Palladium-pincer complex catalyzed C-C coupling of allyl nitriles with tosyl imines via regioselective allylic C-H Bond Functionalization

Article abstract of DOI:10.1021/ol801070n

(Chemical Equation Presented) A mechanistically new palladium-pincer complex catalyzed allylation of sulfonimines is presented. This reaction involves C-H bond functionalization of allyl nitriles under mild conditions. The reaction proceeds with a high regioselectivity, without allyl rearrangement of the product. Modeling studies indicate that the carbon-carbon bond formation process proceeds via (η¹-allyl)palladium pincer complex intermediates.

Full text of DOI:10.1021/ol801070n

ORGANIC
LETTERS
2008
Vol. 10, No. 13
2881-2884
Palladium-Pincer Complex Catalyzed
C-C Coupling of Allyl Nitriles with
Tosyl Imines via Regioselective Allylic
C-H Bond Functionalization
Juhanes Aydin and Ka´lma´n J. Szabo´*
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm UniVersity,
SE-106 91 Stockhom, Sweden
kalman@organ.su.se
Received May 9, 2008
ABSTRACT
A mechanistically new palladium-pincer complex catalyzed allylation of sulfonimines is presented. This reaction involves C-H bond
functionalization of allyl nitriles under mild conditions. The reaction proceeds with a high regioselectivity, without allyl rearrangement of the
product. Modeling studies indicate that the carbon-carbon bond formation process proceeds via (η1-allyl)palladium pincer complex intermediates.
Selective palladium-catalyzed allylic C-H bond activation
based substitution reactions are attractive synthetic tools,
generating a large added value in organic transformations.1
Although several catalytic methods using nucleophiles to
perform allylic C-H bond functionalizations have been
reported in the literature,1f–l application of carbon electro-
philes in these processes remained unexplored.
Electrophilic allylation of imines based on this method is
of particular interest, as these transformations would open
inexpensive new routes for synthesis of functionalized amines
(1) (a) Dyker, G. Handbook of C-H Transformations: Applications in
Organic Synthesis; Wiley-VCH: Weinheim, 2005. (b) Tsuji, J. Palladium
and amino acids,2,3 which represent an important class of
natural products and drug intermediates.2e
Reagents and Catalysts. New PerspectiVes for the 21st Century; Wiley:
Chichester, 2004. (c) Dick, A. R.; Sanford, M. S. Tetrahedron 2006, 62,
Our previous studies3 have shown that palladium-pincer
2440. (d) Ritleng, V.; Sirlin, C.; Pfeffer, M. Chem. ReV. 2002, 102, 1731.
complexes4,5 efficiently and selectively catalyze the allylation
(e) Hosokawa, T.; Murahashi, S.-I. Acc. Chem. Res. 1990, 23, 49. (f) Chen,
M. S.; Prabagaran, N.; Labenz, N. A.; White, M. C. J. Am. Chem. Soc.
of imines. However, in these applications organometallic
2005, 127, 6970. (g) Delcamp, J. H.; White, M. C. J. Am. Chem. Soc. 2006,
128, 15076. (h) Fraunhoffer, K. J.; White, M. C. J. Am. Chem. Soc. 2007,
129, 7274. (i) Franze´n, J.; Ba¨ckvall, J.-E. J. Am. Chem. Soc. 2003, 125,
6056. (j) Fix, S. R.; Brice, J. L.; Stahl, S. S. Angew. Chem., Int. Ed. 2002,
41, 164. (k) Bystro¨m, S. E.; Larsson, E. M.; Åkermark, B. J. Org. Chem.
1990, 55, 5674. (l) Zanoni, G.; Porta, A.; Meriggi, A.; Franzini, M.; Vidari,
G. J. Org. Chem. 2002, 67, 6064.
(2) (a) Ding, H.; Friestad, G. K. Synthesis 2005, 2815. (b) Taggi, A. E.;
Hafez, A. M.; Lectka, T. Acc. Chem. Res. 2003, 36, 10. (c) Alvaro, G.;
Savoia, D. Synlett 2002, 651. (d) Fernandes, R. A.; Stimac, A.; Yamamoto,
Y. J. Am. Chem. Soc. 2003, 125, 14133. (e) Viso, A.; Pradilla, R. F. d. l.;
Garcia, A.; Flores, A. Chem. ReV. 2005, 105, 3167
.
10.1021/ol801070n CCC: $40.75
Published on Web 06/05/2008
2008 American Chemical Society

