Palladium-catalyzed C(sp2) - H cyanation using tertiary amine derived isocyanide as a cyano source

Article abstract of DOI:10.1021/ol302372p

An unprecedented palladium-catalyzed cyanation of aromatic C-H bonds by using tertiary amine derived isocyanide as a novel cyano source was developed. Cu(TFA)₂ was used as a requisite stoichiometric oxidant. Mechanistic studies suggest that a tertiary carbon cation-based intermediate is involved following the C-N bond breakage.

Full text of DOI:10.1021/ol302372p

ORGANIC
LETTERS
Palladium-Catalyzed C(sp2)ꢀH Cyanation
Using Tertiary Amine Derived Isocyanide
as a Cyano Source
2012
Vol. 14, No. 18
4966–4969
Jiangling Peng, Jiaji Zhao, Ziwei Hu, Dongdong Liang, Jinbo Huang, and Qiang Zhu*
State Key Laboratory of Respiratory Disease, Guangzhou Institutes of Biomedicine and
Health, Chinese Academy of Sciences, 190 Kaiyuan Avenue, Guangzhou 510530, China
zhu_qiang@gibh.ac.cn
Received August 27, 2012
ABSTRACT
An unprecedented palladium-catalyzed cyanation of aromatic CꢀH bonds by using tertiary amine derived isocyanide as a novel cyano source was
developed. Cu(TFA)2 was used as a requisite stoichiometric oxidant. Mechanistic studies suggest that a tertiary carbon cation-based intermediate
is involved following the CꢀN bond breakage.
The great versatility of aromatic nitriles in chemical trans-
formations1 and their frequent appearance in pharmaceuti-
cals, biologically active products, and functional materials
have motivated substantial efforts toward their preparation.2
Classical methods of introducing a cyano group into
aromatic rings require prefunctionalized arenes as precur-
sors, such as diazonium salts (Sandmeyer reaction),3 aryl
halides (Rosenmundꢀvon Braun reaction and transition-
metal-catalyzed cross-couplings),4 aldehydes (Schmidt
reaction),5 primary carboxyamides and aldoximes (by
dehydration),6 and many others.7 However, these ap-
proaches suffer from multiple synthetic steps, a lack of
atom economy, and/ortheindispensableuseof toxic cyano
sources. Transition-metal-catalyzed direct functionaliza-
tion of aromatic CꢀH bonds, with or without a directing
group, has now been recognized as an atom- and step-
efficient alternative for CꢀC and Cꢀheteroatom bond
formation.8 Compared with other types of CꢀH functio-
nalization reactions, straight cyanation of aromatic CꢀH
bonds is relatively underdeveloped (Scheme 1). In 2006,
Yu’s group disclosed the first example of a Cu-catalyzed
CꢀH cyanation of 2-arylpyridine using TMSCN as a
CN source.9 Later on, CꢀH cyanation by using other
“normal” cyano sources, such as CuCN,10 K4[Fe(CN)6],
and TsN(Ph)CN,11 were reported. Recently, Jiao et al.
(1) (a) Larock, R. C. ComprehensiveOrganicTransformations:AGuide
to Functional Group Preparations; VCH: New York, 1989. (b) Rappoport, Z.
Chemistry of the Cyano Group; John Wiley & Sons: London, 1970;
pp 121ꢀ312.
(2) (a) Fatiadi, A. J. In Preparation and Synthetic Applications of
Cyano Compounds; Patai, S., Rappaport, S. Z., Eds.; Wiley: New York,
1983. (b) Galli, C. Chem. Rev. 1988, 88, 765. (c) Anbarasan, P.; Schareina,
T.; Beller, M. Chem. Soc. Rev. 2011, 40, 5049.
(3) (a) Sandmeyer, T. Ber. Dtsch. Chem. Ges. 1884, 17, 1633.
(b) Hodgson, H. H. Chem. Rev. 1947, 40, 251.
(4) (a) Rosenmund, K. W.; Struck, E. Ber. Dtsch. Chem. Ges. 1919, 2,
1749. (b) Vekicky, J.; Soicke, A.; Steiner, R.; Schmalz, H.-G. J. Am.
Chem. Soc. 2011, 133, 6948. (c) Zhang, G.; Ren, X.; Chen, J.; Hu, M.;
Cheng, J. Org. Lett. 2011, 13, 5004.
