Sciencemadness Discussion Board

unconventional sodium

 Pages:  1    3  4

Polverone - 8-7-2003 at 16:47

Glass, fibrous or otherwise, is going to dissolve in molten NaOH. Are you an engineer by day, Orgy? You seem to have quite the love of Nifty Apparatus.

me stupid !

Organikum - 8-7-2003 at 18:33

Of course you are right Polverone - I reread where I thought to have read about a porous fritte of glass and it says not glass but quartz and it is told in relation with lithium electrolysis.
- I looked the properties of quartz up and the resistance against alkali seems to be very low? Oha!

Well I have no quartz, doesn´t matter....

As iron seems to work well I suppose steelwool to be the material of choice - perhaps it is possible to press it for a semisintered consistence?

A engineer by day? No, by daytime I am sleeping for I would decompose in sunlight......

But yes nifty apparatus are more my turf as formulas and reagents with unspeakable names. But my love to electrochemistry has cooled out since I discovered the world of pyrolysis and - new - had a look in photochemistry and ozone.
Ah! Deadly new flames on the horizon! UV! Ozone!
Electrochemistry for organic synthesis is diappointing I learned. To much effort, this makes no sense exept if used for preparation/regeneration of reagents like manganeseIII compounds (in situ so any possible) . Ozone has also quite restricted use - in KMnO4 synthesis perhaps? But photochemistry is like the hot tube! Endless possibilities.

(there are designs available which are not as demanding in glassware - the quartztubes alone are very expensive here)

please share

Polverone - 8-7-2003 at 19:01

Pyrolysis? Some months back madscientist told me with great excitement of things he was learning about pyrolytic processes from some book or other. Unfortunately, he doesn't do book scans, so I never got to look at what sort of amazing things he was reading about.

I'd also be interested in hearing more about photochemistry. It seems that it won't be as affordable as pyrolysis and tube furnaces, due to all the glass (quartz?) needed and the expense of high output UV lamps. Still, I am always interested in lesser-known synthetic processes, especially those that might be realizable by an amateur.

No doubt it would take some time to write down all you have learned, but maybe you can share the documents that have most excited/enlightened you.

BromicAcid - 15-7-2003 at 19:20

I posted this on sci.chem awhile ago and one of the regular's there said this process sounds dangerous, the information comes from V.A. Plotnikov and Z.A. Yankelevich, Mem. Inst. Chem. Acad. Sci. Ukrain (USSR) 9, 420-33 (1936) ; Chem. Zentr. (1937), I, 3450-1

"... the alkali chloride of bromide is fused with aluminum chloride or bromide and the fused mass dissolved in nitrobenzene and electrolyzed. By this method, lithium, sodium, potassium, and rubidium can be deposited on the cathode. The alkali ion serves as the cation, and the aluminum appears in the complex anion."

I always thought that way was interesting, here's another.

"The mobility of sodium ions in a soda-lime-silica glass at elevated temperatures is fairly high; if an evacuated bulb of such a glass is dipped into molten sodium nitrate and electrolysis is brought about by bombarding the inside of the bulb with electrons, the circuit being completed with an electrode in the sodium nitrate, then metalic sodium appears in the bulb."

This came from an eBook that I have about lab glass, I can personally vouch for this reaction working although I cannot give volatages, and best of all sodium nitrate melts at a realatively low temperature and is safer then a molten chloride or expecially the hydroxide.

Theoretic - 12-8-2003 at 06:05

The McGraw-Hill "Handbook of inorganic industrial chemicals" says Na reacts with NaOH between 300 and 385 degrees. Don't know why it doesnt go above that.
Check out this. Sodium acetate is soluble in ether, and I'm sure it's ionized, thus electrolytic extraction of sodium can actually be carried out at room temperature! There were other solvents, but they all would react with sodium.

Madog - 20-9-2003 at 11:41

last night i took some dry sodium acetate and added it to some ether in my lab, the fumes were getting me high... anyways, i swirled it around and warmed it and then, not really seeing any disolve, i just poured off the ether into another beaker, hopeing some was disolved. i proceeded to hook up 2 electrodes in it with a car batery charger, it didnt conduct. :-(

madscientist - 20-9-2003 at 13:52

Are you sure the sodium acetate was truly dry? It might have appeared so while actually being a hydrate (probably NaCH<sub>3</sub>COO*3H<sub>2</sub>O).

Theoretic - 29-9-2003 at 04:55
Looks like anhydrous acetate is insoluble in ether :o , but the hydrate is soluble.
On dissolution, the hydrate is likely to give off water as another layer, but water is a few percent soluble in ether and thus will have to be removed-CaCl2 or CaSO4 or something...

[Edited on 29-9-2003 by Theoretic]

Theoretic - 29-9-2003 at 06:16

Synonyms: sodium acetate anhydrous
Molecular formula: CH3COONa
CAS No: 127-09-3
EC No: 204-823-8
Physical data
Appearance: white crystals or powder
Melting point: 58 C
Boiling point: 120 C
Vapour density:
Vapour pressure:
Specific gravity: 1.528
Flash point:
Explosion limits:
Autoignition temperature: 600 C
Water solubility: substantial"
Just the thing for molten electrolysis!

rikkitikkitavi - 30-9-2003 at 11:26

dont think it would help to dissolve sodium acetate in ether , since it is not ionized in this solution, hence non conductive.

some solvents can dissolve ionic components into ions , but not all. It is just a matter of finding the right solvent where Na+ + e- => Na is going to take place at the cathode. Aluminium has been electroplated in some water solvents , but I dont remember what they where .


Theoretic - 2-10-2003 at 04:30

Not ionized? You mean floating around in clusters of anions and cations? Why would it dissolve then if the energy of solvatation isn't released? :o

rikkitikkitavi - 2-10-2003 at 09:27


it is the molecule itself that is solvatized.

Also probably because sodium acetate is not a strong ionic compound as f e x sodium chloride.

If there where free ions , there would be conductivity, since free ions always move in an electrical field, and that will lead to a current flow = conductivity. So I simply deduct that sodium acetate is not ionized in ether.


BromicAcid - 2-10-2003 at 17:22

This may prove to be useful information. Recently I was researching a way to lithium metal production and came across two interesting things involving electrolysis. Electrolysis of a solution of lithium chloride in pyridine yields lithium metal. You can also electrolyze a solution of lithium chloride in ethanol and actually get lithium metal. One source said the reaction of lithium with chilled ethanol is negligible and electrolysis is possible after a certain saturation of lithium in the solution is achieved. I highly doubt that any alcohol could be used for electrolysis of a sodium salt but maybe pyridine. Also there was an article I posted back when the mentioned electrolysis of sodium chloride with aluminum chloride in nitrobenzene. Although obtaining nitrobenzene can be expensive.

Marvin - 5-10-2003 at 19:34

Ive been thinking a lot about organic solvents, and the problem thats been plagueing me, is that as soon as you have a useful cathode reaction for produce sodium you then have to worry about anode reactions.

In ethanol things might be workable, but what is being produced at the anode, and how are these products kept seperate from the sodium compartment without killing the electrolysis. Furthurmore for me if its destorying the solvent at the same rate its producing sodium and this is something exotic like pyirdine or nitrobenzene, that for me makes it unworkable.

One phrase bothers me, what is actually meant by the solution becoming saturated in lithium?

aluminum plating in water

Mr. Wizard - 5-10-2003 at 20:02

Originally posted by rikkitikkitavi
dont think it would help to dissolve sodium acetate in ether , since it is not ionized in this solution, hence non conductive.

some solvents can dissolve ionic components into ions , but not all. It is just a matter of finding the right solvent where Na+ + e- => Na is going to take place at the cathode. Aluminium has been electroplated in some water solvents , but I dont remember what they where .
I was filling some balloons with hydrogen by reacting aluminum electrical cable with lye and water. The reaction wasn't going as fast as I wanted, so I added some copper wire scrap to provide contact with the aluminum and form an electric couple and speed up the reaction. I didn't notice any difference in hydrogen production, but on emptying the bottle of its liquid, I found the copper was plated with aluminum! I never tried it again, but your comment made me remember it. Maybe the nascent hydrogen had something to do with it?

not electrolytic - but back to making sodium!

Magpie - 4-11-2003 at 16:48

I'm fascinated by the Castner process and want to try it sometime but I'm wondering if there isn't an easier way to make sodium. My baby chemistry book states that sodium was originally made by the reduction of sodium carbonate to Na and CO using carbon and heat. I did some thermodynamic calculations to see what temperature would be needed to get a spontaneous reaction (negative free energy change). This came out about 1200 deg C. Sodium boils at 892 deg C so I don't think you'd want to go much higher than the melting point of sodium carbonate which is 851. Evolution of CO would help drive the reaction. An argon or N2 blanket would be needed to protect the floating liquid Na. I realize this would take a muffle furnace and an argon bottle, but electrolysis doesn't sound real easy either.

