Full Paper
et al. converted benzaldehyde into benzonitrile by using triflic
acid and sodium azide.[15b] Thus, none of the methods involve
transformation via DG intermediacy generated in situ and may
not find application in ortho CÀH functionalisation.
failed to initiate the transformation (Table 1, entries 2–4). Re-
placing either Cu-salt with another oxidant such as Cs(OAc)
and phenyl iododiacetate (PIDA) or DCE with other solvents
such as trifluoroacetic acid (TFA), toluene, N,N-dimethylform-
amide (DMF) or acetonitrile furnished 4a in diminished yields
(entries 5–10). Thus, these findings led us to believe that DCE
was the best solvent for the transformation and that it occurs
via iminophosphorane 2a generated in situ as an intermediate.
This was further confirmed by treating pre-synthesised imino-
phosphorane 2a with Ru and Cu-salt to afford the correspond-
ing benzonitrile 4a in 72% yield (Scheme 1).
Results and Discussion
We commenced our studies by using benzoyl azide and the
unactivated alkene styrene as model substrates to optimize
the coupling conditions. Initially, 1a was treated with triphe-
nylphosphine (TPP) in 1,2-dichloroethane (DCE) at 608C for
30 min to generate iminophosphorane 2a. Styrene 3a was
then added to the reaction mixture, which was then stirred at
808C with the addition of catalyst [Ru(p-cymene)Cl2]2 and an
oxidant Cu(OAc)2·H2O in 5.0 and 100.0 mol%, respectively.
After 12 h, a new product, 4a, was isolated in 93% yield
(Table 1, entry 1). The structure of the resulting compound was
identified as a cyano-containing compound benzonitrile in-
stead the desired ortho CÀH alkenylated derivative 5 (Table 1).
Scheme 1. Transformation of pre-synthesised iminophosphorane 2a into
benzonitrile 4a.
However, ruling out any possible involvement of
benzamide generated in situ, the latter failed to un-
dergo transformation to yield 4a. The scope and limi-
tations of the method were demonstrated by synthe-
sising ten nitrile derivatives from their corresponding
acyl azides (Table 2) with minimal variation in isolated
yields. However, the presence of an electron-with-
drawing group in the aryl ring failed to facilitate con-
version into the corresponding benzonitrile deriva-
Table 1. Transformation of benzoyl azide 1a into benzonitrile 4a.[a]
tives. Similarly, aliphatic acyl azides such as 2-phenyl
acetyl azide also failed to furnish the desired benzo-
nitrile derivative. This may be attributed to the de-
crease in the electron density on the nitrogen of the
iminophosphorane moiety, which, in turn, prevents
its chelation with the metal and subsequent transfor-
mation into benzonitrile derivative.
Entry
Cat.
[5 mol%]
Oxidant
[100 mol%]
Solvent
Yield of 4a
[%][c]
1
2[b]
3[d]
4[e]
5[f]
6
7
8
9
10
[Ru(p-cymene)Cl2]2
–
[Ru(p-cymene)Cl2]2
–
Cu(OAc)2·H2O
–
–
Cu(OAc)2·H2O
–
Cs(OAc)
PhI(OAc)2
Cu(OAc)2·H2O
Cu(OAc)2·H2O
Cu(OAc)2·H2O
DCE
DCE
DCE
DCE
TFA
DCE
DCE
toluene
DMF
CH3CN
93
n.d.
n.d.
n.d.
n.d.
46
63
42
28
36
–
[Ru(p-cymene)Cl2]2
[Ru(p-cymene)Cl2]2
[Ru(p-cymene)Cl2]2
[Ru(p-cymene)Cl2]2
[Ru(p-cymene)Cl2]2
After successfully optimising the transformation,
we then studied the DG-assisted CÀH alkenylation
with activated alkenes. We embarked with the appli-
cation of iminophosphoranes generated in situ from
acyl azides as a DG for ortho CÀH alkenylation with
activated alkenes and subsequent transformation of
the DG into a nitrile group to furnish compounds
based on 6 (Figure 2a). Although there are several re-
ports dealing with oxidative aromatic CÀH alkenyla-
[a] Reaction conditions: 1a (1.02 mmol), 3a (1.02 mmol), catalyst (5.0 mol%), oxidant
(100 mol%), solvent (5 mL), 808C, 12 h. [b] Conducted without PPh3. [c] Yield of isolat-
ed product. n.d.=not detected. [d] Conducted without Cu(OAc)2·H2O. [e] Conducted
without [Ru(p-cymene)Cl2]2. [f] Reaction carried out in TFA for 12 h.
The structure was established base on NMR spectroscopic anal-
ysis and also by comparison with reported data.[16] Although
a survey of the literature revealed a single report involving
transformation of iminophosphoranes through aza-Wittig reac-
tion, in our hands, we failed to observe such a transforma-
tion.[17]
tion by employing numerous DGs[18] (Figure 2b), to our knowl-
edge, there are no reports dealing with the synthesis of com-
pounds based on 6 except for the traditional Mirozoki–Heck
reaction[19] (Figure 2c).
We commenced our studies with the condensation of ben-
zoyl azide (1a) and the activated alkene tert-butyl acrylate (3b)
as model substrates for optimising reaction conditions for the
synthesis of 2-alkenyl benzonitriles 6 (Table 3). We treated 1a
with TPP (100 mol%) in DCE at 608C for 30 min to generate
iminophosphorane 2a in situ. Alkene 3b (1.02 mmol) was then
added and the reaction mixture was stirred at 808C with
the addition of catalyst [Ru(p-cymene)Cl2]2 and oxidant
Our strategy of transformation of azides into nitriles has an
advantage over traditional cross-coupling methods because it
expands the scope of the reaction to include ortho CÀH func-
tionalisation. This prompted us to initially study the transfor-
mation of acyl azides into nitriles. Carrying out the reaction
either in the absence of PPh3 or Ru-catalyst/Cu(OAc)2·H2O
Chem. Eur. J. 2015, 21, 11807 – 11812
11808
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim