Sciencemadness Discussion Board - Extracting zirconyl chloride from roll on antiperspirant

Extracting zirconyl chloride from roll on antiperspirant

blogfast25 - 8-4-2010 at 11:56

After two half-hearted and failed attempts at alkaline cracking of pottery grade Zircon (ZiSiO4) (better attempts to be continued later), I came across an article that explains the use of Aluminium-Zirconium chloride glycine (Gly) complexes in roll on antiperspirants. With the goal of trying to extract some small amounts of zirconyl chloride (ZrOCl2 - zirconium oxychloride) from it, I bought 200 ml of 'Mitchum 48 protection - women' (which includes such a complex in its formula) roll on antiperspirant.

Here's what I did.

I first diluted the stuff, a sticky, thickish liquid, opaque in appearance, to about 400 ml. The pH was just under 3 at that point. I added small aliquots of NaOH, the pH went up to about 4 and lumps started to form. At about pH = 6 the liquid had turned into a solid mass, unstirrable. By gradually working more alkali into the mass, the thickness reduced greatly at about pH = 11. Presumably amphoterics like alumina had gone back into solution.

The liquid proved unfiltrable at least without a Buchner. I diluted the by now about 600 ml to about 3 L, maintaining pH at about 11. Still filtering was basically impossible but the solids did decant quite easily.

I let them decant completely and siphoned off the supernatant liquid and replaced it with more alkalised tap water (pH = 11), stirring and allowing to decant once again. I then siphoned off the supernatant liquid again. This should have gotten rid of most of the aluminium as aluminate.

Adjusting the pH of the remaining, decanted slurry to pH = 1 with 20 % HCl, some temperature was developed and the slurry thinned out much noticeably. Sadly, much of the slurry did not dissolve and that part must be considered the water insoluble part of the complex lipophobic/lipophylic emulsion these products appear mostly to be made off.

Some attempts at 'breaking' the emulsion have failed so far: adding more acid, adding kerosene and strong alcohol all have no effects on the unfiltrable emulsion that contains also the zirconyl chloride in the watery phase.

Does anybody here have any ideas on how to 'curdle' or 'cream' this fatty material emulsion and free up the zirconyl chloride containing watery phase?

[Edited on 8-4-2010 by blogfast25]

zed - 12-4-2010 at 16:58

Ummm. Sounds messy, but it might be simple. I'd be tempted to try sticking a wick in it. In other words, I might attempt to burn off all of the "organic" components.

Thereafter, it might be possible to extract Zirconium compounds from the residual ash.

blogfast25 - 13-4-2010 at 06:28

Thanks Zed, but I doubt if that would work: there too much of an aqueous phase present. It'd be a bit like trying to combust milk...

Tonight I'm going to put some of it in a blender to see if I can get the fatty matter to really clump together.

Sedit - 13-4-2010 at 09:40

Quote: Originally posted by blogfast25

Thanks Zed, but I doubt if that would work: there too much of an aqueous phase present. It'd be a bit like trying to combust milk...

Then dry it up and torch it.

I think if you are working with this it would be the first order of business to destroy all organics since they in no way are helping your cause. Iv done plenty of OTC extractions of products since figuring out pathways like this are what im good at.

I don't suppose you have a list of the ingrediants do you because I went looking and can't find a complete list anyware. It appears that burning off all the organics FIRST would yeild you a much more pleasent starting material to work with by removing any emulsifing agents and more then likely cutting the bulk of material nearly in half or more.

Quote: Originally posted by blogfast25

Tonight I'm going to put some of it in a blender to see if I can get the fatty matter to really clump together.

Odds are your just going to stir it up into a frothed emulsion resembling shaving cream and your never going to get it out then. Can't say for sure though because im not 100% on what all the material is.

blogfast25 - 13-4-2010 at 12:34

Drying the product and burning off the organics is definitely one possibility, albeit more of a last resort, IMHO. Definitely not to be forgotten. One would end up with very impure zirconia which of course could be purified to required degree.

But at this stage 'emulsion breaking' (e.g. the 'creaming' of milk to obtain butter) would be a neater way to separate the watery, zirconyl chloride containing phase from the fatty matter.

I do have a list of ingredients and will post it here:

water, Aluminum Zirconium Tetrachlorohydrex Gly, Glyceryl Stearate, Laureth-23, Magnesium Aluminum Silicate (clay), Polysorbate-20, Behentrimonium Metosulphate, Laureth-4, Cetearyl Alcohol, Lauric Acid, EDTA, Parfum, Alpha Isomethyl Ionone, Linalool, Citronellol, Hexyl Cinnamal, Butylphenyl Methylproprional, Geraniol.

