Cationic complexes of dirhodium(II) with 1,8-naphthyridine: Catalysis of reactions involving silanes

Article abstract of DOI:10.1016/j.jorganchem.2006.04.030

The synthesis of cationic dirhodium(II) complexes by partial or total substitution of the acetate groups of [Rh₂(OAc)₄] with different homoleptic neutral bidentate ligands has been attempted. The ligand 1,8-naphthyridine gave the bes

Full text of DOI:10.1016/j.jorganchem.2006.04.030

Journal of Organometallic Chemistry 691 (2006) 3464–3471
www.elsevier.com/locate/jorganchem
Cationic complexes of dirhodium(II) with 1,8-naphthyridine:
Catalysis of reactions involving silanes
a,
a
a
a
b
Marino Basato *, Andrea Biffis , Gianluca Martinati , Cristina Tubaro , Claudia Graiff ,
b
c
c
Antonio Tiripicchio , Laura A. Aronica , Anna M. Caporusso
a
`
Dipartimento di Scienze Chimiche, Universita di Padova, and ISTM-CNR, via Marzolo 1, I-35131 Padova, Italy
b
`
Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica, Universita di Parma,
Parco Area delle Scienze 17/A, I-43100 Parma, Italy
c
`
Dipartimento di Chimica e Chimica Industriale, Universita di Pisa, via Risorgimento 35, I-56126 Pisa, Italy
Received 24 February 2006; received in revised form 20 April 2006; accepted 25 April 2006
Available online 30 April 2006
Abstract
The synthesis of cationic dirhodium(II) complexes by partial or total substitution of the acetate groups of [Rh2(OAc)4] with different
homoleptic neutral bidentate ligands has been attempted. The ligand 1,8-naphthyridine gave the best results: substitution of one as well
as of all four acetate ligands is possible, giving rise to mono-, di- and tetra-cationic complexes. One of the resulting tetrasubstituted com-
plexes has been structurally characterised and found to exhibit the expected lantern-shaped structure. All cationic complexes have been
investigated as catalysts in different reactions involving silanes: promising results have been obtained, particularly in the silylformylation
of alkynes.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Rhodium; 1,8-naphthyridine; Hydrosilylation; Silylformylation; Silane alcoholysis
1. Introduction
recognised fundamental importance of the electrophilicity
of these complexes in determining their activity and selec-
tivity in catalytic reactions, it is quite surprising that up
to now no systematic study on the catalytic efficiency of
related cationic complexes has been carried out.
We have an ongoing research program aimed at the prep-
aration of different kinds of cationic dirhodium(II) com-
plexes and at the evaluation of their catalytic performance
in reactions involving diazocompounds or silanes. Com-
plexes of this kind can be prepared starting from simple
dirhodium(II) acetate upon partial or complete substitution
of the acetate moieties with neutral ligands. Although cat-
ionic dirhodium(II) complexes in which the acetate ligands
have been substituted by coordinated solvent molecules,
most commonly acetonitrile, are quite well known [6], much
less common are examples in which the acetate ligands have
been substituted by bidentate neutral ligands [7,8].
Neutral dinuclear complexes of rhodium(II) are known
to be catalytically active in a number of synthetically useful
transformations [1]. The most popular reactions are metal-
mediated decompositions of diazo derivatives, which upon
formation of highly reactive metal carbene intermediates
lead to cyclopropanations, insertions of the carbene moiety
into C–H, C–C or heteroatom–H bonds, ylide formations
or dipolar cycloaddition reactions; efficient asymmetric
variants of many of these reactions have been developed
as well [2,3]. Furthermore, neutral, electron poor dirho-
dium(II) complexes, such as dirhodium(II) perfluorocarb-
oxylates do also efficiently catalyse many interesting
reactions involving silanes, such as silane alcoholysis,
hydrosilylations or silylformylations [4,5]. Despite the
*
We have recently reported on cationic dirhodium(II)
complexes with oxothioether molecules which exhibit a
Corresponding author. Tel.: + 39 049 8275217; fax: +39 049 8275223.
E-mail address: marino.basato@unipd.it (M. Basato).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.04.030

M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
3465
rather unusual bidentate O–S coordination, in which each
bidentate ligand occupies one equatorial and one axial site
in the complex; preliminary tests indicate that such com-
plexes are promising catalysts of both the silylformylation
and the hydrosilylation of 1-hexyne with dimethylphenylsi-
lane [9].
exception to this general behaviour was observed with the
1,8-naphthyridine ligand (4). When Rh2(OAc)4 was reacted
at room temperature with one or two equivalents of (4), no
product derived from simple ligand addition was observed.
