Unexpected activation of carbon-bromide bond promoted by palladium nanoparticles in Suzuki C-C couplings

Article abstract of DOI:10.1039/c0dt00201a

Dihydroanthracene derivatives (1-6) containing imide (1-3) and amine (4-6) functions have been used for the stabilization of palladium nanoparticles, starting from Pd(0) and Pd(ii) organometallic precursors. Well-dispersed nanoparticles of mean size in the range ca. 1.9 to 3.6 nm could be obtained using Pd(0) precursors (PdLc and PdLd, where L = 1-6 and c and d mean the organometallic precursor involved, [Pd₂(dba)₃] and [Pd(ma)(nbd)] respectively). With the aim to evaluate the behaviour of homogeneous species and nanoparticles used as catalytic precursors, palladium complex coordinated to the diamine 6, [Pd(OAc)₂(κ²- N,N-6)], was prepared, reporting for the first time the X-ray diffraction structure of a metallic complex containing a ligand with a 9,10- dihydroanthracene backbone. Palladium systems were evaluated in Suzuki C-C coupling reactions and relevant differences were observed comparing the reactivity of the homogeneous systems in relation to that obtained using palladium nanoparticles as starting catalyst in relation to the activation of the C-Br bonds for deactivated substrates.

Full text of DOI:10.1039/c0dt00201a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Unexpected activation of carbon–bromide bond promoted by palladium
nanoparticles in Suzuki C–C couplings†
Delphine Sanhes,a Eva Raluy,a Ste´phane Re´tory,a Nathalie Saffon,b Emmanuelle Teumaa and
Montserrat Go´mez*a
Received 24th March 2010, Accepted 2nd August 2010
DOI: 10.1039/c0dt00201a
Dihydroanthracene derivatives (1–6) containing imide (1–3) and amine (4–6) functions have been used
for the stabilization of palladium nanoparticles, starting from Pd(0) and Pd(II) organometallic
precursors. Well-dispersed nanoparticles of mean size in the range ca. 1.9 to 3.6 nm could be obtained
using Pd(0) precursors (PdLc and PdLd, where L = 1–6 and c and d mean the organometallic precursor
involved, [Pd2(dba)3] and [Pd(ma)(nbd)] respectively). With the aim to evaluate the behaviour of
homogeneous species and nanoparticles used as catalytic precursors, palladium complex coordinated to
2
the diamine 6, [Pd(OAc)2(k -N,N-6)], was prepared, reporting for the first time the X-ray diffraction
structure of a metallic complex containing a ligand with a 9,10-dihydroanthracene backbone.
Palladium systems were evaluated in Suzuki C–C coupling reactions and relevant differences were
observed comparing the reactivity of the homogeneous systems in relation to that obtained using
palladium nanoparticles as starting catalyst in relation to the activation of the C–Br bonds for
deactivated substrates.
simultaneous p,p-interaction of the phenyl and pyridine fragments
of 4-(3-phenylpropyl)-pyridine at the ruthenium nanoparticles
surface.6
Introduction
Palladium represents one of the most used catalysts in organic
synthesis, mainly applied in versatile methods for the formation
of C–C bonds, finding numerous industrial applications in fine
chemicals production.1 In the last decades, the synthesis of metal
nanoparticles has become of great interest in the field of catalysis
due to their size and high ratio of surface area to volume.2
Under wet conditions, metallic nanoparticles become soluble
heterogeneous catalytic species, showing the likely advantages of
both homogeneous and heterogeneous classical catalysts.3 Taking
advantage of this multifaceted reactivity, we have recently proved
the dual catalytic behaviour of palladium nanoparticles acting as
reservoir of molecular species and also as heterogeneous catalyst
(single-site and surface-like reactivity) by one sequential process
involving two simple benchmark reactions, Heck C–C coupling
followed by hydrogenation.4
With these results in mind, we conceived ligands containing
a 9,10-dihydroanthracene skeleton to be applied as palladium
nanoparticles stabilizers (Scheme 1). Hydroanthracene derivatives
have been employed in biochemistry (in anticancer chemotherapy,7
as precursors of pyrrolines8 and pharmacophores such as a,b-
unsatured lactam or lactone compounds,9 etc.) and also in catalysis
(three-component Mannich reaction10 or addition of diethylzinc
to arylaldehydes11). However, to the best of our knowledge, this
kind of ligands has not been previously reported as stabilizers of
metallic nanoparticles.
Herein we report the synthesis and characterisation of new
palladium nanoparticles coming from decomposition of different
organometallic precursors in the presence of dicarboximides (1–3)
and the corresponding heterocyclic amines (4–6) linked to a 9,10-
dihydroanthracene backbone (Scheme 1). These nanocatalysts
have been tested in C–C Suzuki cross-couplings between 4-
substituted bromobenzene derivatives and phenyl boronic acid,
showing the specific activation of aryl substrates by means of their
interaction with the metallic surface, in contrast to the reactivity
observed using molecular homogeneous systems.
In particular, the ligands involved in the stabilization of palla-
dium nanoparticles play a crucial catalytic role in selective organic
transformations.5 For this reason, we are interested in the study
of the coordination chemistry at the metallic surface. Actually, we
provided evidence of a new mode of ligand coordination involving
aLaboratoire He´te´rochimie Fondamentale et Applique´e UMR CNRS 5069,
Universite´ Paul Sabatier, 118 route de Narbonne, 31062, Toulouse cedex 9,
France. E-mail: gomez@chimie.ups-tlse.fr.; Fax: +33 561558204; Tel: +33
561557738
Results and discussion
Synthesis of ligands
bUniversite´ de Toulouse, UPS, Structure Fe´de´rative Toulousaine en Chimie
Mole´culaire, FR2599, 118 route de Narbonne, 31062, Toulouse cedex 9,
France
The condensation of 9,10-dihydroanthracene-9,10-a,b-succinic
acid anhydride with the appropriate primary amine in toluene
at reflux resulted in the formation of dicarboximides 1–3 in good
yields (71–87%) (Scheme 1), based on the methodology previously
reported by us.12,13
† Electronic supplementary information (ESI) available: Calculated struc-
tures for complex C6 (Fig. S1), TEM micrographs for Pd1a, Pd2a, Pd4a,
Pd5a, Pd1b, Pd2b, Pd4b, Pd5b, Pd1d, Pd3d, Pd4d and Pd6d (Figs. S2,
S3 and S4), X-ray powder diffraction data for Pd2c, Pd6c, Pd2d and
Pd5d (Fig. S4) and IR data for PdNP and free ligands (Table S1). CCDC
reference number 770513. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0dt00201a
The heterocyclic amines 4–6 were obtained by reduction of the
carbonyl groups with LiAlH4 from the corresponding isolated
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9719–9726 | 9719
©

View Article Online
Scheme 1 Synthesis of dicarboximides 1–3 and amines 4–6.
2
Scheme 2 Synthesis of [Pd(OAc)2(k -N,N¢-6)], C6.
imides 1–3, following the procedure described for 5 (Scheme 1).13
Amines were obtained in quantitative yields.
coordination to the amine fragments. The coordination around
the metal center is roughly square planar, with bond angles ca.
90◦, except for the bite angle N1Pd1N2 (86◦). The four bond
lenghts around the metal center, Pd–O and Pd–N bonds, are quite
Synthesis of [Pd(OAc)2(j2-N,N-6)], C6
˚
similar (ca. 2.03 A). The metallacycle adopts a distorted envelope
A coordination chemistry study was carried out with the diamine
conformation, where N1, N2, C4 and Pd1 atoms are quasi-
coplanar (torsion angle C4–N2–N1–Pd1 = 12◦) and the methylene
carbon atom C3 is out of plane (torsion angle C3–C4–N1–Pd1 =
39◦). The two five-membered cycles linked by the nitrogen atom
N2 exhibit a quasi orthogonal spiro arrangement (torsion angle
Pd1–C4–C22–C5 = 80◦). The orientation of the metallacycle
in relation to the 9,10-dihydroanthracene backbone places the
Pd and acetate groups close to one of the aromatic groups;
the opposite conformation with the methylene groups pointing
to the aromatic fragment of the backbone was not observed.
Modelling study for both conformers at DFT level (B3LYP, using
6-31G* polarization basis sets and pseudopotentials) showed a
lower formation energy for the conformer observed by X-ray
diffraction analysis (energy difference: +3.5 kcal mol-1) (Fig. S1 in
Supplementary information†).
6. Reaction of Pd(OAc)2 with 1 equivalent of 6 in toluene cleanly
proceeded to give [Pd(OAc)2(k -N,N-6)] (C6) in 60% isolated yield
(Scheme 2).
2
Yellow monocrystals were obtained from slow diffusion of
dietheyl ether on a solution of the complex in dichloromethane
(Fig. 1). Even though crystal data have been reported concern-
ing imides derived from 9,10-dihydroanthracene backbone,14 C6
represents the first coordination compound characterized in the
literature by X-ray diffraction containing a ligand with this kind
of skeleton.15 The palladium atom is four-coordinated to two
oxygen atoms of the acetate groups and two nitrogen atoms
of ligand 6, giving a five-membered metallacycle by bidentated
This complex was also characterized by elemental analysis, mass
1
spectrometry, IR and NMR spectroscopies. Concerning the H
and 13C NMR spectra, only one set of signals was observed. The
protons corresponding to the two methylene groups located in the
five-membered metallacycle are down-field shifted in relation to
those corresponding to the free ligand (3.23 and 2.44 for complex
C6, and 2.93 and 1.39 ppm for free ligand 6). The two methyl
groups bonded to the nitrogen atom are observed as two singlets
1
in the H NMR spectrum of the palladium complex (1.43 and
1.81 ppm), consistent with the bidentate coordination of the
diamine 6.
Synthesis of palladium nanoparticles
Both dicarboximides (1–3) and amines (4–6) were used as sta-
bilizers of palladium nanoparticles (PdNP). Nanoparticles were
prepared from decomposition of organometallic precursors under
hydrogen atmosphere in order to reduce the metallic source
and/or hydrogenate the ligands coordinated to the metal, in
the presence of the appropriate stabilizer (1–6), based on the
Fig. 1 View of the molecular structure of complex C6. Thermal ellipsoids
are drawn at 50% probability level. Hydrogen atoms are omitted for
˚
clarity. Selected data: Pd1–N1 = 2.036(3); Pd1–N2 = 2.031(3) A; Pd1–O1 =
˚
˚
2.034(2) A; Pd1–O3 = 2.035(3) A; N2–Pd1–O1 = 91.84(11); O1–Pd1–O3 =
91.03(10); N2–Pd1–N1 = 86.19(12); N1–Pd1–O3 = 90.99(11)◦.
9720 | Dalton Trans., 2010, 39, 9719–9726
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
methodology described by Chaudret’s group (Scheme 3).16 Pal-
3
ladium(II), [PdCl2(cod)] and [Pd(h -C3H5)Cl]2, and palladium(0),
[Pd2(dba)3·CHCl3] and [Pd(ma)(nbd)], precursors were used in
the synthesis of PdNP. The metallic nanoparticles were isolated by
precipitation with pentane, giving black solids.
Scheme 3 Synthesis of PdNP stabilized by ligands 1–6 (cod = 1,5-cy-
clooctadiene; dba = dibenzylidenacetone; ma = maleic anhydride; nbd =
norbornadiene).
Starting with palladium(II) organometallic precursors.
[PdCl2(cod)] in the presence of l, 2, 4 and 5 was treated
with hydrogen in order to obtain the corresponding nanoparticles
(Pd1a, Pd2a, Pd4a and Pd5a). Unfortunately, only agglomerates
in the range of 5 to 100 nm could be observed in any case (Fig.
S2 in Supplementary Information†). Their IR spectra (Table S1
in Supplementary Information†), recorded as KBr pellets in the
range of 4000–400 cm-1, exhibited the presence of the ligands
in the composition of these materials. Comparing the main
absorption bands of these PdNP to those of the corresponding
free ligands, low shifts were observed, pointing to a weak
interaction between the metallic surface and the ligand (Table S1,
see Supplementary Information†). Similar results were obtained
3
starting from [Pd(h -C3H5)Cl]2 as organometallic precursor
Fig. 2 TEM micrographs of PdNP from [Pd2(dba)3·CHCl3] and dispersed
in THF: a) Pd1c; b) Pd2c (1.9 0.5 nm [Pd247]); c) Pd4c (3.6 1.2 nm
[Pd1678]); d) Pd5c (2.4 0.9 nm [Pd497]); e) Pd6c (2.0 0.8 nm [Pd288])
and f) size histogram for Pd6c. (In square brackets, the estimated cluster
composition considering a compact packing of spherical nanoparticles
showing a fcc arrangement).
(Pd1b, Pd2b, Pd4b and Pd5b). Spherical agglomerates in the
range of 10–200 nm were observed (Fig. S3 in Supplementary
Information†).
These results indicate that Pd(II) precursors could not lead to the
formation of well-defined nanoparticles, even for ligands 4 and 5
containing N-donor groups which should favour the coordination
to the metallic surface.
1
up of these syntheses were analysed by H NMR, showing the
Starting with palladium(0) organometallic precursors.
[Pd2(dba)3·CHCl3] in the presence of dicarboximides 1–3
(Pd1c–Pd3c) or amines 4–6 (Pd4c–Pd6c) under hydrogen
conditions, led in all cases to the formation of palladium
nanoparticles. The solutions coming from the work-up of these
syntheses were analysed by 1H NMR, showing the presence
of dibenzylidenacetone (dba) and products corresponding to
the partial reduction of dba. TEM micrographs corresponding
to Pd2c, Pd4c, Pd5c and Pd6c showed the formation of small
spherical nanoparticles with mean diameters in the range of 1.9
and 3.6 nm, depending on the ligand nature (Fig. 2).
The best dispersion of nanoparticles was observed for PdNP
stabilized by the diamine 6, Pd6c (Fig. 2e). Pd1c gave nanoparticles
not homogeneous in size and Pd3c exhibited a sponge-like super-
structure, pointing to aggregation of small individual particles.
[Pd(ma)(nbd)], containing labile and easily reducible ligands
by hydrogenation, was also used as organometallic precursor for
the synthesis of PdNP. This complex was prepared following
the methodology described in the literature.17 Pd1d–Pd6d were
isolated by precipitation, washed with diethyl ether and dried
under reduced pressure. The solutions coming from the work-
presence of maleic anhydride, succinic anhydride and norbornane.
TEM micrographs for materials prepared in the presence of
dicarboximides 1 and 3 or amines 4 and 6, showed the formation
of sponge-like superstructures constituted of aggregations of small
individual particles (Fig. S4 in Supplementary Information†).
In contrast, ligands 2 and 5 containing hydroxyl groups gave
well-dispersed spherical nanoparticles, with mean diameters of
1.41 and 1.93 nm, respectively (Fig. 3). However, size distributions
were relatively large. For these materials, Pd/ligand ratios of ca.
1/0.2 were found (from the elemental analyses) according to the
synthesis conditions, but for Pd3d, Pd4d and Pd6 a lower content
in ligand (Pd/L = 1/0.1) was determined, in agreement with the
big aggregates observed by TEM.
X-ray powder diffraction (XRD) of Pd2c, Pd6c, Pd2d and Pd5d
showed the diffraction pattern corresponding to palladium show-
ing a face-centered cubic packing (fcc) (Fig. S5 in Supplementary
Information†).
To sum up, palladium nanoparticles of mean sizes ca. 2–
4 nm could be obtained in the presence of imide and amine
derivatives starting from palladium(0) organometallic precursors.
The absence of any ligand leads to the formation of agglomerates.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9719–9726 | 9721
©

View Article Online
Table 1 Suzuki C–C cross-coupling of bromobenzene derivatives with
phenyl boronic acid using PdNP as catalytic precursors (Scheme 4)a
Entry Catalytic precursor
R
Conversion (%)b Selectivity (%)b
1
2
3
4
5
6
7
8
Pd2c
Pd2c
Pd2d
Pd2d
Pd5c
Pd5c
Pd5d
Pd5d
OCH3
CF3
OCH3 81
CF3
OCH3 50
CF3
OCH3 58
CF3 62
0
73
—
92c
95c
100
100
100
100
92c
100
43
a Reaction conditions: Pd:aryl bromide:phenyl boronic acid:sodium car-
bonate = 1 : 100 : 150 : 200 mmol in 3 cm3 of toluene and 1 cm3 of water; all
reactions in duplicate. b Arylbromide conversion and selectivity in cross-
coupling product determined by GC. c Benzene was detected as the only
by-product by GC.
Fig. 3 TEM micrographs of PdNP from [Pd(ma)(nbd)] and dispersed in
THF: a) Pd2d and b) Pd5d and their corresponding size histogram.
observed in the case of 4-bromoanisole than that related to 4-
bromobenzotrifluoride due to the less activated carbon-bromide
bond for the oxidative addition (entry 3 vs. 4, Table 1). Pd2d
was also tested as catalytic precursor for the coupling between
4-chlorobenzotrifluoride and the phenyl boronic acid, but no
activation of C–Cl bond was observed. On the other hand, Pd2c
activated 4-bromobenzotrifluoride, but it was inefficient for the
methoxy analogous substrate (entry 1 vs. 2, Table 1).
According to the mechanism generally accepted for the Suzuki
C–C cross-coupling catalyzed by molecular palladium species,
better conversions are obtained using substrates containing
electron-withdrawing groups than those containing electron-
donor groups.19,20 Thus, the results obtained using Pd2c and Pd2d
as catalytic precursors, mainly for Pd2c, indicate that the reactivity
of these nanoparticles is rather similar to that expected using
molecular complexes. Pd2c and Pd2d seem to act as a reservoir
of catalytically active molecular species released from the metallic
surfaces.21
The presence of hydroxyl (2 and 5) and amine (6) moieties in the
stabilizer favours the dispersion of nanoparticles. The differences
in size observed for nanoparticles containing the same stabilizer
(2 or 5) but prepared from different Pd(0) precursor, observing
smaller particles when [Pd(ma)(nbd)] is involved (1.9 nm for Pd2c
vs. 1.41 nm for Pd2d and 2.4 nm for Pd5c and vs. 1.95 nm for
Pd5d), could be related to a faster nucleation process starting with
[Pd(ma)(nbd)] than that with [Pd2(dba)3·CHCl3], in agreement
with the higher ability to be hydrogenated the coordinated ligands
maleic anhydride and norbornadiene than that observed for
dibenzylidenacetone under the same conditions (see above 1H
NMR analyses of work-up solutions). This fact could lead to an
important concentration of nuclei in the case of [Pd(ma)(nbd)] and
consequently favour the formation of smaller particles by the steric
hindrance induced by the stabilizers around the nanoclusters.18
Catalytic results
In contrast, the catalytic activity of Pd5c and Pd5d remained
unchanged for both aryl bromide substrates used (entry 5 vs.
6 and entry 7 vs. 8, Table 1). These results indicate a higher
activity for 4-bromoanisole than that expected on the basis of the
molecular Pd-catalyzed mechanism. This activation enhancement
for 4-bromoanisole could be due to the interaction of the methoxy
group with the metallic surface inducing a higher activation of
the C–Br bond than that expected only for intrinsical electronic
reasons of the substrate.
Palladium nanoparticles obtained from decomposition of Pd(0)
organometallic precursors were used as catalytic precursors for
Suzuki C–C cross-coupling reactions, in particular those showing
better dispersions containing ligands 2 and 5: Pd2c, Pd5c,
Pd2d and Pd5d. The catalytic reaction between bromobenzene
substrates (R = OMe, CF3) and phenyl boronic acid, was carried
out under basic toluene–H2O biphasic conditions (Scheme 4). The
catalytic results are summarized in Table 1.
The differences observed between nanoparticles containing
ligands 2 and 5 can be related to a higher stability of PdNP when
the heterocyclic amine 5 is used as stabilizer than that produced
by the corresponding imide (ligand 2), being faster the interaction
of the substrate with the surface than the leaching of molecular
species for PdNP stabilized by amine 5. Therefore, for Pd5c and
Pd5d, the methoxy substrate interacts with the metallic surface
before the oxidative addition could take place, inducing a higher
reactivity of the C–Br bond, similar to that observed for the
trifluoromethyl-substituted aryl substrate for which the interaction
with the metallic surface is not favoured.
Scheme 4 Palladium nanoparticles catalyzed C–C cross-coupling be-
tween bromobenzene derivatives and phenyl boronic acid.
In all cases, the main product corresponded to the expected
cross-coupling compound, with chemoselectivities higher than
92%. The best activities were achieved using PdNP resulting of
[Pd(ma)(nbd)] decomposition in the presence of the dicarbox-
imide 2, Pd2d (entries 3 and 4, Table 1). Weaker activity was
In order to compare the surface-like and molecular reactivity,
nanoparticles containing the diamine 6 (Pd6c) and the correspond-
ing molecular complex C6 (see above) were tested as catalytic
precursors in the Suzuki C–C cross-coupling reactions (Scheme 4),
9722 | Dalton Trans., 2010, 39, 9719–9726
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 2 Suzuki C–C cross-coupling of 4-R-bromobenzene derivatives
with phenyl boronic acid using Pd6c nanoparticles and C6 complex as
catalytic precursorsa
Conclusions
In summary, we could isolate palladium nanoparticles stabilized
by new functionalized ligands with a 9,10-dihydroanthracene
backbone (1–6) by decomposition of palladium(0) organometallic
precursors. Heterocyclic amines 5 and 6 containing hydroxy
and dimethylamino groups gave metallic nanoparticles stable
under catalytic C–C Suzuki conditions. This fact favoured the
interaction of the substrates with the metallic surface, inducing
a higher activation of the C–Br bond for the 4-bromoanisole
substrate than that produced for 4-bromobenzotrifluoride, on the
contrary to the reactivity expected for molecular homogeneous
Entry
Catalytic precursor
R
Conversion (%)b
1
2
3
4
5
6
7
8
C6
CF3
99c
99
83
64c
83
5
66
99
6
C6
OPh
SMe
OMe
CF3
OPh
SMe
OMe
SMe
SMe
C6
C6
Pd6c
Pd6cd
Pd6c
Pd6c
9
10
Pd6c+6e
Pd6c+6+KOAcf
2
systems. For this purpose, the complex [Pd(OAc)2(k -N,N¢-6)]
28
containing the diamine 6 was prepared and fully characterized.
Actually, the homogeneous system favoured the coupling when
4-bromobenzotrifluoride was used as substrate. Therefore, the
activation enhancement of the C–Br bond for 4-bromoanisole
using PdNP as catalytic precursors is probably promoted by
the interaction of the methoxy group with the metallic surface
diminishing its electron-donor character, and in consequence
favouring the oxidative addition of the C–Br bond with the
metal. Furthermore, 4-bromodiphenylether did not lead to the
formation of the cross-coupling product when PdNP were used as
catalytic precursor in contrast to the high activity observed when
the molecular complex was involved. These facts demonstrate
that the substrate interacts with the metallic surface and the
activation of C–Br bond is feasible depending on the substrate
substituent; substrate coordination and steric effects can strongly
tune the catalytic reactivity. A dramatic decrease on the activity
was observed for 4-bromothioanisole when extra ligand and/or
acetate salt were added, showing the stability of the PdNP in the
reaction medium, protected by ligands and salts which avoid the
approach of the substrate at the metallic surface. The nature of
the catalytic species after substrate activation remains uncertain
(Scheme 5).
a Reaction conditions: Pd:aryl bromide:phenyl boronic acid:sodium car-
bonate = 1 : 100 : 150 : 200 mmol in 3 cm3 of toluene and 1 cm3 of water; all
reactions in duplicate. b Arylbromide conversion and selectivity in cross-
coupling product determined by GC. c Arylbromide conversion and selec-
1
tivity in cross-coupling product determined by H RMN. d Conversion =
5% after addition of 1 equivalent of ligand; conversion = 5% after addition
of 2 equivalents of potassium acetate; conversion = 5% after addition of 1
and 2 equivalents of ligand and potassium acetate respectively. e Addition
of 1 equivalent of ligand 6. f Addition of 1 equivalent of ligand 6 and 2
equivalents of potassium acetate.
enlarging the number of substrates (4-R-bromobenzene deriva-
tives: R = OMe, CF3, OPh, SMe) in order to evidence the effect of
the activation of the C–Br bond by the metallic surface.
Both catalytic systems gave an excellent chemoselectivity,
yielding exclusively the corresponding cross-coupling product for
bromobenzene derivatives where R = CF3, OPh, SMe and OMe
(entries 1–4 for molecular catalytic precursor C6 and entries 5–
8 for palladium nanoparticles catalytic precursor Pd6c, Table 2).
For the molecular catalytic system C6, the best activities were ob-
tained using 4-bromobenzotrifluoride and 4-bromodiphenylether
(entries 1 and 2, Table 2), as expected for the substrates con-
taining electron-withdrawing and poorly electron-donor groups;
consequently, the molecular catalytic system was less active when
4-bromothioanisole and 4-bromoanisole were involved (entries
3 and 4, Table 2). However using Pd6c as catalytic precursor,
for anisole (entry 8, Table 2), thioanisole (entry 7, Table 2)
and diphenylether (entry 6, Table 2) bromobenzene substrates,
the group showing a higher electron-donor ability (methoxy
derivative) exhibited higher activity in relation to OPh and SMe
moieties; in particular 4-bromodiphenylether (entry 6, Table 2)
was practically inactive, probably due to the p,p-coordination of
both aromatic rings at the metallic surface as previously observed
using Ru nanoparticles.6 For this catalytic reaction, the addition
of extra ligand and/or the addition of acetate salt did not induce
an activity increase of the system. Furthermore, addition of free
ligand and potassium acetate in the catalytic reaction involving
SMe-containing substrate triggered a decrease on the activity
(entries 7, 9 and 10, Table 2), which means that the substrate is less
available to the metallic surface and in consequence the activity
decreases. Even under these catalytic conditions, the molecular
leaching is not favoured (entry 3 vs. 10, Table 2).
