Synthesis of carbosilane dendrimers containing up to four metal layers

Article abstract of DOI:10.1002/chem.200802479

Metallodendrimers are an important class of materials with valuable properties and applications in a large number of areas. Herein, we report a highly efficient route for the synthesis of carbosilane dendrimers containing multimetal layers. The procedure

Full text of DOI:10.1002/chem.200802479

DOI: 10.1002/chem.200802479
Synthesis of Carbosilane Dendrimers Containing up to Four Metal Layers
Inma Angurell,* Oriol Rossell, and Miquel Seco*[a]
Abstract: Metallodendrimers are an
important class of materials with valua-
ble properties and applications in a
large number of areas. Herein, we
report a highly efficient route for the
synthesis of carbosilane dendrimers
containing multimetal layers. The pro-
cedure involves the use of heteroditop-
ic and tritopic P,N ligands as connec-
tors of the metal layers. The synthetic
strategy is based on the ability of the
latter ligands to react selectively with
ACHUTGTNRENNUG(cod)]ACHTUNGTRENN(UGN OTf) (cod: 1,5-cyclooctadiene;
OTf: triflate). In this way, metalloden-
drimers containing trimetallic Ru-Au-
Pd or tetrametallic Ru-Au-Au’-Pd
layers have been formed and character-
ized. The trimetallic dendrimer 9 can
be selectively deconstructed by cleav-
the metal complexes [AuCl
tetrahydrothiophene),
ACHTUNGTRENNUNG
CTHUNGTRENNUNG
cymene)}2] and [Pd(h3-2-MeC3H4)-
Keywords: carbosilanes · dendrim-
ers · layered compounds · metala-
tion · N,P ligands
À
À
age of the Ru N or Au N bonds by re-
action with chloride or iodide salts, re-
spectively.
Introduction
regularly decorated around the periphery with ferrocenyl
and (h6-aryl)tricarbonylchromium fragments.
Metallodendrimers are the subject of increasing interest due
to both the basic science involved[1] and their potential ap-
plications.[2] It is accepted that these species can combine
the catalytic,[3] electrochemical, optical and magnetic proper-
ties of organometallic or transition-metal units, with the
benefits generally attributed to dendrimers, such as site iso-
lation,[4] precise steric environments, recyclability[5] and po-
tential cooperative effects. Other promising aspects of met-
allodendrimer chemistry are found in supramolecular
chemistry,[6] nanosciences,[7] biology and medicine.[8]
We have long researched the chemistry of homo- and het-
erometallodendrimers.[12] It is surprising that, to date, no
processes have been described for the synthesis of metallo-
dendrimers containing more than two different metal cen-
ters in controlled positions. Recently, we published
a
method for the construction of carbosilane dendrimers con-
taining two metal layers: an internal one composed of ruthe-
nium moieties and another, on the periphery of the mole-
cule, composed of gold, palladium or rhodium fragments.[13]
The crucial step in our process is the selective binding of the
The vast majority of metallodendrimers are homometallic
systems for which a plethora of synthetic methods have
been developed.[9] There has been far less research into the
incorporation of two different metal centers into the same
dendritic molecule, despite the idea that the combination of
metal atoms could result in different physical properties and
reactivity of the dendritic system.[10] One of the most impor-
tant strategies was described recently by Cuadrado et al.[11]
It allows the synthesis of carbosilane dendrimers that are
4-pyridyldiphenylphosphine (4-pyPPh2) ligand to the [RuACHTUNGTRENNUNG(p-
cymene)]2+ unit through the nitrogen atom donor
(Scheme 1).
Using these findings, and in the belief that dendrimers
composed of organometallic complexes in each generation
or layer will emerge as a new type of hybrid material,[14] we
explored the possibility of increasing the number of metal
layers in the dendritic system. We also studied the incorpo-
ration of new and different metal centers into the systems
because these molecules may be used in catalysis, particular-
ly in cascade or tandem reactions.[15] Herein, we report the
efficient synthesis of a series of heterometallocarbosilane
dendrimers containing two (Ru-Au), three (Ru-Au-Pd) or
four (Ru-Au-Au’-Pd) metal layers.
[a] Dr. I. Angurell, Prof. O. Rossell, Dr. M. Seco
Departament de Quꢀmica Inorgꢁnica
Universitat de Barcelona
Martꢀ i Franquꢂs 1–11, 08028 Barcelona (Spain)
E-mail: inmaculada.angurell@qi.ub.es
miquel.seco@qi.ub.es
2932
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2932 – 2940

FULL PAPER
Scheme 1. Selective coordination of the 4-pyPPh2 ligand at the ruthenium dendrimer surface.
Results and Discussion
A
2ACHTUNGTREN(NUNG p-cymene)}2] and
A
ACHTUNGTRENNUNG
Our strategy for the construction of multimetallic dendrim-
ers was based on the growing, layer by layer, of the starting
OTf: triflate) in CH2Cl2. In all cases metalation took place
almost instantaneously and the final products 4–6 were iso-
ruthenium
dendrimer
1
(Scheme 1) by addition of ap-
propriate ligands or metalloli-
gands. Such a process required
the prior synthesis of a new
ligand, 4-amino
ACHTUNGTRENNUNG[N,N’-bis(meth-
ACHTUNGTRENNUNG
dine (3), and some related met-
allo derivatives.
Synthesis of 4-amino
G
lated in good yields and fully characterized spectroscopical-
no)]pyridine (3) and related transition-metal complexes:
This ligand was obtained as a yellow oil by treatment of a
mixture of hydroxymethyldiphenylphosphine (obtained in
situ from HPPh2 and (HCHO)n) and 4-aminopyridine
(Scheme 2). Ligand 3 was characterized spectroscopically
and by mass spectrometry (MS). The molecular ion peak
(chemical ionization (CI) MS) was detected at m/z 491.0
ly. The nACTHNGUTRENNU(G C=N) stretching of the complexes did not change
significantly from that of the free ligand, thus indicating se-
lective coordination through the phosphorus atoms. This is
confirmed by the large shift of the d31P to low fields. Both
the 1H and 13C{1H} NMR spectra of the ruthenium com-
pound 5 showed two groups of signals for the p-cymene
ligand. Their formation probably results from the high steric
congestion of the molecule, which would prevent the arms
from rotating freely. The full signal assignment for this com-
[M+H]+, whereas d31P=À27.1 ppm and the IR n
ACHTUNGTNER(NUNG C=N)
stretching frequency of 1589 cmÀ1 are the most characteristic
1
spectroscopic data for 3.
plex was obtained from a NOESY H–1H spectrum (see Ex-
perimental Section).
