Expanding the family of mesoionic complexes: Donor properties and catalytic impact of palladated isoxazolylidenes

Article abstract of DOI:10.1039/c0dt00027b

Abnormal isoxazolylidene complexes, a new subclass of mesoionic complexes containing an isoxazolium-derived carbene type ligand, have been synthesised via oxidative addition and compared to structurally related mesoionic complexes by using ³¹P NMR spectroscopy as a convenient probe for their donor ability and in catalytic cross-coupling reactions.

Full text of DOI:10.1039/c0dt00027b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/dalton | Dalton Transactions
Expanding the family of mesoionic complexes: donor properties and catalytic
impact of palladated isoxazolylidenes†
Manuel Iglesias and Martin Albrecht*
Received 9th December 2009, Accepted 12th March 2010
First published as an Advance Article on the web 12th April 2010
DOI: 10.1039/c0dt00027b
Abnormal isoxazolylidene complexes, a new subclass of
mesoionic complexes containing an isoxazolium-derived car-
bene type ligand, have been synthesised via oxidative addition
and compared to structurally related mesoionic complexes by
using 31P NMR spectroscopy as a convenient probe for their
donor ability and in catalytic cross-coupling reactions.
diates. Thus, quaternisation of commercially available 4-iodo-
3,5-dimethylisoxazole with MeOTf or [Me3O]BF4 gave the isoxa-
zolium salts 1 in excellent yields (Scheme 1). Subsequent oxidative
addition to Pd(PPh3)4 afforded the carbene complexes 2 as air-
stable, off-white solids.‡ Successful metallation was indicated by
1
the pertinent 13C{ H} NMR signal of the palladium(II)-bound
carbon atom, which appeared as a triplet at 168.0 ppm (2JPC
=
3.9 Hz). The singlet in the 31P NMR spectrum (dP 21.3) is in
agreement with a mutual trans conformation of the two phosphine
ligands in 2.
The discovery and isolation of the first stable carbene by Bertrand
two decades ago,1 and the subsequent extension of this work
to N-heterocyclic carbenes (NHCs) by Arduengo2 triggered the
exploitation of these species as versatile ligands for transition
metal complexes. Pioneering work by Herrmann, Grubbs, and
Nolan illustrate the outstanding impact of NHCs on many
homogeneous catalysts, often outperforming more ubiquitous
phosphine ligands.3 Improved catalyst stability and reactivity
is generally assigned to the more covalent and hence stronger
bonding of NHCs to metal centres, and to the higher donor ability
of carbenes.4 To date a great diversity of NHCs has been developed,
ranging from the normal imidazolylidenes5 to expanded-ring
NHCs6 and to carbenes with reduced heteroatom-stabilisation like
cyclic alkyl amino carbenes7 and abnormal NHCs.8
The absence of heteroatoms adjacent to the carbene carbon
reduces the stability of the free ligand. Simultaneously, the
inductive effects of the heteroatoms are less pronounced, which
enhances the donor ability of the ligand considerably. As a
consequence, carbenes with low heteroatom stabilisation increase
the electron density at the coordinated metal centre substantially,
thus evoking new reactivity patterns.9 Here we report on isoxa-
zolylidenes as a new member of the family of abnormal remote
NHCs,8 i.e., NHCs with no neutral carbene resonance structure
and with no heteroatom adjacent to the carbene carbon.10 The
precursor isoxazolium salts are readily accessible via versatile
synthetic protocols including [2 + 3] cycloadditions,11 and they
constitute—according to the evaluations shown here and in line
with predictions12 based on computational analysis—a class of
ligands that are among the strongest neutral donors known to
date.
Scheme 1 Synthetic pathway for the preparation of palladium(II) isoxa-
zolylidene complexes 2.
A single crystal X-ray diffraction analysis of 2a confirmed the
global connectivity pattern. However, pronounced disorder in the
OTf- anion precluded the refinement to converge. Better structural
-
data were obtained upon exchanging the counteranion to BF4 .