Incidentally, has anyone tried a standard car battery charger (6v/12v) as a power supply for electrolysis?

alkali metals sans electricity

Polverone - 4-11-2003 at 17:54

For the carbothermic production of potassium as practiced in the 19th century, see and the following page. See also for the modifications pertinent to making sodium.

Lithium carbonate can be reduced with aluminum powder at a lower temperature (in vacuum or under inert gas blanket) to yield a mixture of lithium and carbon, IIRC.

carbothermic metal production

Magpie - 4-11-2003 at 19:46

Thank you Polverone for those most interesting references to the pioneer making of Na and K. What a thrill it must have been to first lay eyes on those metals! We should always remember that these honored men (Davy, Gay-Lussac, etc)were the "Mad Scientists" of their day.

It appears that indeed it does take a very high temperature to drive these reactions as they are talking about a "white hot" reaction vessel.

chemoleo - 18-11-2003 at 15:45

How about the electrolysis of Na-salts of long-chain aliphatic carboxy acids? such as sodium stearate, oleate, palmitate, or even proprionate, butyrate etc?
According to Kabooms post in another thread ( ) ,the Kolbe reaction should take place. But what happens at the Kathode?? Na might actually sequester there!
Comments anyone?

Theoretic - 19-11-2003 at 08:01

In a non-aqueous, non-ionizable solvent nothing else can happen unless there's foreign cations in the solution. :cool::cool:
I suggest ether, acetone or alcohol. Ass cheap, non-ionisable and dissolve the salts you want to use.

chemoleo - 20-11-2003 at 09:26

I guess I should have been clearer, I actually meant electrolysis of sodium carboxylic acid salts in their *molten*states, rather than dissolving it in EtOH etc.
For instance, sodium stearate should be liquid at 80 deg C. ...but I doubt the conductivity is great. However, I am sure the conductivity gets better the shorter the chainlenght... anyway, this is what I was thinking of :)

KABOOOM(pyrojustforfun) - 25-11-2003 at 20:40

in molten state it probably produces a mix of R-R , RCOO-R , CO<sub>2</sub> & leaves Na<sub>2</sub>CO<sub>3</sub> behind.
edit: just found what ya meant. carbonate wont form due to lack of water. you may get Na @ cathode.
it really worths trying:)

[Edited on 27-11-2003 by KABOOOM(pyrojustforfun)]

Theoretic - 27-11-2003 at 07:25

Molten carboxylic acid salts is what I proposed long ago. :mad: Sodium acetate, to be specific. It melts at 324 C, so a hotplate will do the job, never mind a gas flame. Another MSDS states the mp as being 58 C, but 324 C is supported by two MSDS's, while the 58 C figure only by one. NEVER TRUST MSDS'S ON TEMPERATURES (this is not my first experience of MSDS's not agreeing with each other and being self-contradictory - e.g. in two MSDS's the mp of urea is stated as 133 C... ...and the bp as 135 C!). I do agree that the longer the hydrocarbon chain, the lower the mp, but the viscosity is likely to increase as well, thus slowing down gas bubbles that escape and, with very long chains, turning into froth upon electrolysis. :(

chemoleo - 27-11-2003 at 10:00

Lol theoretic, sorry, I must have overlooked this. You didnt specifically mention Na salts of long chain aliphatic carboxylic acids though. I wasn't thinking of Naacetate, as indeed this has been covered. THe strangely low melting point of Naacetate probably refers to the decahydrate, the 'anhydrous' being a mistake. This one is obviously useless.
Hmm, 324 deg and a hotplate? that gotta be a damn hot one at that!! you are speaking of getting to lead melting temperatures!
Anyway, I think I may try this, with palmitate or stearate.... but also, maybe Na-phenolate!!! (NaOC6H5).
pz :)

PS yes, kaboom that's what I thought too, carbonate shouldnt really form at all! It should be at the anode, unless some strange electrochemical reactions are happening that I am not aware of ;)

[Edited on 27-11-2003 by chemoleo]


Organikum - 1-12-2003 at 13:38

the iron screen is not absolutely necessary in a NaOH-cell. In a smaller setup the use of higher voltages is favorable anyways for to provide a better heating by the electricity so a wider distance between the electrodes doesnt hurt at all.
The concrete is for mechanical/chemical resistance mainly - the glasswool/stonewool is for the insulation. The bottom is left uninsulated as this might be a bad idea - I usually apply the propane/petroleum burner at this place ya know?

Well Gosh darn it all to heck!

Hermes_Trismegistus - 1-12-2003 at 14:54

If it has worked for you in practice, I would be a fool, (not an uncommon state for me to be in) to say that it is not possible.

If Organikum has a digital camera available, I sure we would all love to see his in action.

Unfortunately, up here in Canada, Na is a scheduled chemical so all my posts regarding a Down's cell have been theoretical (of course!)

Looking forward too! ;)

However I still beleive that the more traditional Downs cell to hold alot of promise, not least of which is the evolution of Cl gas.

I see the immediate and best use of the Cl gas to be to bubble it through a large volume of brine solution made up of Iodized salt.

2I- + Cl2 => 2Cl- + I2

The NaCl left over would be extremely pure and suitable for use in the down's cell and a VERY nice byproduct would be the evolution of Iodine.



Let's say we stick with NaOH as a source of sodium and we still wanted to use NaCl as a source AND isolate Iodine in the process......(greedy aren't I :D)

We COULD electolyze strong brine to evolve the Cl gas, run it through more brine to collect our Iodine....

then use the NaOH produced in the above reaction to liberate poor unfortunate Na from the grips of the evil duo OH-


It is true, that in both reactions, a great deal of HCl would be available :P

Now I know what you're thinking.....All that HCl and WHAT TO DO?!?:(

I was thinking since we're playing with on the far left of the table anyway.......:P

why not drip the HCl onto some CaCO3
CaCO3 + 2HCl => CaCl2 + H2O + CO2

then add a little of the precious sodium:P

CaCl2 + 2Na => Ca + 2NaCl

and PRESTO!!:P

Calcium metal and ultrapure non-iodized salt (useful in some obscure reactions)


[Edited on 2-12-2003 by Hermes_Trismegistus]

IODINE ??????

Organikum - 2-12-2003 at 11:54

you might inform yourself how much iodine is in iodinized salt. Tip: less, much much less.

And Na is not allowed for privates here where I live also. So you will have to miss pictures on this forever because I wouldnt make something fOrBiDDeN never. :P

Tintetrachloride is not forbidden here and doesnt belong in this thread but my distorted mind NEEDS to get offtopic at least a little bit.

The NaOH process is older than the NaCl process btw. For being much easier. Chloride is conveniantly made by bleaching powder and muriatic acid, washed with water and dried with H2SO4 for further abuse.
The Cl2 from a Downs cell is mucho shitty contaminated and completely USELESS for utmost everything my dear Hermes.

But you should try it by any way as I believe you will learn it only by the hard way.

Teaser you. ;)

And non-iodinized ultrapure salt is available for "dishwasher regeneration" cheap and without any hassle.

btw. if you add your precious sodium to some water you will get INSTANT NaOH! And vulture has an receipt for turning GOLD to LEAD! ;););)

[Edited on 2-12-2003 by Organikum]

Sodium thermite

Tacho - 5-12-2003 at 08:41

I mixed aluminum powder with solid NaOH and heated it in an iron crucible with a blowtorch until the NaOH melted. Inicially small sodium-orange sparks showed here and there. Eventually a strong reaction happened at once, with orange fire and white smoke. I believe most of the sodium evaporated and oxidized to white sodium oxide fumes, but in the crucible there was a gray concrete-like leftover. It showed strong effervescence when thrown in water, but I can't say it's not just fine aluminum and excess NaOH reacting. It absorbs humidity from air and becomes a bubbling gunk.

I find the sodium salt in nitrobenzene electrolysis a very interesting idea. Why would it be so dangerous? Any ideas? Would it be the risk of fire/explosion in a poisonous media or some cancerigenous mutagenic byproduct? The first option I can manage, but not the second.

Polverone - 5-12-2003 at 08:56

I've read detailed accounts of experiments done in nitrobenzene solution electrolysis of alkali metals. Short summary: the cathodic deposits are impure, not terribly high yield, and explosive.

On the other hand, electrolysis of lithium chloride in pyridine gives clean lithium metal rapidly and easily, but it's not so easy for me to get or work with pyridine. Amyl alcohol and acetone are also supposed to work for the LiCl electrolysis, though not as well.

unionised - 5-12-2003 at 14:31

Last I heard, lithium reacted with alcohols (and I'm not sure about pyridine of acetone too).

