Self-assembled bidentate ligands for the nickel-catalyzed hydrocyanation of alkenes

Article abstract of DOI:10.1002/anie.200805092

(Chemical Equation Presented) Well-stocked library: A catalyst with interesting activity, regioselectivity (rs), and functional-group tolerance could be identified from a 5x4 library of isoquinolone-or aminopyridine-derived self-assembled bidentate ligand

Full text of DOI:10.1002/anie.200805092

Angewandte
Chemie
DOI: 10.1002/anie.200805092
Homogeneous Catalysis
Self-Assembled Bidentate Ligands for the Nickel-Catalyzed
Hydrocyanation of Alkenes**
Michiel de Greef and Bernhard Breit*
Nitriles[1] are versatile intermediates in organic synthesis as
they can be transformed into a wide variety of useful
compounds, such as amines, aldehydes, ketones, and a range
of carboxylic acid derivatives. A particularly appealing
method for their synthesis consists of the formal addition of
[2]
=
HCN across a C C double bond, commonly referred to as
hydrocyanation. Although the reaction is most commonly
catalyzed by catalysts incorporating (bidentate) phosphite or
phosphinite ligands, the use of bidentate phosphine ligands
has attracted considerable interest.[3–8] Van Leeuwen and co-
workers showed that bidentate ligands with bite angles of
approximately 1058 have a beneficial effect on the activity of
catalysts for the hydrocyanation of terminal alkenes.[9] The
same research group was able to demonstrate that the use of
electron-deficient ligands also leads to increased activity.[5]
However, catalytic hydrocyanation remains underdeveloped
and most of the research in the area has focused on the
DuPont adiponitrile process.[10] In order to transform hydro-
cyanation into a synthetically useful reaction, it is of prime
importance to develop more efficient catalyst systems.
We recently introduced the concept of self-assembly of
bidentate ligands for combinatorial homogeneous catalysis.[11]
Self-assembly of two complementary species through hydro-
gen bonding is a widely occurring phenomenon in nature, as
exemplified by DNA base pairing between adenine (A) and
thymine (T; (Scheme 1).[12]
Scheme 1. Self-assembly through hydrogen bonding of the adenine–
thymine and aminopyridine–isoquinolone systems. m and n represent
the number of library components.D=donor atom.
In order to explore whether the heterodimeric chelate
bonding mode operates under hydrocyanation conditions, the
activity of [(cod)Ni(3-dpicon)(6-dppap)] (cod = cycloocta-
diene) in the hydrocyanation of styrene with acetone
cyanohydrin was compared to the activities of similar Ni0
complexes that contain other combinations of mono- and
bidentate ligands (Table 1).
Nickel(0) complexes that incorporate two equivalents of
either 3-dpicon (Table 1, entry 10), or 6-dppap (entry 11)
were barely active and gave yields comparable to those
Table 1: Nickel-catalyzed hydrocyanation of styrene.[a]
Beyond the benefit of simplified ligand synthesis, the
potential of our strategy lies in the possibility for the
generation of combinatorial libraries of bidentate ligands
through the simple mixing of suitably functionalized mono-
dentate precursors.[13] This approach allowed us to identify
uniquely active and selective catalysts for hydroformyla-
tion,[14] anti-Markovnikov hydration of terminal alkynes,[15]
and asymmetric hydrogenation[16] and hydration of nitriles.[17]
Interestingly, with bite angles close to 1058 and being
relatively electron poor, our self-assembled bidentate ligands
display all of the features that seem to favor nickel-catalyzed
hydrocyanation.
Entry
Additive
b [%][b]
l [%][b]
[c]
[c]
1
2
3
4
5
6
–
PPh3
P(OPh)3
dppe
xantphos
dpephos
dpephos
3-dpicon
3-dpicon
3-dpicon
6-dppap
–
PPh3
P(OPh)3
–
–
–
–
6-dppap
6-dppap
3-dpicon
6-dppap
–
–
–
–
–
–
–
–
–
[c]
7
14
1
88
92
89
87
49
7
1
–
[c]
1
3
6
3
1
1
7[d]
8
9
10
11
AcOH
–
AcOH
–
–
[*] Dr. M. de Greef, Prof. Dr. B. Breit
Institut fꢀr Organische Chemie und Biochemie
Freiburg Institute for Advanced Studies (FRIAS)
Albert-Ludwigs-Universitꢁt Freiburg
Albertstrasse 21, 79104 Freiburg i. Brsg. (Germany)
Fax: (+49)761-203-8715
[c]
7
–
[a] Reaction conditions: styrene/acetone cyanohydrin/[Ni(cod)2]/L1/L2 =
1:1.25:0.05:0.05:0.05 in toluene (2 mL) for 20 h at 608C. [b] GC yield.
[c] No reaction product detected. [d] The reaction was carried out 458C.
Dppe=1,2-bis(diphenylphosphino)ethane, xantphos=4,5-bis(diphenyl-
phosphino)-9,9-dimethylxanthene, dpephos=bis(2-diphenylphosphino-
phenyl)ether, 3-dpicon=3-diphenylphosphinoisoquinolone, 6-dppap=
6-diphenylphosphino-N-pivaloyl-2-aminopyridine.
E-mail: bernhard.breit@chemie.uni-freiburg.de
[**] This work was supported by the Alexander von Humboldt-
Foundation (fellowship to M.d.G.), the DFG (IRTG 1038) and the
Krupp Foundation.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805092.
Angew. Chem. Int. Ed. 2009, 48, 551 –554
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
551

Communications
Table 2: Ligand matrix (5ꢂ4) of aminopyridine- or isoquinolone-derived
self-assembled bidentate ligands in the nickel-catalyzed hydrocyanation
of styrene.[a]
observed for other monodentate ligands such as PPh3
(entry 2) and P(OPh)3 (entry 3). However, upon mixing of
equimolar amounts of [Ni(cod)2], 3-dpicon, and 6-dppap, a
highly active hydrocyanation catalyst was obtained (Table 1,
entry 8). The total yield of hydrocyanation products was as
high as 90% and the branched reaction product was obtained
as the major isomer (b:l = 97:3). This result strongly supports
the assumption that the heterodimeric 3-dpicon/6-dppap
ligand system acts like a bidentate ligand that is held together
by a hydrogen-bonding network. Further evidence came from
the observation that the heterodimeric 3-dpicon/6-dppap
platform proved less efficient as a ligand system when the
polarity of the reaction medium was increased (entry 9). This
result may be related to the fact that hydrogen-bond
formation becomes problematic under these conditions. As
a consequence, the ligand system loses part of its bidentate
character and displays catalytic activity comparable to that of
systems made up of monodentate ligands. Under similar
conditions, the activity of [(cod)Ni(dpephos)] remained
unchanged (entry 7).
Ligand geometry is another important parameter in
determining catalyst performance and deviations from the
ideal geometry may lead to considerable loss of activity and/
or selectivity. For example, Ni0 complexes that incorporate a
bidentate ligand with a small bite angle such as dppe (Table 1,
entry 4) showed no catalytic activity. On the other hand, the
use of xantphos (Table 1, entry 5) or dpephos (entry 6) gave
rise to catalyst systems displaying activities and selectivities
comparable to those observed for our heterodimeric 3-
dpicon/6-dppap system. When the reaction was run without
phosphorus-containing ligands (Table 1, entry 1) or in the
absence of any metal source, hydrocyanation did not occur at
all.
1a
1b
1c
1d
2a
2b
2c
2d
2e
96%[b]
(97:3)[c]
97%
(97:3)
90%
(96:4)
62%
(95:5)
25%
95%
(97:3)
90%
(97:3)
79%
(95:5)
52%
96%
(96:4)
94%
(96:4)
85%
(95:5)
64%
93%
(95:5)
89%
(96:4)
78%
(94:6)
55%
(95:5)
19%
(95:5)
21%
(95:5)
19%
(96:4)
(96:4)
(96:4)
(96:4)
Reaction conditions: styrene/acetone cyanohydrin/[Ni(cod)2]/L1/L2 =
1:1.25:0.05:0.05:0.05 in toluene (2 mL) for 20 h at 608C. [b] Combined
GC yield of branched and linear hydrocyanation products. [c] Regiose-
lectivity (branched:linear).
the electron density on the phosphorus atom decreases from
the top to the bottom and from the left to the right of the
table. The major trends that can be seen in this study are that
the catalytic activity decreases down a column and depends
more on the nature of the isoquinolone than on the nature of
its aminopyridine counterpart. According to the former trend,
the most efficient catalysts are expected to be located in the
upper (left) part of the table. Indeed, catalysts that incorpo-
rate ligands 1a/2a, 1a/2b, 1b/2a, and 1d/2a appeared to be
excellent hydrocyanation catalysts and were more active than
the parent system 1b/2b. These results are in contrast with
those of Van Leeuwen and co-workers, who observed that
decreased electron density on the donor atom led to increased
activity.[5] This discrepancy may be accounted for by the fact
that the isoquinolone ligand is in itself electron deficient.