substrates,3 such as stannanes and boronates, were employed
as allylating reagents (eq 1).3 We have now found that allyl
nitriles can be used directly for regioselective allylation of
sulfonimines using pincer-complex catalysts 1a-c5 (eq 2)
under mild conditions (typically at rt) in the presence of a
weak base, NaHCO3, and molecular sieves. Although several
pincer complexes with weakly coordinating counterions
(entries 1-3) displayed high catalytic activity, we concen-
trated on exploring the synthetic scope of 1a,5a as a large
variety of analogue PCP complexes,5a,c including chiral
ones,3c,f,5b have recently become available and employed in
organic synthesis.3–5
Table 1. Palladium-Pincer Complex Catalyzed Coupling of
Allyl Nitriles with Sulfoniminesa
Allyl nitrile (2a) reacted rapidly with various aromatic
(3a-d,g), vinyl (3e), and alkyl (3f) sulfonimines (Table 1).
The reaction times required to complete the catalytic allylic
substitution reactions were strongly dependent on the sul-
fonimine substrates. Sulfonimines with electron-withdrawing
groups (3b,d) reacted faster than 3a (cf. entries 1 and 5),
while methoxy derivative 3c (entry 6) was allylated as
quickly as the parent compound 3a. The required reaction
times with vinyl and alkyl sulfonimines (3e and 3f) were
considerably longer (entries 8 and 9) than with aromatic ones.
The regioselectivity of the reaction is excellent, as the
branched allylic product is formed exclusively in the
substitution reactions.
The only exception is application of Cs2CO3 (or other
strong bases) instead of NaHCO3 (entry 4), where the primary
coupling product (4a) undergoes allylic rearrangement to give
4b.
The allylation reactions could be extended to substituted
allyl nitriles 2b-d incorporating internal double bonds
(entries 10-16). The catalytic process with 2b and 2d
(entries 10-13 and 16) proceeded slower than with 2a and
2c. However, the high regioselectivity and the integrity of
the allyl system could be maintained. For symmetrical
substrate 2c, the catalytic reaction could be stopped after
(3) (a) Solin, N.; Kjellgren, J.; Szabo´, K. J. J. Am. Chem. Soc. 2004,
126, 7026. (b) Solin, N.; Kjellgren, J.; Szabo´, K. J. Angew. Chem., Int. Ed.
2003, 42, 3656. (c) Aydin, J.; Kumar, K. S.; Sayah, M. J.; Wallner, O. A.;
Szabo´, K. J. J. Org. Chem. 2007, 72, 4689. (d) Solin, N.; Wallner, O. A.;
Szabo´, K. J. Org. Lett. 2005, 7, 689. (e) Wallner, O. A.; Szabo´, K. J. Chem.
Eur. J. 2006, 12, 6976. (f) Wallner, O. A.; Olsson, V. J.; Eriksson, L.;
Szabo´, K. J. Inorg. Chim. Acta 2006, 359, 1767. (g) Wallner, O. A.; Szabo´,
K. J. Org. Lett. 2004, 6, 1829.
(4) Leading reviews: (a) The Chemistry of Pincer Compounds; Morales-
Morales, D., Jensen, C. M., Eds.; Elsevier: Amsterdam, 2007. (b) Albrecht,
M.; van Koten, G. Angew. Chem., Int. Ed. 2001, 3750. (c) Boom, M. E. v.
d.; Milstein, D. Chem. ReV. 2003, 103, 1759. (d) Dupont, J.; Consorti, C. S.;
Spencer, J. Chem. ReV. 2005, 105, 2527. (e) Dupont, J.; Pfeffer, M.; Spencer,
J. Eur. J. Inorg. Chem. 2001, 1917. (f) Szabo´, K. J. Synlett 2006, 811. (g)
Singleton, J. T. Tetrahedron 2003, 59, 1837. (h) Bedford, R. B. Chem.
Commun. 2003, 1787. (i) Beletskaya, I. P.; Cheprakov, A. V. J. Organomet.
Chem. 2004, 689, 4055. (j) Benito-Garagorri, D.; Kirchner, K. Acc. Chem.
Res. 2008, 41, 201.
a Catalyst 1 (5 mol %), NaHCO3, 2, and 3 in THF were reacted at the
given temperatures and reaction times. b Diastereomeric ratio. c Isolated yield
(%). d Cs2CO3 was employed as base. e E/Z ratio.
(5) Preparation and use of 1a-c: (a) Bedford, R. B.; Draper, S. M.;
Scully, P. N.; Welch, S. L. New J. Chem. 2000, 24, 745. (b) Baber, R. A.;
Bedford, R. B.; Betham, M.; Blake, M. E.; Coles, S. J.; Haddow, M. F.;
Hurthouse, M. B.; Orpen, A. G.; Pilarski, L. T.; Pringle, P. G.; Wingad,
R. L. Chem. Commun. 2006, 3880. (c) Morales-Morales, D.; Grause, C.;
Kasaoka, K.; Redo´n, R.; Cramer, R. E.; Jensen, C. M. Inorg. Chem. Acta
2000, 300, 958. (d) Yao, Q.; Kinney, E. P.; Zheng, C. Org. Lett. 2004, 6,
2997. (e) Yao, Q.; Sheets, M. J. Org. Chem. 2006, 71, 5384. (f) Dupont,
J.; Beydoun, N.; Pfeffer, M. J. Chem. Soc., Dalton Trans. 1989, 1715. (g)
Aydin, J.; Kumar, K. S.; Eriksson, L.; Szabo´, K. J. AdV. Synth. Catal. 2007,
349, 2585. (h) Johansson, R.; Wendt, O. Dalton Trans, 2007, 488. See also
ref 3.
substitution of one of the allylic carbons. Thus, desymme-
trization of 2c (entries 14 and 15) could be achieved,
affording 4k,l in high yields. Because of the mild conditions,
the reaction tolerates many functional groups (CN, NTs, F)
including even the nitro group (entry 13).
Application of NaHCO3 (1 equiv) is a very attractive
feature of the process, as stronger bases induce allylic
rearrangement of the product (e.g., entry 4) and degradation
2882
Org. Lett., Vol. 10, No. 13, 2008

