Sodium!

Patu! You realy should post more; you seem to have lots of interesting ideas. Perhaps you just read 99% of the time.

Thanks for the information. I think I would just stay far away for the first few mins!

Edit: Are you running it on a hot plate or a flame? Do you hold it just above the mp of NaOH or much hotter?

[Edited on 1-5-2004 by Quantum]

BromicAcid's use of resistance heating

Look! For I have found good infomation

SAS Chemistry Guide to making small amounts of Sodium

They decribe in good detail the proccess that patu uses . They use a burner instead of a plate but they even use a 'tin' can. Worth reading to all interested in making small amounts of sodium.

Im going to try it tommorow!

Bromic's technique with patu's wire loop

I replaced the nickel sheet metal cathode with a loop of iron wire (.35 cm dia.). This allowed me to capture almost all visible Na. I hold the loop with a gloved hand--yes, molten NaOH is nasty-- and work it around the melt to get the Na inside. I then lift it out and slap it against the test tube containing motor oil (just because it was nearby). I ran the setup this way for an hour, removing the little balls of Na and quickly returning the loop to the melt. I was usually fast enough to get it back before the NaOH solidified. I kept the propane torch on low flame nearby so if I was too slow, I could re-melt and be making Na within a few seconds. this technique seems to work acceptably well (at least for me) to make small amounts of Na. I'm still amazed that a mere 35 watts in ( I'm running this setup at 10V and 3.5 A) is enough to keep the melt at 315C without any attempt to insulate from ambient. I have not yet perfected a way to clean up the Na that is produced. I've melted it under paraffin, xylene and ordinary paint thinner, all three of which result in shiny balls of Na but there is still come contamination (NaOH) visible each time. Ideas welcome.

Chemical burns!

I tried the sodium making using a spoon on my hotplate. After it melted I stuck a 9v transformer wires into it but nothing happened. I was convenced that it was dead so I put the 2 wires on my tough to see if it would 'tingle'. Sadly the wires had a little bit of NaOH on them! I had to wash my mouth out and now I have a little burn. I used a diffrent PSU(5v 20a) and I saw a little sodium but I could not collect it. I think its redisolving in the NaOH even though I turned off the plate and lifted the spoon off it when I applied the current.

In a way I failed but I did see a speck of shiny Na!

You freak!!! Dont lick live wires, no matter what the voltage is!!

To test if a PSU is working just shove the wires into tap water and see if it bubbles

[Edit:] Im sure you know this already, but you should try to *show* it

[Edited on 9-5-2004 by Saerynide]

I don't know what I was thinking! The sad thing was that I had a multimeter sitting on a bench 5 feet from me the whole time. I didn't get much sleep last night.

My Results

Sodium is much more soluble in cooler hydroxide then hot. For example, 25.3 g of Na will dissolve in 100 g of NaOH (l) @ 480C but at 800C only 6.9 g will dissolve. The papers I have say the solubility increases consistently to the solidification point of the hydroxide melt and therefore it is more economical to remove the Na before cooling the melt. (However if you run your melt too high you increase cell corrosion and decrease yeild by having the sodium react with it's hydroxide so there is some give an take, however I believe it is the general consensus to run the cell at as low a temperature possible.)

To remove your sodium you could try a chilled iron wire, upon touching the surface the sodium should freeze to it and be able to be scraped off, sorry if I've mentioned this method before.

Some information relating to the actual electrochemical process and reason for low yeilds:

"The hydroxide is electrolyzed
NaOH ---> Na + OH-
the sodium appears at the cathode and, at the anode, the hydroxyl is resolved into water and oxygen. The hydrogen that appears at the cathode is the product of a secondary chemical reaction, between the sodium at the cathode and water formed at the anode. It is therefore possible to have both hydrogen and oxygen liberated at the anode, and explosions may result..... The sodium which diffuses to the anode may also react with the oxygen there being evolved, forming sodium peroxide, and the later may react with more sodium forming the monoxide."

