Sciencemadness Discussion Board - Easy anhydrous AlCl3 production

Easy anhydrous AlCl3 production

greenimp - 21-7-2009 at 11:13

I wanted to post something I have been thinking about recently. I have spent alot of tie trying to produce 100 gram scale quantities of anhydrous AlCL3. I have experimented with Al powder is a variety of solvents, with dry HCL gas bubbled through the solution. I have achieved success with this procedure, but rejected it for larger scale production.

Al and Cl2 gas has also rejected for reason of scale/cost

I know someone will say, anhydrous AlCl3 is not that expensive, or hard to purchase. This is true, but the hazardous shipping costs are often greater than 50% the cost of the item. A method that allows medium scale production at low cost would be a boon to my pocket book.

Recently, this patent, US patent 3471250, link has come to my attention. It has been mentioned here a few years ago in relation to creating anhydrous FeCl3. After reading the patent, I think it might suit my needs nicely. One can make hexahydrate AlCl3 easily enough from HCl and Al foil.

Moreover, the AlCl3-DMSO precipitate that forms is also a really nice way to store AlCL3 long term, as it can quickly be heated under vacuum to generate fresh anhydrous AlCl3 as needed.

I will post my experiences with the process soon. If anyone has experience with this method, and wants to comment , have at it.

Klute - 21-7-2009 at 11:32

Nice find! Using triethylamine looks like a good alternative, with a lower bp, and it's pretty cheap...

It's pretty surprising that even the dihydrate of magnesium can be deshydrated without hydrolysis..

Let us know how it goes!

greenimp - 21-7-2009 at 12:04

To be true it was not me that found the patent. It was bought to my attention by another.

I have a gallon of 99% DMSO that is left over from other experiments, so I will be trying it first.

Plus I can probably recover most of it, if I do a large batch.

chemrox - 21-7-2009 at 16:05

I also found that one .. right after finding a vendor with a decent price. Did you make the 99% DMSO from cruder stuff by distillation or buy it that pure?

greenimp - 21-7-2009 at 16:30

My local health food store stocks the 250ml and 500ml bottles of 99% DMSO. I asked the manager if they could order me a gallon, because I was buying so much, and they did it.

I have used it when I was going crazy trying to get my methyleneation experiment to work. It works fine in those reactions.

If this procedure works I am going to be really upset, because I spent a lot of weekends with my gas generator with very little to show for it, save a lot of spent solvent.

This is just silly easy if it works. Oh, and it was 2bfrank that brought the patent back to the surface. He should get the credit for tracking it down.

starman - 21-7-2009 at 18:10

I'm assuming you have a capable vacuum pump.The one example they ran at less than savage vacuum(the one @3mm Hg) required 10 hours on a scale 1/100th what you are looking at.
Do let us know how you go.DMSO is at least readily available.

Polverone - 21-7-2009 at 18:21

Anhydrous aluminum chloride sublimes pretty readily even at atmospheric pressure. It seems a lot more volatile than any of the example compounds in the patent.

greenimp - 21-7-2009 at 19:26

Good point. I encountered that with my previous experiments. Actually, you might be able to use the sublimation to purify it. Heat it under vacuum, get it to sublime, and recover the vapor. I'll have to get my notebook out, but I think I saw sublimation around 60C with my vacuum at full for AlCl3. Sublimation might not work at the large scale, but for 50-100g it might work just fine.

I also wounder it just re-suspending the AlCl3-DMSO in something like DCM would dissociate the two? Maybe another solvent?

Also the vacuum times seem to change with the halide, so 10 hours might not be norm.

Also, having read the patent again in more detail, it says up to a 50% aqueous solution of DMSO can be used. So apparently a little water in the drying reaction is not too critical.

len1 - 21-7-2009 at 21:20

AlCl3 is a stronger Lewis acid - due to the fact that Al3+ ion charge to size ratio is greater, therefore it is harder to dehydrate - than any of the other Lewis acids used in the examples. I do not think thats an accident, but might be worth a try. My advice to many here is - do not boast of what you are going to do, do first, post later, to avoid looking silly.