Sedit - 13-4-2010 at 13:57

LOL well... Theres always a Dildo....

They work very well for breaking emulsions perhaps sonication if you are able could produce a simular effect. It might make it worse though so maybe test a small amount if your able.

I still will not change my mind that the first thing you should do is take a brand new container and burn off all the organics on the grill or something. You see that almost everything is organic so all you will be left with is Clay and your good which would cut back your amount of work(and cost) extremely.

I don't know what the product of heating Aluminum Zirconium tetrachlorohydrex Gly would be but its something to take into consideration before processing the ash that would be left over. More then likely would result in some oxide would be my guess.

Ozone - 13-4-2010 at 14:38

Now that's a bit of unorthodox apparatus.

How about partitioning against saturated NaCl(aq) or Na2SO4(aq)?

O3

not_important - 13-4-2010 at 22:33

Hmmm...

Starting with the original product, oxidise it using chlorine bleach with a Ni salt catalyst. This would chew up the glycerol of the 1st two ingredients, reducing the total bulk, and convert the PEO end groups and long chain alcohols to carboxylic acids and thus to their salts. I don't think OCl/Ni will chop PEO chains, but if it does all the better. Then do the further basification, disperse the solids in aqueous alcohol (rubbing alcohol) and take the pH to around 3 which should be acid enough to give the free carboxylic acids while leaving the Zr hydroxide as such. Dilute with acetone to get most of the fatty acids and other lipophiles into the alcohol-acetone phase, leaving Zr and Al in the wet solid phase.

Alternatively treat the starting material with concentrated hydrochloric acid to get the metals as aqueous chloride complexes, and extract off the more hydrophobic stuff with solvents. Go for solvents with as much density difference from water as possible.

Yet another alternative would be to get it slightly alkaline to get all the Ze as hydroxide/hydrated oxide, evaporate off s much water as possible, disperse in alcohol + some toluene/xylene and azeotrope off the remaining water. Hopefully much of the organics would go into the alcohol phase, the solids would be disperseed in fresh alcohol, heated, and a less polar solvent such as acetone added to pull off most remaining organics.

A possible problem with burning off the organics would be the formation of Al-Zr oxides that would be rather unreactive to acids or alkali, taking you back to the problem of starting with ZrO2 or ZrSiO4. ZrOCl2 itself will decompose somewhere between below the boilinh point of water up to 150 C, giving oxychlorides with lower Cl percentages on up to hydrated ZrO2.

blogfast25 - 14-4-2010 at 07:22

@Ozone:

Osmosis, eh? Definitely hadn't thought of that. Suitable OTC membrane?

@not-important:

Certainly I agree that it would have been wiser to pre-treat the product before inducing hydrolysis of the Al/Zr/Cl Glycine complex. Either by mechanically 'breaking' the emulsion or by chemical treatment.

What about trying to oxidise much of the organics with permanganate or dichromate? Stand well back, I'd say...

Crude burning off of the organics destroys the ZrOCl2, leaving one with a residue of zirconia, alumina (possible aluminates) and some clay. Messy too...

[Edited on 14-4-2010 by blogfast25]

blogfast25 - 17-4-2010 at 13:23

Mechanically breaking ('creaming') the emulsion I simply couldn't get to work.

I then tried chemically breaking it with K2Cr2O7 and separately with commercial bleach.

To 50 ml of the emulsion, 25 ml of saturated dichromate (acidified) were added. Upon boiling no distinct visual changes were noted but the fatty matter did drop out of the solution very quickly. I may test the supernatant liquid for the presence of Cr [+III] later.

Then 50 ml with 25 ml of commercial bleach (no nickel salt, haven't got any at the ready) were boiled up. Not much to see either but here too the fatty matter dropped out of the emulsion quickly.

Presumably these oxidisers knock out part of the sophisticated emulsifier system.

The whole of the remaining emulsion has now been 'bleached' and is decanting overnight.

not_important - 17-4-2010 at 18:44

Glyceryl Stearate is the mono-ester of glycerol with technical stearic acid - a mix of saturated fatty acids. Laureth-# are polyethylene-oxide esters of fatty acids. Polysorbates are polyethylene oxide ethers of sorbatol derivatives, and cetearyl alcohol is long chain primary alcohol made from fatty acids. All are neutral (no acidic or basic groups) surface active agents. Oxidising them converts terminal HO-CH2 groups into CO2H, in alkaline solution CO2(-) groups which are not as easy to pull into nonpolar layers; with acid dichromate yu often get chromium complexes of the organic acids which can be a real bother to "de-chromate". The glycerol may be oxidised fully leaving simple fatty acids, to the same effect.

UnintentionalChaos - 17-4-2010 at 19:13

Am I the only one who's thinking that the zirconium is going to form gelatinous precipitates and refractory oxides if heated, just like the aluminum?

not_important - 17-4-2010 at 19:36

It does, but the texts on extracting Zr from zircon or crude zircronia tell how to get granular type precipitates, and it only goes refractory if heated considerably hotter that wet processes use. The hydrated oxides formed can be dissolved in mineral acids, halides and nitric preferred as the sulfates have low solubility.

densest - 17-4-2010 at 19:44

Since you've tried bleach, perhaps something like stronger like TCCA applied carefully might do it, or Ca(OCl)2 as a solid since the desired product is already chlorinated. Messy and likely to spatter - I'd add the TCCA in 1g quantities in a tall vessel with stirring outdoors. According to Merck articles, Zr acetate and oxychloride are quite soluble. The acetate is less stable in dilute solutions and more stable if heated past about 65C. So adding conc. acetic acid, stirring well, heating, and decanting should produce a usable product. Zr sulfate is also soluble. Centrifuging might break the emulsion as well.

Many zirconium salts form microcrystalline/gelatinous/gummy solids with polymeric structure (also according to Merck) but concentrated solutions are reasonably stable.

watson.fawkes - 18-4-2010 at 07:18

Quote: Originally posted by densest

Centrifuging might break the emulsion as well.

I was about to mention this myself. It also seems promising that some chemical additive that's incapable of breaking the emulsion on its own might do so with a bit of added artificial gravity.

blogfast25 - 18-4-2010 at 07:51

Centrifuging would really have been the ticket but for that you a, erm... centrifuge! But you still need to break the emulsion first, bar perhaps with an ultracentrifuge.

I'm relying on good old 9.8 m/s^2. The 'fatty matter' has neatly collected at the bottom, although the supernatant liquid is still slightly turbid and will need filtering...

Diluted from fairly dilute solution Zr(OH)4.n H2O should be fairly gelatinous but less so than freshly precipitated alumina, at a guess. The aluminium had been previously separated out.

densest - 18-4-2010 at 08:37

@blogfast - yes, the old G works well given time.

I got a small clinical centrifuge for under $100 - it has swinging tube holders for 6 15-20ml tubes. One maker sells them under a lot of brand names. My guess is 500xG? Daigger has similar ones for $250-400 so used ones are not expensive.

I was guessing that a stronger oxidizer might destroy more of the organic emulsifiers since the ones you've tried have made a dent in them.

According to Merck, ZrO dissolves in 66% H2SO4 with heating. The sulfate is moderately soluble in water: 52.5g/100ml @18C. As long as the oxide doesn't get calcined to utter unreactivity, drying, removing organics with solvents or fire, and redissolving should work. The oxide does seem to be more reactive than, say, Al2O3. I've seen mention of alkali metal zirconates forming in basic conditions and have no idea of their solubility or reactivity.

blogfast25 - 18-4-2010 at 11:55

The solubility of these oxides really depends a lot on the degree of calcining: e.g. commercially calcined TiO2 only rally dissolves in conc. hot H2SO4 (and that takes some time too), fused with excess bisulphate and probably in anhydrous HF. I wouldn't really want to take on calcined ZrO2 or natural ZrO2.

In the mean time I've had the first glimpse of Zr(OH)4. n H2O by alkalising about 20 mL of filtered sample to pH = 13.5 (thus keeping any residual Al in solution). The sample was about 10 x diluted with respect to original 200 mL roll on antiperspirant product. A small amount of almost opaque zirconium hydroxide formed and it precipitated neatly to the bottom of the test tube very quickly.

[Edited on 18-4-2010 by blogfast25]

blogfast25 - 21-4-2010 at 07:48

The first dilute solution of ZrOCl2 is now in a bottle awaiting crystallisation. All Zr(OH)4 has been precipitated and most will be filtered and washed tonight. I noticed that a more compact precipitate was obtained at higher pH (13.5) than at lower (11.5).

The current ZrOCl2 was obtained by 'squeezing' the remaining acid 'fatty matter' residue till fairly dry. A couple of hundred mL was obtained and taken to pH = 13.5, yielding about 50 mL Zr(OH)4 precipitate. This was filtered (it filters reasonably well but a Buchner would be great), once washed with pH = 10 NaOH, then 5 times with tap water (the filtrate then ran neutral) and once with distilled water.

The Zr(OH)4 didn't dissolve easily in cold 6 M (20 w%) HCl: I used a 100 mL first, then added another 100 mL but dissolution was incomplete still. I then heated it up, stirring, and at about 90 C the solution was completely clear.

