Heteroatom-directed alkylcyanation of alkynes

Article abstract of DOI:10.1021/ja1017078

Alkanenitriles having a heteroatom such as nitrogen, oxygen, and sulfur at the γ-position are found to add across alkynes stereo-and regioselectively by nickel/Lewis acid catalysis to give highly substituted acrylonitriles. The heteroatom functionalities likely coordinate to the nickel center to make oxidative addition of the C-CN bonds of the alkyl cyanides kinetically favorable, forming a five-membered nickelacycle intermediate and, thus, preventing β-hydride elimination to allow the alkylcyanation reaction.

Full text of DOI:10.1021/ja1017078

Published on Web 07/01/2010
Heteroatom-Directed Alkylcyanation of Alkynes
Yoshiaki Nakao,* Akira Yada, and Tamejiro Hiyama*,§
Department of Material Chemistry, Graduate School of Engineering, Kyoto UniVersity,
Kyoto 615-8510, Japan
Received September 3, 2009; E-mail: yoshiakinakao@npc05.mbox.media.kyoto-u.ac.jp;
thiyama@z06.mbox.media.kyoto-u.ac.jp
Abstract: Alkanenitriles having a heteroatom such as nitrogen, oxygen, and sulfur at the γ-position are
found to add across alkynes stereo- and regioselectively by nickel/Lewis acid catalysis to give highly
substituted acrylonitriles. The heteroatom functionalities likely coordinate to the nickel center to make
oxidative addition of the C-CN bonds of the alkyl cyanides kinetically favorable, forming a five-membered
nickelacycle intermediate and, thus, preventing ꢀ-hydride elimination to allow the alkylcyanation reaction.
Introduction
the unwanted ꢀ-hydride elimination by forming a chelate
intermediate. This strategy to facilitate otherwise challenging
Stereoselective construction of acyclic tri- and tetra-substi-
tuted ethenes is a major challenge in modern organic synthesis,
because they are ubiquitous in many natural products and
materials and serve as a building block for branched alkanes
with vicinal stereocenters through hydrogenation, epoxidation,
and cyclopropanation of the substituted double bond.1
Classical approaches to such structures involve the Wittig,
Honer-Wadsworth-Emmons, Julia, and Peterson reactions, that
often result in low stereoselectivity with substituents of a small
steric and/or electronic difference.2 Representative alternative
protocols to address this issue are stereo- and regioselective
addition reactions across alkynes. Especially, carbometalation
followed by trapping with carbon electrophiles has been a
reliable method to access stereochemically well-defined
olefins.1b Nevertheless, use of prefunctionalized organometallic
reagents and electrophiles as well as stoichiometric metal
residues as a byproduct limits the potential of the protocols,
particularly for large-scale production.
We have developed insertion reactions of alkynes into the
C-CN bonds of nitriles, namely, carbocyanation reaction, as a
novel strategy to access substituted ethenes.3 While many nitriles
have been found to participate in the transformation with high
stereo- and regioselectivity, use of alkanenitriles has been
severely limited due to relatively low reactivity of their
C(sp3)-CN bonds toward oxidative addition and competitive
ꢀ-hydride elimination of an alkylnickel intermediate, resulting
in contamination of hydrocyanation products.4 We envisaged
that coordinating functionalities in alkyl cyanides could suppress
activation of C-C bonds other than C-CN bonds by transition
metals5 has been demonstrated in both stoichiometric6 and
catalytic7 reactions. We report herein chelation-assisted C-CN
bond activation to realize the addition reactions of γ-aza(oxa
or thia)alkanenitriles across alkynes in the presence of a nickel/
Lewis acid (LA) cocatalyst to give highly substituted and
functionalized acrylonitriles with high stereo- and regioselec-
tivity and atom economy.
Results and Discussion
The problem associated with ꢀ-hydride elimination was
previously solved in part by employing highly bulky ligands
such as SPhos8 in the reaction of propionitrile to improve the
yield of the cis-ethylcyanation product,4a,c whereas butyronitrile
still suffered from competitive hydrocyanation products afforded
propylcyanation product in low yield.4c A dramatic improvement
of the product selectivity was observed by introducing a
secondary amino group at the γ-position of butyronitrile. Thus,
the reaction of aminobutyronitrile 1a (1.0 mmol) with 4-octyne
(2a, 2.0 mmol) in the presence of Ni(cod)2 (10 mol %), SPhos
(20 mol %), and AlMe3 (40 mol %) in toluene at 50 °C for 9 h
gave the corresponding cis-alkylcyanation product 3aa in 86%
yield and no trace amount of hydrocyanation products (entry 1
of Table 1). The observed effect of the amino group, however,
(5) For reviews, see: (a) Crabtree, R. H. Chem. ReV. 1985, 85, 245. (b)
Milstein, D.; Rybtchinski, B. Angew. Chem., Int. Ed. 1999, 38, 871.
(c) Murakami, M.; Ito, Y. In ActiVation of UnreactiVe Bonds and
Organic Synthesis; Murai, S., Ed.; Springer: Berlin, 1999; pp 97-
129. (d) Jun, C.-H. Chem. Soc. ReV. 2004, 33, 610.
§ Current address: Research & Development Initiative Chuo University,
1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan. E-mail: thiyama@
kc.chuo-u.ac.jp.
(6) (a) Suggs, J. W.; Jun, C.-H. J. Am. Chem. Soc. 1984, 106, 3054. (b)
Suggs, J. W.; Jun, C.-H. J. Chem. Soc., Chem. Commun. 1985, 92.
(c) Suggs, J. W.; Jun, C.-H. J. Am. Chem. Soc. 1986, 108, 4679. (d)
Gozin, M.; Weisman, A.; Bendavid, Y.; Milstein, D. Nature 1993,
364, 699.
(1) (a) Faulkner, D. J. Synthesis 1971, 175. (b) Flynn, A. B.; Ogilvie,
W. W. Chem. ReV. 2007, 107, 4698.
(2) Modern Carbonyl Olefination; Takeda, T., Ed.; Wiley-VCH: Wein-
heim, Germany, 2004.
(7) (a) Liou, S. Y.; van der Boom, M. E.; Milstein, D. Chem. Commun.
1998, 687. (b) Jun, C.-H.; Lee, H. J. Am. Chem. Soc. 1999, 121, 880.
(c) Chatani, N.; Ie, Y.; Kakiuchi, F.; Murai, S. J. Am. Chem. Soc.
1999, 121, 8645. (d) Dreis, A. M.; Douglas, C. J. J. Am. Chem. Soc.
2009, 131, 412.
(3) (a) Nakao, Y.; Hiyama, T. Pure Appl. Chem. 2008, 80, 1097. (b)
Na´jera, C.; Sansano, J. M. Angew. Chem., Int. Ed. 2009, 48, 2452.
(4) (a) Nakao, Y.; Yada, A.; Ebata, S.; Hiyama, T. J. Am. Chem. Soc.
2007, 129, 2428. (b) Yada, A.; Yukawa, T.; Nakao, Y.; Hiyama, T.
Chem. Commun. 2009, 3931. (c) Yada, A.; Yukawa, T.; Idei, H.;
Nakao, Y.; Hiyama, T. Bull. Chem. Soc. Jpn. 2010, 83, 619.
(8) Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2004, 43, 1871.
9
10024 J. AM. CHEM. SOC. 2010, 132, 10024–10026
10.1021/ja1017078  2010 American Chemical Society

