unconventional sodium

For the diehards wanting to get to sodium metal, I enclose the following patent, which describes the use of a divided cell (only using fibreglass matting) and the use of an ionic liquid (methanesulfonyl chloride) to realise sodium metal <200C:

http://www.freepatentsonline.com/6787019.html

I realise methanesulfonyl chloride may be expensive and hard to acquire, so I hereby provide access to the means of producing the same:

http://www.freepatentsonline.com/4997535.html

Uses methane, sulfur dioxide and chlorine gasses, in a glass tube, under UV light (specifically stating that low vapour mercury lamps work).

BTW, Heres another method by which to get to the borohydride:

http://www.iop.org/EJ/abstract/0953-2048/17/12/L01/

In this synthesis they used magnesium powder and powdered elemental boron, in an aluminium container in the microwave to make magnesium diboride, which can be turned into x.BH4 (where x=Na, Ca, K, (NH4)2, etc.) by introducing the diboride to a concentrated solution of x in water (yeilds suck according to the patent), releasing copious quantities of hydrogen and some x.BH4. I assume anything else in the reaction vessel would be reduced however.... (In which case, fuck the x.BH4)

I read somewhere that boron can be made from boric acid via reduction (at some considerable temp's) with carbon (although be careful, boron carbide will form if you go too hot) and also by electrolysis (presumably of the boron-trihalides - also perhaps BF (DON'T DO THAT ONE ANYWHERE NEAR ME)).

[Edited on 26-3-2007 by tupence_hapeny]

Yup, that was a fuckup.... LNG is what you need.... Town gas as it called in the patent

EDITED FROM HERE ON 4 April 2007 by tupence-hapeny

The other ionic liquid (the phosphoryl one) is an absolute bitch to make, given that the only synthesis that I could find requires (as a starting material) PCl5. Anyhow, here is the synthesis:

http://answers.google.com/answers/threadview?id=433962

From what I have read, white phosphorus is reacted with HCl (not for the fainthearted) and then reacted further with Chlorine to give the PCl5 As the easiest low temp route to phosphorus is via red phosphorus (~290C it supposedly distills over - let me suggest an inert atmosphere) this would appear to be unworkable and unfeasible.

Unfortunately, the low temp electrolysis patent does not detail high yeilds (or high current attempts) with the methanesulfonyl chloride. However, here is a more detailed version of the original patent (unsure if it is a co-application or a later application) which seems similar but includes better details:

http://www.freepatentsonline.com/20030094379.html

Here is another method of making methanesulfonic acid:

http://www.freepatentsonline.com/6207025.html

[Edited on 5-4-2007 by tupence_hapeny]

Ah, so instead of sodium combining with the solvent (which does happen, just really slowly), the chlorine does? Could very well be.

Electrolysis of sodium amide is probably straightforward. But that doesn't help any. Hydroxide probably works, but that can be done somewhat easier in its own melt.

Tim

You DON'T want to mix chlorinated solvents with alkali metals.

Sodium-potassium alloy in an ampoule inside a test tube of chloroform is a classic "flash-bang" (and shrapnel) device.

Why do you want your sodium to float on the surface of your solvent?

If that's really what you want, propylene carbonate might fit the bill.

[Edited on 10-31-2007 by Eclectic]

Yes thats what I thought, the H2 formed is meant to protect the metal, just as in the electrolytic method of K production from KOH. But in this case the K is heavier than the bath, so as soon as the reaction starts subsiding you can turn the stirrer off and the K, if it actually existed, would sink to the bottom. There would be no need for Ar at the end, if it wasnt required in the beginning.

These little details differentiate real work from fantasy. I noticed this oddity, but didnt put too much weight on it before I tried the experiment. Now I see its true import.

I didnt know about the legal status of this patent. How does one check that? To be honest I feel emotionally gutted, having been taken for a ride like that. I have been prepairing for this experiment for over a year.

I really dont believe anything can save this patent. Grignards are hard to start because of moisture and a protective MgO layer over the Mg. I believe the ultrasound is used to form a breach in this layer and expose fresh Mg to the organic halide. This reaction IS meant to happen with water present, the Mg acts as a dehydrating agent and supposedly violently. Well the Mg did react with water - it became a dull colour, there was no problem with initiation. Its just that it didnt react with the KOH. i guess I should have known - try heating Mg with KOH powder to 200C - they dont react. So stirred in a hydrocarbon its likely to be even more so. The fact that the supposedly measured quantities in the patent were actually calculated, and the elementary error that the reaction goes to MgO resulting in an impossible 'measurement' of H2 released in the 100-130 C phase, the lack of any K in my experiments, conducted exactly as in the patent, and the sluggish as one would expect, rather than 'violent' reaction of the Mg with water solidifies my belief that its all rubbish. The guy never did the exp.