of sensitive substrates, such as 3f,g.6 In fact, functionalized
allyl nitriles and related compounds extremely easily undergo
base-catalyzed allylic rearrangement.6 The applied NaHCO3
is one of the few very weak bases which does not induce
such a rearrangement, and therefore, the integrity of the allyl
system in 4a,c-m can be maintained under the applied
reaction conditions. The reactions proved to be rather
sensitive to moisture, and therefore, we employed molecular
sieves, which efficiently dried the reaction medium. The
catalytic transformations still proceed in the absence of
molecular sieves or with substoichiometric amounts of
NaHCO3. However, the reaction times are elongated. In fact,
the reaction of allyl nitrile (2a) and imine 3a (c.f., entry 1)
could also be achieved in the absence of base; however, these
reactions were very slow and poorly reproducible.
Although the reaction proceeds with excellent regioselec-
tivity (entries 1-16) and trans stereoselectivity for the double
bond (entries 10-15), the diastereoselectivity is poor. A
similarly low diastereoselectivity was reported3g for the
related allylstannane based-reactions (eq 1), when the ally-
lating reagent incorporated a nitrile functional group. The
low diastereoselectivity can be explained by either the low
level of stereodiscrimination in the coupling step or by
epimerization of product 4 under the applied reaction
conditions. When a diastereomerically enriched product (such
as 4l) was exposed to the reaction conditions of the coupling
reaction, further change of the diastereomeric ratio could not
be observed. Therefore, it can be concluded that the low
diastereoselectivity of the reaction can be ascribed to the
similar activation energy of the formation of the two
diastereomers of 4 under the pincer-complex catalyzed
coupling reaction.
prerequisite of the successful coupling reaction. The mech-
anism of the coupling reaction (eq 3) probably involves
coordination of the allyl cyanide to the pincer complex
catalyst affording complex 5. This process is certainly aided
by NaHCO3, facilitating the deprotonation of 2a. Strong bases
are known to deprotonate allyl nitriles and related species.6
However, NaHCO3 alone is not able to deprotonate 2a (even
less 2b or 2d), and therefore, the deprotonation process is
assisted by palladium. We have considered 5a-d as possible
structures for the allyl nitrile anion coordinated pincer
complex intermediates. DFT modeling studies (B3PW91/6-
31G(d) level) indicate that the η1-allylpalladium structures
(5a-c) are much more stable (up to 6.7 kcal·mol-1) than
the N-coordinated form 5d. Considering that in the most
stable forms (5a,b), the nitrile group is in the terminal
position, the new C-C bond is created3e between the
substituted γ-carbon of the allyl moiety and 3 (6), which
explains the observed regiochemistry of the process. The
mechanism and selectivity of η1-allylpalladium complex medi-
ated electrophilic substitutions is well established.3a,d,e,5e,h In
these processes other potential electrophiles may also react
with 5.3a Therefore, it is very important to exclude moisture
from these processes, which can be achieved by using
molecular sieves.
Interestingly, under the applied mild reaction conditions
(e.g., entry 1) formation of 4a could not be detected at all
