Synthesis, characterization, and reactivity of N-heterocyclic carbene palladium(II) hydroxide dimers

Article abstract of DOI:10.1021/om200579h

Palladium hydroxide dimers were prepared by reactions of CsOH with monomeric [Pd(NHC)(η³-allyl)Cl] starting materials. The dimers represent species proposed as intermediates in important cross-coupling reactions. These complexes were tested for their catalytic activity in Suzuki-Miyaura coupling and Buchwald-Hartwig aryl amination. The hydroxide complexes were susceptible to hydrogenolysis on exposure to H₂ gas. Following H₂O elimination and β-hydride elimination of an additional cinnamyl ligand, a Pd(0) species is formed which can react with other available ligands.

Full text of DOI:10.1021/om200579h

COMMUNICATION
pubs.acs.org/Organometallics
Synthesis, Characterization, and Reactivity of N-Heterocyclic Carbene
Palladium(II) Hydroxide Dimers
Jonathan D. Egbert, Anthony Chartoire, Alexandra M. Z. Slawin, and Steven P. Nolan*
EaStCHEM School of Chemistry, University of St. Andrews, St. Andrews, United Kingdom KY16 9ST
S Supporting Information
b
ABSTRACT: Palladium hydroxide dimers were prepared by
3
reactions of CsOH with monomeric [Pd(NHC)(η -allyl)Cl]
starting materials. The dimers represent species proposed as
intermediates in important cross-coupling reactions. These
complexes were tested for their catalytic activity in Suzukiꢀ
Miyaura coupling and BuchwaldꢀHartwig aryl amination. The
hydroxide complexes were susceptible to hydrogenolysis on
exposure to H gas. Following H O elimination and β-hydride
2
2
elimination of an additional cinnamyl ligand, a Pd(0) species is
formed which can react with other available ligands.
1
ransition-metal hydroxides are receiving increasing attention
bound ligands to take on a η bonding mode is consistent with
1
T
due to their importance as intermediates in catalytic cycles.
Of interest to us is the recent attention devoted to isolation of
organometallic PdꢀOH complexes in order to study their
literature precedent and facilitates the desired square-planar
geometry. The palladium atoms and those atoms attached
9
directly to the metal all exist in a single plane. The additional
atoms of the cinnamyl ligands exist on the same side of this plane.
The imidazolium moieties of the NHCs are both twisted 63° in
opposite directions of the metalꢀligand plane.
2
ꢀ4
structures and reactivity.
Additionally, the recent emergence
of N-heterocyclic carbenes (NHCs) as ligands has led to the
invention of many catalysts with superior activity and stability
compared to those of their phosphine analogues, for which they
The resonance resulting from the cinnamyl proton nearest the
1
5
phenyl (6.12 ppm) in the H NMR spectrum of 1 is close to the
were once thought to be simple mimics. Our targeted system
resonance of the free alkene (6.37 ppm), as opposed to the
resonance of the chloro starting complex (5.07 ppm). Addition-
ally, in 1 the two cinnamyl protons farthest from the phenyl give a
single resonance at 1.75 ppm. This is in contrast to the case for
the chloro starting complex, in which these two hydrogens are
nonequivalent and give two separate resonances at 1.80 and 3.02
ppm. These cinnamyl resonances are consistent with both the
solution and solid-state structures having the same configuration.
Overall the two complexes appear to be nearly identical by
comparison of the atoms bound to Pd: bond lengths of the two
complexes PdꢀO (1, 2.10 Å; 2, 2.07 Å) and PdꢀC(cinnamyl)
combines these two burgeoning areas of research. This contribu-
tion reports on the reactivity of newly prepared NHCs contain-
ing PdꢀOH dimers.
Palladium(II) halides, including the PdꢀCl monomers from
which our PdꢀOH dimers are derived, often serve as precatalysts
6
for cross-coupling reactions that require a Pd(0) active species.
This often requires initiation by either alkoxide or hydroxide
6
,7
bases. One proposed mechanism for this activation involves an
ꢀ
ꢀ
inner-sphere attack by RO with elimination of Cl (Scheme 1,
left).
In order to isolate a well-defined congener of compound A
(
1 and 2, 2.05 Å). The NHC ligands within the two complexes
(
(
(
Scheme 1), excess CsOH was reacted with [(L)PdCl
6
1
also have very similar steric requirements, from calculation of the
cinnamyl)], yielding the new complexes [(L)Pd(η -cinnamyl)
1
0
0
percent buried volumes of complexes 1 (39.4%) and 2 (41.8%).
μ-OH)] (L = IPr = N,N -bis(2,6-(diisopropyl)phenyl)imidazol-
2
0
For complex 2, the crystal structure was found to contain
bridging Cl atoms in place of OH in approximately 10% of the
complex. This impurity formed during the reaction with CsOH,
and attempts to separate the very similar complexes have been
unsuccessful.
Both 1 and 2 were tested as catalysts for BuchwaldꢀHartwig
aryl aminations and SuzukiꢀMiyaura type couplings. We
had previously reported the use of [Pd(NHC)(cinnamyl)Cl]
2
-ylidene (1), SIPr = N,N -(bis(2,6-diisopropylphenyl)imidaz-
olidine (2)) (Scheme 2).
1
The H NMR spectra of 1 and 2 contain upfield resonances at
ꢀ
4.25 and ꢀ4.29 ppm, respectively, corresponding to the proton
of the bridging hydroxide moieties. Crystals of either 1 or 2
suitable for X-ray diffraction can be grown from concentrated
pentane solution at ambient temperature. Structural data confirm
8
that both 1 and 2 are dimers in the solid state (Figure 1).
1
3
The cinnamyl ligands have also transitioned to η from the η
bonding mode adopted in the chloro precursors. The tendency
of allyl ligands bound to group 10 metals containing three other
Received: July 1, 2011
Published: August 03, 2011
r 2011 American Chemical Society
4494
dx.doi.org/10.1021/om200579h
|Organometallics 2011, 30, 4494–4496

Organometallics
NHC = IPr, SIPr) as precatalysts for both reactions. In both
COMMUNICATION
6
(
performs most efficiently. Both 1 and 2, however, lead to signi-
ficant amounts of dehalogenated aryl chloride forming as a side
product, a side reaction that was avoided in the chloro systems by
0
reactions the active [Pd (NHC)] catalyst must first be formed by
t
reaction of the chloro/allyl precatalyst with KO Bu. With that in
6
mind, we were interested in learning what effect having a basic
using cinnamyl instead of the allyl fragment.
ꢀ
ꢀ
HO in place of the Cl in the precatalyst would have on catalytic
activity.
Additionally, a potentially valuable yet rarely documented
catalytic step is the release of H O from MꢀOH in the presence
2
1
b,11
Compounds 1 and 2 were tested for their ability to carry out
aryl amination reactions (eq 1). Using 0.25 mol% of 1 as catalyst
loading, the reaction between p-chloroanisole and morpholine at
of molecular hydrogen to form a metal hydride.
Model
3
systems demonstrating this step, however, remain rare. The
PdꢀOH complexes reported here were exposed to H in order to
2
5
0 °C yielded 4-(4-methoxyphenyl)morpholine in 92% GC con-
version. The same reaction using 2 gave >99% conversion by GC
89% isolated). Additionally, the catalytic activity was tested for a
observe this reaction.
The hydroxide complex 1 is susceptible to hydrogenolysis in
(
the presence of molecular hydrogen. Reaction of 1 with H gas
2
1
SuzukiꢀMiyaura coupling (eq 2). Using 0.25 mol% of 1 at 50 °C,
the reaction between p-chloroanisole and phenylboronic acid
yielded 4-methoxybiphenyl in 78% conversion, as quantified by
GC (64% isolated yield). Additionally, 21% of the reduction
product methoxybenzene was observed. The same reaction using
(1 atm) when monitored by H NMR spectroscopy shows the
formation of H O, [Pd(IPr) ] (3), and propylbenzene, with the
2
2
last two products in a 1:2 ratio (eq 3). In this reaction palladium
black is also precipitated. Presumably the initial hydrogenolysis
forms [Pd(IPr)(cinnamyl)(H)], which may then reductively
ꢀ
ꢀ
2
yielded58% couplingproduct and 36% reduction product(GC).
eliminate to form unstable 12e [Pd(IPr)]. The 12e species would
then decompose to form Pd(0) and free IPr. The IPr ligand is
ꢀ
then available to react with subsequently formed 12e [Pd(IPr)].
