Cationic carboxylato complexes of dirhodium(II) with oxo thioethers: Catalysts for silane alcoholysis under homogeneous and liquid-liquid biphasic conditions

Article abstract of DOI:10.1002/adsc.200700167

A set of cationic dirhodium(II) complexes with oxo thioethers was prepared and employed as catalysts for the silane alcoholysis reaction. The complexes were found to be highly active under homogeneous conditions, both in the absence and in the presence of a solvent, including coordinating solvents such as N,N-dimethylformamide ; the catalysts could be conveniently employed in concentrations as low as 0.01 mol%, and a maximum TON of 30000 was recorded after 24 h. The same catalysts were also employed under liquid-liquid biphasic conditions with an ionic liquid as the catalyst-containing phase: comparable catalytic activity was observed under these conditions, and the catalyst-containing phase could be recovered and recycled. A chiral cationic dirhodium(II) complex was also prepared in the frame of this work; kinetic resolution of a racemic alcohol was attempted with this catalyst, unfortunately without success.

Full text of DOI:10.1002/adsc.200700167

FULL PAPERS
DOI: 10.1002/adsc.200700167
Cationic Carboxylato Complexes of Dirhodium(II) with Oxo
Thioethers: Catalysts for Silane Alcoholysis under Homogeneous
and Liquid-Liquid Biphasic Conditions
Andrea Biffis,a,* Marino Basato,a Marianna Brichese,a Luca Ronconi,a
Cristina Tubaro,a Alessandra Zanella,a Claudia Graiff,b and Antonio Tiripicchiob
a
Dipartimento di Scienze Chimiche, Università di Padova, and ISTM-CNR, via Marzolo 1, 35131 Padova, Italy
Phone: (+39)-049-827-5216; fax: (+39)-049-827-5223; e-mail: andrea.biffis@unipd.it
Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica, Università di Parma, Parco Area
b
delle Scienze 17A, 43100 Parma, Italy
Received: April 2, 2007; Revised: July 17, 2007
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: A set of cationic dirhodium(II) complexes with an ionic liquid as the catalyst-containing phase:
with oxo thioethers was prepared and employed as comparable catalytic activity was observed under
catalysts for the silane alcoholysis reaction. The com- these conditions, and the catalyst-containing phase
plexes were found to be highly active under homoge- could be recovered and recycled. A chiral cationic
neous conditions, both in the absence and in the dirhodium(II) complex was also prepared in the
presence of a solvent, including coordinating solvents frame of this work; kinetic resolution of a racemic
such as N,N-dimethylformamide; the catalysts could alcohol was attempted with this catalyst, unfortu-
be conveniently employed in concentrations as low nately without success.
as 0.01 mol%, and a maximum TON of 30000 was
recorded after 24 h. The same catalysts were also
employed under liquid-liquid biphasic conditions Keywords: alcohols; catalysis; rhodium; silanes
Introduction
onstrated only in the case of hexamethyldisilazane
(HMDS).[2] A more general and even more atom eco-
Silylation undoubtedly plays a major role among the nomical silylation protocol is represented by silane al-
routinely employed methodologies for the protection coholysis [Eq. (3)], where H2 is the only by-product.
of hydroxy groups in organic synthesis.[1] The reaction Remarkably, the field of application of this reaction
is usually accomplished utilising a chlorosilane to in- has been recently extended far beyond protection/de-
troduce the silyl group [Eq.(1)]. However, this invari- protection strategies to embrace asymmetric synthe-
ably requires the presence of a base in order to neu- sis,[3] as well as the preparation and functionalisation
tralise the hydrochloric acid formed as by-product, of silicones.[4]
which implies the production of one equivalent of salt
A catalyst is needed in order for silane alcoholysis
per equivalent of product. Use of disilazanes together to proceed at a synthetically useful rate.[5] Over the
with a suitable catalyst in place of chlorosilanes is in years, quite a number of catalysts have been proposed
principle much more atom economical, ammonia in the literature, ranging from heterogeneous metal
being the only by-product [Eq. (2)], but the useful- catalysts[6] to organic compounds[7] and transition
ness of this synthetic strategy has been thus far dem- metal complexes.[8] The latter appear to be currently
Adv. Synth. Catal. 2007, 349, 2485 – 2492
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2485

Andrea Biffis et al.
FULL PAPERS
the most productive and versatile catalysts. However,
most of these catalysts are quite unreactive with elec-
tron-rich trialkylhydrosilanes, which incidentally yield
the most commonly employed protecting silyl ether
groups.[1] Furthermore, in many cases such catalysts
require strictly anhydrous reaction conditions and the
exclusion of atmospheric oxygen. Finally, they present
the problem of the separation of the catalyst from the
reaction products and of its recycling.
