petroleum, for several hours, to release the Na (which should melt) but it didn't work at all.
If I was you I wouldnt pursue it because I'd doubt it would work, and I'd be too worried about saftey (open flame and mineral spirits).

I think it's very interesting what Tacho mentioned about reaction with alcohols (i.e. ethanol) - he described evolution of gas (H2?), where the product still reacted violently with water.
May I ask whether it was absolute alcohol??

If yes, it sounds like a very decent method for preparing alcoholates, I wish I had tried it.

Does anyone have some experience in purifying/playing with alcoholates?

For instance, Tacho, once you react your thermite product with alcohol, what did you do with it? Did you try to react it with something, to confirm you had truly some NaOC2H5??? Please elaborate!

PS On the note of my past incidents... I will elaborate on taht one day... in the 'most serious injury' thread. Never found the guts to do so so far :P but dont worry I am not disfigured or anything... you cant see it.... although it's left me with a healthy permanent scar of being cautious. And you, Cyrus, sound just like me when I was 16... and I paid a dear price for that. Be warned, seriously!! I didnt get such a warning (I didnt know of the internet then), while you DO! Be careful, man! no more thermites with Bunsens igniting it by hand! I mean it! A tiny bit of neglect can cause you more harm than you can imagine.

[Edited on 24-6-2004 by chemoleo]

Tacho - 23-6-2004 at 17:20

Interesting, chemoleo, it’s the second time you say you doubt the thermite. Since I claimed to have obtained sodium by it (dirty small nuggets), you put me in a awkward position! Well, who cares! Here is my methoxide experience:

I reacted the condensates that I say are sodium and the residues of the reaction (NaOH+ NaCl+NaCO3+Al, the NaCl makes all the difference, I BET you never done it using NaCl and NaCO3 as I described) with anhydrous methanol, they bubbled and fizzled( I just poured the methanol in the ss cup that held the thermite and added the condensed powdery sodium into it).

After it settled, I poured the methanol in another flask and added water to the ss cup (to wash it). Hell broke loose: bubbling, fizzing and boiling out of the flask.

Then I boiled away most of the methanol and added IPA, to precipitate some NaOH. Some white powder really precipitated. The decanted IPA was very, very alkaline (correct if I am wrong, but IPA dissolves very little NaOH). The reaction I tried that used methoxide failed, but I believe it was not the methoxide fault. I think I got that part right.

I'll be glad to repeat the methoxide thing it in a smaller scale if anyone wants to test an specific theory! Some decent theory please.

[Edited on 24-6-2004 by Tacho]

chemoleo - 23-6-2004 at 18:44

No no,
this is not about doubting you or anyone, it's about finding out what you tried!
Regarding the Na nuggets - I have doubts simply because I didnt get any. So I am trying to find out why that is so.
I am very intrigued by you methoxide experiment, nonetheless. Can anyone imagine a simpler DIY method for preparing it??
I wish there was some way to discern methoxides/alcoholates from other products (that is, I can't think of any that would behave similarly...)

Tacho - 24-6-2004 at 03:14

Did you try it using NaCl as I said? Well, try another run and react ALL the products of the reaction with methanol.

There is a catch here, you have to wait until it cools, on the other hand, they absorb water from the atmosphere quickly, so keep the condensing bowl on top until all is below the bp of methanol. This also helps prevent the products (which I say contain lots of elemental sodium) from catching fire when exposed to fresh oxygen. Yes, it happened.

Cyrus, I also doubt that you will distill sodium that way. If any is produced, it will stick to the walls along with by-products.

Why don't you try the can & water bowl procedure to know the reaction a bit better?

Know what? I'll try to make this thing work with al foil, just to see if someone else can obtain my results, I'm feeling a bit lonely here.
Well, soon...

[Edited on 24-6-2004 by Tacho]

Cyrus - 24-6-2004 at 13:49

Tacho, as soon as my concrete blast chamber (It actually will have some concrete fiberboards) is finished, I'll attempt to duplicate your reaction with Mg, not Al, using the lower and upper bowls.

As for Na distilling, I think it sounds improbable too, but isn't that how the "ancients" made Na a couple hundred years ago? As in NaOH or Na2CO3 and Fe powder, in a charcoal furnace, the product vaporized and was condensed in a flask of some organic liquid, I forget which.

BromicAcid - 24-6-2004 at 14:05

The product was actually condensed in liquid lead. From there it was distilled yet again under an inert atmosphere. This is the old carbonate process that Du Pont used.

Cyrus - 24-6-2004 at 14:28

I just remembered, I think it was naptha.

BromicAcid, you may be right, I cannot check, I am missing encyclopedia #16, where I saw the process.

Tacho - 24-6-2004 at 17:08

Cyrus, did you take a look at the papers about production of sodium that Bromic Acid made available some time ago?

There is a link somewhere in this thread. If you can't find it, PM me.

That is a must-read.

Edit: BTW, I tried to make some Al power using Al foil, a NaCl slurry and a blender. The result was disapointing. It was not a powder, more like crumpled flakes. IT DID NOT WORK in the thermite with NaCl and NaOH.

Edit2: Also, I did a small run of the thermite (1 spoon of NaCl, 1 of NaOH, 1of Al ), but this time, since I have recently bough a bag of ativated carbon powder, I added half a spoon of it to the mix, to see what happens.

I have the impression that the deposited sodium was less powdery, and I scraped some soft metal from a "drop" formed in the condenser. Maybe the improvement was just an impression, maybe it was because I used no carbonate, but if you have carbon powder available, may worth a try.

I keep having this metal that melts under the xilene in many about 2mm blobs, mixed with the powder (residue), but just won't aglutinate to one big blob. It's not an amalgam, because when it fizzes and disappears in water (no flames for small bits), it leaves no residue.

Guess I have to try paraffin as a solvent. Any other ideas?

[Edited on 25-6-2004 by Tacho]

[Edited on 25-6-2004 by Tacho]

Cyrus - 25-6-2004 at 08:39

I started reading them, my computer is more than slightly slow though, (Win 95!)
I didn't get very far, but I will try again!

Tacho, melting the Na/impurities under an inert atmosphere might help. (if this is in a centrifuge it might help too) Then putting it in one of the perforated ladles described elsewhere will remove the NaOH, perhaps other impurities too. Maybe you could mount the ladle onto the centifuge. :P

Molten sodium filtering! That ought to work, but Na has a lot of surface tension, so it would have to be under pressure to drive the Na through the filter. Or under a vacuum to pull it through.

Sorry if I am rambling.

vulture - 25-6-2004 at 13:16

Do not use teflon when working with alkali metals. The only thing teflon reacts with is with alkali metals and then it acts as an oxidizer. And a rather powerful one...

BromicAcid - 25-6-2004 at 14:37

To add to what vulture said, when PTFE and alkali metals combine free fluorine can be evolved in significant quantites.


Cyrus - 25-6-2004 at 15:10

Doesn't teflon react with molten alkali hydroxides also?

Edit: Just made some sodium, using an upper ice-cold bowl and the lower bowl to do the reaction in. I didn't measure anything out, just added NaOH and Mg till it looked right. I mixed these together a little in the lower bowl, and set this in my reaction chamber.

I was being ridiculously safe on this

Then after heating for a minute or two, there were bright orange flames, which lasted about 10 seconds. I heated for another minute, just to make sure the reaction was done, and let it sit for 3 minutes. On the bottom of the lower bowl, there was a very thin white covering/paste, which when I scraped it into xylene, caught on fire in the air, some of it falling to the ground in burning spheres, some catching the xylene on fire. DON'T WORRY, I was prepared. I covered the can, the fire went out. Now xylene ought not to eat sodium, but the mixture started fizzing and frothing-this is from residual moisture I assume, could it be water from the fire itself? So that was all wasted. :mad: Will vegitable oil work?