Polverone - 5-12-2003 at 17:00

It does react with alcohols, but the reaction with amyl alcohol isn't especially fast. In any case, you're continually working "uphill." If enough current is delivered, metal will form faster than the alcohol can dissolve it. But high current density seems to lead to malformed and impure metal deposits. This is why pyridine is especially nice, because it can be used with iron wire as cathode and relatively low current density/voltage.

chemoleo - 5-12-2003 at 17:24

Tacho, that's very interesting.
What is the reaction taht is *supposed* to happen when you react NaOH with Al? (or other salts of Na, is that also possible?)
Else, how about doing this in a large container that is filled with an inert gas such as Argon or helium, to protect Na? Problem obviously seems to be be that this are Na vapors, and they have to condense... to collect them!

Polverone, on another note... this reaction of Li in pyridine, is it, in a modified form, trnasferable to other alkali metal salts such as NaCl/KCl?

BromicAcid - 5-12-2003 at 19:17

Slightly off topic but still on lithium "Lithium has also been isolated by electrolysis of somlutions of lithium salts in non-aqueous solvents. A solution of lithium chloride in anhydrous pyridine give satesfactory results with a cathode current desity of 0.002 - 0.003 ampere per sq. cm. at 14 volts (Kahlenberg, J. Physical Chem. 1899, 3, 603). Solutions of the chloride in alcohols have also been used, but are less satisfactory because the lithium reacts slowly with the alcohol (Patten and Mott, ibid. 1904, 8, 170)"

Darn, it's a photocopy, but I didn't copy down from where so no citation. I wonder if lithium reacts slowly enough with ethanol to facilitate electrolysis of lithum chloride in that medium, it's solubility is 24.28 g/100g solvent at 20C, I know it wouldn't be possible to get sodium in this way though seeing as how it is the alkali metal of choice to produce alkoxides. If you could make lithium though, you could get sodium, you know:

Li(l) + NaCl ----> LiCl + Na(g)

The equilibrium would be to the right with the sodium boiling at 892 C compared with lithium at somewhere around 1317 C (lithium vapor is red, beautiful!). But then again that would be impossible both due to the fact borosilicate glass is not a good idea for that temp, plus when liquid lithium touches glass, watch out!

"Liquid lithium is the most corrosive material known. For example, if a sample of lithium is melted in a glass container, it reacts spontaneously with the glass to produce a hole in the container, the reaction being accompanied by the emission of an intense, greenish white light." (Descriptive Inorganic Chemistry 3rd edition, Geoff Rayner-Canham; Tina Overton)

BTW: For the link above relating to the molten salt database I didn't mean for people just to look at my examples, but to actually type in NaCl or whatever sodium salt you have availible and look at all the eutectic mixtures, 652 of them if you type in NaCl!

Polverone - 5-12-2003 at 20:01

Ethanol reacts too fast for it to work for lithium electrolysis. At one point xoo1246 posted a fascinating patent about producing alkali metals by heating their hydroxides with magnesium in a high-BP hydrocarbon. Unfortunately, he appears to have erased all his posts! If anybody has the patent noted I'd appreciate it if you could post it here.

Tacho - 6-12-2003 at 09:21

I didn't think much about the reaction.
I tried Al powder based on the refs. that said that sodium was produced reacting iron powder with NaOH. Al is more reactive then Fe in this case. If they had cheap Al in those days, they would probably use it.

Inert gas can be provided by a plastic bag with steel wool that has been soaked in dilute acid. If left overnight, only nitrogen is left. I then squeeze it into the "chamber".

I plan to do it soon.

Sodium at last!

Tacho - 6-12-2003 at 14:31

I repeated the sodium thermite experiment and made metallic sodium this time.

I put about 3g of NaOH in a small stainless steel cookie form. Then added half a teaspoon of powdered Al. Some fizzing happened because the hydroxide had already absorbed water from air and was attacking the Al. I heat with the blowtorch and the hydrogen from the fizzing burns with orange fire. But don’t be mislead, this fire is not the reaction.

I cover the cookie form with a stainless steel dish for condensation(thanks chemoleo) and keep heating until the pan steel turns glowing red.

.|________________| dish
..........|_______| cookie pan (5cm diameter)
............... () fire

After some seconds, a really strong orange fire comes out from where the cookie pan touches the dish. The gas is cut, the reaction goes by itself. There was not enough air inside, so it has to be the Al stealing the oxygen from the hydroxide.

Everything gets really hot, with glowing red spot in the center of the dish. I put drops of water on top of the dish to cool it. They fizz jumping around.

When I removed the dish, some gray stuff was stuck where I expected the sodium to condense . Just some grains, but was a good sign. This gray stuff quickly turned white, the way I would expect sodium to oxidize. Very good sign. I drop drops of water on the grey-now-white stuff and…orange sparks! Sodium!


1- Yes, it is some kind of NaOH +Al -> AlOH +Na themite like;
2- This could be improved to become a practical way of making Na;

The Hydroxide absorbs water really fast, reacting with the Al and becoming a bubbling gunk, so don’t mix them much , be quick and use coarse NaOH.

Next I will do some tests with copper pipes, if they don’t get corroded too fast, it’s the way to go. Steel pipes are too thick and would be hard do heat enough to start the reaction.

I also bet the grey stuff left in the cookie pan has lots of elemental sodium.

I have a headache now, maybe some ultraviolet in that fire. My googles are transparent.

unionised - 6-12-2003 at 15:42

An interesting experiment. Seems like a fairly effective way of making DIY sodium (I presume that's what you have got; I guess it could be some finely divided Al that's reactive enough to burn or some such. Come to think of it you have the elements to make NaAlH4.)
Most goggles will block UV quite well, BTW.

The_Davster - 6-12-2003 at 16:35

That was a facinating experiment, and by the way where did the sodium condense?


chemoleo - 6-12-2003 at 19:57

I think I will try this during Christmas holidays.
First I will grind the NaOH, in a glass without access to air, or more importantly, moisture.
Then, I will mix stoichiometric amounts. I have to think of the reaction equation first, however, to figure that one. Either Al2O3 is produced, and hydrogen, or, which is probably more likely, Aluminumhydroxide/oxide ...
I wonder whether this can be set off in a thermite like manner, like with Mg/NaClO3 etc. Will test this in 2 weeks from now or so!
More importantly, tacho, maybe fill the steel dish with water to increase the cooling effect, and hence the effect of condensation of Na vapours onto it!
If you have enough, why dont you scrape it off, and analyse it a little? i.e. add ethanol to it, and see if it reacts, while the Naethanolate forms (hydrogen bubbles should evolve)! Or, collect it, and heat it in a SEALED tupe with N2 (you can get rid of the oxygen by incubating a bit of steel wool with a small amount of acid (i.e. HCl) in a closed container, while shaking it.. there are many methods, anyone got additional info on it?)
Anyway, in an inert gas atmosphere, Na should melt quickly, more easily than Lead! This is a definite test!
Admittedly, evaporated NaOH will melt early too, so, evaporation onto the dish toghether with Na will yield a grey substance indeed.
I guess we have to find a way for separating NaOH from the Na.
I remember that Na dissolves in liquid ammonia NH3, but thats hardly applicable to us!
Anyway, great experiment, cant wait to try it myself!

Edit: how about you use some high container, like a small can, for heating your NaOH/Al, which is covered by a steel dish? this way you have a crude temperature gradient, where you collect the most volatile material (i.e. Na, NaOH) at the top!

Edit2: what is DIY sodium?????

[Edited on 7-12-2003 by chemoleo]

The_Davster - 6-12-2003 at 22:40

DIYsodium= do it yourself sodium

unionised - 7-12-2003 at 04:50

Aluminium hydroxide is not stable on heating. Why do you think it is more probable than the oxide?

Tacho - 8-12-2003 at 02:27

NaAlH4 is a possibility, unfortunately. Does it give orange sparks with water?

Chemoleo, do not grind the NaOH or mix it too well with the Al. It absorbs water from air and its enough to begin a very exotermic reaction. NaOH will have to melt long before the "thermite" begins, which bring us to another caveat: Don't use glass. You have to heat the mix to glowing orange before the reaction begins. I think that's way beyond glass capabilities (sp?). Besides, molten NaOH is a nice solvent for glass.

I will give it a try in larger scale, but it has to be outside the house, and it's raining these days. I don't have to say that this is all quite dangerous in a larger scale. Molten NaOH is... well, nasty. And, yes, I will put water in the dish.

You can give it a try in very small scale first, with a stainless steel spoon, pliers, a blowtorch, gloves and googles. Just to see how it happens.

Can we post pictures in this forum?

Saerynide - 8-12-2003 at 02:54

About what temperature do the reactants have to be heated to?

Tacho - 8-12-2003 at 06:06

The references that said sodium was obtained reacting iron powder and NaOH state temperatures above 1000ºC. I had to heat the crucible (ss cookie pan) to orange, I believe that’s around 1000ºC. But I gess only a portion of it has to be so hot, since the reaction generates a lot of heat.


