Sciencemadness Discussion Board

Action of acids on titanium oxides

hkparker - 11-4-2011 at 18:16

I recently made silicon by reducing sand, however, I've been told I will be stuck with an impurity of titanium oxides. I made a thermite of sand aluminum and sulfur, hydrolisized the aluminum sulfide and then dissolved away the aluminum hydroxide with HCl. If I now have a mix of silicon and titanium oxides my question is what is the most effective way to purity it? Heating (refluxing?) with HCl? Or will nitric acid be more effecting while not reacting with the silicon. Thanks!

A video of what I've done.
http://www.youtube.com/watch?v=B1eO3PCNvDI

plante1999 - 12-4-2011 at 05:03

best way to disolve this oxide:

TiO2:HF , NaHSO4
Ti2O3:HF , HCl
Ti3O5:HF, HCl
Ti4O7;none , this is used as an anode in HF media and this is very corrosion resistant , can be reducted with Mg.

the best i think is to re made the thermite with glassware powder that have lets sitted in HCl or reagent grade SiO2.

magority sand have a lot of TiO2 in.

blogfast25 - 12-4-2011 at 05:17

Quote: Originally posted by hkparker  
I recently made silicon by reducing sand, however, I've been told I will be stuck with an impurity of titanium oxides. I made a thermite of sand aluminum and sulfur, hydrolisized the aluminum sulfide and then dissolved away the aluminum hydroxide with HCl. If I now have a mix of silicon and titanium oxides my question is what is the most effective way to purity it? Heating (refluxing?) with HCl? Or will nitric acid be more effecting while not reacting with the silicon. Thanks!




Firstly, unless your mix contained titanium dioxide to begin with you cannot find any titanium in the reaction products. A mix of TiO2 and sand, reduced the way you’re doing, would probably yield an alloy of Ti and Si.

Secondly, your mix burns way too slowly for any decent yield of silicon to be had. For the metal(loid) silicon to separate out your reaction product mix (Al2O3, Al2S3 and Si metal) that mix has to be completely molten at the end of the reaction. The silicon will then coalesce out of the mix into more or less globular crystalline chunks.

You can increase the speed of the burn by means of either finer ingredients or more additional Al and sulphur. All your reaction products need to collect at the bottom of your crucible in a neat molten puddle. For that you need to reach at least the MP of alumina (just over 2,500 C)!

Also, what you call ‘aluminum hydroxide’ doesn’t really come into it: it is a alumina/aluminium sulphide mix that’s formed here. The Al2S3 dissolves readily in HCl, generating lots of H2S (Al2S3 +6 HCl --- > 2AlCl3 + 2 H2S). This dissolution causes the alumina/aluminium sulphide mix to break up, thereby setting your silicon metal free. But the alumina really doesn’t dissolve in the HCl because it’s annealed Al2O3. In your condition, most of the silicon is present as fines, powder even. Pretty much useless and hard to separate from the alumina.

About that H2S: it is very, very toxic (a case of mild intoxication has been reported on this board by ‘fluke’, look it up) and you’re generating lots of it. It would be much better to capture most of it with a bleach scrubber: lead your off gases through commercial (thin!) bleach, which oxidises the H2S to elemental sulphur. Bar that, make sure you don’t send that toxic cloud in the direction of a house or a person.

I bet the smell of rotten eggs still follows you around today: it gets into your clothes, into the silicon and can be perceived even at very low concentrations… trust me, I’ve done this many times but quit because of the smell.

blogfast25 - 12-4-2011 at 05:31

Quote: Originally posted by plante1999  

magority sand have a lot of TiO2 in.


plante, I'm sorry but that is sheer nonsense. Most sand simply doesn't contain ANY TiO2, never mind 'a lot'. What you wrote is highly misleading. If the 'majority' of sands contained 'a lot' of TiO2' we'd be extracting it from there like mad!

[Edited on 12-4-2011 by blogfast25]

plante1999 - 12-4-2011 at 07:54

blogfast25; take not that if you check composition of white sand you will find that 5-20% is TiO2 , the rest is magoritary SiO2 also if the termite dont have flux , TiO2 is only reducted to Ti(X)O(X2-1) and finish with ebonex (Ti4O7) some speci have been found like Ti5O9 (i will sand u a document about hydrogen , aluminium and carbon reduction of TiO2 to get the reducted oxide.

ScienceSquirrel - 12-4-2011 at 08:09

There are rutile sands that do contain a lot of titanium dioxide but they are the exception rather than the rule and they are mined as ores.
The majority of sands are mainly silica.

blogfast25 - 12-4-2011 at 08:23

Damn, SS, you beat me to it!

Plante, if you’re referring to rutile beach sand, then yes, but that is positively rare. Most sands, including white beach sands don’t contain any rutile at all. There are some other exotic ‘sands’ too but again you won’t find them at the majority of beaches.