So wouldn't it be possible to add more water to the mix to react away the Na2O2 and such thereby increasing overall yields? Although adding H2O to molten hydroxide would be a bad thing, commercial hydroxide has appreciable water in it so just adding some of that could add the necessary water to hydrolyze the oxides of Na to the hydroxides and keep the yield high correct? However this would only be a matter of concern in large scale projects.

(500th post, I am the first International Hazard!)

[Edited on 6/7/2004 by BromicAcid]

Lol, that's a mistake that has been done before (see sodium acetate etc)

Your sodium borate contains crystal water, Na2B4O7 . 10H2O - and you can imagine what happens to the nascent sodium once it contacts water...

Here's a slight mod to Patu's method I tried with promising results. I've put in pics, hope you can access them. (PS. please can somebody u2u me the procedure for placing a pic directly onto the post using the code in the FAQ or upload it to the board, thanx)

This may have been discussed before but I've found that current density plays a major role in results. I used a 5V 50A supply and the more cathode I have in the molten NaOH, the less Na I get. It seems that the more voltage I need.

In the first pic youll see the setup I used. The cathode is a ceramic tube about 10mm ID with a piece of fence wire through it with the tip just protruding from the business end. (2nd & 3rdpic).

This was inserted into the melt as deep as possible without touching the tin obviously. After 5mins I carefully removed the tube and gave it a shake over some paraffin. Out popped a generous blob of sodium. (4th pic).

pic2

pic3

pic 4

Sorry about the mess Vulture.

Today was running the Castner Cell day!

The design, a cast iron pot with a hole drilled in the bottom though which a steel pipe was put, it had a large nut on one side and the other had an end cap that I cut in half, in between I used a silver washer. I put a nickel wire into this pipe and held it in place with some two part epoxy then when it hardened I melted some NaOH separately and poured it into the hole. I drilled a hole though my electric range burner to accommodate the pan sitting on it with the whole pipe going into it, in the picture above on the left you can see just the nickel wire coming out of the bottom of the range. That is where I connected the wire for the cathode.

I had half of a lid on the top with an adjustable nickel anode. Also in the top left hand picture you can see my 'bell' that had a top held on by a hinge.

Initial heating was attempted with a MAPP gas torch but it was not going fast enough plus the CO2 and H2O rising up from the combustion was seemingly quickly absorbed by the NaOH so I just set it on the hot plate and heated on high heat.

Surprisingly the hot plate managed to melt some NaOH in the bottom and I plunged the anode into it and started electrolysis which in turn heated the mix with resistance heating. Now nothing seemed to happen for a while, it did react with the coating on the pan though, it turned the NaOH slightly red.

Eventually everything melted and the anode depth was adjusted so that it could run at 12V and not blow a fuse. The amps were over 15. But the bell shorted out and Na was forming on the outside without my knowledge. That brings me to picture #2. Well, at least it was pretty.

For the rest of the experiment explosions occasionally blew the top off the bell. I got a small amount of Na recovered, ~2g but it does look like a lot considering the low density, but it's nothing considering the 1.5 kg of NaOH used in the cell.

Then it started raining bad, see attachment. Rain falling into molten NaOH was NOT my idea of a fun time, but what the hey, I kept going for over an hour.

But when I decided to stop I took the super hot pan and tried to pour it into a different pan. This was a pain because I forgot there was a pipe running though the hot plate, and it was cast iron so the handle was ultra hot. When I finally finagled it out I poured into the pan which now had significant rain water in it and NaOH exploded out! Molten NaOH spattered on my arms and shirt and table, the table promptly charred and attempted to catch on fire.

Even though it was burning my hand I forced the cast iron pan back into the burner and gave up for a bit.

Much later when it stopped raining I cleaned up my pan, the cast iron, after 4 hours on the hot plate with molten NaOH in it appeared to have suffered no massive effects, just some slight rusting.

But what I have done is not the important part, it's what I've learned.

1. Current density is everything, therefor both electrodes should be short and posses relatively little surface area.
   
2. Cast iron is resistant to NaOH to a suprising degree.
   
3. A torch is not a good item with which to heat large quantities of NaOH.
   
4. Constantly burning the H2 coming off with a constant running torch rather then allowing it to accumulate and ignite for sodium laying around is a good idea.
   
