D. Wang et al. / Tetrahedron Letters 55 (2014) 7130–7132
7131
dichloromethane gave both the highest yield (53%) and regioselec-
tivity (6:1, entry 6). We then investigated the use of oxalyl chloride
and DMF, which is an alternative way to generate the desired
Vilsmeier reagent. However, a decreased yield and regioselectivity
were detected (entry 10).
Nu
A-X
A.
R
R
R
N+
O
N+
O-
X-
N
Nu
6
5
1
A
POCl3, DMF(cat.)
It was hypothesized that the regioselectivity arose from prior
coordination to the N-oxide and delivery of the chloride anion
intramolecularly through a quaternary ammonium salt intermedi-
ate (please see the detailed mechanism as presented in Scheme 3).
Therefore, we proposed that running the reaction at lower concen-
tration would improve the regioselectivity. As expected, more
dilute condition (0.1 M) led to 60% isolated yield of the desired
compound with 10:1 regioselectivity (entry 11). Slower conversion
and decreased yields were observed when more dilute conditions
were applied (entries 12 & 13). Since we proposed that DMF serves
as a catalyst in the reaction, less equivalent of DMF was investi-
gated. We were pleased to find out that half equiv of DMF gave
the highest yield (73%, entry 14), which is also the optimized
reaction condition we found. Less equiv of DMF resulted in slower
conversion and lower yields (entries 15 & 16).
B.
R
R
N+
O-
DCM, 0oC~r.t.
(43%~96%)
N
2
Cl
1
Scheme 2. (A) Functionalization of N-oxides via activating agent/nucleophile. (B)
Novel method for mild and efficient chlorination.
chlorination of heteroarenes utilizing the method.7 Those reported
methods have not found wide applications, most likely due to the
side reactions associated. Since most of the reaction side products
were associated with the activating agent, we focused on searching
for suitable alternatives. Taking advantage of the same strategy but
combining the activating agent and the chloride anion, we hoped
to obtain the C2-chlorinated heteroarenes while minimizing the
aforementioned competing side reactions.
With optimized conditions in hand, the scope of the chlorination
As all of the known activating reagents belong to electrophiles,
searching for good electrophiles with chloride source became our
target. To assess the feasibility, we attempted the chlorination of
quinoline N-oxide 7 (Table 1). Unfortunately, none of the regular
reagents gave satisfactory result. No desired product was detected
for SOCl2, MsCl, or TsCl, while slow conversion and low yield was
observed for POCl3 (entry 1). Vilsmeier reagent, a good electrophile
with chloride counteranion, came to our attention then. Although
it is well known that Vilsmeier reagent reacts with electron-rich
aromatic or heteroaromatic compounds to give aldehydes,8
N-oxides should be more reactive than the remaining aromatic
parts of the molecules. When a solution of 7 was treated with com-
mercially available Vilsmeier reagent, we were delighted to obtain
8 as the major product of the reaction, isolated in 55% yield (entry
5). Although the selectivity is not great (C2:C4 ꢀ 7:1), no aldehyde
was detected. Considering Vilsmeier reagent as expensive and
highly hygroscopic, we attempted synthesizing it in situ. Gratify-
ingly, when a solution of the substrate and POCl3 was treated with
DMF, generating Vilsmeier reagent in situ, a similar result was
observed (entries 6 vs 5). A solvent screen demonstrated that
with
a range of quinoline and isoquinoline derivatives was
examined. We were pleased to observe modest to excellent yields
in all cases examined, and with excellent regioselectivity for
most cases (Figs. 1 and 2). Both electron-rich and electron-poor
substrates showed good reactivity. In the case of halo-substituted
derivatives no interconversion with the existing halogen was
observed. It is worth noting that little conversion was observed
for 9j (to 10j) and 11h (to 12h), indicating that the reaction is highly
selective for the desired C2-chlorination.
The proposed mechanism for this transformation is outlined in
Scheme 3. Prior coordination of Vilsmeier reagent to the N-oxide 9
afforded the activated quinoline complex 13. Nucleophilic attack of
the chloride anion intramolecularly gave a quaternary ammonium
salt intermediate 14. Subsequent rearomatization of 14 affords the
desired 2-chloroarenes 10 and DMF, which proceeds into the next
catalytic cycle. In this transformation, Vilsmeier reagent functions
as both the activating agent and the external chloride source,
delivering the chloride at C2 regioselectively.
Another remarkable feature of our developed method is
scalability. The chlorination of 7 was performed on a ten gram
scale with similar regioselectivity and in 86% isolated yield (see
Supplementary material for details). However, the extension of
the method to the chlorination of simple substituted pyridine-
N-oxides has not been successful despite our best efforts.
Since both the oxidation and chlorination reactions were
conducted under acidic conditions, the potential extension of this
method to a one-pot process was explored. Employing quinoline
Table 1
Reaction optimization for the 2-chlorination of quinoline-N-oxides
conditions
r.t.
N
Cl
N
O
8
7
Entry
Additive
Solvent
[M]
Equiva
Ratiob
Yield
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
POCl3
SOCl2
MsCl
DCM
DCM
DCM
DCM
DCM
DCM
THF
Et2O
CH3CN
DCM
DCM
DCM
DCM
DCM
DCM
DCM
0.1
0.1
0.1
0.1
0
0
0
0
3:1
36%c
N/A
N/A
N/A
55%
53%
30%
51%
48%
45%
60%
32%c
47%c
73%
63%c
40%c
O
POCl3
N/A
N/A
N/A
7:1
6:1
4:1
5:1
3:1
4:1
10:1
10:1
12:1
11:1
12:1
11:1
N
H
TsCl
R
Vilsmeier
POCl3
POCl3
POCl3
POCl3
(COCl)2
POCl3
POCl3
POCl3
POCl3
POCl3
POCl3
0.1
0
N
10
Cl
0.28
0.28
0.28
0.28
0.28
0.1
0.05
0.02
0.1
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.5
0.3
0.1
Cl
H
N+
Cl-
R
R
N
O
Cl
H
N+
O-
R
14
N+
O
9
0.1
0.1
Cl-
N+
13
Cl
Cl-
a
b
c
The equiv of DMF.
Ratio based on the crude products of 1H NMR.
Still some SM remained after stirring for one day.
N
Scheme 3. Proposed reaction mechanism.