1070
X. Fan et al. / Tetrahedron Letters 55 (2014) 1068–1071
Table 2 (continued)
Entry
Ether
Product
t (h)
Yieldb (%)
81
OMe
CN
10
11
12
3.5
1j
2j
C6H13
OMe
C6H13
CN
2k
CN
2l
1.5
1.5
99
67
1k
OMe
1l
OMe
CN
13
14
15
1.5
3.5
1.5
65
82
86
1m
2h
Ph
Ph
1n
2n
2o
Ph
OMe
Ph
CN
Ph
Ph
OMe
CN
1o
OMe
CN
16
17
1.5
36
1p
2p
F
F
OMe
CN
1q
2q
n.r.
n.r.
1.5
1.5
OMe
CN
1r
2r
18
a
Reaction conditions: benzylic ether (0.35 mmol), TMSCN (0.525 mmol), DCM (3 mL), FeCl3 (10 mol %), room temperature, in air.
Isolated yield.
b
(Table 1, entry 16), which suggests that the present reaction is cat-
that the allylic secondary ether 1m ((E)-1-methoxy-1-phenylbut-
2-ene) led cleanly to the thermodynamically favored regioisomeric
cyanation product 2h ((E)-2-methyl-4-phenyl-3-butenenitrile,
Table 2, entry 13), suggesting that the reaction most likely proceeds
via a carbocation pathway. In order to verify this proposal, enantio-
enriched (R)-1a (63% ee) was subjected to this reaction. After 1.5 h,
the racemic cyanation product 2a was obtained in 95% yield. This
result is consistent with our previously mechanistic study in
iron-catalyzed N-alkylation.11 Moreover, benzylic ethyl ether such
as 1-(1-ethoxypropyl)-4-methoxybenzene was also tested to this
reaction. However, only trace amount of cyanation product 2a
was obtained,15 which might be attributed to the difficulty in the
formation of a carbocation intermediate from benzylic ethyl ether.
Based on the above experimental results and previously re-
ported studies by us and Ding group,9a,11 a tentative reaction path-
way was proposed in Scheme 1. We speculate that the reaction
might involve activation of the C–O bond through coordination
of FeCl3 with the methoxy group in ethers. This delivers the
Fe(III)-coordinated species A, which then undergoes an iron-cata-
lyzed sp3 C–O ether bond cleavage to give the benzylic cation
alyzed by Fe(III) species. It is worth noting that FeCl3 has been
demonstrated to be an incompetent catalyst for direct cyanation
of benzylic alcohols,9a,c however, it showed excellent catalytic abil-
ity in our reaction.
With the optimized reaction conditions in hand (Table 1, entry
2), the substrate scope and limitations of this reaction were then
investigated on a series of primary, secondary, and tertiary ben-
zylic and allylic methyl ethers (Table 2). It was found that the pres-
ent reaction conditions showed good substrate compatibility with
a wide variety of secondary and tertiary ethers, providing the cor-
responding nitriles in good to excellent yields within a short period
of reaction time at room temperature (Table 2, entries 1–15). Pri-
mary benzylic and allylic methyl ethers, such as 1q and 1r, could
not undergo this transformation even under harsh reaction condi-
tions (Table 2, entries 17–18),14 presumably because of the lower
stability of the primary carbocation intermediates derived from
1q and 1r. Functional groups, e.g., aromatic fluoro, or methoxy
groups, C@C double bond, cyclopropyl, and acetal, tolerate the
reaction conditions very well (Table 2, entries 2, 4, 5, 7, and 16).
Overall, the reaction is facilitated with electron-rich benzylic
methyl ethers and gives the corresponding cyanation products in
high yield (Table 2, entries 1, 4, and 6). In contrast, the use of elec-
tron-poor benzylic ether 1p, which bear a fluorine atom at the
para-position of the phenyl ring, resulted in a quite lower yield
of product 2p (36%, Table 2, entry 16). With respect to the steric
properties of the ethers, no obvious influence on the reaction
was observed, tertiary benzylic ethers also served as efficient elec-
trophiles in this transformation (Table 2, entries 14–15). For exam-
ple, the higher sterically hindered triphenylmethyl methyl ether
1n reacted smoothly with TMSCN at room temperature to give
the desired tertiary nitrile 2n in high yield (Table 2, entry 14). Note
O
FeCl3
R2
O
Ar
R2
Ar
CN
A
FeCl3
Ar
R2
Ar
R2
CH3OFeCl3
TMSCN
CN
CH3OTMS
FeCl3
B
Scheme 1. Proposed mechanism.