Alkylcyanation of Alkynes
A R T I C L E S
Table 1. Nickel/AlMe3-Catalyzed Alkylcyanation of 4-Octyne
Table 2. Carbocyanation of Alkynes with Alkanenitriles Having a
Coordinating Group
a Isolated yields based on 1. b Run at 80 °C. c Run with 60 mol % of
AlMe3. d Estimated by GC using n-dodecane as an internal standard.
e Run with 0.50 mmol of 2a.
did not work at all with aminopropionitrile 1b (entry 2), whereas
the addition of amino-substituted valeronitrile 1c and hexaneni-
trile 1d across 2a took place exclusively at the γ-position of
the pyrrolidyl group to allow addition of secondary alkyl groups
(entries 3 and 4). We observed that γ-aminonitrile 1a reacted
much faster than propionitrile based on the results from their
competitive reactions with 2a (entry 5).
The amino effect for promotion of the alkylcyanation reaction
was further tested under slightly modified conditions using P(2-
MeO-C6H4)3 as a ligand (Table 2). Use of this ligand gave 3aa
in 88% yield with 1.4 mmol of 2a, 3 mol % of Ni(cod)2, and
12 mol % of AlMe3 at 80 °C for 9 h. Other cyclic and acyclic
amino moieties were equally effective (entries 1 and 2) even
with strained and thus labile9 aziridine-containing substrate 1g
(entry 3). The formation of 3ca and 3da (Table 1) prompted us
to examine secondary alkyl cyanides, challenging substrates for
the alkylcyanation.4b,c To our delight, a range of R-substituents
in 1a tolerated to give branched carbocyanation products in
modest to good yields (entries 4-7). Nevertheless, attempted
addition of optically active (S)-1i of 86% ee10 resulted in 3ia
of 34% ee due to background racemization of (S)-1i, and that
of tertiary alkyl cyanides was futile even with the aid of an
amino group. The pyridyl sp2-nitrogen of 1l also served as a
directing group (entry 8), whereas the corresponding 4-pyridyl
variant did not participate in the reaction (Figure 1). These
results prove that the effect of the heteroatom is derived
primarily from its coordination to nickel but not from its
inductive influence. Moreover, ether, thioether, and acetal
functionalities assisted the reaction to give the corresponding
alkylcyanation products (entries 9-13), whereas alkyl cyanides
having arylamine, sp2-nitrogen-containing five-membered het-
erocycles, epoxide, ester, amide carbonyl oxygens, and thioacetal
did not give the corresponding adduct due to either no reaction
or decomposition of the nitriles, or apparent catalyst decomposi-
a Isolated yields based on 1. b Run with P(t-Bu)3 as a ligand. c Run
with 60 mol % of AlMe3. d Contaminated with 9% of regioisomer 3′id.
e Contaminated with <5% of regio- and/or stereoisomers. f Run with
AsPh3 (6 mol %) and B(C6F5)3 (12 mol %). g Run with slow addition of
2g over 7.5 h and additional stirring for 2.5 h.
tion observed with the thioacetal substrate (Figure 1). The scope
of alkynes was examined briefly with 1a and 1i as nitrile
substrates. In addition to other symmetrical dialkylacetylenes
(entries 14 and 15), internal alkynes with sterically biased
substituents reacted successfully with stereo- and regioselec-
tivities similar to common alkyne-carbocyanation reactions,3 and
adducts are produced having a larger alkyne-substituent and the
cyano group bound to the same sp2-carbon (entries 16 and 17).
Use of less electron-donating triphenylarsine as a ligand was
(9) Aziridines undergo oxidative addition to nickel(0), see: Lin, B. L.;
Clough, C. R.; Hillhouse, G. L. J. Am. Chem. Soc. 2002, 124, 2890.
(10) See Supporting Information for the synthesis of (S)-1i.
9
J. AM. CHEM. SOC. VOL. 132, NO. 29, 2010 10025