[Edited on 22-7-2009 by len1]

chloric1 - 22-7-2009 at 14:50

This is completely off the DMSO track but here is another approach. Heating aluminum powder with anhydrous cupric chloride. Anhydrous cupric chloride is easily obtained from the dihydrate by simply heating to 102 Celsius. The reaction is fast and thermite-like but if you have an efficient means of collecting the AlCl3 fumes, you are good to go.

kclo4 - 22-7-2009 at 15:25

If the AlCl3 is to hard to get from the AlCl3-DMSO via vacuum and heating, maybe a compound that could trap the DMSO and free the AlCl3 could be used/found?

I think a lewis acid that has a significantly higher sublimation temperature could allow the reaction to shift towards the free AlCl3 as it is heated, having the AlCl3 sublimate away leaving the lewis acid-DMSO behind.

AlCl3-DMSO + X = X-DMSO + AlCl3

X = suitable lewis acid, such as FeCl3, MgCl2, etc.

Also would the AlCl3-DMSO itself work as a lewis acid at all? It seems it may if it were in boiling xylene or toluene, as at 110+*C, as it may break down a bit giving up free AlCl3 and DMSO, this might allow for some other method of separating it.

greenimp - 23-7-2009 at 19:52

Well gave this a try today.

7g AlCl3-6H20 was dissolved in 50 ml 95% EtOH. To this was added slowly 50 ml of 99.9% DMSO. The DMSO was added slowly in 1ml aliquots to see if precipitation would occur. After 50 ml had been added, no precipitate had formed. The solution was clear at the end of the DMSO addition.

The solution was then heated to remove the EtOH at atmospheric pressure. As the EtOH was removed, the solution took on a yellow color that increased as the EtOH was removed. When about 45 ml of EtOH had been evaporated, the solution turned cloudy and a fluffy/snowy white precipitate formed. The was removed with vacuum filtration. The weight if the vacuum material was 16.5g.

The 8g of this material was then placed in a round bottom flask and vacuum of 26 inHg was applied while heating to 120C for 2 hours. Noticeable vapor was seen leaving the material can condensing on the inside of the flask. Later inspection of the condensate showed it to be DMSO. After 2H some of the material had taken a yellow color, and the rest was a crunchy white solid, where as the starting material was a fluffy white-soft solid.

Also I should add the precipitate will re suspend in water.

I am going to try a longer vacuum treatment in a smaller flask tomorrow.

[Edited on 24-7-2009 by greenimp]

Klute - 23-7-2009 at 19:59

You should have tried to sublim the solid! So is it anhydrous AlCl3? What was the weight loss?

greenimp - 23-7-2009 at 20:46

No I don't think it is anhydrous yet, I think all I did was remove some more un-complexed DMSO that had not been removed during filtration. I suspect all my vacuum treatment did was dry the material further.

But I do think I was able to get the AlCl3-DMSO complex to form. To what extent will require more testing.

The material does not react "violently" with water, it just re suspends in it.

Maja - 25-7-2009 at 01:42

What about heating AlCl3-DMSO complex in a flask till the complex decomposed, then pour some dry organic solvent to extract DMSO ? And you are left with AlCl3 with some organic solvent(in most cases AlCl3 is suspended in solvent, so I don't even know who would like to evaporate solvent from AlCl3)?

kmno4 - 25-7-2009 at 06:58

Many years ago I did experiment with heating ( 300-400 C as far i remember) of Al pieces (~2x2x2 mm) in stream of HCl (from H2SO4/NaCl) in glass pipe. I prepared about 50g AlCl3 in this way, just to see if it works at all.
Simple, rapid and the cheapest method I know.
Next I threw this AlCl3 into water in water-closet ....
Good thing for a show

I also remember specific smell of anhydrous AlCl3.

AlCl3 is strongy coordinated especially by substances containing oxygen, so preparing it by DMSO method is rather losing the time. But for same experimenting.... why not.
ps. 80% of patents (especially old ones) is a bullshit.

[Edited on 26-7-2009 by kmno4]

Vogelzang - 25-7-2009 at 15:23

Look at the aluminum chloride patents here:

https://www.hyperlab.info/inv/index.php?s=309f6d356ac3dee658...

Хорошая мысль!

