Solventless silane alcoholysis catalyzed by recoverable dirhodium(II) perfluorocarboxylates

Article abstract of DOI:10.1002/adsc.200303247

We have developed a novel reaction protocol for the highly efficient and sustainable catalysis of the silane alcoholysis reaction. The catalysts of choice are dirhodium(II) perfluorocarboxylates bearing long perfluoroalkyl chains, which are easily prepare

Full text of DOI:10.1002/adsc.200303247

FULL PAPERS
Solventless Silane Alcoholysis Catalyzed by Recoverable
Dirhodium(II) Perfluorocarboxylates
Andrea Biffis,* Mirko Braga, Marino Basato
¡
Dipartimento di Scienze Chimiche, Universita di Padova, via Marzolo 1, 35131 Padova, Italy
Fax: ( ‡ 39)-049-8275223, e-mail andrea.biffis@unipd.it
Received: September 18, 2003; Accepted: February 27, 2004
Abstract: We have developed a novel reaction proto- plished through a novel strategy based on fluorous
col for the highly efficient and sustainable catalysis of chemistry. Indeed, perfluorinated catalysts of this
the silane alcoholysis reaction. The catalysts of choice kind are easily adsorbed on silica which has been
are dirhodium(II) perfluorocarboxylates bearing long previously functionalized at its surface with perfluoro-
perfluoroalkyl chains, which are easily prepared in alkyl chains. Use of such supported catalysts (bonded
one step from commercial precursors. Under opti- fluorous phase catalysts) allows an easy and almost
mized reaction conditions, these catalysts exhibit up complete catalyst separation and recycling with
to about 50 times higher activity and 100 times higher improved catalytic performance.
productivity than analogous dirhodium(II) complexes
described in the prior art. Furthermore, the reaction
can be run in a completely solventless fashion. Finally, Keywords: catalysis; fluorous chemistry; green chem-
heterogenization of these catalysts has been accom- istry; rhodium; silanes
Introduction
one equivalent of salt per equivalent of product. On the
contrary, with silane alcoholysis H2 is the only by-
The recent history of chemistry has witnessed the product. Quite remarkably, the field of application of
development of a myriad of homogeneous transition the silane alcoholysis reaction has been recently ex-
metal catalysts for the most diverse reactions.[1] Often tended to the preparation and functionalization of
enough, such catalysts exhibit very positive features in silicones.[6]
terms of activity and selectivity under mild conditions,
but in most cases their technological application is
hampered by the need for their efficient recovery and
recycling. In fact, incomplete catalyst recovery and
recycling results in losses of (precious) metal, losses of
(expensive) ligands, and pollution of the reaction
products. Consequently, it is not surprising that quite a A catalyst is needed in order for silane alcoholysis to
number of strategies have been developed over the last proceed at a synthetically useful rate.[7] Over the years,
three decades to overcome this problem,[2] and that this quite a number of catalysts have been proposed in the
kind of research has enjoyed a further increase of literature, ranging from heterogeneous metal catalysts[8]
interest over the last few years, particularly due to the to organic compounds[9] and transition metal com-
growing concern for the cleanness and efficiency of plexes.[10] The latter, most notably those of ruthenium,
chemical processes which generally goes under the rhodium, iridium, manganese and platinum appear to be
name of ™green chemistry∫.[3]
currently the most productive and versatile catalysts.
Recently, we have started a project aimed at the However, they usually tolerate only a limited range of
development of recoverable catalysts for silane alcohol- reaction solvents, i.e., non-polar, non-coordinating sol-
ysis, Eq. (1).[4] This reaction is a practical alternative to vents such as benzene or dichloromethane, which are of
the classical silylation protocol which is routinely concern for the environment and health. Furthermore,
employed in organic synthesis for the protection of in most cases they require strictly anhydrous reaction
hydroxy groups.[5] The reaction is usually accomplished conditions and the exclusion of atmospheric oxygen.
utilising a chlorosilane to introduce the silyl group, Finally, they present the problem of the separation of the
Eq. (2). However, this invariably requires the presence catalyst from the reaction products and of its recycling.
of a base in order to neutralize the hydrochloric acid In this contribution, we describe in full length our results
formed as by-product, which implies the production of on the discovery of a solventless protocol for efficient
Adv. Synth. Catal. 2004, 346, 451 458
DOI: 10.1002/adsc.200303247
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
451

Andrea Biffis et al.
