Sciencemadness Discussion Board - Manganese Thermite with Lime

Manganese Thermite with Lime

blogfast25 - 12-7-2013 at 12:34

Another attempt at making the MnO2/Al Goldschmidt reaction yield more and better metal largely failed. The purpose here was to test lime (CaO) as:

1) A heat sink to cool the reaction somewhat
2) A slag fluidiser: CaO and Al2O3 form a series of lower melting calcium aluminates. Lower slag viscosity should lead to better metal/slag separation and thus better yields

Using prilled MnO2, a fine grade of Al powder and granite mortar finely ground technical lime the following formulation was tested:

MnO2 __________ 1.05 mole
Al _____________ 4/3 mole
CaO ___________ 1/3 mole

Total batch weight was 153 g.

Initially I wanted to test CaO = 2/3 mole but looking at the volume of ground CaO, I thought it might be too much and cut it back by half. The thermite was lit with Al/KClO3 stoich. mix and a piece of Mg ribbon.

Shell type assembly prior to ignition:

This burnt furiously, with a high flame. A couple of minutes after burn out:

The thermite had expelled quite a bit of material from the crucible, including a lot of fine drops of Mn metal. But the rest of the material seemed to settle quietly at the bottom of the crucible, with a crust of slag at the top.

Breaking into the solidified slag puddle:

Up: clearly porous slag, almost like swiss cheese. A clear sign of volatilisation

Middle: a green area, probably rich in MnO (partly reduced MnO2), small in volume

At the fore: three pieces of manganese metal, one about 1 cm.

Total about 8.4 g or a measly 18 % yield. There's probably a bit more that could have been recovered with more effort.

Considering how furious this burn was it will be interesting to go back to the originally intended formulation of 2/3 moles of CaO. But I’m out of Al powder right now...

[Edited on 13-7-2013 by blogfast25]

Pok - 12-7-2013 at 12:54

I don't think that a good reaction is even possible with MnO2. As I said earlier, Mn3O4 is by far the best and an established solution. I sometimes had Mn reguli > 5cm in size. Only due to fracture at very fast cooling smaller pieces resulted.

citing the (german) attachment:
"Manganese thermite: MnO2 (Braunstein) reacts to heavily and has to be converted to the lower oxide Mn3O4. For this purpose heat 80 g MnO2 for 1 hour at 800 - 900 °C. Mix the product with 20 g Al coarse powder."

Most people can't reach 800-900 °C for 1 hour. So they can buy the needed Mn3O4 here

Attachment: Thermite - Boron Chromium Manganese Silicon.pdf (204kB)
This file has been downloaded 1057 times

[Edited on 12-7-2013 by Pok]

blogfast25 - 12-7-2013 at 13:29

Hey pok!

What do you mean by 'an established solution'?

Where you write that a good reaction is not possible I really don't know what you mean. If anything the reaction with MnO5 is TOO good, in the sense that it generates too much heat and and then starts kicking stuff out of its pram.

And that size reguli ('> 5 cm'): from what size thermites?

I will certainly try Mn3O4 but in terms reaction enthalpy between Mn3O4 and MnO2 there really isn’t much difference. Also, changing the oxide doesn’t change the fundamental problem: that the BP of Mn metal and the MP of alumina are very close together. I’ve already experimented with blends of MnO2 and MnO and found little was to be gained from it.

The trick is burn rate, I think. And Mn3O4 is helpful there. Also coarser Al powder and CaF2 and or CaO as heat sinks.

The text you uploaded is interesting but hardly authorative: it’s a text for schools. Their suggestion of lining the crucible with CaF2 for instance is nonsense. Nobody does that anymore because it doesn’t do anything.

[Edited on 12-7-2013 by blogfast25]

Pok - 12-7-2013 at 16:59

I called it an established process due to concrete mention of MnO2 and Mn3O4 in the literature. Yes, school texts aren't that reliable. But I have idendical information here:

S. Chand (2005) Advanced Inorganic Chemistry, Vol.2, p. 316-317.

There is a simple reason why MnO2 is not good despite similar reaction enthalpy: it releases oxygen at very low temperatures > 800 °C ("low" in the scale of thermite temperatures). This is as if you put water to the mixture. At 2000 °C both substances behave the same: they escape explosively from the mixture. Of course, most of the oxygen is catched by the aluminium but even if a very small amount ist NOT catched, this amount can lead to an eruption of the mixture thus leading to a low yield and small Mn particles.