Further decrease of the electron density may lead to a
situation in which oxidative addition of HCN to Ni0 becomes
troublesome. In all cases, hydrocyanation occurred with yields
greater than 95% and regioselectivities of up to 97:3 in favor
of the branched reaction product. When the reaction was
carried out at temperatures lower than 608C, the catalyst-
incorporating ligand 1a/2a displayed the best overall perfor-
mance and was selected for optimization purposes (Table 3).
In order to obtain more information on the catalyst
structure, a solution of [Ni(cod)2], 3-dpicon, and 6-dppap in
degassed C6D6 was analyzed by NMR spectroscopy. The
31P NMR spectrum of the solution featured doublets at d =
36.0 ppm and at d = 40.5 ppm with a 2JP–P-coupling constant of
3.3 Hz and confirmed the presence of two nonequivalent
phosphine ligands coordinated to the same metal center.
1
Additionally, H NMR spectra showed a substantial shift of
the NH signals of the 6-dppap and 3-dpicon ligands to low
field after addition of [Ni(cod)2]. These observations provide
strong evidence for the expected heterodimeric chelate
formation through a hydrogen-bonding network similar to
Watson–Crick base pairing of adenine and thymine in DNA.
The next logical step involved the generation of a library
of self-assembled bidentate ligand systems analogous to the 3-
dpicon/6-dppap system in order to probe the influence of
electronic effects on activity, selectivity, and stability of the
ligands. It was in this step that our methodology was used to
its full potential as a tool for the combinatorial synthesis of a
library of bidentate ligands. From five phosphinoisoquinolone
and four aminopyridyl phosphine ligands, a set of 20 bidentate
ligand combinations was generated by simply mixing the
components and [Ni(cod)2] in toluene. The resulting catalysts
were explored with respect to their potential to catalyze the
hydrocyanation of styrene with acetone cyanohydrin
(Table 2). The ligands are represented in such a manner that
Table 3: Influence of the reaction conditions on the nickel-catalyzed
hydrocyanation of styrene with acetone cyanohydrin, using the 1a/2a
platform as a ligand system.[a]
Entry
1a
[%]
2a
[%]
[Ni]
[%]
[Styrene]
[HCN]
T
[8C]
Yield
[molLÀ1
]
[molLÀ1
]
[%][b,c]
1
2
3
4
5
5
5
5
10
5
5
5
5
10
5
5
5
5
5
0.50
1.0
1.0
1.0
1.0
0.63
1.3
1.5
1.5
1.5
45
45
35
35
35
95 (97:3)
99 (95:5)
97 (94:6)
98 (98:2)
90 (99:1)
2.5
[a] Reactions were run for 20 h in toluene. [b] GC yield. [c] The value in
parentheses refers to the observed b:l ratio.
552
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 551 –554

Angewandte
Chemie
Better results were obtained when the reaction was
carried out at a styrene concentration of 1.0 molLÀ1 instead
of 0.50 molLÀ1. In this case, the total yield of hydrocyanation
products was essentially quantitative (Table 3, entry 2).
Furthermore, when the reaction was carried out with
1.5 equivalents instead of 1.3 equivalents of acetone cyano-
hydrin, the temperature could be lowered to 358C without a
significant reduction in yield (Table 3, entry 3). Interestingly,
at this temperature, the reaction mixture remained homoge-
neous throughout the reaction. This observation suggested
that the catalyst did not decompose during the reaction and
could be used in a second reaction cycle. Indeed, in an
experiment that used a catalyst with ligand system 1a/2a, the
total yield of hydrocyanation products was as high as 97%
after two reaction cycles. When a catalyst that incorporated a
well-established bidentate ligand such as dpephos was used,
the total yield of hydrocyanation products was not higher than
70% after two cycles. This result underlines the exceptional
(entry 6) and -F (entry 7), were tolerated. Interestingly, the
hydrocyanation of (E)-1-phenylbuta-1,3-diene was also suc-
cessfully accomplished and led to the exclusive formation of
(E)-2-methyl-4-phenyl-3-butenenitrile (Table 4, entry 8).
It has been shown for the first time that heterodimeric
self-assembled bidentate ligands can form complexes with
Ni0. Such complexes appeared to be promising catalysts for
the hydrocyanation of styrene. From a 5 ꢀ 4 library of
isoquinolone- or aminopyridine-derived self-assembled
bidentate ligands, a catalyst with interesting activity, regiose-
lectivity, and functional group tolerance was identified.