FULL PAPERS
silane alcoholysis using dirhodium(II) perfluorocarbox-
ylate catalysts, as well as on the development of novel
strategies for complete catalyst recovery and recycling.
Results and Discussion
The application of a dirhodium(II) perfluorocarboxy-
late as catalyst for silane alcoholysis was originally
reported by Doyle et al.,[10i] who utilized 1 mol %
dirhodium(II) perfluorobutyrate in dichloromethane
at room temperature. This rhodium catalyst does not
Figure 1.
apparently require the strict exclusion of water and solvent at the end of the reaction. This preliminary
oxygen, and its selectivity for the alcoholysis reaction investigation was performed using the triethylsilane
product is very good even with unsaturated alcohols, alcoholysis with 1-octanol as a model reaction. Relevant
which easily undergo metal-catalyzed hydrogenation results are reported in Table 1.
along with silane alcoholysis. However, the amount of
In the absence of other solutes, the investigated
catalyst which has to be employed is intolerably high for catalysts are only soluble in organic solvents able to
commercial applications. Therefore, it is mandatory to coordinate to the free apical positions of the complex,
increase the productivity of this catalytic system, for most notably in solvents containing oxygen donor atoms
example, by efficiently recovering and recycling the such as ethers. These solvents, however, deactivate the
catalyst. In a previous publication,[4a] we have intro- catalyst, since they block its active site (Table 1,
duced dirhodium(II) perfluorocarboxylates bearing Entry 2). On the other hand, the solubility of the
long perfluoroalkyl chains (i.e., from 7 to 13carbon
catalyst in non-coordinating solvents such as dichloro-
atoms long, Figure 1) as catalysts for this reaction. Such methane is promoted by the presence of a primary
highly fluorous dirhodium(II) complexes can be easily alcohol substrate such as 1-octanol. In fact, the alcohol is
prepared in one step from common commercial pre- able to interact with the free apical positions of the
cursors and are air-stable, easy to handle solid com- catalyst as well. The resulting complex adduct is less
pounds which are preferentially soluble in fluorinated fluorous, hence more prone to be solubilized by a
solvents.[4,11] Consequently, we have employed them as conventional organic solvent.
silane alcoholysis catalysts in a dichloromethane/Fluo-
The catalyst could be easily and quantitatively recov-
¾
rinert FC-77 fluorous biphasic system,[12] which ensures ered from the dichloromethane solution by extraction
¾
the easy recovery and recycling of the catalyst. How- with a fluorous solvent such as Fluorinert FC-77.
ever, their catalytic activity in the fluorous biphasic However, the observed catalytic activity was only
system was found to be lower than in the related slightly higher than in the biphasic system and still
homogeneous system, which was attributed to mass- significantly lower than that of the benchmark literature
transport limitations between the two liquid phases.[4a]
catalyst dirhodium(II) perfluorobutyrate under compa-
In the search for a more efficient way to run the rable reaction conditions (compare Entries 1, 3and 4 in
reaction while maintaining the possibility of an easy Table 1). We attribute the lower catalytic activity of our
recovery and recycling of the catalyst, we have been catalysts under these conditions to some degree of
looking for a reaction medium in which the reaction aggregation of the fluorous catalyst molecules in the
could be run in a monophasic fashion at room temper- organic phase. Aggregation of dirhodium(II) perfluor-
ature, and from which the catalyst could still be easily ocarboxylates due to intermolecular interaction be-
separated by liquid-liquid extraction with a fluorous tween the oxygen atoms of a carboxylate group on one
Table 1. Triethylsilane alcoholysis with 1-octanol catalyzed by Rh2(pft)4 using different solvent systems.
Entry
Catalyst
Solvent
Yield[a] [%]
TON
68
TOF [hÀ1]
11
1[b]
2
3Rh
4
Rh2(pft)4
Rh2(pft)4
2(pft)4
Dichloromethane/FC-77
THF
Dichloromethane
Dichloromethane
68
42323
74
74
96
12
32
[c]
Rh2(pfb)4
96(3h)
Reaction conditions: 1 mmol of each reagent, 0.01 mmol (1 mol %) catalyst, 5 mL solvent, room temperature.
[a]
GC yield after 6 h.
[b]
Data from Ref.[4a]
[c]
Dirhodium(II) perfluorobutyrate catalyst: literature data from Ref.[10i]
452
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 451 458

Solventless Silane Alcoholysis
FULL PAPERS
complex molecule with the free apical position of properly long reaction times good yields can be reached
another complex molecule has been described in the also with this substrate, in contrast to the previously
solid state[13] and there is indirect evidence suggesting its described fluorous biphasic system,[4a] in which mass-
presence also in solution.[14] Such an aggregation could transport limitations virtually stopped the reaction of
be further promoted by the presence of long perfluoro- secondary alcohols at about 20% yield. Nevertheless,
alkyl chains tending to segregate from the organic the significant difference in reaction rate between
medium, thereby leading to clustering of catalyst primary and secondary alcohols can be exploited in
molecules and consequently decreased catalytic effi- competitive silane alcoholysis experiments performed
ciency.
in a deficit of silane, leading to preferential silylation of
In view of the high solubilizing power of alcohols such the primary alcohol function (Entry 7).
as 1-octanol, we finally resolved to try the reaction
After reaction completion, the catalyst finds itself
without any organic solvent. Indeed, coordination of dissolved in the reaction product, which is a much worse
alcohol molecules to the free apical positions of the ligand for the rhodium complex than the alcohol
catalyst renders it soluble in the neat alcohol as well as in substrate. As a consequence, the catalyst does not
a 1:1 mixture of 1-octanol and triethylsilane. Actually, precipitate out of the solution, but it readily migrates
the solubility of the complex in such a mixture is so good into a separated fluorous phase as soon as the product
that it is extracted from a fluorous phase by treatment phase comes in contact with it. The fluorous phase
with it, the extent of extraction depending on the containing the catalyst can be subsequently contacted
catalyst and fluorous solvent employed: for example, with a second batch of liquid reagents mixture, which
¾
Rh2(pfd)4 is completely extracted from a Fluorinert
extracts it back, thereby starting the reaction again.
FC-77 solution, whereas Rh2(pft)4 migrates only parti- Thus, under these reaction conditions, dirhodium(II)
ally. The results reported in Table 2 show that the perfluorocarboxylates represent a further example of
activity of the highly fluorous catalysts in the solventless ™smart∫ homogeneous catalysts, which readily and
system actually equals that of the benchmark literature spontaneously separate from the product phase at the
catalyst in methylene chloride as reported by Doyle.[10i] end of the reaction.[15] Unfortunately, the recovery of the
However, the productivity of the catalyst can be rhodium catalysts using this strategy is not complete.
significantly increased, as apparent from the reported This is partly due to incomplete extraction of the catalyst
turnover numbers, and no hazardous organic solvent is from the product phase and partly to the long reaction
needed for the reaction. No significant difference in time needed for the reaction to reach completion, which
reactivity is observed for catalysts with differently long causes partial catalyst decomposition. Best results
¾
fluorous chains (Entries 1 4), in accordance with the obtained with Rh2(pfd)4 and Fluorinert FC-77 as the
idea that changing the chain length does not influence extractant account for 60% catalyst recovery after the
the reactivity of the active site. Secondary alcohols such first cycle. Nevertheless, it is important to remark that
as 2-octanol react much more slowly (Entry 6), as was the recovered catalyst retains the initial catalytic activity
the case with Doyle×s catalyst.[10i] However, by allowing (Entry 5, Table 2).
Table 2. Triethylsilane alcoholysis catalyzed by dirhodium(II) perfluorocarboxylates under solventless conditions at room
temperature.
Entry
Catalyst
Mol % catalyst
Alcohol
Yield[a] [%]
TON
TOF [hÀ1]
1
2
Rh2(pfo)4
Rh2(pfd)4
2(pfdo)4
Rh2(pft)4
Rh2(pfd)4
Rh2(pfd)4
Rh2(pfd)4
0.1
0.1
0.1
0.1
0.060
0.1
0.1
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
2-octanol
1-octanol
2-octanol
1-octanol
2-octanol
73
75
60
69
730
750
600
690
870
240
710
140
96
30
31
25
29
36
10
30
6
3Rh
4
5[b]
6
52
24[c]
71
7[d]
14
96(3h)
94(14 h)
[e]
8
9
Rh2(pfb)4
Rh2(pfb)4
1
1
32
[e]
94
7
Reaction conditions: see the Experimental Section.
[a]
GC yield after 24 h.
Recycle of Entry 2.
Yield is 66% after 5 days.
[b]
[c]
[d]
Competitive reaction experiment.
[e]
Dirhodium(II) perfluorobutyrate catalyst: literature data from Ref.[10i]
Adv. Synth. Catal. 2004, 346, 451 458
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
453

Andrea Biffis et al.
FULL PAPERS
In order to overcome the problems associated with reversibly released from the support by applying CO2
partial catalyst decomposition, we tried to reduce the pressure.[21] In our case and in the case of Bannwarth,
reaction time by running the reaction at a slightly higher however, the catalyst remains on the solid support for
temperature (508C). We were delighted to observe that the whole duration of the experiment, i.e., it is truly
the catalytic activity increased considerably at this ™heterogenized∫ on the support.
temperature, reaching values one order of magnitude
We have prepared our silica support by derivatization
higher than at room temperature (Table 3). High yields of silica gel following a procedure originally reported by
could be reached in reasonable time with only 0.01% Curran.[22] Functionalization of the silica surface with
catalyst, i.e., 100 times less than in the original paper by 1H,1H,2H,2H-perfluorodecyldimethylchlorosilane re-
Doyle.[10i] However, the extent of catalyst decomposi- sults in a material with high affinity for fluorous
tion apparently increased together with the reaction compounds. Such derivatized silica can be conveniently
rate, so that the amount of catalyst that could be loaded with dirhodium(II) perfluorocarboxylate com-
recovered by extraction after reaction completion was plexes [either Rh2(pfo)4 or Rh2(pft)4] by contacting it
even lower than at room temperature.
with a solution of the complex of choice (see the
The problem of recycling could be solved by devel- Experimental Section); adsorption of the complex on
oping a novel form of supported fluorous catalyst using the fluorous layer covering the silica surface takes place
an approach that we have named ™bonded fluorous in good yields and the resulting product can be easily
phase catalysis∫ (BFPC, Figure 2).[4b] In this approach, a isolated by simple filtration.[4b] In this way, BFP catalysts
fluorous film is built on a solid support by covalently with metal complex loadings of 22 mmol/g fluorous silica
binding perfluoroalkyl chains on its surface. Such for Rh2(pfo)4 and of 12 mmol/g fluorous silica for
materials are not entirely new: they have been used for Rh2(pft)4 could be prepared.
a long time as stationary phases for liquid chromatog-
Preliminary experiments performed at room temper-
raphy (bonded phase chromatography) and have re- ature pointed out that the BFP catalyst derived from
cently enjoyed a considerable revival of interest due to Rh2(pfo)4 was highly active under solventless conditions
their application in the extraction of fluorous-tagged (Table 4, entries 1 and 2), reaching activities up to five
molecules from complex mixtures.[16] Indeed, such
supports can, in principle, also extract fluorous catalysts
from their solution in an organic solvent, thereupon
forming a novel kind of supported catalyst. We have
preliminarily reported on the viability of this strategy in
a very recent communication.[4b]
Related approaches using silica functionalization with
non-fluorous chains were previously proposed by Neu-
mann and Cohen,[17] and Williams et al.[18] Very recently,
Pozzi et al. tried to apply the same BFPC strategy for
supporting fluorous chiral Co(salen) catalysts for the
hydrolytic kinetic resolution of terminal epoxides.[19]
However, the supported catalyst turned out to be
completely inactive. Contemporarily to the publication
of our preliminary results on this methodology, another
example of application of the same strategy was
Figure 2. The ™bonded fluorous phase catalysis∫ (BFPC)
published by Bannwarth.[20] Finally, very recent work
approach. The surface of a solid support S is derivatized
carried out independently by Jessop and Tsang must be
with molecules bearing long fluorous chains; in this way, a
mentioned, in which fluorous silica supports act as a
reservoir of catalytically active species which are
surface layer is created in which a fluorous catalyst can be
embedded.
Table 3. Triethylsilane alcoholysis catalyzed by dirhodium(II) perfluorocarboxylates under solventless conditions at 508C.
Entry
Catalyst
Mol % catalyst
Alcohol
Yield[a] [%]
TON
TOF [hÀ1]
1
2
3Rh
4
5
Rh2(pfd)4
Rh2(pfd)4
2(pfo)4
Rh2(pfdo)4
Rh2(pft)4
0.1
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
100
97
98
100
98
1000
9700
9800
10000
9800
42
404
408
417
408
0.01
0.01
0.01
0.01
Reaction conditions: see the Experimental Section.
[a]
GC yield after 24 h.