Setting it aside for cooling I then noticed something strange: almost immediately a small amount (definitely less than a gram, at a guess) of precipitate formed at the bottom: that was definitely not there previously. It looked amorphous. Further cooling didn't increase the amount.

Could this be some reaction product of the 'fatty matter' and the bleach that somehow slipped through the net? A soap-type product that hydrolised in acid conditions and solidified on cooling? A halogenated fatty acid perhaps? The small amount will be isolated tonight.

zed - 26-4-2010 at 01:59

Seems like a lot of trouble to get a little Zirconium. Might be easier to buy a little piece of cubic zirconia.....Smash it to powder with a hammer, and then dissolve it in acid, or use some other process to convert it to the salt you desire.

Recently, I have been informed that some ceramic cooking knives, have blades that are composed of Zirconium Oxide.

Ephoton - 26-4-2010 at 02:18

some times its just for the fun of doing it

the buety (dislexic man never get that one) of amatuer science :cool:

blogfast25 - 26-4-2010 at 06:03

Quote: Originally posted by zed

Seems like a lot of trouble to get a little Zirconium. Might be easier to buy a little piece of cubic zirconia.....Smash it to powder with a hammer, and then dissolve it in acid, or use some other process to convert it to the salt you desire.

Recently, I have been informed that some ceramic cooking knives, have blades that are composed of Zirconium Oxide.

Well, good luck trying to dissolve annealed Zirconia in anything other than molten alkali or anhydrous HF! Zircon (ZrSiO4) can be 'cracked' by fusing with NaOH but my limited experience is that that isn't so easy either.

@ ephoton: spot on!

The extraction is now almost complete, after I hit a particular snag. Report to follow.

[Edited on 26-4-2010 by blogfast25]

not_important - 26-4-2010 at 21:01

Following up on blogfast25's reply, not only is it difficult to get ZrO2 to dissolve in something, but cubic zirconia is about the most expensive way to get it. Both ZrO2 and ZrSiO4 are some in technical grades for use in metal casting and ceramics for much less than gemstone quality zirconia.

Fairly extreme conditions are needed to dissolve those compounds, fusion at 600 C or higher with NaOH or Na2CO3 is used for the production of hydrated ZrO2, the starting point for many zirconium compounds; hours of fusion is followed by chilling, milling of the solids, and fusion once again. Fusion with potassium pyrosulfate has been used for analytical purposes.

There's a thesis on the subject at http://upetd.up.ac.za/thesis/available/etd-03062006-161327/ and an overview attached

Attachment: 30123025-Zirconium-and-Zirconium-Compounds.pdf (313kB)
This file has been downloaded 1416 times

blogfast25 - 27-4-2010 at 06:27

@not_important:

Is the second link to the wonderful monography by Venables on Zirconium and compounds (or is it a different document?) I'd warmly recommend to download Venables' book for anyone with an interest in Zr.

The first link is highly interesting and I will now download the various parts and sieve for information, so thanks also for that.

I have a *.pdf on a lab study titled:

High Purity Zirconia and Silica Powders via Wet Process: Alkali Fusion of Zircon Sand

By Chieko Yamagata, João B. Andrade, Valter Ussui, Nelson.B. Limaand José O. A. Paschoal

Presented at the Sixth International Latin-American Conference on Powder Technology, November 07-10, Búzios, Rio de Janeiro, Brazil

It's fairly detailed and the crucial part reads:

"Aqueous caustic soda (NaOH) 50 wt% solution was prepared to using as the decomposition agent of zircon sand. The prepared solution and Brazilian zircon sand were mixed in wt/wt ratio of 1.0; 1.2 and 1.5 NaOH/ZrSiO4. The moisture was carefully charged into an iron crucible and set in an electric furnace. The alkaline fusion reaction was conducted following thermal treatment: first heating at rate 20oC/min up to 100oC, remained at this temperature for 1h and then heating at rate 20oC/min up to 200oC, remained for 1h and finally the temperature was increased to 600oC for 2h, at that time the alkali fusion reaction takes place and caustic frit is formed, a mixture of sodium zirconate and sodium silicate. This frit was washed with water, removing the soluble sodium silicate and hydrolyzing the zirconate to an impure hydrous zirconium oxide. The water-leached frit was treated in cold 8M HCl at 80oC for obtaining an impure zirconyl chloride solution."

Yields of up to 87 % zirconia (+ hafnia) recovery are reported using a high NaOH/zircon (1.5:1) ratio.

