Rhodium or ruthenium units peripherally coordinated to carbosilane dendrimers functionalized with P-stereogenic monophosphines

Article abstract of DOI:10.1016/j.jorganchem.2006.10.035

Carbosilane dendrimers containing P-stereogenic monophosphines as terminal groups, Dend-{CH₂PPhR}_n (R = 2-biphenylyl or 9-phenanthryl), were reacted with [RhCl(COD)]₂ or [RuCl₂(p-cymene)]₂ to afford the corresponding chiral metalladendrimers Dend-{CH₂PPhR(RhCl(COD))}_n or Dend-{CH₂PPhR(RuCl₂(p-cymene))}_n, respectively. Attempts to obtain the first generation Ru-dendrimer for R = 2-biphenylyl proved unsuccessful, probably due to the steric hindrance of R. Complete characterization of these species was achieved by multinuclear NMR spectra, including 2D experiments, mass spectrometry, and optical rotation determinations. The catalytic properties of the rhodium dendrimers were tested in the hydrogenation of dimethylitaconate and those of the ruthenium derivatives in the asymmetric hydrogen transfer of acetophenone. The following model chiral compounds, (CH₃)₃Si{CH₂PPhR(RhCl(COD))} and (CH₃)₃Si{CH₂PPhR(RuCl₂(p-cymene))}, were prepared in order to detect potential dendritic effects. All compounds were active in the catalytic conditions tested, but low or null e.e. were found.

Full text of DOI:10.1016/j.jorganchem.2006.10.035

Journal of Organometallic Chemistry 692 (2007) 851–858
www.elsevier.com/locate/jorganchem
Rhodium or ruthenium units peripherally coordinated to carbosilane
dendrimers functionalized with P-stereogenic monophosphines
*
´
Lara-Isabel Rodrıguez, Oriol Rossell , Miquel Seco, Guillermo Muller
` `
´
Departament de Quı´mica Inorganica, Universitat de Barcelona, Martı i Franques, 1-11, 08028 Barcelona, Spain
Received 19 September 2006; received in revised form 16 October 2006; accepted 17 October 2006
Available online 24 October 2006
Abstract
Carbosilane dendrimers containing P-stereogenic monophosphines as terminal groups, Dend-{CH2PPhR}n (R = 2-biphenylyl or
9-phenanthryl), were reacted with [RhCl(COD)]2 or [RuCl2(p-cymene)]2 to afford the corresponding chiral metalladendrimers Dend-
{CH2PPhR(RhCl(COD))}n or Dend-{CH2PPhR(RuCl2(p-cymene))}n, respectively. Attempts to obtain the first generation Ru-dendri-
mer for R = 2-biphenylyl proved unsuccessful, probably due to the steric hindrance of R. Complete characterization of these species
was achieved by multinuclear NMR spectra, including 2D experiments, mass spectrometry, and optical rotation determinations. The
catalytic properties of the rhodium dendrimers were tested in the hydrogenation of dimethylitaconate and those of the ruthenium deriv-
atives in the asymmetric hydrogen transfer of acetophenone. The following model chiral compounds, (CH3)3Si{CH2PPhR(RhCl(COD))}
and (CH3)3Si{CH2PPhR(RuCl2(p-cymene))}, were prepared in order to detect potential dendritic effects. All compounds were active in
the catalytic conditions tested, but low or null e.e. were found.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Dendrimers; Chirality; Rhodium; Ruthenium; Catalysis
1. Introduction
generation chiral phosphorus-containing dendrimer used in
Pd-catalyzed asymmetric allylic alkylation [7]. In addition,
a dendrimer displaying a Ru-BINAP chiral core is also
known [8]. Very recently, we undertook a program consist-
ing of the preparation of the first dendritic systems periph-
erally functionalized with P-stereogenic monophosphines,
(S)-MePPh(2-biphenylyl) and (S)-MePPh(9-phenanthryl),
with the aim to assess their potential catalytic proper-
ties, after grafting into the periphery selected transition
metal fragments. It is well known that the introduction of
chirality within dendritic architectures enables these sys-
tems to be used in enantioselective catalysis. Thus, our first
report in this area described the use of allylpalladium-con-
taining carbosilane dendrimers as catalysts in the asymmet-
ric hydrovinylation of styrene [9]. The excellent results
obtained in terms of selectivity and enantiomeric excess
of 3-phenyl-1-butene prompted us to extend this study to
new chiral metalladendrimers. In the present study, we
wish to report the synthesis and characterization of two
metalladendrimer families, resulting from the grafting of
The preparation and characterization of dendrimers
containing transition metal fragments has received consid-
erable attention in recent years because they have proved to
have the potential to combine the advantage of homoge-
neous and heterogeneous catalysts in one system [1].
Although a number of dendritic catalysts have been
described, so far, relatively few reports on catalytic asym-
metric synthesis are available. This is because chiral ligands
are expensive or difficult to obtain. Some examples include
the amino alcohol-zinc alkyl compounds attached to phos-
phine-functionalized poly(propyleneimine) (PPI) [2], the
Ti-TADDOL systems [3], the chiral Salen systems [4], the
‘‘pyrphos’’-functionalized PPI dendrimers [5], the species
resulting by grafting chiral ferrocenyl diphosphines (‘‘Josi-
phos’’) into the periphery of a dendrimer [6], and the third
*
Corresponding author.
E-mail address: oriol.rossell@qi.ub.es (O. Rossell).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.10.035

852
L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
rhodium, RhCl(COD), or ruthenium, RuCl2(p-cymene)
fragments onto the surface of chiral phosphine-functional-
ized carbosilane dendrimers. The efficiency of the rhodium
complexes for the asymmetric hydrogenation of dimethyli-
taconate and the ruthenium species for the asymmetric
transfer hydrogenation of acetophenone is also described.
Mononuclear metal models were prepared in order to
detect dendritic effects.
Cl
Cl
Cl
CH2Cl2
r.t.
P*
Ru
Ru
Ru
Cl
Cl
P*
Cl
1P2 : Ru-1P2
2P2 : Ru-2P2
for P* = 1P1 : Ru-1P1
2P1 : Ru-2P1
Scheme 1.
