Regioselectivity in Styrene Hydrocyanation
FULL PAPER
as a Lewis acid co-catalyst. Experiments with deuterated
styrene as substrate gave new insights into the mechanism
of the reaction. The scrambling of the hydrogen (the deute-
rium labeling experiments were performed by adding HCN
to the deuterated substrate) in the nitrile products and in
spectra were recorded on an Avatar 360 FTIR instrument in attenuated
total reflectance mode.
Caution! HCN is a highly toxic, volatile liquid (b.p. 278C) that is also
susceptible to explosive polymerization in the presence of base catalysts.
It should be handled only in a well-ventilated fume hood by teams of at
least two technically qualified persons who have received appropriate
medical training for treating HCN poisoning. Sensible precautions in-
clude having available proper first aid equipment as well as HCN moni-
tors. Uninhibited HCN should be stored at a temperature lower than its
melting point (ꢀ138C). Excess HCN may be disposed of by addition to
aqueous sodium hypochlorite, which converts the cyanide to cyanate.
3
the unreacted styrene showed the formation of h -benzyl
and s-alkyl intermediates during the catalytic cycle and a
fast equilibrium between the two species with the styrene
substrate. Furthermore, no CꢀCN bond cleavage was ob-
served for the nitrile products during the isomerization ex-
periments, which suggests the irreversibility of the elimina-
tion step.
Coordination of styrene to AlCl
120 mg) were dissolved in [D ]benzene (1 mL). The solution was stirred
6
for 2 h and analyzed by NMR spectroscopy. H NMR (400 MHz, C D ,
3 3
ACHTUNGETRNNUNG( 1d): Styrene (50 mL) and AlCl
(
6
1
6
2
5
58C, TMS) d=7.21–7.17 (m, 2H), 7.08–6.98 (m, 3H), 6.53 (dd, 1H),
.56 (dd, 1H), 5.03 ppm (dd, 1H); C NMR (500 MHz, C D , 258C,
6 6
DFT provided a molecular-level picture of the catalytic
process considered and allowed an in-depth rationalization
of the experimental observations. It was found that the con-
ventional thermodynamic cycle could not be used for the
correct description of the Ni-catalyzed hydrocyanation reac-
tion, because it lacked important mechanistic details. Forma-
tion of the molecular complex by coordination of a nitrile
with the Ni catalyst, as an intermediate product of the re-
ductive elimination step, is essential for the overall catalytic
performance.
1
3
TMS) d=137.58, 136.95, 128.39, 126.19, 113.27 ppm.
Coordination of 2-phenylpropionitrile to AlCl (2b): 2-Phenylpropioni-
trile (50 mL) and AlCl (120 mg) were dissolved in [D ]benzene (1 mL).
The solution was stirred for 2 h and analyzed by NMR spectroscopy.
3
ACHTUNGTRENNUNG
3
6
1
6 6
H NMR (400 MHz, C D , 258C, TMS) d=7.50–7.47 (m, 3H), 7.46–7.34
1
3
(
m, 2H), 4.49–4.47 (m, 1H), 1.91–1.90 ppm (m, 3H);
400 MHz, C , 258C, TMS) d=149.5, 131.6, 130.2, 129.9, 126.9, 123.9,
9.8, 31.6, 19.7 ppm.
Coordination of bippp to AlCl
AlCl (2.4 mg) were dissolved in [D
stirred for 2 h and analyzed by NMR spectroscopy. H NMR (400 MHz,
CD , 258C, TMS) d=7.51–6.87 (m, 7H), 3.94–2.94 (m, 1H), 1.40–
C NMR
(
9
6
D
6
3
(7): Bippp (16 mg, 0.018 mmol) and
]toluene (1 mL). The solution was
3
8
1
Selectivity patterns were controlled by the relative stabili-
3
C
0
1
6
D
5
3
ties of the h -benzyl and s-alkyl intermediates and the inter-
3
1
.89 ppm (m, 6H); P NMR (162 MHz, C
61.1 ppm (s).
6
D
5
CD
3
, 258C, H
3
PO
4
) d=
mediate products of their decomposition. The degree of sta-
bilization of these species in the presence of the Lewis acid
co-catalyst was not uniform. This determined the qualitative
shifts in the energetics of the competing reaction paths and
altered the regioselectivity of the hydrocyanation reaction.
