Paddywhacker - 28-2-2009 at 05:59
I've never done this reaction, but it looks like it could be very useful.
Either reduce an aldehyde or ketone to an alcohol, or, in reverse, oxidize an alcohol to a carbonyl. And all it needs, besides the catalyst, is IPA
for the reduction or acetone for the oxidation.
The only problem is the catalyst is aluminium alcoxide and I have not seen a prep for it. Does anybody know?
sparkgap - 28-2-2009 at 06:49
http://dx.doi.org/10.1002/047084289X.ra084
Personally, the need for mercuric chloride is a damper...
sparky (~_~)
chemrox - 28-2-2009 at 11:12
You having trouble getting HgCl2 Sparky? Course the expense is an issue too. I don't understand something though. Reading the article it sounds
more like Al amalgam than Al alkoxide is being used. I wonder why the Al alkoxide preps have to go this way?
[Edited on 28-2-2009 by chemrox]
Attachment: Al t-butoxide.pdf (139kB)
This file has been downloaded 1436 times
chemoleo - 28-2-2009 at 11:47
You just activate the Al that way, so only tiny quantities are needed for reaction Al with IPA (250 mg HgCl2, and 250 mg I2 for 50 g of Al).
'Activation' here means that the passivation layer of Al2O3 is permeabilised so that the unprotected Al underneath can react with the alcohol...
See this preparation for instance (german)
http://www.lambdasyn.org/synfiles/aluminiumtriethanolat.htm
[Edited on 1-3-2009 by chemoleo]
chemrox - 28-2-2009 at 16:09
That was a treat! I had forgotten the German for Hg was "quicksilver." "Quecksilber," to be in the language... Thanks!
sparkgap - 28-2-2009 at 21:50
Yes CRX, that was what I was alluding to; corrosive sublimate 
here is hard to
get, not unless you pay a lot and sign not a few papers.
I suppose I could make my own, though...
sparky (~_~)
not_important - 28-2-2009 at 23:41
Boron alkoxides can also function well in M-P-V reductions of ketones and aldehydes. They do fail for aromatic carbonyl compounds, where the benzene
ring is directly attached to the C=O as with benzaldehyde and acetophenone.
I've used this route long ago, and there's some write-ups in the journals on it (one is attacked). The borone alkoxide (borate ester) can be made by
adding boric acid to isopropanol and toluene/xylene then azeotroping off water.
It doesn't always give good yields, but boric acid or borates and isopropanol are both readily accessable so it's worth checking out. It is even
milder than the aluminium version, there's only a few other groups that interact; epoxides are one and active halides sometimes do.
Note that for oxidations the isopropanol formed is removed by fast simple diatillation of the acetone, which carries some of the isopropanol along. A
large excess of acetone must be used. I've never tried it with borate isopropoxide.
Attachment: 1051.pdf (61kB)
This file has been downloaded 662 times
Paddywhacker - 1-3-2009 at 00:16
Thanks for that, n_i. A great article.
But alkyl borates can only be made by the Dean & Stark method if the alcohol boils above that of the water-toluene azeotrope. Maybe there is
something that will azeotrope away water at a lower BP than that of IPA, and that does not also azeotrope away the IPA itself.
Trimethyl borate can be made by batch processing, and then maybe that can be used to transesterify with IPA at a temperature above the BP of methanol
but below the BPs of IPA and trimethyl borate.
not_important - 1-3-2009 at 01:05
toluene-isopropanol-water have a tertiary azeotrope 48.7:38.2:13.1 boiling at 76.3 C As this is only a little below the BP of isopropanol there will
be additional alcohol coming over. It works, I've done it. You could do this in a Soxhlet with molecular sieves in the extraction thimble, and omit
the toluene. C5, C6, & C7 alkanes also work, buy cold weather petrol and fractionate.
MEK-iPr-water boils at 73.5, 88:1:11, but consider what the purpose of making the B(O-iPr)3 is ...
Isopopyl eather would work better than toluene, were it not one of the worst ethers for peroxide formation. MTBE might work, I've no idea if it forms
a useful azeotrope.
Trimethyl borate forms an azeotrope with methanol, roughly 3:1 and boiling at 54 C. As this is 10 C below the BP of the alcohol and ester, making it
the lowest boiling component in the full system you propose, transesterfication isn't going to work.
benzylchloride1 - 1-3-2009 at 20:47
If you have access to anhydrous aluminum chloride and fairly dry isopropy alcohol, aluminum trisopropoxide should be easy to make. 91% isopropanol can
be obtained from WalMart and dried with anhydrous magnesium sulfate. The isopropanol is placed in a large flask. The aluminum chloride is slowly added
with swirlling, while the mixture is heated to boiling. The flask needs to be large and a efficent reflux condenser with a drying tube is attached. A
ice bath should be present to control the highly exothermic reaction. The aluminum chloride forms the isopropoxide, this is a rather expensive method
though. I once prepared aluminum isopropoxide. I amalgameted the aluminum, but the reaction did not start. I heated the mixture and added anhydrous
aluminum chloride. The aluminum chloride removed the remaining water and formed aluminum isopropoxide. The mixture darkened, and the aluminum
dissolved. The reaction became very vigerous. The product after refluxing for several hours and removing the excess sopropanol was distilled under
reduced pressure. The product crystallized and was highly water reactive. I am afraid that organomercury intermediates could be formed, these scare me
to death. One of my professors told me not to work with amalgams in contact with organic materials, several of his co-workers many years ago where
working with sodium amalgam in synthesis. Within a month they became extremly ill and nearly died. I want to reduce 2-butanone to sec-butyl alochol
via this method. Procedures for the preparation of the alkoxide and methods for the reduction can be found in Vogels textbook.
[Edited on 1-3-2009 by benzylchloride1]