Efficient nickel catalyst for coupling of acetonitrile with aldehydes

Article abstract of DOI:10.1039/b505778g

A Ni complex of a diarylamido-based PNP ligand is an efficient and robust catalyst for coupling of acetonitrile with aldehydes. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b505778g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Efficient nickel catalyst for coupling of acetonitrile with aldehydes{
Lei Fan and Oleg. V. Ozerov*
Received (in Berkeley, CA, USA) 24th April 2005, Accepted 9th July 2005
First published as an Advance Article on the web 8th August 2005
DOI: 10.1039/b505778g
preferentially via associative pathways.10 Coupling of acetonitrile
A Ni complex of a diarylamido-based PNP ligand is an efficient
with aldehydes seemed a suitable test application because it is
viewed as challenging even in its racemic version.3–5 The utility of
metal complexes of neutral tridentate ligands such as the ‘‘pybox’’
family as Lewis acid catalysts has been firmly established.11
The explorations of the catalytic applications of anionic pincer
ligands (primarily of the PCP type) have been focused elsewhere.12
We have surveyed (Table 1) three compounds bearing the PNP
ligand (3-Ni, 3-Pd, and 3-Pt) as catalysts for coupling of
p-fluorobenzaldehyde (4a) with acetonitrile. For all three cases,
we have found that triflate in 3 is easily displaced by MeCN upon
dissolution in acetonitrile and so a truly three-coordinate precursor
[(PNP)M]+ is not necessary. We chose p-fluorobenzaldehyde (4a)
as our test substrate because of the convenience of the 19F NMR
spectroscopic probe and because the p-F substituent is ‘‘average’’
(e.g., similar to H) in terms of its electronic effect.13
and robust catalyst for coupling of acetonitrile with aldehydes.
Coupling of carbon-based nucleophiles with the carbonyl electro-
philes is one of the most widely used strategies for carbon–carbon
bond formation. Utilization of a-deprotonated nitriles as nucleo-
philes is an attractive method for the synthesis of b-cyanoalcohols,
useful building blocks in synthesis.1–5 However, the scope of
nitrile-derived nucleophiles is usually limited to those arising from
nitriles with enhanced C–H acidity, such as a-arylnitriles and
a-acylnitriles.1,2 Catalytic utilization of the less acidic simple
alkylnitriles has been problematic. A strong base that would be
required for direct deprotonation1 may not be compatible with
pendant functional groups and may render the process less
economical. Shibasaki et al. recently explored coordination to a
Lewis acidic transition metal complex as a way to activate
acetonitrile towards deprotonation by a mild base.3 The screening
of several candidate Ru, Cu, and Pd complexes yielded
[CpRu(PPh3)(NCMe)2]PF6 as the catalyst of choice. Other recent
approaches include the use of CuOBut and a strong neutral base
proazaphosphatrane as catalysts.4,5
Our group has been exploring the transition metal chemistry of
the diarylamido-based PNP pincer ligands.6,7 These ligands are an
excellent fit for group 10 metals where they form extremely robust
Scheme 2
(PNP)MX (M
5 Ni, Pd, Pt) square planar compounds
(Scheme 1).8 Compounds 3 were synthesized by adaptation of
the previously reported methods. Amido is a weak trans-influence
ligand and this should be reflected in strong binding of 2-electron
donors (L or X2) in a position trans to N. For example, the Pd–Cl