5. Insulation around the cell is a good idea, as is a lid.
   
6. Wood chars easily from the heat of a Castner cell and can start fires.
   

What I would consider to be a good vessel now. A reducing adapter for pipes, preferably with the largest differential. Then a plug at the bottom. Electrodes come in from the sides and almost touch in the center at the bottom of the vessel. In the center just below the surface across the top should be a sheet of stainless steel. Na accumulates on one side, O2 goes off on the other. So much simpler then my design.

I tried to draw some of this up but I'm terrible at using computer drawing programs.

Overall the experience was not bad at all, until I realized that the extra hot NaOH would attack me so readily.

I'd love to be your neighbor!

Thanks Vulture, and I bet my neighbor would not mind in the slightest swapping places with you!

I'm doing better today. The NaOH fumes and burning Na really messed with my eyes, don't know why I wasn't wearing goggles. They're all bloodshot now. Also where I got NaOH (l) on my arms.... It looks bad but it was only in small spatters, as I told my girlfriend "It's like I got liquid delete key splashed on me."

Castner Cell V 3.0 is almost 75% completed (although it no longer looks like a Castner Cell) I've got high hopes for it to continue my trend:

V 1.0 Cost about $90 and produced 0 g of Na
V 2.0 Cost about $30 and produced about 3 g Na
V 3.0 has a projected cost of $10 and should therefore produce >3g Na, right?

Actually this one is shaping up to be a good one. But what I really need is a diaphragm, which is hard to come across, the best material that I have come across is woven graphite thread which you can buy in large sheets for welding protection. But it's expensive and I don't need a large sheet, I guess I can do without...

I've made some hand drawings of my past cell and this one and I'm taking pictures of this one taken apart. Will post tomorrow or the next day.

Sodium Cell V 3.0

My setup for this is picture in frames 1 and 2, it is made from pipe fittings. In English measurements I have a 2 in. to 1 1/2 in. reducing adapter screwed into a 1 1/2 in. to 1/2 in. brushing. This makes the body of the cell. A 1/2 in. plug was bought to fit into the brushing and two holes were drilled at the extreme sides of the plug to accommodate two nickel electrodes.

One electrode was inserted to expose 1.6 cm into the melt and the other inserted to expose 1.2 cm to the melt, this resulted in a calculated current density of 3.51 A/cm2 for the larger electrode and 5.0 A/cm2 for the smaller electrode assuming amperature of 15 which is the Max my battery charger could get to. However the maximum current during the experiment was around 10 A therefore these numbers must be adjusted. The electrodes were spaced roughly 1 cm apart.

They were held in place and the plug was filled with a two part moldable putty advertised as a steel welding compound. It holds up reasonably to the temperatures and extreme conditions associated with this experiment plus it is non-condutive. As you can see in picture two this whole plug was able to be separated (although the electrodes are fixed if the current is too low NaCl can be added in a small amount, this greatly increases the conductivity of the cell, however it supposedly increases cell corrosion on a comparable level, the latter I have not noticed).

In picture one you see the divider I have in the cell. This was a tight fitting piece of sheet nickel that did not extend between the electrodes, stopping just short of them. This was so as the sodium rose to the surface it would stay away from the oxygen rising right next to it. It worked very well for this purpose and the two sides of the cell showed different corrosion afterward as a result of it.

Initially 15 - 20 ml of pre-molten hydroxide was poured into the cell and the current started immediately. However before I could get more NaOH melted (I chose not to heat the cell directly, relying on resistance heating to do the job) the sodium production, which began very rapidly blew all the NaOH out from between the electrodes and solidified the melt after rendering the condutive component ineffective.

It was my mistake that I attempted to add more molten NaOH to this directly as it did not melt the NaOH coating on the electrodes and therefore I was left with a centimeter of NaOH solid over my electrodes. In an emergency pinch I put some water into the mix and put two new electrodes in from the top, from here I used resistance heating to melt my way to the electrodes ( I also added a few pellets of KOH, this shows a definite increase in melting speed from eccentric formation but overall potassium yield is low), which took quite some time, but finally I was able to hook the imbedded electrodes back up and begin again.