Upon opening the lower bowl, there were about 6 different globs that glowed bright yellow. A couple minutes later, the globs had turned a dull pale green. The whole mass of reactants had hardened into a cylinder at the bottom of the lower bowl by now. This was lifted out, several sections looked metallic. Anyways, there were several extremely small balls of sodium that fell out from somewhere,
they fizzled in water. Ahh.

Sorry Tacho, I did not try the reaction with NaCl, Na2CO3 and NaOH. That will have to wait until tomorrow.

[Edited on 26-6-2004 by Cyrus]

[Edited on 26-6-2004 by Cyrus]

Great Work! Congratulations!

Tacho - 26-6-2004 at 08:00

Keep beeing careful!

I have never used Mg, but I can confirm 2 things:

1) Xylene DOES fizz a bit when you put sodium in it, probably due to residual water (maybe that's why they use sodium to dry organic solvents... duh!).

2) the little balls of sodium do NOT catch fire when tossed on water, they just dance on the surface, fizzing, until they disapear.

Also: Silicone oil is not good to keep sodium. It reacts. At least mine did.

I find the yellow and green colors quite surprising! I can't explain that.

Edit: I'm having a feeling that molten paraffin may be the best way to separate sodium. Keeping sodium in solid paraffin also sounds good.

[Edited on 26-6-2004 by Tacho]

[Edited on 26-6-2004 by Tacho]

Cyrus - 27-6-2004 at 15:27

Later I put about 0.3 cc of the residue from the lower bowl into a test tube, and added about 2 ml of H2O. Wham, instantaneously, the test tube heated up, and a jet of orange flame shot out. Ahh, must be lots of Na, or is there? Earlier Tacho said that fire and H2 was not a definite indication of Na. What else could it be?
MgO? nope
Na2O? I don't think so. Na2O + H2O
The links by Polverone to Murspratt sp? on the first page of this topic suggest that other oxides of sodium may be formed-it mentions a dirty green NaO3 IIRC. That is what I believe the green residues are.

Filtering under paraffin might work, (murspratt? suggests dropping the liquid Na through course linen to remove impurities) but Na would be a pain to store that way. Just imagine all the hassle to get a piece out and seal up the jar again.

[Edited on 27-6-2004 by Cyrus]

Tacho - 28-6-2004 at 03:02

Originally posted by Cyrus
(snip)What else could it be?

I'm pretty sure you got sodium. Didn't you see little metalic balls that react with water? Well, I think that's it. Squeeze them, sodium is soft.
Would magnesium little balls react with cold anhydrous alcohol? Maybe that's another test.

The Al-NaOH thermite without NaCl yields a grey mass that reacts with water but has no sodium (doesn't react with dry alcohols). Many people have mentioned hydrides.


Filtering under paraffin might work, (murspratt? suggests dropping the liquid Na through course linen to remove impurities) but Na would be a pain to store that way. Just imagine all the hassle to get a piece out and seal up the jar again.

[Edited on 27-6-2004 by Cyrus]

On the other hand, liquid solvents keep absorbing moisture from air, like xylene, I think paraffin in a test tube would do a better job in keeping it. Just melt it using a flame before use.

[Edited on 28-6-2004 by Tacho]

Cyrus - 29-6-2004 at 13:44

Yes, I got some sodium, but I was wondering what the other stuff was. Sorry for the dumb question I forgot all about hydrides.:(

I am planning to make a charcoal furnace, so then NaCO3 + 2C -> 3CO + Na. The Na vapors will bubble into molten paraffin.


ballzofsteel - 29-6-2004 at 21:23

This is a patent from Castner.
He uses tar as the carbon source,mixed with iron oxide,which is then calcined in the absence of O2.
The reduction is told to take place @ red heat temps.
50%+ yeild from Naoh.


Tacho - 30-6-2004 at 03:47

Originally posted by Cyrus
(snip)Na vapors will bubble into molten paraffin.

Sorry, but I doubt it. Sodium boiling point is 892ºC. It will condensate long before it reaches the paraffin. Or it will do terrible things to the poor paraffin when it gets there at this temperature.

[Edited on 30-6-2004 by Tacho]

Cyrus - 1-7-2004 at 12:55

Ok, I changed the design slightly- if you can decipher the poor image from the charcoal furnace thread, the Na ought to condense into a liquid in the pipe and run into the molten paraffin.

ordenblitz - 15-8-2004 at 10:39

It has been pointed out that dillutents like sodium silicate, sodium chloride or Magnesium oxide can or have been added to the NaOH / Mg Thermite reaction to slow it down.
I was thinking that there could by a variety of coolants that might function better for this application. One that comes to mind is magnesium carbonate. Besides cooling it should also create a blanket of gas to protect the sodium as well. Also might one try consolidating the mixture by compression before ignition to slow it down thereby making it easier to collect the sodium vapor?

Sodium (and others) from Halide salt reduction with CaC2

chemoleo - 2-9-2004 at 14:20

According to US patent 4,105,440, CaC2 reduces alkali/earth alkali metals when their salts or eutectic mixtures are molten, and CaC2 is added to it.
I guess this is only feasible if you have a furnace or similar, as high temps (betw. 700 and 1000 deg C) are required.

The reaction is

CaC2 + 2MeX ---> CaX2 + 2 me + 2 C (finely powdered graphite).

X is a halide, i.e. Cl, Br, and Me is the alkali metal.

I am not quite sure how to extract the (i.e. Na, Ba) from this, I haven't read the patent to the end, but just wanted to add this to this thread for completeness' sake.

[Edited on 2-9-2004 by chemoleo]

JohnWW - 2-9-2004 at 17:17

But, surely, a problem in that patented process, if it is genuine, would be reaction of the alkali metal produced with the graphite byproduct to form an alkali metal carbide? The only way this could be overcome would be by rapid removal of the Na as the vapor (above its boiling-point at the reaction vessel pressure), to be condensed elsewhere, as soon as it is formed. In fact, the whole scheme, involving displacement of Na (more electropositive) by Ca, would be an equilibium reaction depending on volatility and removal of the Na vapor for its success.

John W.

chemoleo - 2-9-2004 at 17:28

No, it is not a simple displacement because your starting point isn't Ca, it's CaC2! It's not just a thermite reaction, using Ca instead of Al!

As to genuineness, read the patent and judge for yourself, before speculating.
As to isolation, Mg can be made this way, and it doesn't seem all that difficult - apart from the heat and the necessity of a protective atmosphere. But then the Mg can be simply decanted off... isnt that cool?

[Edited on 3-9-2004 by chemoleo]

BromicAcid - 2-9-2004 at 17:47

Very true, this is very similar to the reaction that I have seen that Dupont used to use for the production of sodium.


The Dow Chemical Co. has recently patented a process for producing sodium by distillation of a mixture of carbon and sodium carbonate fused in an electric arc furnce at 1200 C [224]. The sodium vapor is condensed by rapid chilling in a lead alloy, containing 5-15% of sodium, at 375-400C. Part of this quenching liquid is continuisly withdrawn to a still in which the sodium is removed at 600C. Good efficencies are claimed by the Dow Co. for this process. Similar processes use high-frequency induction furnaces for the reduction of sodium compounds by granular graphite [26], of for the reduction of sodium chloride by lime and coke [212]. Calcium carbide has been proposed as a reducing agent [65]...