2. Results and discussion
be different, as a consequence of the presence of the chiral
phosphine. For the same reason, the protons of the CH2P,
Si(CH3)2 and the (CH3)2CH groups became diastereotopic.
As expected, the protons of the (CH3)CH and CH3 units of
the p-cymene ligand gave only one signal, which appeared
as a septet by the coupling with the six protons of the
methyl groups, and, as a singlet, respectively. The most
interesting feature in the 13C NMR spectra is the confirma-
tion of the diastereotopic character of the carbon methyl
atoms of the isopropyl group and those of the (CH3)2Si
units. Complete assignation of 1H and 13C signals was
2.1. Synthesis and characterization of ruthenium dendrimers
The starting carbosilane dendrimers peripherally func-
tionalized with (S)-MePPh(2-biphenylyl) or (S)-MePPh(9-
phenanthryl) are represented in Chart 1.
These compounds, as well as the model compounds 1P1
and 1P2, were obtained as described earlier [9]. The ruthe-
nium dendrimers were obtained following the protocol
shown in Scheme 1. The process was monitored by 31P
NMR spectroscopy which evidenced a new signal for the
new species, showing the expected deshielding effect. Com-
plete metallation was deduced from the absence of 31P
NMR resonances due to unreacted phosphine units. In
fact, the process is identical to that described in the synthe-
sis of ruthenium dendrimers with alkyldiphenylphosphino
terminated carbosilane dendrimers [10].
obtained from NOESY H–1H and HSQC H–13C experi-
ments. For example, the HSQC 1H–13C spectra showed the
correlation between the CH3 carbons of the isopropyl
group with the corresponding protons. In the 29Si NMR
spectra, the internal silicon centres of the dendrimers were
observed, as well as the expected doublet (J(Si–
P) = 7.4 Hz) for the external silicon atoms. ES mass spectra
gave in all cases peaks at [MÀCl]+, along with other frag-
ments, confirming the nature of the compounds.
1
1
The metalladendrimers were isolated as red solids in
high yields. Attempts to obtain Ru-3P1 were unsuccessful,
given that only a partial functionalization of the dendrimer
branches was achieved, probably due to the steric hin-
drance of the biphenylyl moiety. The failure of this synthe-
sis was evidenced by its 31P NMR spectrum, which
contained very broad bands. In fact, the singlet observed
for Ru-2P1 was also relatively broad. The new species are
soluble in moderately polar organic solvents (CH2Cl2,
2.2. Synthesis and characterization of rhodium dendrimers
The rhododendrimers Rh-nP1 and Rh-nP2 were cleanly
obtained by reacting [RhCl (COD)]2 with the correspond-
ing dendrimer in CH2Cl2 at room temperature. Recrystalli-
zation in CH2Cl2/diethylether gave the targeted compounds
as yellow solids in good yields. They are soluble in
most common organic solvents and were characterized by
1
CHCl3, acetone, thf) and were characterized by H, 13C,
29Si, and 31P NMR spectroscopy as well as electrospray
1
mass spectroscopy. The H NMR spectra for both groups
1
of compounds, Ru-nP1 or Ru-nP2, were similar, showing,
in all cases, the ring protons of the p-cymene groups to
elemental analyses, H, 13C and 31P NMR spectroscopy,
and ES mass spectrometry (see Scheme 2).
R
Ph
P
Ph
P
R
Ph
R
Si
Si
P
Ph
P
Si
Si
R
Me
R
Ph
P
R
Si
P
Me
Me
Si
Si
Si
Ph
Ph
Si
P
Si
Si
P
Si
Si
Ph
R
R
Si
Si
Si
Si
Si
1P
P
Si
Si
Si
Ph
R
P
R
P
R
Ph
Ph
2P
Si
3P
Si
P = P1
P = P2
for R =
R =
R
P
Ph
P
(only for R = 2-biphenylyl)
R
Ph
Chart 1.

L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
853
the CH2P protons for Rh-1P1 and Rh-2P1 appeared as
two pseudotriplets (J(HH) = J(HP)) while for Rh-3P1 the
same protons appeared as two multiplets. ES mass spec-
trometry showed [MÀCl]+ peaks for the models Rh-1P1
and Rh-1P2 and [MÀRh(COD)Cl–Cl]+ and other fragmen-
tations for the metalladendrimers, confirming the nature of
the products.
P*
Cl
Cl
Cl
CH2Cl2
r.t.
P*
Rh
Rh
Rh
for P* = 1P1 : Rh-1P1 1P2 : Rh-1P2
2P1 : Rh-2P1 2P2 : Rh-2P2
3P1 : Rh-3P1
2.3. Catalytic reactions: hydrogenation of dimethylitaconate
Scheme 2.
One of the most known methods for the asymmetric
hydrogenation of dimethylitaconate under mild conditions
uses rhodium or ruthenium complexes generated in situ via
the addition of a metal complex precursor and a monoden-
tate or bidentate chiral phosphine ligand [11,12], Eq. (1).
In general, rhodium complexes have afforded better
results in terms of efficiencies and enantiomeric excesses
than the ruthenium compounds. This can be attributed to
the extraordinary capacity of ruthenium(II) to generate a
large number of coordination environments, with or with-
out chiral ligand, as well as polynuclear-type complexes
In all cases, the unique phosphorous signal appeared as
a doublet (J(P–Rh) was about 150 Hz). HSQC 1H–13C
experiments allowed us the full assignation of the signals
in both groups of dendrimers. The coordination of the
Rh(COD)Cl moiety to the chiral phosphine induced all
the protons and carbons of the COD to be different. Thus,
although the CH protons trans to the phosphine resonated
at the same d, the cis CH protons appeared, at higher field,
as two singlets. Analogously, the protons of the CH2 units
of the COD appeared as four superimposed multiplets that
1
could be assigned by a H–13C HSQC experiment (Fig. 1).