We have previously shown that these disadvantages
can be overcome using dirhodium(II) perfluorocar-
boxylate catalysts.[9] We have built on previous results
by Doyle et al., who demonstrated that dirhodium(II)
perfluorobutyrate is a robust catalyst which can be
used without special precautions at room temperature
in dichloromethane solution.[8n] We have developed
novel reaction protocols using longer chain dirhodi-
Scheme 1. Structure of the cationic dirhodium(II) complexes
with oxothioether ligands.
um(II) perfluorocarboxylates, which display higher ry results on our attempts to develop chiral variants
catalyst productivities and are easily removed from of cationic dirhodium(II) catalysts with oxothioether
the reaction mixture upon extraction into a perfluori- ligands will be reported.
nated phase or heterogenised by adsorption onto a
perfluorinated fluorous support. Both these strategies
allow to quantitatively recover and recycle the cata- Results and Discussion
lyst; furthermore, with the heterogenised catalyst it is
possible to run the reaction in a completely solvent- We started our work by selecting a suitable set of cat-
less fashion.
ionic dirhodium(II) catalysts. Besides the metal com-
Recently, we have started a related research pro- plexes which we disclosed in our previous communi-
gram aimed at the preparation of different kinds of cation on the subject,[10b] that is, complexes 1 and 2,
cationic dirhodium(II) complexes and at the evalua- we were interested in developing chiral complexes of
tion of their catalytic performance in technologically the same kind in order to test their enantioselectivity
relevant reactions involving silanes (silane alcoholysis, in the silane alcoholysis reaction; indeed, such reac-
hydrosilylations, silylformylations).[10] Our motivation tion is potentially useful for the kinetic resolution of
is that despite the recognised fundamental importance alcohols; it is worth noting that diastereoselective var-
of the electrophilicity of neutral dirhodium(II) com- iants of this reaction have been recently disclosed,
plexes in determining their activity and selectivity in which employ a chiral silane[3a] or a chiral alcohol+
catalytic reactions of this kind,[11] no study on the cat- chiral catalyst.[3b] We started by trying to link a chiral
alytic efficiency of related cationic complexes has functional group to the carboxy group of the oxo-
been apparently carried out. The investigated com- thioester ligands: in particular, we envisaged the sub-
plexes are all derived from simple dirhodium(II) ace- stitution of the methyl group by a (À)-menthyl one or
tate upon partial or complete substitution of the ace- the formation of an amide with the chiral amine (+)-
tate ligands with neutral ligands such as 1,8-naphthyri- 1-methylbenzylamine (Scheme 2, compounds 3 and 4)
dine[10a] or oxothioether molecules.[10b] The latter li- however, although it was as expected very easy to
gands give rise to complexes with a rather unusual bi- prepare the chiral ligands and the corresponding neu-
dentate O-S coordination of the oxo thioether, tral adducts 7 and 8 with dirhodium(II) acetate, the
highlighted in the general structure reported in synthesis of the cationic complexes proved difficult,
Scheme 1; preliminary tests indicate that complex 1 is owing to the apparent instability of the resulting com-
indeed a promising catalyst of the silylformylation of plexes and/or to ligand decomposition (hydrolysis of
1-hexyne with dimethylphenylsilane.[10b]
the ester/amide function under reaction conditions).
In this contribution we report on the catalytic activ- In fact, using our standard protocols for the removal
ity of cationic dirhodium(II) complexes with oxo- of the acetate ligands with HBF4 or Meerweinꢁs salt
+
À [10b]
thioether ligands in the silane alcoholysis reaction, OEt3 BF4 ,
complex mixtures of products were in-
under homogeneous (both with and without solvent) variably obtained. Apparently, steric crowding on the
as well as liquid-liquid biphasic conditions involving carboxylate function negatively affects the stability of
an ionic liquid as the catalyst-containing phase. The the resulting complex.
latter approach potentially allows the easy separation
Better results were obtained by introducing a chiral
and reuse of the liquid phase containing the catalyst. centre on the carbon atom a to the sulfur. Enantio-
This issue will be discussed as well. Finally, prelimina- merically pure ligands possessing a chiral centre in
2486
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2485 – 2492

Cationic Carboxylato Complexes of Dirhodium(II) with Oxo Thioethers
FULL PAPERS
Scheme 2. Chiral ligands 3–6.
Figure 1. ORTEP view of one of the two independent cat-
ionic complexes of 11 (cation B). Ellipsoids are drawn at
30% probability level.
that position (5 and 6) were easily prepared from
enantiomerically pure (S)-lactate following literature
procedures,[12] and readily gave the corresponding
neutral adducts 9 and 10 with dirhodium(II) acetate,
albeit in much lower yields than in the case of un-
branched ligands. The related cationic dirhodium(II)
complexes were still characterised by limited stability,
although the complexes appeared more stable than
their analogues with the chiral information on the car-
boxy function. The cationic complex with ligand 5
was isolated only in impure form; however, we were
able to purify and characterise complex 11, derived
from ligand 6.
Table 1. Selected bond distances [Š] and angles [deg] of
complex (11).