(5) (a) Schmidt, K. F. Z. Angew. Chem. 1923, 36, 511. (b) Rokade,
B. V.; Prabhu, K. R. J. Org. Chem. 2012, 77, 5364.
(8) For recent reviews on CꢀH functionalization, see: (a) Mei, T.-S.;
Kou, L.; Ma, S.; Engle, K. M.; Yu, J.-Q Synthesis 2012, 44, 1778. (b) Shi,
Z.-Z.; Zhang, C.; Tang, C.-H.; Jiao, N. Chem. Soc. Rev. 2012, 41, 3381.
(c) Ramirez, T. A.; Zhao, B. G.; Shi, Y. A. Chem. Soc. Rev. 2012, 41, 931.
(9) Chen, X.; Hao, X. S.; Goodhue, C. E.; Yu, J.-Q. J. Am. Chem.
Soc. 2006, 128, 6790.
(6) (a) Ishihara, K.; Furuya, Y.; Yamamoto, H. Angew. Chem., Int.
Ed. 2002, 41, 2983. (b) Yang, S. H.; Chang, S. Org. Lett. 2001, 3, 4209.
(7) (a) Zhou, W.; Zhang, L.; Jiao, N. Angew. Chem., Int. Ed. 2009, 48,
7094. (b) Qin, C.; Jiao, N. J. Am. Chem. Soc. 2010, 132, 15893. (c) Chen,
J.; Sun, Y.; Liu, B.; Liu, D.; Cheng, J. Chem. Commun. 2012, 48, 449.
(10) (a) Jia, X.; Yang, D.; Zhang, S.; Cheng, J. Org. Lett. 2009, 11,
4716. (b) Reddy, B. V. S.; Begum, Z.; Reddy, Y. J.; Yadav, J. S.
Tetrahedron Lett. 2010, 51, 3334.
(11) (a) Yan, G.; Kuang, C.; Zhang, Y.; Wang, J. Org. Lett. 2010, 12,
1052. (b) Yang, Y.; Zhang, Y.; Wang, J. Org. Lett. 2011, 13, 5608.
r
10.1021/ol302372p
Published on Web 08/31/2012
2012 American Chemical Society

communicated a novel cyanation of indoles by employ-
ing DMF as the sole source of CN.12 An unprecedented
combined source (DMF and NH3) of CN in the cyanation
of 2-arylpyridines was first discovered by Chang’s group.13
A combination of DMSO and an ammonium ion was also
found to be effective in the cyanation of indoles described
by Cheng.14 These methods take advantage of not only
obviating the prefunctionalization of arenes but also
applying novel nontoxic single or combined CN sources.
Nevertheless, alternative approaches for the direct cyana-
tion of arenes applicable to a broader substrate scope being
performed under milder conditions are still desirable.
Herein, we report a novel palladium-catalyzed C(sp2)ꢀH
cyanation of free (NH)-indoles as well as 2-arylpyridines
by using tertiary amine derived isocyanide as an unprece-
dented cyano source via CꢀN bond cleavage.
CꢀH activation followed by isocyanide insertion.18 When
2-phenylindole 1a was used as a substrate, the desired
product N-tert-butyl-2-phenyl-1H-indole-3-carboxamide
3a was obtained in less than 10% yield. In an effort to
improve the yield of 3a by changing the solvent from THF
toHOAc, 3awas isolated in65% yield together withanun-
expected side product, 2-phenyl-1H-indole-3-carbonitrile
4a (5%). The intriguing mechanism prompted us to in-
vestigate this unique transformation in detail.
Table 1. Optimization of Reaction Conditionsa
catalyst
oxidant
time yield
entry (10 mol %) (1.0 equiv)
solvent
THF
(h)
(%)b
Scheme 1. Cyano Sources for CꢀH Cyanation
1
Pd(TFA)2
Pd(TFA)2
Pd(OAc)2
Pd(TFA)2
Cu(TFA)2
Cu(TFA)2
Cu(TFA)2
Cu(OAc)2
3
3
1
3
1
1
1
1
34c
84
82
0d
2
ClCH2CH2Cl
ClCH2CH2Cl
ClCH2CH2Cl
3
4
5e
6f
7g
8h
Pd(OAc)2 Cu(TFA)2 ClCH2CH2Cl
80
82
n.d.