The top of the melt kept solidifying and I had to break though the crust to get to the sodium below. In picture 4 you can see my stirring rod lifted from the melt, shiny blobs of sodium clinging to it. All in all I had a small collection of sodium, about 1 gram, however much more was created but did not get collected for various reasons.

This cell wins my certification for what could be the best medium production cell I have encountered. The most major problem being the electrodes were too close together, this resulted in the sodium forming occasionally shorting out the cell, and localized overheating which eventually nearly lead to the metalloid compound that thwarts some cells. Therefore I would increase the size of the bottom brushing and in turn increase the distance between the electrodes by nearly one centimeter.

Also some outside heating should be available, an electric coil would be ideal, that lacking insulation would do the trick. Better division of the compartments is not necessary but a lid, as shown in the picture is, not to keep gasses out but to eliminate spray from the electrode gasses rising to the surface.

I might experiment with sodium production more but I believe this is the pentacle of my abilities, it is cheap, easy to produce, and you could buy and endcap for the other side to cap it up when not in use for a constant sodium production apparatus.

P.S. Rift, my power source is a car battery charger.

I tried to draw this as well as I could, but I'm still not too good at drawing, hopefully, this combined with the pictures of the real thing above should give a thorough idea of what this thing looks like. [Note: NOT TO SCALE]

1. This is a reducing adapter, it reduces one pipe size to another, the top opening which is the opening to the cell is 2 in. and it reduces to 1.5 inches at the bottom opening. It is threaded at the top and bottom. On the inside it just slopes down gradually after an inch or two. It is stainless steel.
   
2. This is called a brushing. It does the same thing as the reducing adapter but less gradually. It basically acts as a total plug but with an opening of the desired size. This brushing fit the 1.5 inch opening in the reducing adapter (A) and left an opening in the bottom of .5 inches. for the plug (C) to fit into. This is threaded to screw into the bottom of the reducing adapter, and the hole in it is threaded as well for the plug.
   
3. This is the plug. It fits the hole in the brushing. What it is normally is simply a plug, there is no way for liquid to escape and it is for sealing off extra outlets in pipework and such. But in this one I drilled two holes to accommodate the electrodes, you can see this removed with the electrodes inserted in it, in the post that I did above.
   
4. This was the anode, made of nickel. Both the cathode and anode were made from wire of the same length and not cut down. The anode went into the mix deeper because current density on the anode is less important, plus I wanted to make sure there was enough conductive area inside to allow for that much current transfer.
   
5. This was the cathode. Small amount was exposed to the inside of the cell to achieve a high current density (5.0 Amp/cm2) which is significantly higher then the current density mentioned by Organikum at the beginning of this thread. Notice how I had the portion of the anode and cathode at the bottom of the cell different lengths sticking out, this was so I could differentiate between them once the cell was running because it all looks the same from above.
   
6. This is where the putty went, it came up to be even with the bottom of the brushing adapter. It stopped current from traveling between the electrodes too low and insulated them from shorting out with the inside of the cell. As I said before this was some kind of 'cold weld' mix that came in a roll, you cut off a piece and manipulated it till it was of a homogenous consistency then applied it and it hardened rapidly. Not recommended for this environment but it held up.
   
7. Part G is a piece of nickel plate. It was jammed into the cell to divide it in half. Notice how the electrodes are on either side of it, and the divider does not dip between them. This was just so sodium formed, rising up would not encounter much oxygen doing the same because less then 1 cm past the electrodes the rising elements were divided.
   

All the parts where put together without any sort of sealant, just used a pipe wrench. The divider need not be nickel, the electrodes don't have to be nickel either. The electrodes do not have to in that manner, the basic design is there.

This was not the perfected cell as I said before, but it has the ability to be nearly perfect for a middle sodium production level. The loop method actually seems nicer for lower quantity sodium. What my problem was, was the spacing of the electrodes, they did not seem to enter the molten salt far enough as my current didn't rise much above 10 amp, and they were too close, the solution stayed very molten in the bottom but the top was solid, which brings me to my second problem, external heating must be applied to the top.