Table 1 Effect of different Lewis acids, bases, temperature and
additives on the coupling of CH3CN with 4-fluorobenzaldehydea
#
Catalyst
T/uC Base/mol%
Time Yield
Noneb
None
None
None
60
60
45
45
100 DBUc
100 DBU
100d DBUH+
100d DBUH+
100 DBU
100 DBU
100 DBU
100 DBU
100 DBU
100 DBU
100 DBU
10 DBU
18 h 13%
18 h 10%
24 h 0%
24 h 0%
bond in (PNP)PdCl (1-Pd)8 is ca. 0.06–0.12 A shorter than in
1
2
3
4
˚
related (PCP)PdCl complexes where Cl is trans to a strong trans-
influence aryl donor.9 We surmised that a cationic [(PNP)M]+
fragment should behave as a potent Lewis acid. Strong binding of
a substrate in the fourth coordination site of a square planar
complex would not prohibit ligand exchange because ligand
exchange in square planar d8 16-electron complexes occurs
+
5
6
7
5% 3-Ni
5% 3-Nib
5% 3-Pd
60
60
50
18 h 81%
18 h 81%
22 h 11%
14 h 7%
22 h 7%
14 h 5%
22 h 55%
22 h 65%
22 h 75%
22 h 90%
22 h 95%
22 h 0%
24 h 0%
8e 5% 3-Pd + 5% NaBArF 45
4
9
5% 3-Pt
50
10e 5% 3-Pt + 5% NaBArF 45
4
11 5% 3-Ni
12 5% 3-Ni
13 5% 3-Ni
14 5% 3-Ni
15 5% 3-Ni
16 5% 3-Ni
17 5% 3-Ni
50
50
50
50
50
50
45
20 DBU
50 DBU
100 DBU
200 DBU
100 NEt3
100 proton
sponge
18 5% 3-Ni
45
100 DBU
100 DBU
24 h 88%
24 h 90%
Scheme 1
19d 5% 3-Ni + 5% NaBArF 45
4
a
Reaction was carried out with 0.20 mmol of aldehyde in 0.5 mL of
CH3CN (9.6 mmol). Yield calculated in situ using 19F NMR (single
Department of Chemistry, MS 015, Brandeis University, 415 South
Street, Waltham, Massachusetts, USA. E-mail: ozerov@brandeis.edu
{ Electronic supplementary information (ESI) available: characterization
data and experimental details. See http://dx.doi.org/10.1039/b505778g
b
c
runs).
4A molecular sieves.
Triflate counterion. Ar 5 3,5-(CF3)2C6H3.
1,8-Diazabicyclo[5.4.0]undecene.
d
e
4450 | Chem. Commun., 2005, 4450–4452
This journal is ß The Royal Society of Chemistry 2005

View Article Online
aldehydes – 2,6-dichlorobenzaldehyde is one of the most reactive
ones, while o-tolualdehyde was in fact more reactive than
p-tolualdehyde.Analkenylaldehyde,(E)-cinnamaldehyde,exhibited
reactivity comparable to that of p-tolualdehyde. Acetone and
cyclohexenone were unreactive, as was 1-octanal. However,
i-butyraldehyde produced the corresponding b-cyanoalcohol in a
90% yield, albeit rather slowly. The reason behind the higher reac-
tivity of i-butyraldehyde compared with 1-octanal is not clear at this
time. In all reactions except entries 23 and 24 (8% and 10% side-
products), the amount of side-products (see ESI{) was 3% or less.
For some substrates, the reaction did not go to completion even
after extended periods of time. We found that the increase in the
yield of 5 can be achieved by decreasing the volume of CH3CN
used (Table 3), albeit at a cost of the increase in the amount of
side-product (see ESI{).
Scheme 3
The following conclusions can be made from the results
presented in Table 1:
1) Compounds 3-Pd and 3-Pt are not competent as catalysts.
2) Addition of molecular sieves is not necessary.
3) The choice of DBU as base is crucial. Larger concentration of
DBU enhances the rate.
4) 45–50 uC is the optimal temperature range.
5) Addition of Na[B(C6H3(CF3)2)4] has no effect on the
reaction.
6) The reaction is not merely catalyzed by the Brønsted acids or
bases and 3-Ni is critical for selective catalysis. DBU catalyzes a
slow and unselective ‘‘background’’ reaction (see ESI{).
