Titanium (III) Potassium Alum: failed attempt

Alums need little introduction here, I think. Essentially double salts with the generic formula NM(SO4)2.12H2O with N = NH4, Na, K, Rb or Cs and M = trivalent Al, Fe, Tl, Cr and some other exotic metals.

I’ve often wondered if M = Ti³⁺ would be possible. Searching, admittedly not very deep, only yielded one Google book reference (<a href="http://books.google.co.uk/books?id=PpTi_JAx7PgC&pg=PA684&lpg=PA684&dq=titanium+(III)+sulphate+alum" target="_blank">link</a> but it’s quite a general book and perhaps not very authorative on the subject.

So I decided to have a go myself.

About 2.0 g (0.025 mole) of TiO2 was dissolved in an excess of 96 % H2SO4 and the solution (of TiOSO4) diluted slightly to about 40 ml (so about 0.6 M TiOSO4). A small amount of undissolved TiO2 was eliminated by pipetting off the supernatant liquid. To this solution was then added 2.2 g of K2SO4 which was dissolving by vigorous stirring. Slight cloudiness developed on heating (to help dissolve the K2SO4), pointing to hydrolysis of the TiOSO4 and was immediately stopped. Then about 1 g of ultra-fine, very pure zinc powder was added to effectuate the reduction of the TiO²⁺ ion:

TiO²⁺(aq) + ½ Zn(s) + 4 H₃O⁺(aq) → Ti³⁺(aq) + ½ Zn²⁺(aq) + 6 H₂O(l)

The solution turned a very deep purple, characteristic of Ti³⁺ cations, almost immediately and hydrogen evolved.

At that point theoretically the solution contained an estimated over 30 g of KTi(SO4)2.12H2O per 100 g of solution and was carefully double clingfilmed to at least provide a first barrier to air oxygen. But on cooling and overnight ice bath no crystals materialised.

It is possible that the solubility limit of the alum has not been exceeded. It may be difficult however to make solutions of TiOSO4 that are more concentrated without running into hydrolysis problems.

Maybe I should make another attempt with ammonium sulphate? Or maybe this alum just cannot exist…

<!-- bfesser_edit_tag -->[<a href="u2u.php?action=send&username=bfesser">bfesser</a>: fixed external link(s)]

[Edited on 13.10.13 by bfesser]

[Edited on 13-10-2013 by blogfast25]

I really think water can be a big problem in trying to form Ti3+ as Ti really wants to go to TiO2, ie. Ti(IV), the only way to prevent this is to make sure there is enough excess H2SO4 to sink the water. For example, the reaction:

(i) 3TiO(2+) + 6H+ + Ti(s) <=> 4Ti3+ + 3H2O

produces 3 moles of water, if this not sunk by a second reaction then a parallel reaction that can occur is:

(ii) Ti(s) + H2O + 2H+ => TiO(2+) + H2(g)

Note reaction (i) produces water and reaction (ii) requires it, so an effective strategy here is to sink the water by ionisation to H3O+

So you need a further 3 moles of H2SO4 to fully ionise the water produced by reaction (i), the balance equation now becomes:

3TiO(HSO4)2 + Ti(s) + 9H2SO4 => 4Ti(HSO4)3 + 3HSO4(-) + 3H3O+

This way you really force the issue and force the oxidising titanium to be oxidised by TiO(2+) and not H2O!

NOTE: I used HSO4(-) as the speciation for sulfate, not SO4(2-) as it's only the pKa1 of H2SO4 that is really extreme enough to guarantee things go to completion.

Furthermore, if you are using 95% H2SO4, then you also need to account for the 5% water there as this would reduce the amount of H2SO4 species available is probably far less than 95%, i.e. part of it is HSO4- and H3O+.

So if you start with TiO2(s), you need a wopping 11 H2SO4 + whatever excess to correct for the water already in 95% H2SO4, and then to once dissolved, to this you add your titanium strip.