Instead, a well-defined cationic complex was formed stem-
ming from the neat substitution with a 1,8-naphthyridine
ligand of one acetate ligand, which then acted as counter-
ion (Scheme 2). Remarkably, no addition of a strong acid
such as HCl, as reported, for example in Ref. [7], was nec-
essary in order to allow the reaction to proceed. The struc-
ture of the complex was confirmed spectroscopically: the
1H-NMR spectrum clearly reveals the presence of two
kinds of bridging acetate ligands in a 2:1 ratio, as well as
of an acetate counterion. Only the monosubstituted prod-
uct (5) was formed, even in the presence of excess ligand.
The substitution process could be pushed further by forc-
ing the reaction conditions. Thus, by reacting Rh2(OAc)4
with four equivalents of (4) in acetic acid at reflux, com-
plete substitution of the acetate ligands occurred, with for-
mation of complex (6). The resulting complex was however
only dicationic, in that two acetate ligands remained in the
coordination sphere of the complex by occupying its apical
positions, the other two acting as counterions, as confirmed
by NMR analysis. Since the apical positions of dirho-
dium(II) complexes are usually their catalytically active
sites, we expected their occupation by acetate ligands to
have a negative effect on their reactivity. Therefore, we
tried to remove the apical acetates by anion exchange with
NaBPh4. Such treatment only caused the substitution of
the acetate counterions with two BPhꢀ4 anions, but left
the coordinated acetates untouched, yielding complex (7).
The apical acetates could finally be removed by treatment
with Meerwein’s salt triethyloxonium tetrafluoborate,
which ethylated them yielding the tetracationic complex
(8).
We wanted to extend such investigations to cationic
complexes in which the lantern-shaped structure of the
[Rh2(OAc)4] precursor is preserved. Such complexes can
be obtained by partial or total substitution of the acetate
moieties with homoleptic rather than heteroleptic neutral
bidentate ligands, as demonstrated in the literature by the
successful application of ligands based on 1,8-naphthyri-
dine [7,8]. Thus, in this contribution we wish to report on
the development of an improved synthesis of cationic
dirhodium(II) complexes with 1,8-naphthyridine and on
the results of a systematic study on the catalytic activity
of these complexes in different model reactions involving
silanes, such as silylformylations, hydrosilylations and
silane alcoholysis. Furthermore, preliminary results on
our attempts to generalise this strategy for the synthesis
of cationic dirhodium(II) complexes to other neutral biden-
tate ligands will be presented.
2. Results and discussion
In the course of our study, we have investigated four dif-
ferent potential neutral bidentate ligands of N–N, P–P and
S–S types, which are reported in Scheme 1.
The outcome of this investigation was that, in the case of
P–P and S–S ligands, the bidentate ligands gave upon reac-
tion with Rh2(OAc)4 in toluene at room temperature sim-
ple coordination to the free apical positions of the metal
precursor, yielding poorly soluble coordination polymers.
Upon forcing the reaction conditions by prolonged heating
of the resulting products in acetic acid in order to labilise
the acetate ligands and favour their substitution, the coor-
dination polymer derived from (1) dissolved, yielding how-
ever a mixture of products of common stoichiometry
[Rh2(MeSCH2SMe)2(OAc)4], in which the ligand (1) prob-
ably occupies different coordination sites. In contrast, the
coordination polymer derived from ligand (2) remained
virtually insoluble, whereas in the case of the P–P ligand
(3) the mononuclear rhodium(III) complex [Rh(P–P)-
(OAc)3] was formed, in accordance to previous observa-
tions by Cotton et al. with the same diphosphine ligand
but using a different synthetic procedure [10]. The only
The structure of complex (7) Æ 10CH3CN was deter-
mined by X-ray diffraction methods. In the crystals, cations
[Rh2(OAc)2(Naft)4]2+ (Naft = 1,8-naphthyridine), anions
BPhꢀ4 and solvent molecules CH3CN were found. An
ORTEP view of the cation is shown in Fig. 1, together with
the atomic labelling scheme. A selection of the most impor-
tant bond distances and angles is listed in Table 1. The cat-
ionic complex is centrosymmetric with the inversion center
on the midpoint of the Rh–Rh bond. The coordination of
each Rh atom is essentially octahedral, with four nitrogen
atoms of the four naphthyridine ligands lying on the
equatorial plane, the acetate ion and the other Rh atom
on the apical sites completing the octahedral coordination.
˚
The Rh–Rh bond distance of 2.448(1) A falls within the
MeS
SMe
PhS
SPh
expected range for dirhodium(II) complexes and is slightly
longer than that reported for the complex [Rh2(OAc)4-
(H2O)2] [11]; the four naphthyridine ligands bridge the
(2)
(1)
Rh atoms through the N atoms with Rh–N bond lengths
˚
Ph2P
PPh2
ranging from 2.050(3) to 2.077(3) A, typical for N(pyridil)
–
N
N
Rh coordination [12]. The acetate anions are bound to
the Rh through an oxygen atom (Rh1–O1 bond length of
(3)
(4)
0
˚
Scheme 1. Sketch of the employed bidentate ligands.