Suitable crystals of 2b were grown by slow diffusion of hexanes
into a saturated CH2Cl2solution. The molecular structure (Fig. 1)
reveals the expected square planar geometry of the complex, with
the two phosphines situated in mutual trans position. Interestingly,
the geometry around nitrogen is indicative for sp2 hybridisation,
as no pyramidalisation was observed. This fact combined with
˚
the short C–N bond length (C3–N1 1.27(3) A) suggests that the
resonance structure A contributes more significantly to the ground
state of the remote isoxazolylidene in complex 2 than structures
B–E (Scheme 2).
Even though the free carbene route has been successfully
applied for the synthesis of specific abnormal and remote carbene
complexes,13 we have concentrated on an oxidative addition
protocol,14 which avoids the manipulation of sensitive interme-
Scheme
2
Most relevant resonance structures contributing to
3,5-dimethylisoxazol-4-ylidene.
The donor strength of the isoxazolylidene in 2 and its impact
on palladium mediated-catalysis was compared to different types
of isostructural NHC ligands. In an attempt to minimise stereo-
electronic effects, complexes [Pd(NHC)(PPh3)2I]OTf 3–7 were
synthesised (Fig. 2), all comprising NHCs with CH3 groups
ortho to the metal-bound carbon. As a consequence, the steric
School of Chemistry and Chemical Biology, University College Dublin,
Belfield, Dublin 4, Ireland. E-mail: martin.albrecht@ucd.ie; Fax: (+)353-
17162501
† Electronic supplementary information (ESI) available: Experimental
and analytical details. CCDC reference number 757620. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/c0dt00027b
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5213–5215 | 5213
©

View Article Online
Table 1 Palladium-catalyzed Suzuki–Miyaura cross-coupling
Entry
Catalyst
Substrate
T/◦C
t/h
Conv.
1a
2a
3
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Bromobenzaldehyde
4-Chloroacetophenone
4-Chloroacetophenone
4-Chloroacetophenone
4-Chloroacetophenone
4-Chloroacetophenone
4-Chloroacetophenone
4-Chlorotoluene
20
20
20
20
20
20
140
140
140
140
140
140
140
19
19
19
19
19
19
2
2
2
2
2
89%
80%
79%
84%
30%
1%
85%
90%
80%
67%
84%
79%
< 10%
2a
3a
4
4a
5
5a
6
6a
7
Fig. 1 Solid state molecular structure of 2b (30% probability, hydro-
7b
2a
3
-
gen atoms, BF4 counterion and second independent molecule omit-
8b
◦
˚
ted for clarity). Selected bond lengths (A) and angles ( ): I1–Pd1
2.6466(18), Pd1–C4 2.037(19), Pd1–P1 2.330(5), Pd1–P2 2.339(5),
O1–C5 1.34(2), O1–N1 1.40(2), N1–C3 1.27(3), C3–C4 1.41(3),
C4–C5 1.37(3); I1–Pd1–P1 89.50(13), I1–Pd1–P2 93.25(13), I1–Pd1–C4
176.1(6), P1–Pd1–P2 176.63(19), P1–Pd1–C4 88.1(5), P2–Pd1–C4
89.3(5), O1–N1–C6 113.7(19), C3–N1–C6 136(2), O1–N1–C3 109.6(16),
C3–C4–C5 101.8(18).
9b
4
10b
11b
12b
13b
5
6
7
2
2
2a
a General conditions: ArBr (1.0 mmol), PhB(OH)2 (1.2 mmol), catalyst
(1 mol%), K2CO3 (1.5 mmol), H2O (3 mL). b General conditions: ArCl
(1.0 mmol), PhB(OH)2 (1.2 mmol), catalyst (1 mol%), K2CO3 (1.5 mmol),
Bu4NCl (1.5 mmol) DMA (3 mL); all conversions determined by 1H NMR
spectroscopy, average of at least two runs.
impact about the carbene core should be essentially identical in
all complexes 2–7 and differences may therefore be attributed
predominantly to electronic modulations.