If the electrodes are too close you can get a hot spot that can cause the metalloid to form, too far away and electrolysis becomes difficult. In my setup here they were very close, too close as my final conclusion. But other then that it worked really well.

The lid for the cell can be anything, I just set it on there to prevent the spray of molten hydroxide that comes up during electrolysis, and partially to keep the hydrogen coating over my molten sodium. All I used was a sheet of metal that did not even cover the top entirely. But it would be nice to have a plug for the top too since it is already threaded, easy storage and easy to render 'safe' should an emergency arise, just remember to turn off the electrolysis first.

I never got a good method to get the molten Na out though, I used a little scoop I made like a ladle but it was not good, brought over hydroxide, probably introduced impurities, hard to get the Na out. I personally wanted to try an eye dropper in preheated mineral oil, before taking sodium out sucking up some mineral oil to heat the dropper and pushing it back out, then using the dropper to suck up the sodium.

I put the electrodes at the bottom for two reasons, A) I couldn't find a good diaphragm to separated them that was not metal and I didn't want risk shorting the cell out. When the electrodes were at the bottom I could put the separation right above and it worked. B) When the electrodes come from the bottom there is lots of working room above the mix, you are not trying to avoid bumping electrodes and such.

Now that I've made the grey mixture I have a better grasp of it. If your electrodes are too close, localized high heating, therefore grey mixture forms, and it spreads and eliminates the constructive conductivity of the cell. As I've said, the electrode spacing is the most important part especially since the electrodes are immobile. Too close and you get overheating, sodium coming between the electrodes and shorting them out, and possible NaOH/Na mix formation. Too far and the actual work performed drops and the electrolysis is sluggish. The Na formed under the electrolyte, rose to the top, and collected under the hardened crust, when I broke the crust the liquid Na collected on my stirring rod (a nickel wire) and I took the picture.

Steels of all sorts have decent resistance to NaOH, copper is attacked rapidly by slightly wet NaOH when heated, haven't tried it in anhydrous. NaOH fresh from the molten mix on a stirring rod reacts quite violently with water (no surprise). If you have a stirring rod of sorts get some pliers to break off the NaOH that makes unwieldily coatings on it though usage.

I will answer more questions as they come

Great job!

Great job on the diagram, I think this is what we all needed!
I do have a few questions though

For one thing, how do you avoid the electrodes from shortcutting through the plug C? I guess plug C is made of metal right? In that case presumably the holes through which the electrodes are threaded are greater than the diameter of the electrodes, and the insulation is provided by the 'putty'? In that case, what is this mysterial putty made of that it resists molten 310 deg C hot NaOH (and is an insulator too, at those temps?)?

Regarding the grey NaOH/Na mix - this is the SOLUTION to a great problem! Everytime I tried making Na, I had this particular problem. How certain are you that the temperature is the cause of this problem? I.e. you wouldnt possibly be able to take the temperature between the electrodes? If temperature is such a crucial factor to this - then I wonder how they got it done in the original Castner Tiegel description.
The only way I can see the whole temperature issue to work out (without major investment in equipment) is to provide for decent insulation. I was thinking along the lines of resting the whole cell in a Al2O3/SiO2 putty, with a few Kanthal wires inside. The Al2O3/SiO2 would then serve as an 1) insulating material, 2) which is heat resistant and shows no great thermal expansion coefficients 3) also an electrical heating mantle. I guess then power could be applied to the heating wires within, and the optimum power would have to be determined simply by trial and error. But once established - Na production would run at the max!

Also, I was thinking about the separation of the electrodes. O2 and H2 interfere with the reaction, so there needs to be a decent separation between cathode and anode, as you have done with Ni-plate G. Great idea btw. I wonder where u get all that nickel from. Anyway - is there a chance to shape this into some Y shape - i.e. to ensure even greater separation of the gases? Or - even better - to have a relatively small exit hole (i.e. 2 cm diameter) for each electrode, so that 1) heat loss would be avoided, and this problem of crust formation and 2) to increase separation of the gases forming at the electrodes? Of course, then this would have to be a removable device, else the Na couldnt be taken off.