[26] B.P. 486930
[65] French P. 828712
[212] U.S.P. 2200906
[224] U.S.P. 2391728


The thermal reduction methods in general utilize carbon or a carbide as the reducing agent. [22]

6NaOH + 2C ---> 2Na2CO3 + 2Na + 3H2

A mixture of rubidium chloride or cesium chloride with calcium carbide heated to 700-900C in vacuo gives a 75% yield of the alkali metals. [23] With sodium chloride a temperature of 950C is used. [24] The production of potassium is reported using silicon or calcium carbide as the reducing agent at a temperature of 100-1150C.

2KF + CaC2 ---> 2K + CaF2 + 2C

Part of the KF may be substitued by K2CO3 or K2SiO3 without any loss in yield.

2K2CO3 + 3Si + 6CaO ---> 4K + 2C + 3(2CaO*SiO2)

These methods usually require good vacuums at high temperatures.

[22] G.L. Putnam, Ind. Eng. Chem. 30, 1138 (1938).
[23] V.D. Polyakov and A. A. Fedorov, J. Applied Chem. (USSR) 13, 1833-8 (1940) [C. A. 35, 5049 (1941)]
[24] P. V. Gel'd et al., J. Applied Chem. (USSR) 20, 800-8 (1947) [C. A. 42, 4478 (1948)]

From "Comprehensive Inorganic Chemistry".

BTW, sodium carbide is only stable up to 400C, therefore removal before reaction is not a problem, same with hydrogen, sodium hydride is only stable upto a lower temperature, it can safely be held under a hydrogen atmosphere in a Castner cell without reaction, it maintains a mirror sheen to it in such a case.

Ohhhh... I'm posting this without spell checking it, bad boy, bad! :D

Edit: Spelling.... ;)

Oh, and... Cyrus
Na vapors will bubble into molten paraffin.

Run the vapors into a copper spiral condenser immersed in heated mineral oil to the melting point of sodium, the sodium will condense and the liquid sodium can run into your paraffin :D

[Edited on 9/3/2004 by BromicAcid]

mick - 5-1-2005 at 12:18

I have just received a book through Ebay, signed by Miss Phebe E. Travis, (Cou/nton?+ something I cannot read) Sept. 28th. 1857.

Preparation of potassium.
(a quick search has not shown anything similar)
My typing is not to good but here goes.

The expensive and troublesome method of procuring this metal by galvanism, has been replaced by a much more convenient and productive furnace operation, founded on the decomposition of potash at white heat by charcoal. For the pupose carbonate of potash is mingled with charcoal. This mixture is best prepared by ignited cream of tartar in a covered crucible; a black mass is then obtained, commonly known as black flux, consisting of carbonate of potassa in intimate mixture with charcoal derived from the burning of the organic acid. This mass is finely powdered, and 1/10 of charcoal in small fragments is added. The mixture is then placed in an iron bottle V ( fig. 348) laid horizontally in the furnace. The bottle should be about 3/4 full, and well protected with a refractory lute of 5 parts fine sand and 1 part fire-clay, laid on moist, and well dried in the sun.
(If anyone is interested I can copy the diagram and post it)
The cover of the furance (M) admits the fuel, the draft (O) is regulated by a damper, and a temporary front (r,n) closes the side-opening. A short iron tube (a,o) connects the retort with a copper condensing chamber (A, B, C) containing naphtha, and supported on (T, P S, from diagram). The heat is gradually raised to the most intense whitness. Decomposistion of the carbonate of potash ensues, the free carbon takes the oxygen of the carbonate, carbonic oxyd is evolved, and the potassium distils over in metallic globules, which condense in the receiver (A).

The receiver is made of copper and the potassium is collected under naphtha. It also tells you that one of the pipes can become blocked and how to keep it clear.

The comment about sodium is that it can be made the same way as potassium


Some of the spelling mistakes above are not mine, I copied them.

Edit mick

[Edited on 5-1-2005 by mick]

BromicAcid - 1-5-2005 at 12:01

10 grams of potassium hydroxide was mixed with 5 grams of magnesium metal and placed into a pipe with a one way gas exit ball check valve. The mixture in the pipe was heated from the bottom with a propane torch and after a few minutes a reaction commenced evidenced by a popping sound from within the container, upon hearing this the exit gasses were ignited and they burned for a mere thirty seconds. Heating was continued for fifteen minutes and discontinued.

After an hour the vessel was opened and after removing the cartridge portion the upper half caught on fire (possibly from potassium that condensed therein). Gobules of potassium were at the bottom of the vessel but for the most part there was a tower of unreacted material in the middle (should have incorporated my KOH with my Mg better). A gobule was removed with a metal rod, it came up cleanly, potassium.... And ignited, actually everything ignited, even cold it seemed to be quite reactive. I dumped out the contents, there were areas of grey at the bottom and green areas in the matrix, from what I don't know. When the 'gunk' at the bottom was broken apart there was small amounts of potassium in the matrix.

Allow me to reenerate that I do these reductions between magnesium and potassium hydroxide because literature states (as do experiments by other members) that the reaction between NaOH and Mg is violent. So next time I have to remember to pour in mineral oil before I start poking around in the reaction products.

ordenblitz - 18-6-2005 at 16:18

After reading the unconventional sodium thread again last night I decided to test my idea of using magnesium carbonate as a coolant to slow the reaction of either NaOH +Mg or KOH+Mg down so as to be able to better collect the vapor.

I ground, under argon in a mortar and pestle, 5.6gm KOH then added 2.4gm Mg as -200 mesh granular powder.

Before adding any MgCO3 I wanted to get an idea of just how fast this mix would combust without any coolant. So I put a .05gm +or- on a spatula and passed it into a flame. It didn't really impress me. So I placed roughly .1 gram in a 25 ml erlenmeyer touched it off with a loop of hot nichrome.

This time it was even more sedate, burning rather slowly and in a controllable manner. Since it happened much slower in the flask, under argon then did the small amount I burned in the open. I figured atmospheric oxygen might be having a greater effect then anyone had guessed. Was a coolant even needed?

Looking at the reaction: NaOH + Mg --> MgO + Na + 0.5H2
There isn't much hydrogen generated and so it might be possible to totally confine the combustion and greedilly capture all the alkali thus produced.
I remembered that a while ago, I picked up an old Parr oxygen combustion bomb kit from LabX for almost nothing. It has since been gathering dust waiting for something to do.

Same as above, I ground 1.12gm KOH with .48gm Mg under argon and loaded it into the parr bomb. I threaded 7cm of nichrome on the studs and capped it while still in the inert atmosphere. Not knowing exactly what would happen I took it outside and placed the aptly named "bomb" behind two steel blocks, connected the wires and made a hasty retreat. I pressed the button and nothing…. not a whimper. Disappointed I approached the bomb and touched it and to my surprise it was warm ~90c.
I let it cool in the freezer for a bit and then used the wrenches to crack the top. There was only a weak little hiss as my .02014 grams of hydrogen escaped. I placed the bomb in mineral spirits to finish the opening and let the contents flood. The interior was remarkably clean and most of the contents had stuck together in one porous chunk. The MgO probably floated on the potassium, that more or less pooled to the bottom of the steel cup. I dumped the contents and poked around a bit to see what was what. You can see in the pic which end of the chunk was down.

I will try an even larger quantity tomorrow and probably tap the bomb while still hot, in an attempt to consolidate the hot metal into one lump. Then I will try it all again with sodium.

I don't know if it really is necessary exclude all oxygen in the bomb but it probably does help in keeping the combustion pressures lower. I think I should try a smaller load, with normal atmosphere to see.

One could probably cobble together a suitable combustion chamber out of commonly available materials. The wire pass through might be a problem. I think it's a phenolic sleeve on the Parr.