[Edited on 25-7-2009 by Vogelzang]

len1 - 25-7-2009 at 18:34

The fact that far too many patents are bullshit can be vouched for from my many experimental pages on this site. But the above patent I dont believe lies in this class. Good idea indeed, why didnt I think of that. AlCl3 must sublime out of a melt mixture of low b.p. chlorides, such as the sodium-calcium eutectic of the downs cell, and aluminium sulphate by LeChatliers. Who wants to mess with HCl at 700C? On the contrary, the DMSO method, apart from being economically silly, does not have thermodynamics on its side either.

kmno4 - 26-7-2009 at 02:12

700 C is too much, there is no need for such heating and it is above melting point of aluminum.
Al quickly reacts with hot HCl, AlCl3 sublimates above 200 C, surface impurities of Al (mainly oxide) just leave in pipe, product is pure and ready to use. The only messy thing is strong corrosive effect of HCl when it is out of aparaturus. Just every metal item close to HCl sourse will cover with rust soon. Unfortunately it was tested

But it happens only if you let HCl go out.
Al, H2SO4, NaCl - what can be cheaper ?
Glass pipe - I used laboratory glass pipe 3 cm in diameter, ~40 cm long, filled in central part with Al pellets (cut 2mm wire), heating wire around the pipe....
One of the earliest experiments of mine, so cannot be difficult to do. I do not force anyone to do this but it seems to be the only good method for preparing AlCl3 on multigram scale.

len1 - 26-7-2009 at 02:59

You are right, I exaggerated for effect. Its more like, 400-500C in a tube oven.

But to answer your question, NaCl, CaCl2, and Al2SO43 is cheaper. You dont have to adjust HCl rate with possibility of generating too much effluxing you product and corroding everything. The former method you can do completely hermatically on a large scale in a box oven, the later as you say needs a tube so yield is limited. Plus this method is new. AlCl3 often comes from reputable suppliers (like my last batch from sigma, partly hydrated. Its a problem if you want it to make LiAlH4 to get around the high cost of the latter. This is in fact the topic for which it was suggested, but my understanding is noone tried it out.

Sedit - 26-7-2009 at 16:22

This appears to be a nice way to produce AlCl3 len and I wouldn't mind attempting a small amount just for fun but I must ask about Aluminum sulfate and how does one obtain it anhydrous? Will simple heating suffice or should desiccation or other means be used?

len1 - 26-7-2009 at 19:32

Thats very easy. Al sulphate melts and undergoes various hydrates of decreasing water content, starting at 86C, until by 250C or so its essentially anhydrous.

But its even easier than that, because alum, the dirty variety sold in gardening supplies will do even if you want your AlCl3 99%. The AlCl3 will purify itself by fractional distillation.

Sedit - 26-7-2009 at 19:45

Very nice, Thank you much len.

I can't help but wounder why this is not more popular though considering everything that gets spoke of on these forums talk about feeding dry HCl over hot Al and condensing the vapors IIRC. Seems rather trouble some when put into perspective against this process.

len1 - 27-7-2009 at 01:08

Ah yes, I was rather thinking that two. Two possibilities come to mind

1) The patent is obscure - so those passing HCl over Al dont know better
2) It doesnt work and the patent author is a lier - my concern is the Al sulphate getting covered by insoluble CaSO4 stopping reaction.

Which is it? You cant rule this one out on theory, so the only way is to do it. If you or someone else dont get there first, Ill find out myself. Thats why science is more fun than watching crap on TV.

[Edited on 27-7-2009 by len1]

Magpie - 27-7-2009 at 07:45

Quote:

Thats why science is more fun than watching crap on TV.

TV's crap is about to reach a new low. Nadya Suleman is getting a "reality" show where each of her 14 children will be paid $250/day.

Sorry for dumping in this important thread, but I couldn't help it.

Sedit - 27-7-2009 at 18:33

Don't know how useful they might be to anyone here because they both seem out of my league(allthough the last one seems doable to some extent) but I thought I would post them anyway because they are different then the methods I hear about the most.

Quote:

(c) removing all soluble iron chloride from said liquor to obtain a substantially iron free, aluminum hloride-containing pregnant liquor;

(d) crystallizing said pregnant liquor to produce a separable slurry of phosphorous or magnesium-containing aluminum hexahydrate (ACH) crystals;

(e) heating said phosphorous or magnesium-containing ACH crystals from said slurry to a temperature of about 200° C. to 450° C. to produce partially calcined ACH (PCACH);

(f) chlorinating said PCACH in the presence of chlorine and a reductant to produce said anhydrous aluminum chloride;

Reference:http://www.patentstorm.us/patents/4465566/claims.html

US Patent 4096234 - Production of anhydrous aluminum chloride from clay using catalyst and recycling of silicon chloride

Quote:

This invention relates to the production of aluminum chloride from clay. More particularly, this invention relates to an improved process for the chlorination of clays containing silicon oxide as well as aluminum oxide whereby the aluminum oxideis chlorinated while the chlorination of the silicon oxide is suppressed.