Experimental
General
All operations were carried out using standard Schlenk or Fischer–
Porter bottle techniques. The organic solvents were purified
by standard procedures and distilled under argon atmosphere:
toluene, diethyl ether and THF over sodium benzophenone
and pentane over calcium hydride. Ligands 1,12 2 and 5,13 and
[Pd(ma)(nbd)]17 were obtained following the procedure previously
described. NMR spectra were recorded on an Advance 500 Bruker
spectrometer equipped with a CryoFlowProbe (1H, standard
1
SiMe4) in CH3OD. The H NMR monitoring was performed in
the NMR tube (Wilmad NMR tube w/J. Young valve, 5 mm).
IR spectra were performed on a Perkin Elmer IR-FT 1760-X
spectrophotometer at the “Service Commun d’Infrarouge de la
Structure Fe´de´rative de Toulouse” using KBr pellets after isolation
of the particles as solids. Elemental analyses were carried out at
the “Service central d’analyse du CNRS de Vernaison”. TEM
analyses were performed on a JEOL JEM 1011 transmission
electron microscope running at 100 kV with a point resolution
These results corroborate the interaction of 4-R-bromobenzene
at the metallic surface, both improving the reactivity of the C–
Br bond for deactivated substrates such as 4-bromoanisole and
inhibiting the cross-coupling by strong coordination on the surface
such as in the case of 4-bromodiphenylether.
˚
of 4.5 A, at the “Service Commun de Microscopie Electronique
de l’Universite´ Paul Sabatier, TEMSCAN”. A drop of colloidal
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9719–9726 | 9723
©

View Article Online
Scheme 5 Activation of 4-bromoanisole promoted by the metallic surface of palladium nanoparticles.
solution was deposited onto a holey carbon covered copper grid.
The digital acquisition of the images was recorded by a camera
height of column wide angle SIS (Megaview III). The size of the
particles was determined by means of the software of treatment
of images “Image-J”. GC analyses were carried out on an Agilent
GC6890 with a flame ionization detector, using a SGE BPX5
column composed by 5% of phenylmethylsiloxane. Powder X-ray
diffraction measurements were performed at room temperature
placing the samples in capillary tubes, on a Panalytical MPDPro
diffractometer equipped with a multi-shell mirror working in
transmission mode and with a fast linear X’Ce´le´rator detector;
the analyses were carried out at the “Service de diffraction de
rayons X du Laboratoire de Chimie de Coordination” in Toulouse.
Modelling studies have been carried out using the following
software: Spartan ’06 for Windows and Linux. Wavefunction, Inc.
18401 Von Karman Avenue, suite 370. Irvine, CA 92612 USA.
at 0 ◦C and LiAlH4 (750 mg, 20.0 mmol) was slowly added,
giving a white suspension. The mixture was heated at reflux for
48 h. Diethylether (20 cm3) and a saturated aqueous solution of
Na2SO4 were then added at 0 ◦C. The addition of the aqueous
solution was slowly performed and stopped when effervescence
was no longer observed in the reaction mixture. The white product
was then filtered over Celite and washed several times with a
mixture of CH2Cl2–MeOH (9 : 1). The organic phase was washed
with water (3 ¥ 20 cm3), dried with anhydrous Na2SO4, filtered
and concentrated at reduced pressure, yielding the corresponding
amine. Compound 4 was obtained as brown oil (0.46 g, 99%).
mp 137 ◦C; (Found: C, 88.5; H, 7.1; N, 3.4. Calc. for C26H21NO2:
C, 88.9; H, 7.2; N, 3.9%); nmax(KBr pellet)/cm-1 3062 ( CH),
2927 (C–H), 1654 and 1457 (C C); dH(300 MHz; CDCl3; 298
K) 1.10 (3H, t, J 6.0, CH3), 1.72 (2H, m, CH2N), 2.58 (4H,
m, CH2N, 2CHCH2N), 2.82 (1H, m, CHN), 4.01 (1H, d, J 3.0,
CH–Ph), 4.06 (1H, d, J 3.0, CH–Ph), 7.05 (13H, m, CH CH);
dC(100.6 MHz; CDCl3; 298 K) 22.8 (CH3), 44.1 (CHCH2N), 44.2
(CHCH2N), 47.7 (CH–Ph), 47.8 (CH–Ph), 56.0 (CH2N), 56.1
(CH2N), 65.6 (CHN), 123.5–128.1 (CHaromatic), 142.0–144.2
Synthesis of ligands 9,10,11,15-tetrahydro-(11S,15S)-(N,N-
dimethylethylenediamine)-9,10-pyrollanthracen-12,14-dione, 3
A solution of 9,10-dihydroanthracene-9,10-a,b-succinic acid an-
hydride (1.00 g, 3.6 mmol) and N,N-dimethylethylene-diamine
(0.4 cm3, 3.6 mmol) in toluene (50 cm3) was refluxed for 72 h
∑
(Caromatic); m/z (CI, NH3) 352.2 (M + H+ C26H21NO2+ requires
352.5).
˚
in the presence of molecular sieves 4 A. The reaction mixture
was then cooled, filtered and the solvent removed under reduced
pressure. The residue obtained was dissolved in dichloromethane
(20 cm3) and washed with ammonium chloride saturated aqueous
solution (3 ¥ 20 cm3). The combined organic layers were dried on
anhydrous Na2SO4, filtered and the solvent was evaporated under
vacuum leading to a white powder (0.97 g, 77%). Mp 203 ◦C;
9,10,11,15-tetrahydro-(11S,15S)-(N,N-dimethyl-ethylenedia-
mine)-9,10-pyrollanthracene, 6
Compound 3 (200 mg, 0.58 mmol) was dissolved in freshly
distilled THF (25 cm3) and the resulting suspension was vigorously
stirred until total dissolution. The reaction mixture was then
cooled at 0 ◦C and LiAlH4 (330 mg, 8.7 mmol) was slowly
added, giving a white suspension. The mixture was heated at
reflux for 24 h. Diethylether (20 cm3) and a saturated aqueous
solution of Na2SO4 were then added at 0 ◦C. The addition of
the aqueous solution was slowly performed and stopped when
effervescence was no longer observed in the reaction mixture.
The white product was then filtered over Celite and washed
several times with CH2Cl2. The organic phase was washed with
a saturated aqueous solution of NaCl (3 ¥ 20 cm3), dried with
anhydrous Na2SO4, filtered and concentrated at reduced pressure,
yielding the corresponding amine. Compound 6 was obtained as
n
max(KBr pellet)/cm-1 3072, 3042, 3026 and 3012 ( CH), 2977,
2951, 2859 and 2815 (C–H), 1704 (CO), 1479 and 1465 (C C),
1148 (C–N); dH(300 MHz; CDCl3; 298 K) 1.66 (2H, t, J 6.0,
CH2N(CH3)2), 2.08 (6H, s, 2CH3), 3.13 (2H, t, J 6.0, CH2N(CO)2),
3.17 (2H, s, 2CHCON), 4.71 (2H, s, 2CH–Ph), 7.08 (4H, m,
2CH CH), 7.21 (4H, m, 2CH CH), 7.31 (4H, m, 2CH CH);
dC(75.5 MHz; CDCl3; 298 K) 35.7 (CH2N(CH3)2), 45.0 (2CH–Ph
or 2CH-CON or 2CH3), 45.6 (2CH–Ph or 2CH-CON or 2CH3),
46.9 (2CH–Ph or 2CH-CON or 2CH3), 55.2 (CH2N(CO)2), 126.9–
124.2 (CHaromatic), 138.8 (Caromatic), 141.5 (Caromatic), 176.4
∑
(CO), 176.8 (CO); m/z (EI) 346 (M+ C22H22N2O2+ requires 346.4).
◦
a white powder (0.17 g, 91%). mp 145 C; nmax(KBr pellet)/cm-1
3065, 3039 and 3017 ( CH), 2927, 2860, 2840, 2823, 2800 and
2774 (C–H), 1466 and 1456 (C C), 1140 (C–N); dH(300 MHz;
CDCl3; 298 K) 1.40 (2H, t, J 6.0, CH2N(CH3)2), 2.10 (6H, s,
2CH3), 2.23 (4H, m, 2CHCH2N), 2.62 (2H, m, 2CHCH2N), 2.93
(2H, t, J 6.0, CHNCH2), 4.05 (2H, s, 2CH–Ph), 7.01 (4H, m,
CH‘CH), 7.17 (4H, m, CH CH); dC(100.6 MHz; CDCl3; 298 K)
9,10,11,15-tetrahydro-(11S,15S)-(N-(S)-methylbenzylamine)-
9,10-pyrollanthracene, 4
Compound 1 (500 mg, 1.30 mmol) was dissolved in freshly distilled
THF (25 cm3) and the resulting suspension was vigorously stirred
until total dissolution. The reaction mixture was then cooled
9724 | Dalton Trans., 2010, 39, 9719–9726
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
44.4 (CHCH2N), 45.8 (CH3), 47.0 (CH–Ph), 54.4 (CHCH2N),
57.5 (CHNCH2, CH2N(CH3)2), 58.0 (CHCH2N), 123.7–125.9
(CHaromatic), 141.9–144.2 (Caromatic); m/z (CI, NH3) 319.4
The hydrogen was replaced by argon and the volatiles removed
under reduced pressure. The resulting black solid was further
washed with pentane (3 ¥ 5 cm3), concentrated in vacuum and
then dispersed in 2 cm3 of THF for TEM analyses. The isolated
particles were also analysed by IR spectroscopy and elemental
analysis.
∑
(M + H+ C22H26N2 requires 319.5).
+
Synthesis of [Pd(OAc)2(j2-N,N-6)] complex, C6
(Pd2c: Found: C, 40.5; H, 4.7; N, 1.6; Pd, 39.4. Calc. for
Pd5(2)(THF)4.6(H2O)3.2: C, 40.5; H, 4.9; N, 1.1; Pd, 39.5%.
idealized formula taking into account the size determined by TEM
and considering a compact packing arrangement of spherical
nanoparticles: [Pd247(2)49(THF)227(H2O)158]; Pd3c: Found: C, 30.4;
H, 3.2; N, 1.6; Pd, 45.4. Calc. for Pd5(3)0.7(THF)3.9(H2O)4: C, 34.9;
H, 4.9; N, 0.6; Pd, 45.8%.; Pd4c: Found: C, 24.1; H, 2.1; N, 1.6;
Pd, 60.7. Calc. for Pd5(4): C, 35.3; H, 2.8; N, 1.6; Pd, 60.2%;
idealized formula taking into account the size determined by TEM
and considering a compact packing arrangement of spherical
nanoparticles: [Pd1678(4)336]; Pd5c: Found: C, 26.8; H, 2.4; N, 0.9;
Pd, 57.3. Calc. for Pd5(5)0.6(THF)1.2: C, 29.4; H, 3.0; N, 1.0; Pd,
62.1%; idealized formula taking into account the size determined
by TEM and considering a compact packing arrangement of
spherical nanoparticles: [Pd497(5)60(THF)119]; Pd6c: Found: C, 29.6;
H, 3.5; N, 1.4; Pd, 45.3. Calc. for Pd5(6)0.6(THF)4: C, 34.7; H,
4.7; N, 1.7; Pd, 52.6%; idealized formula taking into account
the size determined by TEM and considering a compact packing
arrangement of spherical nanoparticles: [Pd288(6)35(THF)230])
Amine 6 (100 mg, 0.314 mmol) was dissolved in 15 cm3 of toluene
and Pd(OAc)2 (63 mg, 0.284 mmol) was then added. This mixture
was stirred for 1 h at 70 ◦C. The solvent was then evaporated to give
a green solid. Recrystallization from dichloromethane:diethylether
◦
2
(1 : 5) at 4 C afforded [Pd(OAc)2(k -N,N-6)] complex as yellow
crystals (91 mg, 0.167 mmol, 60%).(Found: C, 56.1; H, 6.2; N,
4.8. Calc. for C26H32N2O4Pd: C, 57.5; H, 5.9; N, 5.1%); nmax(KBr
pellet)/cm-1 2960–2851 (C–H), 1625–1590 (C O), 1020 (C–O);
dH(300 MHz; CDCl3; 298 K) 1.43, 1.81 (6H, s, 2CH3), 2.44, (2H,
m, CH2NCH3), 2.47 (4H, s, 2CH2N), 2.52 (6H, s, 2CH3CO), 2.67
(2H, m, 2CHCH2N), 3.23 (2H, m, CH2NCH2), 4.06 (2H, s, 2CH–
Ph), 7.20–7.00 (8H, m, CH CH). dC(100.6 MHz; CDCl3; 298 K)
22.8, 23.4 (2C, NCH3), 42.1 (2C, CHCH2N), 45.4 (2C, CHC C),
50.7 (2C, CH3–CO), 58.2 (1C, CHCH2N), 60.1 (1C, NCH2CH2),
62.7 (1C, CHCH2N), 64.4 (1C, CH2NCH3), 126.9–123.8 (8C, CH
aromatic), 143.6–139.6 (4C, Caromatic), 178.1 (2C, C O); m/z
(CI, NH3) 483 (M-OAc+. C24H29N2O2Pd requires 483.9).
General procedure for synthesis of Pd nanoparticles
From decomposition of [Pd(ma)(nbd)]. 0.17 mmol of palladium
precursor [Pd(ma)(nbd)] (50 mg) and 0.2 equivalents of ligand
(0.034 mmol; 13.6 mg for 1; 14.3 mg for 2; 11.7 mg for 3; 11.8 mg
for 4; 13.4 mg for 5; 10.7 mg for 6) were dissolved in 80 cm3 of
dried and degassed THF, and the system was then pressurised with
three bar of hydrogen in a Fischer–Porter bottle. The reagents were
dissolved in 80 cm3 of dry THF and the system was pressurised
with three bar of dihydrogen. In few minutes, the initially cream
solution became grey. The reaction mixture was then stirred at
room temperature for 18 h. The hydrogen was replaced by argon
and the volatiles removed under reduced pressure. The resulting
black solid was further washed with diethyl ether (3 ¥ 5 cm3),
concentrated in vacuum and then dispersed in 2 cm3 of THF for
TEM analyses. The isolated particles were also analysed by IR
spectroscopy and elemental analysis.
(Pd3d: Found: C, 18.7; H, 2.1; N, 0.6; Pd, 71.0. Calc. for
Pd5(3)0.2(THF)1.3(H2O)4.2: C, 15.0; H, 2.1; N, 0.7; Pd, 69.1%;
Pd4d: Found: C, 18.3; H, 1.6; N, 0.7; Pd, 65.4. Calc. for
Pd5(4)0.4(THF)0.6(H2O)5.7: C, 18.7; H, 0.7; N, 1.6; Pd, 65.0%;
Pd5d: Found: C, 36.2; H, 2.9; N, 1.2; Pd, 40.3. Calc. for
Pd5(5)(THF)2.5(H2O)9.5: C, 34.7; H, 5.2; N, 1.1; Pd, 41.6%;
idealized formula taking into account the size determined by TEM
and considering a compact packing arrangement of spherical
nanoparticles: [Pd259(5)52(THF)130(H2O)492]; Pd6d: Found: C, 12.3;
H, 1.3; N, 0.6; Pd, 77.8. Calc. for Pd5(6)0.15(THF)(H2O)2): C, 12.7;
H, 2.3; N, 0.6; Pd, 77.4%.
From decomposition of [PdCl2(cod)]. 0.35 mmol of palladium
precursor [PdCl2(cod)] (100 mg) and 0.2 equivalents of ligand
(0.07 mmol; 26.5 mg for 1; 30.0 mg for 2; 12.2 mg for 4; 13.9 mg
for 5) were dissolved in 80 cm3 of dried and degassed THF, and
the system was then pressurised with three bar of hydrogen in a
Fischer–Porter bottle. In few minutes, the initially yellow solution
became black. The reaction mixture was then stirred at room
temperature for 18 h. The hydrogen was replaced by argon and
the volatiles removed under reduced pressure. The black solid was
dispersed in 2 cm3 of THF and some drops were used for TEM
analyses.
From decomposition of [Pd(g3-C3H5)Cl]2. 0.27 mmol of palla-
3
dium precursor [Pd(h -C3H5)Cl]2 (50 mg) and 0.2 equivalents of
ligand (0.055 mmol; 20.7 mg for 1; 23.2 mg for 2; 19.2 mg for 4;
21.7 mg for 5) were dissolved in 80 cm3 of dried and degassed THF,
and the system was then pressurised with three bar of hydrogen
in a Fischer–Porter bottle. In few minutes, the initially yellow
solution became black. The reaction mixture was then stirred at
room temperature for 18 h. The hydrogen was replaced by argon
and the volatiles removed under reduced pressure. The black solid
was dispersed in 2 cm3 of THF and some drops were used for TEM
analyses.
From decomposition of [Pd2(dba)3·CHCl3]. 0.19 mmol of pal-
ladium precursor [Pd2(dba)3·CHCl3] (100 mg) and 0.2 equivalents
of ligand (0.039 mmol; 14.6 mg for 1; 16.4 mg for 2; 13.5 mg for
3; 15.3 mg for 4; 13.4 mg for 5; 13.6 mg for 6) were dissolved
in 80 cm3 of dried and degassed THF, and the system was then
pressurised with three bar of hydrogen in a Fischer–Porter bottle.
In few minutes, the initially yellow solution became black. The
reaction mixture was then stirred at room temperature for 18 h.
In few minutes, the initially red solution became black. The
reaction mixture was then stirred at room temperature for 18 h.
General procedure for catalytic Suzuki C–C coupling
The preformed nanoparticles prepared as described above
(0.01 mmol, 1.0 mol%), phenyl boronic acid (183 mg, 1.5 mmol,
1.5 eq) and Na2CO3 (212 mg, 2.0 mmol, 2 eq) dissolved in
1.0 cm3 of deoxygenated water were solubilised in 3 cm3 of dried
toluene. 1.0 mmol of substrate was then added (0.13 cm3 for
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9719–9726 | 9725
©

View Article Online
Table 3 Crystal data for C6·3H2O
Scientifique (CNRS) and the Universite´ Paul Sabatier for financial
support. D. S. is grateful to the Ministe`re de l’Enseignement
Supe´rieur et de la Recherche for a PhD research grant.
C6·3H2O
Empirical formula
Formula weight
T/K
C26 H38 N2 O7 Pd
596.98
173(2)
0.71073
Monoclinic
P21/c
11.2894(4)
11.9637(5)
20.1191(8)
90
101.045(2)
90
2667.01(18)
4
Notes and references
˚
Wavelenght/A
1 J. Tsuji, in Palladium reagents and catalysts, John Wiley & Sons, Sussex,
United Kingdom, 2004.
Crystal system
Space group
2 For selected recent reviews, see: L. Duran Pacho´n and G. Rothenberg,
Appl. Organomet. Chem., 2008, 22, 288–299; D. Astruc, Angew. Chem.
Int. Ed., 2005, 44, 7852–7872; B. Chaudret, Top. Organomet. Chem.,
2005, 16, 233–259; M. Moreno-Man˜as and R. Pleixats, Acc. Chem.
Res., 2003, 36, 638–643; A. Roucoux, J. Schultz and H. Patin, Chem.
Rev., 2002, 101, 3757–3778; H. Bo¨nnemann and R. M. Richards,
Eur. J. Inorg. Chem., 2001, 2455–2480; J. A. Widegren and R. G. Finke,
J. Mol. Catal. A: Chem., 2003, 198, 317–341.
3 J. Durand, E. Teuma and M. Go´mez, Eur. J. Inorg. Chem., 2008, 3577–
3586 references therein.
4 S. Jansat, J. Durand, I. Favier, F. Malbosc, C. Pradel, E. Teuma and M.
Go´mez, ChemCatChem, 2009, 1, 244–246.