A R T I C L E S
Nakao et al.
Scheme 1. Plausible Mechanism
with 1d through multiple ꢀ-hydride elimination-hydronickelation
sequences to finally give 3da through C (R ) Et). The amino
group can also interact with AlMe3, but the resulting species H
would not be involved in the present catalytic cycle and in
equilibrium with cyano-coordinating one I, which can participate
in the catalysis.
Figure 1. Nitriles inapplicable to the alkylcyanation reaction.
found to be effective for the addition across terminal alkynes
(entries 18 and 19); nickel catalysts with an electron-donating
phosphine were prone to induce trimerization and/or oligomer-
ization of terminal alkynes.
All the data described above suggest a catalytic cycle
involving 5-membered azanickelacycle C as a key intermediate
generated by rapid oxidative addition of the C-CN bond of 1a
to nickel(0) through coordination of the amino group to nickel(0)
(A) and intramolecular η2-coordination of the cyano group (B),
wherein the cyano nitrogen is bound to AlMe3 (Scheme 1).11
Subsequent ligand exchange (D), alkylnickelation (E), and
reductive elimination afford 3aa and regenerate A. No observed
adduct derived from 1b may be attributed to lack of the
possibility of a 5-membered chelate, while a possible 6-mem-
bered nickelacycle F derived from 1c would be reluctant to
proceed through the subsequent elemental steps and undergo
ꢀ-hydride elimination (G) followed by hydronickelation in an
opposite direction to give 5-membered intermediate C (R )
Me), which appears to be responsible for the formation of 3ca.
Similar rearrangement of nickelacycles was reported by Rovis
and co-workers.12 This isomerization would also be operative
Conclusion
In conclusion, we have demonstrated that regio- and stereo-
selective alkylcyanation of alkynes is achieved by introducing
a coordinating heteroatom in alkanenitriles and using nickel/
LA catalysts. Accordingly, the scope of the alkylcyanation
reaction is broadened significantly to allow stereoselective
synthesis of various tri- and tetra-substituted ethenes having an
alkyl group containing various heteroatom functionalities that
allow further elaboration of the adducts. The concept to suppress
ꢀ-hydride elimination presented herein will be applicable to
other carbocyanation reactions especially across alkenes.
Acknowledgment. We thank Mr. Eiji Morita for experimental
assistance. This work has been supported financially by a Grant-
in-Aids for Creative Scientific Research, Scientific Research (S),
Young Scientists (A), and Priority Areas “Molecular Theory for
Real Systems” from MEXT and by General Sekiyu Research &
Development Encouragement & Assistance Foundation. A.Y.
acknowledges the JSPS for a predoctoral fellowship.
(11) (a) Brunkan, N. M.; Brestensky, D. M.; Jones, W. D. J. Am. Chem.
Soc. 2004, 126, 3627. (b) Acosta-Ram´ırez, A.; Mun˜oz-Herna´ndez, M.;
Jones, W. D.; Garc´ıa, J. J. J. Organomet. Chem. 2006, 691, 3895. (c)
Acosta-Ram´ırez, A.; Mun˜oz-Herna´ndez, M.; Jones, W. D.; Garc´ıa,
J. J. Organometllics 2007, 26, 5766. (d) Ate¸sin, T. A.; Li, T.; Lachaize,
S.; Brennessel, W. W.; Garc´ıa, J. J.; Jones, W. D. J. Am. Chem. Soc.
2007, 129, 7562. (e) Nakao, Y.; Ebata, S.; Yada, A.; Hiyama, T.;
Ikawa, M.; Ogoshi, S. J. Am. Chem. Soc. 2008, 130, 12874.
(12) O’Brien, E. M.; Bercot, E. A.; Rovis, T. J. Am. Chem. Soc. 2003,
125, 10498.
Supporting Information Available: Detailed experimental
procedures including spectroscopic and analytical data (PDF).
This material is available free of charge via the Internet at http://
pubs.acs.org.
JA1017078
9
10026 J. AM. CHEM. SOC. VOL. 132, NO. 29, 2010