2.297(2) A and O1–Rh1–Rh1 bond angle of 173.94(5)ꢁ).

3466
M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
O
O
N
Rh
O
N
Toluene, r.t.
(OAc)
Rh
O
[Rh2(OAc)4] +
O
O
N
N
(5)
N
N
AcO
N
N
Rh
Acetic acid
N
(OAc)2
Rh
[Rh2(OAc)4] + 4
N
reflux, 8 h
OAc
N
N
N
N
(6)
Na(BPh4)
MeOH/ Tol 4/1
N
N
N
N
AcO
MeCN
N
N
(Et3O)(BF4)
MeCN
Rh
Rh
N
N
(BPh4)2
N
N
(BF4)4
Rh
Rh
N
N
OAc
MeCN
N
N
N
N
(8)
(7)
Scheme 2. Synthesis of the complexes.
In the crystals several CH3CN solvent molecules are
located in the voids left by the ions, forming with them
an intricated network of weak Van der Waals interactions.
A view of the crystal packing is reported in Fig. 2 (view
along the crystallographic a-axis).
The catalytic activity of the various complexes was eval-
uated in a series of technologically relevant reactions
involving silanes, such as the hydrosilylation of alkynes,
the silylformylation of alkynes and the alcoholysis of sil-
anes. The model reactions employed are reported in
Scheme 3. In order to have some benchmark catalysts with
which to critically evaluate the activity of the synthesised
complexes, we involved in our study two additional cata-
lysts, namely the parent dirhodium(II) acetate and the cat-
ionic complex [Rh2(OAc)2(MeCN)6](BF4)2 bearing simple
acetonitrile molecules as acetate substituents. Although
this complex has long been known in the literature [6], it
has to the best of our knowledge never been employed as
catalyst before. The results of the various tests are reported
in Table 2.
The hydrosilylation of 1-hexyne with dimethylphenylsi-
lane was run at 90 ꢁC in a solventless fashion using an
excess of 1-hexyne as the solvent (catalyst/silane/hexyne
ratio 1/1000/4000). The best results were obtained with
the simple Rh2(OAc)4 precursor: (Z)-(1-dimethylphenylsi-
lyl)-1-hexene was formed in 87% yield with 98% selectivity
after 6 h, other products stemming mainly from the formal
cis-addition of the silane to the triple bond. The cationic
reference catalyst showed higher conversion but also
poorer selectivity for the main reaction product. The less
thermodynamically stable Z-product remained however
largely predominant. This is in contrast with other reports

M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
3467
Fig. 2. Crystal packing of [Rh2(OAc)2(Naft)4](BPh4)2 Æ 10CH3CN. View
along the crystallographic a-axis.
Fig. 1. View of the cation [Rh2(OAc)2(Naft)4]2+
numbering scheme.
, with the atomic
lower temperature than the hydrosilylation reactions,
namely at room temperature, under 10 atm of CO, using
a 1/1000 catalyst/substrate ratio. In the case of this reac-
tion the reactivity order of the various complexes changed
quite radically. In fact, the parent Rh2(OAc)4 exhibited
very poor productivity, whereas the cationic complexes
were much more effective. Only the Z isomer of the silyl-
formylated product was observed, which highlights the
high regio- and stereoselectivity that can be obtained with
these catalysts; other by-products stemmed almost exclu-
sively from hydrosilylation of the triple bond, which is
known to be the main competitive reaction. Particularly
interesting was also the reactivity trend with the tetrasub-
stituted naphthyridine complexes. Whereas complexes (6)
and (7) were almost inactive, complex (8) exhibited a com-
parable conversion (90% yield) and a higher selectivity
(99%) than the cationic complex with acetonitrile ligands.