454
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 451 458

Solventless Silane Alcoholysis
FULL PAPERS
times higher than those observed with the unsupported almost quantitative yield; the purity of the product was
analogues. Furthermore, the catalysts could be sepa- assessed to be>95% by gas chromatography, NMR
rated by simple filtration with < 3% rhodium leaching spectroscopy and elemental analysis.
and could be recycled.[4b] In contrast, the BFP catalyst
Primary alcohols such as 1-octanol and benzyl alcohol
based on Rh2(pft)4 was found to be much less active react more rapidly than secondary alcohols such as 2-
(Entry 3). We speculate that such lower activity may be octanol also under BFP catalysis (Table 4, Entry 7).
due to the lower degree of dispersion of the latter Consequently, competitive silane alcoholysis experi-
complex on the fluorous silica surface, i.e., to the ments can be performed in this case as well, leading to
presence of aggregates of catalyst molecules.
preferential silylation of the primary alcohol function
We then moved to the evaluation of the performance (Entry 8). However, cyclic secondary alcohols such as
of the supported catalysts at 508C. Table 4 shows the cyclohexanol are also quite efficiently silylated (En-
relevant results. It can be appreciated that also in this try 10).
case the activities spring up dramatically for both
Finally, a preliminary evaluation of the effect of the
supported catalysts and are fully comparable with the nature of the silane on the reactivity of BFP catalysts has
values for the unsupported complexes, at least in the been performed (Table 5). It is interesting to note that
case of BFP-Rh2(pfo)4. Furthermore, the supported the nature of the organic substituents on the silane may
catalysts remain separable by simple filtration and are have an influence both on the overall activity of the
recyclable. Complete recycling in the small laboratory- catalytic system as well as on its selectivity towards
scale experiments performed up to now is somewhat primary alcohols in competitive experiments. Indeed,
difficult due to the small amount of solid catalyst phenyldimethylsilane displays a much lower selectivity
involved, which makes it difficult to collect the whole of than triethylsilane while reacting at a comparable rate,
the catalyst batch employed in one run. Nevertheless, we whereas t-butyldimethylsilane is much more selective
have been able to show that the recovered BFP catalysts but also quite less reactive. The differences in reactivity
maintain their catalytic activity at least for one recycling observed with the BFP catalyst are much more pro-
(Table 4, Entry 5). Even higher catalytic activities (50 nounced than those observed with the homogeneous
times higher than the originally reported homogeneous literature catalyst;[10i] we are currently looking for a
catalysts) can be reached by further increasing the suitable explanation of this phenomenon.
temperature to 808C, while the catalyst still remains
perfectly recyclable (Table 4, Entries 11 and 12). A
preparative experiment performed with 1 mL (1 equiv- Conclusion
alent) of each reagent using the reaction conditions of
Entry 11 yields after 16 hours reaction time and catalyst In conclusion, we have developed a novel reaction
separation by simple filtration the silyl ether product in protocol for silane alcoholysis catalyzed by dirhodi-
Table 4. Solventless triethylsilane alcoholysis using BFPC at different reaction temperatures.
Entry
Catalyst
Mol % catalyst
Alcohol
T [ 8C]
Yield[a] [%]
TON
TOF [hÀ1]
1[b]
BFP-Rh2(pfo)4
BFP-Rh2(pfo)4
2(pft)4
BFP-Rh2(pfo)4
BFP-Rh2(pfo)4
BFP-Rh2(pft)4
BFP-Rh2(pfo)4
BFP-Rh2(pfo)4
0.1
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
1-octanol
2-octanol
1-octanol
2-octanol
benzyl alcohol
cyclohexanol
1-octanol
1-octanol
RT
RT
RT
50
50
50
100
39
4
93
63180500
55
18
65
15
83
1000
3900
400
42
162
17
2[b]
0.01
0.01
0.01
0.0060
0.01
0.01
0.01
3BFP-Rh
4
9300
388
5[c]
6
43
5500
1800
6500
1500
8300
7700
9500
11360
229
75
271
63
346
321
1580
1890
7
50
50
8[d]
9
10
BFP-Rh2(pfo)4
BFP-Rh2(pfo)4
BFP-Rh2(pfo)4
BFP-Rh2(pfo)4
0.01
0.01
0.01
0.0081
50
50
80
80
77
11
95[e]
92[e]
12[f]
Reaction conditions: see the Experimental Section.