Reference: http://www.patentstorm.us/patents/4096234/description.html

@Len I have been meaning to ask you but never remebered to do so. You called the DMSO method "economically silly" but I fail to see how it is so. The DMSO is recovered and is only used to complexe with the AlCl3 long enough to dehydrate it and the two are seperated with heat to recover the dehydrated AlCl3 and DMSO. Since the only thing that leaves the reaction is H2O I don't see how this is a non economical means to dry your AlCl3.

Am I missing something?

kmno4 - 28-7-2009 at 04:34

Kirk-Othmer Encyclopedia of Chemical Technology - for those who want to know how AlCl3 is produced idustrially. Brauer etc. - for laboratory methods.
Patents - is somebody likes stories in kind of "making hexogen or opium from sugar, water, DMSO, muriatic acid and simple kitchen devices". Sometimes it works, but only at times.

Bolt - 28-7-2009 at 11:29

Quote: Originally posted by len1

Ah yes, I was rather thinking that two. Two possibilities come to mind

1) The patent is obscure - so those passing HCl over Al dont know better
2) It doesnt work and the patent author is a lier - my concern is the Al sulphate getting covered by insoluble CaSO4 stopping reaction.

Which patent is this?

Sedit - 28-7-2009 at 13:47

The one posted on Hyperlab with the link supplied by Vogelzang,

Its ironic because I was thinking about something along the same lines using Aluminum, H2SO4 and NaCl to attempt to produce it when I first heard of the DMSO patant but starting with the sulfate on this one seems like a better way to go.

Ill link the one PDF here,

Patent US1818839:
Process for manufacturing anhydrous aluminum chloride

Attachment: Aluminum Chloride.pdf (140kB)
This file has been downloaded 879 times

SunlightChemist - 10-8-2009 at 23:38

Thanks guys, from what i hear, it would seem that making it by simply passing chlorine gas through hot/molten aluminium is the easiest way to go. I'll do that me thinks.

- T.

not_important - 11-8-2009 at 04:41

Using molten aluminum is overkill, hot solid aluminium will ignite in chlorine, running Cl2 into molten Al should be reasonably exciting.

Simply take aluminium in the form of chips, turnings, or chopped wire, washing it free from grease with acetone, then fill most of the length of a borosilicate tube a cm or two in diameter with the Al, using borosilicate glass wool or the similar high temperature wool used in kilns. Mount the tube horizontally, arrange to heat it, except for the end section (which should be empty), to about 300 C. One end should exit through a stopper into a wide mouth bottle, which will be cooled with an air blast, and with a second tube in the stopper for removing excess HCl gas. The other end is connected to a generator of dry HCl, and has an additional input to pass in dry CO2 or N2.

Turn on the N2 or CO2, pass it through the apparatus for 5 or 10 minutes, then turn on the heat to raise the Al containing section of the tube to ~300 C, and again wait a few minutes. Then start passing dry HCl through the tube. Watch out for AlCl3 condensing inside the exit end of the tube, apply heat using a burner or hot air gun to sublimate any solid forming, forcing it into the condensing bottle.

It should be noted that the gases leaving the condensing bottle via the exit tube will contain H2 and HCl, so must be properly vented or if you are a clever lad, washed and the H2 collected.

Turn off the HCl feed when there is only a few cm of unreacted Al metal remaining, while providing a flow of dry N2 or CO2. This section serves as a trap for alloying metals in the aluminium that have volatile chlorides, most of their chlorides will react with the Al metal and be reduced back to their original metallic form.

Keep the dry N2 flowing, turn off the heaters, and allow the apparatus to cool somewhat. While still warm remove the condensing bottle with its load of AlCl3, and tightly stopper it.

There's an older version of this in Thorp's Inorganic Chemical Preparations, which is out of copyright and can be downloaded from the Internet Archive or Google Books.