Another important intermediate compound required to
synthesize the desired multimetallic dendrimers is the den-
dron-like molecule 7, composed of three 3 ligands, two gold
atoms and two palladium-allyl units. It was synthesized by
mixing complexes 4 and 6 in the presence of AgOTf, as a
halide abstractor, in CH2Cl2 at low temperature. It is re-
markable that in this process no traces of oligomerization of
4 were observed. The selective attack of 6 to 4 should be un-
derstood on the basis of steric factors. Thus, the free rota-
Scheme 2. Synthesis of ligand 3.
To assess whether the ligand could coordinate to transi-
tion metals, we provoked reactions between 3 and [AuCl-
Chem. Eur. J. 2009, 15, 2932 – 2940
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2933

M. Seco, I. Angurell, and O. Rossell
dendrimers. The strategy that we first designed consisted of
the progressive growing of the ruthenium dendrimer 1 by
successive reactions with some of the units prepared above.
Our first attempts were addressed to introduce the ditopic
ligand 3 to 1; unexpectedly, the coordination of 3 was not se-
lective, in contrast to our earlier results carried out with 4-
pyPPh2,[13] and a mixture of products resulting from the co-
ordination of the ligand to 1 by nitrogen or phosphorus
atoms was obtained. We reasoned that the use of metal
complexes, such as 5, instead of free ligand 3, could over-
come this problem since in such derivatives the phosphorus
donor centers are blocked. However, treatment of 1 with
the ruthenium compound 5 did not result in the formation
of the expected dendrimer and 1 was recovered unaltered. It
is evident that steric effects play an important role here. We
therefore decided to start the process of growing multime-
tallic dendrimers with the ruthenium dendrimer 8, which
tion of the arms containing the gold metal in 4 would re-
strict the approach of the molecules to give oligomerization
products. In contrast, the rigid six-membered metallocycle
in 6 permits the pyridine group to react freely with 4, in the
absence of steric congestion.
had been prepared earlier from
(Figure 2). Thus, 8 would be the core dendrimer from and
2 and [AuClACHTUNGTRENNUNG
(tht)][13]
The variable-temperature 31P{1H} NMR spectrum of this
molecule indicates dynamic behaviour (Figure 1). The spec-
trum in [D3]CH3NO2 shows two signals at room temperature
Figure 2. Dendrimer 8 with two metal layers.
around which we would try to construct the new three- or
four-metal-layer dendrimers. We expected that the presence
of the 4-pyPPh2 ligand in 8, now simultaneously acting as a
ditopic molecule and as a spacer, would improve its capacity
to accept additional layers, because the AuCl terminal
groups are known to react with a number of ligands or met-
alloligands. The characteristic spectroscopic 31P{1H} NMR
data of 8 used to monitor subsequent reactions were
d31PRu=31.2 ppm and d31PAu =32.3 ppm.
The dendrimer 9 containing three metal layers (Ru-Au-
Pd; Figure 3) was constructed by reaction of 8a (resulting
from treatment of 8 with AgOTf) with 6. The process was
fast and dendrimer 9 was obtained as a yellow solid with a
91% yield. The incorporation of the palladium unit into 8
slightly shifted the resonances of the initial phosphorus nu-
cleus to d31PRu=28.5 ppm and d31PAu =29.5 ppm and a
new signal for the PPd nucleus emerged. This signal is shift-
Figure 1. Variable-temperature 31P{1H} NMR spectrum of compound 7 in
CD3NO2.
in the gold region and another two in the palladium region.
Upon warming, these lines begin to collapse, and above
360 K the spectrum displays only two signals (d31PAu =15.5;
d31PPd=6.4 ppm). These spectral changes are reversible on
re-cooling the sample. The reason for this behaviour proba-
bly lies in the strong steric interactions between the two
arms of the molecule, in good accord with the results de-
scribed above for complex 5.
Multimetallic dendrimers: Once the different pieces of the
puzzle were synthesized, we proceeded to build the metallo-
2934
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2932 – 2940

Carbosilane Dendrimers
FULL PAPER
behaviour observed in dendri-
mer 9. The ESI mass spectrum
of 10 showed the [MÀOTf]+
(m/z 1994.4) peak. Interestingly,
in this case the fragmentation
gives the same pattern as that
observed for the related dendri-
mer 9, that is, cleavage of the
À
molecule through the Ru N
bond ([{Ru}]+ (m/z 585) and
[{Au}+{Pd}+OTf]+
(m/z
1260)).
The dendrimer containing
four metal layers, 11, was con-
structed by a similar procedure;
we treated CH2Cl2 solutions of
8a with 7 at low temperature.
The yellow solid 11 was ob-
tained after solvent evaporation
with 88% yield (Figure 4).
An alternative procedure in-
volving two successive reactions
was also studied. In the first,
the triflate derivative 8a was
Figure 3. Dendrimer 9 with three metal layers.
reacted at À108C with a solu-
tion of the gold complex 4a
(obtained in situ from the reac-
ed from that of the free compound to lower fields, which in-
dicates coordination to the dendritic core. Furthermore, co-
ordination of this fragment to the dendrimer causes dynamic
effects in solution, and two signals in the 31P{1H} NMR spec-
trum for the PPd unit were observed, with different intensi-
ty and width depending on the temperature. The ESI mass
spectrum clearly showed fragmentation of 9 in two parts
tion of 4 with AgOTf). The second step was the addition of
a stoichiometric amount of 6 to create the Pd layer. After
appropriate workup, 11 was isolated with a yield close to
that obtained by using the first procedure. Scheme 3 summa-
rizes the different synthetic steps carried out for the con-
struction of the metallodendrimers. The 31P{1H} NMR spec-
trum of 11 in [D6]acetone clearly revealed the four different
types of phosphorus nucleus: d31PRu or Au=30.4 ppm;
d31PRu or Au=29.7 ppm; d31PAu’=17.3 ppm; and d31PPd=
7.8 ppm (Figure 5). Note that the proximity of the PRu and
À
through the Ru N bonds: the peak due to the core
[{Si5Ru4}+3OTf]+ (m/z 3164) (along with the ones resulting
from the sequential loss of OTf anions), and the intense
peak at m/z 1260, correspond-
ing to the arms [{Au}+{Pd}+
OTf]+ (see Experimental Sec-
tion for assignment notation).