Cation A
Cation B
À
Rh1 Rh2
Rh1 O7
2.438(1)
2.284(7)
2.316(6)
2.274(3)
2.267(3)
1.208(14)
1.207(12)
1.816(12)
1.856(11)
1.532(18)
1.454(15)
81.32(20)
81.44(17)
2.435(1)
2.311(6)
2.323(6)
2.260(2)
2.286(3)
1.207(11)
1.217(12)
1.815(9)
1.848(11)
1.510(14)
1.486(16)
80.92(16)
80.33(20)
À
À
Rh2 O8
À
Rh1 S1
À
Rh2 S2
À
O7 C16
À
The far-infrared spectrum of complex 11 presents O8 C28
an intense band at 281 cmÀ1, corresponding to the Rh-
À
C14 S1
À
C26 S2
carboxylic oxygen stretching, in the range of similar
compounds present in the literature,[13] and a medium
À
C16 C14
À
[14]
C26 C28
band at 180 cmÀ1, corresponding to n
(Rh Rh). The
A
À
À
À
S1 Rh1 O7
last signal was absent in the parent neutral dirhodium
complex, since in that case the vibration is IR silent
because of the centrosymmetric structure of the com-
À
À
S2 Rh2 O8
1
pound. As expected, in the H NMR spectrum two
distinct set of signals for the bridging acetato ligands
The cationic complex has an approximate C2 sym-
are present with a 2/1 intensity ratio. Furthermore in metry and the coordination of each Rh atom is octa-
the 13C NMR spectrum the carboxylic carbon of the hedral with three acetate ligands and two chiral (R)-
thioether ligand is shifted well downfield from the PhCH2SCH(CH3)C(O)OEt ligands respectively bridg-
À
value observed in the neutral axial adduct, as the con- ing and chelating the two Rh atoms. The Rh O bond
sequence of the oxygen coordination.
distances involving the acetate ligands range from
À
Crystals of 11 suitable for X-ray structure determi- 2.024(6) to 2.048(6) Š and the Rh Rh ones of
nation were grown upon slow evaporation of a solu- 2.438(1) and 2.435(1) Š fall within the expected range
tion of the complex in acetone:n-hexane (1:1). In the for dirhodium(II) complexes. The two chiral ligands
crystals, two independent but very similar cations chelate through the sulfur and the carbonyl oxygen
À
À
[Rh
2(OAc)3[(R)-PhCH2SCH
(CH3)C(O)OEt)]2]+ (la- atoms, forming five membered rings. The S Rh O
G
À
belled A and B) and anions BF4 were found. Cations bite angles span from 80.92(16) to 81.44(17)8. The
A and B have slightly different conformations but the chelation rings adopt an envelope conformation. The
same configuration of all chiral centers. An ORTEP absolute configuration of chiral C14, C26, S1, S2
view of the cation (B) is shown in Figure 1, together atoms is R,R,S,S.
with the atomic labelling scheme. A selection of the
The cationic dirhodium(II) complexes 1, 2 and 11
most important bond distances and angles for A and were subsequently employed as catalysts in the silane
B is listed in Table 1.
alcoholysis reaction. In order to gain a first insight
Adv. Synth. Catal. 2007, 349, 2485 – 2492
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2487

Andrea Biffis et al.
FULL PAPERS
Table 2. Solventless triethylsilane alcoholysis catalysed by
cationic dirhodium(II) complexes with oxothioethers.[a]
Entry
Catalyst [mol%]
Rh2(OAc)4 (0.1)
1 (0.1)
1 (0.1)
Alcohol
Yield [%]
1
2
3
4
5
6
7
8
9
A
benzyl alcohol
benzyl alcohol
1-octanol[b]
benzyl alcohol
benzyl alcohol
1-octanol[b]
1-phenylethanol
benzyl alcohol
1-phenylethanol
2-octanol
10
100
100
100
100
96
100
93
61
2 (0.1)
11 (0.1)
11 (0.1)
11 (0.1)
11 (0.01)
11 (0.01)
11 (0.01)
10
>99
[a]
Reaction conditions: see the Experimental Section.
Catalyst only partially soluble.
[b]
into the reactivity of the complexes, we chose the re-
action between benzyl alcohol and triethylsilane as a
suitable test reaction. We initially employed reaction
conditions which already proved useful in our previ-
ous work with dirhodium(II) perfluorocarboxylates,
Figure 2. Conversion profile of the reaction with 1-phenyl-
ethanol and catalyst 11 (Table 2, entry 7)
running the reaction in a solventless fashion, with 0.1 catalyst solubility in reaction mixtures containing a
mol% catalyst, at 508C for 24 h. As it is apparent hydrophobic alcohol. An obvious way to cope with
from Table 2, entries 2, 4 and 5, all catalysts provided this problem is to add a conventional organic solvent
essentially quantitative yields of the silyl ether. In to the reaction mixture, and we selected 1,2-dichloro-
contrast, simple dirhodium(II) acetate gave only 10% ethane (DCE) as a suitable candidate. Addition of
reaction yield under the same reaction conditions DCE had no significant effect on catalyst reactivity;
(entry 1). Thus, it was proved that the presence of a yields fully comparable or even slightly superior to
positive charge on our catalysts greatly enhanced the tests run without solvent were reached under oth-
their reactivity in this reaction.