Regards the ‘partially’ reduced Ti oxides, it’s a story I’ve heard many times before and I consider it mostly a fairy tale. At best these are very exotic mixed valence oxides, at worst a mixture of Ti and TiO2/Ti2O3. I’ve reduced TiO2 many times before, only to obtain the metal, each and every time, never any mixed oxide. Thermodynamics don’t favour them: they favour the reduction all the way down to Ti(0), for highest ΔG.

Even Ebonex is likely to be something more like a cermat: a composite of metal and oxide in this case, not an oxide in its own right and in the commonly understood sense of the word. The fact that it’s electrically conductive points strongly in that direction.

In very special non-stoichiometric non-equilibrium conditions maybe such species can be created. But presenting them here as if poor htparker is going to have to deal with them is folly.

hkparker - 12-4-2011 at 14:12

@blogfast25
Quote:

Secondly, your mix burns way too slowly for any decent yield of silicon to be had. For the metal(loid) silicon to separate out your reaction product mix (Al2O3, Al2S3 and Si metal) that mix has to be completely molten at the end of the reaction. The silicon will then coalesce out of the mix into more or less globular crystalline chunks.


It was pretty darn molten, I'm not sure how well it picked up on camera but I think it was molten enough, since I got some bigger pieces out of it.
Quote:

You can increase the speed of the burn by means of either finer ingredients.


Looks like more sulfur and aluminum. My reactants were dusty fine, probably just not mixed well enough.
Quote:

Also, what you call ‘aluminum hydroxide’ doesn’t really come into it: it is a alumina/aluminium sulphide mix that’s formed here. The Al2S3 dissolves readily in HCl, generating lots of H2S (Al2S3 +6 HCl --- > 2AlCl3 + 2 H2S). This dissolution causes the alumina/aluminium sulphide mix to break up, thereby setting your silicon metal free. But the alumina really doesn’t dissolve in the HCl because it’s annealed Al2O3. In your condition, most of the silicon is present as fines, powder even. Pretty much useless and hard to separate from the alumina.


Help me out here. Aluminum sulfide hydrolisizes readily, so it should be destroyed in water (well, almost all of it). It produces a mix of aluminum hydroxide and oxide. I know the hydroxide will dissolve in HCl, but why won't the oxide?
Quote:

About that H2S: it is very, very toxic (a case of mild intoxication has been reported on this board by ‘fluke’, look it up) and you’re generating lots of it. It would be much better to capture most of it with a bleach scrubber: lead your off gases through commercial (thin!) bleach, which oxidises the H2S to elemental sulphur. Bar that, make sure you don’t send that toxic cloud in the direction of a house or a person.

I bet the smell of rotten eggs still follows you around today: it gets into your clothes, into the silicon and can be perceived even at very low concentrations… trust me, I’ve done this many times but quit because of the smell.


Yea, I would have wanted to capture it but didn't have the bleach/glassware requires without lots of work. While doing it my house smelled a little, the next day it was only my lab, now its only the silicon. I read fluke's post before doing this, scary stuff.


In the future I would likely use cab-o-sil as a source of silicon dioxide. Its very fine and to my understanding quite pure.

@plante Ground up glassware would contain B2O3

bob800 - 12-4-2011 at 15:02

Quote: Originally posted by hkparker  
Yea, I would have wanted to capture it but didn't have the bleach/glassware requires without lots of work.


Even if you don't get ill, your neighbors may notice the pungent fumes and, (unlikely but possible), call the authorities if suspicious. Even if that wasn't the case, they could mistake the H<sub>2</sub>S odor as a gas leak and call the fire department, as has happened before.

hkparker - 12-4-2011 at 15:57

Yea I've read the thread. Fortunately they didn't notice, and are well aware of what I do and fully ok with it (I really am luck!)

blogfast25 - 13-4-2011 at 04:51

Your alumina doesn’t dissolve in HCl because of the conditions in which it was obtained (melting -- > freezing). Calcined oxides are generally far more resistant to acid attack than freshly prepared (precipitated) ones. For the industrial production of Al the ore Bauxite is mainly used, which is chiefly alumina, contaminated with Fe2O3 (and some silica). To purify it, the ore is finely crushed, then subjected to strong NaOH in autoclaves (to be able to heat beyond the atmospherical BP of the solution), the alumina then enters solution as aluminate over the course of several hours (the Fe2O3 doesn’t dissolve at all).

One form of natural, crystalline alumina is red corundum, known as Ruby: would you expect such gems to dissolve in HCl? ;)

Yes to Cab-O-Sil and equivalents! Much finer and purer than ball milled beach sand. Easier too.

So did you get any significant yield of silicon metal than?

hkparker - 13-4-2011 at 14:09

Ok, thanks for the help! Ill run this with cab-o-sil in the future and do a follow up.

As for this product, I took some flakes from the dried filter, added anhydrous NaOH and poured on a little water (my reason for isolating silicon was to try this pretty exciting reaction) in a test tube. It bubbled, and got really hot, but not close to what I've seen with pure silicon. Some of this I think is due to it not being a fine powder (I read the silicon has to be very fine), so I will grind it up and try again. There is definitely a good amount of silicon in there but still pretty crude.