Monometallic fragments at m/z
651 [{Pd}]+ and 723 [{Au}]+
were also identified. To help
detect fragments in the ESI
mass spectrum and to unambig-
uously assign the signals in the
1H NMR spectrum, we synthe-
sized the model compound 10
(see Experimental Section) by
the same procedure used for 9.
In this case, the 31P{1H} NMR
spectrum showed three sharp
signals, one for each type of
phosphorus nucleus, which con-
firmed that steric congestion
was the cause of the dynamic
Chem. Eur. J. 2009, 15, 2932 – 2940
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2935

M. Seco, I. Angurell, and O. Rossell
uted to the triflate ligands coor-
dinated to the ruthenium atoms
in the interior of 11 and to the
non-coordinate anion, respec-
tively. The IR spectrum sup-
ported the assigned structure of
11 through the presence of
peaks at 1635 and 1614 cmÀ1,
which corresponded to the vi-
brations n(C=NpyPPh ) and n(C=
2
N
ligand3), respectively. The ESI
mass spectrum of 11 showed
important fragmentation of the
dendrimer at any condition of
skimmer voltage. Nevertheless,
the structure and electronic
properties of the building
blocks of the assembled mole-
cule clearly determine the frag-
mentation patterns. Like in 9,
here we can observe a peak due
to
the
dendrimer
core
[{Si5Ru4}+3OTf]+ (m/z 3164)
and the corresponding arm
[{Au}+{Au’2}+2{Pd}ÀpyPPh2 +
3OTf]2+ (m/z 1435). Intense
peaks due to reorganization of
the fragmentation units are ob-
served at m/z 1260, [{Au}+
{Pd}+OTf]+ and m/z 555,
Figure 4. Dendrimer 11 with four metal layers.
[{Au}+{Pd}]2+
.
Given that the ESI mass
spectrum for 11 was not conclu-
sive, we became interested in
synthesizing the model com-
pound 12 bearing in mind that
it can be seen as an arm of den-
drimer 11. Therefore, we ex-
pected that the mass spectra of
both species would be similar.
Indeed, the ESI mass spectrum
for 12, along with the
[MÀ3OTf+H]2+ peak (m/z
1839), showed at low skimmer
voltages the peak correspond-
ing to the fragment [{Au}+
{Au’2}+2{Pd}ÀpyPPh2 +
3OTf]2+ (m/z 1435), which had
been observed in the ESI mass
spectrum of 11. Moreover, ad-
ditional fragments derived from
fragmentation and reorganiza-
tion processes, such as that at
m/z 1260 [{Au}+{Pd}+OTf]+,
were also present in this case, as in the spectrum of 11.
These results confirm the proposed structure for dendrimer
11.
Scheme 3. Synthetic routes to construct multilayer dendrimers.
PAu signals does not permit an unambiguous assignment. In
the 19F NMR spectrum two signals for the OTf groups were
seen at d=À78.1 (br) and À78.6 ppm (s), which were attrib-
2936
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2932 – 2940

Carbosilane Dendrimers
FULL PAPER
ple of a selective deconstruction process for organometallic
dendrimers. However, it should be noted that geometric dis-
assembly of dendritic structures is a known concept in or-
ganic chemistry, with interesting implications in areas such
as drug delivery.[16]
Conclusion
The synthesis of heterometallodendrimers containing two
(8), three (9), or four (11) metal layers of ruthenium, gold
or palladium has been accomplished by using the ditopic 4-
pyPPh2 and tritopic 3 ligands as linkers between the metal
layers. The strategy consisting of the progressive growing of
the ruthenium dendrimer 1 by successive reactions with the
previously designed metalloligands 4, 6, and 7 shows two
crucial steps: the selective binding of the bifunctional 4-
pyPPh2 ligand to the ruthenium atom through the N donor
atom and the specific reaction of 4 and 6 to give 7, without
by-products resulting from the oligomerization of the gold
complex. For the first time, the selective deconstruction of
one organometallic dendrimer, 9, has been observed by its
reaction with chloride or iodide salts. This interesting pro-
cess potentially opens up the possibility of using such kinds
of organometallic dendrimers in many areas, for example, as
prodrug systems.
Figure 5. 31P{1H} NMR spectrum of dendrimer 11 in [D6]acetone.
Decomposition of 9: selective decoordination of metal frag-
ments: The construction of metallodendrimers 8, 9, and 11
is based on acid–base interactions between the metal ions
(Ru, Au, or Pd) and the donor atoms (N or P) of the ligands
pyPPh2 and 3. We became interested in finding out the
strength of this kind of interaction as well as the possibility
of performing a selective and controlled deconstruction of
them. For this, we studied the addition of four equivalents
of (PPh4)Cl to a CH2Cl2 solution of 9. This reaction led to
the rapid replacement of the triflate ion by the chloride, as
indicated by the 31P{1H} NMR spectrum. Moreover, moni-
toring of the resulting solution revealed the gradual decoor-
dination of the metalloligand block [pyPPh2-Au-(NP2)-Pd-
ACHTUNGTRENNUNG(Me-allyl)] from the ruthenium center (Scheme 4). This be-
Experimental Section
haviour had been previously observed for dendrimer 8.[13] In
contrast, the results obtained by addition of (PPh4)I were
different. In this case, the reaction of an equimolar amount
of the iodide salt (PPh4)I to a solution of 9 promoted a se-
General data: All manipulations were performed under an atmosphere
of dry nitrogen by means of standard Schlenk techniques, and gold com-
plexes were synthesized with protection from light to avoid metal reduc-
tion. All solvents were distilled from appropriate drying agents prior to
use. 1H, 13C{1H}, 31P{1H}, and 19F NMR, NOESY 1H–1H and HSQC
1H–13C spectra were recorded on Bruker 250 and Varian Mercury 400
spectrometers. Chemical shifts are reported in ppm relative to external
À
lective attack on the Au N bond resulting in the release of
the cationic palladium complex 6. To sum up, the iodide
À
anion produces the cleavage of the Au N bond instead of
1
standards (SiMe4 for H and 13C, 85 % H3PO4 for 31P, CF3COOH for 19F)
À
the Ru N bond. The same reactivity is observed for the
and coupling constants are given in Hertz. Infrared spectra were record-
ed with an FTIR 520 Nicolet spectrophotometer (KBr pellets). Elemental
analyses of C, H and N were carried out at the Serveis Cientꢀfico-Tꢂcnics
de la Universitat de Barcelona. ESI mass spectra were recorded with an
Agilent LC/MSD time-of-flight spectrometer. The notation used in the
ESI mass spectral assignments is shown in Figure 6.