Marvin - 19-6-2005 at 04:24

I think if you are happy to swap magnesium metal for sodium or potassium its worth reading US4725311.

The patent was dug up by someone who has since removed all his posts. Its a way of achieveing the same result without making close relatives of flash powder mixtures and setting them off inside sealed containers. I like the look of the solvent method a lot and plan to try it when I get some cheap block magnesium. I would be inclined to wonder if you had diluted with magnesium carbonate if that would have been reduced by the magnesium itself. I have the nasty feeling that would have been a rather better flashpowder what with the effect of carbondioxide on magnesium fires.

ordenblitz - 20-6-2005 at 17:08

Yesterday, I decided to try making sodium using the pressure bomb method. I wanted to see the difference between Na and K when reacted with Mg in an inert atmosphere. Some have suggested that Na + Mg makes a near flash powder. But just as my potassium experiments showed the previous day, that only seems to happen out in the open and not in a negative oxygen environment.

2.43gm Mg + 4gm NaOH were ground to a fine powder in a mortar and pestle while under argon. A small amount ~.1 gm was transferred to a small flask that had been flushed with Ar, placed in a hood and lit with a hot wire loop. It was not much more impressive than my similar KOH+Mg experiment.

On to the confined reaction… I placed ~2gm of the above NaOH+Mg mix in the parr bomb and sealed while under Ar.

This then was wired up and set out to fire. I still use the blocks even though from the outside, nothing much seems to be happening. Picture taken 30 seconds after ignition.

After allowing the bomb to cool for 10 min or so, I cracked the seal while waiting for a hiss of escaping hydrogen but none was heard as before in my K experiments. Very interesting I thought, where had it gone? The contents after opening under mineral spirits and washing with 1, 4 dioxane, were very different then the potassium experiment. You can see light and dark silver/grey and blue coloration.

Sodium hydride.
Merck says: prepared by passing hydrogen into molten sodium dispersed in oil or mixed with a catalyst such as anthracene above 250º. Reacts explosively with water, violently with lower alcohols, ignites spontaneously upon standing in moist air.
So this method isn't going to work if one is after solid sodium metal but I think it has promise for potassium since it does not form hydrides in this reaction. I am still working on improving the Kbomb and will post the further results as I have them.

The upper layers in the parr cup were lighter in color and certainly contained a greater percentage of MgO and were a brittle crunchy mass. The lower layers were more metallic yet not like sodium is supposed to look. The picture is of the solid upper mass that came out of the cup.

Upon tossing the chunk into water this is what happened.
Not nearly as fun as what happened when small portions of the lower layer were tossed in.

For those who don't happen to have an oxygen combustion bomb lying around, one could make a heavy metal, threaded capped tube and drill and tap one end for a small spark plug. One could then attach a fine nichrome wire between the center electrode and the ground tang. Voila instant reaction chamber.

Marvin, of course I would trade magnesium for K or Na since I have plenty of Mg lying around and none of the latter. I read the patent you referenced and thought I might try a small version in test a tube and see if it might work. I placed a stoichiometric mix of NaOH + Mg in heavy mineral oil and began heating. The contents did become very frothy and I continued this heating for roughly 20 minutes. After that time I added a few ml of 2-propanol as a catalyst as suggested in the patent. I continued heating for another 30 minutes. After heating the contents settled and no visible difference in the magnesium was noted. I washed the contents on filter paper with 1,4 dioxane and tossed it into some water looking for any reaction. There was none. It's possible that the mineral oil interfered or the reaction times were too short.

Rxninoil.JPG - 44kB

Marvin - 22-6-2005 at 06:32

I think for any reasonable preperative amounts that reaction is going to be difficult to handle even aside from problems extracting the sodium or potassium product.

Treat the process in the patent more like a grignard reaction. Started with an iodine crystal or two a dry alcohol will devour magnesium until one or the other runs out, when this starts its obvious and the rest of the reaction should follow. I'm not certain what you mean by mineral oil but it needs to dissolve the alkoxide and be able to reflux at a fairly low temperature. I'd be inclined to follow the process as closely as possible for the first few tests. The only major aggro with this method seems to be seperating the metal from magnesium oxide in the ingot in the end, hense the trickery with controlling the density of the solvent on some of the examples.

S.C. Wack - 23-6-2005 at 11:38

This is Winkler's 14 page article, published in 1890. It presents several Mg reduction experiments. He seems to have had good results with potassium, as has been mentioned before. From Gallica.

Any highlights worth translating?

Attachment: ber_23_44_1890.pdf (808kB)
This file has been downloaded 2999 times

ordenblitz - 23-6-2005 at 18:09

Can anyone who has read the above tell me what procedure was used in the separation of the MgO from the K.

garage chemist - 24-6-2005 at 04:16

@ S.C. Wack: Wow, that is very interesting! I'll translate some parts of it when I have time (need to go now).

garage chemist - 25-6-2005 at 02:29

- They use magnesium for the reduction of alkali metal compounds, the compound and magnesium powder are first dried and then ground together in a mortar. Then the mixes are heated in a glass tube sealed at one end. A test to the behavior of the mix is first carried out with very small amounts.

-The heating of a mixture of 74 parts Li2CO3 and 72 parts Mg produces a violent reaction, and the glass tube is usually shattered.

-Heating a mix of 106 parts Na2CO3 and 72 parts Mg produces a yellow flame from the glass tube, and the inside of the tube becomes coated with a mirror of metallic sodium. All the sodium is vaporized from the mixture because of the heat of reaction.

- Heating a mixture of 138 parts K2CO3 and 72 parts Mg produces no vigorous reaction, but it still reacts and the inside of the tube becomes coated with a mirror of metallic potassium.
The potassium can be isolated from the reaction mixture by conducting the reaction in a glass tube which is open at both ends and after reaction distilling off the potassium (bp 667°C, easily reachable with a bunsen) in a slow stream of hydrogen. The potassium vapor is green and coats the inside of the tube with a mirror in the cooler parts.

The mentioned stochiometric composition of the K2CO3/Mg mix must be correct and the components have to be mixed thoroughly, otherwise, expecially when there is a lack of magnesium, Kohlenoxydkalium (the potassium salt of hexahydroxybenzene/dihydroxyacetylene, a grey dust) forms, which is highly explosive and very dangerous.
The mix also must be heated rapidly and not gradually, also to prevent the formation of this dangerous compound.

The production of potassium from KOH and magnesium is more convenient. The KOH must be dehydrated before use by melting and heating it to red glow for some time.
56 parts of this are mixed with 24 parts Mg and heated in the glass tube, the reaction is more vigorous than with K2CO3, but no carbon is formed and, more importantly, Kohlenoxydkalium cannot form.
A large portion of the potassium is evaporated from the reaction mixture by the reaction heat, the rest can again be separated by distilling it off in a stream of hydrogen (this sounds more difficult than it is- simply do the reaction in a glass tube, after the reaction slowly let in dry hydrogen from one side and heat the mirror and the reaction mix and the K will distill off to form a clean mirror, the K can be collected by dipping the mirrored part of the tube into hot paraffin oil to melt the K).
Doing this Reaction on a larger scale is dangerous because of the vigor of the reaction, but it can be moderated by mixing in some MgO, and using a mixture of 56 parts KOH, 24 parts Mg and 56 parts MgO is without any danger, even on a large scale (kilograms).
The potassium remains in the reaction mixture and can be conveniently distilled off by the method mentioned above.

- Production of rubidium from the carbonate is as easy as potassium from the hydroxide, but the reaction is much slower and less dangerous. The Rb can be distilled off in a stream of hydrogen.
Reduction of RbOH is even easier.