[a]
GC yield after 24 h.
[b]
Data from Ref.[4b]
[c]
Recycle of Entry 4.
Competitive reaction experiment.
Yield after 6 h.
Recycle of Entry 11.
[d]
[e]
[f]
Adv. Synth. Catal. 2004, 346, 451 458
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
455

Andrea Biffis et al.
FULL PAPERS
Table 5. Competitive solventless silane alcoholysis using 0.01 mol % BFP-Rh2(pfo)4 catalyst at 508C.
Entry
Silane
Alcohol
Yield[a] [%]
Relative
reactivity
1
2
3
Et3SiH
1-octanol
2-octanol
1-octanol
2-octanol
1-octanol
2-octanol
65
15
48
38
11
1
4.3
1.3
11
Me2PhSiH
t-BuMe2SiH
Reaction conditions: 1 equivalent of each reagent, catalyst, 24 h.
[a]
GC yield after 24 h.
filtered off, thoroughly washed with toluene and dried under
vacuum to constant weight. The mother liquor and the solution
from the washings were collected and the solvent was removed
at reduced pressure. The residue was dissolved in 6 mL hot
aqua regia and the resulting solution was diluted to 100 mL
with water. The Rh content in the solution was quantitatively
estimated by ICP-AAS, and from this result the Rh content in
the BFP catalyst was determined by difference. The yield in
supported complex turned out to be 22 mmol/g fluorous silica
(86% in respect to the initial amount of complex).
BFP-Rh2(pft)4: Rh2(pft)4 (25 mg) was dissolved in 2 mL
THF. Fluorous silica (0.5 g) was then added and the resulting
mixture was stirred at room temperature for 4 hours, after
which the supported catalyst was filtered off, thoroughly
washed with THFand dried under vacuum to constant weight.
The Rh content in the resulting BFP catalyst was determined as
described above for BFP-Rh2(pfo)4. The yield in supported
complex turned out to be 12 mmol/g fluorous silica (78% in
respect to the initial amount of complex).
um(II) perfluorocarboxylates which, in comparison
with the one originally described in the literature, allows
us to reach up to 50 times higher catalytic activity and
100 times higher catalytic productivity. The new reaction
protocol uses down to 100 times less catalyst than the
original one and runs in a completely solventless
fashion. Furthermore, heterogenization of these cata-
lysts has been accomplished by adsorbing them on silica
which has been previously functionalized at its surface
with perfluoroalkyl chains. Use of such supported
catalysts (bonded fluorous phase catalysts) allows an
easy and almost complete catalyst separation and
recycling with improved catalytic performance. We are
currently pursuing the application of these novel
supported catalysts to other alcohol substrates of
synthetic interest as well as to other technologically
relevant reactions which are known to be catalyzed by
dirhodium(II) perfluorocarboxylates.[23]
Catalytic Tests
Experimental Section
Reactions in organic solvents: Catalytic tests were run in a
Schlenk tube equipped with a magnetic stirring bar. The
Schlenk tube was charged with the required quantity of
catalyst, evacuated and filled with argon. The anhydrous
organic solvent (5 mL) was then added under argon, followed
by 162 mL (1 mmol) 1-octanol and 173 mL (1.05 mmol) trie-
thylsilane. The resulting solution was vigorously stirred at
room temperature. Samples of 0.5 mL were taken after 6 h
reaction time and analyzed by GC following the previously
established procedure.[4a]
Solventless reactions with homogeneous catalysts: In a
typical procedure, a Schlenk tube equipped with a magnetic
stirring bar was charged with the required quantity of catalyst,
evacuated and filled with argon. 1-Octanol (1 mL) and
triethylsilane (1.1 mL, 1.08 equivalents) were then added,
and the resulting solution was vigorously stirred at room
temperature or at 508C (thermostatted oil bath) for 24 hours.
Samples of 0.1 mL were withdrawn after this period, diluted
with 1 mL dichloromethane and analyzed by GC following the
previously established procedure.[4a] Catalyst recycling was
performed by adding 1 mL fluorous solvent to the reaction
mixture, stirring for a few minutes to allow partition to come to
equilibrium, removing the product phase and adding fresh
reagents before resuming the stirring. The amount of rhodium
remaining in the product phase was determined by ICP-AAS
Solvents (Carlo Erba), fluorinated solvents and reagents
(ABCR) and common reagents (Aldrich) were used as
received unless otherwise stated. The preparation of dirho-
dium(II) perfluorocarboxylates has been previously report-
ed.[4a,11]
Preparation of Fluorous Silica
The fluorous silica was prepared starting from high purity silica
gel 60 (Fluka) and 1H,1H,2H,2H-perfluorodecyldimethyl-
chlorosilane following a procedure fully analogous to that
reported by Curran.[22] The degree of silica functionalization,
determined gravimetrically and by elemental analysis, was
found to be 0.81 mmol/g.