erwise identical reaction conditions (Table 3). Using
We then moved to test other alcohols in the same only 0.001 mol% catalyst, a remarkable maximum
reaction, such as 1-octanol and 1-phenylethanol. Al- TON of 30000 was reached in 24 h with benzyl alco-
though the yields with these substrates were almost hol (Table 3, entry 10). This represents the highest
quantitative as well, we observed that, in the case of TON ever reported for the silane alcoholysis reaction,
hydrophobic alcohols such as 1-octanol as the reac- significantly higher than that with our heterogenised
tant, the cationic catalysts were only partially soluble dirhodium(II) perflurocarboxylate system (maximum
in the mixture of the reagents. This observation could TON with benzyl alcohol 8300).[9c] The correponding
appear as a limitation of our synthetic protocol, but average TOF also ranks among the highest ever mea-
on the other hand it made it apparent that the soluble sured, significantly better results having to the best of
fraction of the catalyst was already sufficient to drive our knowledge been obtained only with Crabtreeꢁs
[8o]
the reaction to completion in the given reaction time. catalyst [IrH2(Solv)2(PPh3)2]SbF6 (see, for example,
G
We further corroborated this hypothesis by determin- Table 5 in ref.[8d,l]). Interestingly, preliminary results
ing the conversion profiles of the various reactions. In obtained in polar, coordinating aprotic solvents such
Figure 2, the conversion profile of the reaction with 1- as DMF indicate that the catalysts are highly active
phenylethanol and catalyst 11 is reported. It is appar- under these conditions as well (Table 3, entry 17). Use
ent that even with this comparatively less reactive al- of a polar, coordinating solvent is known to cause a
cohol the reaction reaches completion in just one decrease in the reactivity of most transition metal cat-
hour.
alysts for silane alcoholysis, with very few excep-
Consequently, we carried out some additional test tions;[6a,8a,b] in particular, partial catalyst deactivation
reactions with complex 11 using only 0.01 mol% cata- in even slightly coordinating solvents, evidently due
lyst; as reported in Table 2, entries 8–10 the yields to solvent competition with the reagents for coordina-
were still very good.
tion to the free apical positions of the complex, was
Having established the high reactivity of our cata- observed using dirhodium(II) perfluorocarboxylate
lysts in the silane alcoholysis reaction, we set out to catalysts.[9c] This finding is important in connection to
devise means for overcoming their shortcomings. The the possible application of these catalysts to the pro-
first important limitation to be tackled was the poor tection of hydroxy groups in hydrophilic polyols such
2488
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2485 – 2492

Cationic Carboxylato Complexes of Dirhodium(II) with Oxo Thioethers
FULL PAPERS
Table 3. Triethylsilane alcoholysis in DCE catalysed by cat-
(BMIM BF4) and n-methyloctylimidazolium tetra-
fluoroborate (MOIM BF4). All catalysts proved to be
very well soluble in these media, whereas reagents
and products built up a second organic phase when
contacted with the ionic liquid. However, the reaction
outcome observed with different alcohols and cata-
lysts under these conditions was not very satisfactory.
Only 20–30% yields were reached with 0.1 mol% cat-
alyst at 508C after 24 h. Furthermore, extensive de-
composition of the catalyst was observed in the
course of the reaction, with formation of rhodium col-
loids which then partly migrated into the reagent and
product phase. Consequently, high contents of leached
rhodium, usually amounting to one third to one half
of the total rhodium present, were measured in the
reagents and product phase.
We reasoned that the lower reactivity of the cata-
lysts in the liquid-liquid biphasic system could be
caused by the chosen ionic liquids. For example, it is
well-known that impurities present in commercial imi-
dazolium ionic liquids (water, halide anions, free 1-al-
kylimidazole species) may heavily affect the perfor-
mance of catalysts dissolved therein.[16] Indeed, by
switching to higher purity ionic liquids containing the
more stable hexafluorophosphate anion (BMIM PF6)
much better catalyst performances in the reaction be-
tween benzyl alcohol and triethylsilane could be ach-
ieved in 24 h at 508C with only 0.01 mol% of catalyst
11. The obtained reaction yield was, however, found
ionic dirhodium(II) complexes with oxothioethers.[a]
Entry
Catalyst [mol%]
Alcohol
Yield [%]
1
2
3
4
5
6
7
8
1 (0.1)
1 (0.01)
1 (0.1)
benzyl alcohol
benzyl alcohol
1-octanol
1-octanol
2-octanol
cyclohexanol
1-phenylethanol
benzyl alcohol
benzyl alcohol
benzyl alcohol
1-octanol
99
75
99
95
80
91
26
100
100
30
99
85
96
33
96
90
100
99
93
62
95
64
8
1 (0.01)
1 (0.01)
1 (0.01)
1 (0.01)
2 (0.1)
2 (0.01)
2 (0.001)
2 (0.01)
2 (0.01)
2 (0.01)
2 (0.01)
11 (0.1)
9
10
11
12
13
14
15
16
17[b]
18
19
20
21
22
23
2-octanol
cyclohexanol
1-phenylethanol
benzyl alcohol
benzyl alcohol
benzyl alcohol
1-octanol
1-octanol
2-octanol
cyclohexanol
1-phenylethanol
t-butyl alcohol
11( 0.01)
N
11 (0.1)
11 (0.1)
11 (0.01)
11 (0.01)
11 (0.01)
11 (0.01)
11 (0.01)
[a]
Reaction conditions: see the Experimental Section.
DMF as solvent.