- Reduction of Cs2CO3 with magnesium is not possible, even at temperatures where the Mg begins to evaporate. Only traces of K and Rb vapor are given off, which might be a method for purification of Cs salts from these contaminants.

@ S.C. Wack: I posted a link to your attachment in a german forum, don't be surprised if the download counter goes up.

IrC - 3-7-2005 at 18:21

I like the patent that Marvin posted, especially since I have a bunch of blocks of Mg. Since I was involved long ago in a galaxy far away with work on electroink processes, I know the chemistry of liquid toner copy machines very well. The question being since the dispersants used in say the Savin (TM) line of liquid copy machines are probably the most highly refined and pure isoparaffins you can buy, and quite easy and cheap to obtain, would simple copier dispersant be suitable for use in the process? Looking at the patent they were showing 92% purity mixed with other things, and it seems to me the more pure a substance is the better the process. Has anyone thought of using this source for the solvent?

Cyrus - 18-7-2005 at 06:54

I've caught the Na/K bug again. I'd like to try the Mg + K2CO3 method, except for this "Kohlenoxydkalium" problem. Since this stuff has H in it, I was wondering if its formation could be prevented by not using H2 gas, but CO2 or He, or a vacuum. I like the idea of a vacuum because Na and K volatize somewhere around 250-300 C (O. C. Braur's Prep. Inorg. Chem.) which would probably be obtained just by the reaction's heat. So there would be no messing around with H2 and no secondary heating.

The mentioned stochiometric composition of the K2CO3/Mg mix must be correct and the components have to be mixed thoroughly, otherwise, expecially when there is a lack of magnesium, Kohlenoxydkalium (the potassium salt of hexahydroxybenzene/dihydroxyacetylene, a grey dust) forms, which is highly explosive and very dangerous.
The mix also must be heated rapidly and not gradually, also to prevent the formation of this dangerous compound.

Following US patent 4725311

BromicAcid - 30-10-2005 at 12:46

[Note, the amounts that I used are accurate to within plus or minus 1 gram, due to my scale only reading to the gram level]

16 grams of potassium hydroxide was placed into a 250 ml erlemeyer flask along with 175 ml kerosene procured from my local gas station. To this was added 8 grams of magnesium in the form of shavings and the flask was fitted with a vigurex collum to help keep the liquid in the mixture even close to the boiling point, the top of the vigurex was attached to a hose which lead into some KOH pellets to try to keep water out of the reaction medium though air could still make it in.

The amounts that I use corresponeded to a 25% run of the run in the patent example number 1. The kerosene was clear and upon addition of a small piece of sodium small bubbles appeard on the surface after some time but for the most part it was unreactive. The patent also gives procedures for sodium, however it takes a 15 hour reflux as opposed to the 4 hour reflux for potassium and I just didn't have the time.

Continuing on, the mixture was heated slowly as I was wary of what was described as "a violent eruption of H<sub>2</sub>" the solvent got fairly hot and still no reaction, the magnesium shavings took up a large volume and didn't fit underneath the liquid level. Trying to give the reaction a little start without it going full scale I dissolve ~5 ml isopropyl alcohol in 100 ml kerosene and added 2 drops with no reaction. The mixture continued to heat under magnetic stirring until suddenly some bubbles came to the surface, they looked smokey. The heat was turned to low and the reaction picked up slightly but lacking furhter heating subsided (Note: this was about 1 hour into a gradual heating).

Now that I was fairly sure the reaction wasn't going to run away on me as badly as I feared I turned the heat back up and some time later the reaction started again. At first it looked like the KOH pellets became clear and they formed a film on the bottom of the flask, the stirring bar didn't have the power to dislodge them. Eventually a nice steady reaction was taking place right at the bottom of the flask where the magnesium met the KOH. This continued for some time and it was evident the magnesium was reacting well, white powdery MgO was being produced which was at first somewhat sticky but at the end was free flowing. More of the magnesium passed beneath the surface and more reacted until it was just flakes of magnesium swirling in the reaction medium but the reaction slowed down greatly. During this time 25 drops of the isopropyl alcohol/kerosen mixture was added and the vigor of the reaction increased slightly for some time then slacked off again. This was 4 hours into the reaction and approximately 75% of the magnesium had reacted.

Being that I was running out of daylight I had to shut down the reaction. I left it stirring with the heat off for a few hours (which was much longer then I intended) while it cooled and then stoppered the reaction for the night. When I came back the next day I again turned on the stir bar and noticed that the mixture was now brownish as well as the powder. I'm assuming too much oxygen got into it and made some potassium oxides or maybe due to the basicity of the reaction medium and such the isopropyl alcohol underwent some reaction.

To the reaction mixture was added a small amount more of the kerosene/isopropyl alcohol mixture and it was again heated in hopes of reaction more magnesium and also to liquify any potassium presnt to allow for me to filter it. However after another hour no apparent reaction had taken place and the brown discoloration had become more apparent. I decided to allow it to cool some more and filter it.

A buchner funnel was lined with fiberglass insulation and the hot kerosene was poured through it to heat the funnel. Then the solid was poured in all at once despite my fears that the potassium present would ignite and ignite the kerosene and ignite me. However things went well and the mixture was filtered but no potassium was in the filtrate. So I looked at the solid and it wasn't bursting into flames. Curious I took some more kerosene on the side and added a fair amount of isopropyl alcohol to it and grabbed some of the discolored magnesium shavings off the top and tossed them into the alcohol mix, vigorous fizzing ensued and more was added with the same result, I noticed that the funnel was now making popping/crackling sounds like it was threatening to catch on fire so I tossed it into a 5 gallon bucket of water causing incredible fizzing.

No potassium was recovered. My reasons for failing, the isopropyl alcohol has such a low boiling point and I was running the reaction at the boiling point of kerosene, the patent recomends t-butanol which I did not have, this was part of my problem. Additionally I think I should have run the reaciton at a higher temperature, intially I was too afraid to go all out and boil it for fear of an incredible gas evolution but it never got that bad, I should have ran it at boiling like the patent recomended. Finally I should have never left it open to the atmosphere for so long, I think the brown discoloration was due to oxygen going into the reaction mixture and destroying my elemental potassium. Also filtration was not the best option, in theory the kerosene should have been distilled off and dioxane added which is more dense then the potassium and causes it to rise to the surface, then it is either skimmed off or filtered off, the solid all fell out at once so I think that stopped efficent filtering from happening.

All in all it was an easy reaction using OTC materials (for the most part) and assuming I get a good alcohol I believe I could get a good yield from the reaction, I believe the inital hydrogen evolution is the first stage of the reaction and the alcohol pushes the complete reduction of the magnesium but when I was done I had 2 grams of magnesium left, so the reaction hadn't gone to completion. I would be greatly interested with someone else trying this reaction with sodium hydroxide.

[Also, does anyone have a method or source for t-butyl alcohol?]

[Edited on 10/30/2005 by BromicAcid]

Marvin - 6-11-2005 at 04:52

Volatility should not have been much of a problem. The alcohol should quickly turn into alkoxide. Potassium alkoxide contacting magnesium metal, producing potassium metal and magnesium alkoxide, and then magnesium alkoxide reacting with potassium hydroxide to form MgO or Mg(OH)2 and pottassium alkoxide again.

I would be inclined to suggest starting the reaction off first - small amount of magnesium metal, quantity of alcohol, trace of carbon tet to kick it off. This should go without heating and if it doesnt work the rest of the reaction will probably not fly.

Then adding the kerosene dilutant and then the rest of the magnesium followed by the KOH in small amounts (while the reaction works to destroy the water).