Future efforts will focus on the extension of our methodology
to enantioselective hydrocyanation.[18–22] Heterodimeric self-
assembled bidentate ligands bearing chiral information were
recently applied successfully to rhodium-catalyzed asymmet-
ric hydrogenation.[16]
stability of our catalyst. On the other hand, the increase in Experimental Section
General procedure: A resealable Schlenk flask was charged with
concentration seemed to have a negative influence on the
regioselectivity (Table 3, entry 1 versus entry 2). This issue
was addressed by using a twofold excess of both 1a and 2a
with respect to the metal (Table 3, entry 3 versus entry 4).
Furthermore, the yield only slightly diminished when the
catalyst load was lowered from 5.0 to 2.5 mol% (Table 3,
entry 4 versus entry 5). As before, the reaction mixture
remained homogeneous throughout the reaction, which
suggested that the catalyst was stable under the reaction
conditions. We therefore had good reason to believe that this
minor loss of yield could be dealt with by increasing the
reaction time. In summary, [Ni(1a/2a)2] turned out to be
capable of catalyzing the hydrocyanation of styrene under
mild conditions with excellent yield and regioselectivity and
at a catalyst load of 2.5 mol%.
The synthetic scope of [Ni(1a/2a)2] was subsequently
explored. In most cases, reacting 1 equivalent of alkene with
1.5 equivalents of acetone cyanohydrin in the presence of 2.5
mol% of [Ni(1a/2a)2] for 25 hours in toluene at 358C gave
the expected hydrocyanation products in high yields
(Table 4). In all cases, the reactions proceeded with essentially
complete regioselectivity in favor of the branched isomer.
Various functional groups, including -OMe (Table 4, entry 1),
-OAc (entry 2), -Me (entry 3), -Ph (entry 5), -CO2Me
[Ni(cod)2] (0.050 mmol), 1a (0.10 mmol), and 2a (0.10 mmol) in a
glovebox. The Schlenk flask was then connected to a conventional
Schlenk line and toluene (1 mL) was added. The dark orange solution
was stirred for 60 min at room temperature. A solution of substrate
(2.0 mmol) in toluene (1 mL) was then added, followed by acetone
cyanohydrin (1.5 mmol). The Schlenk flask was sealed and the
reaction mixture stirred for 25 h at 358C. After cooling to room
temperature, the homogeneous mixture was concentrated and the
residue purified by flash chromatography on SiO2.
Received: October 17, 2008
Published online: December 12, 2008
Keywords: alkenes ·homogeneouscatalysis·hydrogenbonding·
.
nickel · self-assembly
[1] P. Pollak, G. Romeder, F. Hagedorn, H. Gelbke in Ullmanꢀs
Encyclopedia of Industrial Chemistry, Vol. A17, Wiley-VCH,
Weinheim, 1985, pp. 363 – 376.
[2] K. Huthmacher, S. Krill in Applied Homogeneous Catalysis with
Organometallic Compounds, Vol. 1 (Eds: B. Cornils, W. A.
Hermann), Wiley-VCH, Weinheim, 2002, pp. 465 – 486.
[3] M. Kranenburg, P. C. J. Kamer, P. W. N. M. van Leeuwen, D.
Vogt, W. Keim, J. Chem. Soc. Chem. Commun. 1995, 2177 – 2178.
[4] W. Goertz, P. C. J. Kamer, P. W. N. M. van Leeuwen, D. Vogt,
Chem. Commun. 1997, 1521 – 1522.
[5] W. Goertz, W. Keim, D. Vogt, U. Englert, M. D. K. Boele, L. A.
van der Veen, P. C. J. Kamer, P. W. N. M. van Leeuwen, J. Chem.
Soc. Dalton Trans. 1998, 2981 – 2988.
[6] E. Burello, P. Marion, J.-C. Galland, A. Chamard, G. Rothen-
berg, Adv. Synth. Catal. 2005, 347, 803 – 810.
[7] L. Bini, C. Mꢁller, J. Wilting, L. von Chrzanowski, A. L. Spek, D.
Vogt, J. Am. Chem. Soc. 2007, 129, 12622 – 12623.
[8] A. Acosta-Ramꢂrez, A. Flores-Gaspar, M. Muꢃoz-Hernꢄndez,
A. Arꢅvalo, W. D. Jones, J. J. Garcꢂa, Organometallics 2007, 26,
1712 – 1720.
[9] Z. Freixa, P. W. N. M. van Leeuwen, Dalton Trans. 2003, 1890 –
1901.