The study was then extended to a number of other aldehydes
using 3-Ni as the catalyst (Table 2). We find that in general
electron-poor and ‘‘average’’ aromatic aldehydes react at . 90%
conversion in , 24 h to give the desired b-cyanoalcohols with
little or no side products. The more reactive aldehydes are
converted to the products in high yield even at 22 uC. The electron-
rich aldehydes (bearing Me and p-MeO substituents) reacted
more slowly. On the other hand, steric factors appear to play
Initially we used acetonitrile that was rigorously dried by
distillation from CaH2. The components used in this reaction were
stored in a glovebox over molecular sieves for reasons of
convenience (and/or out of habit). However, utilization of the
‘‘off the shelf’’ acetonitrile as is did not result in an experimentally
significant change in conversion (Table 2, entry 16). As well, for
several reactions the components were combined in the air without
apparent detriment. 3-Ni is stable in the air.
We performed a series of kinetic experiments14 in an attempt to
elucidate the mechanism of the catalysis using p-FC6H4CHO as a
convenient reporter (via its 19F NMR resonance) substrate. In
contrast to [CpRu(PPh3)(NCMe)2]PF6 as catalyst, the isotope
effect was found to be close to unity (k(CH3CN)/k(CD3CN) 5
1.13(9)). We undertook a study of the initial rates as a function of
the concentrations of the relevant components. The rate of the
aldehyde consumption was found to be first order in [Ni], and of
order 0.46(10) in DBU. The rate dependence on [p-FC6H4CHO]
appeared to be negligible within the experimental error.15 It seems
to us appropriate to remark, however, that the same kinetics may
not necessarily apply to all substrates and conditions.
at most
a limited role in the reactivity of aromatic
Table 2 Coupling of CH3CN with various aldehydes catalyzed by
3-Ni
#
Aldehyde, R 5
DBU/mol% T/uC Time/h Yield (%)b
1
2
3
4
5
6
7
8
9
p-MeC6H4 (1a)
100
100
5
100
5
100
100
100
5
100
5
100
5
100
5
100
100
5
100
100
5
45
45
45
45
45
45
45
RT
45
45
45
45
45
45
45
45
45
45
45
RT
45
45
45
45
RT
45
45
45
24
24
39
24
39
24
24
24
16
24
47
24
47
24
39
22
24
23
24
24
23
24
42
42
16
23
72
72
75
70e
35
40
25
p-MeC6H4– (1a)
p-MeC6H4– (1a)
p-MeOC6H4 (1b)
p-MeOC6H4 (1b)
p-MeOC6H4 (1b)
2,4-Cl2C6H4 (1c)
2,4- Cl2C6H4 (1c)
2,4- Cl2C6H4 (1c)
42e
.95
95c
.95
90
Although we cannot definitively rule out all other mechanistic
possibilities, we tentatively propose a mechanism (Scheme 4)
related to that delineated by Shibasaki et al.3 The only species
observed by 31P NMR at the outset of the catalytic reaction (and
the major component at the end of the reaction) is the complex 6.
The exchange between 6 and 7 is slow on the NMR timescale as
evinced by the observation of two distinct 31P NMR resonances in
10 o-MeC6H4 (1d)
11 o-MeC6H4 (1d)
12 Ph (1e)
66
76
80
88
13 Ph (1e)
14 p-FC6H4 (1f)
15 p-FC6H4 (1f)
16 p-FC6H4 (1f)
17 p-CF3C6H4 (1g)
18 p-CF3C6H4 (1g)
19 2-furyl (1h)
20 2-furyl (1h)
21 2-furyl (1h)
22 n-C7H16 (1i)
23 (E)-PhCHCH (1j)
24 (E)-PhCHCH (1j)
75
90d
.95
.95
.95
89c
.95
traces
46
Table 3 Coupling of CH3CN with selected aldehydes in a smaller
volume of CH3CN catalyzed by 3-Nia
Conversion Product
(%)b,d
#
Aldehyde, R 5
3-Ni (%)b,c
Yieldb,e
100
5
100
1
2
3
4
5
6
7
8
a
p-MeC6H4 (1a)
p-MeOC6H4 (1b)
p-MeOC6H4 (1b)
o-MeC6H4 (1d)
Ph (1e)
5%
5%
10%
5% .95
5% .95
5% .95
95
70
82
90
68
70
92
90
89
75
80
5%
,3%
12%
6%
6%
9%
73
95
95
52
25 p-MeO2CC6H4 (1k) 100
26 2,6-Cl2C6H4 (1l) 100
27 2,6-Me2C6H4 (1m) 100
28 Isopropyl (1n) 100
p-FC6H4 (1f)
(E)-PhCHCH (1j)
90
10% 94
2,6-Me2C6H4 (1m) 10% .95
19%
15%
a
Reaction was carried out on a 0.20 mmol scale of aldehyde with
5% (MePNPi-Pr)NiOTf in 0.5 mL CH3CN (9.6 mmol) unless noted
otherwise. NMR yield (single runs) calculated using 1,4-dioxane as
0.20 mmol aldehyde in 0.1 mL CH3CN with 0.20 mmol of DBU at
b
b
45 uC for 24 h. Yields measured by NMR versus the internal
integration standard (1,4-dioxane). As 100% - fraction of starting
aldehyde. NMR yield of 5. NMR yield of the side product(s)
(see ESI).
c
d
c
an internal integration standard. Isolated yield. CH3CN ‘‘off the
shelf’’. Reaction was performed with 5% (MePNPi-Pr)NiOTf and 5%
e
d
e
NaBArF
.
4
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 4450–4452 | 4451

View Article Online
Table 4 Coupling of CH3CN with 2-furaldehyde catalyzed by 3-Ni at
low loading
#
T/uC
DBU
Time/h
3-Ni
Yielda
TON
1
2
3
a
45
45
45
5%
100%
100%
23
23
48
0.1%
0.1%
0.5%
36%
61%
81%b
360
610
162
b
NMR yield of 5 (single runs). Conversion is 100%, 19% side
products (see ESI).
In summary, we report a robust and easy to handle catalyst for
coupling of aldehydes with acetonitrile. The catalysis proceeds
under mild conditions and is applicable to a broad spectrum of
aldehydes. While pincer complexes of Pd have been applied to a
number processes such as the Heck reaction,12 pincer complexes of
Ni have not found as much application so far despite the fact that
Ni is cheap and environmentally benign. The present work thus
represents a surprisingly rare example of catalytic utilization of a
pincer-ligated Ni complex.
Scheme 4
The authors are grateful to Brandeis University and Research
Corporation for support.
Scheme 5
Notes and references
1 S. Arseniyadis, K. S. Kyler and D. S. Watt, Org. React., 1984, 31, 1.
2 K. Motokura, D. Nishimura, K. Mori, T. Mizugaki, K. Ebitani and
K. Kaneda, J. Am. Chem. Soc., 2004, 126, 5662; S.-I. Murahashi,
H. Takaya and T. Naota, Pure Appl. Chem., 2002, 74, 19; H. Takaya,
T. Naota and S.-I. Murahashi, J. Am. Chem. Soc., 1998, 120, 4244;
Y. Yamamoto, Y. Kubota, Y. Honda, H. Fukui, N. Asao and
H. Nemoto, J. Am. Chem. Soc., 1994, 116, 3161; M. Sawamura,
H. Hamashima and Y. Ito, J. Am. Chem. Soc., 1992, 114, 8295.
3 N. Kumagai, S. Matsunaga and M. Shibasaki, J. Am. Chem. Soc., 2004,
126, 13632.
an independently prepared mixture of 6 and 7. Thus the
concentration of 7 in the catalytic mixture can only be very small.
It seems reasonable at this juncture to propose that the
displacement of DBU by MeCN is the rate-limiting step or at
least a large contributor to the rate limiting step by way of pre-
equilibrium. This is consistent with the observed isotope effect of
near-unity and with the effective lack of dependence of the rate on
the concentration of [p-FC6H4CHO]. The partial order in [DBU] is
more difficult to rationalize. In the proposed mechanism DBU
impedes the formation of 7, but is necessary for the deprotonation
of 7. It is possible that these two contributions are responsible for
the apparent partial positive order in [DBU].
