metalresearcher
National Hazard
Posts: 758
Registered: 7-9-2010
Member Is Offline
Mood: Reactive
|
|
Next attempt in obtaining sodium metal
The Chemicals:
NaOH + Mg metal snippets
The Retort:
The vessel (top right) can be unscrewed from the bent pipe.
As the reaction temperature (at most 600oC) is a lot lower than Na2CO3 + C (at least 1200oC), lower requirements are needed for this retort as it is
just mild steel.
Next weekend I am going to test it. I will put the duct pipe in a bath of lamp oil but only shallowly to prevent flashback of lampoil into the hot
vessel.
I'll keep you informed.
|
|
|
metalresearcher
National Hazard
Posts: 758
Registered: 7-9-2010
Member Is Offline
Mood: Reactive
|
|
YES !!! I got it Sodium metal !!!
The first tiny amounts of Na metal after opening thevessel.
Watch here how.
<iframe sandbox width="425" height="349" src="http://www.youtube.com/embed/OUn8LY7AVxs?hl=en&fs=1" frameborder="0"
allowfullscreen></iframe>
|
|
|
smaerd
International Hazard
Posts: 1262
Registered: 23-1-2010
Member Is Offline
Mood: hmm...
|
|
Excellent!
That hydrogen gas explosion is not for the faint of heart hahaha! Maybe there is a way to build a chimney instead of the beaker. So that the hydrogen
gas can be vented well above the heating source  ? Looks a little scary to scale
this reaction up hehehe.
|
|
|
m1tanker78
National Hazard
Posts: 685
Registered: 5-1-2011
Member Is Offline
Mood: No Mood
|
|
Nice demo! If you succeed in collecting and cleaning up the sodium, please refer to the following thread to share or discuss your procedure.
http://www.sciencemadness.org/talk/viewthread.php?tid=16225
Is the oil dip really necessary?
Tank
|
|
|
Takron
Harmless
Posts: 47
Registered: 19-5-2010
Member Is Offline
Mood: No Mood
|
|
Getting Metals
The science behind making the metals are fun, but if you want some in order to actually do stuff with.
Sodium
Potassium, Gallium, and a lot more.
|
|
|
metalresearcher
National Hazard
Posts: 758
Registered: 7-9-2010
Member Is Offline
Mood: Reactive
|
|
I tried again with KOH + MG but no K :-(
This morning I repeated my experiment of Friday evening but now with KOH, furthermore, same circumstances (it was a new clean retort).
But no K metal was formed .... Why does it yield Na but no K ?
<iframe sandbox width="425" height="349" src="http://www.youtube.com/embed/dTUo0aRKVR0?hl=en&fs=1" frameborder="0"
allowfullscreen></iframe>
|
|
|
Neil
National Hazard
Posts: 556
Registered: 19-3-2008
Member Is Offline
Mood: No Mood
|
|
Did you look in the elbow of your retort?
|
|
|
#maverick#
Hazard to Others
Posts: 209
Registered: 7-4-2011
Member Is Offline
Mood: hybridised
|
|
I am experimenting with getting the sodium separated from the slag will post a video/notes if I manage to do it successfully. Let's just say this
weeks science shopping list includes brake fluid and Baby oil
|
|
|
DougTheMapper
Hazard to Others
Posts: 145
Registered: 20-7-2008
Location: Michigan, USA
Member Is Offline
Mood: Energetic
|
|
Alright, here's an idea as well as a few interesting thoughts:
Sodium boils at roughly 880C, presumably hotter than the reaction. It can be assumed that sodium will be lost through vaporization. The logical way
around this is to condense the sodium out of the hydrogen stream, but this creates obvious collection issues.
What if you built a closed steel container which would contain all the reaction products while being rated to contain the volume of hydrogen produced
under pressure? It's not as suicidal as it sounds: Per my calculations, 2 moles of NaOH and 2 moles Mg (roughly 120 grams of material total) are put
into a 0.5L steel vessel, something probably looking like a "pipe bomb" (a 24-inch section of 2" steel pipe with caps). The reaction is initiated
electronically, producing 1 mole of H2 gas. Per the ideal gas law, the internal pressure would stabilize at something like 41 atm once it had cooled
down. 2-inch sch40 steel pipe has a burst pressure of almost 550atm and this is the thinnest available wall.
Of course this raises questions as to the relationship between the strength of steel pipe vs temperature. I would imagine it would be best to initiate
the reaction electronically while the pipe is submerged in cooling water; this would not only maintain the strength of the pipe but reduce collateral
damage should rupture occur. I don't think this will happen though since the working pressure is 105 atm anyway.
