Modular syntheses of chiral and achiral C,N-chelated Pd(ii)-pyridinylidenes

Article abstract of DOI:10.1039/b915085d

Modular, high-yielding routes to achiral and chiral, C,N-chelated pyridinylidene complexes of Pd(ii) have been developed which rely on oxidative addition of either chloropyridinium imines or aldehydes to Pd(0), respectively, and, in the latter case, subsequent Schiff base formation with chiral primary amines. These protocols succinctly circumvent many of the problems currently associated with the synthesis of chelating pyridinylidenes.

Full text of DOI:10.1039/b915085d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/dalton | Dalton Transactions
Modular syntheses of chiral and achiral C,N-chelated Pd(II)-pyridinylidenes†
Elizabeth T. J. Strong, Jacquelyn T. Price and Nathan D. Jones*
Received 27th July 2009, Accepted 17th August 2009
First published as an Advance Article on the web 3rd September 2009
DOI: 10.1039/b915085d
Modular, high-yielding routes to achiral and chiral, C,N-
chelated pyridinylidene complexes of Pd(II) have been de-
veloped which rely on oxidative addition of either chloropy-
ridinium imines or aldehydes to Pd(0), respectively, and, in
the latter case, subsequent Schiff base formation with chiral
primary amines. These protocols succinctly circumvent many
of the problems currently associated with the synthesis of
chelating pyridinylidenes.
(e.g., in D). In principle, the carbene donor may also be “remote”,
i.e., meta and para to N (e.g., in C). Furthermore, these types
of pyridinylidenes are expected to possess enhanced s-donor and
p-acceptor properties compared to both pyridin-2-ylidenes and
NHCs derived from imidazolium cations.9
We have now uncovered two new modular routes to families of
chiral and achiral, chelating pyridin-n-ylidenes that rely on oxida-
tive addition of chloropyridinium aldehydes (n = 2; Scheme 1) or
imines (n = 3; Scheme 2) to Pd(0).
Pyridinylidenes (A, Fig. 1) are N-heterocyclic carbenes (NHCs)
formally derived by deprotonation of pyridinium precursors.‡
They are stronger s-donors and p-acceptors1 than “Arduengo-
type” NHCs derived from imidazolium cations (B) because they
have only one, as opposed to two, N-atom alpha to the carbene
centre.2 These enhanced properties make them attractive ancillary
ligands for applications in catalysis.3,4 However, selective and
convenient routes to metal–pyridinylidenes are rare, especially for
chelating varieties.5
Fig. 1 Compounds relevant to this work.‡
Scheme 1 Synthesis of chiral, C,N-chelated Pd(II)–pyridin-2-ylidenes;
all anions (not shown) are trifluoromethanesulfonate, TfO-. (i)
Pd2(dba)3·CHCl3 (0.5 eq.), MeCN, 60–70 ◦C, 1 h; (ii) Me3SiOTf (1 eq.),
MeCN, rt, ca. 2 h; (iii) (1) Pd2(dba)3·CHCl3 (0.5 eq.), MeCN, 60–70 ◦C,
To the best of our knowledge, only a single chiral pyridin-4-
ylidene has been reported (C, also incidentally bidentate),6 and
there are no chiral pyridin-n-ylidenes (n = 2, 3). Because C is based
specifically on nicotine, it does not point the way to a general
modular route for the synthesis of families of chiral, chelating
pyridinylidenes.
˚
1 h, then (2) Me3SiOTf, rt, 1 h; (iv) NH2R (1.2 eq.), MeCN, 4 A molecular
sieves, rt, ca. 12 h.
Bidentate pyridinylidenes are troublesome to make because cre-
ation of the necessary pyridinium cation requires treatment with
an electrophile that indiscriminately attacks the other donor (e.g.,
in methylations of 4¢,2-bipyridine).7 This necessitates laborious
protection and deprotection steps and/or separations. The most
successful circumvention of this problem has been to employ
2-pyridylpyridiniums, the syntheses of which simultaneously in-
stall both the pyridinium cation and second donor.8 However,
this strategy limits the second donor to pyridine and forces the
carbene donor ortho to N in the resultant metal–pyridinylidene
Department of Chemistry, The University of Western Ontario, London,
Ontario, Canada N6A 5B7. E-mail: njones26@uwo.ca; Fax: +1 (1)519
661 3022; Tel: +1 (1)519 661 2111 ext. 84456
† Electronic supplementary information (ESI) available: Synthetic details
and full characterization data for all new compounds. CCDC reference
numbers 732628–732630. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/b915085d
Scheme 2 Synthesis of C,N-chelated Pd(II)–pyridin-3-ylidenes; all anions
(not shown) are TfO . (i) p-anisidine (1.2 eq.), MeCN, 4 A molecular
sieves, rt, 3 h; (ii) Pd2(dba)3·CHCl3 (0.5 eq.), MeCN–toluene, rt, 5 h;
(iii) Me3SiOTf (1 eq.), MeCN, rt, ca. 3 h.