Finally, I think the issue of sluggish kinetics of titanium metal reaction will be overcome iff TiO(2+) is in high concentration making the solution oxidising AND H2O is trace (ok if it's there as H3O+) else TiO2 can form which hampers the kinetics. Besides with such concentrated H2SO4, making the solution really hot is possible should the need arise that should take care of kinetic issues, though this would be also be significantly more hazardous!

Sorry for the long discussion, just found the topic interesting

[Edited on 14-10-2013 by deltaH]

Today I tried to dissolve some 'pyro grade' titanium powder (about 60 mesh, what the pyros like to call 'titanium sponge') in 50 w% H2SO4.

2.4 g of Ti powder (0.05 mole), 15 g of water and 15 g of 96 % H2SO4 (twice the stoichiometric amount) were combined in a 100 ml RBF. Reaction started surprisingly quickly and was very vigorous, much more than I expected. The solution became purplish/blue very quickly.

Unfortunately it became clear that also some TiO2 was being formed. Presumably H2SO4 is too oxidising at 50 %.

After about 10 minutes near BP, I added some 5 g of K2SO4, simmered for some more and then allowed to cool. Quite quickly the mass solidified into a two phase system, with a clear, colourless top layer (K2SO4? KHSO4?) and a purplish, opaque mushy mass:



Some of the bottom layer was dissolved in 37 % HCl and it was clear that there was undissolved white matter, presumably TiO2, in that layer.

[Edited on 17-10-2013 by blogfast25]

Quote: Originally posted by deltaH

Sounds very promising indeed and loved the colours! A small cooler box is a handy place to put a flask if you want it to cool down very slowly (over a day)... would help the crystallisation!

2.4 g (0.05 mole) of the same Ti powder and 25 ml of 37 w% HCl (about twice the stoichiometric) was loaded into a 100 ml RBF and refluxed with gentle heating on a hot plate. A primitive gazometer served as HCl scrubber and indicator of hydrogen evolution (right large test tube):



Reaction was very swift, taking approx. 30 min for the gas evolution to more or less die down. For most of the time 5 or 6 large bubbles of hydrogen could seen floating to the top at the same time.

Somewhat before hydrogen evolution had ceased, 0.025 mole of K2SO4 was added and the solution simmered for a few more minutes. When gas evolution had all but ceased heating was stopped and the solution would have been approx. 2 M in Ti³⁺ but not enough sulphate was present for KTi(SO4)2.12H2O stoichiometry. So after a bit more cooling about 5 ml of 96 w% H2SO4 was added very slowly to make up for the deficiency in sulphate molarity.

Unfortunately this caused the solution which was hitherto clear (but very dark) to lighten in colour and to cloud over strongly. I suspect again oxidation of the Ti³⁺ by the sulphuric acid, to a hydrolysed Ti(IV) species. After cooling this was obtained:



I’ll know for sure tomorrow if the precipitate is TiO2 but it sure looks like it. Hopes of salvaging anything from this attempt are slim.

Another attempt tomorrow.

Quote: Originally posted by deltaH

If the addition of the acid did indeed oxidise the Ti3+ species, then it must have also produced H2 upon this addition, which it seems it didn't.

Interesting, I did not know that, I must admit that Ti in any state other than IV is completely foreign to me, which is why I find this so fascinating

But your thread has made me realise that I prepared a Ti(III) complex accidentally once. I was condensing a quinone with bis(trimethylsilyl)carbodiimide to form the quinone-dicyanoimide and this was done by first reacting the quinone with TiCl4, then adding the imide, so that the titanium would pull the oxygen from the quinone to form some solid white Ti(OCl)x mess that precipitates. Anyhow, somehow when washing glassware, I noticed these blue ink-like stains all over the sink from those white solids the next day.

Hang on... maybe what you have is that same titanium oxychloride mess.

[Edited on 18-10-2013 by deltaH]

Well, well, looks like deltaH could be right.

This material does not appear to be TiO2 (IV).