A second run on a new batch of 200 mL of roll on antiperspirant ("Mitchum") has now started, this time by attacking the virgin emulsion straight away with commercial bleach to try and free up the watery phase immediately.

not_important - 27-4-2010 at 07:06

The bleach should destroy the glycine used in the complex, and attack some of the oxidisable stuff too. I'd recommend finishing it by adding some nickle, even if that means dissolving a copper-nickle alloy coin in acid. Use lots of bleach, you can evaporate off water, cool things down, and add fresh bleach.

If that seems to give a reasonably behaved mess, then consider evaporating it to near dryness, mushiness, and adding concentrated hydrochloric acid to make it strongly acidic. That should pull the Zr and Al into solution as chloride complexes, and allow the extraction of some of the organics such as the fatty acids into a non-polar solvent. After that the strong alkali treatment you used before to remove the Al into solution.

PDF is from Kirk-Othmer Encyclopedia of Chemical Technology. And I knew of that paper you quoted from too, but forgot to mention as the controlled temperature processing makes it more fussy than just tossing a zircon-soda ash mix into the kiln.

[Edited on 27-4-2010 by not_important]

blogfast25 - 27-4-2010 at 08:20

@not_important:

Oh, that second *.pdf is great too. All this info should keep me sweet for days!

I don't have a single nickel compound at hand right now, what is it supposed to do? Our coins don't contain any, IIRW. But boiling the 200 mL of emulsion with 200 mL commercial bleach for 10 minutes caused clear separation with the hydrophobic phase sitting neatly on top of the clear (well, very slightly turbid) watery phase. Separating completely will be easy and done tonight.

But I'm weary to reduce the watery phase down to almost dryness at this point. Will this not Diels-Alder some of the remaining fragile organics into a carbonised mess? I'm not entirely sure but I seem to have had this problem at the first attempt. One way to get rid of carbonised matter is boiling with strong alkali of course.

Right now I'm more tempted to thoroughly filter the watery phase, dilute it and eliminate the aluminium at pH = 10 - 11 with strong NaOH. Any remaining fatty acids should saponify and dissolve.

A second treatment with strong bleach is certainly tempting.

I might just test both routes separately on small quantities...

[Edited on 27-4-2010 by blogfast25]

not_important - 27-4-2010 at 17:28

Nickle salts added to sodium hypochlorite form 'nickle peroxide' which functions as a catalyst for oxidising a number of substrates, not all of which behave as well or react as completely as without the nickle catalyst.

I'd not worry about Diels-Alder, if for no other reason than ingredients list generally go from major to minor components, those complex organics are fragrances and only present in small quantities. Diels-Alder isn't noted for carbonising, if you mean dehydration A) most of the H-O-C stuff should have been oxidised by the hypochlorite, and B) we're talking evaporating as you would on a steam bath, and only until the stuff gets mushy from the NaCl crystallising out. Bleach is dilute enough that you want to get rid of a goodly percent of the water, which also decomposes most of the remaining hypochlorite; the result after adding con HCl should salt out the fatty acids and some of the other now oxidised surfactants in the mixture while holding Al and Zr tied up in MCl4<sup>-</sup> type complexes in the aqueous phase.

blogfast25 - 28-4-2010 at 12:33

I mentioned Diels-Alder in relation to a problem that occurred with the first batch: after dissolving the crude Zr(OH)4 in hot 6 M HCl and boiling off the liquid, I noticed the solution turned noticeably brown near the end. Without absolute evidence I believe this may be due to carbonisation of organics picked up from the paper filter, upon contact with strong HCl. I have some reason to believe the presence of ZrOCl2 actually aggravates the problem. Unfortunately the brown stuff doesn't precipitate but it does co-precipitate when re-precipitating the Zr(OH)4 with alkali. I thus have a batch of Zr(OH)4 (from the first attempt) that is slightly tinged with a beige colour (pure Zr(OH)4 is snow-white) and may be in need of the bleach treatment.

To avoid this, all filtering of acid solutions is now done after partial neutralisation to pH ≈ 4.

On the watery phase of the second batch, after filtering (it's still very slightly turbid), two treatments have been applied:

Treatment A: eliminating aluminium first.

100 mL of the crude, filtered solution was slightly diluted and alkalised with 5 M NaOH to pH = 11. Upon standing something fairly typical of Zr happened. After a few hours the clear supernatant started to slightly cloud over. Taking pH again it was near neutral! Presumably the cloudiness was due to hydrolising aluminate. Due to polymerisation Zr water soluble compounds tend to show some 'sponginess' before settling for a final pH. Here I adjusted the pH to about 13 - 14 and left to stand overnight. It was hoped that most fatty acids would saponify into water soluble soaps.