5 M. Tamura and H. Fujihara, J. Am. Chem. Soc., 2004, 125, 15742–
15743; S. Jansat, M. Go´mez, K. Philippot, G. Muller, E. Guiu, C.
Claver, S. Castillo´n and B. Chaudret, J. Am. Chem. Soc., 2004, 126,
1592–1593; I. Favier, M. Go´mez, G. Muller, M. R. Axet, S. Castillo´n,
C. Claver, S. Jansat, B. Chaudret and K. Philippot, Adv. Synth. Catal.,
2007, 349, 2459–2469.
6 I. Favier, S. Massou, E. Teuma, K. Philippot, B. Chaudret and M.
Go´mez, Chem. Commun., 2008, 3296–3298.
7 J. A. Seetula, J. Mol. Catal. A: Chemical, 2005, 231, 153–159; A. Bisi,
S. Gobbi, A. Rampa, F. Belluti, L. Piazzi, P. Valenti, N. Gyemant and
J. Molnar, J. Med. Chem., 2006, 49, 3049–3051.
˚
a/A
˚
b/A
˚
c/A
a (◦)
b (◦)
g (◦)
3
˚
Volume/A
Z
Dc/Mg m-3
1.487
0.742
1240
Absorption coefficient/mm-1
F(000)
Cyrstal size/mm
q range for data collection/◦
Index ranges
0.10 ¥ 0.05 ¥ 0.05
5.12 to 24.71
-11< = h < = 13, -13< = k < = 13,
-18< = l < = 23
12249
Reflections collected
Independent reflections [Rint
Refinement method
]
4468 [0.0631]
Full-matrix least squares on F2
Observed reflections [I > 2s(I)]
Data/Parameters [restraints]
Goodness-of-fit on F2
4468/18 [347]
0.827
R1 = 0.0371, wR2 = 0.0547
R1 = 0.0715; wR2 = 0.0606
0.461/-0.411
Final R indices [I > 2s(I)]
R indices (all data)
8 W. K. Anderson and A. S. Milowsky, J. Org. Chem., 1985, 50, 5423–
5424.
-3
˚
Largest diff. peak/hole/e A
9 A. Sanyal, Q. Yuan and J. K. Snyder, Tetrahedron Lett., 2005, 46,
2475–2478; K. L. Burgess, N. J. Lajkiewicz, A. Sanyal, W. Yan and
J. K. Snyder, Org. Lett., 2005, 7, 31–34.
10 A. Sasaoka, Md. I. Uddin, A. Shimomoto, Y. Ichikawa, M. Shiro and
H. Kotsuki, Tetrahedron: Asymmetry, 2006, 17, 2963–2969.
11 Y. Okuyama, H. Nakano, M. Igarashi, C. Kabuto and H. Hongo,
Heterocycles, 2003, 59, 635–643.
12 D. Sanhes, I. Favier, N. Saffon, E. Teuma and M. Go´mez, Tetrahedron
Lett., 2008, 49, 6720–6723.
13 D. Sanhes, A. Gual, S. Castillo´n, C. Claver, M. Go´mez and E. Teuma,
Tetrahedron: Asymmetry, 2009, 20, 1009–1014.
4-bromoanisole, 0.14 cm3 for 4-bromobenzotrifluoride, 0.18 cm3
for 4-bromodiphenylether, 0.20 cm3 for 4-bromothioanisole). The
resulting biphasic system was heated at 65 ◦C during 6 h and
then cooled at room temperature. The organic phase (0.1 cm3) was
diluted in 2 cm3 of diethyl ether and filtered on Celite. The mixture
was consecutively washed with 2 cm3 of 1 M NaOH and water,
and the organic phase was dried on Na2SO4, filtered and analysed
by gas chromatography to determine the substrate conversion.
14 A search on Cambridge Structural Database (CSD, Version 5.31
updates February 2010) reveals 30 related structures for imide deriva-
tives containing 9,10-dihydroanthracene backbone, some of them
crystallized with different solvent molecules.
Crystal structure determination
15 We previously reported palladium and rhodium complexes containing
diphosphite ligands where the phosphorous moieties are bonded to a
9,10-dihydroanthracene backbone, but we could not crystallize them
(ref. 13).
16 K. Philippot and B. Chaudret, C. R. Chimie, 2003, 6, 1019–1034.
17 K. Itoh, F. Ueda, K. Hirai and Y. Ishii, Chem. Lett., 1977, 6, 877–
880.
18 J. Turkevitch, Gold Bull., 1985, 18, 125–131; M. A. Watzky and R. G.
Finke, J. Am. Chem. Soc., 1997, 119, 10382–10400; A. Watzky and
R. G. Finke, Chem. Mater., 1997, 9, 3083–3095; M. A. Watzky, E. E.
Finney and R. G. Finke, J. Am. Chem. Soc., 2008, 130, 11959–11969.
19 Y.-C. Lai, H.-Y. Chen, W.-C. Hung, C.-C. Lin and F.-E. Hong,
Tetrahedron, 2005, 61, 9484–9489.
20 A. Jutand and A. Mosleh, Organometallics, 1995, 14, 1810–1817.
21 C. C. Cassol, A. P. Umpierre, G. Machado, S. I. Wolke and J. Dupont,
J. Am. Chem. Soc., 2005, 127, 3298–3299; J. G. de Vries, Dalton Trans.,
2006, 421–429; F. Ferna´ndez, B. Cordero, J. Durand, G. Muller, F.
Malbosc, Y. Kihn, E. Teuma and M. Go´mez, Dalton Trans., 2007,
5572–5581.
Yellow crystals of C6·3H2O obtained from a dichloromethane/
ether dissolution of the complex were selected and mounted
on
a
Bruker-AXS APEXII diffractometer with graphite-
˚
monochromatized Mo-Ka radiation (l = 0.71073 A) at 173 K.
Crystal data are summarized in Table 3. The structure was
solved by direct methods22 and all non hydrogen atoms were
refined anisotropically using the least-squares method on F2.23
CCDC 770513 contains the supplementary crystallographic data
for this paper.† These data can be obtained free of charge
via www.ccdc.cam.ac.uk/conts/retrieving.htcm-3 (or from the
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223
336033; e-mail: deposit@ccdc.cam.ac.uk).
Acknowledgements
22 SHELXS-97, G. M. Sheldrick, Acta Crystallogr. 1990, A46, 467-473.
23 SHELXL-97, Program for Crystal Structure Refinement, G. M.
Sheldrick, University of Go¨ttingen 1997.
The authors thank C. Pradel for his helpful contribution con-
cerning TEM analyses, and the Centre National de la Recherche
9726 | Dalton Trans., 2010, 39, 9719–9726
This journal is
The Royal Society of Chemistry 2010
©