The different nature of the carbon atoms of the CH2 units
was also corroborated in the 13C NMR spectra. On the
other hand, both the 1H and the 13C NMR spectra revealed
the diastereoscopic character of the protons and carbon
atoms of the CH3Si and CH2P groups, due to the chiral
character of the phosphines. In particular, for example,
[Rh]
MeOOC
MeOOC
COOMe
COOMe
*
H2 (g)
Eq. 1. Hydrogenation of dimethylitaconate.
CH3Si
CH2Si
CH2P
CH2 cod
CH cod cis
cis P
P
CH cod trans
Rh
Cl
trans P
p
5.0
0.0
ppm (f2)
Fig. 1. 1H–13C HSQC spectrum of compound Rh-2P2.

854
L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
[13]. With these precedents, we tested the catalytic capabil-
ity of the rhododendrimers for the asymmetric hydrogena-
tion of dimethyl itaconate. Our interest was to establish the
relationship between the size/generation of the dendrimer
and its catalytic properties. As the standard reaction condi-
tions we chose a substrate:catalyst (catalytic site) ratio of
500:1, 10 bar hydrogen pressure, and a reaction tempera-
ture of 20 ꢁC. The conversion was monitored by GC. We
found that the precursors were active and that compounds
belonging to the P2-family were the most active, although
the enantiomeric excess proved to be zero in all cases. As
can be seen from Table 1, the activity decreases regularly
upon going from the model compound Rh-1P1 to the first
generation dendrimer, Rh-3P1. The reduction in activity
observed may be due to the reduced accessibility of the
metal centres in Rh-3P1. The effect of increasing approach
of peripheral groups in dendrimers at higher generations
has been demonstrated in a large number of cases [14]. This
effect is also observed for the P2-containing family of com-
pounds, for which the Rh-1P2 model was the most active
(Table 2).
In order to improve the enantioselectivity of the process,
we decided to apply the strategy by Reetz et al. [15,16] in
combinatorial asymmetric transition-metal-catalyzed reac-
tions, consisting in the formation of a rhodium species con-
taining simultaneously two different chiral monodentate
phosphines, La and Lb. This hetero-combination, RhLaLb,
was shown to be more active, in terms of enantioselectivity
in the Rh-catalyzed olefin hydrogenation, than its related
homo-combinations, RhLaLa or RhLbLb. Thus, trying to
take advantage of the anchored chiral phosphine that
avoids the formation of the homo-combinations, we added,
in this case, non-chiral PPh3 to our two dendritic systems,
in order to evaluate the results of the hydrogenation of the
dimethylitaconate. The most notorious effect observed was
the sharp decrease of the activity of the rhododendrimers,
as expected, but, unfortunately, the enantiomeric excess
was not improved, remaining null. At this point, it is clear
that much more work is needed to understand the catalytic
properties of the dendritic systems reported in this paper.
2.4. Catalytic transfer hydrogenation of acetophenone
The catalytic hydrogenation of organic substrates by
hydrogen transfer reactions avoids the use of gaseous
hydrogen and allows the use of standard reflux techniques.
A drawback is the often low catalytic activity and produc-
tivity. The hydrogen is usually supplied by 2-propanol or
formic acid and the process is catalyzed by transition met-
als, ruthenium being one of the most widely used so far
[17]. For this purpose, several ruthenodendrimers have
been also employed [18]. In this paper, we describe the cat-
alytic behaviour of our derivatives in the asymmetric trans-
fer hydrogenation of acetophenone (Eq. (2)), by using
6 · 10À3 M solutions of the ruthenium compounds and a
Ru/acetophenone/tBuOK ratio of 1/200/10. The reaction
solution was stirred at 82 ꢁC.
OH
*
O
[Ru]
OH
O
+
+
iPrOH, 82ºC
Table 1
Ru/S/ t BuOK 1/200/10
Hydrogenation of dimethylitaconate using catalytic precursors Rh-nP1
(n = 1,2,3)
Eq. 2. Catalytic transfer hydrogenation of acetophenone.
Run
Catalytic
precursor
t (h)
Conversion
(%)
TOF
e.e.
(%)
(hÀ1
)
The time course curves of the hydrogenation acetophe-
none for both families of compounds are represented in
Fig. 2. It can be seen that the species containing 2-biphenylyl
(P1) are more active than those having 9-phenanthryl (P2), in
clear contrast with the results observed in the hydrogenation
of dimethylitaconate. In addition, it is particularly notewor-
thy that some positive dendritic effects were observed. Thus,
the zeroth generation dendrimers were more active than the
model compounds for both series of products, although the
enantiomeric excesses were low (about 15%) and difficult to
reproduce. After 22 h of refluxing, the conversion of ace-
tophenone was not total and the catalysts appeared deacti-
vated, as can be deduced from the green color of the
solution and the precipitation of a solid.
1
2
3a
Rh-1P1
1
2
2
61.2
98.1
16.8
306
245
42
ꢀ0
ꢀ0
ꢀ0
4
5a
Rh-2P1
Rh-3P1
2
2
94.4
11.2
236
28
ꢀ0
ꢀ0
6
2
68.6
171
ꢀ0
Conditions: ratio [Rh]/substrate 1/500; T = 20 ꢁC; 10 bar H2; 20 ml thf.
a
Ratio [Rh]/substrate/PPh3 1/500/1.
Table 2
Hydrogenation of dimethylitaconate using catalytic precursors Rh-nP2
(n = 1,2)
Run
Catalytic
precursor
t (h)
Conversion (%)
TOF
(hÀ1
e.e.
(%)
)
3. Conclusion
1
2a
Rh-1P2
1
1
85.5
13.5
428
68
ꢀ0
ꢀ0
In summary, two series of novel chiral rhodium and
ruthenium dendrimers have been prepared and character-
ized and their catalytic properties in the hydrogenation of
dimethylitaconate and acetophenone, respectively, have
3
Rh-2P2
1
1
64.0
7.4
320
37
ꢀ0
4a
ꢀ0
Conditions: ratio [Rh]/substrate 1/500; T = 20 ꢁC; 10 bar H2; 20 ml thf.
a
Ratio [Rh]/substrate/PPh3 1/500/1.