-
˚
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 9123–9125 | 9123
©

View Article Online
This general approach—a hybrid of Raubenheimer’s synthe-
sis of monodentate Pd(II)–pyridinylidenes3,4 and Uozomi’s so-
called “metal-introduction” route to N,C,N-pincer complexes10—
conferred advantages hitherto not fully realized in metal–
pyridinylidene chemistry: (i) it gave control over the location of the
carbene donor (see below for n = 3); (ii) by temporally separating
pyridinium generation from the installation of the second donor,
it obviated the need for pre-protection of this donor and/or for
laborious separations; (iii) it was synthetically high-yielding; and
(iv) in the specific case of n = 2, it placed the diversification step at
the end of the synthetic sequence which, in principle, allowed the
trivial generation of libraries of chiral Pd(II) complexes based on
the pool of readily available chiral primary amines.
was oriented such that the O-atom made a short contact with the
˚
˚
metal (ca. 2.88 A; RvdW Pd + O = 3.15 A).
The molecular structure of the cation of 7 is shown in Fig. 3.¶
In this compound, the Pd coordination and pyridinylidene planes
were approximately coincident on account of the coordinated
˚
imine. The Pd–C bond length was 2.004(4) A, which was slightly
longer than that of 3. The unfavourable steric interaction between
the N–Me group and the cis-coordinated MeCN ligand resulted
in the distortion of this MeCN–Pd axis from linearity (167.6(4)◦)
and the enlargement of the angle between these two ligands
˚
(101.56(17) A).
Pyridinium [1–Cl][OTf] was readily accessed by treatment of
2-chloro-3-formylpyridine with MeOTf in Et2O at rt. Oxidative
addition of [1–Cl][OTf] to Pd2(dba)3 (dba = dibenzylidene ace-
tone) in MeCN at 60–70 ◦C gave the Pd(II)–pyridin-2-ylidene
chloride 2, from which the halide could be removed by the action of
AgOTf or Me3SiOTf to give 3. Alternatively, 3 could be generated
readily in a single pot without the isolation of 2 by sequential
addition of Pd2(dba)3 and Me3SiOTf to [1–Cl][OTf]. Subsequent
condensation of 2 and 3 with (R)-(+)-a-methylbenzylamine and
(R)-(+)-1-(1-naphthyl)-ethylamine gave the chiral C,N-chelated
Pd(II)–pyridinylidenes 4–7 in moderate to high yields (70–93%).
Both geometrical isomers of 4 and 5, in which the chloride and
pyridinylidene ligands were trans- or cis-disposed, were observed
spectroscopically (see ESI†). It was not possible to conduct the
condensation reactions prior to installation of the metal because
these reactions gave products of nucleophilic aromatic substitution
in addition to the intended imines.
Fig. 3 ORTEP representation of the molecular structure of the cation of
7 (30% ellipsoids) with all H-atoms except for the one associated with the
˚
chiral centre removed for clarity. Selected bond distances (A) and angles
(◦): Pd1–C31 2.004(4), Pd1–N1 2.086(4), Pd–N2 2.028(4), C31–Pd1–N2
101.56(17), C31–Pd1–N4 80.85(17), N4–Pd1–N1 93.85(15), N2–Pd1–N1
83.83(16).