4 Y. Suto, N. Kumagai, S. Matsunaga, M. Kanai and M. Shibasaki, Org.
Lett., 2003, 5, 3147.
5 P. Kisanga, D. McLeod, B. D’Sa and J. Verkade, J. Org. Chem., 1999,
64, 3090.
6 R. C¸ elenligil-C¸ etin, L. A. Watson, C. Guo, B. M. Foxman and
O. V. Ozerov, Organometallics, 2005, 24, 186 and references within.
7 Other groups have been utilizing similar ligands as well: B. C. Bailey,
J. C. Huffman, D. J. Mindiola, W. Weng and O. V. Ozerov,
Organometallics, 2005, 24, 1390; M.-H. Huang and L.-C. Liang,
Organometallics, 2004, 23, 2813; A. M. Winter, K. Eichele, H.-G. Mack,
S. Potuznik, H. A. Mayer and W. C. Kaska, J. Organomet. Chem.,
2003, 682, 149; S. B. Harkins and J. C. Peters, J. Am. Chem. Soc., 2005,
127, 2030.
8 O. V. Ozerov, C. Guo, L. Fan and B. M. Foxman, Organometallics,
2004, 23, 5573; L. Fan, L. Yang, C. Guo, B. M. Foxman and
O. V. Ozerov, Organometallics, 2004, 23, 4778; L. Fan, B. M. Foxman
and O. V. Ozerov, Organometallics, 2004, 23, 326.
9 D. Morales-Morales, C. Grause, K. Kasaoka, R. Redon, R. E. Cramer
and C. M. Jensen, Inorg. Chim. Acta, 2000, 300–302, 958; R. J. Cross,
A. R. Kennedy and K. W. Muir, J. Organomet. Chem., 1995, 487, 227.
10 J. P. Collman, L. S. Hegedus, J. R. Norton, and R. G Finke, in
Principles and Applications of Organotransition Metal Chemistry,
University Science: Mill Valley, CA, 1987, p. 241.
11 G. Desimoni, G. Faita and P. Quadrelli, Chem. Rev., 2003, 103, 3119.
12 M. E. van der Boom and D. Milstein, Chem. Rev., 2003, 103, 1759.
13 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165.
14 See Electronic Supplementary Information for details.
15 Interestingly, Shibasaki et al.3 reported that the rate of the reaction was
also of zeroth order in aldehyde, but first order in DBU and partial
order in the metal catalyst.
Compound 8 would thus be a key catalytic intermediate.
However, attempts at its independent preparation resulted
(Scheme 5) only in the formation of the C-bound isomer 10
(coupling to two 31P nuclei is exhibited by both the 13C and the 1H
nuclei of the Ni–CH2 moiety). 10 does not catalyze the title
reaction, does not react with (DBU)H+, is not formed in a separate
reaction between 7 and DBU, and is not observed in the
catalytically active reaction mixtures. Perhaps the isomer 8 is
stabilized in some way in the reaction mixture via hydrogen
bonding with (DBU)H+. It is also possible that the reaction of 8
with an aldehyde is much faster than the isomerization to 10.16
In contrast to the Ru catalyst,3 we do not observe unselective
decomposition of 3-Ni in the catalytic mixture. Experiments with
low catalyst loading (Table 4) show that high turnover numbers
can be achieved, although at low Ni loading and high DBU
concentration the unselective background reaction becomes
competitive. Catalysis by 3-Ni does not require additives such as
NaPF6 or molecular sieves which were critical for the performance
of [CpRu(PPh3)(NCMe)2]PF6.3
16 We thank one of the referees for this insightful suggestion.
4452 | Chem. Commun., 2005, 4450–4452
This journal is ß The Royal Society of Chemistry 2005