Remarkably, the 31P NMR shifts of complexes 2–7 show a clear
trend that reflects the expected ligand basicity: stronger electron
donors result in a lower frequency of dP (Fig. 2). This behavior
is independent of the solvent (DMSO-d6 or CD2Cl2) and may be
rationalised by considering the paramagnetic contribution spara to
the isotropic shielding constant. Recent theoretical calculations
predict that more basic carbene ligands induce a smaller HOMO–
LUMO gap,15 which increases the population of the paramagnetic
triplet state and hence spara. Consequently, more basic carbene
ligands effect a shift of the dP values to lower field.
more direct probe, perhaps with a somewhat better resolution,
dP analysis is generally rapid and highly convenient due to the
high natural abundance and sensitivity of the 31P nucleus. Due
to these advantages, 31P NMR probing may provide a general
method for determining the relative donor strength of a variety
of carbene subclasses, provided the metal–carbon core is sterically
comparable. In addition, this method is complementary to the
frequently used nCO method based on IR stretch vibrations of
carbonyl ligands.18 Least square regression of the obtained data
gives agoodlinear fit whichallows for anestimate of theTEPbased
on 31P NMR chemical shifts according to the equation TEP =
2125–4.01 ¥ dP.
The catalytic impact of the electronically different carbene
ligands was tested in palladium-catalysed Suzuki–Miyaura cou-
pling reactions using aryl bromides (Table 1). Under moderate
conditions, the abnormal carbene complexes (2a–5) show higher
activity than the (abnormal) pyridylidene complex 6 or the
classical carbene complex 7 (entries 1–6). This enhanced catalytic
performance may originate from the enhanced basicity of the
ligand, which is expected to facilitate the rate-limiting oxidative
addition of the aryl halide to the palladium centre.
According to this 31P basicity scale, isoxazol-4-ylidene are
stronger donors than the normal carbene, pyridin-2-ylidene, and
even abnormal carbene (cf. dP of 4–7), but not as strong as the
pyrazol-4-ylidene reported previously (cf. 3).14b This difference
parallels the weaker donor properties of 2-oxazolylidenes as
compared to normal imidazolylidenes16 and corroborates the
higher inductive effect exerted by oxygen than by nitrogen. It is
worth noting that the 31P NMR scale is consistent with recently
computed Tolman electronic parameters for this class of ligands,12
and also with a scale based on 13C NMR chemical shifts of
a trans located NHC.17 While dC measurement may provide a
Fig. 2 Measured 31P NMR chemical shifts for [Pd(NHC)(PPh3)2I][CF3SO3] complexes (CD2Cl2) and calculated TEP values (from ref. 12; slightly
different substituents in remote positions were used in calculations for the pyrazol-4-ylidene in 3 and for the imidazol-4-ylidne in 5).
5214 | Dalton Trans., 2010, 39, 5213–5215
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
1
Activated aryl chlorides were converted only with limited
success under similar conditions using 4-chloroacetophenone as
substrate, yet moderate 22–37% yields were obtained at 80 ◦C
and in the presence 1.5 eq. Bu4NBr as additive. No clear
correlation between the donor ability and the catalytic activity
was evident, though palladium black was formed in these runs and
the catalytic activity ceased after 2 h. Substantial improvements
were achieved upon changing the solvent system from H2O to
N,N-dimethylacetamide (DMA). In the presence of Bu4NCl as
additive and at elevated temperatures, activated chlorides were
arylated within three hours. In order to establish the impact of the
carbene, reactions were stopped before reaching full conversion
(Table 1, entries 7–12). Interestingly, the most basic carbene-
type ligands show the best performance (entries 7, 8), though
the correlation between ligand basicity and catalytic activity is
only modest. Non-activated aryl chlorides such as chlorotoluene
were not fully converted even after prolonged reaction times (entry
13). Further ligand optimization, especially addressing the steric
demand of NHCs for efficiently promoting oxidative addition
reactions,12,19 constitutes an obvious strategy for further enhancing
the catalytic activity of the complexes and for transforming also
more challenging substrates such as deactivated aryl chlorides.