L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
855
The product Rh-1P1 was obtained as a yellow solid. Yield:
90
80
70
60
50
40
30
20
10
0
0.304 g (89%). 31P{1H} NMR (101.3 MHz, CDCl3, 298 K),
d
(ppm): 24.9 (d, 1JPRh = 147.3 Hz). 1H NMR
(400.1 MHz, CDCl3, 298 K), d (ppm): 8.32–7.15 (m, Ar,
14H), 5.45 (s(br), CHtrans, 2H), 3.17 (m, CHcis, 1H), 2.84
(m, CHcis, 1H), 2.45–1.81 (m, CH2, 8H), 1.49 (pt,
J ꢁ 13 Hz, CH2P, 1H), 0.67 (pt, J ꢁ 14 Hz, CH2P, 1H),
0.02 (s, CH3Si, 9H). 13C–{1H} NMR (100.6 MHz, CDCl3,
298 K), d (ppm): 145.7–126.4 (m, Ar), 102.9 (m, CHtrans),
Ru-1P1
Ru-2P1
Ru-1P2
Ru-2P2
1
1
71.3 (d, JCRh = 14.2 Hz, CHcis), 69.9 (d, JCRh = 14.0 Hz,
CHcis), 33.4 (d, 1JCRh = 3.0 Hz, CH2), 32.7 (d,
1JCRh = 2.5 Hz, CH2), 29.2 (s(br), CH2), 28.7 (d,
1
1JCRh = 1.3 Hz, CH2), 14.0 (d, JCP = 14.6 Hz, CH2P),
3
1.65 (d, JCP = 3.0 Hz, CH3Si). EA: Calcd for C30H37-
0
2
4
t (h)
6
8
ClPRhSi: C, 60.56; H, 6.27. Found: C, 60.42; H, 6.41%.
MS (ES(+), m/z): 559.1 (559.6 calcd) [MÀCl]+, 451.1
(451.4 calcd) [MÀClÀCOD]+.
Fig. 2. Conversion of hydrogenation of acetophenone vs. time using Ru
(II) compounds as precursors.
4.2.2. Synthesis of Rh-2P1
been investigated. In both studies, the dendritic systems
were active, although their behaviour in terms of activity
and enantiomeric excesses were not satisfactory.
This complex was obtained in the same way as Rh-1P1.
Starting from 2P1 (0.160 g, 0.108 mmol) and the rhodium
dimer [RhCl(COD)]2 (0.107 g, 0.217 mmol), a pale yellow
solid was obtained. Yield: 0.256 g (96%). 31P{1H} NMR
4. Experimental
(101.3 MHz, CDCl3, 298 K),
d
(ppm): 25.8 (d,
1JPRh = 147.0 Hz). H NMR (400.1 MHz, CDCl3, 298 K),
d (ppm): 8.26–7.12 (m, Ar, 56H), 5.43 (s(br), CHtrans
1
4.1. Reagents and general techniques
,
8H), 3.18 (s(br), CHcis, 4H), 2.82 (s(br), CHcis, 4H), 2.47–
1.77 (m, CH2, 32H), 1.50 (pt, J = 13.6 Hz, CH2P, 4H),
0.77 (pt, J = 15.0 Hz, CH2P, 4H), 0.25 to (À0,19) (m,
CH2Si, 16H), 0.14 (s, CH3Si, 12H), À0.02 (s, CH3Si,
12H). 13C{1H} NMR (100.6 MHz, CDCl3, 298 K), d
(ppm): 145.8 À126.5 (m, Ar), 102.9 (m, CHtrans), 71.5 (d,
All manipulations were performed under purified nitro-
gen using standard schlenk techniques. All solvents were
distilled from appropriate drying agents. 1H, 13C{1H},
31P{1H} and 29Si{1H} were obtained on Varian Gemini
200, Bruker DRX 250 and Varian Mercury 400 spectrom-
eters. 2D NMR spectra (NOESY 1H–1H and HSQC
1H–13C) were recorded on a Varian Mercury 400 spectrom-
eter. Chemical shifts are reported in ppm relative to exter-
1
1JCRh = 13.9 Hz, CHcis), 70.1 (d, JCRh = 14.1 Hz, CHcis),
33.6 (s, CH2), 32.7 (s, CH2), 29.5 (s, CH2), 28.8 (s, CH2),
1
3
12.7 (d, JCP = 13.8 Hz, CH2P), 10.2 (d, JCP = 3.6 Hz,
CH2Si0), 3.2 (s, CH2Si1), À0.4 (s, CH3Si), À0.8 (s, CH3Si).
29Si{1H} NMR (49.7 MHz, CDCl3, 298 K), d (ppm): 8.98
(s, Si0), 3.34 (s, Si1). EA: Anal. Calc. for C124H152Cl4P4Rh4
Si5: C, 60.53; H, 6.23. Found: C, 60.68; H, 6.34%. MS
(ES(+), m/z): 2177.8 (2178.2 calcd) [MÀRhCODClÀCl]+,
2087.6 (2088.0 calcd) [MÀRhCODClÀCODÀCl+H2O]+.
1
nal standards (SiMe4 for H, 13C and 29Si and 85% H3PO4
for 31P) and coupling constants are given in Hz. MS (ES)
spectra were recorded with a Fisions VGQuattro spectrom-
eter. Enantiomeric excesses and conversions were deter-
mined by GC on a Hewlett-Packard 5890 Series II gas
chromatograph (30-m Chiraldex DM column) with a
FID detector. Elemental analysis (C,H) were performed
´
´
at the Servicio de Microanalisis del Centro de Investigacion
4.2.3. Synthesis of Rh-3P1
y Desarrollo del Consejo Superior de Investigaciones
This complex was obtained using the same procedure as
for Rh-1P1. Starting from 3P1 (0.060 g, 0.017 mmol) and
the rhodium dimer [RhCl(COD)]2 (0.033 g, 0.068 mmol),
a yellow solid was obtained. Yield: 0.085 g (91%).
31P{1H} NMR (101.3 MHz, CDCl3, 298 K), d (ppm):
´
Cientıficas (CSIC). [RuCl2(p-cymene)]2, [RhCl(COD)]2 as
well as the starting chiral carbosilane compounds were pre-
pared as previously described [9]. Other reagents were used
as received from commercial suppliers.
