A non-fluorous copper catalyst for the styrene cyclopropanation reaction in a fluorous medium

Article abstract of DOI:10.1039/b517337j

The complex Tp^Br3Cu(NCMe) (1), containing no fluorine atoms, can be dissolved in the perfluoropolyether FOMBLIN and employed as a catalyst for the styrene cyclopropanation reaction with ethyl diazoacetate, with activities and diastereoselectivities identical to those observed under homogeneous conditions with the advantage of being able to use a fluorous separation technique for catalyst recycling. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b517337j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
A non-fluorous copper catalyst for the styrene cyclopropanation
reaction in a fluorous medium
Juan Urbano,a Rocio Izarra,b Jose Luis Go´mez-Ariza,b Swiatoslaw Trofimenko,c M. Mar D´ıaz-Requejo*a
and Pedro J. Pe´rez*a
Received (in Cambridge, UK) 6th December 2005, Accepted 12th December 2005
First published as an Advance Article on the web 20th January 2006
DOI: 10.1039/b517337j
nine bromine atoms in the Tpx ligand skeleton (Fig. 1), accounting
for ca. 70% of the overall molecular weight of this compound.
With this very high percentage of halide, we wondered if it were
possible to dissolve it in halide-containing media, with the aim of
using it for catalytic purposes. We focused particularly on
perfluoropolyethers, which should provide some degree of polarity
to help dissolve the copper complex. From this search, we found
that 1 could be dissolved in FOMBLIN1, a perfluoropolyether of
general composition CF3O[–CF(CF3)CF2O–]x(–CF2O–)yCF3 and
commercially available due to its use as a vacuum oil. When
dichloromethane or cyclohexane solutions of 1 were stirred for
several hours with FOMBLIN1, some of the copper complex was
absorbed by the fluorous phase, as determined by AAS studies of
the supernatant organic phases at the end of the process. The
initial experiments were carried out using two different types of
FOMBLIN1, with different compositions as well as viscosities.{
The mixtures were stirred for 20 h, after which the organic
(dichloromethane or cyclohexane) phases were studied by AAS to
determine the amounts of Cu still dissolved; the remaining amount
of the initial 0.25 mmol of 1 now being retained in the fluorous
phase. As shown in Table 1 entries 1–4, the more viscous 1400 cSt
(1St = 0.0001 m2 s21; 1cSt = 0.01 St) samples absorbed a higher
quantity of 1 and the cyclohexane solutions also induced higher
phase absorptions.
The complex TpBr3Cu(NCMe) (1), containing no fluorine
atoms, can be dissolved in the perfluoropolyether FOMBLIN1
and employed as a catalyst for the styrene cyclopropanation
reaction with ethyl diazoacetate, with activities and diastereo-
selectivities identical to those observed under homogeneous
conditions with the advantage of being able to use a fluorous
separation technique for catalyst recycling.
With barely ten years since the appearance of the seminal work by
Horva´th and Ra´bai,1 fluorous biphasic catalysis (FBC) has
nowadays become one of the most interesting areas in the field
of catalysis.2 The increasing number of contributions related to this
field has compelled the need for a strict definition of what should
be named with the ‘‘fluorous’’ appellative;3 whether applied to a
reaction medium, catalyst, or separation technique, amongst
others. A common feature of systems tagged as FBC is the
existence of a fluorous medium, in which a catalyst, containing a
certain number of carbon–fluorine bonds, is either soluble or
partially soluble. A second, immiscible phase is employed as a
carrier for the reagents and products. In many cases, the phases
become mixed upon heating, a fact that favours catalytic reactions.