“Gay-Lussac and Thénard (1808) showed that molten caustic potash (KOH) or caustic soda (NaOH) brought into contact with red-hot iron turnings produced the respective alkali metal as a distillate.
Castner (1886) produced sodium on a large scale by heating NaOH with iron and carbon at a temperature of 1000 °C:

Another thing:

The cement-like residue reacts quite violently with water, but generating only hydrogen (?), no sparks. As unionized has pointed out, NaAlH4 is a possible result here. This residue could be a strong reducing/hydrogenating agent. Could anybody suggest an easy test for strong reduction?

I know this is not electrochemistry, but here is where sodium production is being discussed.

chemoleo - 8-12-2003 at 06:45

If we ignore the formation of NaAlH4 for the moment, the theoretical reaction would be:

Al + 3NaOH --> Al(OH)3 + 3Na

2 Al(OH)3 --> Al2O3 + 3H2O


2 Al + 6 NaOH --> Al2O3 + 6 Na + 3H2O

Depending on the temperature though, you may get AlO(OH), which yields Al2O3 with further heat (I was referring to this unionised).
Anyway Tacho, maybe try this with weighed stoichiometric amounts! This in a sealed container, with a small pressure outlet, in an atmosphere that is preferably devoid of O2 etc.
I am sure this wold work nicely!
Anyway, pls test your grey substance with ABSOLUTE ethanol - it should evolve H2 if it contains free Na. If it is NaAlH4, it probably shouldnt (altho i dont know), as Ethanol cant be reduced further...

PS hey Tacho, I played lots with molten NaOH and know of its hygroscopicity, so dont worry for my safety... yet... until I dump 500 g of metallic sodium into my bath tub :D:D
PS yes you can post pictures, look for the 'new feature for sciencemadness' thread.

[Edited on 8-12-2003 by chemoleo]

I think Tacho is on to something here, but...

Hermes_Trismegistus - 8-12-2003 at 10:58

Originally posted by chemoleo
the theoretical reaction would be:

Al + 3NaOH --> Al(OH)3 + 3Na

2 Al(OH)3 --> Al2O3 + 3H2O


2 Al + 6 NaOH --> Al2O3 + 6 Na + 3H2O

[Edited on 8-12-2003 by chemoleo]

Al2O3 + 6 Na + 3H2O as products?!?

wouldn't (6Na)2 + (3H20)2 ==> (6Na(OH)) + 3H2 ???

Since Chemoleo loves poetry so much I feel the need to express this another way..:D

Said a globule of sodium, so sweet
To a droplet of water petite,

“Let’s unite in the air, rearrange ourselves there, and give off a great deal of heat!”

***Thanx Tacho!***

"By 1890, Castner developed a large scale electrolytic method for preparing sodium using a cylindrical iron pot with an iron cathode and a nickel anode"

Now, I've had a little trouble trying to puzzle a way out (in my mind) to electrically insulate the anode's and cathodes from the stainless steel container.

I beleive if I was going to construct a down's cell using carbon anode/cathode's I would have similar trouble with sealing the electrodes to the vessel, both physically and electrically.

It has occured to me to use a ceramic plug to hold the electrodes but then I think that the expansion of the Stainless steel pipe bottom might crack the ceramic inserts spilling molten salt and sodium on the garage floor followed by a little chlorine gas.

In case anyone is wondering, It would seem to me to be easy to use a foot long peice of 3" stainless pipe from online as my reaction vessel. If I was going to attempt to build a Na Cell.
Maybe the Castner heat/charcoal process is a little more realistic than dealing with the juice.

The first Castner Process... 6NaOH + C==>2Na+3H2+2NaCO3 seems quite workable using ceramic crucibles and a single pole (sublimator type)condensor.

[Edited on 8-12-2003 by Hermes_Trismegistus]

good point... with a FATAL flaw :D:D

chemoleo - 8-12-2003 at 12:41

Lol Hermes aka Bob,

you have a point there, thanks for pointing it out! Sadly though, your poetry seems to distract you from thinking this through :D - do what you do best!

Ok, the reaction of the products will occur, as you said this is

2Na + 2H2O --> 2 NaOH + H2.

There you failed to notice the fatal flaw, which is
HYDROGEN IS PRODUCED! You are effectively removing hydrogen out of the system, leaving oxygen only!!

So what does that mean in terms of reaction equations?

1] 2 Al + 6 NaOH --> Al2O3 + 6 Na + 3H2O

2] 2 H2O + 2 Na --> 2 NaOH + H2

As you are eliminating water in this process, multiply 1] times 2, and 2] times 3.

This gives

1'] 4 Al + 12 NaOH --> 2 Al2O3 + 12 Na + 6 H2O

2'] 6 H2O + 6 Na --> 6 NaOH + 3 H2

Overall, after eliminating H2O, we then get the final version:

4 Al + 6 NaOH --> 2 Al2O3 + 6 Na + 3 H2


The beauty of chemistry ey? :D :D

This is why the reaction DOES work!!

Let me rewrite that poem:

Said a globule of sodium, so sweet
To a droplet of water petite
I will eat you up no worry,
but it farted HYDROGEN and it was sorry!!



[Edited on 8-12-2003 by chemoleo]

KABOOOM(pyrojustforfun) - 8-12-2003 at 19:12

<font face=symbol>D</font>H°<sub>f</sub> for NaNO<sub>3</sub> and Al<sub>2</sub>O<sub>3</sub> are -424.8 & -1669.8 KJ
NaNO<sub>3</sub> <s>&nbsp;&nbsp;&nbsp;></s> Na(s) + 0.5 N<sub>2</sub> + 1.5 O<sub>2</sub> - 424.8 KJ (in thermodynamic you can use any theorical equation)
1.5 O<sub>2</sub> + 2 Al <s>&nbsp;&nbsp;&nbsp;></s> Al<sub>2</sub>O<sub>3</sub> + 1669.8 KJ
Na(s) <s>&nbsp;&nbsp;&nbsp;></s> Na(g) &nbsp;&nbsp; <font face=symbol>D</font>H<sub>subl</sub>= +108 KJ
1669.8-424.8-108=1137 so
<font color=green>NaNO<sub>3</sub> + 2 Al <s>&nbsp;&nbsp;&nbsp;></s> Na(g) + 0.5 N<sub>2</sub> + Al<sub>2</sub>O<sub>3</sub> + 1137 KJ</font> :o

Na<sub>2</sub>CO<sub>3</sub> <font face=symbol>D</font>H°<sub>f</sub>= -1130.9 KJ/mol
<font color=green>Na<sub>2</sub>CO<sub>3</sub> + 2 Al <s>&nbsp;&nbsp;&nbsp;></s> 2Na(g) + C + Al<sub>2</sub>O<sub>3</sub> + 430.9 KJ</font>

NaOH <font face=symbol>D</font>H°<sub>f</sub>= -426.7 KJ/mol
<font color=green>NaOH + 2/3 Al <s>&nbsp;&nbsp;&nbsp;></s> Na(g) + 1/3 Al<sub>2</sub>O<sub>3</sub> + 0.5H<sub>2</sub> + 21.9 KJ</font>

Yep it works nicely...

Al Koholic - 8-12-2003 at 20:20

5.4 grams Al powder.
8.6 grams NaOH prills.

Substances placed into an empty Campbells soup can. Can was crimped slightly so pliers could grasp it. An empty tuna can served as a lid and was crimped slightly to grasp the soup can (didn't want it flying off when the H2 ignited).

Holding onto the can with pliers, the container was held over the flame of a bunsen burner for about 2-3 minutes when a nice hissing sound began to grow louder. As the hissing was increasing in intensity, a burst of hydrogen ignited and proceded to burn from around the edges of the tuna can lid. The container which was brought to a dull red by the burner was now a bright orange. The reaction subsided within 30 seconds or so and I let the container cool with the aid of some snow (leaving the lid on of course).

I looked inside and saw a grey, fused mass. This was chipped out of the container with minimal effort and isolated. A small portion was grasped with forceps and placed into a jar of water resulting in vigorous fizzing. I could see the mass falling apart (into NaOH and Al2O3 no doubt) at the bottom of the jar with many bubbles of H2 coming off. The rest of the mass was placed into a pyrex jar for storage. I took the reaction cans outside and poured water on them and got some nice fireballs from the sodium adhering to the sides.

The product weighs 9.2 grams and has gotten coated with a faint white haze. I can see many small spherical, metallic globs sticking out of the grey chunks which must be sodium with an oxide coating. I am also not storing the product under kerosene because I don't care about the oxidation. I am also aware of the non-stoichiometric experiment. This was slightly intentional because I wanted the sodium to be trapped in a lot of Al2O3/Al.

Now to design a larger scale process and then a purification step involving melting the sodium and somehow removing the Al2O3....