We were unable to crystallize either 4 or 5. However, an achiral
analogue (8) was easily accessed by the reaction of 2 and para-
anisidine. Crystals of 8 contained two independent molecules in
the asymmetric unit which had significantly different Pd–C bond
The molecular structure of the cation of 3 is shown in Fig. 2.§
It featured an approximately square planar metal coordination
geometry, to which the pyridinylidene ring was roughly perpendic-
˚
lengths (1.932(4) vs. 1.991(4) A). The molecular structure of one of
˚
ular. This compound contained a short Pd–C bond of 1.979(2) A,
the cations is shown in Fig. 4.ꢀ We have found only a single shorter
consistent with the analogous bonds in Raubenheimer’s Pd(II)–
˚
4
Pd–C bond (1.901(3) A) in the literature, which belonged to a
˚
pyridinylidenes, which were ca. 1.98–2.03 A, and with what
cyclometallated bis(imidazolin-2-ylidene) complex.12 The average
Pd–C bond length in 8 was shorter the corresponding distances
in 3 and 7 and on the short end of those reported for other bi-
and tridentate Pd(II)–pyridinylidenes (ca. 1.942–1.994).6,13 Steric
conflict between the N–Me group and the cis ligand also featured
in this complex and, as a result, the pyridinylidene ring and metal
is typically observed for Pd(II) complexes of Arduengo-type
11
˚
NHCs (avg. Pd–C, 1.99 A). The strong trans influence of the
pyridinylidene donor was clearly evident in the elongated Pd–N
˚
bond for the trans MeCN ligand (2.078(4) A) over those for the
˚
cis (1.989(4) and 1.986(4) A). The corresponding C–C–N–Pd axes
were very nearly linear. In addition, the pendant aldehyde group
Fig. 4 ORTEP representation of the molecular structure of one of the
crystallographically independent cations of 8 (30% ellipsoids) with all
Fig. 2 ORTEP representation of the molecular structure of the cation
of 3 (30% ellipsoids) with◦all H-atoms removed for clarity. Selected bond
˚
H-atoms removed for clarity. Selected bond distances (A) and angles
˚
distances (A) and angles ( ): Pd1–C2 1.979(2), Pd1–N2 1.989(4), Pd1–N3
(◦): Pd2–C21 1.932(4), Pd2–N6 2.030(4), Pd2–N4 2.042(3), C21–Pd2–Cl2
98.51(13), C21–Pd2–N4 79.94(16), N6–Pd2–N4 96.61(14), N6–Pd2–Cl2
84.93(9), N4–Pd2–C21–C25 14.9(3), Cl2–Pd2–C21–N5 26.9(4).
2.078(4), Pd1–N4 1.986(4), C2–Pd1–N2 88.86(14), C2–Pd1–N4 89.72(14),
N4–Pd1–N3 90.94(16), N2–Pd1–N3 90.53(16).
9124 | Dalton Trans., 2009, 9123–9125
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
coordination planes were not coincident (Cl2–Pd2–C21–N5 =
26.9(4)◦).
Research Council of Canada (NSERC) for financial support of
this work.
The synthesis of analogous Pd(II) complexes of pyridin-3-
ylidenes required a different approach (Scheme 2). Although
3-chloro-4-formyl-1-methylpyridinium [9–Cl][OTf] could be made
easily from the corresponding pyridine and MeOTf, attempted
reaction of it with Pd2(dba)3 in MeCN either at room or elevated
temperature gave only Pd black. Therefore, [9–Cl][OTf] was
instead condensed with p-anisidine in the first step to give 10.
Schiff base formation prior to oxidative addition was possible
in this case because of the attenuated susceptibility of [9–Cl]+ to
nucleophilic aromatic substitution compared to [1–Cl]+. However,
we were limited strictly to aromatic amines because attempted
condensations with more strongly nucleophilic aliphatic amines,
such as the chiral ones we had used for the pyridin-2-ylidenes,
also led to aromatic substitution chemistry. Reaction of 10 with
Pd2(dba)3 cleanly gave the desired Pd(II)–chloride, 11, from which
the halide could be removed by standard methods to yield the
target C,N-chelated Pd(II)–pyridin-3-ylidene, 12. Presumably, the
directing effect of the preinstalled imine aided the oxidative
addition reaction in this approach.
All Pd(II)–pyridin-n-ylidenes (n = 2, 3) were yellow or orange
microcrystalline powders that were inert to O2 and H2O: their 1H
NMR spectra did not change after exposure of CD3CN solutions
of the compounds to air for 24 h at rt, even in the presence of added
H2O. Decomposition to Pd black was seen only after 4 d under
these conditions. However, the solid compounds readily absorbed
water from the air and it was necessary to store them under
dry N2.
Notes and references
‡ Deprotonated pyridinium cations may be represented either by ylidic
(zwitterionic) or carbene resonance forms when the N-atom is ortho (I
and II, respectively) and para (I¢ and II¢, respectively) to the site of
deprotonation (both have been used in the literature). However, when the
N-atom is meta (I¢¢), a valid carbene Lewis representation does not exist.
For this reason, and to maintain consistency, the metal–pyridin-n-ylidenes
described in this paper (n = 2, 3) are shown in their ylidic forms.
§ Crystal data for 3: Chemical formula, C15H16F6N4O7PdS2; M, 648.84;
¯
˚
˚
cryst. syst., triclinic; space◦group, P1; a◦, 7.8534(16) A; b, 8.9797(18) A;
◦
3
˚
˚
c, 17.407(4) A; a, 80.77(3) ; b, 77.31(3) ; g , 77.95(3) ; V, 1162.7(4) A ;
T, 150(2) K; Z, 2; reflns collected, 5621; independent reflns (Rint), 4025
(0.0341); final R indices (I > 2s(I)), R1 = 0.0441, wR2 = 0.1084.