I would also be inclined to try barbeque lighter fluid rather than kerosene, this is just a gut feeling though based on my understanding this is often pure hexane.

BromicAcid - 6-11-2005 at 10:01

Finally got around to getting the pictures off my computer (which is somewhat broke now) attached is the reaction sequence, the first picture showing all the reactants in the flask before the heat was applied. The second picture shows nearly 4 hours later, the metal still visible is magnesium, most of it had reacted though. However I had to leave and could not work up my solution and as a result I had to leave it and come back the next day, it was left for a few hours exposed to the atmosphere more then I would have liked because I didn't want to cap boiling kerosene. The final picture is from the next day before re-heating to melt any potassium, note that the color is now brown, potassium peroxides/superoxides, oxidations products of those compounds with the kerosene... etc, definatley something considering the striking color change, and as I noted the magnesium turnings recovered gave an almost violent reaction with isopropyl alcohol dissolved in kerosene.

[Again, although this is somewhat off topic this can be thoretically be applied to sodium as well.]

overtime.jpg - 63kB


neutrino - 16-1-2006 at 07:18

What about electrodeposition of alkali metals from glymes? For those of you who don't know what I'm talking about, see this page.

Ionic salts can easily dissolve in glymes, forming complexed cations and free anions. The cations can then be plated out, or in our case electrowinned. This is used industrially for Cu, Cr, and Ni, so I imagine it must be extensible to sodium. There shouldn't be anything to react with the metal once its produced, assuming we safely remove the anion product.

So, would this work? Electrolyze NaCl in glyme, vent the chlorine, collect the solid Na?

Esplosivo - 4-2-2006 at 06:09

Today I attempted the preparation of potassium following the information from the patent posted by BromicAcid. The following was the pathway taken.

30g of finely ground KOH was mixed with 16g of Magnesium powder. To the mixture, 200mL of toluene GPR was added. A reflux setup (with an oil bath) was used to bring the mixture to the boiling point of toluene (i.e. approx. 111 deg C). The increase in temperature was done slowly. At an oil bath temperature of approx. 50 deg C 12mL of absolute ethanol were added to the mixture. On increasing the temp. effervescence was noted, especially above 70 deg C (oil bath temperature). At an oil bath temperature of approx. 90-100 deg C vigorous gaseous evolution occured. Collection of the sample of the gas, using an inverted tube over the mouth of the condensor, and combustion of the gas immediately gave the expected 'pop' indicating Hydrogen.

Reflux of the mixture was continued for a further 3 hours. The mixture was left to cool to 90 deg C (oil bath temp) to check for gaseous evolution. At the 3rd hour gaseous evolution slowed down and I decided to stop there. On examining the contents of the RBF, little KOH was left, some magnesum hydroxide was present at the surface, whereas tiny granules with a metallic lusture were present (which I assume were potassium).

The yield was extremely low, most probably because of the low boiling point of the hydrocarbon used. Next time I will try distilling some diesel so to purify it and run the experiment again. All chemicals were GPR grade. Pic attached shows the product.

PS. I have some other pictures of the gas evolution but I do not know how to use scipics. If anybody is interested and ready to help please U2U me.

[Edited on 4-2-2006 by Esplosivo]

Edit to change picture size-davster

[Edited on 24-12-09 by The_Davster]

Product.JPG - 48kB

[Edited on 2-16-2010 by Polverone]

neutrino - 4-2-2006 at 11:04

This thread describes the process of putting pictures into posts.

Mineral oil should be worth trying. It has a very high boiling point (well over 300*C) and remains stable at high temperatures. Just don’t try boiling it, it will decompose.

What exactly drives this reaction forward? Is it the lattice energy of magnesium oxide/hydroxide formed?

12AX7 - 4-2-2006 at 11:25

Sure. Works with aluminum too, for all alkalines and earths -- though the latter of course need to be vacuum distilled off under high heat. Although, I forget, isn't lithium or cesium not possible doing this?


neutrino - 11-2-2006 at 08:30

Lithium is the impossible one. The patent describes every alkali except that one.

Why is the vacuum distillation necessary? Couldn't the metal simply be allowed to coalesce into one big piece and taken out?

If you're sure that Al will work, I'll try this with shredded Al foil. First I need to find a source of samarium cobalt magnets for a high-temperature stir bar.

12AX7 - 11-2-2006 at 10:45

Well it takes a lot of heat, so sublimating or distilling the alkali earths out under vacuum is more convienient than blasting the snot out of it at white heat or so.

Aluminum might work with the lower temperature method that's been the recent topic, but it still has that problem of forming oxides.


ordenblitz - 27-3-2006 at 20:34

The last time I worked with the NaOH + Mg --> Na + MgO + H thermite type reaction, the result was a porous and friable mix of magnesium oxide, sodium and sodium hydride that was for all intents and purposes impossible to separate. At the time I had done some thinking about the possibility of using a hydrogen scavenger to bind the H and leave the elemental sodium, however no reasonable ideas surfaced so I put the project on the back burner.

Last evening I came up with the following:
4NaOH + 4Mg + B --> 4MgO + 3Na + NaBH4

This was tested today with good results. I took pictures but didn't have time to resize tonight. Tomorrow I will post those.


1.599 gm - Sodium hydroxide, tech
0.972 gm - Magnesium, (-325m spherical)
0.108 gm - Boron 97%, 5 micron.

The above was ground lightly in a mortar and pestle and then placed in the Parr combustion bomb and sealed. The ignition wires attached and fired. The casing became very hot, ~200c, which was allowed to cool and then opened. No pressure escaped nor was any expected. The contents removed with a lab spoon and dumped into a beaker with kerosene. Several large chunks including one of obviously metallic sodium which had agglomerated and solidified on the bottom of the bomb. The surface was sliced into with a blade revealing very shiny silver metal.

For those of you who stopped thinking about sodium 14 lines ago......
This was also tested today:
NaOH + 4Mg + H3BO3 --> NaBH4 + 4MgO

First the magnesium and hydroxide was ground in a heat dried mortar while in a clear poly bag filled with argon. Then the boric acid was added quickly the previous and all was placed in the Parr bomb that had been placed in the bag with the mortar. This must be done fast as the mix begins heating almost immediately. The wires were connected and the bomb fired.

The result was mostly light gray with white streaks. It did not behave the same as the earlier run for sodium. It seemed stable in air but effervesced nicely when placed in water. More testing certainly necessary but I think I am on to a good synth here.
I however have not decided on an easy way to separate the NaBH4 from the MgO.
Does anyone know if N-Methyl Pyrrolidone might be a suitable solvent for this purpose or have a suggestion for another?

Polverone - 27-3-2006 at 23:59

Couldn't the rapid heating simply be an acid-base reaction between the NaOH and the boric acid? Perhaps you could first react the sodium hydroxide and boric acid, heat to drive off the water formed, then re-powder and add the magnesium. I don't think that NaBH4 should effervesce in cold water. Whether you find a novel way to NaBH4 or simply manage to make the "sodium thermite" useful, either way it's quite interesting.

garage chemist - 28-3-2006 at 01:35

In cold water, the decomposition of NaBH4 is almost too slow to be visible. The solution (of reagent grade NaBH4) is only a little cloudy due to extremely small hydrogen bubbles.

However, if any acid is added, especially HCl or something similarly strong, an extremely violent hydrogen evolution is observed.
I think the effervescence of your NaBH4 product in water was due to residual boric acid which decomposed part of the NaBH4.

You should try to extract your (powdered) NaBH4 reaction product with dilute cold NaOH solution and add some HCl to a sample of the liquid.
If there is any hydrogen evolution, you can be sure that NaBH4 is present.