model compound 10. From these results, we conclude that
the selective deconstruction of 9 and 10 can be rationalized
in terms of the different hard/soft properties of the chloride
and iodide anions. To our knowledge, this is the first exam-
The starting materials [Pd(h3-2-MeC3H4)
[RuCl2A [Ru
(p-cymene)]2,[19] 4-pyPPh2,[20]
(pyPPh2)]
(OTf)[13] and dendrimers 1,[10c] 2,[13] and 8[13] were prepared by
A
R
ACHTUNGTRENNUNG
(tht)],[18]
C
E
G
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
following published procedures. Other reagents were purchased from
commercial suppliers.
Figure 6. ESIMS fragment notation: {Si5Ru4}=Dend-[Ru
cymene)]4; {Au}=pyPPh2 Au ; {Au’2}=py-N(CH2-PPh2 Au)2; {Pd}=py-N-
(CH2-PPh2)2(Pd(Me-allyl)); {Ru}=Ru(OTf)(p-cymene)(PMePh2).
ACHTUNGTRENNUNG(OTf)ACHTUNGTRENNUNG(p-
Scheme 4. Metallo derivatives obtained from the reaction of dendrimer 9
and (PPh4)X (X=Cl, I).
G
R
G
N
ACHTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 2932 – 2940
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2937

M. Seco, I. Angurell, and O. Rossell
Synthesis of 3: A mixture of p-formaldehyde (0.779 g, 24.7 mmol) and di-
phenylphosphine (4.45 mL, 25.9 mmol) was heated at 1208C for 5 h to
yield Ph2PCH2OH. The yellow oil obtained was left to reach room tem-
perature and 4-aminopyridine (0.394 g, 4.10 mmol) was added. The reac-
tion mixture was heated to 1208C for 6 days and the yellow oil obtained
was dissolved in THF and dried with anhydrous Na2SO4. Then, the sol-
vent was evaporated to dryness and the oil obtained was purified by
chromatography (SiO2 column) with ethyl acetate as eluent. Ligand 3
was obtained as a white solid. Yield: 64%; 31P{1H} NMR (101.25 MHz,
CDCl3, 298 K): d=À27.1 ppm (s, PPh2); 1H NMR (250.13 MHz, CDCl3,
298 K): d=8.12 (m, 2H; Ha), 7.30–7.35 (m, 20H; C6H5), 6.47 (m, 2H;
m/z: 801.1 [M+H]+, 651.1 [MÀOTf]+; elemental analysis calcd (%) for
C36H35F3N2O3P2PdS: C 53.97, H 4.40, N 3.50; found: C 53.76, H 4.18, N
3.65.
Synthesis of 7: AgOTf (0.035 g, 0.136 mmol) was added to a mixture of 4
(0.050 g, 0,052 mmol) and 6 (0.084 g, 0.104 mmol) in CH2Cl2 (10 mL) at
À108C and the mixture was stirred at this temperature for 2 h with pro-
tection from light. The solution was isolated by filtration and the solvent
was evaporated to dryness. The residue was washed with diethyl ether to
obtain 7 as a white solid (0.128 g, 88% yield). 31P{1H} NMR (162 MHz,
CD2Cl2, 298 K): d=19.0 (sbr) and 18.2 (s) (PAu), 8.9 (s), 8.6 (s) and
7.9 ppm (s) (PPd); 31P{1H} NMR (162 MHz, [D6]acetone, 298 K): d=18.0
(s) (PAu), 8.4 (sbr) and 7.6 ppm (sbr) (PPd); 31P{1H} NMR (101 MHz,
[D3]CH3NO2, 298 K): d=14.9 (s) and 14.1 (s) (PAu), 5.9 (s) and 5.0 ppm
(s) (PPd); 1H NMR (400 MHz, [D6]acetone, 298 K): d=8.28 (sbr, 2H;
Ha), 8.02–7.62 (m, 64H; C6H5, Ha, Hb), 6.51 (sbr, 4H; Hb), 5.33 (m, 8H;
CH2PPd), 5.09–4.59 (mbr, 4H; CH2PAu), 4.24 (s, 4H; CH2syn), 3.41 (t, 3J-
Hb), 3.85 ppm (d, 2J
(62.90 MHz, CDCl3, 298 K): d=152.2 (d, 3J
AHCTUNGTRENNUNG
ACHTUNGTRENNUG
Ca), 128.7–136.4 (m; C6H5), 107.9 (s; Cb), 52.1 ppm (m; CH2P); IR (KBr):
n˜ =1589 cmÀ1 (vs) (C=N); MS (CI): m/z: 491.0 [M+H]+; elemental analy-
sis calcd (%) for C31H28N2P2: C 75.91, H 5.75, N 5.71; found: C 76.03, H
5.68, N 5.53.