[Edited on 23-12-2007 by len1]

This time I ran the experiment for 6 hours. Changes from the previous runs were as follows

1) The heat output was upped so the reagents were at a vigorous boil at 200C, to achieve this the RB flask was placed in a sand bath maintained at 240C.

2) Freshly cut Mg turnings, from a third source were used and thrown straight into the hot solution - there was no time for the Gringard stopping MgO to form

3) The mixture was stirred very vigorously from a special fast overhead motor

4) As before I did not use my good ground-glass multi-necked flasks, KOH at high T etches glass, and I dont want them stained, especially on an experiment that appears dodgy. I used a cheap flask with a rubber stopper - there was no perceptible leak or interaction with the rubber at 200C.

Result - some white percipitate, which would have to be Mg(OH)2 formed and Mg darkened from interaction with water. No K formed as evidenced by throwing contents of flask into water. Zilch in the language of the patent inventors.

I am now very confiedent this patent is rubbish and am sorry I wasted time on it. Reasons are:

1) No K formed in any experiments despite following patent closely, three different sources of Mg, the last freshy cut

2) The patent clearly 'cooks' the amount of H2 evolved. In water Mg + 2H2O -> Mg(OH)2 + H2, not Mg + H2O -> MgO + H2. SO 278mmol H2O releases 139mmol H2 not the 278 that the patent quotes. The hydroxide doesnt dehydrate at 200C. The very accurate wrong 'measurement' is in exact accord with the incorrect theoretical formula, which shows that it was never measured. The same goes for the H2 released subsequently - construed to add up exactly to 1000mmol, the amount of pure KOH used. If they cooked one part of their patent, they are likely to have cooked the rest.

3) This type of reaction was not really expected. KOH and Mg dont react at 200C the KOH form only an emulsion in the hydrocarbon and as such its interaction with the Mg in this medium is not expected to be better than in the solid state.

In short 'BUSTED'. I feel unclean for having participated in this rort. I look forward to doing something more interesting.

Quote: Originally posted by hashashan

I probably don't understand something but had anyone tried electrolyzing a NaCl solution in glycerol? what will happen?

think I found a new one. it uses amalgumated copper electrodes and a salt solution.

it works for gallium as well

http://www.patentstorm.us/patents/4012298/description.html

production of sodium metal

Potassium made in Germany

Quote: Originally posted by len1

I dont think your results are different. A white precipitate is generated from the reaction of Mg with H2O, that would occur in pure hot water as well. The reaction generates H2 although at half the rate to what the patent authors 'measured'. Thats the effervesence. But the point of the patent is not to form magnesium hydroxide. My request to you or anyone else who thinks this reaction works, is please show the potassium. Anything else is pure speculation. Len [Edited on 27-12-2007 by len1]

len1 and everybody else, i understand that skimming sodium off a solvent is perfect and you get it nearly clean.
however, has nobody thought of dipping a strong magnet into any nps and thus recovering sodium from MgO or Al2O3 or whatever acids and salts?

Or am i just clumsy?

Quote: Originally posted by Alastair

dipping a strong magnet

Quote: Originally posted by watson.fawkes

What substance in the reaction mixture would respond to a magnetic field, and how would it respond?

Quote: Originally posted by Alastair

Sodium is paramagnetic, so "is only attracted when in the presence of an externally applied magnetic field" (wiki)

Quote: Originally posted by watson.fawkes

The strength of that interaction is minuscule. It also decreases with temperature. Plus you're talking about a liquid, not solid, so there's yet another source of spin disordering and an even lower strength of interaction.

Quote: Originally posted by Alastair

I won't start a pointless discussion about sodium being a solid.

Quote: Originally posted by watson.fawkes

That's good, because in all the elemental sodium syntheses I've seen here, the sodium comes out at elevated temperatures, as a liquid.

That's good, now you can start a chemistry/physics revolution and change all the literature on sodium, and everything else that liquifies just below 100C. Good luck

Quote: Originally posted by Alastair

That's good, now you can start a chemistry/physics revolution and change all the literature on sodium, and everything else that liquifies just below 100C.