0
Pd(TFA)2
ꢀ
ꢀ
ClCH2CH2Cl
ClCH2CH2Cl
ClCH2CH2Cl
Cu(TFA)2
Pd(OAc)2
ꢀ
a Conditions: 1a (0.20 mmol), 2a (0.24 mmol), catalyst (10 mol %),
oxidant (1.0 equiv), solvent (1.0 mL), air, sealed tube, 70 °C. b Yield of
isolated 4a. c 46% of 3a was also isolated. d N-Acetyl-N-tert-butyl-2-
phenyl-1H-indole-3-carboxamide 3b was obtained in 36% yield. e 5 mol %
of Pd(OAc)2 was used. f 1.0 equiv of Pd(TFA)2 was used. g Homocoupling
of 1a at C3 occurred. h 1.0 equiv of Pd(OAc)2 was used. 21 mg (36%) of 3a
and 34 mg (51%) of 3b were isolated.
We started to optimize the reaction conditions by using
2-phenylindole 1a and tert-butylisocyanide 2a as sub-
strates in the presence of 10 mol % of Pd(TFA)2 as a
catalyst and 1 equiv of Cu(TFA)2 as an oxidant. When the
reaction was performed in THF at 70 °C, the isolated yield
of 4a was improved to 34%, together with a significant
amount of 3a (46%, entry 1, Table 1). To our delight, the
formation of 3a was completely suppressed when the
solvent was changed to ClCH2CH2Cl, and the yield of 4a
was increased dramatically to 84%. Pd(OAc)2 was also
effective in catalyzing this reaction, producing 4a in a
compatible yield (82%, entry 3). It was intriguing that
when the oxidant was switched from Cu(TFA)2 to Cu-
(OAc)2, N-acetyl-N-tert-butyl-2-phenyl-1H-indole-3-car-
boxamide 3b was observed to be the only isolatable
product,19 indicating the vital role played by the counter-
anion incyanation (entry 4). Reactionswithotheroxidants
led to either unidentified reaction mixtures or recovery of
1a (see Supporting Information). The loading of Pd(OAc)2
can be reduced to 5 mol %, and the yield of 4a was almost
maintained (entry 5). A control reaction with a stoichio-
metric amount of Pd(TFA)2 in the absence of any copper
The distinctive reactivity of isocyanides makes them
well-known in Lewis or Brønsted acid promoted Passerini
and Ugi mutlicomponent reactions and others.15 They also
serve as isoelectronic equivalents of CO in transition-
metal-catalyzed imination, carboxyamidation, and amidi-
nation reactions of prefunctionalized aryl halides16 or
aromatic CꢀH bonds in the presence of directing groups.17
Recently, we reported an acid-controlled carboxyamida-
tion of 2-unsubstituted indoles via palladium-catalyzed
(12) Ding, S.; Jiao, N. J. Am. Chem. Soc. 2011, 133, 12374.
(13) Kim, J.; Chang, S. J. Am. Chem. Soc. 2010, 132, 10272.
(14) Ren, X.; Chen, J.; Chen, F.; Cheng, J. Chem. Commun. 2011, 47,
6725.
(15) For recent literature reviews on isocyanide chemistry, see: (a)
Wilson, R. M.; Stockdill, J. L.; Wu, X.; Li, X.; Vadola, P. A.; Park, P. K.;
Wang, P.; Danishefsky, S. J. Angew. Chem., Int. Ed. 2012, 51, 2834. (b)
€
van Berkel, S. S.; Bogels, B. G. M.; Wijdeven, M. A.; Westermann, B.;
Rutjes, F. P. J. T. Eur. J. Org. Chem. 2012, 3543. (c) Tobisu, M.; Chatani,
N. Chem. Lett. 2011, 40, 330.
(16) (a) Jiang, H.; Liu, B.; Li, Y.; Wang, A.; Huang, H. Org. Lett.
2011, 13, 1028. (b) Vllaar, T.; Ruijter, E.; Znabet, A.; Janssen, E.; de
Kanter, F. J. J.; Maes, B. U. W.; Orru, R. V. A. Org. Lett. 2011, 13, 6496.
(c) Qiu, G.; He, Y.; Wu, J. Chem. Commun. 2012, 48, 3836.