1
1
26.9 (d, JPRh = 146.9 Hz). H NMR (400.1 MHz, CDCl3,
298 K), d (ppm): 8.27–7.10 (m, Ar, 112H), 5.43 (s(br),
CHtrans, 16H), 3.16 (s(br), CHcis, 8H), 2.81 (s(br), CHcis,
8H), 2.40–1.76 (m, CH2, 64H), 1.45 (m, CH2P, 8H), 0.74
(m, CH2P, 8H), 0.50 to (À0,19) (m, CH2Si + CH3Si,
148H). 13C{1H} NMR (100.6 MHz, CDCl3, 298 K), d
(ppm): 145.6 À126.4 (m, Ar), 102.9 (m, CHtrans), 71.3 (d,
4.2. Synthesis and characterization data
4.2.1. Synthesis of Rh-P1
The model compound 1P1 (0.200 g, 0.574 mmol) was
dissolved in 10 ml of CH2Cl2 and the rhodium dimer
[RhCl(COD)]2 (0.141 g, 0.287 mmol) was added. After stir-
ring for 20 min, the solvent was removed and the resulting
pasty solid was recrystallized from CH2Cl2/diethyl ether.
1
1JCRh = 14.9 Hz, CHcis), 69.9 (d, JCRh = 13.6 Hz, CHcis),
33.5 (s, CH2), 32.6 (s, CH2), 29.3 (s, CH2), 28.7 (s, CH2),

856
L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
12.2 (m, CH2P), 10.9–2.5 (m, CH2Si), À0.6 (s, CH3Si3),
À0.8 (s, CH3Si3), À4.1 (s, CH3Si1), À6.3 (s, CH3Si2).
29Si{1H} NMR (49.7 MHz, CDCl3, 298 K), d (ppm): 7.90
(s, Si2), 5.71 (s, Si1), 3.32 (s, Si3) EA Anal. Calc. for
C276H372Cl8P8Rh8 Si17: C, 60.03; H, 6.79. Found: C,
60.18; H, 6.88%.
31P{1H} NMR (101.3 MHz, CDCl3, 298 K), d (ppm):
22.4 (s). H NMR (400.1 MHz, CDCl3, 298 K), d (ppm):
1
8.56–6.88 (m, Ar, 14H), 5.36–5.29 (m, H–C6H4, 3H), 4.59
(d, 3JHH = 5.8 Hz, H–C6H4, 1H), 2.51 (sep, 3JHH = 6.9 Hz,
CH–(CH3)2, 1H), 2.17 (pt, J = 14.3 Hz, CH2P, 1H), 2.00 (s,
CH3–C6H4, 3H), 1.17 (m, CH2P, 1H), 1.01 (d,
3
3JHH = 7.0 Hz, CH–(CH3)2, 3H), 0.63 (d, JHH = 6.9 Hz,
4.2.4. Synthesis of Rh-1P2
CH–(CH3)2, 3H), À0.15 (s, CH3Si, 9H). 13C{1H} NMR
This procedure was analogous to that used for Rh-1P1.
Starting from 1P2 (0.147 g, 0.394 mmol) and the rhodium
dimer [RhCl(COD)]2 (0.097 g, 0.197 mmol), a yellow solid
was obtained using hexane instead of diethyl ether to pre-
(62.9 MHz, CDCl3, 298 K), d (ppm): 146.1–126.3 (m, Ar),
107.7 (s, C–CH(CH3)2), 101.3 (s, C–CH3), 93.4 (d, JCP
=
6.0 Hz, C6H4), 87.2 (d, JCP = 7.5 Hz, C6H4), 86.8 (d,
JCP = 3.4 Hz, C6H4), 85.4 (d, JCP = 4.2 Hz, C6H4), 29.7
(s, CH(CH3)2), 22.9 (s, CH(CH3)2), 20.0 (s, CH(CH3)2),
cipitate the product. Yield: 0.228 g (93%). 31P{1H} NMR
1
1
(101.3 MHz, CDCl3, 298 K), d (ppm): 19.7 (d, JPRh
=
17.2 (s, CH3), 12.1 (d, JCP = 18.4 Hz, CH2P), 1.7 (d,
147.7 Hz). 1H NMR (400.1 MHz, CDCl3, 298 K), d
(ppm): 8.73–7.30 (m, Ar, 14H), 5.52 (m, CHtrans, 2H),
3.13 (m, CHcis, 1H), 2.98 (m, CHcis, 1H), 2.45–1.73 (m,
CH2 + CH2P, 10H), 0.16 (s, CH3Si, 9H). 13C{1H} NMR
(100.6 MHz, CDCl3, 298 K), d (ppm): 137.5–122.9 (m,
3JCP = 2.5 Hz, CH3Si). EA: Anal. Calc. for C32H39Cl2-
PRuSi: C, 58.71; H, 6.00. Found: C, 58.35; H, 6.32%.
MS (ES(+), m/z): 619.1 (619.2 calcd) [MÀCl]+.
4.2.7. Synthesis of Ru-2P1
1
Ar), 103.5 (m, CHtrans), 72.1 (pt, JCRh = 14.0 Hz, CHcis),
Experimental conditions were identical to those of the
preparation of Ru-1P1 using 0.150 g of dendrimer 2P1
(0.102 mmol) and 0.125 g of the ruthenium dimer
[RuCl2(p-cymene)]2 (0.204 mmol). A deep red compound
was obtained. Yield: 0.271 g (98%). 31P{1H} NMR
(101.3 MHz, CDCl3, 298 K), d (ppm): 21.0 (s (br)). 1H
NMR (400.1 MHz, CDCl3, 298 K), d (ppm): 8.63–6.68
1
1
33.5 (d, JCRh = 2.9 Hz, CH2), 33.4 (d, JCRh = 2.7 Hz,
1
1
CH2), 29.3 (d, JCRh = 1.0 Hz, CH2), 29.0 (d, JCRh
=
1
1.2 Hz, CH2), 15.8 (d, JCP = 13.7 Hz, CH2P), 1.9 (d,
3JCP = 2.8 Hz, CH3Si). EA: Anal. Calc. for C32H37-
ClPRhSi: C, 62.09; H, 6.02. Found: C, 62.20; H, 6.19.