Quote: Originally posted by watson.fawkes

Are you actually asserting anything? I can't tell. Otherwise I'll have to assume you're trolling.

Quote: Originally posted by Alastair

If you think it is a big problem to cool sodium down to room temperature, you can aswell go on theorizing about it. My point was just to propose an idea for someone who would actually try it, wouldnt hurt i think?

This shit, Self-propagating synthesis, is insane. Microwaves with a susceptor (generally SiC powder packed between two crucibles) have been used to ignite what are essentially thermite mixtures, which burn in a vacuum/gasless environment, producing everything from TiC and other refractories to NiAl Alloy (useful). I suspect they could be harnessed for the preparation of alkali metals, as the volatile components of various Pidgeon-type processes - just put a hole in the top of an old MW, insert a large crucible, then pack it with SiC, then another crucible, with a cover that can hold a vacuum (no MW are reaching this, so bolt-down lids would be a good idea).

This is some sick shit, thermite in a MW - perhaps we can make Si without needing Sulfur (YAY).

Quote: Originally posted by aliced25

This is some sick shit, thermite in a MW - perhaps we can make Si without needing Sulfur (YAY).

Sure. That thermite is slightly exothermic, so if you heat it enough you'll end up with molten alumina and molten Si.

The Standard Enthalpy of Reaction for SiO2 + 4/3 Al === > Si + 2/3 Al2O3 is about - 212 kJ/mol, that's quite a bit for free. Provide the missing heat (to get to about 2,500 C) with MW and Bob should be your uncle. Just don't expect chip grade!

Quote: Originally posted by aliced25

Don't want chip grade, what I want is a reductant for various other metals, including lithium, possibly strontium, and especially calcium.

Quote: Originally posted by watson.fawkes

If all you want is a source of silicon for reduction, there's no need to use elemental silicon. If crude silicon works for you, then ferrosilicon should as well. There are four low-melting eutectic points, that is local minima on the phase diagram. The one with the greatest percentage of silicon is the eutectic of FeSi and Si, at around 60% Si. But the whole range from ~ 50% Si to 60% Si has a melting point at 1207 - 1220 °C, a couple hundred degrees lower than that of Si itself at 1414 °C. This is rather easier to get to, as the T4 ratio is 1.8 (that's ratio of radiative heat loss). And it's a lot easier to make an arc furnace than screwing around with microwaves. Welding rods, castable refractory, and an ammeter (to know when to push the rods in) is all you really need. At the materials are dirt cheap: sand, carbon, and steel scrap.

Ferrosilicon was mentioned in the other thread.

Ferrosilicon can be made easily by mixed thermite (Fe2O3 + SiO2 in the desired ratio, + Al and CaF2). The Fe2O3 acts as the heat booster to provide the 'missing heat' for the straight SiO2 thermite. I've done this with good results a few years ago. Because various FeSi mixtures are more dense than pure Si, you also get much better gravitational separation which tends to be a problem with sulphur assisted SiO2 thermites. Not to mention the H2S!

FeSi globules from aluminothermy: still quite uncorroded after so many years:

Quote: Originally posted by aliced25

What is your idea, first go with a propane kiln and something like this (http://www.scribd.com/doc/123531537/Magnesium-silicothermic-...)? A simple rotary-vane pump should get enough vacuum, if not we need additional heat. Working with thermite mixtures we run into the issue that even under vacuum they start a self-sustaining combustion reaction (http://en.wikipedia.org/wiki/Self-propagating_high-temperatu...) which will rapidly reach and breach the ability of most metals to contain the reaction with some products.

Precisely, these reactions aren't self-sustaining, they need vacuum in order to proceed. @Watson.Fawkes, we are trying to avoid the need for an arc, for one thing it is going to seriously complicate building a vacuum container and for other, external heat is applied to cause the reaction to kick in.

For mine, if we can make pure silicon, we can make pure silicon, it is about the same price on pyro sites as Ferrosilicon anyway, the only thing that has put people off it in the past is the slag forming H2S. If external heating can be used to kick the fucker off, then a simple Al/SiO2 reaction with Si and Al2O3 as the slag would be fucking brilliant. The less crud in the pot the better when trying something new out is how I look at it (although Pidgeon, et al, did use Ferrosilicon for the reduction of both Lithium and Magnesium, because it was cheaper than Al - not our problem, for us Al is significantly cheaper, as is Mg for that matter).