This ligand scrambling is similar to the Pd hydrogenolysis reac-
3
b
tion recently reported by Fulmer et al., in which a “PCO”
pincer ligated a Pd hydroxide moiety. On exposure to H2,
In the case of aryl amination, the SIPr ligand containing 2
proved to be the more viable catalyst. This is in agreement with the
observed catalytic activity for the chloro/allyl complexes. Most
notably, for SuzukiꢀMiyaura couplings, the IPr-containing 1
Scheme 1. Plausible Mechanisms of Pd(0) Formation
1
Figure 1. Crystal structure of [Pd(IPr)(η -cinnamyl)(μ-OH)] (1)
2
with thermal ellipsoids drawn at the 50% probability level. Non-
hydroxide hydrogen atoms have been omitted, and isopropyls have
been abbreviated for clarity. Selected bond distances (Å) and angles
(
(
8
deg): PdꢀO = 2.10 (average), Pd1ꢀC(carbene) = 1.951(6), Pd1ꢀC-
cinnamyl) = 2.049(6), Pd1ꢀPd2 = 3.1720(13); O2ꢀPd1ꢀO1 =
1.80(14), C(carbene)ꢀPdꢀC(cinnamyl) = 91.2(2).
Scheme 2. Synthesis of Pd(II) Hydroxides
4
495
dx.doi.org/10.1021/om200579h |Organometallics 2011, 30, 4494–4496

Organometallics
COMMUNICATION
giving X-ray crystallographic data for 1 and 2. This material is
available free of charge via the Internet at http://pubs.acs.org.
’
AUTHOR INFORMATION
Corresponding Author
*
E-mail: snolan@st-andrews.ac.uk.
’
ACKNOWLEDGMENT
This work was supported by the ERC (Advanced Investigator
Award to SPN-FUNCAT). S.P.N. is a Royal Society Wolfson
Research Merit Award holder.
’
REFERENCES
(
1) (a) Bryndza, H. E.; Tam, W. Chem. Rev. 1988, 88, 1163.
b) Wenzel, T. T. Stud. Surf. Catal. 1991, 66, 545–554. (c) Jessop,
P. G.; Joo, F.; Tai, C.-C. Coord. Chem. Rev. 2004, 248, 2425–2442.
d) Boogaerts, I. I. F.; Nolan, S. P. J. Am. Chem. Soc. 2010,
32, 8858–8859. (e) Boogaert, I. F. F.; Fortman, G. C.; Furst,
M. R. L.; Cazin, C. S. J; Nolan, S. P. Angew. Chem., Int. Ed. 2010,
4, 8674–8677. (f) Ramon, R. S.; Gaillard, S.; Poater, A.; Cavallo, L.;
Slawin, A. M. Z.; Nolan, S. P. Chem. Eur. J. 2011, 17, 1238–1246.
g) Boogaerts, I. I. F.; Nolan, S. P. Chem. Commun 2011, 47, 3021–3024.
2) Roesky, H. W.; Singh, S.; Yusuff, K. K. M.; Maguire, J. A.;
(
2 2
Figure 2. Moles of H present in reactor calculated from H pressure:
4
ꢀ
(
a) reactor loaded with 84.7 mg of 1 (0.67 ꢁ 10 mol); (b) reactor
ꢀ4
(
1
loaded with 83.8 mg of 1 (0.66 ꢁ 10 mol).
[
Pd(PCO)(OH)] eliminates H O with a final product having two
2
4
of the PCO ligands only bound through their P atoms in a new
Pd(0) complex. The isopropylbenzene presumably forms via hydro-
genation of 1-benzyl-1-propene by 3, which is similar to the case for
the known hydrogenation catalyst [Pd(SIPr)(PCy )], which
interacts with H . Alternatively, the 12e [Pd(IPr)] intermediate
(
(
12
3
Hosmane, N. S. Chem. Rev. 2006, 106, 3813–3843.
13
ꢀ
2
(3) (a) Fulmer, G. R.; Muller, R. P.; Kemp, R. A.; Goldberg, K. I.
J. Am. Chem. Soc. 2009, 131, 1346–1347. (b) Gerber, R.; French, C. M.
Chem. Eur. J. 2010, 16, 6771–6775. (c) Fulmer, G. R.; Kaminsky, W.;
Kemp, R. A.; Goldberg, K. I. Organometallics 2011, 30, 1627–1635.
may have acted as the active catalyst in the hydrogenation of
1-benzyl-1-propene.