Preparation of BFP Catalysts
BFP-Rh2(pfo)4: Rh2(pfo)4 (25 mg) was suspended in 5 mL
toluene. Fluorous silica (0.5 g) was then added and the
resulting mixture was briefly heated at reflux with efficient
stirring, upon which the metal complex quantitatively dis-
solved. The mixture was allowed to cool to room temperature
while stirring was continued, after which the green solid was
456
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 451 458

Solventless Silane Alcoholysis
FULL PAPERS
on an aliquot of the product phase (0.5 mL), which was treated
with 6 mL hot aqua regia and diluted to 100 mL with water
before the measurement.
[3] P. T. Anastas, J. Warner, Green Chemistry Theory and
Practice, Oxford University Press, Oxford, 1998.
[4] 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.
Solventless reactions with BFP catalysts: In a typical
procedure, a Schlenk tube equipped with a magnetic stirring
bar was charged with the required quantity of catalyst,
evacuated and filled with argon. 1-Octanol (1 mL) and
triethylsilane (1.1 mL, 1.08 equivalents) were then added,
and the resulting suspension was vigorously stirred at the
required reaction temperature (thermostatted oil bath) for 24
hours. Samples of 0.1 mL were withdrawn after this period,
diluted with 1 mL dichloromethane and analyzed by GC
following the previously established procedure.[4a] Catalyst
recycling was performed by filtering off the catalyst under air,
washing it with a little dichloromethane and drying under
vacuum before starting the new run. Rhodium leaching was
determined by ICP-AAS on an aliquot of the filtrate (0.5 mL),
which was treated with 6 mL hot aqua regia and diluted to
100 mL with water before the measurement.
Competitive reaction experiments: A Schlenk tube equip-
ped with a magnetic stirring bar was charged with the required
quantity of catalyst, evacuated and filled with argon. 1-Octanol
(1 mL), 2-octanol (1 mL, 1 equivalent) and silane (1 equiv-
alent) were then added, and the resulting mixture was
vigorously stirred at the required reaction temperature (ther-
mostatted oil bath) for 24 hours. Samples of 0.1 mL were
withdrawn after this period, diluted with 1 mL dichlorome-
thane and analyzed by GC following the previously established
procedure.[4a]
Preparative silane alcoholysis with BFP-Rh2(pfo)4 cata-
lyst: A Schlenk tube equipped with a magnetic stirring bar was
charged with 27.6 mg BFP-Rh2(pfo)4 (0.6 mmol catalyst,
0.01 mol %), evacuated and filled with argon. 1-octanol
(1 mL, 6.35 mmol) and 1 mL (6.35 mmol, 1 equivalent) trie-
thylsilane were then added. The resulting suspension was
placed in a thermostatted oil bath preheated at 808C and
vigorously stirred for 16 hours. The solid catalyst was filtered
off affording the triethylsilyl ether product; yield: 1.5 g (97%).
The product was analyzed without further purification by GC,
NMR spectroscopy and elemental analysis, and found to
be>95% pure. Anal. calcd. for C14H32OSi: C 68.73, H 13.35;
found: C 68.78, H 13.19.
[5] T. W. Greene, P. G. Wuts, Protective Groups in Organic
Synthesis, 2nd Edn., Wiley, New York, 1991.
[6] a) B. P. S. Chauhan, T. E. Ready, Z. Al-Badri, P. Boud-
jouk, Organometallics 2001, 20, 2725; b) Y. Li, Y.
Kawakami, Macromolecules 1999, 32, 3540; c) R. Zhang,
J. E. Mark, A. R. Pinhas, Macromolecules 2000, 33, 3508
and references cited therein.
[7] E. Lukevics, M. Dzintara, J. Organomet. Chem. 1985,
295, 265.
[8] 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.
[9] 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. Mur-
akami, Tetrahedron Lett. 1994, 35, 8413.
