Thermochemical dioxane approach

From Sciencemadness Wiki
Jump to: navigation, search

The thermochemical dioxane approach is the name given to a procedure developed by NurdRage to separate sodium metal from the sodium-magnesium oxide (Na-MgO) slag obtained by reducing sodium hydroxide with magnesium powder in a thermite-like reaction.

Summary and history

Magnesium metal is capable of reducing sodium hydroxide/oxide to sodium metal and magnesium oxide, since the formation of magnesium oxide is more favorable than the formation of sodium oxide/hydroxide:

2 NaOH + 2 Mg → 2 Na + 2 MgO + H2

This route was first brought to the attention of the general public by NightHawkInLight in April 2011, when he made the aggregate by reducing sodium hydroxide with magnesium powder, in a steel can. He was able to separate some of the sodium by adding the aggregate in a glass container with water-mineral oil two phase system. The aggregate reacts with water, but most of the sodium will rise from the aggregate, due to the bubbling formed around the sodium particles, and remain suspended in the mineral oil. The sodium metal particles are then collected and fused into a single mass.[1]

However, this method has some serious drawbacks, such as most of the sodium is destroyed during the separation in the water-oil system. Likewise, the surviving sodium is extremely impure, and requires significant purification. The yield is also not great.

Since then, many other separation ideas have been suggested: distillation of sodium metal from the aggregate in a furnace, melting the sodium in a high boiling point solvent (mineral oil, paraffin wax, toluene, xylene), extraction with liquid mercury, ion exchange, etc.

At the beginning of April 2017, NurdRage was using this Na-MgO aggregate to dry some dioxane that he made from ethylene glycol, during his pyrimethamine synthesis project. While distilling the dioxane, he discovered that the sodium metal from the aggregate coalesced into large metallic spheres, and was able to achieve separation of the sodium metal from the aggregate in good (though not spectacular) yield.


First, in a steel can add a mixture of equimolar amount of sodium hydroxide and magnesium metal (turnings or powder). To ignite it, a sparkler or fuse with magnesium ribbon is used and to limit the evaporation of sodium metal, cover the can with a heavy object. The thermite-like reaction will produce a sodium-magnesium oxide slag, which is very reactive and moisture sensitive. Large yellow flames are produced during this step. After it has cooled, grind the slag to a relative fine powder and dump it in a flask containing some freshly prepared and dried dioxane (avoid using old dioxane as it may contain peroxides). Connect the flask to a distillation setup and distill off the dioxane. As the level of dioxane in the flask decreases, sodium metal droplets will begin to appear at the surface of the solution and start to coalesce. As the level of the dioxane further decreases, the sodium droplets will start to coalesce in larger pieces. Heat the flask until all the dioxane has been removed. Now remove the droplets of sodium metal either directly from the flask or just dump the content of the flask in a dish and remove the sodium from the slag this way. The sodium pieces are then added in mineral oil to limit oxidation. To clean the sodium metal, simply heat the mineral oil until the sodium melts and then add a few drops of isopropanol until all the sodium has coalesced in a single clean droplet. To get more sodium metal from the slag, repeat the process until no more sodium can be obtained. Yield of this process is 41%.[2]


Since no electrolysis is required, this procedure avoids the use of highly corrosive molten sodium hydroxide or extremely hot molten sodium chloride or sodium/calcium chloride eutectic mixture, which require high temperature for long periods of time. By not requiring electricity (save for distillation apparatus and cooling fluid circulation), no electrodes are used, so this method can be done outside the lab. All reagents used in this process can be made from domestically available chemicals: dioxane from antifreeze and acid drain cleaner, sodium hydroxide from NaOH pellets, magnesium from firestarters/car parts/boiler anodes, mineral oil from baby oil or beekeeping stores.


The reducing process generates large amounts of heat, which causes some of the resulting sodium to evaporate, which ignites in a massive yellow fireball. This part of the process is a serious fire hazard, and even after the fire is extinguished and the aggregate has cooled, the resulting slag is still highly reactive, and can even ignite in open air, if scratched against a hard surface. For better separation, the aggregate must be grounded to a more fine form, and for this, a coffee grinder or a ball mill. During the grinding process, the resulting particles become smaller, and thus more reactive, which increases the chance of a fire hazard, while also destroying some of the sodium from air oxidation. This can be limited by either grinding the aggregate under inert gas or after being soaked in dioxane. Since dioxane is highly flammable, adding it to the potential pyrophoric aggregate is dangerous, and the fire hazard is only eliminated when the entire aggregate is immersed under dioxane and placed in a flask with a ground glass stopper. The yield of this procedure is not great, only around 30-40%, so you will need to use larger amounts of precursors to obtain any useful amounts of sodium metal.

Other alkali metals


Doesn't work, due to lithium metal being more reactive than magnesium.


Works, however, the resulting aggregate is pyrophoric, which makes the separation extremely dangerous.


Not tested so far, but given the high reactivity of the potassium-magnesium oxide aggregate, it's assumed that the resulting aggregate in this case will be even more reactive, and separation might be next to impossible, unless the reaction is performed in an inert gas chamber or in vacuum.


Same as rubidium.

See also



Relevant Sciencemadness threads