These results highlight the importance of having free apical
coordination sites in the complex catalyst. Compared with
the previously reported cationic dirhodium(II) complexes
with oxothioethers, we have obtained comparable conver-
sions but much higher selectivities with the best catalysts
presented in this work. Although not optimised, the cata-
lytic performance of these complexes compares favourably
with that of other rhodium-based catalysts such as dirho-
dium(II) perfluorobutyrate [4d,4e], Rh4(CO)12 [15,16] or
solvated rhodium atoms [15]. Thus, it can be stated that
the presence of a positive charge on the dirhodium(II) cat-
alyst does significantly promote the silylformylation
reaction.
from the literature where cationic rhodium-based catalysts
were shown to exhibit a preference for cis hydrosilylation
leading to the E isomer [13], but in accordance with results
obtained with neutral, electron poor dirhodium(II) perfluo-
robutyrate [4b]. The performance of the naphthyridine
complexes appeared to be much worse, both in terms of
conversion and selectivity, although the Z-isomer remained
predominant. Furthermore, two of the complexes turned
out to be insoluble in the reaction mixture and were there-
fore not evaluated. Since also the performance of cationic
dirhodium(II) complexes with oxothioethers was found to
be inferior to that of simple Rh2(OAc)4 [8], it can be con-
cluded that the presence of a positive charge on the dirho-
dium(II) catalyst does not appear to enhance its reactivity
in this reaction [14]. We are currently attempting to find a
rational explanation for the exceptional productivity
observed with complex [Rh2(OAc)2(MeCN)6](BF4)2.
The silylformylation of 1-hexyne with dimethylphenylsi-
lane could be performed in dichloromethane at a much
Table 1
˚
Selected bond distances (A) and angles (deg) in compound
[Rh2(OAc)2(Naft)4](BPh4)2 Æ 10CH3CN
Rh1–Rh10
Rh1–N3
Rh1–N20
2.448(1)
2.050(3)
2.050(2)
2.057(2)
89.88(9)
88.16(9)
90.06(9)
91.73(9)
95.99(9)
94.45(8)
88.79(8)
Rh1–N40
Rh1–O1
O1–C25
O2–C25
N40–Rh1–O1
N3–Rh1–Rh10
N20–Rh1–Rh10
N1–Rh1–Rh10
N40–Rh1–Rh10
O1–Rh1–Rh10
2.077(3)
2.297(2)
1.269(3)
1.245(3)
87.03(9)
89.31(7)
88.49(6)
88.44(6)
87.67(7)
173.94(5)
Rh1–N1
N3–Rh1–N20
N3–Rh1–N1
N20–Rh1–N40
N1–Rh1–N40
N3–Rh1–O1
N20–Rh1–O1
N1–Rh1–O1
Finally, the silane alcoholysis of triethylsilane with ben-
zyl alcohol was run at 50 ꢁC in a solventless fashion using a
1/1000 catalyst/substrate ratio. None of the employed cat-
alysts was found to be very productive in this reaction.
There is, however, a significant difference in the catalytic
efficiency of the various complexes. In fact, whereas the
Symmetry transformations used to generate equivalent atoms: ꢀx + 2,
ꢀy + 1, ꢀz + 1.

3468
M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
Bu
H
H
Bu
H
RT, DCM, 10 bar CO
0.1 mol% cat., 6h
Bu
C
CH
+
Me2PhSiH
+
SiMe2Ph
OHC
SiMe2Ph
Bu
H
SiMe2Ph
OSiMe2Ph
Bu
SiMe2Ph
Bu
H
H
90 ˚C, no solvent
0.1 mol% cat., 5h
Bu
C
CH
+
Me2PhSiH
+
H
H
SiMe2Ph
Bu
PhMe2Si
H
H
50 ˚C, no solvent
0.1 mol% cat., 24h
+ H2
CH OH
+ Et3SiH
CH2OSiEt3
2
Scheme 3. Catalytic reactions investigated in the present work.
Table 2
Conversions and selectivities (in brackets) of the catalytic reactions with the various complexes
Catalyst
Hydrosilylation
Silylformylation
Silane alcoholysis
Rh2(OAc)4
87(98)
100(90)
47(77)
31(84)
n.s.a
6(59)
93(94)
73(93)
0
7(64)
90(99)
10(100)
25(100)
19(100)
19(100)
n.s.a
[Rh2(OAc)2(MeCN)6](BF4)2
[Rh2(OAc)3(Naft)](OAc) (5)
[Rh2(OAc)2(Naft)4](OAc)2 (6)
[Rh2(OAc)2(Naft)4](BPh4)2 (7)
[Rh2(MeCN)2(Naft)4](BF4)4 (8)
n.s.a
3(100)
Reaction conditions: see Scheme 3.
a
Catalyst not soluble.
Rh2(OAc)4 precursor reacted very slowly but did not
apparently undergo decomposition in the course of the
reaction, the cationic reference catalyst appeared to be
extremely active, causing a sudden burst of dihydrogen
production in the reaction mixture as soon as the silane
was added; this catalyst, however, apparently deactivated
very rapidly, thus finally giving an unsatisfactory yield.
The naphthyridine complexes also exhibited poor perfor-
mance, with the tetracationic complex (8) quite surprisingly
being the worst catalyst in this reaction. A kinetic curve
obtained with catalyst (6) and prolonged reaction times
(Fig. 3) showed that the behaviour of such complexes is
similar to the parent dirhodium acetate, in that the low
yield is due to the intrinsically low catalyst activity and
not to its rapid deactivation.