I think there is no possibility for a moderate reaction with MnO2 at all. Because if you have a small size Al powder, most of the erupting oxygen will be catched. But this will lead to a fast reaction (eruption, low yield, small Mn). If you have a large size Al powder, you will lower the reaction speed. But not all of the oxygen from the MnO2 decomposition will be catched, because the oxygen can develop "far away" from the next Al particle (also eruption, low yield, small Mn).

The 5 cm size reguli were obtained from larger batches. I don't remember the mass of the batch but one large regulus contained about 70-80% of the total yield (not of the theoretical yield, but of the practical yield in my cases).

S.C. Wack - 12-7-2013 at 20:02

Quote: Originally posted by blogfast25

The trick is burn rate, I think.

As soon as the reaction has started add the remainder of the charge from an iron spoon, not too much at one time, but still rapidly enough so that the mass in the crucible is kept in a state of brilliant incandescence until the reaction is ended. It is advisable for the operator to wear blue glasses and a heavy glove

http://www.mediafire.com/view/4l5z29t8749ate9/Laboratory_Met...

blogfast25 - 13-7-2013 at 04:26

S.C.: C. K. Gupta in one of his many books on extractive metallurgy describes an industrial process for aluminothermic manganese that is very similar. But that has GOT to be tricky to execute!

I’m nonetheless completely miffed at the choice of Mn3O4. From a very reputable source, here are the respective Standard Heats of Formation for MnO2, Mn3O4 and Mn2O3: - 850 kJ/mole; - 1384 kJ/mole and – 1386 kJ/mole.

(Source *.pdf: http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s...)

For the respective (stoichiometric) reduction reactions with aluminium the respective Standard Heats of Reaction then are (MnO2, Mn3O4 and Mn2O3): - 597 kJ/mole; - 850 kJ/mole and – 290 kJ/mole (all per mole of oxide).

That would make Mn3O4 the worst (as in 'the hottest'), followed by MnO2. Mn2O3 would be ideal to reduce generated heat as much as possible. Mn3O4 may be the lower oxide in terms of Mn oxidation state but it still contains the most oxygen per mole of oxide.

The Kremer pigment that pok linked to is unlikely to be Mn3O4: look up ‘Caledonian Brown’ (which is what it’s advertised as).

Edit: Here’s a *.pdf by Gupta:

http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s...

From p. 407 (391 from table of content):

<i>”Using manganese dioxide gives a
great excess of heat and entails the risk of the reduction reaction being explosive. Even
Mn3O4, when used as the starting material, gives excess heat while the heat is insufficient
if manganese monoxide (MnO) is used. A mixture of MnO and Mn3O4, however, can give
sufficient heat for reduction; thus the process can be made self-sustaining by choosing a
value of the manganese to aluminum ratio between 1.5 and 1.125. These calculations and
estimations have been carried out for an adiabatic reactor. In actual reactors there would be
heat losses and the appropriate value of the ratio would be closer to 1.125.
In practice, the reduction is conducted in an open magnesite crucible. The reaction is
started by igniting a small amount of a mixture of manganese dioxide (MnO2) and granular
aluminum. To this incandescent charge Mn3O4, aluminum granules and powdered lime
are added in calculated quantities to sustain the controlled reaction. The addition of charge
is continued at regular intervals until the crucible is filled with the reacted mass consisting
of a lime-alumina slag and manganese metal.”</i>

[Edited on 13-7-2013 by blogfast25]

12AX7 - 13-7-2013 at 13:48

*Cough*, Mn3O4 is the hottest by mole because its moles are bigger. Try dividing those by the mixture's molar density (mole/cc, let's say).

Tim

Pok - 13-7-2013 at 14:29

Quote: Originally posted by blogfast25

The Kremer pigment that pok linked to is unlikely to be Mn3O4

I know that is Mn3O4. The MSDS says: "Chemical Composition: Manganese oxide (Mn3O4) - CAS No.: 317-35-7". I made Manganese with it in 10 % excess mixture with coarse Al powder. No CaO or CaF2 or anything. And it always worked fine.

