Protecting group for pinacol
Glen Brizius
acceptable protecting group for my compound, a substituted pinacol (1, 2
diol, two benzene rings, some groups off the benzenes, etc). I've tried
what I view as the standard, ie acetonide formation, with poor yield (even
when attempting to remove water). I've also tried protecting using
SiCl2(i-Pr)2, but the resulting cylic bis-ether is quite unstable to both
acid and water and simply falls apart on my silica gel column. The same is
true surprisingly for SiCl2(phenyl)2. I'm looking at some pretty
unappealing choices, i.e. di(t-butyl)SiCl2 (should be more stable, but it's
prohibitively expensive), carbonate formation (rock solid to acids, but
tricky to do because of dangerous reagents...trichloromethyl chloroformate
is no toy, and phosgene is no better), and maybe methylene ketal formation
with paraformaldehyde / conc. H2SO4 / heat, which I'm afraid will introduce
a pinacol rearrangement, and which even if made successfully is more
difficult to remove.
Does anyone have any ideas? I can't use a simple monofunctional protecting
group like TMSCl, because part of my reaction scheme involves "tethering"
the two oxygens together to trap the correct orientation for the next step.
So I need one protecting group to tie up both alcohols.
Comments gratefully accepted
Flames sent back to hell
_______________________________
Glen Brizius
Dept. of Chemistry, USC Columbia
http://www.cosm.sc.edu/~brizius
"Man, that Raney cobalt really *is* pyrophoric!"
--
Paul J. Franklin (moderator - sci.chem.organic.synthesis)
http://chemistry.gsu.edu/post_docs/koen/wnewshp.html
Georgia State University <che...@panther.gsu.edu>
Atlanta, GA
Uncle Al
Glen Brizius wrote:
>
> Hi all. For the last 3 weeks I've been patiently trying to find an
> acceptable protecting group for my compound, a substituted pinacol (1, 2
> diol, two benzene rings, some groups off the benzenes, etc). I've tried
> what I view as the standard, ie acetonide formation, with poor yield (even
> when attempting to remove water). I've also tried protecting using
> SiCl2(i-Pr)2, but the resulting cylic bis-ether is quite unstable to both
> acid and water and simply falls apart on my silica gel column. The same is
> true surprisingly for SiCl2(phenyl)2. I'm looking at some pretty
> unappealing choices, i.e. di(t-butyl)SiCl2 (should be more stable, but it's
> prohibitively expensive), carbonate formation (rock solid to acids, but
> tricky to do because of dangerous reagents...trichloromethyl chloroformate
> is no toy, and phosgene is no better), and maybe methylene ketal formation
> with paraformaldehyde / conc. H2SO4 / heat, which I'm afraid will introduce
> a pinacol rearrangement, and which even if made successfully is more
> difficult to remove.
>
> Does anyone have any ideas? I can't use a simple monofunctional protecting
> group like TMSCl, because part of my reaction scheme involves "tethering"
> the two oxygens together to trap the correct orientation for the next step.
> So I need one protecting group to tie up both alcohols.
Instead of azeotroping off the water during acetonide formation (or
using an azeotrope plus a Soxhlet filled with activated 3A molecular
sieves), you might try running the thing while slowly adding enough
Me3SiCl to scavange all the water. I had a hindered carbonyl that
would not form an ethylene glycol ketal, but the TMSCl pushed it
along. Byproduct was TMS-O-TMS and some other stuff.
The acid sensitivity of the pinacol is a problem. Maybe you can
buffer the thing along with pyridinium triflate or methanesulfonate as
catalyst, with excess pyridine to grab the HCl. Or use an acid
pyridinium salt as catalyst and a 2,6-hindered pyridine or poly(vinyl
pyridine) to go after the HCl.
--
Uncle Al Schwartz
http://uncleal.within.net/
http://www.mazepath.com/uncleal/
http://www.ultra.net.au/~wisby/uncleal/
http://www.guyy.demon.co.uk/uncleal/
(Toxic URLs! Unsafe for children and most mammals)
"Quis custodiet ipsos custodes?" The Net!
Gabriel Tojo
[Your display is set for the "US-ASCII" character set]
[Some characters may be displayed incorrectly]
Glen Brizius wrote:
>
> Hi all. For the last 3 weeks I've been patiently trying to find an
> acceptable protecting group for my compound, a substituted pinacol (1, 2
> diol, two benzene rings, some groups off the benzenes, etc).
Either the diol is quite unreactive and you need harsh reaction conditions, or it is
quite sensible and decomposes, and you need milder conditions. Which is your case?