Thus, it would appear that cationic dirhodium(II) com-
plexes are worse catalysts for this reaction than neutral,
electron poor dirhodium(II) complexes, such as dirhodium
perfluorocarboxylates. The latter catalysts were recently
reported to reach almost complete conversion in model
reactions under the same reaction conditions, even with a
1/10000 catalyst/substrate ratio [5b,5c]. However, very
recent results from our group show that cationic dirho-
dium(II) complexes with oxothioether ligands exhibit even
superior catalytic activities under similar reaction condi-
tions (A. Biffis, M. Brichese, to be published). An explana-
tion for such widely different results may be traced back to
the different ligand set used. In fact, it is known that dirho-
dium(II) catalysts are able to coordinate both reaction
partners of the silane alcoholysis reaction; however, coordi-

M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
3469
pared according to the literature [6a]. Unless otherwise
noted, solvents were dried before use and the reaction appa-
ratus carefully deoxygenated; reactions were performed
under argon and all operations were carried out under an
inert atmosphere. The solution 1H- and 13C{1H}-NMR
spectra were acquired on Bruker Avance 300 MHz at room
temperature. The chemical shifts were determined by refer-
ence to the residual solvent peaks, using tetramethylsilane
as internal standard. The FTIR spectra were recorded on
a Biorad FT S7 PC spectrophotometer at 2 cmꢀ1 resolution
in KBr disks.
60
50
40
30
20
10
0
3.2. Synthesis of the complexes
0
1000 2000 3000 4000 5000 6000 7000
3.2.1. [Rh2(OAc)3(Naft)](OAc) (5)
time (m)
1,8-naphthyridine (70 mg, 0.54 mmol) was added to a
suspension of Rh2(OAc)4 (200 mg, 0.45 mmol) in toluene
(25 mL); the reaction mixture was stirred for 3 h at room
temperature and then evaporated to small volume under
reduced pressure. Treatment with diethyl ether gave a
brown precipitate, which was filtered and dried under vac-
uum (70% yield). Anal. Calc. for C16H18N2O8Rh2
(M = 572.14): C, 33.59; H, 3.17; N, 4.90. Found: C,
Fig. 3. Kinetic curve of triethylsilane alcoholysis with benzyl alcohol and
catalyst (6).
nation of the silane leads to substrate activation and there-
fore to the catalytic event, whereas coordination of the
alcohol leads to catalyst deactivation [4a]. Although the
presence of a positive charge on the complex should
increase its electrophilicity and favour silane activation, it
may also lead to an increase in the ‘‘hardness’’ of the rho-
dium center, hence to the preferential coordination of oxy-
gen donor ligands such as the alcohol. The choice of a
proper ligand set should greatly help in modulating the
character of the rhodium center, hence in promoting the
catalytic activity.
1
32.98; H, 3.16; N, 5.09%. H NMR (CDCl3): d 1.81 (br,
3H, CH3COꢀ), 1.94 (s, 6H, CH3COꢀ), 2.60 (br, 3H,
2
2
CH3COꢀ), 7.66 (m, 2H, Naft), 8.16 (m, 2H, Naft), 11.14
2
(m, 2H, Naft). 13C{1H} NMR (CDCl3): d 24.8, 25.0 and
26.3 (CH3COꢀ), 125.1, 126.2, 140.4, 163.1 and 164.0
2
(Naft), 191.8 and 192.1 (COꢀ2 ). FT IR (KBr, cmꢀ1): 3054,
2853, 1559, 1412, 787 and 708.
In conclusion, we have shown that the presence of a
positive charge on a dirhodium(II) complex affects its cat-
alytic efficiency in a manner that strongly depends on the
target reaction. In particular, cationic dirhodium(II) com-
plexes with 1,8-naphthyridine ligands appear to be promis-
ing catalysts for the silylformylation of alkynes, whereas
they are much less efficient in the hydrosilylation of alky-
nes. In the case of silane alcoholysis, the catalytic efficiency
appears to have a stronger dependence on the particular
ligand set used rather than on the complex charge. In order
to rationalise all these experimental observation, a thor-
ough mechanistic study of the underlying chemical pro-
cesses is needed. We are currently engaged in designing
target experiments which should clarify the role of the com-
plex charge in these reactions, particularly in the case of the
silylformylation reaction, for which the mechanism of
action of dirhodium(II) catalysts is still quite obscure [4e].