The_Davster - 8-12-2003 at 20:49

Why couldent you just heat the unupurified sodium/Al2O3/Al mix(in an inert gas) untill the sodium melts and then pour off the molten sodium (the Al2O3 and Al shoud sink due to their greater density).
Can anyone see a problem with this?

Polverone - 8-12-2003 at 21:27

Even better, heat and prod/stir it under xylene, toluene, kerosene, etc. Sodium has a nice low MP so this should be feasible and easier than working with inert gas.

Congratulations Al Koholic, and thank you Tacho. Given how easily NaOH attacks aluminum, I wonder if a more common, cheaper form of the metal might work (like foil, filings, or turnings).

Orgy: your electrolytic cell may work marvelously, but does it glow with heat and belch burning hydrogen? Does it?! I thought not. Tacho's method gets more Mad Science Points for that reason. However, you may recapture the glory if you show us pictures of an operating Castner cell :D.

Tacho - 9-12-2003 at 02:39

Thank you all,

Congratulations for those who could figure out the equations!


1 - The gray residue does NOT react with anhydrous acetone, anhydrous IPA, anhydrous methanol nor 99,5% ethanol. After 12 hours , there was a bit of white haze in the alcohols, the acetone has evaporated (duh).

2 - I could not make zinc powder react.

3 - About electrolisys of molten NaOH (I mixed a bit of NaCO3 too) using porous ceramic: bad news, it takes an incredible short time for the molten NaOH to dissolve the ceramic and, worse , the residue in the ss reactor is very dificult to clean because, if you think about it - I didn't- you are kind of making... glass.

Saerynide - 9-12-2003 at 02:57

This sounds like something to try over xmas break :D

Theoretic - 9-12-2003 at 05:05

I'm a bit dazed... :D:D:D The thread wasn't too interesting when it was solely electrolytic, but the chemicalmethods surely livened it up! ;) I don't believe my eyes!
Just one problem I see with the Na2CO3/NaNO3 + Al method... I think it would rather react to produce CO/N2 and NaAlO2...

Marvin - 9-12-2003 at 05:17

Polverone, Neuron reminded me the patent number a few months ago when I couldnt find it, its US4725311 I think. It deserved more attention than it got.

Magnesium and sodium hydroxide being heated together without a solvent, themite style will also work but this is very exothermic and can prevent a reasonable yeild.

Sodium hydroxide + aluminium will yeild a small amount of sodium. Mostly you get sodium aluminates and metaaluminates.

Magnesium does not form 'magnesates' which is why it doesnt have the same problem.

vulture - 9-12-2003 at 05:22

Won't the NaOH decompose into Na2O before the reaction starts, driving off H2O vapor?

Theoretic - 9-12-2003 at 07:16

No, NaOH doesn't do that. Not at the temperatures available to a home chemist, anyway. Even sodium carbonate requires 1000 C to decompose.

KABOOOM(pyrojustforfun) - 9-12-2003 at 20:59

<blockquote>quote:<hr>Just one problem I see with the Na2CO3/NaNO3 + Al method... I think it would rather react to produce CO/N2 and NaAlO2...<hr></blockquote>CO reacts with Al
3CO + 2Al <s>&nbsp;&nbsp;&nbsp;></s> 3C + Al<sub>2</sub>O<sub>3</sub> &nbsp;&nbsp; <font face=symbol>D</font>H<sub>c</sub>=-1251 kJ
excess Al must be used with those methodes. even though I included sublimation energy for Na the reaction is still vety exothermic for NaNO3 methode. the hardest part is to get a single piece steel retort :D<!--

[Edited on 10-12-2003 by KABOOOM(pyrojustforfun)]

Sodium nitrides

chemoleo - 9-12-2003 at 21:10

just wondered, how easily is sodium nitride produced (if at all) , analogous to Magnesium Nitride Mg3N2?
Surely this will interfere with the desired reactions if a N2 atmosphere (devoid of O2) is employed....adding an other impurity to the already mentioned NaAlO2, NaAlH4 etc...

I guess the smell would tell upon dissolution in water, due to the formation of ammonia....

Hermes_Trismegistus - 10-12-2003 at 01:55

Thanx Marvin for that sweet patent, I am not sure about the solvent he used for the actual example, but it seems that if the kerosene he suggested worked I think you're in the money.


the Castner Tiegel cell shows definite promise but I am dismayed both by the copious use of Nickel (200 bucks for the cheapest, thinnest 200 grade sheet I could find, and that was only 12 inch by 12 inch)

It also puzzled me with the Condensed material at the bottom of the chamber and the wooden gasket?

I wondered if my babbling fish is really beginning to babble?

and Still !!! the thought of a down's cell gets my motor running...

I think I may be getting closer to the prize (ingots of sodium the size of a cpu tower)

I have ALWAYS wanted to throw one into a lake!!!

I just found out that Beer keg's are actually high grade stainless!


vulture - 11-12-2003 at 08:18

Originally posted by KABOOOM(pyrojustforfun)
<font face=symbol>D</font>H°<sub>f</sub> for NaNO<sub>3</sub> and Al<sub>2</sub>O<sub>3</sub> are -424.8 & -1669.8 KJ
NaNO<sub>3</sub> <s>&nbsp;&nbsp;&nbsp;></s> Na(s) + 0.5 N<sub>2</sub> + 1.5 O<sub>2</sub> - 424.8 KJ (in thermodynamic you can use any theorical equation)
1.5 O<sub>2</sub> + 2 Al <s>&nbsp;&nbsp;&nbsp;></s> Al<sub>2</sub>O<sub>3</sub> + 1669.8 KJ
Na(s) <s>&nbsp;&nbsp;&nbsp;></s> Na(g) &nbsp;&nbsp; <font face=symbol>D</font>H<sub>subl</sub>= +108 KJ
1669.8-424.8-108=1137 so
<font color=green>NaNO<sub>3</sub> + 2 Al <s>&nbsp;&nbsp;&nbsp;></s> Na(g) + 0.5 N<sub>2</sub> + Al<sub>2</sub>O<sub>3</sub> + 1137 KJ</font> :o

Na<sub>2</sub>CO<sub>3</sub> <font face=symbol>D</font>H°<sub>f</sub>= -1130.9 KJ/mol
<font color=green>Na<sub>2</sub>CO<sub>3</sub> + 2 Al <s>&nbsp;&nbsp;&nbsp;></s> 2Na(g) + C + Al<sub>2</sub>O<sub>3</sub> + 430.9 KJ</font>

NaOH <font face=symbol>D</font>H°<sub>f</sub>= -426.7 KJ/mol
<font color=green>NaOH + 2/3 Al <s>&nbsp;&nbsp;&nbsp;></s> Na(g) + 1/3 Al<sub>2</sub>O<sub>3</sub> + 0.5H<sub>2</sub> + 21.9 KJ</font>

That's not correct IMHO. You need to substract the formation enthalpy of the reagents from the formation enthalpy of the reaction products.


A + B ---> C + D

DHf = HfC + HfD - HfA - HfB

So you should leave the minus sign infront of all the enthalpies. Recalculation shows exothermic values.

BromicAcid - 12-12-2003 at 15:22

I really like the Castner Cell and all, but with all the talk about chemical reduction to acheive sodium metal I decided to give it a whirl. Well, kind of.... "The hydroxides or carbonates of the alkali metals -- 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.." [Inorganic and Theoretical Chemistry] So I chose potassium hydroxide and magnesium turnings to start...

1 Shows the reactants prior to any disturbance, 5 g of Magnesium turnings and 11 g of potassium hydroxide in prill form.
2 Shows the reaction in progress, I ran it in a soup can because I had all the reagents on hand and it was cold and I didn't plan and it was easy to find. The top was closed but hydrogen evolution quickly gave a dull thud and poped it off and flames started rising above it. The reaction started almost immediately as the bottom was heated with a bunsen burner. Some smoke and characteristic violet emmision of light.
3 Flames died down and I took a picture, you can't see it well but there were little potassium globules, when you blew on them they would burst into flames, I closed the top and let it cool, being that it was 15 F outside it did not take long.
4 After it cooled down I moved the lid off, it caught on fire again when the air hit it. The top of the mass is greenish and molten looking but quite solid on probing, I can break it up and potassium is visible in the matrix.
5 I slammed the container down on the picnic table to dump out the contents. It immediately burst into flames giving the characteristic violet spectrum of potassium and continued burning for over a minute leaving a white ash that reacted with water releasing gas (probably oxygen, potassium forms appreciable peroxide and superoxide when exposed to air).
6 On reaction with water of the solid left over with potassium in matrix it released hydrogen, caught fire, precipitated magneisum hydroxide, you can see it at the bottom of the beaker.

