Angewandte
Communications
Chemie
barely converted (not shown; see the Supporting Information
for details). The chemoselectivity of the reaction with respect
to sulfonate against bromide removal is clearly in favor of the
tosylate (6!7; Scheme 2, bottom). The above reactions all
showed full conversion within a reaction time of 12 h.
Conversely, tosylate 3a, which was derived from the corre-
sponding secondary alcohol, furnished hydrocarbon 4 in
diminished yield owing to competing elimination; bromide 3c
reacted cleanly in near-quantitative yield (3!4; Scheme 2).
Unlike acyclic substrates 1 and 3, cyclic tosylate 5 exclusively
underwent elimination.
With the initial finding that B(C6F5)3 catalyzes the
reductive cleavage of primary tosylates, we turned towards
the chemoselective deoxygenation of orthogonally protected
1,n-diols. We opted for TBDMS protection on one end of the
diol for its ease of deprotection. It must be noted, however,
that the less bulky EtMe2Si group is also fully compatible with
the reductive removal of the tosylate.[10a] Diols with longer
carbon chains underwent chemoselective cleavage of the
tosylate in high yields (8/10/12!9/11/13; Scheme 3, top).
Figure 1. Mechanistic explanation of rearrangements proceeding by
anchimeric assistance.
This reaction is likely to involve the aforementioned three-
membered silyl oxonium ion (Figure 1, right).
We next focused on the chemoselective removal of
tosylates in the presence of aryl ethers (Scheme 4). It must
be emphasized that Gevorgyan, Yamamoto, and co-workers
have already reported on B(C6F5)3-catalyzed reductive ether
cleavage with Et3SiH at room temperature.[2b] Applying our
Scheme 4. Chemoselective cleavage of tosylates in the presence of aryl
ethers. [a] Determined by 1H NMR spectroscopy with 1,3,5-trimethoxy-
benzene added after the reaction. [b] Yield of isolated product after
flash column chromatography on silica gel. [c] Et3SiH (2.4 equiv).
standard procedure, we tested ethylene glycol derivative 21a,
which reacted with moderate selectivity (21a!22a). How-
ever, high yields were achieved for 21b–d, which are derived
from the higher diol hexane-1,6-diol. Substituents at the aryl
ring that are often prone to interfere with transition-metal-
catalyzed or radical reductions were tolerated (21b–d!22b–
d). A CF3 group was also compatible but the corresponding
product was obtained in low yield (21e!22e). Tosylate 21 f
with an additional methyl ether did not afford the desired
compound; both tosylate removal and methyl-ether cleavage
occurred while the methylene linkage was left intact. The free
phenol was still isolated in decent yield with double the
amount of Et3SiH (21 f!22 f).
Scheme 3. Chemoselective cleavage of tosylates in the presence of silyl
ethers. [a] Determined by 1H NMR spectroscopy with 1,3,5-trimethoxy-
benzene added after the reaction. [b] Yield of isolated product after
flash column chromatography on silica gel. TBDMS=tert-butyldime-
thylsilyl.
However, the 1,2- and 1,3-diols 14 and 16 were transformed
into the rearranged products 15 and 17, respectively, with
neighboring-group participation (14/16!15/17; Scheme 3,
middle). These defunctionalization reactions pass through
phenonium ions (Figure 1, left) rather than three- or four-
membered cyclic silyl oxonium ions[10a] (not shown) as
confirmed by deuterium labeling with Et3SiD (14/16![2H]-
15/[2H]-17). The situation changes with aliphatic 18 where
anchimeric assistance by an adjacent aryl group is not
possible; the regioisomers 19 and 20 were formed in an
almost equimolar ratio (18!19 and 20; Scheme 3, bottom).
The defunctionalization also proceeded smoothly in the
presence of an alkene without competing hydrosilylation
(23!24; Scheme 5, top). The cyclic ether in tosylated
tetrahydrofurfuryl alcohol remained untouched (25!26;
Scheme 5, bottom). To highlight the utility of the new
method, we selected a carbohydrate, TBDMS-protected 1,2-
3
ꢀ
deoxy-d-glucose 27, with several delicate C(sp ) O linkages
(Scheme 5, bottom). The standard setup furnished 28, which
was directly converted into 29 by a known procedure[13] (27!
28!29). The unprotected tetrahydropyran 29 was isolated in
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 4
These are not the final page numbers!