Upon cooling back to room temperature (or below), separation of
the phases takes place, and recycling of the catalyst and removal of
the products can then be performed. A usual rule of thumb for a
catalyst employed in FBC is that it should contain at least a 60 wt%
of fluorine, although some exceptions have been found.4
The aim of this first series of experiments was to evaluate
whether or not complex 1 could be dissolved in this fluorous
phase. However, the concentrations of 1 in FOMBLIN1 shown in
Table 1 do not necessarily correspond to the distribution
equilibrium. Because of this, we have also studied the effect of
time on the absorption process, choosing 60 cSt FOMBLIN1 as
the fluoro phase. A series of identical experiments were performed
using cyclohexane as the carrier for 1 and the FOMBLIN1 (60 cSt)
sample over 8, 30 and 45 h. Fig. 2 displays a plot of [1] vs.
employed absorption time. The existence of a saturation limit for
the amount of the complex TpBr3Cu(NCMe) that can be dissolved
in this medium is observed. Regarding the intimate nature of the
In the recent years we have reported the catalytic properties of a
family of complexes of general formula TpxCu(L) (Tpx
=
homoscorpionate ligand)5 towards the decomposition of ethyl
diazoacetate and subsequent transfer of the carbene unit
:CHCO2Et to several saturated and unsaturated substrates under
homogeneous conditions.6–8 After these results, we decided to
investigate the potential of these catalysts, under heterogeneous
conditions, to be subject to the well-known and advantageous
processes of catalyst recovery, recycling and product separation.
Among these TpxCu(L) complexes, that containing the hydro-
tris(3,4,5-tribromopyrazolyl)borate ligand TpBr3Cu(NCMe) (1)
showed very interesting catalytic properties.7,8 Complex 1 includes
aLaboratorio de Cata´lisis Homoge´nea, Departamento de Qu´ımica y
Ciencia de los Materiales, Unidad Asociada al CSIC, Universidad de
Huelva, Campus de El Carmen, 21007-Huelva, Spain.
E-mail: mmdiaz@dqcm.uhu.es; perez@dqcm.uhu.es;
Fax: (+34) 959 219942; Tel: (+34) 959 219956
bDepartamento de Qu´ımica y Ciencia de los Materiales, Universidad de
Huelva, Campus de El Carmen, 21007-Huelva, Spain
cDepartment of Chemistry and Biochemistry, University of Delaware
Newark, Delaware, 19716, USA
Fig. 1 The hydrotris(3,4,5-tribromopyrazolyl)borate ligand, TpBr3
.
1000 | Chem. Commun., 2006, 1000–1002
This journal is ß The Royal Society of Chemistry 2006

View Article Online
Table 1 Absorption of complex TpBr3Cu(NCMe) (1) in FOMBLIN1
at room temperature
corresponding to a mixture of diethyl fumarate and maleate.
This diastereoselectivity was nearly identical, within experimental
error, to that obtained when the experiment was carried out under
homogeneous conditions in dichloromethane (cis : trans, 58 : 42).
The amount of copper that leached into the hexane solution was
determined to be 1.5% of the initial amount absorbed, therefore
98.5% of 1 remained in the fluorous phase. In good accordance
with this observation, a second cycle was run under identical
conditions (Table 2, entry 2), with yields and diastereoselectivities
very similar to those in the first test. The catalyst was recovered
and recycled three more times (Table 2, entries 3–5) with the same
behaviour; copper leaching being maintained within the 1–1.5%
range.
Amount of
1 absorbed/
Time/ FOMBLIN1
h
/
[1] in
FOMBLIN1/M
Entry
cSta
Solvent
mmolb
1
2
3
4
5
6
7
20
20
20
20
8
60
1400
60
1400
60
60
CH2Cl2 4.2 6 1023 1.05 6 1023
CH2Cl2 5.7 6 1023 2.85 6 1023
C6H12 5.9 6 1023 1.48 6 1023
C6H12 8.0 6 1023
4.0 6 1023
C6H12 4.3 6 1023 1.07 6 1023
C6H12 7.0 6 1023 1.75 6 1023
C6H12 7.3 6 1023 1.82 6 1023
30
45
60
a
The volume of FOMBLIN1 employed was 4 mL for the 60 cSt
b
samples and 2 mL for the 1400 cSt samples. Determined by AAS.
ð1Þ
The observation of catalyst leaching must unequivocally be
related to some loss of activity, though not appreciably within the
reaction time (1.5 h) shown in Table 2. To observe this, we
monitored the evolution of nitrogen as a probe for the reaction
rate.{ Fig. 4 displays the variation of nitrogen evolution, at room
temperature, during the recycling process, where a certain decrease
is clearly observable in the reaction rate from one cycle to the next.