An externally heat-ignited thermite mixture, would presumably have very similar effects on crucibles as the normal thermite reactions, so I would suggest that any vacuum vessel be designed to have a single-use crucible, that is cheap and expendable and can be placed in there easily (ie. the thing should be able to be pulled apart, in fact modular would be great). The difficulty is, how to we get the external heat through an expendable crucible to the reagents, sufficient to raise them to initiation temperatures? How do we get a metal that will allow us to push that much heat through it without melting/deforming under vacuum? I know the Russian's used a Tungsten wire to initiate the SHS mixtures, which could be conceivably pushed through the vacuum housing and potentially through a small hole in the base of the expendable crucible. But that gets away from the propane idea into electrodes.

A lot of questions to be answered yet, not to worry, we have the collected literature of NASA, the US Military and the USSR's best Scientists to help us answer it. A straight crucible, made of something like alumina (or porcelain - such as a porcelain vase/cup) that costs fuck all, would be the first part. SS to contain it, which would need to be vacuum sealed to the main chamber (containing air/water cooled condenser and collection cup). I'm thinking flanges here, bolted flanges joining the two components with a gasket.

I'm thinking if we keep the original small, then catastrophic failure of any component would be no more dangerous than a normal thermite reaction. A vertical design for the area going into the furnace, simplifying the system and probably an angled condenser coming out of the area above it to both the collection cup and vacuum take-off (like a normal pot-still). I think given the range of materials this will be tested with, from alkali metals to phosphorus, SS is going to be the best option, I'm just worried about it's ability to cop the heat at the furnace end.


[Edited on 5-6-2013 by aliced25]

Quote: Originally posted by aliced25

we are trying to avoid the need for an arc, for one thing it is going to seriously complicate building a vacuum container and for other, external heat is applied to cause the reaction to kick in.

Probably the best way to work this is to try out the preparation of Calcium metal from a much cheaper ore, similar reaction, similar product, just a cheaper ore.

Attachment: Di.Wang.Tao.Feng.A.Novel.Vacuum.Aluminothermic.Reduction.Lithium.Process.pdf (305kB)
This file has been downloaded 2180 times

Attachment: Lebedev.Shishkina.Dmitriev.Preparation.of.48Ca.and.its.Use.in.Experiments.on.Synthesis.of.New.Superheavy.Elements.pdf (150kB)
This file has been downloaded 921 times

Attachment: Loomis.The.Production.of.Metallic.Calcium.by.Thermal.Reduction.pdf (1.3MB)
This file has been downloaded 1120 times

I am planning to reutilize my micro-castner soon and post some pictures because I'm gonna retry the chlor-alkali - combination process with molten NaCl:CaCl 70:30 at 650C. The reactor design is following (and later you see picture of the reactor I have constructed earlier at springtime) - ATTACHMENT. The reactor essentially consists of anode from underside and cathode around the reactor from upside, and I will contain it within clay an layers of rockwool to make it a lot more energy efficient. I'ts made out of 316 steel and I'll be using graphite anode and iron cathode with 5V at about 50-100A depending on the conductor's cross diameter and EC values. This is rather low power, but since it could procude sodium and clean chlorine gas for chlorination processes, I'll give it a go-code. The chlor-alkali will be waiting for polymer membrane or 2 kilograms of mercury until procceeding.

The reactor volume is about 4 liters so it should ideally hold about 6-7kg of chloride salts, which of about 5kg is sodium based and will be electrolyzed, and of this, in theory, about 2000 grams of sodium and 3000 grams of chlorine gas can be obtained. Heating should be carried out using 6kW propane torch.

I made test run with my electric heater. This is molten sodium chloride from my reactor, 500 grams, the color is from temperature.

I was thinking this type:

http://images.flatworldknowledge.com/averillfwk/averillfwk-f...

The anode shall be put through the bottom with plaster fitting that insulates it from the bottom, and the cathode through the top flange that is lifted off when the system is opened.

How the sodium shall interfere with the cathode? What it does? Should I consider this structure straight away?

https://s3.amazonaws.com/readers/2010/02/17/downscellschemat...



[Edited on 28-10-2013 by testimento]

Quote: Originally posted by testimento

I have my old power source rated at 400A at 6VDC, but with rewinding it can go as high as 600 amps[...]

Quote: Originally posted by testimento

I made test run with my electric heater. This is molten sodium chloride from my reactor, 500 grams, the color is from temperature.

Quote: Originally posted by PirateDocBrown

If I had to guess, using Mg to reduce NaOH, and then some means of purifying that.