3.2.2. [Rh2(OAc)2(Naft)4](OAc)2 (6)
1,8-naphthyridine (270 mg, 2.09 mmol) was added to a
suspension of Rh2(OAc)4 (200 mg, 0.45 mmol) in acetic acid
(30 mL); the reaction mixture was stirred for 8 h under reflux
and then evaporated to small volume under reduce pressure;
treatment with diethyl ether afforded an orange solid, which
was filtered and dried under vacuum (61% yield). Anal. Calc.
for C40H36N8O8Rh2 Æ 4HOAc (M = 1202.79): C, 47.93; H,
1
4.35; N, 9.31. Found: C, 47.50; H, 4.24; N, 9.49. H NMR
(CDCl3): d 1.79 (s, 9H, CH3COꢀ and HOAc), 2.65 (s, 3H,
2
CH3COꢀ), 7.78 (m, 4H, Naft), 8.67 (m, 4H, Naft), 9.20 (br,
2
2H, HOAc), 10.31 (m, 4H, Naft). 13C{1H} NMR (CDCl3):
d 22.9 and 27.4 (CH3COꢀ and HOAc), 125.0, 126.1, 143.0,
2
159.7 and 161.3, 175.4 (COꢀ2 and HOAc), 179.3 (CO2ꢀ). FT
IR (KBr, cmꢀ1): 3420, 3059, 2851, 1711, 1582, 1373, 1325,
841 and 793.
3.2.3. [Rh2(OAc)2(Naft)4](BPh4) (7)
2
3. Experimental section
A solution of excess NaBPh4 in methanol (5 mL) was
added dropwise and under vigorous stirring to a solution
of complex (6) (200 mg, 0.17 mmol) in a mixture metha-
nol/toluene 4/1 (10 mL); the yellow solid which precipi-
tated was filtered, washed with methanol (2 · 5 mL) and
dried under vacuum (71% yield). Anal. Calc. for
C84H70B2N8O4Rh2 (M = 1482.96): C, 68.03; H, 4.76; N,
7.56. Found: C, 67.76; H, 4.89; N, 7.48. 1H NMR
3.1. General procedures
The reagents (Aldrich-Chemie) were high purity prod-
ucts and generally used as received. 1,8-Naphthyridine
(Naft) was prepared following the method of Skraup [17].
The cationic complex [Rh2(OAc)2(MeCN)6](BF4)2 was pre-

3470
M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
(CD3CN): d 2.60 (s, 3H, CH3COꢀ), 6.80, 6.96, 7.24 (m,
Table 3
2
Crystallographic data for the compound [Rh2(OAc)2(Naft)4](BPh4)2 Æ
10CH3CN
20H, BPh4), 7.66 (m, 4H, Naft), 8.36 (m, 4H, Naft),
10.38 (m, 4H, Naft). 13C{1H} NMR (CD3CN): d 27.4
(CH3COꢀ), 122.7 (Ph), 125.4 (Naft), 126.6 (Ph), 127.1
Formula
C104 H100 B2 N18 O4 Rh2
2
Formula weight
Crystal system
Space group
1893.46
Triclinic
P-1
12.485(3)
14.119(5)
15.764(5)
(Naft), 136.6 (Ph), 143.0 (Naft), 161.2 (Naft), 162.3 (Naft),
165.1 (Ph), 179.1 (CO2ꢀ). FT IR (KBr, cmꢀ1): 3053, 2985,
1578, 1371, 1515, 1324, 839, 789, 732 and 704.
˚
a (A)
˚
b (A)
˚
3.2.4. [Rh2(MeCN)2(Naft)4](BF4)4 (8)
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
109.87(5)
109.50(5)
97.83(3)
2363.8(13)
1, 1.33
982
33.18
9248
8891 (Rint = 0.0241)
8531
591
R1 = 0.0407, wR2 = 0.1110
R1 = 0.0419, wR2 = 0.1124
(Et3O)(BF4) (1 M solution in CH2Cl2, 0.73 mL,
73.00 mmol) was added to a solution of complex (7)
(180 mg, 12.1 mmol) in CH3CN (15 mL). The reaction mix-
ture was stirred for 24 h, then treated with diethyl ether
(20 mL) to precipitate the product, as a green solid, which
was filtered and dried under vacuum (88% yield). Anal. Calc.
for C36H30N10B4F16Rh2 (M = 1155.67): C, 37.41; H, 2.61;
3
˚
V (A )
Z, Dcalcd (g cmꢀ3
F(000)
l, cmꢀ1
Refl. collected
Refl. unique
Obs. Refl. [I > 2r(I)]
Parameters
)
1
N, 12.11. Found: C, 36.90; H, 2.51; N, 12.07. H NMR
(CD3CN): d 1.95 (s, 3H, CH3CN), 7.82 (m, 4H, Naft), 8.58
(m, 4H, Naft), 9.66 (m, 4H, Naft). 13C{1H} NMR (CD3CN):
d 126.8 (Naft), 128.1 (Naft), 144.4 (Naft), 160.7 (Naft), 162.6
(Naft), acetonitrile signals not detected. FT IR (KBr, cmꢀ1):
3441, 2200, 1606, 1513, 1056, 839 and 788.