The next day the crude Zr(OH)4 was amply washed with dilute NaOH (pH about 10) and ample amounts of hot tap water until the filtrate ran almost neutral.

The precipitate was then dissolved in 100 mL of hot 6 M HCl but the solution remained very turbid, presumably because of organics. 100 mL of bleach was added and the solution turned a chlorinish light green-yellow! This solution was then boiled down to about 100 mL and until no chlorine was detected by smell. It was then re-diluted. The colour remains and some whitish precipitate collected at the bottom. This will now be part-alkalised to pH ≈ 4 and filtered. The filtrate will be diluted and alkalised with 9 % ammonia to about pH ≈ 7 for re-precipitation of the Zr(OH)4.

Treatment B: without eliminating aluminium first

To 50 mL of the crude, filtered solution was added 50 mL of bleach and a small dash of 6 M HCl. This solution was boiled down to almost dry and then 50 mL of 6 M HCl was added: this caused precipitation as well as the appearance of the chlorinish light green-yellow colour. It could not a priori be excluded that much of the precipitate might be ZrOCl2 because the latter is poorly soluble in conc. HCl (a preferred method of re-crystallisation, apparently). So another 50 mL 6 M HCl and water up to about the 200 mL mark was added and the solution brought briefly to the boil again: much of the precipitate, but not all, seemed to redissolve. This will now be part-alkalised to pH ≈ 4 and filtered. The filtrate, which still contains the aluminium, will be diluted and alkalised with strong NaOH to about pH ≈ 13 - 14 for re-precipitation of the Zr(OH)4 and elimination of the aluminium as aluminate.

blogfast25 - 4-5-2010 at 12:38

With some delays about half of the ZrOCl2.8H2O has now been crystallised. A first estimate shows that 100 mL of the roll on antiperspirant contains about 15 g of the zirconylchloride octahydrate. This is only a rough estimate because the yield varied somewhat from batch to batch, Treatment B for instance giving higher yield than Treatment A (but B is easier with only one precipitation). I'm not sure what those differences are due to but poor definition of the hydrate may be a cause.

The crystallites were all obtained by dissolving the purified zirconia hydrate (Zr(OH)4.nH2O) in an appropriate amount of 6 M HCl and boiling till almost dry. When the first onset of crystallisation is observed, near dryness, heat is then switched off and on cooling the compound then crystallises out. But not all of it may be the octahydrate. Hygroscopicness seems also to vary. I may need to cut the heat slightly earlier.

All crystallites are coloured yellow (to varying degrees) and KSCN shows the culprit is ferri(III)chloride, picked up from the commercial bleach (also shown by KSCN to contain Fe3+). This may explain also the varying degrees of hygroscopicness.

When all the ZrOCl2 has been collected, I will try and recrystallise the whole lot to try and eliminate the FeCl3. ZrOCl2 can be recrystallised from strong solutions by adding conc. hydrochloric acid. Alternatively, from strongly acid solutions the ziroconyl chloride can also be precipitated by adding water (acc. F. Venables).

Unfortunately the sources I have on ZrOCl2 say little about solubility or precise methods of recrystallisation. So a little trial and error will be required.

blogfast25 - 18-5-2010 at 13:06

Although I've still not completely finished with this method of extracting a water soluble Zr source for an OTC material like a roll on antiperspirant (aerosol antiperspirants contain no Zr, by law, not sure why), I've certainly learned a lot about it...

Firstly, no matter what method it became clear I was always doing something wrong:

1. product yields varied wildly despite very similar treatments,
2. product didn't really behave like ZrOCl2.8H2O is described: there was hydrolysis immediately upon contact with water but product did dissolve easily when small amounts of HCl were added,
3. each time I tested for Al, some was found.

I began to believe that the product was severely contaminated with Al (as well as with Fe, from the bleach, see bottom of post).

In total about 65 g of variable 'product' had been obtained from 400 mL of antiperspirant. I decided to try another procedure and treated 100 mL of virgin antiperspirant with 100 mL of commercial bleach and some 50 mL of 6 M HCl. On boiling, the phases separated completely and the hydrophobic phase sat on top. Decanting and filtering I obtained the clear watery phase (about 200 mL) containing the aluminium zirconium chlorohydrex Gly.

At this point, a word needs to be said about these aluminium zirconium chlorohydrex Gly complexes (AZGly). Here's the datasheet of a typical AZGly, in this case Zirconal ® AP4G (pdf):

http://www.bk-giulini-pcg.com/dyndata/AP-AP4G.pdf

It's chemical formula is Al4Zr(OH)12Cl4Glyx.nH2O

Note the atomic ratio of Al/Zr = 4 in this case. A bit more research showed that ratio can vary from 2 to 10 for various AZ complexes (which may or may not contain Glycine, depending on brand)!