[b]
as carbohydrates, which are not soluble in apolar or- to be significantly dependent on the source of the em-
ganic solvents.[6a,8a]
ployed ionic liquid: best results were obtained with a
No substantial difference in catalytic efficiency was home-made sample kindly made available to us by
observed between the various catalysts; catalyst (1) the group of Prof. Valeria Conte, University of Roma
was only slightly less active than the other two, which “Tor Vergata” (93% yield), whereas commercial high
exhibited fully comparable efficiency. The catalysts purity samples from Fluka and Solvent Innovation
were found to be active with primary as well as with (>98.5% and 99% purity, respectively) gave signifi-
secondary alcohols. Secondary alcohols often gave cantly lower yields (57 and 73%, respectively) under
slightly lower yields, but the observed differences in identical reaction conditions.
reactivity were anyway far less marked than with dir-
Using commercial BMIM PF6 from Solvent Innova-
hodium(II) perfluorocarboxylate catalysts.[8n,9] Poor tion, we carried out a screening of different alcohols
reactivity was instead observed with tertiary alcohols and silanes in order to assess the applicability of the
like tert-butyl alcohol (Table 3, entry 23). We also per- liquid-liquid biphasic system. As can be seen from
formed a preliminary evaluation of the enantiomeric Table 4, entries 1–8, most primary and secondary alco-
discrimination of 11 in the kinetic resolution of race- hols and silanes could be converted using 0.01 mol%
mic 1-phenylethanol. Thus, the alcohol was reacted catalyst of 11, with catalytic efficiencies in many in-
with 0.5 equivalents of triethylsilane under the homo- stances fully comparable to those observed in the ho-
geneous conditions reported in Table 3. Unfortunate- mogeneous system (Table 3), in spite of the fact that
ly, no enantioselection was observed in the reaction.
the employed ionic liquid was not the optimal one.
We also wanted to check the possibility of catalyst Only the reactivity of 2-octanol was found to be unex-
recovery and recycling, and envisaged to do so by pectedly low in comparison to other tested secondary
switching from homogeneous to liquid-liquid biphasic alcohols, although a somewhat lower reactivity was
conditions, using a second liquid phase in which only observed with this alcohol also using the homogene-
the catalyst was soluble. Given the cationic nature of ous system. Considering the silane screening,
our catalysts, ionic liquids offered themselves as a Me2PhSiH (Table 4, entry 7) displayed a reactivity
very natural choice.[15] We initially employed commer- similar to Et3SiH, while sterically encumbered silanes
cial ionic liquid media from Fluka (>97% purity) such as t-BuMe2SiH (Table 4, entry 8) gave unsatisfac-
such as n-butylmethylimidazolium tetrafluoborate tory yields; these results are in accordance with those
Adv. Synth. Catal. 2007, 349, 2485 – 2492
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2489

Andrea Biffis et al.
FULL PAPERS
Table 4. Silane alcoholysis in ionic liquids catalysed by cat-
ionic complex 11.[a]
positive charge on dirhodium(II) complexes can be
considered as a viable strategy for rendering them ef-
ficient electrophilic catalysts in respect to more con-
ventional ligand substitution with electron-withdraw-
ing groups. The peculiar nature of the employed thio-
ether ligands, that are expected to act as hemilabile li-
gands with the carbonyl group being displaced by the
silane in solution, may also positively contribute to
the catalytic efficiency of these catalysts. Use of these
catalysts under liquid-liquid biphasic conditions
allows a convenient separation of the catalyst from
the products; on the other hand, partial catalyst de-
composition under the employed reaction conditions
prevents at present its efficient recycling. We are cur-
rently aiming at developing more robust cationic dir-
hodium(II) complexes and optimised reaction condi-
tions in order to overcome this limitation.
Entry
Alcohol
Silane
Yield [%]
1
2
3
4
5
6
7
8
benzyl alcohol
2-phenylethanol
1-phenylethanol
1-octanol
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Me2PhSiH
t-BuMe2SiH
Et3SiH
73
99
45
50
14
99
65
17
57
20
2-octanol
cyclohexanol
benzyl alcohol
benzyl alcohol
benzyl alcohol
benzyl alcohol
9[b]
10[c]
Et3SiH
[a]
Reaction conditions: see the Experimental Section.
Recycle of entry 1.
Recycle of entry 9.
[b]
[c]
obtained with dirhodium(II) perfluorocarboxylate cat-
alysts.[9c]
Experimental Section
We also preliminarly evaluated the ease of recovery
and recycling of the catalyst. After the first catalytic Synthesis of the Neutral Complex {Rh2(OAc)4[(R)-
G
run under the conditions of Table 4, entry 1, the ionic PhCH2SCH(CH3)C(O)OEt]2} (10)
E
liquid phase containing the catalyst was successfully
Rh2
ACHTREUNG
separated by simple decantation and extraction with
diethyl ether. ICP-AAS analysis of the Rh content in
the organic phase demonstrated that 90% of rhodium
was recovered in this case. The catalyst-containing
phase was recycled twice (Table 4, entries 9 and 10)
but a significant decrease in the reaction yield, most
notable after the second recycle, was apparent. This is
most probably the consequence of partial catalyst de-
composition, as qualitatively confirmed by the pro-
gressive darkening of the catalyst-containing phase.
Thus, although the use of commercial BMIM PF6 ena-
bles an efficient recovery of the precious metal from
the reaction mixture, it does not allow us at this stage
to perform repeated recycles of the same catalyst
batch. On the other hand, we are confident that be-
sides the efficiency of the catalytic system also the ef-
fectivness of the recovery/recycling procedure can be
further improved to a significant extent through opti-
misation of the employed high-purity ionic liquid.
