S. K. Sahoo / Tetrahedron Letters 57 (2016) 3476–3480
3477
Table 1
prevents the catalyst from interacting with pyridine C(3)AH and
Control experiment for the exact role of reagentsa,b,c
C(4)AH bonds. In 2013, Kanai and co-workers reported a challeng-
ing C4- selective direct alkylation protocol of pyridine substrates
through hydrometalation/nucleophilic addition/rearomatization
sequence catalyzed by Co(II) and hydride sources.6a In addition,
hypervalent iodine mediated oxidative amination to arene,8
alkene,9 and alkyne10 based substrates have been reported. In line
with these observations, a hypothesis to apply pyridine based
molecules for C-4 based amination reaction by using metal free
hypervalent oxidizing reagent has been envisaged (Scheme 1a).
To test the hypothesis on the regioselective C-4 amination of
pyridine scaffolds, N-fluorobenzenesulfonimide (NFSI) and iodo
benzene diacetate (IBDA) were initially employed to react with
Pyridine (Py) at 100 °C using various solvents like DCE, Dioxane,
DMSO, DMF, THF, EtOH, and toluene. Unfortunately, under these
conditions, the desired product was not observed. However grati-
fyingly, in the presence of toluene solvent, an unexpected diben-
zylic product 1a was observed instead of the expected C-4
aminated product 10a (Scheme 1). Following this observation, a
thorough search for the importance of dibenzylic product resulted
in many more of its applications and synthetic utility under robust
and toxic metal catalyzed conditions11 including a side dibenzylic
product reported in metal catalyzed and unfriendly peroxide con-
dition.12 Delightfully, this is an interesting methodology to observe
homo coupling sp3C–sp3C bond formation using different oxidizing
reagents (Scheme 1b).
CH3
CH3
IBDA, NFSI
Py, 100 0C, 15 h
H3C
H3C
(2)
(2a)
Entry
NFSI (mmol)
IBDA (mmol)
Py (mmol)
NMR yield (%)
1
2
3
4
5
6
7
1
1
1
X
1
X
X
2
2
X
1
X
1
X
1
X
1
1
X
X
1
72
02
01
01
01
01
01
X = Not used.
Bold refer to better yield compare to other method.
a
p-Xylene (2) 30 mmol, 100 °C, 15 h.
Yield was calculated comparing the amount of NFSI used.
Tetra chloro ethane (TCE) internal standard for NMR yield.
b
c
CH3
2% Grubbs 1
DCM, 40 oC
Overnight H3C
H3C
(2')
(2'a) 30% yield
The reaction conditions were then implemented to p-xylene (2)
substrate to generalize whether the reaction promotes the benzylic
homo coupling dimerized product. Gratifyingly, almost major
amount of bibenzylic product 2a was observed along with unchar-
acterized trace amount of side product (Table 1, entry 1). For cross
verification, the formation of bibenzylic product was further con-
firmed by the reduction of (E)-1,2 dip-tolylethene 20a substrate
H2 Baloon
Pd on activated
carbon 5 wt%
EtOAc (0.5 mL), RT
Over night
CH3
(Scheme 2).13
.
H3C
A controlled experiment was performed to ascertain the exact
role of all NFSI, IBDA, and Py reagents on this reaction condition
which confirmed that all the three abovementioned reagents are
playing vital role in this reaction (Table 1). The reaction was fur-
ther standardized by varying the stoichiometry of the three
reagents. It is found that combination of 1 mmol NFSI, 2 mmol
IBDA, and 1 mmol Py treated with p-xylene (2) gave better yield
(NMR yield 72%, isolated yield 70%) compared to other conditions
(see Supporting information (SI, Table S2.1)). It is worth mention-
ing that the yield of benzylic dimerized product (2a) is calculated
compared to the amount of NFSI as oxidizing reagent used in the
corresponding reaction condition. In this reaction condition, excess
amount of p-xylene (2) was used (for the role of both solvent and
substrate); hence NFSI was used as a limiting reagent. The reaction
condition was tested with various solvents (see SI Table S2.2) to
decrease the amount of p-xylene (2). Unfortunately, there was no
improvement in the yield of the product. To find out the role of
IBDA, the reaction was tested with different hypervalent reagents
and oxidants (see SI Table S2.3), and it was concluded that IBDA
(2a) 92% yield
Scheme 2. Indirect synthesis of dibenzylic product 2a.
is a better oxidant under the present reaction condition. Further,
the reaction was treated with various fluorinating, halogenated,
and aminating reagents (see SI Table S2.4) and it was observed that
NFSI is a more efficient reagent for the formation of the corre-
sponding dimerized product 2a. Further, Py proved to be a better
supporting reagent for the formation of dimerized product 2a com-
pared to other pyridine derivative reagents (see SI Table S2.5).
Subsequently, it was found that 100 °C is the optimum temper-
ature for this reaction (see SI Table S2.6). Likewise, the reaction
was optimized with different time interval and found that 15 h is
required (see SI Table S2.7) to furnish better yield of the product
2a. Another noteworthy observation was that only 30 equiv
(15 equiv of dimer product compare to NFSI) of the substrate 2a
was required to produce the expected dimerized product 2a in
quantitative yield (see SI Table S2.8).
Having acquired this decent optimization condition, the sub-
strate scope was tested with a variety of toluene derivatives
(Table 2). The substrates m-xylene (3) and o-xylene (4) also gave
their corresponding homo coupled sp3C–sp3C dimerized products
3a and 4a in good yields. An interesting observation in case of sub-
strate mesitylene (5) was the coupling occurred at only one methyl
position to furnish the dimerized product 5a while the other two
methyl substituents remained intact. The substrate 4-tert-butyl
toluene (6) showed well tolerance under this condition giving
moderate yield of the corresponding homo coupled dimerized
product 6a. Most importantly, the compound 4-methylanisole (7)
having two possible positions for radical formation, one at
IBDA
OAc
OAc
a)
b)
+
NFSI
NR2
1'a
IBDA = Ph
I
Toluene
(solvent)
N
(Py)
N
(
)
C-4 amination
SO2Ph
SO2Ph
CH3
NFSI =
F
N
IBDA, NFSI
Py
1a
(
)
1
(
)
Scheme 1. (a) Working hypothesis (sp2CAN bond formation). (b) Reaction obser-
vation (metal free sp3C–sp3C bond formation).