(17) (a) Wang, Y.; Wang, H.; Peng, J.; Zhu, Q. Org. Lett. 2011, 13,
4604. (b) Wang, Y.; Zhu, Q. Adv. Synth. Catal. 2012, 354, 1902. (c) Zhu,
C.; Xie, W.; Falck, J. R. Chem.;Eur. J. 2011, 17, 12591.
(18) Peng, J.; Liu, L.; Hu, Z.; Huang, J.; Zhu, Q. Chem. Commun.
2012, 48, 3772.
(19) Hu, Z.; Liang, D.; Zhao, J.; Huang, J.; Zhu, Q. Chem. Commun.
2012, 48, 7371.
Org. Lett., Vol. 14, No. 18, 2012
4967

salt provided 4a in an equal yield of 82%. However, a
reaction with only 1 equiv of Cu(TFA)2 resulted in no
desired product formation, indicating that a palladium
catalyst was indispensable. Finally, a reaction with a
stoichiometric amount of Pd(OAc)2 was performed. Inter-
estingly, secondary and tertiary carboxyamidated indole
derivatives 3a and 3b were isolated in 36% and 51% yields,
respectively. This result further proved that the counter-
anion trifluoroacetate (CF3COOꢀ) was necessary in the
desired cyanation product formation (entry 8).
Scheme 2. Scope of Indole Substratesa
Next, the scope of indole derivatives was investigated
under the optimal reaction conditions involving Pd(OAc)2
(5 mol %) and Cu(TFA)2 (1 equiv), in ClCH2CH2Cl at
70 °C. 2-Phenyl substituted indoles with N-methyl and
N-benzyl groups were cyanated smoothly in 76% and 72%
yields, respectively (4b and 4c, Scheme 2). However, N-Boc
protected 2-phenylindole furnished only a trace amount of
the product even when the reaction time was extended to
30 h (4d), suggesting that the electron density on indole was
crucial. 5-Methyl-2-phenylindole was cyanated in 78%
yield (4e) under the standard conditions. In contrast, for
an electron-deficient 5-fluoro substituted analogue, addi-
tion of PivOH (1.0 equiv) was necessary to drive the
reaction to completion at 80 °C (4f, 51%). To some extent,
the reaction was sensitive to methyl substitution on the
ortho position of the 2-phenyl ring, delivering 4i in a much
lower yield (51%) than a meta (4h, 78%) or para (4g, 84%)
counterpart. Other functional groups on the 2-phenyl ring,
including Cl and OMe, also survived the reaction condi-
tions, giving corresponding indole-3-carbonitriles 4j and
4kin82% and 76% yields, respectively. Moreover, 2-alkyl-
substituted indoles were also suitable substrates in the
current C3 cyanation protocol, albeit in lower yields.
Various acyclic and cyclic 2-alkyl-1H-indole-3-carboni-
triles (4mꢀs) were produced in moderate to good yields.
2-(Thiophen-2-yl)-1H-indole was cyanated to its corre-
sponding indole-3-carbonitrile 4t in 69% yield. However,
in the caseof electron-deficient pyridine substituedindoles,
only 2-(pyridin-2-yl)-1H-indole was able to be cyanated in
21% yield (4w), probably due to coordination of the
pyridine nitrogen with the palladium catalyst.
a Reaction conditions: 1 (0.4 mmol), 2a (0.48 mmol), Pd(OAc)2
(5 mol %), Cu(TFA)2 (1.0 equiv), ClCH2CH2Cl (2.0 mL), air, sealed
tube, 70 °C. bPivOH (1.0 equiv) was added, 80 °C.
Scheme 3. Scope of Electron-Deficient Arenesa
a Reaction conditions: 5 (0.4 mmol), 2a (0.60 mmol, 1.5 equiv),
Pd(OAc)2 (5 mol %), Cu(TFA)2 (1.2 equiv), TsOH (0.5 equiv), toluene
(2.0 mL), air, sealed tube, 120 °C, isolated yields of 6.
To explore the feasibility of the current protocol in the
cyanation of electron-deficientarenes, 2-phenylpyridine 5a
was first used as a model substrate. Harsher reaction
conditions employing the same palladium catalyst and a
copper oxidant in the presence of 0.5 equiv of TsOH in
toluene at 120 °C were identified, furnishing a monocya-
nated product 6a in a moderate yield (41%, Scheme 3).