So, it was pretty interesting. I chose magnesium over aluminum because it does not form compounds with aluminum as stated before: "With aluminum metal in place of magnesium some alkali aluminate is formed, and the yeild is reduced considerably."
For example the carbonate:

3Na2CO3 + 2Al ---> Al2O3 + 6Na + (3CO2)
which competes with
3Na2CO3 + 2Al ---> 2Na3AlO3 + C + (CO2 + CO)

So that explains my preference for magnesium over aluminum and before I explained why I picked potassium hydroxide over sodium hydroxide although you could probably use them interchangeably. Supposedly the mixture does not get hot enough to volitize the potassium, I can not vouch for this but if this is the case the reaction vessel can be made from 2 inch wide pipe maybe 5 inches long with an end cap and a reducing piece to put it down to say 3/4 in. diameter and then you can put a 45 deg. piece in place and make a metal retort. Heat the reaction mixture and continue to do so after reaction and distill over the sodium metal. Supposedly the sodium vapor is cooled in molten lead then distilled from that later, a bit impractical but what the hey.

Alkali - 12-12-2003 at 17:29

Congratulations for your experiment, it's really interesting!!

Just one suggestion for you to try it once again: Why not covering the mixture (magnessium turnings + KOH) with sand so you could preserve the metallic Potassium formed in a better way so you could then add mineral oil and pass potassium pieces on a separate container with mineral oil!!??

I have plenty of Sodium metal at home (more than 1 Kg of it) and 1 Kg Strontium Chloride. I'm planning to obtain Strontium metal reducing the chloride with my sodium by adding it inside an inverted container over the fused strontium chloride so the strontium metal does not oxidise once it's being formed. What do you think? Would be a really interesting experiment to perform, don't you agree?

[Edited on 13-12-2003 by Alkali]

[Edited on 13-12-2003 by Alkali]

unconventional sodium

unionised - 13-12-2003 at 03:35

Sand will react with Mg or with KOH or with K.
Perhaps something innert would be better. MgO for example.

Tacho - 13-12-2003 at 04:29

Congratulations BromicAcid! I’m stunned! Potassium using Mg, that’s great!

My experiments, in the other hand are not so impressive. In fact, they are very disappointing.

I tried the NaOH/Al thermite on a larger scale to confirm metallic sodium deposition.
I took pictures of the setup, but I still don’t know how to post them here. I could not use the FTP to this server (?). Any easy way? Whats that attachment button for? Whats the insert picture button for? Sorry, but I'm sure I am missing something obvious here.

I used a lager stainless steel cup and dish (dish more like a bowl)

..|..................| bowl
........|_._| cup

(aargh!, dots everywhere, I can't get even ascII drawings right here)

I used about four spoons of NaOH and four of Al powder. I covered the cup with the bowl and put ice/water in it (the bowl). The reaction started very soon after the blowtorch was set (about 2 min.) and lasted for about one minute, yellow/orange flames coming out. By the way, there was a loud “pop” before the reaction began, it was the hydrogen inside burning, anyone trying this reaction in larger scale should remember that hydrogen. After the reaction was over, I quickly poured the water and dipped the dish bottom in xylene to prevent oxidation and then scraped the deposits to the xylene. Here things began to go wrong. The deposit was brittle and not metallic. I tranfered the deposit to a pyrex tube and tried to melt the sodium. Funny things happen here: The xylene boils and stirs everything inside, making a sedimentation of molten sodium (if there was any) impossible, besides, the few droplets of water that went into the xylene (ice in the dish=condensation) go to the bottom, happily reacting with anything they find.

I got no metallic sodium, only the gray cement-like residue.

It is a fact that, soon after the reaction is over, the deposits on the walls of the cans, pans, whatever, react with water to produce yellow/orange flames but, sadly, this is not the idiot-proof method of producing sodium I though it would be.


Organikum - 13-12-2003 at 06:34


Dont forget the license fee for the soup-can. :D

BromicAcid - 13-12-2003 at 12:06

My number one reason to post the pictures.... they looked neat. To get this reaction to work right I would use a 5 inch long 3 inch diameter pipe threaded at both ends with end caps. I would drill a hole about 1/8 inch or so in diameter on one and and leave the other.

These pipes are exceedingly thick and resillant to pressure. The hole serves as an exit for the hydrogen gas but I would buy a piece of pipe exactly the size and force it in, I would pack the smaller pipe full of steel wool to try and condense out sodium vapor and allow more cooling surface.

Immediately upon initiation of the reaction I would attempt to light the exit gasses so they would burn off as the reactoin proceeded, being under hydrogen pressure some of the sodium would convert to the hydride. Experimentation or additional calculations would have to be done to make sure the hole for the exit gas was wide enough to pervent an excessive hydrogen pressure on sodium, one atm of hydrogen = no reaction usualy, several and we might have a problem. But if you're really hyped up about putting some inert substance over top of the reaction mixture give magnesium oxide a shot.

Anyways, when the reaction started to die down you could put a weight over the exit area exactly like a pressure cooker and that would keep an inert atmosphere inside. Wait to cool, then remove endcap and put contents into xylene, heat to mp of sodim with stirring, breaking up the chunks with a stirring rod then hopefully with it's low density it will rise to the top of the reaction byproducts and seperation will be feasable.

But mainly I just did this to see what it would look like, if anyone wants to continue this further I would highly recomend the use of magnesium in place of aluminum for the reduction. Local scrap yards will sell magnesium to an individual for $.70 a pound then just sand it to a shiny finish, take a large drill bit to it and collect the shavings, very very little atmospheric oxidation.

Personally any further work I do will be on the Castner Cell although I am looking foreward to seeing if anyone can perfect this method as well.

KABOOOM(pyrojustforfun) - 13-12-2003 at 21:27

vulture: I put the realized energy into right side of the equation . eg 1137 KJ is realized from the nitrate flash mix which means <font face=symbol>D</font>H = -1137 kJ
anyway I shall do the calculation for Mg
HBr! I love that purplish flame;)

Theoretic - 15-12-2003 at 05:03

I think that aluminates are actually intermediates in the process, since first aluminium mreacts with the hydroxide forming oxide and aluminate, which react to produce a different aluminate. Then the sodium ions can finally be reduced to sodium.

Reduction of Aluminates

chemoleo - 15-12-2003 at 10:27

with reference to Marvin's statement:
/ Aluminium and sodium hydroxide is supposed to go mainly to Na3AlO3, a mixture that was formerly used to defrost oil wells and which produces no sodium directly, formation of sodium is a side reaction with the coproduction of sodium metaaluminate. Magnesium seems the only feasable metalic reducing agent which limits most peoples ability to use this.

I agree that side products *might* be the formation of aluminates. However, what I dont understand is why an excess of Al would not reduce that, too.

After all, why is this reaction

Al(ONa)3 (or Na3AlO3) + Al --> Al2O3 + 3 Na

not possible???

Those sodium aluminium oxides could only persist (under conditions of free Al and heat) when the reduction of the Na+ to Na would require more energy than the oxidation of an additional Al produces!
Now is that the case?

[Edited on 15-12-2003 by chemoleo]

vulture - 15-12-2003 at 11:31

Bless the great Ur of Al2O3 which makes this all possible...

Look up Born-Haber proces.

KABOOOM(pyrojustforfun) - 17-12-2003 at 18:33

MgO <font face=symbol>D</font>H°<sub>f</sub>= -601.8 kJ/mol
NaNO<sub>3</sub> + 3Mg <s>&nbsp;&nbsp;&nbsp;></s> 3MgO + Na(g) + 0.5N<sub>2</sub> <font face=symbol>D</font>H=424.8+108-(3*601.8)=-1272.6 kJ
Na<sub>2</sub>CO<sub>3</sub> + 3Mg <s>&nbsp;&nbsp;&nbsp;></s> 3MgO + 2Na + C <font face=symbol>D</font>H=1130.9+108-(3*601.8)=-566.5 kJ
NaOH + Mg <s>&nbsp;&nbsp;&nbsp;></s> MgO + Na + 0.5H<sub>2</sub> <font face=symbol>D</font>H=426.7+108-601.8=-67.1 kJ
I don't know the heat capacity of the products at different temperatures. if I had the data I could calculate their T
Marvin: lol I forgot "his" Os

Saerynide - 17-12-2003 at 22:05


MgO DH°f= -601.8 kJ/mol
NaNO3 + 3Mg > 3MgO + Na(g) + 0.5N2 DH=424.8+108-(3*601.8)=-1272.6 kJ

Holy hell!! Thats A LOT of heat :o I think I now understand the power of nitrates...