MS (ES(+), m/z): 583.1 (583.6 calcd) [MÀCl]+.
3
(m, Ar, 56H), 5.27 (pt, JHH = 5.6 Hz, H–C6H4, 8H),
3
4.2.5. Synthesis of Rh-2P2
5.21 (d, JHH = 6.0 Hz, H–C6H4, 4H), 4.80 (d (br),
This complex was obtained using the same procedure as
for Rh-1P2. Starting from 2P2 (0.110 g, 0.070 mmol) and
the rhodium dimer [RhCl(COD)]2 (0.069 g, 0.140 mmol),
a yellow solid was obtained. Yield: 0.170 g (95%).
31P{1H} NMR (101.3 MHz, CDCl3, 298 K), d (ppm):
3JHH ꢁ 5 Hz, H–C6H4, 4H), 2.22 (s (br), CH-(CH3)2,
4H), 2.00 (pt, J = 14.0 Hz, CH2P, 4H), 1.92 (s, CH3–
3
C6H4, 12H), 1.34 (m, CH2P, 4H), 0.96 (d, JHH = 7.2 Hz,
3
CH–(CH3)2, 12H), 0.71 (d, JHH = 6.8 Hz, CH–(CH3)2,
12H), 0.14 to (À0.16) (m, CH2Si, 16H), À0.04 (s, CH3Si,
12H), À0.25 (s, CH3Si, 12H). 13C{1H} NMR
(100.6 MHz, CDCl3, 298 K), d (ppm): 146.3–126.5 (m,
Ar), 107.2 (s, C–CH(CH3)2), 96.7 (s (br), C–CH3), 91.1 (s
(br), C6H4), 87.3–86.7 (m, C6H4), 29.8 (s, CH(CH3)2),
23.4 (s, CH(CH3)2), 20.8 (s, CH(CH3)2), 17.7 (s, CH3),
13.7 (s (br), CH2P), 9.8 (s, CH2Si0), 3.1 (s, CH2Si1), À0.4
(s, CH3Si), À1.3 (s, CH3Si). 29Si{1H} NMR (49.7 MHz,
CDCl3, 298 K), d (ppm): 3.77 (s, Si1). EA: Anal. Calc. for
1
17.8 (d, JPRh = 147.1 Hz). 1H NMR 1H (400.1 MHz,
CDCl3, 298 K), d (ppm): 8.66–7.40 (m, Ar, 56H), 5.44
(s(br), CHtrans, 8H), 3.12 (s(br), CHcis, 4H), 2.90 (s(br),
CHcis, 4H), 2.40–1.60 (m, CH2 + CH2P, 40H), 0.40–0.19
(m, CH2Si, 16H), 0.21 (s, CH3Si, 12H), 0.08 (s, CH3Si,
12H). 13C{1H} NMR (100.6 MHz, CDCl3, 298 K), d
(ppm): 136.3–122.8 (m, Ar), 103.2 (m, CHtrans), 72.1
1
(d(br), JCRh = 13.6 Hz, CHcis), 33.5 (s, CH2), 32.7 (s,
CH2), 29.3 (s, CH2), 28.9 (s, CH2), 14.3 (d, 1JCP = 12.9 Hz,
C132H160Cl8P4Ru4 Si5: C, 58.74; H 5.97. Found: C, 58.80;
3
CH2P), 10.2 (d, JCP = 3.1 Hz, CH2Si0), 3.2 (s, CH2Si1),
H, 6.09%. MS (ES(+), m/z): 2666.7 (2663.5 calcd)
[MÀCl]+, 2359.7 (2357.3 calcd) [MÀRuCl2(p-cymene)À
Cl]+, 2048.2 (2051.1 calcd) [MÀ2(RuCl2(p-cymene))ÀCl]+,
1313.3 (1314.0 calcd) [MÀ2Cl]2+, 1160.9 (1160.9 calcd)
3
3
À0.5 (d, JCP = 2.7 Hz, CH3Si), À0.8 (d, JCP = 1.5 Hz,
CH3Si). 29Si{1H} NMR (49.7 MHz, CDCl3, 298 K), d
(ppm): 9.50 (s, Si0), 3.98 (s, Si1). EA: Calcd for
C132H152Cl4P4Rh4 Si5: C, 62.02; H, 5.99. Found: C,
[MÀRuCl2(p-cymene)À2Cl]2+
4.2.8. Synthesis of Ru-1P2
.
62.14; H, 6.08%. MS (ES(+), m/z): 2026.9 (2027.8 calcd)
[MÀ(RhCODCl)2ÀCl]+.
Experimental conditions and workup were identical to
those of the preparation of Ru-1P1 using 0.130 g of com-
pound 1P2 (0.349 mmol) and 0.107 g of the ruthenium
dimer [RuCl2(p-cymene)]2 (0.174 mmol). Yield: 0.225 g
(95%). 31P{1H} NMR (101.3 MHz, CDCl3, 298 K), d
4.2.6. Synthesis of Ru-1P1
The model compound 1P1 (0.150 g, 0.430 mmol) was
dissolved in 10 ml of CH2Cl2 and the ruthenium dimer
[RuCl2(p-cymene)]2 (0.132 g, 0.215 mmol) was added. After
stirring for 20 min, the solvent was removed and the result-
ing solid was washed with diethyl ether. The product Ru-
1P1 was obtained as a red solid. Yield: 0.253 g (90%).
1
(ppm): 23.2 (s). H NMR (400.1 MHz, CDCl3, 298 K), d
3
(ppm): 8.85–7.11 (m, Ar, 14H), 5.22 (d, JHH = 6.6 Hz,
3
H–C6H4, 1H), 5.01 (d, JHH = 6.2 Hz, H–C6H4, 2H), 4.39

L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
857
3
(d (br), JHH = 4.5 Hz, H–C6H4, 1H), 2.77 (pt, J = 4.2 Hz,
thermostated at 20 ꢁC. H2(g) was admitted until a pressure
of 10 bar was reached. After the time indicated for each
reaction, the autoclave was slowly depressurized and the
quantitative distribution of products and their e.e. were
determined by GC analysis.