In summary, we have developed a straightforward approach to
palladium(II) complexes comprising novel mesoionic carbene-type
ligands that are derived from isoxazolium salts. Evaluation of the
donor ability of the 4-isoxazolylidene ligand using 31P NMR as a
probe situates this NHC at the more basic edge, thus enlarging the
toolbox for the synthesis of new, highly electron-rich metal centres.
As a first application, a protocol for the arylation of aryl chlorides
has been developed, which is remarkably efficient, especially
when considering the optimisation potential, for example through
steric modification of the ligand scaffold. Besides introducing a
convenient ligand basicity scale for a variety of NHC subclasses,
these results may pave the way for the synthesis of more efficient
homogeneous catalysts.
1.90, 1.86 (2 ¥ s, 3H, C–CH3). 13C{ H} NMR (100 MHz, CD2Cl2): d 168.0
(2JPC = 3.9 Hz, C–Pd), 160.9 (3JPC = 1.9 Hz, C–Me), 134.7 (2JPC = 6.2 Hz,
Car), 131.3 (Car), 130.8 (1JPC = 25.1, Car), 128.5 (3JPC = 5.3, Car), 120.5
(1JFC = 321.1 Hz, CF3), 38.2 (N–CH3), 14.3, 13.7 (2 ¥ C–CH3). 31P NMR
(202 MHz, CD2Cl2): d 21.3 (PPh3). Anal. Calcd for C43H39 F3INO4P2PdS
(1018.12): C, 50.73; H, 3.86; N, 1.38. Found: C, 50.91; H, 4.03; N, 1.48.
Crystal data for 2b: yellow plate, C42H41BF4INO3P2Pd, M = 989.81,
˚
triclinic, a = 13.3917(9), b = 15.7897(10), c = 20.5894(15) A, a = 92.386(6),
◦
3
˚
b = 99.693(6), g = 90.623(5) , U = 4287.1(5) A , T = 173(2) K, space
¯
group P1, Z = 4, 18 577 measured reflections, 9996 unique (Rint = 0.0708),
R1 = 0.1155, wR2 = 0.3298 for I > 2s(I).
1 A. Igau, H. Grutzmacher, A. Baceiredo and G. Bertrand, J. Am. Chem.
Soc., 1988, 110, 6463.
2 A. J. Arduengo III, R. L. Harlow and M. Kline, J. Am. Chem. Soc.,
1991, 113, 3122.
3 (a) T. M. Trnka and R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18;
(b) W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41, 1290; (c) S.
Diez-Gonzalez, N. Marion and S. P. Nolan, Chem. Rev., 2009, 109,
3612.
4 D. Bourissou, O. Guerret, F. P. Gabbai and G. Bertrand, Chem. Rev.,
2000, 100, 39.
5 F. E. Hahn and M. C. Jahnke, Angew. Chem., Int. Ed., 2008, 47,
3122.
6 M. Iglesias, D. J. Beetstra, J. C. Knight, L.-L. Ooi, A. Stasch, S. Coles,
L. Male, M. B. Hursthouse, K. J. Cavell, A. Dervisi and I. A. Fallis,
Organometallics, 2008, 27, 3279.
7 J. Vignolle, X. Cattoen and D. Bourissou, Chem. Rev., 2009, 109,
3333.