Substituents with varied electronic properties, such as
methyl (6b), acetyl (6c), chloro (6d), and phenyl (6e), on
the meta and para positions of the phenyl ring were
tolerable under the reaction conditions, albeit in low to
acceptable yields. Quinoline could act as a directing group
as well, delivering the corresponding aryl nitrile 6g in 45%
yield. In addition, benzo[h]quinoline demonstrated the high-
est efficiency in the current cyanation reaction (66%, 6h).
To confirm the origin of the carbon in the cyano group,
13C-labeled 1-adamantanylisocyanide 2b (see Supporting
Information) was synthesized and subjected to the standard
reaction conditions in place of 2a. As anticipated, 4a with
complete 13C incorporation was formed in 86% yield,
indicating that the CN in the product was derived from
isocyanide. Another tertiary amine derived isocyanide 2c
was also a suitable substrate to deliver a CN group to C3 of
1a (eq 1, Scheme 4). The rest of 2c was detected as an
inseparable mixture of 1-(4-fluorophenyl)-2-methyl-pro-
pene 7 and 3-(4-fluorophenyl)-2-methyl-propene 8 in a 2:1
ratio as indicated by 1H NMR, indicating a tertiary carbon
cation might be an intermediate. In contrast, when second-
ary amine and aniline derived isocyanides were used as
substrates, only the corresponding carboxyamidated pro-
ducts were obtained (see Supporting Information).
Substrate 1x bearing a phenolic hydroxy group on the
meta position of the 2-phenyl ring was employed as a sub-
strate under the standard conditions (eq 2). Surprisingly,
4968
Org. Lett., Vol. 14, No. 18, 2012

lead the reaction pathway to direct reductive elimination
from an intermediate similar to B, giving a secondary and/
or tertiary carboxyamide. Oxidation of Pd(0) by Cu(II)
regenerates the active Pd(II) catalyst.
Scheme 4. Mechanistic Studies
Scheme 5. Proposed Reaction Mechanism
In summary, we have developed an unprecedented
palladium-catalyzed cyanation of aromatic CꢀH bonds
by using tertiary amine derived isocyanide as a novel cyano
source. Cu(TFA)2 was used as a requisite stoichiometric
oxidant, in which the trifluoroacetate counteranion plays
a vital role in promoting cyanation product formation.
Both electron-rich 2-alkyl(aryl)indoles and electron-deficient
2-arylpyridine derivatives/analogues can be cyanated.
Mechanistic studies suggest that a key electrophilic imidoyl
palladium(II) intermediate is involved followed by the CꢀN
bond breakage to give the cyanation product and a tertiary
carbon cation-based fragment. The carbon cation inter-
mediate can be partially trapped by an inbuilt phenolic
nucleophile.
together with the normal cyanated product 4x, another
cyanated product 4x0 with a tert-butyl group attached to
the phenol oxygen was also isolated in 10% yield, most
likely, resulting from the tert-butyl carbon cation being
trapped by the free phenol. In addition, substrate 1y
containing a secondary alcohol was initially designed to
trap the carbon cation intermediate. Instead of cyanation
product formation, cyclic imidic ester 10 was isolated in
61% yield, suggesting that a cyclic imidoyl palladium(II)
intermediate was formed rather than the in situ generation
of a cyanide anion (eq 3).
Based on the results of mechanistic studies, a plausible
reaction mechanism is proposed in Scheme 5. Initially,
electrophilic palladation of Pd(II) on C3 of the indole gives
the σ-indolylpalladium(II) intermediate A. The following
migratory insertion of isocyanide generates the key im-
idoyl palladium(II) intermediate B, in which the loosely
coordinated trifluoroacetate makes the palladium center
more electrophilic. The high electrophilicity of Pd(II) and
the stability of the tertiary carbon cation are the driving
forces to break the CꢀN bond, leading to the cyanation
product, palladium(0) species, and 2-methylpropene by
proton elimination via a tert-butyl carbon cation inter-
mediate. The lack of any one of the “driving forces” will
Acknowledgment. We are grateful to the National
Science Foundation of China (21072190) for financial
support of this work.
Supporting Information Available. Experimental pro-
cedure, and characterization data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 18, 2012
4969