3
CH2P, 1H), 2.66 (sep, JHH = 6.8 Hz, CH–(CH3)2, 1H),
2.01 (m, CH2P, 1H), 1.73 (s, CH3–C6H4, 3H), 0.99 (d,
3JHH = 7.0 Hz, CH–(CH3)2, 3H), 0.96 (d, JHH = 6.9 Hz,
3
CH–(CH3)2, 3H), À0.32 (s, CH3Si, 9H). 13C{1H} NMR
(100.6 MHz, CDCl3, 298 K), d (ppm): 134.6–123.2 (m,
Ar), 107.5 (s, C–CH(CH3)2), 95.4 (s, C–CH3), 92.5 (s,
4.3.2. Hydrogen transfer
2
C6H4), 89.0 (s (br), C6H4), 88.6 (d, JCP = 2.3 Hz, C6H4),
82.7 (s (br), C6H4), 30.5 (s, CH(CH3)2), 23.3 (s, CH(CH3)2),
21.1 (s, CH(CH3)2), 18.0 (s, CH3), 17.0 (s (br), CH2P), 0.9
The precursor complex (6 · 10À6 mol [Ru]) was dis-
solved in 3 ml of a freshly prepared solution 0.02 M of t-
BuOK in propan-2-ol at room temperature. The resulting
solution was stirred for 30 min. Then 10 ml of a 0.12 M
solution of acetophenone in propan-2-ol was added. The
solution was stirred at 82 ꢁC. Aliquots of the catalytic reac-
tion were passed through a short column (silica gel) eluted
with ethyl acetate before evaluation by GC of the quantita-
tive distribution of products and their e.e.
3
(d, JCP = 2.0 Hz, CH3Si). EA: Anal. Calc. for C34H39-
Cl2PRuSi: C, 60.17; H, 5.79. Found: C, 60.29; H, 5.93%.
MS (ES(+), m/z): 643.0 (643.2 calcd) [MÀCl]+.
4.2.9. Synthesis of Ru-2P2
Experimental conditions and workup were identical to
those of the preparation of Ru-1P1 using 0.100 g of dendri-
mer 2P2 (0.064 mmol) and 0.078 g of the ruthenium dimer
[RuCl2(p-cymene)]2 (0.127 mmol). Yield: 0.173 g (97%).
31P{1H} NMR (101.3 MHz, CDCl3, 298 K), d (ppm):
Acknowledgements
Financial support for this research was provided by the
DGICYT (Projects BQU2003-01131 and CTQ2004-01546).
L.-.I.R. is indebted to the Ministerio de Ciencia y Tec-
1
23.2 (s). H NMR (400.1 MHz, CDCl3, 298 K), d (ppm):
8.80–7.10 (m, Ar, 56H), 5.23 (d, JHH = 6.0 Hz, H–C6H4,
3
3
´
4H), 5.06 (d, JHH = 6.0 Hz, H–C6H4, 4H), 4.82 (d,
nologıa for a scholarship.
3JHH = 6.0 Hz, H–C6H4, 4H), 4.44 (d (br), JHH ꢁ 4 Hz,
3
H–C6H4, 4H), 2.80 (pt, J = 14.4 Hz, CH2P, 4H), 2.60
3
(sep, JHH = 6.8 Hz, CH–(CH3)2, 4H), 2.00 (m, CH2P,
4H), 1.71 (s, CH3–C6H4, 12H), 0.94 (d, JHH = 6.8 Hz,
CH–(CH3)2, 12H), 0.88 (d, JHH = 7.0 Hz, CH–(CH3)2,
References
3
3
[1] (a) For reviews of dendrimer catalysis, see: D. Astruc, F. Chardac,
Chem. Rev. 101 (2001) 2991;
12H), 0.12 to (À0.15) (m, CH2Si, 16H), À0.23 (s, CH3Si,
12H), À0.68 (s, CH3Si, 12H). 13C{1H} NMR (100.6 MHz,
CDCl3, 298 K), d (ppm): 134.6–122.9 (m, Ar), 107.7 (s,
C–CH(CH3)2), 94.7 (s, C–CH3), 92.5 (s (br), C6H4), 89.6
(s, C6H4), 88.3 (s (br), C6H4), 82.7 (s (br), C6H4), 30.5 (s,
CH(CH3)2), 23.3 (s, CH(CH3)2), 21.1 (s, CH(CH3)2), 18.0
(s, CH3), 15.6 (s (br), CH2P), 9.8 (s, CH2Si0), 2.9 (s,
CH2Si1), À1.0 (s, CH3Si), À2.3 (s, CH3Si). 29Si{1H}
NMR (49.7 MHz, CDCl3, 298 K), d (ppm): 3.62 (d,
(b) R.M. Crooks, M. Zhao, L. Sun, V. Chechik, L.K. Yueng, Acc.
Chem. Res. 34 (2001) 181;
(c) L.J. Twyman, A.S.H. King, I.K. Martin, Chem. Soc. Rev. 31
(2002) 69;
(d) G.E. Oosterom, J.N.H. Reek, P.C.J. Kramer, P.W.N.M. van
Leeuwen, Angew. Chem., Int. Ed. 40 (2001) 1828;
(e) J.P. Majoral, A.M. Caminade, Chem. Rev. 99 (1999) 845;
(f) B. Romagnoli, W. Hayes, J. Mater. Chem. 12 (2002) 769;
´
(g) B. Helms, J.M.J. Frechet, Adv. Synth. Catal. 348 (2006) 1125.
[2] M.S.T.H. Sanders-Hoven, J. Jansen, J. Vekemans, E. Meijer, Polym.
Mater. Sci. Eng. 210 (1995) 180.
[3] (a) D. Seebach, R. Marti, T.Y. Hintermann, Helv. Chim. Acta 79
(1996) 1710;
(b) P.B. Rheiner, D. Seebach, Polym. Mater. Sci. Eng. 77 (1997) 130.