(
4) Cross, W. B.; Daly, C. G.; Acherman, R. L.; George, I. R.; Singh,
K. Dalton Trans. 2011, 40, 495–505.
5) (a) Nolan, S. P. J., Ed. N-Heterocyclic Carbenes in Synthesis; Wiley:
(
Additional confirmation of the reaction stoichiometry was
garnered from in situ measurement of the hydrogen pressure
change. The change in hydrogen pressure over the course of the
New York, 2006. (b) Glorius, F., Ed. N-Heterocyclic Carbenes in Transi-
tion Metal Catalysis; Springer: New York, 2007. (c) Cazin, C. S. J., Ed. N-
Heterocyclic Carbenes in Transition Metal Catalysis and Organocatalysis;
Springer: New York, 2011. (d) Díez-Gonz ꢀa lez, S., Ed. N-Heterocyclic
Carbenes; RSC Publishing: Cambridge, U.K., 2011. (e) Díez-Gonz ꢀa lez,
S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009, 109, 3612–3676.
reaction showed that 3 equiv of H gas were consumed for every
2
1
mol of 1 loaded into the reactor (Figure 2).
Reaction of 1 with PCy in THF at reflux yields the previously
3
(6) (a) Marion, N.; Navarro, O.; Mei, J.; Stevens, E. D.; Scott, N. M.;
described Pd(0) complex [Pd(IPr)(PCy )] (4). This, however,
3
Nolan, S. P. J. Am. Chem. Soc. 2006, 128, 4101–4111. (b) Navarro, O.;
Marion, N.; Mei, J.; Nolan, S. P. Chem. Eur. J. 2006, 12, 5142–5142.
(c) Marion, N.; Nolan, S. P. Acc. Chem. Res. 2008, 41, 1440–1449.
(7) Suzuki, A.; Miyaura, N. Chem. Rev. 1995, 95, 2457–2483.
(8) CCDC-831869 (1) and CCDC-831870 (2) contain supplemen-
tary crystallographic data for this contribution. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
is not an efficient reaction, due to thermal decomposition of 1.
This also is in contrast with previous reports on similar phosphine
complexes of the form [Pd(PR )(OH)L] , in which phosphine
3
2
addition resulted inphosphine ligation toformanequilibrium with
14
monomeric species, of the form [Pd(PR ) (OH)L]. Hydroge-
3
2
nolysis was also performed in the presence of added PCy to give
3
complex 4 in quantitative yields in reactions monitored by NMR
spectroscopy. Attempts to scale this reaction for an improved
synthesis of 4, however, were unsuccessful, and on a larger scale
reactions resulted in mixtures of 3 and 4.
In summary, the use of PdꢀOH complexes as precatalysts for
BuchwaldꢀHartwig aryl amination and SuzukiꢀMiyaura cou-
plings confirms the possibility of PdꢀOR species being inter-
mediates in the formation of catalytically active Pd(0) from
Pd(II)ꢀCl complexes. Additionally, the newly reported PdꢀOH
complexes are susceptible to hydrogenolysis, modeling this
potentially valuable catalytic step. Studies aimed at exploring
further the role of MꢀOH species in catalysis are ongoing.
(
9) Barloy, L.; Ramdeehul, S.; Osborn, J. A.; Carlotti, C.; Taulelle, F.;
De Cian, A.; Fischer, J. Eur. J. Inorg. Chem. 2000, 2523–2532.
10) Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.;
Scarano, V.; Cavallo, L. Eur. J. Inorg. Chem. 2009, 1759–1766.
11) (a) Goeden, G. V.; Caulton, K. G. J. Am. Chem. Soc. 1981,
(
(
103, 7353–7355. (b) Thompson, J. S.; Bernard, K. A.; Rappoli, B. J.;
Atwood, J. D. Organometallics 1990, 9, 2727–2731. (c) Thompson, J. S.;
Randall, S. L.; Atwood, J. D. Organomatallics 1991, 10, 3906–3910.
(
d) Bohler, C.; Avarvari, N.; Schonberg, H.; Worle, M.; Grutzmacher, H.
Helv. Chim. Acta 2001, 84, 3127–3147.
12) Jurcik, V.; Nolan, S. P.; Cazin, C. S. J. Chem. Eur. J. 2009,
5, 2509–2511.
13) Fantasia, S.; Egbert, J. D.; Jurcik, V.; Cazin, C. S. J.; Jacobsen,
H.; Cavallo, L.; Heinekey, D. M.; Nolan, S. P. Angew. Chem., Int. Ed.
(
1
(
2
009, 48, 5182–5184.
’
ASSOCIATED CONTENT
(14) Grushin, V. V.; Alper, H. Organometallics 1996, 15, 5242–5245.
S
Supporting Information. Text and figures giving experi-
b
mental details and characterization data for 1 and 2 and CIF files
4
496
dx.doi.org/10.1021/om200579h |Organometallics 2011, 30, 4494–4496