It's structure is reminiscent of the real structure of ZrOCl2.8H2O, which does not contrary to popular belief, contain the cation ZrO2+ but far more likely the polymeric cation [Zr4(OH)8(H2O)16]8+.

Going back to the treatment, the 200 mL of watery phase were then mixed with 100 mL of water in which about 15 g of solid NaOH had been dissolved. This was stirred strongly and heated in an oven at 150 DC for 2 hours, then filtered and the filtrate checked for Al by neutralising: there was a lot of precipitate, at least as much as the precipitate on the filter (which I assumed to be Zr(OH)4.nH2O). The precipitate was first amply washed with boiling 1 M NaOH, then with ample hot tap water, then re-dissolved in 200 mL of 6 M HCl, then re-precipitated from a more dilute solution and at much lower pH (about 5.5), to avoid occlusion of the NaCl in the zirconia hydrate.

Redissolving again in 6 M HCl and boiling down to about 100 mL, I left the solution to cool and a white precipitate formed: most of the liquid had in fact turned into a slushy semi-solid. Taking a teaspoon full of sample and adding water, it dissolved effortlessly into a perfectly clear solution, without any hydrolysis. This looked more like the real thing! I then checked the dissolved sample for Al by adding quite some NaOH, boiling, filtering and neutralising the filtrate: it still contained Al! Comparing the precipitate on the filter and the precipitated alumina hydrate in the filtrate, I'd say the 'real thing' contains still at least 10 -30 % Al!

This is a little puzzling: is it possible that the large amount of NaOH wasn't sufficient to convert all the Al into Al(OH)4 (-)? Was the amount of water too small to dissolve all the aluminate? How then to explain it wasn't washed out with the hot 1 M NaOH?

Another explanation is that when treated with NaOH the AZGly does not split easily into aluminate and zirconium hydrate. I found a reference that states that to split an AZGly with strong acid (e.g. for titrometric determination of Al or Zr with EDTA) prolonged boiling with acid is necessary. Could this be true also of the complex in alkaline conditions? Is it possible that the complex tends to lose some Al, thereby being transformed in another complex, this one with a lower Al/Zr ratio? Either way, it would appear so far that using 'my' method, the 'alkali fusion' may need to be repeated until all Al has been removed...

Another problem is the contamination with Fe3+, which has proved hard to remove. I hoped firstly that recrystallisation from slightly acid solution would concentrate the FeCl3 in the mother liquor but the solubilities of ZrOCl2.8H2O and FeCl3 are both quite high. Recrystallisation by adding conc. HCl (37 %) should be possible but my 20 % doesn't cut it.

Separating out the Fe (after reduction to [+II]) as FeS, Fe(OH)2 or FeCO3 in conditions where Zr doesn't precipitate (pH < 4) isn't possible. Complexation to ferrocyanate (Fe(CN)6 4-) or ferricyanate (Fe(CN)6 3-) would probably do the trick but I've no cyanides. The ferri thiocyanate (FeSCN2+) complex isn't stable enough I think.

So, I'm a bit stuck on that.

This afternoon I carried out the first proper attempt at fusing zircon (ZrSiO4) with NaOH according to the Brazilian paper's wet method (see above). The fusate is now soaking in some water overnight, to loosen it up. Hopefully the fusion will have been successful. Fingers crossed...

not_important - 18-5-2010 at 21:18

Part of the purpose of using hypochlorite is to destroy the glycine, removing issues of the complex changing solubility patterns.

see http://www.biochemj.org/bj/030/0484/0300484.pdf

I suspect that your NaOH concentration may be too low to fully dissolve all the aluminium, the concentration more than the total amount of NaOH. Also stirring while heating would be better. On the other hand 2 hours seems excessive, I suspect you would do better with a shorter heating and then do a second extraction with NaOH solution. You should be aware that you may be forming extremely fine precipitate that is getting through your filter paper.

I see three ways of cleaning up the iron

1) to a nearly neutral solution of the crude product, add some pieces of zinc. Iron should plate out on the zinc, allowing both to be filtered off. Zn(OH)2 will dissolve in NaOH, like Al(OH)3, and even in strong aqueous ammonia, so it can be leached out of the Zr upon precipitation (or reprecipitation).