[4] R. Breinbauer, E. Jacobsen, Angew. Chem., Int. Ed. Engl. 39 (2000)
3604.
[5] (a) G.D. Engel, L.H. Gade, Chem. Eur. J. 8 (2002) 4319;
(b) Y. Riboudouille, G. Engel, M. Richard-Plouet, L. Gade, Chem.
Commun. (2003) 1228.
2JSiP = 7.4 Hz, Si1). EA: Anal. Calc. for C140H160
-
Cl8P4Ru4 Si5: C, 60.16; H, 5.77. Found: C, 58.35; H,
5.89%. MS (ES(+), m/z): 2758.5 (2759.6 calcd) [MÀCl]+,
1362.9 (1362.0 calcd) [MÀ2Cl]2+
4.3. Catalytic reactions
.
4.3.1. Hydrogenation
[6] (a) C. Ko¨llner, A. Pugin, A. Togni, J. Am. Chem. Soc. 120 (1998)
10274;
Hydrogenation reactions were performed in a stainless-
steel autoclave fitted with an external jacket connected to
an isobutanol bath. The temperature was controlled using
a thermostat to 0.5 ꢁC. Internal temperature was moni-
tored using a thermopar coupled to a digital recorder,
whereas the internal pressure was continuously measured
as a function of time with a Linseis L-200 recorder.
(b) R. Schneider, C. Ko¨llner, L. Weber, A. Togni, Chem. Commun.
(1999) 2415;
(c) A. Togni, N. Bieler, U. Burckhardt, C. Ko¨llner, G. Pioda, R.
Schneider, A. Schnyder, Pure Appl. Chem. 71 (1999) 1531;
(d) C. Ko¨llner, A. Togni, Can. J. Chem. 79 (2001) 1762.
[7] R. Laurent, A.M. Caminade, J.P. Majoral, Tetrahedron Lett. 46
(2005) 6503.
[8] (a) Q. Fan, Y. Chen, X. Chen, D. Jiang, F. Xi, A. Chan, Chem.
Commun. (2000) 789;
Dimethyl itaconate (0.01 mol), the precursor complex
(2 · 10À5 mol [Rh]) and, eventually, PPh3 (2 · 10À5 mol)
were dissolved in 20 ml of thf. The resulting solution was
immediately placed in the autoclave, which had previously
been purged by successive vacuum/nitrogen cycles and
(b) G. Deng, Q. Fan, X. Chen, D. Liu, A. Chan, Chem. Commun.
(2002) 1570;
(c) G. Deng, Q. Fan, X. Chen, D. Liu, J. Mol. Catal. A., Chem. 193
(2003) 21.

858
L.-I. Rodrı´guez et al. / Journal of Organometallic Chemistry 692 (2007) 851–858
´
[9] L.-I. Rodrıguez, O. Rossell, M. Seco, A. Grabulosa, G. Muller, M.
[16] A. Duursma, R. Hoen, J. Schuppan, R. Hulst, A.J. Minnaard, B.L.
Feringa, Org. Lett. 5 (2003) 3111.
Rocamora, Organometallics 25 (2006) 1368.
[10] I. Angurell, G. Muller, M. Rocamora, O. Rossell, M. Seco, Dalton
Trans. (2004) 2450.
[17] (a) See, for example: J. Hannedouche, G.J. Clarkson, M. Wills, J.
Am. Chem. Soc. 126 (2004) 986;
[11] M.J. Burk, Acc. Chem. Res. 33 (2000) 363.
(b) C. Thoumazet, M. Melaimi, L. Ricard, F. Mathey, P. Le Floch,
Organometallics 22 (2003) 1580;
´
[12] I. Serrano, M. Rodrıguez, I. Romero, A. Llobet, T. Parella, J.M.
´
´
Campelo, D. Luna, J.M. Marinas, J. Benet-Buchholz, Inorg. Chem.
45 (6) (2006) 2644 (and references therein).
(c) V. Cadierno, P. Crochet, J. Garcıa-Alvarez, S.E. Garcıa-Garrido,
J. Gimeno, J. Organomet. Chem. 663 (2002) 32;
´
[13] C.E. Housecroft, C.M. Che, T.C. Lau, in: J.A. McCleverty, T.J.
Meyer (Eds.), Comprehensive Coordination Chemistry, vol. 5,
Elsevier Pergamon, Oxford, UK, 2004 (Chapters 5.5 and 5.6).
(d) M. Gomez, S. Jansat, G. Muller, M.C. Bonnet, J.A.J. Breuzard,
M. Lemaire, J. Organomet. Chem. 659 (2002) 186;
(e) P. Crochet, J. Gimeno, S. Garcıa-Granda, J. Borge, Organomet-
´
´
[14] (a) M. Bardajı, A.M. Caminade, J.P. Majoral, B. Chaudret,
allics 20 (2001) 4369;
Organometallics 16 (1997) 3489;
(b) N.J. Hovestad, E.B. Eggeling, H.J. Heidbuchel, J.T.B.H. Jastr-
¨
zebski, U. Kragl, W. Keim, D. Vogt, G. van Koten, Angew. Chem.,
Int. Ed. 38 (1999) 1655;
(c) E.B. Eggeling, N.J. Hovestad, J.T.B.H. Jastrzebski, D. Vogt, G.
van Koten, J. Org. Chem. 65 (2000) 8857.
(f) C. Standfest-Hauser, C. Slugovc, K. Mereiter, R. Schmid,
K. Kirchner, L. Xiao, W. Weissensteiner, Dalton Trans. (2001)
2989;
(g) R. Noyori, S. Hashiguchi, Acc. Chem. Res. 30 (1997) 97.
[18] (a) Y. Chen, T. Wu, J. Deng, H. Liu, Y. Jiang, M.C.K. Choi, A.S.C.
Chan, Chem. Commun. (2001) 1488;
[15] M.T. Reetz, Y. Fu, A. Meiswinkel, Angew. Chem., Int. Ed. 45 (2006)
1412 (and references therein).
(b) P.N. Liu, Y.C. Chen, X.Q. Li, Y.Q. Tu, J.G. Deng, Tetrahedron:
Asymmetry 15 (2004) 2481.