KABOOOM(pyrojustforfun) - 19-12-2003 at 21:09

that was what I wanted to point out. more Mg is consumed than the other methods but its being so hot is very useful, as the chemical ways of preparation of Na are best down at the highest T possible. it should propel gaseous Na! which can be distilled all without use of an outside heat source (it would be kinda hard if one has to distill the impure Na he's already made)

BromicAcid - 19-12-2003 at 21:39

Here is an idea that Ive been rolling around regarding the recovery of sodium product from a chemical reduction. Apparatus, two 5 inch pieces of 2 in wide pipe, threaded on both ends, one end of each pipe sealed with an endcap. They are hooked together by a step down sizer going from 2 to 3/4 inch diameter with a 1 in 3/4 in pipe connecting the two together. In one of the large pipes, a hole is drilled approx. 1/4 inch diameter near the endcap and a small length of pipe is run about 4 inches up.

Fill pipe A without hole with enough reaction mixture to stay safely below the hole connnecting it to the pipe on the other side when on its side. Pack pipe B with steel wool fairly full being sure to put a plug of it in the opening of the little pipe on the inside put apparatus together being sure that smaller pipe is sticking up and trying not to get any of the reactiong mixture into pipe B.

So start the reaction by applying heat. Sodium metal is reduced and with additional heat applied vaporized. Hydrogen pressure prevents excessive reaction. Sodium metal condenses out on the steel wool in the second pipe and hydrogen gas comes out of the smaller pipe where it would be ignited. Additional heat would be supplied to the pipe where the reaction would be taking place after initial reaction subsided to facilitate additional recovery. Cover top of smaller pipe to prevent atmosphere from entering and wait to cool.

Finally after it is cool unscrew pipe and quickly dump steel wool into Xylene. Carefully heat with stirring and pull out the steel wool from the molten sodium. Allow to cool and store ingot under xylene or kerosene.

Of course this all rests on sodium not alloying with the steel. ;) Now I'm planing on not posting here till I fish my castner cell, my downs cell, and perfecting my chemical reduction and comparing them this summer, so don't draw me back! ;)

unionised - 20-12-2003 at 04:12

Na doens't alloy with steel. OTOH if that little hole gets bunged up you have made a complicated pipe bomb.

Theoretic - 24-12-2003 at 06:01

About the NaOH/Al method... Apparently sodium oxide's lattice energy is so low and Al2O3's is so high that the reduction of Na2O by Al is exothermic! And combination of sodium oxide + oxygen to form sodium peroxide is exothermic too!
This is what I think happens:
1) Al reacts with NaOH:

2Al + 6NaOH => 2NaAlO2 + 2Na2O + 3H2

2) Then Al reacts with Na2O:

2Al + 3Na2O => 6Na + Al2O3

3) Finally, Al2O3 combines with Na2O:

Al2O3 + Na2O => 2NaAlO2


Al + 2NaOH => Na + H2 + NaAlO2

Maybe aluminium does reduce sodium aluminates - my previous version.

guaguanco - 13-1-2004 at 11:07

Originally posted by chemoleo
If we ignore the formation of NaAlH4 for the moment,
[Edited on 8-12-2003 by chemoleo]

That's safe to do, because I really,really doubt you've produced LiAlH4. LiAlH4 tends to be pyrophoric in air; I'm certain that hot LiAlH4 would react instantly with O2, and possibly even with N2.

Experimentals on Sodium Thermites.

chemoleo - 16-1-2004 at 11:56

During holidays, I did two experiments on sodium thermites.

1. 6 g Al grains (filings, 0.1-0.5 mm) and NaOH prills (9 g) were mixed, in accordance with Al koholics post earlier. The experiment was pretty much done as he did, a small iron soup can (roasted so that the label would burn away) whcih contained the mix and a tuna can on top that served as a lid. This was heated with a strong propane gas flame, from below. After a minute or so, reaction ensued, gas came out between the two cans (lots) and ignited instantaneously from the propane flame below. Colour was nicely yellow. At the same time, the bottom part of the can started to glow red-orange, indicating that a reaction really was taking place. After 15 seconds or so, gas evolution stopped, and the red glow disappeared.
Being very pleased with the result, I allowed the can to cool in the snow. When I opened it later, it did NOT ignite or anything. Neither, to my disappointment, did I see any sodium metalic globules on the side of the container :( - in fact, everything was still in the bottom of can, a grey hard substance. This I removed from the can, and put a chunk into water. Wow, lots of gas evolved, which sometimes ignited UNDER the water! Then I placed the rest into hot petroleum and boilded it at 110 deg C for 30 minutes (sodium melts at 98 deg C or so). Sadly, nothing happened to it at all, no metal sodium floating out of the grey chunk :( ...
so, what is this stuff really? I am thinking the Na may be so intimately mixed with the Al2O3 that it could not be melted out... alternatively, maybe the excess of Al caused the formation of NaAlH4 rather than free Na.
To test this, I repeated the experiment, and used stoichiometric amounts of NaOH and Al. Same story again, the reaction product looked the same after the reaction, and again did not liberate free Na after heating it in petroleum :(... Oh, and then I took a small piece of the grey stuff and held it into a Bunsen flame. It didnt burn, not even glow really, so I am pretty sure there wasnt much free Na or NaAlH4 around. Yet, the fact that it reacted so violently with water contradicts this.... I am confused :(

2. The same story, but 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!! Whoa, I was lucky that I didnt get burned! Strange though that this was such a violent reaction! I struggle to believe it was only due to the finer grade of Mg, as the NaOH prills have at least a 1 mm diameter anyway! I was thinking this is a great mix for easy flashpowder (ahem, flashprills), doesnt need any oxidiser, just needs simple NaOH!
Needless to say, I didnt isolate any sodium from that experiment either....

Slightly discouraging, those two experiments, I know... I am yet waiting to see someone isolate appreciable amounts of Na that way!

Tacho - 18-1-2004 at 09:32

Well chemoleo, I was waiting for someone to do the Mg-NaOH thing, since I don't have any Mg.

I really expected it to work after what Bromic acid did.


I had the same problems with Al-NaOH, read my earlier post. As I said there, I don't think anyone will manage to make any Na that way, but with Mg... Well, maybe someone can improve it.

Wish I had some Mg.


Organikum - 24-1-2004 at 09:34

You should have grinded your NaOH as fine as possible - carefully and preventing uptake of water from the air of course and mixed this with the Al/Mg powder. Wrapping the can with rockwool might be favorable too.

A simple question of heat and heattransfer.

Thats my suggestion here.

[Edited on 24-1-2004 by Organikum]

Tacho - 5-2-2004 at 05:04

I have been thinking about the magnesium-NaOH. BromicAcid says it is described as a very energetic/explosive. Chemoleo, by experience, describes it as a dangerous explosive flash. Well...

I gave up finding magnesium: disk drives structures are not made of it (I tested many kinds), I can’t find any scrap chainsaw structure, and the soldering rods for car weels are simply not used anymore, nobody sells them here. Chemical supply is too expensive.

Seems I'm out...

If anybody wants to give it a try (carefull!), I would sugest two things: use Mg and NaOH in pellets, about 5mm diameter and think about using some inert low-melting-point salt as a “dilluent” for the NaOH. My ignorant intuition keeps me thinking about anhydrous sodium silicate. What would be its melting point?

chemoleo - 5-2-2004 at 15:56

The diluent idea may be worth pursuing! I wish I had some Mg here right now, and I would try it!
I dont think grain size is an issue - as I used NaOH pellets (1mm diameter) anyhow - and yet it produced a very sensitive flash!
Tacho, have you checked ebay? They sometimes sell Mg filings, in bulk. Keep us up to date with your endeavours!

hodges - 5-2-2004 at 17:26

Originally posted by Tacho
I have been thinking about the magnesium-NaOH. BromicAcid says it is described as a very energetic/explosive. Chemoleo, by experience, describes it as a dangerous explosive flash.

Does anyone have the equations for the reactions that are presumed to take place? If so, I could calculate the amount of heat liberated per gram. I tend to think it is going to be less than most thermites due to the higher heats of formation of the oxides alkali metals than metals such as Fe/Cu. Although if hydrogen is produced at some point in the reaction that could explode when mixed with air.


Tacho - 6-2-2004 at 05:48

Just wanted to show the picture of the setup I used to do the thermite ignition/condensation. Its attached.

BTW- everything is made of stainless steel.

[Edited on 6-2-2004 by Tacho]

Sodium1.jpg - 13kB

Saerynide - 6-2-2004 at 07:04

The glass doesnt break?? :o

Tacho - 6-2-2004 at 08:29


I just had time to read the text you posted and it's absolutely fantastic! I'm embarassed I haven't read it before. Anyone interested in sodium making should read it, so I post the link again:


E-bay is not an option. Delivery overseas would make it too expensive, or, at least,more expensive than a local chemical supply which sells reagent grade stuff.


I edited my post. No glass involved, just ss.

I think I made a sodium alloy using the themite.

Tacho - 7-2-2004 at 08:40

I will be careful this time, no more false alarms.

