J . Org. Chem. 1996, 61, 8315-8316
8315
An Efficien t P r ep a r a tion of
R*,R*)-1,2-Dicycloh exyleth a n e-1,2-d iol, a
Su p er ior Ch ir a l Dir ector for Syn th esis
w ith Bor on ic Ester s
proportion of 2 that was dehydroxylated to 1,2-dicyclo-
hexylethanol was never more than 2% with any com-
mercial rhodium catalyst, including the relatively active
rhodium on carbon. Rhodium trichloride hydrate costs
less per gram atom than the commercial catalysts, is
(
easily recyclable, and yielded similar results when used
William C. Hiscox and Donald S. Matteson*
together with alumina.5 Treatment of the crude hydro-
genation product 3 (and/or 4) with aqueous 2 M sodium
hydroxide and pentaerythritol yielded free diol 5 (95%
isolated).
Department of Chemistry, Washington State University,
Pullman, Washington 99164-4630
Received J uly 15, 1996
(R*,R*)-1,2-Dicyclohexylethane-1,2-diol (5) is a valu-
able chiral director for the insertion of a chloromethyl
group into boronic esters with (dichloromethyl)lithium
1
,2
to form chain-extended R-chloro boronic esters, and the
recent Wang-Sharpless preparation3 of (R*,R*)-1,2-
diphenylethane-1,2-diol (1) from stilbene makes this
precursor to 5 readily available in kilogram lots.4 The
original procedure for hydrogenation of 1, 14% in metha-
nol, over rhodium on alumina at 60 °C, 7 atm, by
1
Hoffmann and co-workers yielded 84% of 5, and with
the best of the variable lots of catalyst we encountered,
we obtained 98% of 5.
However, several obstacles remained to the routine
preparation of large batches of 5. Large batches of 1
could be hydrogenated at 1 atm and 25 °C but the
reaction required weeks to complete.2 Some batches of
1
poisoned the catalyst even after extensive efforts at
Instead of making 5, the dioxaborolane 3 can be used
directly in several ways. With simple boronic esters such
as methyl or isopropyl alkylboronates, transesterification
is fast and quantitative to yield the corresponding 2-alkyl-
2
purification. Addition of a small amount of acetic acid
sometimes seemed to help and sometimes did not, and
sometimes the catalyst became poisoned when the hy-
drogenation was partially complete.
Some batches of commercial rhodium on alumina were
active and yielded clean 5, but others showed low or no
activity, and some yielded several percent of the benzylic
4
,5-dicyclohexyl-1,3,2-dioxaborolanes. If 2 is first purified
6
by distillation and rigorous care is taken to exclude
water, the hydrogenation product 3 can be distilled.
Ethylation of 3 with ethylmagnesium bromide has been
hydroxyl hydrogenolysis products 1,2-dicyclohexylethanol
used to prepare 2-ethyl-4,5-dicyclohexyl-1,3,2-dioxaboro-
1
[distinguished by H NMR at δ 0.70-0.95 (m) and 3.37-
lane.2 However, if moisture is not excluded at all stages
1
3
0
.49 (m)] and 1,2-dicyclohexylethane [ H NMR δ 0.70-
of the hydrogenation, a significant amount of high boiling
oxybis(dioxaborolane) 4 (R ) cyclohexyl) is formed as a
byproduct if distillation is attempted.
A brief study with a simple boronic ester, (4R,5â)-2-
methyl-4,5-diphenyl-1,3,2-dioxaborolane (the analog of 2
.95 (m)]. Rhodium on carbon was a much more active
catalyst than rhodium on alumina, but yielded nearly
0% hydrogenolysis products. Rhodium trichloride was
3
also active, but caused 5-10% benzylic cleavage.
In the hope that benzylic hydrogenolysis might be
suppressed by conversion of the diphenylethanediol (1)
to a stable cyclic borate ester, excess trimethyl borate
was added to 1 to form (4R,5â)-2-methoxy-4,5-diphenyl-
having CH
3
-B in place of CH O-B), resulted in even
3
cleaner hydrogenation, with no visible NMR evidence of
hydrogenolysis to 1,2-diphenylethanol.
Recovery of most of the rhodium as the trichloride from
spent catalyst can be effected conveniently with hydro-
chloric acid. Hydrogen ion alone has insufficient oxidiz-
1
,3,2-dioxaborolane (2). Because 2 is a liquid, it could
be used in higher concentrations than 1.
This approach proved successful. Up to 27% 2 by
volume has been hydrogenated successfully at 50 °C
under 11 atm of hydrogen within 3 days, and the
7
ing potential to attack rhodium, and air oxidation of the
finely divided metal is likely involved.
Our improved hydrogenation procedure has allowed
the use, simple recovery, and regeneration of a reproduc-
ible and active rhodium catalyst, rhodium chloride
hydrate, which is soluble in methanol and deposits
rhodium metal on alumina in situ when the vessel is
pressured with hydrogen. Without alumina, the rhodium
plated out on the walls of the glass liner in the hydro-
genation vessel and was inactive. Alternatively, with
(1) Hoffmann, R. W.; Ditrich, K.; K o¨ ster, G.; St u¨ rmer, R. Chem. Ber.
1
989, 122, 1783-1789.
(2) (a) Matteson, D. S.; Man, H.-W. J . Org. Chem. 1993, 58, 6545-
6
1
547. (b) Matteson, D. S.; Man, H.-W.; Ho, O. C. J . Am. Chem. Soc.
996, 118, 4560-4566.
(3) (a) Wang, Z.-M.; Sharpless, K. B. J . Org. Chem. 1994, 59, 8302-
8
303. (b) Some samples of potassium osmate have proved insoluble
and inert under the conditions described in (ref 3a), but can be
dissolved in water (heated as necessary) to make a 1% solution and
then added over a period of ∼15 min. (We thank R. P. Singh for this
observation.) Solutions of osmium tetraoxide in 1:1 water/tert-butyl
alcohol also work well if added over the course of 1 h, but Wang and
Sharpless have noted, and our inexperienced helper has reconfirmed,
that the mixture overheats and gives a poor ee if the osmium is added
all at once.
(5) However, hydrogenation of 2 at 25 °C with rhodium trichloride
hydrate and charcoal required more than 1 week to complete and
resulted in 6% debenzylation products.
(6) Attempted vacuum distillation of 4, R ) phenyl, at ∼250 °C,
resulted in pinacol rearrangement to diphenylacetaldehyde. The
thermal stability of 4, R ) cyclohexyl, has not been tested.
(7) CRC Handbook of Chemistry and Physics; Weast, R. C., Lide,
D. R., Astle, M. J ., Beyer, W. H., Eds., 1989; pp D154-D156.
(4) Unfortunately, (R*,R*)-1,2-diphenylethane-1,2-diol is not a useful
chiral director for this purpose, as reported in ref 1 and replicated in
our laboratory by G. D. Schaumberg.
S0022-3263(96)01337-0 CCC: $12.00 © 1996 American Chemical Society