Industrially, NaBH4 is made by reaction of trimethyl borate with NaH suspension in mineral oil. It gives an aqueous solution of 3 mol NaOH and 1 mol of NaBH4 as an intermediate product.

From this solution, the NaBH4 is isolated by extraction with isopropylamine (2- aminopropane) as solvent.

Isopropylamine could be made by reductive amination of acetone. I think that you'll be able to find something on the general procedure for reductive aminations on Rhodium.
After all, amphetamine is made by reductive amination of phenylacetone.
You'll be doing the same, just without the phenyl.

jimmyboy - 28-3-2006 at 02:37

so boron is the way to go? is boron easy to make? what i have read so far is melt down boric acid for the oxide then reduce with magnesium - so we are basically using the magnesium twice.. first to make the boron then the sodium metal

too bad boron is so friggin expensive :(

[Edited on 29-3-2006 by jimmyboy]

ordenblitz - 28-3-2006 at 18:29

Pictures as promised.

This is the magnesium and boron I used for the reaction producing mostly sodium.

The first shot is of a chunk of reaction mix pulled up from the bottom of the Parr cup. You can see the sodium has pooled and solidified there. I gently tapped the assembly not long after firing while still very hot.
I scraped the bottom with a knife exposing a fresh silvery layer.
A shot from the side where you can see the magnesium oxide layers over the metallic layer.

Tossing selected pieces into some water.

This was the result from the NaBH4 tests. As you can see the material is more powdery right out of the cup. It is not brittle as the above mixes came out. The color is lighter as well.

As Garage Chemist suggested, I extracted the powder with a cold NaOH solution.
The beaker on the left contains 10% HCL solution, the vacuum flask on the right contains the suspect NaBH4 extract in NaOH solution.
The result after they were added together.
A good deal of gas immediately evolved smelling like hydrogen and HCL.

Magpie - 29-3-2006 at 11:38

Ordenblitz this is some really fine work - producing elemental sodium and NaBH4 all before supper. This is a quantum leap for MadScience! ;)

I'm intriqued by your Parr bomb calorimeter. I have never seen one before. Do you use it in support of your "real" work? What size is it? Can you show us a picture of its parts in more detail and describe their features? I looked on the Parr website but they don't give much for construction details.

ordenblitz - 29-3-2006 at 17:05

Thanks but I was only building on other peoples work.

Bromic's posts in the NaBH4 thread got me to thinking about my work on sodium that was again an expansion on his fine experiments. There may be other scavengers for the H that would work better or be easier to obtain than Boron and that should be the next direction for the evolution of this sodium process.

The big problem in making NaBH4 via this method is identifying available solvents or processes for its extraction. Isopropylamine has been suggested but I am afraid that the synthesis would be quite involved. Maybe someone with some NaBH4 could do some solvent testing.

I wanted to post more on this in the NaBH4 thread but the two methods are so interconnected in my process keeping it together seemed appropriate.

As I mentioned earlier in this thread I acquired the Parr setup on Labx for next to nothing.
It is pretty old, circa 1950. I have been to Parr's site as well and really cant find anything that is similar there. I think these things fell out of favor when machines like TGA and DSC came along. Parr still makes some specialized bombs for oxygen combustion work though but they are very expensive. I'm sure there are a few more orphan units like mine, lurking in university labs looking for a good mad scientist to take it home and love.
Making your own out of SS pipe and a modified spark plug would be just as good or better even since it would have far more internal space for actually doing some production. One would only need to drill and tap out an end cap for a small diameter, long reach spark plug... think small engine etc. I would bend up the ground tang then get some fine diameter nichrome wire and wrap around the center electrode and the ground tang leaving a loop for contact to the powder. Fill, cap and spark it up!

The_Davster - 29-3-2006 at 17:49

So thats why a bomb calorimeter is used and not an open reaction vessel....

I added 0.4g NaOH prills, 0.8g Mg filings and 0.5 g boric acid to a 100mL beaker and mixed well with a stirring rod. I then added two scoops of this mix to a testtube which was then heated over an alcohol burner. White fumes were given off then a orange fireball shot out of the tube, bounced once, then came to a stop leaving a glowing pile of something. The glowing pile(after it cooled) was added to cold water, the aqeuous part decanted and acidified. No bubbles formed. The remainder of the residue in the tube was heated for a few more seconds on the burner, then removed and allowed to semi cool. Cold water was added to the tube and some hissing was observed, it might have been some sodium in there reacting with water or it might have just been really hot still. The aqueous layer of the tube was decanted and acidified. Strong bubbling occured.:)

Seems simple enough, but now I gotta air out my basement....:P

But in any case, this is a cool method for making borohydride or sodium, I just gotta get some pipe parts I guess. What are the chances of by doing this reaction in a sealed container that it could rupture and kill me? Or is the pressure produced minimal?

[Edited on 30-3-2006 by rogue chemist]

IrC - 29-3-2006 at 18:26

I would add a tee to your pipe and screw in one on those emergency blow off valves for hot water heaters just to be safe.

ordenblitz - 29-3-2006 at 18:29

It’s a very pretty yellow flame huh Rogue!
If you measure reactants properly this should be a net gasless reaction.
The O is grabbed by the Mg and I suspect pretty fast. If the NaBH4 is forming as we have assumed, the H goes to the borohydride. Maybe next time you try this in an open test tube you might try warming the contents then igniting it, say by dipping in with a steel wire heated red on a propane torch. This should keep the reactants in the tube as it would burn from the top down not the other way round. It is possible that the H would get scavenged before getting away but of this I have no idea. You may be able to determine weather a sealed chamber is necessary or not.
I have had trouble getting the contents to ignite in my bomb from the resistance wire when using magnesium of a mesh size larger than 325. Using the coarser mix you have described, one would have to preheat the chamber before firing to get the thing to start.

I have done the reaction 5 times with varying amounts of mix to better than 3/4 full by volume in the bomb and never did I hear any escaping gas when opening the bomb. One could, for only a few dollars more fit a reducer in one end of the pipe and install a small ball valve to safely vent the chamber before attempting to open it. From what I have experienced so far this is a relatively tame reaction and probably not necessary but for a few dollars of peace of mind, I would.

[Edited on 30-3-2006 by ordenblitz]

The_Davster - 29-3-2006 at 18:47

Sounds like you have had some experiance with the flame as well eh;).

I am thinking the white fumes were just MgO being lofted by hot air in the tube or even potentially water vapour from NaOH absorbing water from the air.

When I have more time(ie not a few weeks before exam time like I am now) I would like to try this again with a tin can with a diameter a couple inches at least and a lid on top held down with a brick. The reactants could be arranged in an inverted conical pile such that in the centre of the bottom there is little reactants and along the outsides the piles grow steeper. Heating would be started in the centre of the bottom of the can. Hopefully such a design could prevent the ignition from lofting everything everywhere.

Also this is an outdoor or garage experiment next time:P

By the way, the sodium borohydride MSDSs list its decomposition as 400, what exactly does it decompose into at this temp?

12AX7 - 29-3-2006 at 20:14

Igniting a mix from the bottom seems like a bad idea...

Don't forget that sodium is gaseous at the reaction temperature. The bomb keeps it pressurized in liquid form. That would explain the homely blob of sodium found in the first test.

So lemme see here,
NaOH + Mg = Na + MgO + 1/2 H2 = NaH + MgO. Lye reacts with magnesium to produce thermodynamically favored magnesium oxide (or likewise with aluminum), yielding sodium vapor and hydrogen gas. However, the reaction products can react further, producing hydride, useless for the goal of sodium metal.