[Edited on 11-8-2009 by not_important]

Eclectic - 11-8-2009 at 21:09

I have seen a patent involving the purification of AlCl3 by distilling the AlCl3:Et2O adduct. I'm having some trouble finding it as http://gb.espacenet.com seems to be having issues. Ether solutions maybe a more convenient way to store and use AlCl3.

bbartlog - 22-3-2010 at 12:06

As I recently prepared about 500g of cupric chloride (CuCl2), I wondered whether it would be possible to produce anhydrous AlCl3 via reaction of aluminum with anhydrous CuCl2. There are of course various chlorides that could theoretically by reduced by aluminum, but CuCl2 is fairly easy both to prepare and to dry. If this approach were practical it would be one way to avoid the use of gaseous chlorine and/or HCl.
The difference in heats of formation (706kJ/mol for AlCl3 versus 358kJ/mol for CuCl2) suggests that the reaction would be exothermic but not to my mind dangerously so. My thought would be to add some aluminum foil and powdered CuCl2 to one glass vessel, connect this by glass tube to a receiver, and put the entire arrangement in an oven with the receiver set in water to keep its temperature down around 100C. The oven being possibly necessary to keep the AlCl3 from condensing before it reaches the receiver.
The first question that occurred to me was whether I should use a mole ratio of 3:2 (CuCl2 to Al) or 3:1. While the aluminum should be able to fully reduce the copper, it occurs to me that the reaction might not proceed to completion if all that is left is aluminum coated with copper alongside CuCl and AlCl3, so a 3:1 mole ratio is probably better.
On doing further searching, I came across this paper: http://www.dtic.mil/cgi-bin/GetTRDoc?AD=AD277616&Locatio...
It's not directly applicable to my interest, but it has some interesting data on a mixture of AlCl3 and CuCl (which I will surely end up with at some point during this reaction). Notably, the vapor pressure of the mixed AlCl3/CuCl is higher than expected (see table 5 on page 16). This at first seems advantageous (the AlCl3 should vaporize at a lower temperature), but on further reflection it sounds like there may be a minimum boiling azeotrope of these two chloride salts. In which case this approach would result in some mixture of CuCl and AlCl3 rather than pure AlCl3 condensing in the receiver.
Can anyone point me to further information on possible azeotropes of AlCl3 and CuCl, or for that matter just some reading on azeotropes of chloride salts generally? I have found numerous references to eutectics but almost nothing on azeotropes. Also, besides this possible problem of separating CuCl and AlCl3, can anyone point out problems in the proposed approach for preparing AlCl3?

[Edited on 22-3-2010 by bbartlog]

Polverone - 22-3-2010 at 12:17

Some years ago I was able to make a self-sustaining exothermic mixture that burned with visible flame from anhydrous CuCl2 rolled inside aluminum foil. This may be because I had a mixture of oxygen-containing species along with the chloride after simply heating CuCl2 to dryness. Still, I would be cautious not to underestimate the potential exotherm.

bbartlog - 22-3-2010 at 12:27

Interesting. I suspect that oxygen was involved, that is, the chloride reaction served mainly to continuously strip the passivation from the aluminum, and most of the energy came from the formation of oxides. In that case I'll make sure to start small, and try to exclude oxygen as best I can...

peach - 31-8-2010 at 01:33

Quote: Originally posted by len1

Thats why science is more fun than watching crap on TV.

I can forgive any negative vibes emanating from len in other posts due to this, the rest of his posts here and particularly the mention of a spectrum within the hard / soft definitions, something I've become very familiar with. Not forgetting the DIY sodium.

I have a liter of DMSO handy, so I could give it a go, but I'd prefer to go to the effort of whipping some ether up so I have that around if I need it for anything else. I expect it'll also be far easier to work with than DMSO given the massive BP difference. Producing AlCl3 by a hot method isn't the issue for me, it's dealing with the clogs and lumps it produces. The clogs mean problems. Poking around with sticks is asking for broken glass and it'll never be clean or particularly free of BroLow acids with that much atmospheric exposure.

These AlCl3 threads are getting busy. Don't forget FeCl3, which can be dried to anhydrous under a stream of HCl(g) without it subliming all over the place.

If you don't need a hard acid, things get even easier. Metal oxides feature Lewis acid sites. Microwave irradiation of some aryls, particularly those with halides attached to the opposite side of the ring to the ether linkage, show potentially useful charge shifting in response to microwaves, which appears to allow softer Lewis acids to attack it's lone pair when they normally wouldn't be bothered; hard Lewis acids attack hard Lewis bases, and soft attack soft, preferentially. By making something like the lone pair on an oxygen appear softer than it would normally, softer acids attack more efficiently.

[Edited on 31-8-2010 by peach]