Final R indices [I > 2r(I)]
Final R indices all data
2
2 1=2
R1 ¼ RkF o j ꢀ j F ck=R j F o j; wR2 ¼ ½R½wðF 2o ꢀ F c2Þ ꢁ=R½wðF o2Þ ꢁꢁ
.
charged with the rhodium catalyst (0.008 mmol), evacuated
and filled with argon. 0.8 mL (7.89 mmol) benzyl alcohol
were then added and the resulting solution was heated with
stirring to 50 ꢁC in a thermostated oil bath. After addition
of 1.9 mL (11.81 mmol, 1.5 eq.) triethylsilane the solution
was further stirred at 50 ꢁC for 24 h. 0.1 mL samples were
withdrawn after this period, diluted with 1 mL dichloro-
3.3. Catalytic tests
3.3.1. Silylformylations
The silylformylation reaction was performed in a 25 mL
stainless steel autoclave fitted with a Teflon inner crucible
and a stirring bar. Me2PhSiH, (0.31 mL, 2 mmol), 1-hexyne
(0.23 mL, 2 mmol) and rhodium catalyst (0.002 mmol)
were dissolved in dichloromethane (2 mL) under CO atmo-
sphere in a Pyrex Schlenk tube. The obtained solution was
introduced in the autoclave, previously placed under vac-
uum (0.1 mmHg), by a steel siphon. The reactor was pres-
surised with 10 atm CO and the mixture was stirred at
room temperature for 6 h. After removal of excess CO
(fume hood), the reaction mixture was diluted with pentane
(10 mL), filtered on celite and concentrated under vacuum.
The composition of the reaction mixture was determined
1
methane and analyzed by H-NMR.
3.4. X-ray structure determination of
[Rh2(OAc)2(Naft)4](BPh4)2 Æ 10CH3CN (7)
Crystals of compound (7) suitable for X-ray analysis
were grown by slow evaporation of a solution of the com-
plex in CH3CN. Data were collected on a Enraf Nonius
CAD 4 single-crystal diffractometer (Cu Ka radiation,
1
˚
k = 1.5418 A). Because of the presence of several crystalli-
by GLC, GC-MS and H-NMR analysis [15].
sation solvent molecules, the crystals were unstable and
decomposed rapidly in air at room temperature so the data
were collected at 173 K, after coating the crystal with per-
fluoropolyether oil. Details for the X-ray data collection
are collected in Table 3. The structure was solved by direct
methods with SHELXS-97 and refined against F2 with
SHELXS-97 [19], with anisotropic thermal parameters for
all non-hydrogen atoms. A correction for absorption was
made (maximum and minimum value for the transmission
coefficient was 1.000 and 0.636) [20]. Idealised geometries
were assigned to the hydrogen atoms.
3.3.2. Hydrosilylations
The hydrosilylation reaction was run in a Pyrex Carius
tube fitted with a Corning Rotaflo tap. Me2PhSiH
(0.31 mL, 2 mmol) and 1-hexyne (0.92 mL, 8 mmol) were
added, under argon atmosphere via syringe, to the rhodium
catalyst (0.002 mmol). The suspension was stirred at 90 ꢁC
for 5 h and then subjected to GLC analysis to determine
the conversion of the silane. The reaction mixture was fil-
tered on celite and concentrated under vacuum in order
to remove the excess alkyne. The isomeric composition of
1
the reaction products was determined by H-NMR analy-
4. Supporting information available
ses [18].
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic Data
Centre, CCDC No. 294808 for compound (7). These data
3.3.3. Silane alcoholysis
The silane alcoholysis reaction was run in a Schlenk
tube equipped with a magnetic stirring bar. The tube was

M. Basato et al. / Journal of Organometallic Chemistry 691 (2006) 3464–3471
3471
[6] (a) G. Pimblett, C.D. Garner, J. Chem. Soc. Dalton Trans. (1985)
1977;
can be obtained free of charge via www.ccdc.cam.ac.uk/
data_request/cif.
(b) G. Pimblett, C.D. Garner, W. Clegg, J. Chem. Soc. Dalton Trans.
(1986) 1257.
[7] (a) W.R. Tikkanen, E. Binamira-Soriaga, W.C. Kaska, P.C. Ford,
Inorg. Chem. 22 (1983) 1147;
Acknowledgement
(b) W.R. Tikkanen, E. Binamira-Soriaga, W.C. Kaska, P.C. Ford,
Inorg. Chem. 23 (1984) 141.