[10] Selected recent examples: a) L. D. Field, B. A. Messerle,
M. Rehr, L. P. Soler, T. W. Hambley, Organometallics
2003, 22, 2387; b) M.-K. Chung, G. Ferguson, V. Rob-
ertson, M. Schlaf, Can. J. Chem. 2001, 79, 949; c) S.
Chang, E. Scharrer, M. Brookhart, J. Mol. Catal. A:
Chem. 1998, 130, 107; d) X. Wang, W. W. Ellis, B.
Bosnich, Chem. Commun. 1996, 2561; e) C. Lorenz, U.
Schubert, Chem. Ber. 1995, 128, 1267; f) B. T. Gregg,
A. R. Cutler, Organometallics 1994, 13, 1039; g) T. C.
Bedard, J. Y. Corey, J. Organomet. Chem. 1992, 428, 315;
h) D. H. Barton, M. J. Kelly, Tetrahedron Lett. 1992, 33,
5041; i) M. P. Doyle, K. G. High, V. Bagheri, R. J. Pieters,
P. J. Lewis, M. W. Pearson, J. Org. Chem. 1990, 55, 6082;
j) X.-L. Luo, R. H. Crabtree, J. Am. Chem. Soc. 1989,
111, 2527; k) W. Caseri, P. S. Pregosin, Organometallics
1988, 7, 1373.
[11] a) A. Endres, G. Maas, Tetrahedron Lett. 1999, 40, 6365;
b) A. Endres, G. Maas, J. Organomet. Chem. 2002, 643
644, 174.
Acknowledgements
[12] I. T. Horvath, Pure Appl. Chem. 1998, 31, 641.
[13] F. A. Cotton, E. V. Dikarev, X. Feng, Inorg. Chim. Acta
1995, 237, 19.
[14] M. P. Doyle, S. N. Mahapatro, A. C. Caughey, M. S.
Chinn, M. R. Colsman, N. K. Harn, A. E. Redwine,
Inorg. Chem. 1987, 26, 3070.
Financial support from the University of Padova ™Progetti di
Ateneo 2002∫ is gratefully acknowledged. We wish to thank
Dr. Giulia Zanmarchi, University of Padova, for the ICP-AAS
measurements.
[15] a) V. K. Dioumaev, R. M. Bullock, Nature 2003, 424, 530;
b) G. Hourdin, A. Germain, C. Moreau, F. Fajula, J.
Catal. 2002, 209, 217; c) X. Zuwei, Z. Ning, S. Yu, K. Li,
Science 2001, 292, 1139; d) A. Biffis, M. Zecca, M.
Basato, J. Mol. Catal. A: Chemical 2001, 173, 249 and
references cited therein.
References
[1] Applied Homogeneous Catalysis with Organometallic
Compounds, (Eds.: B. Cornils, W. A. Herrmann), 2nd
Edn., Wiley-VCH, Weinheim, 2002.
[16] D. P. Curran, Synlett 2001, 1488.
[17] R. Neumann, M. Cohen, Angew. Chem. Int. Ed. 1997, 36,
1738.
[2] See, for example, the recently published thematic issue
on ™Recoverable Reagents and Catalysts∫: Chem. Rev.
2002, 102, Issue 10.
Adv. Synth. Catal. 2004, 346, 451 458
asc.wiley-vch.de
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
457

Andrea Biffis et al.
FULL PAPERS
[18] M. S. Anson, A. R. Mirza, L. Tonks, J. M. J. Williams,
Tetrahedron Lett. 1999, 40, 7147.
[22] D. P. Curran, S. Hadida, H. Mu, J. Org. Chem. 1997, 62,
6714.
[19] M. Cavazzini, S. Quici, G. Pozzi, Tetrahedron 2002, 58,
3943.
[20] C. C. Tzschucke, C. Markert, H. Glatz, W. Bannwarth,
Angew. Chem. Int. Ed. 2002, 41, 4501.
[23] a) M. P. Doyle, T. Ren, Prog. Inorg. Chem. 2001, 49, 113;
b) M. P. Doyle, K. G. High, C. L. Nesloney, Chem. Ind.
(Dekker) 1992, 47, 293 .
[21] a) C. Eckert, P. G. Jessop, C. L. Liotta, PCT Int. Appl.
WO 02/096550, 2002; b) P. M. Jenkins, A. M. Steele, S. C.
Tsang, Catal. Commun. 2003, 4, 45.
458
¹ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 451 458