Generally I am not such a great fan of the idea of taking off Na every few minutes/tens of minutes. Dont you think it would be neater to have the Na accumulate to a certain extent (i.e. 10s of grams worth), then to take it out, add more NaOH, and continue? I mean - what I don't like about this is that it reminds me too much of the loop method - where one gets a scoop of Na every few minutes which weighs a small fraction of a gram. Surely this can be improved?

Anyway - great job once again - you are on the path to eternal Na-fication!

[Edited on 26-8-2004 by chemoleo]

Tonight I've been flicking together a cell myself.

The pictures should speak for themselves.

The idea is to collect the sodium formed at the cathode inside the pipe to reduce loss. There is a problem however; the pipe is made of galvanized steel. Now I know this isn't ideal and that it's probably going to shortcut the cell when the molten sodium collects in it, but at the moment I have nothing better at my disposal.

I'm planning to use thick copper wire both as anode and cathode material.







The outside ring of insulation is mainly to keep (possible) combustion heat from reaching the electrodes and such.

I'm going to test it without the electrodes but with the pipe as soon as the weather clears up (blasted belgian weather!!!) to see if it can actually melt NaOH and how the pipe will react.

EDIT: The vessel is a tobacco can. It's very similar material as a soupcan. We'll see if it holds up.

EDIT2: Bugger. The resistance between the inner and outer wall of the pipe is 0,4 Ohms at roomtemp. I wonder if passivation in HNO3 could solve this.

I'll be using an old ATX power supply that delivers 18A @ 5V.






[Edited on 28-8-2004 by vulture]

I've put the apparatus on an electrical hotplate (500W) without the electrodes, just to see if I can reach the required temperature. Currently the temperature on the outside of the can, approx halfway the apparatus is 305C. I still have a few other problems to solve, like inserting the electrodes, because the NaOH formed a rather hard crust.

I've exchanged the cathode holder by a piece of insulating material.

The main conclusion of this test is: insulation is everything

I got stuck somewhere around 250C with the setup like it was in the previous pics.




EDIT: Scrap that, temperature is currently 324C... There are some transparent spots developing in the NaOH too...

EDIT3: Ok, sofar the weather is holding out.
The NaOH melted to a black mass, possibly because of contamination from the can.
Let's see how the steel pipe holds up in there...



[Edited on 29-8-2004 by vulture]

Some testing yesterday with the steel pipe in the setup was rather dissapointing, the pipe seems to cause huge heatloss, I couldn't get above 266C. Alas, I borrowed my fathers hot air gun and activated it at 530C .

Twenty seconds...hehe...

The only problem I have now is that I keep shortcutting my electrodes either way through the can or by touching the pipe. And the ATX power supply has a rather sensitive shortcut sensor, cuts off immediatly.

Does CuO dissolve in hot NaOH? I have copper pipe which would be ideal, if I oxidize the outside to CuO I could prevent nasty shortcuts and sodium forming on the outside.

Some further observations...

I've been using a copper pipe now instead of the galvanized steel one. It is oxidized by the heat but immediatly reduced back to pure Cu when electrolysis starts.

Furthermore, the electrodes should be seperated atleast 2cm if the current rises above 2A. My run when they were to close resulted in something that could best be described as a piece of artillery...
BANG[pause]BANG[pause]BANG[pause] etc which was quite nerve wrecking.

Despite the electrolysis was barely enough to keep the NaOH molten I still end up with the grey stuff.

[Edited on 3-9-2004 by vulture]

Success. I managed to make myself some drops of sodium using the wire slope setup. However I found out that the NaOH solidifies to fast using such small quantities and stainless steel eggcup.

By the way. Corrosion was massive and rendered the dark but clear NaOH a brown "mush". Stainless steel was used both as cathode and anode. (Cathode was made out of "steelwire" and it didn't seem to be zinkplated, I guess this is stainlesS?). Anyways, the anode (stainless eggcup) was heavily corroded and a yellow salt was formed on it's side above the NaOH surface.