So, you try something like,
NaOH + Mg + B = Na + MgO + BH3 (assuming boron is the stronger 'hydriphile' so to speak). Which works pretty well, but requires boron, so you might use more common materials:
NaOH + Mg + B(OH)3 = Na + MgO + BH3 [unbalanced], but there's too much oxygen and hydrogen to go from boric acid directly: NaOH + 4Mg + B(OH)3 = 4MgO + NaBH4. Same results as with plain lye, except boron has been reduced too. Of course if you want sodium hydride or borohydride, you can use either method, (somehow) extract them from the magnesium oxide, and smile.
One good way to reduce the hydrogen is to dehydrate things:
NaOH + 4Mg + B2O3 = 4MgO + 2B + Na + H, my this has some strong potential, let me rewrite that,
6NaOH + 9Mg + B2O3 = 9MgO + 6Na + 2BH3
Wonderful, it doesn't even need much boric oxide it would seem!

You could also start with sodium (or potassium? :D ) borate, probably a fused product (as with the boric anhydride) to ensure it is anhydrous.

Lemme see, borax is sodium "tetra"borate, which really means Na2B4O7 IIRC, so you'd have to add a good bit of lye even to that already.

Heyyy, when you melt things, gaseous anions tend to go away.. (at least, when glasses are analyzed, you only see oxides listed). You might be able to melt borax (or boric acid) with sodium carbonate, get it good and hot until it stops bubbling, then pour the molten glass into a bucket of mineral oil (obviously, something other than water) to quench it while remaining anhydrous, then wash with solvent, grind and you should have very little hydrogen whatsoever! Why, then you'd get a bunch of plain boron because there would be no hydrogen... oh, so you could reduce the boron present... ah, but the assumption is that sodium carbonate or hydroxide won't turn to sodium oxide on its own, so some must remain...?


[Edited on 3-30-2006 by 12AX7]

jimmyboy - 30-3-2006 at 12:07

boron oxide would be alot easier/cheaper if it worked - and not near as expensive - just add extra magnesium

hmm maybe this can be applied to phosphorus as well?

[Edited on 31-3-2006 by jimmyboy]

The_Davster - 1-4-2006 at 20:37

2g NaOH, 4.8g of Mg filings and 3 g boric acid were mixed in a soup can. The can was tilted adn then returned to normal such that the reactants were slanted inside the can. The lit was placed on the can and secured with a single strip of ducttape. I could not find a brick so I used a slice of railway track on top of the can. A big plastic flower pot was placed over everything and the can heated with a propane torch on the side with the least reactants. After a few seconds of heating there was a hissing sound and the entire apparatus glowed orange...smoke excaped the can and floated up to the garage ceiling. I extinguished the flaming ducttape with some water and opened the garage door....that smoke can't be healthy.... I left and came back 5 min later when the garage was mostly vented and the can had cooled down. The big flower pot was white inside from the white smoke. 400mL of ice/wate was added to the can slowly...there was a small orange explosion so some sodium must have been formed. A sample of the water extract was acidified and strongly bubbled. It is being filtered currently of all the other crud in there. Is there any way that an aqueous solution of borohydride coul d be used to make isoproplamine via reductive amination of acetone, the places I saw were unclear on this?

EDIT: Weird...the aqueous filtrate does not bubble when acid is added, but the residue in the filter does...I thought borohydride was soluble in water? And it worked for you ordenblitz...weird...

EDIT2: Is it possible that boron could form? I got a good deal of black insoluble flakes...

[Edited on 2-4-2006 by rogue chemist]

borohydride.JPG - 23kB

neutrino - 2-4-2006 at 06:45

In response to the question about thermodynamics, the lattice energy of aluminum/magnesium oxide drives the reaction forward.

4Al + 6NaOH --> 2Al<sub>2</sub>O<sub>3</sub> + 6Na + 3H<sub>2</sub>

ΔH = -797.9 kJ

ordenblitz - 2-4-2006 at 09:12

After the reaction, I dumped the contents of the parr cup into a beaker. To this I added a 5% NaOH solution. I also saw a few yellow flashes as well, obviously from some sodium in the mix but there wasn't all that much. I stirred a bit then removed the solution through a 1.3 micron filter. When I added the HCL solution, as you can see from the pictures frothed quite nicely.

It makes sense that the sealed chamber is necessary since after all, the very reactive products of the in initial reaction are not going to politely hang around to form the borohydride without some encouragement from confinement.

I do not know much of anything about how NaBH4 is used in mad chemistry. The problem it seems is how to separate magnesium oxide and the borohydride. Is it possible that the NaBH4/MgO complex could be added to any reductive reaction and the MgO not substantially interfere?

I was wondering something

lacrima97 - 4-4-2006 at 05:36

What if you took lithium and dropped it into a solution of NaCl and some solvent other than water. Something that the Lithium and Natrium were unreactive to? I'm not sure what type of solvent would work though. Could something like diethyl ether be unreactive to the metals, or would this just react horribly. Maybe sodium could be precipitated this way somehow.

[Edited on 4/4/2006 by lacrima97]

12AX7 - 4-4-2006 at 20:48

NaCl is insoluble in ether. You need something better, like uh, isn't it vaguely soluble in pyridine or somethin? Problem is reactivity tends to follow polarity, probably your best solvent (water) is unfortunately the worst!


BromicAcid - 4-4-2006 at 21:03

True, there are not many solvents that would be able to preform this trick, anhydrous liquid ammonia would be a possibility but only 3 grams of NaCl is soluble in 100 g of the stuff so you'd need a lot of ammonia though lithium chloride might be significantly more soluble allowing more the NaCl to be added slowly and in ever increasing amounts. Provided you had the proper reaction vessel you could also make it the same way potassium is made from sodium, via distillation. Distilling sodium from a mixture of sodium chloride and lithium metal would be one fun route. But why waste lithium metal, it burns prettier then sodium and has a better electropositive value. Still, I guess it is one possible viable way to sodium.

Marvin - 6-4-2006 at 04:35

Sodium metal is soluable in liquid ammonia, as are most of the alkali metals. The solution eventually decomposes to hydrogen and the hydride but for a while, its blue.

Also the 'lithium is stronger' is not really true, the unusual ordering only applies to the ions in water.

BromicAcid - 6-4-2006 at 07:16

It decomposes to the amide and hydrogen via:

2Na + 2NH<sub>3</sub> ---> 2NaNH<sub>2</sub> + H<sub>2</sub>

However if your ammonia is very pure this decomposition is retarded and one can attain the alkali metal by evaporation of the ammonia, first getting a golden solution and finally the metal itself. Still, you need good ammonia and it takes lithium metal, I know the potentials are only good for aqueous solutions but it's still something to go by.

skullandfeather - 12-9-2006 at 21:21

thats an amazing borohydride synth!

could lead be used instead of Hg in brine electrolysis?
i know that Na-Pb alloy is used to dry ether, could it be used as sodium amalgam is as well?

12AX7 - 13-9-2006 at 06:35

I doubt it, it's notoriously difficult to get liquid water together with liquid lead (or bismuth).

It would happen in an autoclave under intense pressure. I doubt the little sodium content would not react under that kind of heat.

Easier, but still under pressure, would be the tin-bismuth eutectic. If your goal is distilling sodium from it, this would be easy as tin and bismuth have high boiling points.

You might take an alloy such as Wood's metal, which melts on par with hot water, but there are volatile elements like cadmium and zinc which would distill off with the sodium, if your goal is seperating sodium. If you just want amalgam, the other metals may cause trouble.


 Pages:  1  2    4