The same 0.5L vessel with 1mol H2 inside at 1000C would be at just 119atm and at 2000C, 283atm . I believe this will not pose too much of a problem as
long as the pipe is kept cool. I have yet to look up tables for the vapor pressure of sodium but this will also be a contributing factor.
Anyway, the pipe would be allowed to cool whereupon the hydrogen is vented by loosening a bleed valve installed on one of the caps. (Also perhaps
substituting a reducer for a cap on one end, threading a small plug into the reducer, and loosening this to bleed the H2) It will likely catch fire
since H2 heats when expanded through the Joule–Thomson effect, and should be performed in an adequate environment with suitable safety gear.
After the inside has reached atmspheric pressure, the pipe is then flooded with mineral oil or wax, is resealed, and goes into a boiling salt water
bath. Since sodium melts at 97C, the sodium should dislodge form the slag and agglomerate in the liquid state with vigorous shaking of the vessel.
After this, the oil is drained and the apparatus cooled and disassembled to remove (theoretically) 46g of sodium metal which will have agglomerated at
the bottom. It can then be melted and purified under further molten paraffin baths. If this process works, it will prove even more cost effective than
the electrolysis cell found here: http://www.sciencemadness.org/talk/viewthread.php?tid=9797 as well as much faster.
Oh, and it would probably be advantageous to purge the vessel with an inert gas prior to ignition to prevent the H2-O2 reaction from forming water,
increasing pressure, and damaging yields. Also, adding a few steel ball bearings to the pipe will help pulverize the slag during the
melt-agglomeration phase while also making the whole device way more suspicious with regard to its intent. Add them with the oil I guess.
As another bonus, with some simple plumbing the H2 can be collected in homemade low-pressure cylinders for use in torches, balloons, hydrogenation or
whatever rather than vented to the atmosphere. Of course, this will require careful purging to avoid an explosion with O2, but is certainly possible
with a little care. Heck, you might as well do something with half a liter of hydrogen at 600psi.
Here's some info on the pressure ratings of steel pipe: http://www.ihservice.com/PDF's/Pipe%20Selection%20Chart.pdf
But that's just me. What do you guys think?
-DTM
Victor Grignard is a methylated spirit.
|
|
|
Neil
National Hazard
Posts: 556
Registered: 19-3-2008
Member Is Offline
Mood: No Mood
|
|
DTM you only have 87 posts, you are supposed to wait until you are 100 posts before you are a hazard to others...
Hot thermite reactions in closed vessels make face go away from head.
Also hot hydrogen makes strong steel weak
its a bit like suggesting that a CO2 cartridge filled with charcoal and saltpetre won't explode because when it cools it will be well within the burst
strength of the CO2 cartridge.
|
|
|
Dr.Bob
International Hazard
Posts: 2734
Registered: 26-1-2011
Location: USA - NC
Member Is Offline
Mood: No Mood
|
|
Plus, if there is any air in the tube beforehand, it will mix with the hydrogen to make H2O (steam) and more heat, which could also create a failure.
This whole thing would be much better if done under an inert gas (helium or argon are best for Mg and Na, nitrogen and CO2 are reactive with Mg.)
Neat experiment, but I would not be putting my unprotected hand near a oil bath which might explode or catch fire. Use a longer tube and just blow
some helium through it to dilute H2 as well as keep inert atmosphere.
Bob
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
Congradulations MR, you have certainly put the effort in! 
| Quote: | | Is the oil dip really necessary? |
The dip tube is there to prevent the magnesium and sodium reacting with the atmosphere. I've seen people making it with lithium and NaOH before, in
the open air, and it works, but it's not likely to produce much yield.
| Quote: | | Sodium boils at roughly 880C, presumably hotter than the reaction. It can be assumed that sodium will be lost through vaporization.
|
It's possible, due to the mode of heating and the reaction taking place, that some of the sodium was blown out of the dip tube with the hydrogen. That
may have even been the cause of the impressive fireball. I'd have shit my pants if I'd tried that indoors, certainly a stand well back, outside,
experiment.
| Quote: | | The same 0.5L vessel with 1mol H2 inside at 1000C would be at just 119atm and at 2000C, 283atm |
Just 119 atm (1,700psi) at 1000C? That is a big, big ask for something made at home, particularly with regards to what else it
contains and the method of heating. Most steels (stainless included) begin loosing a lot of their strength well below 1000C, and the
service temperatures are all around or below that point.
| Quote: | | What do you guys think? |
Don't try what you've suggested.
Have a look at Parr Reactors for an idea of what can be done in terms of pressure and heat.