It was filtered off on Buchner and washed profusely with cold water and didn’t lose any colour. It’s a sandy, non-gelatinous, light blue powder. Here it is just after sucking dry, the filtrate of Ti(III) can also be seen:



Here’s another photo of it, after having recovered it from the filter:



Some tests were carried out on the material. In four test tubes about a quarter of a teaspoon of material was loaded. Then various additions were made:



1) Strong ammonia solution: precipitates black Ti(OH)3. After a while gas started to evolve.
2) Strong NaOH solution: precipitates black Ti(OH)3
3) Some drops of 37 % HCl, some water and a few drops of 35 % H2O: oxidation of Ti(III) to Ti(IV) and formation of red Ti(IV) peroxo complex
4) Pure 37 % HCl: no change, no dissolution

All this strongly points to the material being a Ti(III) compound, insoluble or very poorly soluble.

A pinch of it was put on Al foil on a max setting hot plate. The material darkened somewhat but no substantial change was observed.

Then about 2 g of the material was treated with 10 ml of 33 % NH3 and the Ti(OH)3 filtered off. To a blank (virgin NH3 solution) and a few ml of the clear and colourless filtrate, a ml of Ba(NO3)2 solution (2 g in about 10 ml of water) was added:



Left the blank, right the filtrate: the latter tests very strongly positive for sulphate ions. This suggests there’s plenty bound sulphate in the material. Remember that the precipitate had been washed carefully with DIW.

I now assume that the precipitate obtained in the post above was in all likelihood identical to this material. That would suggest it will not contain any potassium.

Some further tests will now be carried out to test for chloride and potassium. If a sulphate based insoluble Ti(III) compound exists, the preparation of a Ti(III) alum may not be possible or very difficult to do.



[Edited on 19-10-2013 by blogfast25]

Quote: Originally posted by deltaH

I suggest you repeat the test with 6g Ba(NO3) to be on the safe side. [Edited on 19-10-2013 by deltaH]

Quote: Originally posted by woelen

Maybe you have a similar effect with your titanium salt. It may be anhydrous.

Quote: Originally posted by bismuthate

I did see that however that mixture also contained HCl which may have helped in the reaction. Maybe an oxidiser like a chlorate would react in a dry mixture.

Even though I had taken out a little insurance for the sulphate/chloride test by holding back a bit of the Ba(NO3)2 solution and then testing the filtrate with it (it tested negative), the test for chlorides was repeated as follows.

1.0 g of product was treated with 10 ml of 33 % NH3 and the Ti(OH)3 filtered off. The clear filtrate was acidified with glacial acetic acid, diluted to about 100 ml and heated up to near BP. To this was then added a 50 ml solution of 5 g Ba(NO3)2 in water, also hot. The hot slurry was then filtered and allowed to cool.

To it was added a strong solution of about 5 g AgNO3 in about 20 ml of water and a strong precipitation occurred:



The quantity of precipitate is in line with yesterday’s observation and is difficult to explain other than by bound chloride in the product.

The remaining material (about 7 g) is now being dried carefully at low heat till constant weight, for use in more quantitative tests.

Quote: Originally posted by bismuthate

since it does contain chloride I wonder if it will react with H2SO4. Also you still haven't tested it with a dry reducing agent/ product mixture. You may want to test its solubility in different solvents.

Testing the product with oxidisers (that is what you meant, I suppose?) is useless: we know what happens.

Re. solvents, I’m more interested in volatile, non-reactive anti-solvents, for drying purposes. Acetone and glacial acetic acid seem to fit the bill so far.

Thanks deltaH for your comments.

***************

Another batch of the product was attempted, this time leaving out the K2SO4. No precipitate was obtained. So 4.4 g of K2SO4 was then added anyway, and some precipitate formed but much less than in the initial run.

Sensing that perhaps the potassium wasn’t important but the sulphate in the K2SO4 was, another run was made, this time with no K2SO4 but 10 ml (instead of 5 ml) of 96 w% H2SO4 added when the titanium dissolution was all but over. No precipitate formed whatsoever, indicating again the K2SO4 plays some part in its formation, but after a bit of cooling something interesting happened: over the course of about 5 minutes a deep purple crystalline product slowly separated out, eventually filling most of the volume:



The photo doesn’t do the colour justice, my camera is colour blind (possibly due to TL lighting in my lab): it’s more like a lighter shade of chromium alum and very different from the sandy light blue precipitate.