I found out that the intense gas release ejected the formed droplets of Na in my slope which made extraction very hard. And further on I can tell you that sewing-machine oil is not inert to Na.

Need to get myself some charcoal lighter fluid. Thanks everyone for giving me the inspiration and providing information. I'll be getting back to this experiment in the near future.

A pic of my entire setup (I used a computer PSU the 12V line was used- with 5V, not enough current was flowing).

[Edited on 30-1-2007 by chemoleo]

The molten NaOH

[Edited on 30-1-2007 by chemoleo]

The sodium pieces under paraffin, prior to melting.

I have no pics of the actual process, sorry.
I have some videos of the process (and melting together the Na, plus a video where the Na reacts with water, it caught fire), but they are way too big to upload.

[Edited on 30-1-2007 by chemoleo]

my own Downs cell

Wow, sounds neato

Calcium is insoluble, heavier and has a higher melting point than sodium, and more than the electrolyte hopefully (hence using enough CaCl2 for the low melting point eutectic). It also has a somewhat more negative reduction potential than sodium, so it probably does react to produce sodium metal. Hence the calcium chloride ought to be inert.

Tim

Wow, tumadre, that's a huge container for sodium production!

Are you sure you need that much insulation? (I suppose you've calculated everthing out, and I haven't, but it does seem like a lot of insulation. Have you looked into using kaowool insulation?)

Hmm, ceramics that resist chlorine. I don't know... what would the reaction be? BromicAcid, why are MgO or Al2O3 containing ceramics best here? (just curious as usual) I did a bit of googling on chlorine and ceramics, but never found anything. Molten NaCl/CaCl2 shouldn't be a problem at 550 deg. C. or whatever it was, but it will leak through porous ceramics noticably.

I still want to do this reaction too, as soon as I get a good power supply. I suppose I could rig up something using the MOT I got. As for heating it, sticking the apparatus inside of a furnace would work, although it wouldn't be too precise. But I don't think NaCl electrolysis is as sensitive to changes in temperature as NaOH is.

Molten NaOH in an ALUMINUM container!!! Now that's confusing, especially because of the thermite reactions that occur with lye and aluminum. (Tacho was doing experiments with these)

"Ceramics uneffected by chlorine"

Ahh! I typed a long response and the computer ate it. To put it simply, I used patu's method with an ATX 12v power supply, red devil lye, and a copper anode and cathode. I used a propane torch to melt the NaOH around the anode and cathode, and then kept only the center of the NaOH molten during the reaction. This prevents ALL corrosion from the outside container and prevents the sodium from overheating and forming that "grey stuff". If the sodium hydroxide gets too hot, it will just melt more NaOH.

The copper anodes and cathodes held up much better than iron, and although the molten NaOH was turned blue, blue is a beautiful color (much better than a brownish rusty color), and it didn't seem to affect anything.

There was splattering and popping, mostly at the beginning, after the loop on the cathode was mostly submerged, the popping was almost gone. I got a few miniscule splats of NaOH on my gloves and arms (and face shield) but only one really hurt. It took off a few mm of skin in as many seconds.

I really recommend this method to make a bit of sodium- it's easy, inexpensive, and really pretty safe. Sure, you can get a tiny bit of NaOH on you, but the reaction is very slow and controllable and so there's no danger of any deflagrations or anything spewing flaming mineral oil onto people's heads...

Here's a diagram.

Cyrus

[Edited on 30-7-2005 by Cyrus]

"Liver of sulfur"

Ah. I'll have to slap my contact then

Tim

i am still here!

If nothing else, if you assume it as AlCl3 in solution of molten NaCl, any Na formed will reduce the Al, forming Al metal, which I'm assuming is your goal. If kinda inappropriate for this thead...

So yeah you can't make an NaX eutectic with anything but alkali and alkaline earth salts with a higher reduction potential, namely, Li, K, Rb, Cs, Ca, Sr and Ba.

Hmm, most of those aren't that hard to come by. A quaternary or pentanary eutectic between Li, Na, K, Ca, Sr and Ba may melt as low as 200C.

Tim