¶ Crystal data for 7: Chemical formula, C29H30F6N6O6PdS2; M, 843.1;
˚
cryst. syst., orthorhombic; space group, P212121; a, 12.968(3) A; b,
3
˚
˚
˚
13.049(3) A; c, 22.681(5) A; V, 3838.0(13) A ; T, 150(2) K; Z, 4; reflns
collected, 8748; independent reflns (Rint), 8748 (0.0000); final R indices (I >
2s(I)), R1 = 0.0508, wR2 = 0.1325.
ꢀ Crystal data for 8: Chemical formula, C36H37Cl2F6N7O8Pd2S2; M,
¯
˚
1157.55; cryst. syst., triclinic; space group, P1; a,◦ 13.107(3) A; b,
◦
◦
˚
˚
13.283(3) A; c, 14.155(3) A; a, 98.15(3) ; b, 115.18(3) ; g , 98.29(3) ; V,
3
˚
2149.3(7) A ; T, 150(2) K; Z, 2; reflns collected, 11 969; independent reflns
(Rint), 7561 (0.0208); final R indices (I > 2s(I)), R1 = 0.0400, wR2 = 0.0965.
1 R. Gleiter and R. Hoffmann, J. Am. Chem. Soc., 1968, 90,
5457.
2 C. Boehme and G. Frenking, Organometallics, 1998, 17, 5801; C.
Kocher and W. A. Herrmann, J. Organomet. Chem., 1997, 532, 261.
3 S. K. Schneider, C. F. Rentzsch, A. Kruger, H. G. Raubenheimer and
W. A. Herrmann, J. Mol. Catal. A: Chem., 2007, 265, 50.
4 S. K. Schneider, P. Roembke, R. J. Julius, H. G. Raubenheimer and
W. A. Herrmann, Adv. Synth. Catal., 2006, 348, 1862.
5 O. Schuster, L. Yang, H. G. Raubenheimer and M. Albrecht, Chem.
Rev., 2009, 109, 3445.
6 M. Albrecht and H. Stoekli-Evans, Chem. Commun., 2005, 4705.
7 P. Castan, F. Dahan, S. Wimmer and F. L. Wimmer, J. Chem. Soc.,
Dalton Trans., 1990, 2971.
8 J. S. Owen, J. A. Labinger and J. E. Bercaw, J. Am. Chem. Soc., 2004,
126, 8247.
We investigated some reactions of these compounds with
organometallic reagents because of the scarce reports of metal–
pyridinylidene complexes bearing organic ligands.8,14 Allyl com-
plexes 13 and 14 were made by equimolar metathesis reactions
between nBu3Sn(allyl) and the chlorides 5 and 11, respectively
(Fig. 5). They were brown and orange microcrystalline powders
that were much less stable than the corresponding chloro and
acetonitrile compounds, 4–7, 11 and 12, and showed loss of
the allyl group upon exposure to normal atmosphere. Prolonged
heating resulted in their decomposition to Pd black (within ca. 1 h
at 50–60 ◦C).
9 G. Heydenrych, M. von Hoffgarten, E. Stander, O. Schuster, H. G.
Raubenheimer and G. Frenking, Eur. J. Inorg. Chem., 2009, 1892.
10 K. Takenaka, M. Minakawa and Y. Uozumi, J. Am. Chem. Soc., 2005,
127, 12273.
11 E. Baba, T. R. Cundari and I. Firkin, Inorg. Chim. Acta, 2005, 358,
2867.
12 N. Stylianides, A. A. Danopoulos, D. Pugh, F. Hancock and A. Zanotti-
Gerosa, Organometallics, 2007, 26, 5627.
13 A. Poulain, A. Neels and M. Albrecht, Eur. J. Inorg. Chem., 2009, 1871;
H. Meguro, T. Koizumi, T. Yamamoto and T. Kanbara, J. Organomet.
Chem., 2008, 693, 1109; N. A. Piro, J. S. Owen and J. E. Bercaw,
Polyhedron, 2004, 23, 2797.
Fig. 5 Pd(II) allyl complexes of pyridin-n-ylidenes.
14 G. Song, Y. Zhang, Y. Su, W. Deng, K. Han and X. Li, Organometallics,
´
2008, 27, 6193; E. Alvarez, S. Conejero, P. Lara, J. A. Lopez, M.
We thank Dr Michael Jennings (UWO) for collection of
crystallographic data, and the Natural Sciences and Engineering
Paneque, A. Petronilho, M. L. Poveda, D. del Rio, O. Serrano and
E. Carmona, J. Am. Chem. Soc., 2007, 129, 14130.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 9123–9125 | 9125
©