AHCTUNGTRENNUNG
Synthesis of 4: Ligand 3 (0.047 g, 0.10 mmol) was dissolved in CH2Cl2
(10 mL) and [AuCl
G
ACHTUNGTRENNUNG
was stirred at room temperature for 1 h with protection from light. The
solvent was evaporated to dryness and the solid residue was washed with
Et2O (3ꢄ5 mL) and dried under reduced pressure. Compound 4 was ob-
tained as a white solid with 98% yield. 31P{1H} NMR (101.25 MHz,
CDCl3, 298 K): d=19.0 ppm (s; PPh2); 1H NMR (250.13 MHz, CDCl3,
298 K): d=7.92 (sbr, 2H; Ha), 7.30–7.89 (m, 20H; C6H5), 6.23 (sbr, 2H;
Hb), 4.87 ppm (sbr, 4H; CH2P); 13C{1H} NMR (100.61 MHz, CDCl3,
N
ACHTUNGTRENNUNG
(CH3); IR (KBr): n˜ =1614 cmÀ1 (s) (C=N); MS (ESI): m/z: 2699.4
elemental
analysis
calcd
(%)
for
C
AHCTUNGTRENNUNG
298 K): d=153.4 (sbr; Cg), 149.1 (s; Ca), 134.0 (d, 2J
A
1
C6H5), 133.1 (s; p-C6H5), 131.3 (d, J
U
3J
ACHTUNGTRENNUNG(C,P)=10.1 Hz; m-C6H5), 109.4 (s; Cb), 51.0 ppm (sbr;
(KBr): n˜ =1588 cmÀ1 (vs) (C=N); MS (ESI): m/z: 957.2 [M+H]+, 919.3
[MÀCl]+, 723.3 [MÀAuCl+H]+, 439.4 [MÀCl2]2+; elemental analysis
calcd (%) for C31H28Au2Cl2N2P2: C 38.97, H 2.95, N 2.93; found: C 38.67,
H 2.99, N 2.75.
Synthesis of 5: Complex [RuCl2ACTHNUTRGNE(UNG p-cymene)]2 (0.182 g, 0.30 mmol) was
added to a solution of 3 (0.151 g, 0.31 mmol) in CH2Cl2 (15 mL) and the
mixture was stirred at room temperature for 1 h. Then, the solvent was
evaporated to reduced volume (5 mL) and Et2O was added to obtain an
orange solid. This solid was isolated by filtration, washed with Et2O (3ꢄ
10 mL) and dried under reduced pressure. Compound 5 was obtained as
an orange solid with 82% yield. 31P{1H} NMR (101.25 MHz, CDCl3,
298 K): d=22.9 ppm (s; PPh2); 1H NMR (250.13 MHz, CDCl3, 298 K):
d=6.93–7.99 (m, 22H; C6H5 +Ha), 5.98 (d, 3J
5.26 (d, 3J(H,H)=5.8 Hz, 2H; (CH)cymene(1)), 5.13 (d, 3J
(H,H)=6.0 Hz, 2H; (CH)cymene(1)), 4.92 (d,
(H,H)=6.0 Hz, 2H; (CH)cymene(2)), 4.36–4.71 (m, 4H; CH2P), 2.48 (sep,
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
2H; (CH)cymene(2)), 5.06 (d, 3J
G
G
ACHTUNGERTN(UNGN H,H)=6.7 Hz, 12H; CHACHNUTRTGENNGNU ACHTNUGERTN(NUGN H,H)=6.7 Hz,
(CH3)2), 0.84 (d, 3J
3J
3J
ACHTUNGTRENNUNG
3
A
U
ACHTUNGTRENNUNG
6.9 Hz, 6H; CH
N
E
N
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
(C,P)ꢀ15 Hz; CH2allyl), 51.1
1
CCH3), 88.7, 88.6 (s; (CH)cymene(2)), 86.0, 85.5 (s; (CH)cymene(1)), 48.6 (d, J-
ACHTUNGTRENNUNG(C,P)=16.8 Hz; CH2P), 30.3 (s; CHAHCTNURTEGN(GNUN CH3)2), 22.3, 22.0 (s; CHCAHTNUGTRENUN(NG CH3)2),
17.3 ppm (s; CCH3); IR (KBr): n˜ =1589 cmÀ1 (s) (C=N); MS (ESI): m/z:
1104.9 [M+H]+, 1068.9 [MÀCl]+; elemental analysis calcd (%) for
C51H56Cl4N2P2Ru2: C 55.54, H 5.12, N 2.54; found: C 55.29, H 5.34, N
2.48.
N
2
Synthesis of 6: Complex [Pd(h3-2-MeC3H4)
ACTHUNRTGENN(GU cod)]ACHUTNGTREN(NUGN OTf) (0.052 g,
0.12 mmol) was added to a solution of 3 (0.064 g, 0.13 mmol) in CH2Cl2
(15 mL) and the mixture was stirred for 1 h. Then, the solvent was evapo-
rated to dryness and the residue washed with Et2O (3ꢄ5 mL) and dried
under vacuum. Compound 6 was obtained as a light pink solid (96%
yield). 31P{1H} NMR (101.25 MHz, CDCl3, 298 K): d=5.0 ppm (s; PPh2);
1H NMR (250.13 MHz, CDCl3, 298 K): d=7.38–7.82 (m, 22H; C6H5 +
Synthesis of model 10: Complex [Ru
(pyPPh2)](OTf) (0.031 g, 0.031 mmol) was dissolved in CH2Cl2 (5 mL)
and [AuCl(tht)] (0.010 g, 0.031 mmol) was added. The solution was
ACHTUNGTREN(GNUN OTf)ACHTUNTREGNN(GUN p-cymene)ACHTUNGTRENNUNG(PMePh2)-
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
stirred at room temperature for 5 min and the solvent was evaporated to
dryness. Then, the residue was washed twice with hexane/diethyl ether
(1:1), and the solid obtained was dried under vacuum. The solid was dis-
solved in CH2Cl2 (5 mL) and AgOTf (0.012 g, 0.047 mmol) was added.
The solution was stirred for 30 min, filtered and placed in a bath at
À108C. Then, 6 (0.021 g, 0.031 mmol) was added to the solution and the
mixture was stirred for 30 min. The solvent was evaporated to dryness
and the residue washed with diethyl ether. The yellow solid was dried
under vacuum (0.059 g, 89% yield). 31P{1H} NMR (101 MHz, CD2Cl2,
Ha), 5.98 (d, 3J
2H; CH2syn), 3.07 (t, 3J
CH3); 13C{1H} NMR (100.61 MHz, CDCl3, 298 K): d=153.5 (s; Cg), 149.8
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(s; Ca), 129.6–133.8 (m; C6H5), 108.5 (s; Cb), 73.7 (s; CH2allyl), 50.9 (sbr
;
CH2P), 24.2 ppm (s; CH3); 19F NMR (376.48 MHz, CDCl3, 298 K): d=
À78.8 ppm (s; CF3SO3); IR (KBr): n˜ =1586 cmÀ1 (s) (C=N); MS (ESI):
2938
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2932 – 2940

Carbosilane Dendrimers
FULL PAPER
298 K): d=28.8 (s; PAu), 19.8 (s; PRu), 7.7 ppm (s; PPd); 1H NMR
(400 MHz, CD2Cl2, 298 K): d=8.60 (sbr, 2H; Ha), 7.70–7.17 (m, 44H; Ph,
(Cb), 91.9, 85.5, 83.9, 83.7 ((CH)cymene), 74.5 (sbr; CH2allyl), 51.0 (sbr; CH2P),
31.2 (CH), 23.9 (CH3allyl), 22.4, 21.2 (CH3), 18.6 (CH3), 13.1 ppm (d, 1J-
Hb, Ha), 6.28 (d, 3J
5.91 (d, 3J
G
A
A
U
(sbr), À79.1 ppm (s) (OTf); IR (KBr): n˜ =1639 (w) (C=NpyPPh ), 1616 cmÀ1
2
(CH)cymene), 5.66 (d, 3J
G
(s) (C=Nligand3); MS (ESI): m/z: 1839 [MÀ3OTf+H]2+, 1260 [{Au}+
CH2P), 4.02 (s, 2H; CH2syn), 3.05 (sbr, 2H; CH2anti), 2.35 (d, 2J
{Pd}+OTf]+, 1996 [{Ru}+{Au}+{Pd}+2OTf]+, 1435 [{Au}+{Au’2}+
10.4 Hz, 3H; CH3P), 2.19 (sep, J
2{Pd}ÀpyPPh2 +3OTf]2+
;
elemental
analysis
calcd
(%)
for
3H; CH3cymene), 1.69 (s, 3H; CH3allyl), 1.02 (d, 3J
C148H139Au3F21N7O21P8Pd2RuS7: C 43.06, H 3.39, N 2.38; found: C 43.38,
H 3.02, N 2.21.
A
R
ACHTUNGTRENNUNG
Reactivity of dendrimer 9 with halides: (PPh4)I (0.010 g, 0.022 mmol) or
(PPh4)Cl (0.008 g, 0.022 mmol) was added to a solution of dendrimer 9
(0.050 g, 0.0056 mmol) in CH2Cl2 (3 mL) at À108C, and the yellow solu-
tion was stirred for 30 min. Analysis of the solution by 31P{1H} NMR
spectroscopy showed: a) reaction with (PPh4)I: 31P{1H} NMR (101 MHz,
298 K): d=36.7 (m; PAu), 30.1 (PRu), 22.7 (PPh4+), 5.0 ppm (s; PPd);
b) reaction with (PPh4)Cl: 31P{1H} NMR (101 MHz, 298 K): d=31.3
(PAu), 24.5 (PRu), 22.7 (PPh4+), 6.7 ppm (sbr; PPd).
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
(d, 1J
ACHTUNGTRENNUNG
[Ru]+, 1260 [{Au}+{Pd}+OTf]+, 651 [{Pd}]+, 723 [{Au}]+; elemental
analysis calcd (%) for C79H76AuF12N3O12P4PdRuS4: C 44.25, H 3.57, N
1.96; found: C 44.52, H 3.51, N 2.19.
Synthesis of dendrimer 11: A solution of dendrimer 8a (0.007 mmol) in
CH2Cl2 (5 mL), obtained as described above for dendrimer 9, was cooled
to À108C and a solution of 7 (0.077 g, 0.028 mmol) in CH2Cl2 (3 mL) was
added. The mixture was stirred at this temperature for 1 h. The solvent
was evaporated to dryness and the residue was washed with diethyl
ether. Dendrimer 11 was obtained as a yellow powder (0.102 g, 88%
yield). 31P{1H} NMR (101 MHz, CD2Cl2, 298 K): d=30.8 (br; PAu), 29.8
(s; PRu), 18.3 (s; PAu’), 8.4–8.0 ppm (m; PPd); 31P{1H} NMR (101 MHz,
[D6]acetone, 298 K): d=30.4 (s; PRu or PAu), 29.7 (s; PRu or PAu), 17.3
Acknowledgement
We acknowledge support from the Ministerio de Ciencia y Tecnologꢀa
(Project CTQ2006-02362/BQU).
1
(s; PAu’), 7.8 ppm (s; PPd); H NMR (400 MHz, CD2Cl2, 298 K): d=8.60
(sbr, 8H; Ha), 7.94–6.81 (m, 360H; Ph, Hb, Ha), 6.14 (d, 8H; Hb), 6.03 (d,
8H; Hb), 5.67, 5.49, 5.37 (m, 16H; (CH)cymene), 5.01–4.57 (m, 48H;
CH2P), 4.00 (sbr, 16H; CH2syn), 2.98 (m, 16H; CH2anti), 2.14–1.68 (m,
48H; CH, CH2Pdend, CH3, CH3allyl), 1.01 (sbr) and 0.83 (sbr) (24H; CH3),
0.00 to À0.74 ppm (m, 40H; CH2Si, CH3Si); 13C{1H} NMR (100 MHz,
CD2Cl2, 298 K): d=154.8 (sbr; Ca), 150.0, 149.8 (Ca), 137.0–129.0 (C6H5),
[1] See, for example: a) G. R. Newkome, F. Vçgtle, C. N. Moorefield,
Dendrimers and Dendrons, Wiley, New York, 2001; b) D. A. Toma-
lia, H. D. Durst in Top. Curr. Chem. Vol. 165 (Ed.: E. Weber),
Springer, Berlin, 1993; c) G. R. Newkome, E. He, C. N. Moorefield,
Chem. Rev. 1999, 99, 1689–1746; d) A. W. Bosman, H. M. Janssen,
E. W. Meijer, Chem. Rev. 1999, 99, 1665–1688; e) J. P. Majoral,
A. M. Caminade, Chem. Rev. 1999, 99, 845–880; f) C. H. M. Amijs,
G. P. M. van Klink, G. van Koten, Dalton Trans. 2006, 308–327;
g) P. A. Chase, R. J. M. K. Gebbink, G. van Koten, J. Organomet.
Chem. 2004, 689, 4016–4054; h) M. Marcos, R. Martꢀn-Rapffln, A.
Omenat, J. L. Serrano, Chem. Soc. Rev. 2007, 36, 1889–1901.
[2] S.-H. Hwang, C. D. Shreiner, C. N. Moorefield, G. R. Newkome,
New J. Chem. 2007, 31, 1192–1217.
120.9 (q, 1J
ACHTUNGTRENNUNG(C,F)=319 Hz; OTf), 113.3 (s; Cb), 108.8 (sbr; Cb), 91.1, 85.9
(CH)cymene), 74.3 (m; CH2allyl), 51.2 (m; CH2P), 30.7 (s; CH), 23.9 (s;
CH3allyl), 21.9 (s; CH3), 21.8 (s; CH3), 17.9 ppm (s; CH3); 19F NMR
(376.4 MHz, [D2]CH2Cl2, 298 K): d=À78.1 (sbr), À78.6 ppm (s) (OTf); IR
(KBr): n˜ =1635 (m) (C=NpyPPh ), 1614 cmÀ1 (s) (C=Nligand3); MS (ESI):
2
m/z: 3164 [{Si5Ru4}+3OTf]+, 1435 [{Au}+{Au’2}+2{Pd}ÀPyPPh2 +
3OTf]2+, 1260 [{Au}+{Pd}+OTf]+, 555 [{Au}+{Pd}]2+
.
Synthesis of model 12: Complex [Ru
(pyPPh2)](OTf) (0.022 g, 0.022 mmol) was dissolved in CH2Cl2 (3 mL)
and [AuCl(tht)] (0.007 g, 0.022 mmol) was added. The mixture was stirred
at room temperature for 5 min, protected from light, and the solvent was
evaporated to dryness. Then, the residue was washed twice with hexane/
diethyl ether (1:1) and the solid obtained was dried under vacuum. In the
next step, the product was dissolved in CH2Cl2 (5 mL) and AgOTf
(0.009 g, 0.035 mmol) was added. The solution was stirred for 30 min, fil-
ACHTUNGTREN(GNUN OTf)ACHTUNTREGNN(GUN p-cymene)ACHTNUGTREN(NUGN PMePh2)-
[3] For overviews on catalytically active metallodendritic complexes,
see: a) D. Astruc, F. Chardac, Chem. Rev. 2001, 101, 2991–3023;
b) G. E. Oosterom, J. N. H. Reek, P. C. J. Kamer, P. W. N. M. Van
Leeuwen, Angew. Chem. 2001, 113, 1878–1901; Angew. Chem. Int.
Ed. 2001, 40, 1828–1849; c) R. van Heerbeek, P. C. J. Kamer,
P. W. N. M. van Leeuwen, J. N. H. Reek, Chem. Rev. 2002, 102,
3717–3756; d) G. D. Engel, L. H. Gade, Chem. Eur. J. 2002, 8,
4319–4329; e) A. M. Caminade, P. Servin, R. Laurent, J. P. Majoral,
Chem. Soc. Rev. 2008, 37, 56–67; f) R. AndrØs, E. de Jesffls, J. C.
Flores, New J. Chem. 2007, 31, 1161–1191; g) D. MØry, D. Astruc,
Coord. Chem. Rev. 2006, 250, 1965–1979.
[4] L. J. Twyman, A. S. H. King, I. K. Martin, Chem. Soc. Rev. 2002, 31,
69–82.
[5] R. van Heerbeek, P. C. J. Kamer, P. W. N. M. van Leeuwen, J. N. H.
Reek, Chem. Rev. 2002, 102, 3717–3756.
[6] a) J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995; b) F. Zeng, S. C. Zimmerman, Chem. Rev.
1997, 97, 1681–1712.
[7] a) Special issue: Dendrimers and Nanoscience (Ed.: D. Astruc) C. R.
Chim. 2003, 6, 709; b) D. K. Smith, Chem. Commun. 2006, 34–44.
[8] a) C. C. Lee, J. A. Mackay, J. M. J. FrØchet, F. C. Szoka, Nat. Biotech-
nol. 2005, 23, 1517–1526; b) M. Gingras, J.-M. Raimundo, Y. M.
Chabre, Angew. Chem. 2007, 119, 1028–1035; Angew. Chem. Int.
Ed. 2007, 46, 1010–1017.
G
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
tered and placed in
a
bath at À108C.