(25 mL). (R)-PhCH2SCH
AHCTREUNG
16.8 mmol) was then added, and the reaction mixture was
stirred for 3 h at room temperature, evaporated to a small
volume under reduced pressure, and treated with diethyl
ether to afford the product as a purple solid, which was fil-
tered and dried under vacuum. Yield: 25%. Anal. calcd. for
C32H44O12Rh2S2 (M=890.6): C 43.55, H 4.98, S 7.20; found:
1
C 43.69, H 5.27, S 7.14%; H NMR (CDCl3): d=1.25 (t, 3H,
CH3CH2), 1.64 (d, 3H, CH3CH), 1.88 (s, 6H, CH3COOÀ),
3.94 (q, 1H, CH), 4.04 (q, 2H, CH3CH2), 4.32 (AB system,
2H, CH2Ph), 7.25–7.55 (m, 5H, Ph); 13C NMR (CDCl3): d=
14.0 (CH3CH2), 15.5 (CH3CH), 23.8 (CH3COOÀ), 36.2
(CH2Ph), 42.7 (CH), 61.2 (CH3CH2), 127.2–130.5 (Ph), 171.9
[C(O)OEt], 191.4 (CH3COOÀ); FT-IR (KBr): n=3056–
2882, 1734, 1589, 1435, 1235, 381, 351, 329 cmÀ1.
Synthesis of the Cationic Complex {Rh2
PhCH2SCH(Me)C(O)OEt]2}BF4 (11)
N
A
N
0.23 mmol) was dissolved in dichloromethane (25 mL) and
42 mL of a 54% w/w solution of HBF4 in diethyl ether
(0.3 mmol) were added to the resulting solution. The reac-
tion mixture was stirred for 3 h, during which time the color
changed from purple to green; evaporation to a small
volume under reduced pressure and treatment with diethyl
ether afforded a green compound, which was filtered off
and dried under vacuum. Yield: 84%. Anal. calcd. for
C30H41BF4O10Rh2S2 (M=918.4): C 39.24, H 4.50, S 6.98;
Conclusions
In conclusion, we have demonstrated that cationic dir-
hodium(II) complexes with oxothioether ligands are
highly efficient catalysts for silane alcoholysis. In com-
parison with dirhodium(II) perfluorocarboxylates, the
catalysts exhibit higher reactivity, in particular with
secondary alcohols. Furthermore, the catalysts appear
to work well under solventless conditions, in the pres-
ence of organic solvents (including polar, coordinating
ones), and also under liquid-liquid biphasic condi-
1
found: C 39.67, H 4.88, S 6.48%; H NMR (CDCl3): d=1.42
(t, 6H, CH3CH2), 1.54 (d, 6H, CH3CH), 1.96 (s, 6H,
CH3COOÀ), 2.47 (s, 3H, CH3COOÀ), 3.80 (AB system, 4H,
CH2Ph), 4.04 (q, 2H, CH), 4.64 (q, 4H, CH3CH2), 7.35–7.55
(m, 10H, Ph); 13C NMR (CDCl3): d=14.0 (CH3CH2), 17.5
tions. These findings clearly point out that placing a (CH3CH), 22.6 and 24.5 (CH3COOÀ), 38.6 (CH2Ph), 51.2
2490
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2485 – 2492

Cationic Carboxylato Complexes of Dirhodium(II) with Oxo Thioethers
FULL PAPERS
(CH), 66.2 (CH3CH2), 129.1–131.2 (Ph), 181.5 [C(O)OEt],
188.3 and 193.3 (CH3COOÀ); FT-IR (KBr): n=3061–2850,
1657, 1570, 1447, 1254, 1082, 1003, 376, 359, 322, 281,
173 cmÀ1.
(10 mmol) were added and the resulting mixture was heated
with stirring to 508C in a thermostated water bath. After ad-
dition of triethylsilane (15 mmol, 1.5 equivs.) the mixture
was further stirred at 508C for 24 h. After this period, the
organic phase was separated and the ionic liquid phase was
extracted with diethyl ether (3ꢂ5 mL). The organic phases
were combined and the solvent was removed under reduced
pressure. The residue was analysed for the determination of
the yield as described above. The extent of rhodium leach-
ing was determined by digestion in 6 mL hot aqua regia of
an 0.5 mL aliquot of the organic phase in the reaction mix-
ture. The resulting solution was diluted to 100 mL with
water and its rhodium content was determined by ICP-AAS.
Solventless Catalytic Tests; General Procedure
The reaction was run in a Schlenk tube equipped with a
magnetic stirring bar. The tube was charged with the rhodi-
um catalyst (0.01 mmol, 0.1 mol%), evacuated and filled
with argon. The alcohol (10 mmol) was then added and the
resulting solution was heated with stirring to 508C in a ther-
mostated water bath. After addition of triethylsilane
(15 mmol, 1.5 equivs.) the solution was further stirred at
508C for 24 h. After this period, the yield was determined
spectroscopically or by gas chromatography. In the case of
benzyl alcohol and 1-phenylethanol, yields were determined
by NMR analysis of a drop of the reaction mixture dissolved
in CDCl3. In the case of the other alcohols, 0.1 mL samples
of reaction mixture were instead withdrawn, diluted with
1 mL dichloromethane and analysed by gas chromatography
(on a Shimadzu GC-8 A instrument equipped with a 25 m
OV-1701 capillary column: 1008C isotherm for 60 s followed
by heating at 168CminÀ1 to 2008C). The GC system was
previously calibrated by determining the retention times
and the response factors of the alcohol reagents and of the
silylated products. Experiments with 0.01 mol% catalyst
were run as described above with 0.002 mol catalyst,
20 mmol alcohol and 30 mmol triethylsilane.