[8] (a) C.S. Campos-Fernandez, X. Ouyang, K.R. Dunbar, Inorg. Chem.
39 (2000) 2432;
Financial support from the University of Padova (post-
doctoral ‘‘assegno di ricerca’’ to C.T.) is gratefully
acknowledged.
(b) C.S. Campos-Fernandez, L.M. Thomson, J.R. Galan-Mascaros,
X. Ouyang, K.R. Dunbar, Inorg. Chem. 41 (2002) 1523.
[9] M. Basato, A. Biffis, G. Martinati, M. Zecca, P. Ganis, F. Benetollo,
L.A. Aronica, A.M. Caporusso, Organometallics 23 (2004) 1947.
[10] F.A. Cotton, K.R. Dunbar, C.T. Eagle, L.R. Falvello, S.-J. Kang,
A.C. Price, M.G. Verbruggen, Inorg. Chim. Acta 184 (1991) 35.
[11] F.A. Cotton, B.G. De Boer, M.D. La Prade, J.R. Pipal, D.A. Ucko,
Acta Cryst. B27 (1971) 1664.
[12] (a) K.M. Kadish, T.D. Phan, L. Giribabu, E. Van Caemelbecke, J.L.
Bear, Inorg. Chem. 42 (2003) 8663;
(b) J.F. Berry, F.A. Cotton, C. Lin, C.A. Murillo, J. Cluster Sci. 15
(2004) 531.
[13] See for example J.W. Faller, D.G. D’Alliessi, Organometallics 21
(2002) 1743.
[14] For another example of successful application of [Rh2(OAc)4] as
catalyst in hydrosilylation reactions, see T. Murai, F. Kimura, K.
Tsutsui, K. Hasegawa, S. Kato, Organometallics 17 (1998) 926.
[15] L.A. Aronica, S. Terreni, A.M. Caporusso, P. Salvadori, Eur. J. Org.
Chem. (2001) 4321.
[16] I. Matsuda, A. Ogiso, S. Sato, Y. Izumi, J. Am. Chem. Soc. 111
(1989) 2332.
[17] T.J. Kress, W.W. Paudler, J. Org. Chem. 31 (1966) 3055.
[18] I. Ojima, N. Clos, R.J. Donovan, P. Ingallina, Organometallics 9
(1990) 3127.
References
[1] (a) C.A. Merlic, A.L. Zechman, Synthesis (2003) 1137;
(b) M.P. Doyle, T. Ren, Prog. Inorg. Chem. 49 (2001) 113;
(c) M.P. Doyle, M.A. McKervey, T. Ye, Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds, Wiley, New York,
1998;
(d) R.D. Adams, F.A. Cotton, Catalysis by Di- and Polynuclear
Metal Cluster Complexes, Wiley, New York, 1998.
[2] (a) D.C. Forbes, M.C. McMills, Curr. Org. Chem. 5 (2001) 1091;
(b) M.P. Doyle, D.C. Forbes, Chem. Rev. 98 (1998) 911.
[3] (a) H.M.L. Davies, Ø. Loe, Synthesis (2004) 2595;
(b) H.M.L. Davies, R.E.J. Beckwith, Chem. Rev. 103 (2003) 2861;
(c) H.M.L. Davies, E.G. Antoulinakis, Org. React. 57 (2001) 1.
[4] (a) M.P. Doyle, K.G. High, V. Bagheri, R.J. Pieters, P.J. Lewis,
M.W. Pearson, J. Org. Chem. 55 (1990) 6082;
(b) M.P. Doyle, K.G. High, C.L. Nesloney, T.W. Clayton Jr., J. Lin,
Organometallics 10 (1991) 1225;
(c) M.P. Doyle, G.A. Devora, A.O. Nefedov, K.G. High, Organo-
metallics 11 (1992) 549;
(d) M.P. Doyle, M.S. Shanklin, Organometallics 12 (1993) 11;
(e) M.P. Doyle, M.S. Shanklin, Organometallics 13 (1994) 1081.
[5] (a) A. Biffis, E. Castello, M. Zecca, M. Basato, Tetrahedron 57
(2001) 10391;
[19] G.M. Sheldrick, SHELXS-97. Programs for Crystal Structure Analysis
(Release 97-2), Gottingen, Germany, (1997).
[20] (a) N. Walker, D. Stuart, Acta Crystallogr. A39 (1983) 158;
(b) F. Ugozzoli, Comput. Chem. 11 (1987) 109.
(b) A. Biffis, M. Zecca, M. Basato, Green Chem. 5 (2003) 170;
(c) A. Biffis, M. Braga, M. Basato, Adv. Synth. Catal. 5 (2004) 170.