Also interesting is the fact that the supernatant liquid is completely colourless, indicating it contains no Ti(III) (but Ti(IV) cannot be excluded).

I have not tested the crystalline matter yet but I’m hoping it is water soluble. With there being more than enough sulphate present, this could even be a simple Ti2(SO4)3 hydrate. That would raise hopes of a Ti(III) alum again. But isolating and drying this material may prove challenging. Paper filtration media won't do here...

[Edited on 21-10-2013 by blogfast25]

Quote: Originally posted by Poppy

Have you tried simply evaporating the mixture? I had a table showing double salts compatibility but its not on this computer, also, it may be the case that the proportion of the reagents is exceeding/ requesting some molar ratio, thus failure is prominent. Have you nuts to wait a month or so and make some evaporites??

Quote: Originally posted by bfesser

Evaporating under a vacuum is trivial. Jam in a one-hole stopper with a tube connected to vacuum, and swirl gently in a warm water bath.

I repeated the experiment with 50 % H2SO4 and late addition of K2SO4 from above but with (NH4)2SO4 instead. So, 2.4 g of Ti powder, 15 g of water, 15 g of 96 % H2SO4 and addition of 3.3 g of (NH4)2SO4 when gas evolution started to wane.
After cooling to RT it looked like this:



The light blue precipitate is again intrinsically blue (not just coloured by the deep blue supernatant liquid) and filters to a filter cake of this colour:



Interestingly, assuming this precipitate is the same as the one reported above, this was obtained in the absence of chloride anions.

Another test will now concentrate on trying lower H2SO4 concentrations.

Quote: Originally posted by deltaH

Oh my... just looked up ionic radii of these metals on wiki, turn out Ti(III) is 81 pm versus Al(III)'s 67.5, so taking K/Al alum as the base case, looks like you need something slightly smaller, not bigger!!!! I'd say go for sodium at 116pm compared to potassium's 152pm

Indeed, well if CsTi(III) is known, then you can't exactly argue with that, now can you

Quote: Originally posted by blogfast25

[...]Of Al the NH4, Na and K alums are known (not sure about the Rb and Cs ones) [...]

Quote: Originally posted by woelen

[ Cs-alums tend to be much less soluble than their K-counterparts. I would not want to use the rare CsOH for this experiment, if you have CsCl or CsNO3, then use that in combination with a concentrated solution of Ti2(SO4)3, without any added K2SO4 or (NH4)2SO4. [Edited on 27-10-13 by woelen]

Quote: Originally posted by deltaH

Out of madscience interest, can one prepare alums from things like choline and a metal? The reason that I ask is because I happen to have quiet a bit of choline left over from my choline soap experiments? [Edited on 27-10-2013 by deltaH]

Quote: Originally posted by woelen

Smart-elements sells Cs at EUR 79 per 100 gram at ultra high purity.

Thanks, woelen.

Well, well: standing overnight in an ice bath did yield a crop of crystals, from yesterday's 33 % batch:



Again, the colour is false, it’s really purple.

Although this could be NH4Ti(SO4)2.12H2O, it could also be plain (NH4)2SO4 doped with a bit of Ti2(SO4)3. The quantity (volume, visual estimate) is decidedly more than the volume of (NH4)2SO4 I added though (but it’s subjective). The crystals are small but well formed and I’ll try and take a look under my (crappy) microscope.

The only firm decider here is to determine the Ti content.

[Edited on 28-10-2013 by blogfast25]

Quote: Originally posted by deltaH

Are you still going to try the cesium version?

It's low on my list. Ti assay of the last product, even higher concentrations of Ti2(SO4)3 and the KTi alum (assuming the NH4Ti is a real alum) are more interesting than confirming what we already know (that the CsTi alum exists).