Kinetic Resolution of Alcohols
The reaction was run in a Schlenk tube equipped with a
magnetic stirring bar. The tube was charged with the rhodi-
um catalyst (11) (7 mg, 0.008 mmol), evacuated and filled
with argon. 1-Phenylethanol (0.9 mL, 8 mmol, 1 equiv.) and
triethylsilane (0.6 mL, 4 mmol, 0.5 equivs.) were then added
and the resulting solution was stirred at room temperature.
10 mL samples of the reaction mixture were withdrawn at
regular intervals, diluted with 0.6 mL CDCl3 and analysed
by NMR, until the conversion of the silane was complete.
The enantiomeric excess was determined by 1H NMR in
CDCl3 solution after addition of 0.2 equivs. of the chiral
shift reagent tris[3-[(heptafluoropropyl)hydroxymethylene]-
(À)-camphorato]praseodymium
A
Supporting Information Available
Catalytic Tests in Homogeneous Solution; General
Procedure
Synthetic procedures for the preparation of ligands and
complexes, summary of X-ray crystallographic data for
The reaction was run in a Schlenk tube equipped with a
magnetic stirring bar. The tube was charged with the re-
quired amount of rhodium catalyst (0.01 mmol, 0.1 mol%),
evacuated and filled with argon. The solvent (1,2-dichloro-
ethane or dimethylformamide, 6.5 mL for experiments with
0.1 mol% of catalyst, 0.7 mL with 0.01 mol% catalyst,
0.5 mL with 0.001 mol% of catalyst) and the alcohol
(10 mmol) were added and the resulting solution was heated
with stirring to 508C in a thermostated water bath. After ad-
dition of triethylsilane (15 mmol, 1.5 equivs.) the mixture
was further stirred at 508C for 24 h. After this period, the
solvent was removed under reduced pressure and the resi-
due was analysed by NMR (for benzyl alcohol, 1-phenyl-
ethanol and tert-butyl alcohol) or by GC (for the 1-octanol,
2-octanol and cyclohexanol) to determine the yield, as de-
scribed above. Experiments with 0.01 mol% catalyst were
run as described above with 0.002 mol catalyst, 1.6 mL sol-
vent, 20 mmol alcohol and 30 mmol triethylsilane. Experi-
ments with 0.001 mol% catalyst were run as described
above with 0.0001 mol catalyst (0.1 mL of a 1m solution in
the solvent of choice), 0.5 mL solvent, 10 mmol alcohol and
15 mmol triethylsilane.
the
structure
determination
of
[Rh2(OAc)3((R)-
G
PhCH2SCH(Me)C(O)OEt)2]BF4.
X-ray crystallographic files are available in CIF format
(CCDC 653126). These data can be obtained free of charge
from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data request/cif.
Acknowledgements
Financial support from the University of Padova (postdoc-
toral stipend to C.T.) is gratefully acknowledged. We wish to
thank Prof. Valeria Conte, Università di Roma “Tor Verga-
ta”, Italy, for helpful discussions and for samples of high-
purity ionic liquids. We also wish to thank Dr. Giulia Zan-
marchi, Università di Padova, for the ICP-AAS measure-
ments.
References
Catalytic Tests in Liquid-Liquid Biphasic Systems;
General Procedure
[1] T. W. Greene, P. G. Wuts, Protective Groups in Organic
Synthesis, 2nd edn., Wiley, New York, 1991.
The reaction was run in a Schlenk tube equipped with a
magnetic stirring bar. The tube was charged with the rhodi-
um catalyst (0.001 mmol, 0.01 mol%), evacuated and filled
with argon. The ionic liquid (0.7 mL) and the alcohol
[2] Recent examples: a) H. Firouzabadi, N. Iranpoor, S.
Sobhani, S. Ghassamipour, Synthesis 2005, 595 and ref-
erences cited therein; b) B. Kharimi, B. Golshani, J.
Org. Chem. 2000, 65, 7228.
Adv. Synth. Catal. 2007, 349, 2485 – 2492
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2491

Andrea Biffis et al.
FULL PAPERS
[3] a) S. Rendler, G. Auer, M. Oestreich, Angew. Chem.
Int. Ed. 2005, 44, 7620; b) D. R. Schmidt, S. J. OꢁMalley,
J. L. Leighton, J. Am. Chem. Soc. 2003, 125, 1190.
[4] a) B. P. S. Chauhan, T. E. Ready, Z. Al-Badri, P. Boud-
jouk, Organometallics 2001, 20, 2725; b) Y. Li, Y. Ka-
wakami, Macromolecules 1999, 32, 3540; c) R. Zhang,
J. E. Mark, A. R. Pinhas, Macromolecules 2000, 33,
3508, and references cited therein.