After I read the wonderful BromicAcid`s refs, I decided to try to reduce NaCl+NaOH using Al. I mixed equal volumes of the salts and added Al powder until “it looked right”. I did not use the big cup that shows in the picture of my previous post, but a smaller “cookie pan”.

Cool water in the dish... Blowtorch at the pan... Orange flames...Blowtorch off... Cool a bit and...

At the botton of the dish, on the condensation area, there was a gray deposit. Scratching this deposit , I noticed It was powdery, not metallic. Anyway I gathered a decent amount of it ( 1g?) and slowly added to water. Orange sparks popped every time the dust grains touched the water. A bigger chunk fizzled floating around the water in flames, the way I would expect sodium to do.

I repeated the experiment with the same results.

I would guess there is at least 30% elemental sodium there.

I’m trying to be cool, since I was wrong last time, but I am very excited.

I tried to reduce pure NaCl with Al, but there was no ignition, so I did not look for condensates. Maybe its a slow thing. We’ll see.

chemoleo - 7-2-2004 at 08:52

Hmm, but that's exactly what I got too!
A grey substance that would ignate on water contact! But I failed to isolate any pure Na - see my post above!
I think, much better may be to do the same thing with Mg, and add NaCl as a diluent to calm down the reaction. I wish I had some Mg here :(

I have made sodium alright!

Tacho - 7-2-2004 at 10:18

I mixed NaOH+NaCO3+NaCl, about a teaspoon of each, plus enough Al powder to make it “look righ”. The thermite was harder to start, but once it started it was stronger than usual.

Beautiful chunks deposited in the condensation plate. Scratched to a 50ml centrifuge tube with about 20ml of xylene. This time it felt like a soft metal being scratched. Metallic bits everywhere. About 2ml of powder.

I am disappointed because it did not melt into one blob under heat. It remained a powder but, oh boy, pick a chunk of that and throw in water!! BIG FAT FLAMES!

If that is not metallic sodium I don’t care. Is good enough for me.

Lye, common salt, pool’s carbonate.
10 minutes, little smoke, indoors, simple reactants.

Believe me chemoleo: this is different! it's not THAT grey thing!

I believe the NaCl and NaCO3 are being reduced by the Al, like BromicAcid's papers mention, the NaOH-Al thermite just gives the heat and produces the gray garbage you mention. THIS IS NOT THE SAME THING!

Try it!

[Edited on 7-2-2004 by Tacho]

[Edited on 7-2-2004 by Tacho]

[Edited on 7-2-2004 by Tacho]

vulture - 7-2-2004 at 10:44

How bout leaching out the sodium with mercury and distilling of the mercury under reduced pressure?

Hazardous and toxic, I know, but hey, fun for sure!

Saerynide - 7-2-2004 at 11:02

WOW, good job :D Now I really want to try this sometime. I just need to find aluminum powder :(

Mumbles - 7-2-2004 at 14:46

Might you have developed any weight ratios as of yet? Add Al until it looks right can be a little hard to reproduce.

Tacho - 7-2-2004 at 15:10


You are right, but I can't offer you proper measures yet. I will post a decent description as soon as possible.

If anyone wants to try, I estimate the Al powder volume in about 2/3 the volume of the salts (1vol NaOH, 1vol NaCl, 1vol NaCO3, 2 vol Al powder). That proportion will surely give you results to work with.

hodges - 7-2-2004 at 15:54

Originally posted by Tacho
I mixed NaOH+NaCO3+NaCl, about a teaspoon of each, plus enough Al powder to make it “look righ”. The thermite was harder to start, but once it started it was stronger than usual.

Beautiful chunks deposited in the condensation plate.

This sounds interesting. How did you ignite the thermite and then get the condensation plate over it in time?


Obtaining metallic sodium nuggets using sodium salts and powdered aluminum.

Tacho - 8-2-2004 at 01:58

Mix 20g powdered aluminum, 6g NaOH, 3g NaCO3, and 5g NaCl in a steel crucible.
Put the crucible over a heat resistant base that allows heating by flame. Cover the crucible with a steel bowl filled with water (check attached picture, but don’t do it over you computer).
Heat the crucible vigorously with a blowtorch or a heavy duty bunsen burner until a reaction takes place, marked by yellow flames escaping from the crucible.
When the reaction is over and the crucible is no longer red, remove the bowl carefully pouring the water away. There should be a light grey deposit at the botton. Scrape this deposit with a spatula to a 50ml vial with some xylene in it using a funnel. Repeat this procedure four times.

A warning here: you now have metallic sodium everywhere! Be careful. When you wash the crucible, the bowl or the funnel, orange flames and loud pops might show. Specially the residues in the crucible: there is a lot of sodium left there. You have to wash and dry them because the residues absorb humidity from air very fast.

Heat the xylene to ebulition and then turn off the heat. While the xylene cools, using a glass rod (I used disposable wood chopstick) keep revolving the sediment, doing compressions as well as stirring. Slowly. I can’t help thinking of butter making. You should end up with half a dozen metallic nuggets and a gray residue.

My yield was 0,3g for five runs. Enough to cover a fingernail with nuggets.


1- The thermite reaction is harder to start than the pure NaOH-Al.
2- The little sodium nuggets are very soft, as expected, and don’t produce flames when tossed in water, they just run around quickly on the surface, fizzing until they disappear with a pop and a spark. Careful with the residue in the crucible though, not only it can ignite again when you remove the bowl, but it can promote some fire when washed.
3- I don’t know the mesh grade of my Al powder, but it looks like grey coarse alumina, or very fine sand. I bough it in a fiberglass supply shop. It is supposed to get mixed with polyester resin for some purpose. It is not that superfine kind that would stick to a glass like paint. It falls off a glass surface leaving just some dirt behind. My NaOH comes in 1-2mm pellets.
4- I tried twice to melt the sodium without the xylene. First time I got a little nugget for 1 run but ruined the glass tube. Second time the sodium ignited and I ruined the glass tube.
5- Everything is extremely empirical. The weights of the reactants are basically based in the volumes seemed to make better results, that is: 1 vol of NaOH, 1 vol NaCl, 1 vol NaCO3 and 2,3 vol Al.

This is all I learned in the last 24 hours. A total of 3.456 improvements can be made to the process.

BTW- Please take a look at my “Versatile miniscale glassware” thread in the reagent aquisition and apparatus forum. I don’t think it’s getting the views it deserves. I wish I had seen something like that when I started my hobby.

Crucible.jpg - 10kB

Pictures of the whole thing.

Tacho - 11-2-2004 at 03:02

Voguel’s third edition says that the way to make sodium POWDER is to heat it in xylene and stir. No wonder I can’t get the sodium to melt together in a big blob. This is the main problem I find. I bet more than half that powder is sodium, just have to find a way to melt it in one piece. Any help would be appreciated.

Here goes the full thing in pictures:

SodiumMaking.jpg - 175kB

darkflame89 - 8-4-2004 at 04:26

Do think i had had a potential termite mixture at home..:oMy drain cleaner has NaOH and Al pieces together in the same bottle. All ihave to do is to pour into a soup can and heat...:o

Tacho - 8-4-2004 at 06:04

To obtain sodium granules you must mix NaCl to the NaOH. Some NaCO3 seems to improve yields. See my post some time ago.

Anyway, I always used aluminum POWDER. I would be surprised if those chunks of aluminium make a violent reaction. If you really want to try, maybe shreded thin Al foil. I Never tried it.

Be very careful, many things can go wrong!

BromicAcid - 28-4-2004 at 14:07

I went out experimenting today and here are the results. I tried chemical reduction of NaOH with Aluminum but the shavings were too big. I was hoping that extensive heating would liquify them but after 15 minutes with the torch I gave up. Oh well. After that I hooked up my power source to try some electrolysis. I know that I've read of only a few solvents that can be used to produce sodium but I decided to do some tests of my own. I tried NaOH in acetonitrile, kerosine, mineral oil, and toluene with no results, wasn't expecting any anyways.

Finally though I tried electrolysis of NaNO3 like I mentioned in the patent a few posts back. So I used the same procedure as before, put electrodes in NaNO3, spritz with water to make conductive, pump up the voltage. Immediately the NaNO3 started turning green, then the water started to evaporate off from the heat and NO2 started coming up. I grabbed a strip of copper and held it above the test tube out of curiosity and sure enough it started eating it, good ol' nitric acid. So I ran it a bit longer and it was really really eating into one of my electrodes. Nickel really isn't resillant to oxidation so it was dying a horrible death.

I let it run a bit more but was getting nervous as it started to melt and work all of the water our of it's system. Regardless of the patent I still have a slight phobia that the Na/NaNO3 mixture that would result would explode like flash so before it totally dried out I discontinued electrolysis.

This weekend I am going to do some major sodium research, I've got two methods that I want to try for mass production. Everyone cross your fingers! ;)

[Edited on 4/29/2004 by BromicAcid]

 Pages:  1    3  4