[10] C. A. Tolman, R. J. McKinney, W. C. Seidel, J. D. Druliner, W. R.
Stevens, Adv. Catal. 1985, 33, 1 – 46.
[11] B. Breit, Angew. Chem. 2005, 117, 6976 – 6986; Angew. Chem.
Int. Ed. 2005, 44, 6816 – 6825.
[12] J. D. Watson, F. H. C. Crick, Nature 1953, 171, 964 – 967.
Table 4: Regioselective hydrocyanation of functionalized alkenes with
[Ni(1a/2a)2] as the catalyst.[a]
Entry
Substrate
b:l
Yield [%][b]
=
1
2
3
4
5
6
7
8
4-OMe-C6H4CH CH2
>99:1
>99:1
>99:1
>99:1
>99:1
>99:1
quant.
quant.[c]
99
=
4-OAc-C6H4CH CH2
=
4-Me-C6H4CH CH2
=
C6H5CH CH2
89
=
4-Ph-C6H4CH CH2
4-CO2Me- C6H4CH CH2
4-F-C6H4CH CH2
(E)-1-phenylbuta-1,3-diene
quant.
quant.
quant.
86
=
=
>99:1
[d]
[a] Reaction conditions: styrene/acetone cyanohydrin/[Ni(cod)2]/L1/L2 =
1:1.5:0.025:0.05:0.05 in toluene (1 mL) for 25 h at 358C. [b] Yield of
isolated product. [c] Yield after 40 h. [d] (E)-2-methyl-4-phenyl-3-butene-
nitrile was the only detectable reaction product.
Angew. Chem. Int. Ed. 2009, 48, 551 –554
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
553

Communications
[13] For alternative approaches to bidentate ligands through self-
[15] F. Chevallier, B. Breit, Angew. Chem. 2006, 118, 1629 – 1632;
Angew. Chem. Int. Ed. 2006, 45, 1599 – 1602.
assembly, see Ref. [11] and: a) J. M. Takacs, D. S. Reddy, S. A.
Moteki, D. Wu, H. Palencia, J. Am. Chem. Soc. 2004, 126, 4494 –
4495; b) M. J. Wilkinson, P. W. N. M. van Leeuwen, J. N. H.
Reek, Org. Biomol. Chem. 2005, 3, 2371 – 2383; c) V. F. Slagt,
M. Rꢆder, P. C. J. Kamer, P. W. N. M. van Leeuwen, J. N. H.
Reek, J. Am. Chem. Soc. 2004, 126, 4056 – 4057; d) V. F. Slagt,
P. W. N. M. van Leeuwen, J. N. H. Reek, Angew. Chem. 2003,
115, 5777 – 5781; Angew. Chem. Int. Ed. 2003, 42, 5619 – 5623;
e) V. F. Slagt, P. W. N. M. van Leeuwen, J. N. H. Reek, Chem.
Commun. 2003, 2474 – 2475.
[16] M. Weis, C. Waloch, W. Seiche, B. Breit, J. Am. Chem. Soc. 2006,
128, 4188 – 4189.
ˇ
[17] T. Smejkal, B. Breit, Organometallics 2007, 26, 2461 – 2464.
[18] A. L. Casalnuovo, T. V. RajanBabu, T. A. Ayers, T. H. Warren, J.
Am. Chem. Soc. 1994, 116, 9869 – 9882.
[19] W. Goertz, P. C. J. Kamer, P. W. N. M. van Leeuwen, D. Vogt,
Chem. Eur. J. 2001, 7, 1614.
[20] J. Wilting, M. Janssen, C. Mꢁller, D. Vogt, J. Am. Chem. Soc.
2006, 128, 11374 – 11375.
[14] a) B. Breit, W. Seiche, J. Am. Chem. Soc. 2003, 125, 6608 – 6609;
b) B. Breit, W. Seiche, Angew. Chem. 2005, 117, 1666 – 1669;
Angew. Chem. Int. Ed. 2005, 44, 1640 – 1643; c) C. Waloch, J.
Wieland, M. Keller, B. Breit, Angew. Chem. 2007, 119, 3097 –
3099; Angew. Chem. Int. Ed. 2007, 46, 3037 – 3039.
[21] B. Saha, T. V. RajanBabu, Org. Lett. 2006, 8, 4657 – 4659.
[22] J. Wilting, M. Janssen, C. Mꢁller, M. Lutz, A. L. Spek, D. Vogt,
Adv. Synth. Catal. 2007, 349, 350 – 356.
554
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 551 –554