when pincer-complex catalyst 1a was replaced by traditional
palladium catalysts, such as Pd(OCOCF3)2, Pd(OAc)2,
Pd2(dba)3, and Pd(PPh3)4. The absence of the reaction with
Pd(OCOCF3)2 is particularly important from a mechanistic
point of view, as it indicates that the activation effects of
trifluoroacetate complex 1a cannot be explained simply by
its Lewis acid activity. If Lewis acid activation would be
sufficient for coupling of 2a with 3a, then Pd(OCOCF3)2
and pincer complex 1a should have displayed a similar
catalytic activity.
There are a few examples in the literature for transition
metal catalyzed C-H bond activation of nitriles.1a,7 In these
literature studies ruthenium,7a rhodium,7b and palladium7c
catalysts were employed to perform Michael addition of
activated alkyl nitriles. As shown above, commonly used
palladium sources are not efficient to catalyze the allylation
of sulfonimine 3 with allyl nitriles 2 under the employed
mild conditions. Thus, application of pincer complex 1 is a
Our previous DFT modeling studies3e on the C-C bond
formation between the allyl moiety of palladium pincer
complexes (such as 5a-c) and sulfonimines revealed that the
stereoselectivity of the reaction is dependent on the geometry
of the double bond in the functionalized (e.g., CN) allyl moiety.
On the basis of these results, the relatively small energy
difference between 5a and 5b may account for the poor
stereoselectivity of the allylation process. Accordingly, a
possible way of increasing the stereoselectivity of this C-C
coupling is the optimization of the ligand size in the side arms
of 1 to increase the energy difference between the 5a and 5b
type of intermediates. Formation of (η1-allyl)palladium com-
plexes 5a,b (required for reaction with electrophiles3a,e ) is
favored by the terdentate pincer ligand architecture,3a,b,e
which explains the fact that traditional palladium (i.e.,
nonpincer, Pd(OCOCF3)2) catalysts are inefficient in the
presented transformations. The allylation process is termi-
nated by formation of 7, which after decomplexation affords
product 4a and regenerates the catalyst.
(6) (a) Kisanga, P. B.; Verkade, J. G. J. Org. Chem. 2002, 67, 426. (b)
Yamaguchi, A.; Aoyama, N.; Matsunaga, S.; Shibasaki, M. Org. Lett. 2007,
9, 3387.
(7) (a) Naota, T.; Taki, H.; Mizuno, M.; Murahashi, S.-I. J. Am. Chem.
Soc. 1989, 111, 5954. (b) Paganelli, S.; Schionato, A.; Botteghi, C.
Tetrahedron Lett. 1991, 32, 2807. (c) Stark, M. A.; Richards, C. J.
Tetrahedron Lett. 1997, 38, 5881
.
Org. Lett., Vol. 10, No. 13, 2008
2883

In summary, we have presented the first palladium pincer-
complex catalyzed C-H bond activation based allylation
reaction. The experimental findings, the modeling studies,
and the fact that traditional palladium sources are ineffective
as catalysts suggest that the key intermediate of the reaction
is an (η1-allyl)palladium pincer complex. The presented
mechanistically new catalytic process opens novel synthetic
routes to palladium pincer-complex catalyzed C-H bond
activation reactions.
Acknowledgment. This work was supported by the
Swedish Research Council (VR).
Supporting Information Available: Experimental pro-
1
cedures, NMR data, as well as H and 13C NMR spectra of
products 7a-m, 8, and 10a-c. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL801070N
2884
Org. Lett., Vol. 10, No. 13, 2008