I've tried
> what I view as the standard, ie acetonide formation, with poor yield (even
> when attempting to remove water). I've also tried protecting using
> SiCl2(i-Pr)2, but the resulting cylic bis-ether is quite unstable to both
> acid and water and simply falls apart on my silica gel column. The same is
> true surprisingly for SiCl2(phenyl)2. I'm looking at some pretty
> unappealing choices, i.e. di(t-butyl)SiCl2 (should be more stable, but it's
> prohibitively expensive),
I would not try with cyclic silanes. They are too sensible to base and acid.
carbonate formation (rock solid to acids, but
> tricky to do because of dangerous reagents...trichloromethyl chloroformate
> is no toy, and phosgene is no better),
Other possibilities are (MeO)2CO/NaOMe (Tetrahedron, 4995, 90) and carbonyldiimidazol
(Tetrahedron, 413, 92)
and maybe methylene ketal formation
> with paraformaldehyde / conc. H2SO4 / heat, which I'm afraid will introduce
> a pinacol rearrangement, and which even if made successfully is more
> difficult to remove.
Yes.
>
> Does anyone have any ideas? I can't use a simple monofunctional protecting
> group like TMSCl, because part of my reaction scheme involves "tethering"
> the two oxygens together to trap the correct orientation for the next step.
> So I need one protecting group to tie up both alcohols.
In JACS 634, 95, there is an example in which a diol does not react with phosgene. You
could have some hint of your problem in this paper.
An interesting approach for diol protection is the p衫ethoxybenzylation of one of the
alcohols followed by treatment with DDQ. DDQ produces a benzylic cation wich is attacked
by the free neighbouring alcohol (Tetrahedron, 2291, 91).
Please, give us more data.
Best regards.
--
Gabriel Tojo
Profesor Titular de Quimica Organica
Royal University of Santiago de Compostela
SPAIN
Ron Forsch
>Hi all. For the last 3 weeks I've been patiently trying to find an
>acceptable protecting group for my compound, a substituted pinacol (1, 2
>diol, two benzene rings, some groups off the benzenes, etc). I've tried
snip
>prohibitively expensive), carbonate formation (rock solid to acids, but
>tricky to do because of dangerous reagents...trichloromethyl chloroformate
>is no toy, and phosgene is no better),
Oh, I think you can find some safe phosgene substitutes. Replace the
chlorine with active leaving groups like N-hydroxysuccinimide or
4-nitrophenol. Many of these are available commercially. A quick
check in Fieser would probably give a lit ref for carbonate formation,
most likely with addition of 4-dimethylaminopyridine.
Kevin Shaughnessy
> Hi all. For the last 3 weeks I've been patiently trying to find an
> acceptable protecting group for my compound, a substituted pinacol (1, 2
> diol, two benzene rings, some groups off the benzenes, etc). I've tried
> what I view as the standard, ie acetonide formation, with poor yield (even
> when attempting to remove water). I've also tried protecting using
>
[snip]
> Comments gratefully accepted
>
> _______________________________
> Glen Brizius
> Dept. of Chemistry, USC Columbia
> http://www.cosm.sc.edu/~brizius
> "Man, that Raney cobalt really *is* pyrophoric!"
>
Glen,
I've had excellent success using 2,2-dimethoxypropane as an acetonide
forming reagent in place of acetone. The reaction is extremely
mild--simple stirring, or possibly heating, in DMF. I've used this for
reduced sugars and 1,3-diols with near quantitative yields. Unfortunately
I've moved on since doing that work so can't offer any references.
Good luck,
Kevin Shaughnessy (203) 432-3943
Dept of Chemistry, Yale University kevin.sh...@yale.edu
P.O. Box 208107
New Haven, CT 06520-8107
Bill Walker
ring?
Instead of using silicon which is extremely sensitive to hydrolysis, perhaps
you could make a cyclic borate or carbonate (like is done with catechols)?
What about the thiocarbonate made with thionyl chloride? (I think this may be
possible)
We've tried to do some oxime formation with the benzil version of your
compound class, and found that simple reacting the reagent with the substrate
in a solventless melt was much more successful than trying to do the reaction
in solvents or with azeotropic removal of water. This trick may work for you
also.
Otherwise, you may decide to purify your cyclic silyl ether by means other
than silica gel, or by using a different adsorbant. What activity grade is
the silica gel? A higher activity may provide chromatographic conditions to
which your compound is stable. Have you considered using reverse-phase silica
gel (there will no longer be the problem of adsorbed water).
Best of luck,
Bill
Glen Brizius wrote:
> Hi all. For the last 3 weeks I've been patiently trying to find an
> acceptable protecting group for my compound, a substituted pinacol (1, 2
> diol, two benzene rings, some groups off the benzenes, etc).
>
> [SNIP]
> Does anyone have any ideas? I can't use a simple monofunctional protecting
> group like TMSCl, because part of my reaction scheme involves "tethering"
> the two oxygens together to trap the correct orientation for the next step.
> So I need one protecting group to tie up both alcohols.
>
--