2) Both Al and Fe can be kept in solution as the pH is raised by the presence of acetate. This is done by bringing the pH just to the point that precipitate starts to form, adding dilute HCl to dissolve that, and adding acetic acid and/or ammonium acetate, then raising the pH in a controlled fashion. NH3 or Na2CO3 are used in preference to stronger bases. You'll find this in older books on inorganic analysis, it's the sort of thing you'd want to do after earlier purification.

3) After precipitating the Zr(OH)4 and washing it, stir it into some quite dilute HCl with the hope of leaching out the Fe while dissolving very little Zr. Or use acetic acid + ammonium acetate, if you can get reasonable strong acetic acid.

Another trick used to remove iron was extraction with ether from HCl solutions, doesn't seem like the best alternative.

blogfast25 - 19-5-2010 at 06:04

Quote: Originally posted by not_important

You should be aware that you may be forming extremely fine precipitate that is getting through your filter paper.

That's possible but I tested the filtrate for amfoterics by toggling the pH from high to neutral to low and back. Zirconia hydrate does not re-dissolve at high pH. So any precipitate formed would be in all likelihood hydrated alumina.

As regards removing the Fe, 1) wouldn't work because zirconyl chloride starts precipitating (hydrolising) at about pH = 5. If a neutral solution could be obtained, I'd have reduced the Fe to +II and precipitated it as Fe(OH)2 (Ks = 4.87 E-17, at pH = 7, [OH-]^2 ≈ E-14).

But I like 2) and 3) which come down to leaching out the Fe3+ and the lowest possible pH without re-dissolving the zirconia hydrate.

Alternatively, avoid Fe altogether, perhaps by using a homemade bleach? Or another strong oxidiser? K2Cr2O7 didn't work and KMnO4 would have the drawback that the Mn2+ wouldn't be easy to get rid of either...

not_important - 19-5-2010 at 06:26

Consider the H+ ion concentration - there's not too much at pH 5, you'd want to get the pH to the point cloudiness just starts to appear, then re-acidify a tiny bit. You'll lose some of the zinc to acid, but the iron still should plate out.

Sorry, H2O2 is the best alternative oxidiser, then maybe hot dilute HNO3. Making your own hypochlorite is a bit of a bother, you could do it right in a mix of the starting goop and aq NaOH and bubble in Cl2, but you may be fighting the Al-Zr hydroxide ppt.

The amount of iron in the bleach has to be small, as the bleach is A) oxidising and B) alkaline, meaning iron should be Fe(III) and drop out as the low solubility hydroxide-hydrated oxide, which as a rather low Ksp.

The acetate trick should work as it is/was used analytically to separate Zr from Al and Fe.

blogfast25 - 20-5-2010 at 09:22

I will attempt to dissolve the contaminated zirconyl chloride into a pH = 5 acetate buffer and plate out the iron with zinc.

blogfast25 - 17-6-2010 at 07:08

Applying what had been learned from the previous experience with extracting zirconyl chloride from roll on antiperspirant (ROAP) I applied the following procedure to get a better idea of zirconyl chloride content:

1. 100 mL ROAP ('Mitchum 48 h - women'), 90 mL of thin commercial bleach and 10 mL of 32 % HCl were mixed and simmered for about 10 min. The phases separated and the watery phase was filtered off.

2. 20 g of NaOH, dissolved in 100 mL of water were slowly added with intensive stirring to the filtrate and simmered under reflux for 1 hour.

3. The precipitate (very crude ziconia hydrate) was filtered and washed with 500 mL of cold 5 M NaOH, then with copious amounts of boiling water until pH of the wash water was about 11.

4. The washed precipitate was dissolved in 70 mL of 32 % HCl, with heating to boiling point. The solution was reduced by part refluxed boiling to about one third of the original 200 mL. This gets rid of a lot of the HCl (thus saving a bit on alkali), without causing hydrolysis of the Zr.

5. The reduced volume was diluted to about 1 L and carefully neutralised to pH = 11 with strong NaOH.

6. The precipitated hydrated zirconia was filtered and washed with copious amounts of boiling water until pH ≈ 8.

7. The washed precipitate was dissolved in 70 mL of 32 % HCl and boiled in (covered) to about 50 mL and allowed to cool and stand overnight. A semisolid mass of ZrOCl2.8 H2O formed, typically slightly yellow by FeCl3 from the HCl.

8. The crude zirconyl chloride was washed with several small amounts of acetone (distilled from nail polish remover) to remove water and FeCl3, then given a final rinse with pharma acetone, allowed to dry and dried further for about 15 h in a CaCl2 dessicator.

About 8.15 g of snow white product was thus obtained. Allowing for minor losses, it would appear 100 mL of ROAP contain about 9 g of zirconyl chloride.