Thus, we conclude that the selectivity towards the linear
Coordination of [Ni
.018 mmol) and bippp (16 mg, 0.018 mmol) were dissolved in
]toluene (1 mL). AlCl (2.4 mg) was added and the solution was
stirred for 2 h and analyzed by NMR spectroscopy. P NMR (400 MHz,
CD , 258C, H PO ) d=146.0 (s, [Ni(bippp)(cod)]), 126. 0 ppm (d,
J=19.4 Hz).
General procedure for the hydrocyanation experiments: A solution of
the ligand (1.2 equiv) in solvent (1 mL) was added to [Ni(cod) ] (9.0 mg,
.033 mmol). Styrene (20 equiv) was added with an Eppendorf pipette,
A
H
U
G
R
N
U
(bippp)
A
H
U
G
R
N
U
G
3
(8): [Ni
A
H
U
T
E
N
N
2
(5 mg,
0
[
D
8
3
3
1
C
6
D
5
3
3
4
A
H
N
T
E
N
N
ACHTUNGTRENNUNG
product 3-phenylpropionitrile in the presence of AlCl is
3
3
due to the higher stability of the intermediate h -benzyl
A
H
U
G
R
N
U
G
2
complex. The selective stabilization of this intermediate in
0
the presence of the Lewis acid leads to the formation of a
followed by n-decane (50 mL) as internal standard and AlCl (1.05 equiv).
The solution was transferred into a 15 mL Schlenk tube equipped with a
Teflon-coated stirring bar.
3
3
“steady state” for the h -benzyl intermediate and indirectly
promotes the formation of the linear product 3-phenylpro-
pionitrile via the s-alkyl intermediate. This proposition was
supported by DFT calculations as well as by experimental
data concerning the conversion and selectivity in styrene hy-
drocyanation at different reaction times and temperatures.
Method A: An excess of HCN was added with an Eppendorf pipette and
the Schlenk tube was warmed to 908C in an oil bath. The mixture was
stirred for 16 h.
Method B: A round-bottomed Schlenk flask was filled with solvent
ꢀ1
(
1 mL) and an excess of HCN (13 mmolmin ), which was collected in a
Different Lewis acids, such as FeCl and CuCN, were ap-
plied in the reaction, which led to lower selectivity and ac-
5 mL syringe and added to the reaction mixture by a syringe pump
during 3 h (closed system). The mixture was stirred for another 2 h.
3
The reaction product was cooled to 08C and flushed with a gentle stream
of argon for 1 min to remove traces of HCN. Samples were analyzed by
GC with n-decane as internal standard. All the reactions were carried
out in duplicate, and showed variabilities for conversion and selectivity
of ꢁ2 and ꢁ1%, respectively.
tivity compared to AlCl . This deeper understanding of the
3
mechanism of the hydrocyanation reaction might also lead
in the future to improved catalysts for the hydrocyanation
of simple alkenes.
[
21]
Computational details: DFT with the B3LYP hybrid exchange-correla-
tion functional was used for the quantum chemical calculations. This
[
22]
method has recently been used successfully for the description of sys-
tems analogous to those considered herein. In addition, our preliminary
calculations (see Supporting Information for the results of the accuracy
tests for the ab initio wave-function-based method and several popular
exchange-correlation functionals) show a very good accuracy of the
Experimental Section
General considerations: Chemicals were purchased from Aldrich, Acros,
and Merck and used as received. Styrene was filtered over alumina neu-
3
method applied to the description of AlCl coordination to nitrile com-
tral and distilled over CaH
preparations were carried out under an argon atmosphere using standard
2
prior to use and was stored at ꢀ308C. All
pounds, which is crucial for the co-catalyst behavior of the Lewis acid in
the hydrocyanation reaction. Full geometry optimizations and saddle-
point searches were all performed using the Gaussian 03 program. The
[19]
[20]
[23]
Schlenk techniques. [Ni
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(cod)
2
]
and HCN were synthesized according
to literature procedures. NMR spectra were recorded on a Mercury 400
and a Varian Unity Inova 500 spectrometer ( H, C{ H}, P{ H}). IR
full-electron 6-31+G(d) basis set was used for the nickel ion, whereas all
1
13
1
31
1
atoms of the styrene molecule, HCN, and Lewis acids (AlCl
3
and CuCN)
Chem. Eur. J. 2009, 15, 8768 – 8778
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8777