[Edited on 13-7-2013 by Pok]

blogfast25 - 14-7-2013 at 04:03

Quote: Originally posted by 12AX7

Try dividing those by the mixture's molar density (mole/cc, let's say).

Tim

The only number that allows a theoretical comparison on an ‘all other things being equal’ basis is the adiabatic end temperature. I’ve done more than enough of those to know by just looking at the numbers that for haussmannite it will exceed 3000 K.

@pok:

Did you by any chance test the Kremer product for iron content, qualitatively or quantitatively?

Pok - 14-7-2013 at 06:15

I didn't test it for it's iron content. I think only a quantitative test would make sense, because manganese oxides always contain at least a tiny amount of iron. I can send you a sample if you want to check it's iron content, if you want.

(I don't have the time) It seems to be from a natural source (ore) from Morocco. So there might be a certain contamination with iron.

blogfast25 - 14-7-2013 at 11:43

Thanks, pok, I might take you up on that. But I bought a manganese oxide on eBay years ago that was advertised as 'high grade MnO2'but it looks suspiciously light in colour, a bit like the Kremer product. I'll see if I can determine the MW of that one.

[Edited on 14-7-2013 by blogfast25]

blogfast25 - 20-7-2013 at 12:36

While waiting for some materials to arrive, two more simple experiments were carried out.

Aluminium household foil was shredded using a kitchen blender and the finest fraction of this shredded Al foil was sieved off using a metal mesh sieve of about 1 x 1 mm mesh size. This material substituted the very fine (400 mesh) Al powder in the formulation above.

This mixture couldn’t be ignited. Clearly the Al was simply too coarse at this scale of things.

Another batch of the same amount but with the 400 mesh Al powder was then prepared and the two batches then mixed. In essence this the same formulation as above but with 50 % Al 400 mesh and 50 % of the very coarse, shredded, 1 mm material.

This lit up very well and burned much more contained. But when breaking into the slag puddle it appeared the results were similar to the 100 % 400 mesh Al powder of above: very porous slag puddle with poor yield of about 25 %, due to a few small but nice reguli of Mn metal.

[Edited on 20-7-2013 by blogfast25]

metalresearcher - 16-8-2013 at 23:20

I tried it as well and it went very fast.

<iframe sandbox width="560" height="315" src="//www.youtube.com/embed/53RJXNFi8N4?list=UUJArDPDoXa4a1vzvNqvbu1A" frameborder="0" allowfullscreen></iframe>

metalresearcher - 17-8-2013 at 01:45

Another attempt: I moderated it with powdered ACC (Autoclaved Cellular Concrete), the same material as this block is made of, 1/2 parts by volume to the Al/MnO2 mixture (stoichiometrically 13 parts MnO2 + 5.4 parts Al).
I driilled a hole 25mm diameter and 50mm deep in the block.
I put a lid onto it so when it would erupt, it keeps inside. But the Mg ribbon failed to ignite, so I ignited it again without lid and it was indeed not needed.

<iframe sandbox width="640" height="360" src="//www.youtube.com/embed/rg9gEAuuxnQ" frameborder="0" allowfullscreen></iframe>

This is what I got after cooling. Frozen blobs which I shattered into this:

[Edited on 2013-8-17 by metalresearcher]

blogfast25 - 17-8-2013 at 04:32

Metalresearcher:

The second video says it's 'private' (can you make it public?)

The first one confirms what we already know: a stoichiometric mixture of fine MnO₂ and fine Al powder without any heat sinks essentially behaves like a flashpowder.

It would have been interesting to see the influence of that filler.

[Edited on 17-8-2013 by blogfast25]

metalresearcher - 17-8-2013 at 04:50

Now I have made it public.

blogfast25 - 17-8-2013 at 09:18

Interesting result. You used 0.5 volume part of 'ytong powder' per 1 volume part of thermite mixture?

It tamed the reaction quite a lot.

Bear in mind that Al can also reduce SiO2, so contamination of the Mn metal with Si is a possibility.

I would repeat the experiment on a larger scale and perhaps weighing the amounts of ytong and thermite. See what quality metal you get...

[Edited on 17-8-2013 by blogfast25]

Fantasma4500 - 17-8-2013 at 10:36

be very careful scaling up MnO2, as it will somehow selfconfine (assuming some advanced physics comes into play here) and downright explode, yes with more power than CuO + Al

for CaO + Al, its calcined
ive heard about CaO + Al from a book about the elements, as the mentioned route to Ca metal, but later was told that you need to calcine it together
on a sidenote.. i think ive found a quite neat thermite ignition mix..
its KNO3 and silicone metal powder, if its packed in al foil with a proper amount per cm, then it burns very well, more easily ignited stuff like BP can be added to make it ignite straight away

perhaps you could get higher metallic yield by putting the thermite in a container of some sort, and putting heat resistant obstacles on top, a guy i talk to told me that it started to rain with iron and aluminium oxide when he did his thermite (200g + -)
so something tells me that alot of the metal boils off, at these temperatures

blogfast25 - 17-8-2013 at 10:49

Quote: Originally posted by Antiswat

be very careful scaling up MnO2, as it will somehow selfconfine (assuming some advanced physics comes into play here) and downright explode, yes with more power than CuO + Al

[snip]

perhaps you could get higher metallic yield by putting the thermite in a container of some sort, and putting heat resistant obstacles on top, a guy i talk to told me that it started to rain with iron and aluminium oxide when he did his thermite (200g + -)
so something tells me that alot of the metal boils off, at these temperatures

Stop spreading misinformation!

Scaling up does not make this thermite more or less 'explosive'. An explosion can only occur when the energy is contained in a shell where pressure can build up until the shell 'breaks' and all energy is then released at once. Ironically this is what you suggest in a round about way a bit further down! :mad:

Manganese thermite has a tendency to boil off its manganese metal, as explained in this thread. Have the courtesy to read it before commenting.

Your comment is a typical case of ill-digested rumours and half truths found on the Tinkerwebs, once again rehashed, this time by you. Very unscientific to say the least!

Don't conflate 'deflagration' with 'explosion'.

[Edited on 17-8-2013 by blogfast25]

Fantasma4500 - 28-8-2013 at 10:12

im sorry for have forgotten to link to where i found it, anyways here it is..

http://www.youtube.com/watch?v=lVdZBbIjdo0

''I did a very extensive research about thermites. Copper thermite is more powerful if you have small amounts of mixture (about 5 grams). In this video, I had about 300 grams of mixtures, and in this circumstances, manganese thermite is much more powerful.''

i cannot say the person is viable, i dont know the person, could be a 4 year old who found a random video, or an expert in this, who knows.

dont like the word explode myself, but i feel it is more powerful than just deflagrate, but yeah, perhaps deflagrate does fit whats seen in the video.. most likely not high explosive

im trying the hardest to be the furthest from 'k3wl', sorry..

metalresearcher - 13-9-2013 at 12:44

Here I did a test using MnO2 with magnesium powder.
It did very well and the resulting black substance dissolved in (hot) 10% hydrochloric acid.

Did anybody else test this ?

With Mg + SiO2 I did get silicon powder.

<iframe sandbox width="640" height="360" src="//www.youtube.com/embed/3Dr0rtIQDis" frameborder="0" allowfullscreen></iframe>

elementcollector1 - 13-9-2013 at 12:54

Quote: Originally posted by metalresearcher

Here I did a test using MnO2 with magnesium powder.
It did very well and the resulting black substance dissolved in (hot) 10% hydrochloric acid.

Did anybody else test this ?

With Mg + SiO2 I did get silicon powder.

If I had to guess, the manganese sublimed and recooled on the upper brick, giving a solid lump of metal. Very good!

What concerns me is the color of your HCl after solution: If I didn't know better, I would have assumed that metal was iron from the way it dissolved. Manganese salts are pink...

[<a href="u2u.php?action=send&username=bfesser">bfesser</a>: removed quoted embedded video]

[Edited on 14.9.13 by bfesser]

12AX7 - 13-9-2013 at 13:12

Mn(II) is extremely pale in solution. Manganese compounds are often very impure, iron being the most prominent impurity. Fe(II) could easily overcome the color of Mn(II), even with relatively pure (~95%?) MnO2.

Tim

bfesser - 13-9-2013 at 16:20

<strong>metalresearcher</strong>, I like this video much more than your recent uploads on Si thermites. The details in the beginning and the wet chemistry at the end are helpful. It would be nice to see more analysis of the yellow solution, though.