8 (a) P. L. Arnold and J. Pearson, Coord. Chem. Rev., 2007, 251, 596;
(b) M. Albrecht, Chem. Commun., 2008, 3601; (c) O. Schuster, L. Yang,
H. G. Raubenheimer and M. Albrecht, Chem. Rev., 2009, 109, 3445.
9 M. Heckenroth, A. Neels, M. G. Garnier, P. Aebi, A. W. Ehlers and M.
Albrecht, Chem.–Eur. J., 2009, 15, 9375.
10 According to IUPAC nomenclature, abnormal carbenes belong to
the class of mesoionic compounds, and the complexes should hence
be termed mesoionic complexes rather than carbene complexes. The
term mesoionic is formally more appropriate than abnormal or remote
carbene, yet it omits the genealogically relationship to normal NHCs
and Fischer carbenes. Based on the data currently available, all
these classes of carbene complexes should be described as mesoionic
complexes.
11 N. R. Candeias, L. C. Branco, P. M. P. Gois, C. A. M. Afonso and A. F.
Trindade, Chem. Rev., 2009, 109, 2703.
12 D. Gusev, Organometallics, 2009, 28, 6458.
13 (a) V. Lavallo, C. A. Dyker, B. Donnadieu and G. Bertrand, Angew.
Chem., Int. Ed., 2008, 47, 5411; (b) E. Aldeco-Perez, A. J. Rosenthal, B.
Donnadieu, P. Parameswaran, G. Frenking and G. Bertrand, Science,
2009, 326, 556.
14 (a) E. Kluser, A. Neels and M. Albrecht, Chem. Commun., 2006, 4495;
(b) Y. Han, H. V. Huynh and G. K. Tan, Organometallics, 2007, 26,
6581.
The authors thank Dr A. Neels and Ms O. Sereda for X-ray
diffraction measurements. This work was financially supported
by the Swiss National Science Foundation, COST D40, and the
European Research Council through a Starting Grant. M. A.
also acknowledges the Alfred Werner Foundation for an Assistant
Professorship.
15 A. A. Tukov, A. T. Normand and M. S. Nechaev, Dalton Trans., 2009,
7015.
16 L. Perrin, E. Clot, O. Eisenstein, Jennifer Loch and R. H. Crabtree,
Inorg. Chem., 2001, 40, 5806.
17 H. V. Huynh, Y. Han, R. Josthibasu and J. A. Yang, Organometallics,
2009, 28, 5395.
18 (a) A. R. Chianese, A. Kovacevic, B. M. Zeglis, J. W. Faller and R. H.
Crabtree, Organometallics, 2004, 23, 2461; (b) R. A. Kelly III, H.
Clavier, S. Giudice, N. M. Scott, E. D. Stevens, J. Bordner, I. Samardjiev,
C. D. Hoff, L. Cavallo and S. P. Nolan, Organometallics, 2008, 27, 202;
(c) S. Wolf and H. Plenio, J. Organomet. Chem., 2009, 694, 1487.
19 (a) Z. Li, Y. Fu, Q.-X. Guo and L. Liu, Organometallics, 2008, 27, 4043;
(b) F. Barrios-Landeros, B. P. Carrow and J. F. Hartwig, J. Am. Chem.
Soc., 2009, 131, 8141.
Notes and references
‡ Typical procedure: solid Pd(PPh3)4 (144 mg, 0.125 mmol) and 4-iodo-
1,3,5-trimethylisoxazolium triflate 1a (50 mg, 0.125 mmol) were dissolved
in dry CH2Cl2 (10 mL) and stirred for 16 h at ambient temperature. After
concentrating the pale orange solution to 3 mL, the product precipitated
by addition of Et2O (10 mL). The precipitate was redissolved into CH2Cl2
(3 mL) and precipitated with Et2O (3¥), and subsequently washed with
Et2O until the solution remained colourless. The residue was dried under
vacuum, affording 2a as an off-white solid (102 mg, 80%). 1H NMR
(360 MHz, CD2Cl2): d 7.65–7.45 (m, 30H, Har), 3.60 (s, 3H, N–CH3),
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5213–5215 | 5215
©