A solution of 4 (0.021 g,
0,022 mmol) in CH2Cl2 (3 mL) was added to this mixture. Then, AgOTf
(0.018 g, 0.070 mmol) was added and the mixture stirred for 5 min. The
next step consisted of the addition of 6 (0.035 g, 0.044 mmol) to the solu-
tion. The mixture was stirred for 30 min, the solvent was evaporated to
dryness and the residue washed with diethyl ether and dried under
vacuum. Compound 12 was obtained as a yellow solid (0.086 g, 95%
yield). 31P{1H} NMR (101 MHz, CD2Cl2, 298 K): d=30.0 (s; PAu), 20.6
(s; PRu), 18.6 and 18.4 (PAu), 8.7–8.4 ppm (PPd); 31P{1H} NMR
(101 MHz, [D6]acetone, 298 K): d=29.7 (s; PAu), 20.0 (s; PRu), 17.4 (sbr
PAu), 7.9 and 7.8 ppm (s; PPd); H NMR (400 MHz, CD2Cl2, 298 K): d=
;
1
8.61 (sbr, 2H; Ha), 7.93 (sbr, 2H; Ha), 7.93–7.17 (m, 78H; Ph, Hb, Ha),
6.28 (d, 3J
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
E
[9] See, for example: a) J. F. Stoddart, T. Welton, Polyhedron 1999, 18,
3575–3591; b) A. M. Caminade, R. Laurent, B. Chaudret, J. P. Majo-
ral, Coord. Chem. Rev. 1998, 178–180, 793–821; c) M. Dasgupta,
M. B. Peori, A. K. Kakkar, Coord. Chem. Rev. 2002, 233–234, 223–
235; d) O. Rossell, M. Seco, I. Angurell, C. R. Chim. 2003, 6, 803–
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Chem. Eur. J. 2009, 15, 2932 – 2940
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2939

M. Seco, I. Angurell, and O. Rossell
817; e) I. Cuadrado, M. Morµn, J. Losada, C. M. Casado, C. Pascual,
B. Alonso, F. Lobete in Advances in Dendritic Macromolecules,
Vol. 3 (Ed.: G. R. Newkome), JAI Press, Greenwich, CT, 1996, 151–
195; f) I. Cuadrado, M. Morµn, C. M. Casado, B. Alonso, J. Losada,
Coord. Chem. Rev. 1999, 193–195, 395–445; g) S. ArØvalo, E. de Je-
sffls, F. J. de la Mata, J. C. Flores, R. Gómez, Organometallics 2001,
20, 2583–2592, and references therein; h) J. A. Camerano, M. A.
Casado, M. A. Ciriano, F. Lahoz, L. A. Oro, Organometallics 2005,
24, 5147–5156.
mora, Organometallics 2006, 25, 1368–1376; e) L.-I. Rodrꢀguez, O.
Rossell, M. Seco, G. Muller, Organometallics 2008, 27, 1328–1333.
[13] I. Angurell, O. Rossell, M. Seco, E. Ruiz, Organometallics 2005, 24,
6365–6373.
[14] a) A. M. McDonagh, C. E. Powell, J. P. Morrall, M. P. Cifuentes,
M. G. Humphrey, Organometallics 2003, 22, 1402–1413; b) K. Onit-
suka, M. Fujimoto, H. Kitajima, N. Ohshiro, F. Takei, S. Takahashi,
Chem. Eur. J. 2004, 10, 6433–6446.
[15] a) L. F. Tietze, Chem. Rev. 1996, 96, 115–136; b) L. F. Tietze, N.
Rackelmann, Pure Appl. Chem. 2004, 76, 1967–1983.
[16] a) F. M. H. de Groot, C. Albrecht, R. Koekkoek, P. H. Beusker,
H. W. Scheeren, Angew. Chem. 2003, 115, 4628–4632; Angew.
Chem. Int. Ed. 2003, 42, 4492–4494; b) R. J. Amir, N. Pessah, M.
Shamis, D. Shabat, Angew. Chem. 2003, 115, 4632–4637; Angew.
Chem. Int. Ed. 2003, 42, 4494–4499; c) M. L. Szalai, R. M. Kev-
witch, D. V. McGrath, J. Am. Chem. Soc. 2003, 125, 15688–15689;
d) K. Haba, M. Popkov, M. Shamis, R. A. Lerner, C. F. Barbas III,
D. Shabat, Angew. Chem. 2005, 117, 726–730; Angew. Chem. Int.
Ed. 2005, 44, 716–720.
[17] D. A. White, Inorg. Synth. 1972, 13, 55.
[18] R. Usón, A. Laguna, Organometallic Synthesis, Vol. 3, Elsevier, Am-
sterdam, 1986, p. 324.
[19] M. A. Bennett, T. N. Huang, T. W. Matheson, A. K. Smith, Inorg.
Synth. 1982, 21, 74.
[20] a) L. Hirsivaara, M. Haukka, J. Pursiainen, J. Organomet. Chem.
2001, 633, 66–68; b) G. R. Newkome, D. C. Hager, J. Org. Chem.
1978, 43, 947–949.
[10] a) A. Juris, V. Balzani, S. Campagna, G. Denti, S. Serroni, G. Frei,
H. U. Guedel, Inorg. Chem. 1994, 33, 1491–1496; b) G. X. Liu, R. J.
Puddephatt, Organometallics 1996, 15, 5257–5259; c) S. Achar, C. E.
Immoos, M. G. Hill, V. J. Catalano, Inorg. Chem. 1997, 36, 2314–
2320; d) C. M. Casado, B. Gonzµlez, I. Cuadrado, B. Alonso, M.
Morµn, J. Losada, Angew. Chem. 2000, 112, 2219–2222; Angew.
Chem. Int. Ed. 2000, 39, 2135–2138; e) I. Angurell, G. Muller, M.
Rocamora, O. Rossell, M. Seco, Dalton Trans. 2003, 1194–1200; f) I.
Angurell, G. Muller, M. Rocamora, O. Rossell, M. Seco, Dalton
Trans. 2004, 2450–2457; g) D. MØry, L. Plault, C. Ornelas, J. Ruiz, S.
Nlate, D. Astruc, J. C. Blais, J. Rodrigues, S. Cordier, K. Kirakci, C.
Perrin, Inorg. Chem. 2006, 45, 1156–1167.
[11] M. Zamora, B. Alonso, C. Pastor, I. Cuadrado, Organometallics
2007, 26, 5153–5164.
[12] a) M. Benito, O. Rossell, M. Seco, G. SegalØs, Organometallics 1999,
18, 5191–5193; b) M. Benito, O. Rossell, M. Seco, G. Muller, J. I.
Ordinas, M. Font-Bardia, X. Solans, Eur. J. Inorg. Chem. 2002,
2477–2487; c) I. Angurell, J. C. Lima, L.-I. Rodrꢀguez, L. Rodrꢀguez,
O. Rossell, M. Seco, New J. Chem. 2006, 30, 1004–1008; d) L.-I. Ro-
drꢀguez, O. Rossell, M. Seco, A. Grabulosa, G. Muller, M. Roca-
Received: November 26, 2008
Published online: February 2, 2009
2940
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2932 – 2940