1992, 33, 5041; n) M. P. Doyle, K. G. High, V. Bagheri,
R. J. Pieters, P. J. Lewis, M. W. Pearson, J. Org. Chem.
1990, 55, 6082; o) X.-L. Luo, R. H. Crabtree, J. Am.
Chem. Soc. 1989, 111, 2527; p) W. Caseri, P. S. Pregosin,
Organometallics 1988, 7, 1373.
[9] a) A. Biffis, E. Castello, M. Zecca, M. Basato, Tetrahe-
dron 2001, 57, 10391; b) A. Biffis, M, Zecca, M,
Basato, Green Chem. 2003, 5, 170; c) A. Biffis, M.
Braga, M. Basato, Adv. Synth. Catal. 2004, 5, 170.
[10] a) M. Basato, A. Biffis, G. Martinati, C. Tubaro, C.
Graiff, A. Tiripicchio, L. A. Aronica, A. M. Caporusso,
J. Organomet. Chem. 2006, 691, 3464; b) M. Basato, A.
Biffis, G. Martinati, M. Zecca, P. Ganis, F. Benetollo,
L. A. Aronica, A. M. Caporusso, Organometallics 2004,
23, 1947.
[11] a) M. P. Doyle, K. G. High, C. L. Nesloney, T. W. Clay-
ton Jr. , J. Lin, Organometallics 1991, 10, 1225; b) M. P.
Doyle, G. A. Devora, A. O. Nefedov, K. G. High, Orga-
nometallics 1992, 11, 549; c) M. P. Doyle, M. S. Shan-
klin, Organometallics 1993, 12, 11; d) M. P. Doyle, M. S.
Shanklin, Organometallics 1994, 13, 1081.
[5] E. Lukevics, M. Dzintara, J. Organomet. Chem. 1985,
295, 265.
[6] a) M.-K. Chung, G. Orlova, J. D. Goddard, M. Schlaf,
R. Harris, T. J. Beverifge, G. White, F. R. Hallett, J.
Am. Chem. Soc. 2002, 124, 10508; b) K. Yamamoto, M.
Takemae, Bull. Chem. Soc. Jpn. 1989, 62, 2111; c) L. H.
Sommer, J. E. Lyons, J. Am. Chem. Soc. 1969, 91, 7061.
[7] a) F. Le Bideau, T. Coradin, J. Henique, E. Samuel,
Chem. Commun. 2001, 1408; b) J. M. Blackwell, K. L.
Foster, V. H. Beck, W. E. Piers, J. Org. Chem. 1999, 64,
4887; c) Y. Tanabe, H. Okumura, A. Maeda, M. Mura-
kami, Tetrahedron Lett. 1994, 35, 8413.
[8] Selected examples: a) M.-K. Chung, M. Schlaf, J. Am.
Chem. Soc. 2005, 127, 18085; b) H. Ito, K. Takagi, T.
Miyahara, M. Sawamura, Org. Lett. 2005, 7, 3001; c) H.
Ito, A. Watanabe, M. Sawamura, Org. Lett. 2005, 7,
1869; d) R. L. Miller, S. V. Maifeld, D. Lee, Org. Lett.
2004, 6, 2773; e) L. D. Field, B. A. Messerle, M. Rehr,
L. P. Soler, T. W. Hambley, Organometallics 2003, 22,
2387; f) S. V. Maifeld, R. L. Miller, D. Lee, Tetrahedron
Lett. 2002, 43, 6363; g) M.-K. Chung, G. Ferguson, V.
Robertson, M. Schlaf, Can. J. Chem. 2001, 79, 949; h) S.
Chang, E. Scharrer, M. Brookhart, J. Mol. Catal. A:
Chem. 1998, 130, 107; i) X. Wang, W. W. Ellis, B. Bos-
nich, Chem. Commun. 1996, 2561; j) C. Lorenz, U.
Schubert, Chem. Ber. 1995, 128, 1267; k) B. T. Gregg,
A. R. Cutler, Organometallics 1994, 13, 1039; l) T. C.
Bedard, J. Y. Corey, J. Organomet. Chem. 1992, 428,
315; m) D. H. Barton, M. J. Kelly, Tetrahedron Lett.
[12] U. Burkard, F. Effenberger, Chem. Ber. 1986, 119, 1594.
[13] a) W. L. Driessen, W. L. Groeneveld, Rec. Trav. Chim.
1969, 88, 977; b) D. M. Adams, Metal-Ligand and Relat-
ed Vibrations, Arnold, London, 1967.
[14] D. F. Shriver, C. B. Cooper, III, Adv. Infrared Raman
Spectrosc. 1980, 6, 127.
[15] For leading references on ionic liquids, see a) R. D.
Rogers, K. R. Seddon (Eds.), Ionic Liquids IIIB: Fun-
damentals, Process, Challenges, and Opportunities, ACS
Symposium Series, ACS, Washington DC, 2005; b) P.
Wasserscheid, T. Welton (Eds.), Ionic Liquids in Syn-
thesis Wiley-VCH, Weinheim, 2003.
[16] See, for example: A. Stark, M. Ajam, M. Green, H. G.
Raubenheimer, A. Ranwell, B. Ondruschka, Adv.
Synth. Catal. 2006, 348, 1934, and references cited
therein.
2492
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2485 – 2492