I can certainly confirm that the suspected NH4Ti alum is completely soluble in water, to a clear solution. And the colour seems far too intense for just Ti³⁺ doped (NH4)2SO4. Just a few more sleeps before I know the definitive answer...

The results of the Ti assay of the suspected NH4Ti alum are in.

The material was first washed with ice cold 10 % H2SO4, then gently patted dry with filter paper and stored in a CaCl2 desiccator. The material came out much lighter in colour, so that made me fear partial oxidation to Ti(IV) had taken place. 6.4 g of dried material was obtained.

But making the stock solution of about 0.1 M concentration in 1 M HCl showed all of it dissolved without insoluble residue. It did not dissolve very fast though and some gentle heating was applied to speed things up. Here’s the stock solution:



This solution was the titrated against 0.1 M ferric ammonium alum:

Ti3+(aq) + Fe3+(aq) === > Ti4+(aq) + Fe2+(aq), using KSCN as an indicator for excess Fe3+.

I found the titanium content of the sample to be 13.2 w%. The theoretical value for NH4Ti(SO4)2.12H2O is 10.1 w%. So a tad too high for comfort.

A couple more qualitative tests were carried out too. The dried material doesn’t appear to dissolve/react with acetone or surgical ethanol.

When mixed with some powdered NaOH and heated in a test tube, the smell of ammonia is unmistakable, pointing to ammonium ions being part of the material. This could also be the basis of a quantitative test: dry distilling off the NH3 into a known amount of standardised HCl, followed by titration of the remaining HCl, to determine ammonium content.

I’ve also prepared another batch of Ti2(SO4)3 solution, this time of even higher concentration, this time aiming at the potassium alum again.

Quote: Originally posted by soniccd123

There is no more progress on this? I'm planing to try some of what is described at this topic with Zirconium in next months (in summer vacations actually, right now i'm at med school so i really don't have time).

Quote: Originally posted by blogfast25

Quote: Originally posted by soniccd123   There is no more progress on this? I'm planing to try some of what is described at this topic with Zirconium in next months (in summer vacations actually, right now i'm at med school so i really don't have time). What do you have in mind?

Quote: Originally posted by soniccd123

Well, I know that Zirconium (III) ions exists and supose that they may have similar properties to the Titanium (III) ions. I have some objetives in mind: - Produce a Zirconium (III) solution and understand how to handle and maybe stabilize it; - Crystalize the Zirconium (III) salt from it; - Crystalize a Zirconium alum from it. I know that may be very hard, even impossible with the equipment i have at home, but i'm really willing to try it. To do this, i suspect that some of the techniques, procedures and informations provided by this thread are going to be really helpful.

Quote: Originally posted by soniccd123

Yeah, you guys are right. I found Zirconium (III) Chloride and other halides names at the CRC Handbook of Chemistry and Physics, then, just for sake of curiosity, i've got to search some articles about it; I found some and had them saved in my computer but didn't read till now. It really exists, but is extremely sensible to moisture as woelen said. My bad people, I should had researched a bit more before posting.

Quote: Originally posted by blogfast25

Quote: Originally posted by soniccd123   Yeah, you guys are right. I found Zirconium (III) Chloride and other halides names at the CRC Handbook of Chemistry and Physics, then, just for sake of curiosity, i've got to search some articles about it; I found some and had them saved in my computer but didn't read till now. It really exists, but is extremely sensible to moisture as woelen said. My bad people, I should had researched a bit more before posting. That's not to say that a bit of Zr chemistry wouldn't be interesting for this forum. I have a few threads on it.

Quote: Originally posted by soniccd123

Have you ever been able to crystalize out pure Ti2(SO4)3?

No. I'm not sure it's possible. Ti³⁺ requires very low pH to prevent hydrolysis or oxidation of water. I have made Ti₂(SO₄₃ solutions of about 30 w% though... with high acid reserve. Maybe with vacuum evaporation these solution would have yielded crystals, who knows?

[Edited on 26-9-2015 by blogfast25]