The typical operating temperature of these is 300 to 600C and 100 to 300 atm. They need building in a very special way to ensure they don't explode,
lining with glass or PTFE to keep the chemicals off them and they are then shrouded as well.
Look at the kinds of alloys they are using. The reactors are machined from a solid bar.
| Quote: | | This whole thing would be much better if done under an inert gas (helium or argon are best for Mg and Na, nitrogen and CO2 are reactive with Mg.)
|
Yep!
I've said that before as well and it would certainly benefit from it, but now MR has a positive result for sodium, he knows it's working and can
improve on it.
I'd go with argon. Anyone with a MIG or TIG welder will have a cylinder of it, it's unreactive, it's cheaper and it's heavier than air, so it sinks in
containers and blanket the contents.
| Quote: | This morning I repeated my experiment of Friday evening but now with KOH, furthermore, same circumstances (it was a new clean retort).
But no K metal was formed .... Why does it yield Na but no K ? |
Could it be that the amount of moisture in the KOH and air in the retort is causing it to go straight back to KOH? Is it a new bottle of KOH and how
much are K are you expecting out?
I would again suggest some argon if only to be sure the air is well displaced before heating it up. Attach a long bit of thin tube to the argon
supply, poke it all the way down into the retort, submerge the outlet and bubble argon in to push the air out. Then pull the argon tube out without
taking the dip tube out of the oil.
I noticed you got no flame in the second attempt.
Try catching some of the gas that comes off and light it to see if there's any hydrogen there, or if those bubbles are just air and water vapour.
[Edited on 27-9-2011 by peach]
|
|
|
DougTheMapper
Hazard to Others
Posts: 145
Registered: 20-7-2008
Location: Michigan, USA
Member Is Offline
Mood: Energetic
|
|
| Quote: |
The retort? How can it be both hot and cold simultaneously?
|
Ah, I was referring to the gas temperature inside the reactor. The reactor itself is kept as cool as possible by submergence in water to maintain its
strength.
I don't think it's really as crazy as all that. I do say "just" 119atm only because that figure is almost within the working pressure of the pipe. The
"water hammer" effect induces spikes in the order of a few thousand psi anyway, so there's quite a bit of safety margin. To reach the burst pressure
with 1 mol H2, (according to some quick calcs) the gas would have to hit 3624C (6500F), something I really doubt would happen. In fact, I'm about to
find the energy of the reaction and specific heats of the products, so assuming no heat transfer to the vessel I can find the theoretical max temp,
and thus pressure, of the H2 gas.
After that, it's only a matter of finding if the pipe strength curve falls below the H2 pressure curve as the hydrogen cools by losing heat to the
vessel. Also keep in mind that the amount of reactants could be halved or some other fraction to significantly reduce pressures. Oh, and ignition will
likely come from a coil of nichrome wire directly buried in a loose pile of reactants.
Regardless, point well taken. This is something to never go about doing and I would not recommend anyone to try it either. There, now
I'm back to being a "hazard to self" hahaha
-DTM
Victor Grignard is a methylated spirit.
|
|
|
metalresearcher
National Hazard
Posts: 758
Registered: 7-9-2010
Member Is Offline
Mood: Reactive
|
|
As @peach said, I used too much KOH in the second attempt of Sep 25.
The stoichiometric amounts are about 2.5 KOH to 1 mass part Mg but commercial KOH contains 15% water so take 2:1.
Tonight I did another attempt with 2.5g Mg and 5g KOH which reacted MUCH more vigorously.
But the amounts of K were still tiny but is was there. Sundayb there was no K at all. But argon is a good idea. Trying to get a cylinder from a TIG
welding shop ?
<iframe sandbox width="425" height="349" src="http://www.youtube.com/embed/UPQEEnWw1YY?hl=en&fs=1" frameborder="0"
allowfullscreen></iframe>
|
|
|
m1tanker78
National Hazard
Posts: 685
Registered: 5-1-2011
Member Is Offline
Mood: No Mood
|
|
I believe argon would be a waste of money in your application (but nice to have in general). It's useful to displace oxygen before the reaction begins
but then there's that pesky 15% water. If you could reliably tackle the moisture problem beforehand, you'd kill 2 birds with 1 stone.
Your yield will increase dramatically.
The reaction will be much less ...um... explosive. In other words, it won't have such a tendency to splatter everywhere.
Still, it's a nice start. Thanks for sharing the vids.
Tank
|
|
|
metalresearcher
National Hazard
Posts: 758
Registered: 7-9-2010
Member Is Offline
Mood: Reactive
|
|
Maybe heating the KOH to, say 1000oC (then it melts) and then let it freeze and crush it ?
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