This is in good accordance with the already mentioned detection of
copper leaching. A very low concentration of 1 (ca. 1024 M) was
employed to amplify the differences between the kinetic curves.
The results displayed in Table 2 demonstrate the catalytic
capabilities of complex 1 in FOMBLIN1 for the carbene transfer
reaction, with the same catalytic activity and diastereoselectivity
being repeated throughout the five cycles studied, and with minor
leaching being observed after each run. This behaviour is identical
to that observed under homogeneous conditions, i.e., the change in
Fig. 2 Variation of [1] in FOMBLIN1 (60 cSt) with the time employed
in the absorption process at room temperature.
interaction between complex 1 and FOMBLIN1 (60 cSt), we can
only assume at this stage of our investigations that it has to be
related to the existence of the bromine atoms in the molecules;
attempts to dissolve measurable amounts of other TpxCuL
complexes being ineffective.
The extraction of complex 1 from this fluorous medium was
performed with various solvents.
A series of samples of
FOMBLIN1 (4 mL), containing an exactly-determined amount
of dissolved 1, were treated with six different solvents: dichlor-
omethane, THF, acetone, ethanol, cyclohexane and hexane. After
5 min of stirring, the organic solution was separated and the
copper content determined by AAS. The results are shown in
Fig. 3. With dichloromethane, THF or acetone, nearly 30% of the
copper was removed from the fluorous phase. Ethanol only
retrieved 10%, whereas cyclohexane and hexane showed 1.5–2% of
copper removal. Therefore, we have chosen hexane as the solvent
for the separation of the products after the catalytic reaction, since
there was no significant loss of catalyst due to it.
Fig. 3 Effect of solvent on the washing, at room temperature, of the
FOMBLIN1 solutions of 1. Initial [1]FOMBLIN1 = 1.48 6 1023 M.
Leaching determined by AAS.
Once the fixation of complex 1 in a fluorous phase was
achieved, we moved on to study its catalytic use in a well-known
probe reaction: styrene cyclopropanation with ethyl diazoacetate
(EDA),9 where two diastereomers, cis and trans, are formed (eqn.
(1)). A ca. 2.4 6 1023 M solution of 1 in FOMBLIN1 60 cSt
(0.0095 mmol in 4 mL) was charged with 0.5 mmol of EDA and
2 mmol of styrene, and stirred for 1.5 h. After this time, 2 mL of
hexane was added, the mixture stirred for 5 min and the organic
phase separated. GC studies revealed the total consumption of
diazo compound. 90% cyclopropanes were found, with a cis : trans
ratio of 60 : 40 (Table 2, entry 1), the remaining 10%
Table 2 Styrene cyclopropanation of EDA using TpBr3Cu(NCMe)
(1) as the catalyst in FOMBLIN1 at room temperature
Yield
(%)
Amount of Cu
leached/mmol (%)
Entry
Cycle
Time/h
cis : trans
1
2
3
4
5
1
2
3
4
5
1.5
1.5
1.5
1.5
1.5
90
90
85
93
91
60 : 40
59 : 41
60 : 40
58 : 42
57 : 43
—
1.40 6 1024 (1.5)
1.07 6 1024 (1.15)
1.24 6 1024 (1.35)
1.23 6 1024 (1.35)
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 1000–1002 | 1001

View Article Online
Notes and references
{ Experimental procedures, all performed at room temperature. (a)
Fixation of 1 in FOMBLIN1, general procedure: A solution of 0.25 mmol
of 17 in 4 mL of the solvent (dichloromethane or cyclohexane) was
transferred into a Schlenk tube containing 4 mL of FOMBLIN1 60 cSt or
2 mL of FOMBLIN1 1400 cSt, purchased from Aldrich and employed
without further purification. After the desired stirring time (from 8 to 45 h),
the mixture was decanted and the upper, organic phase separated. The
volatiles were removed under vacuum and the residue investigated by AAS,
following standard protocols to determine the amount of copper not
absorbed into the FOMBLIN1. (b) Washing with different solvents: A
solution of 1 in FOMBLIN1 (60 cSt), prepared as above and with a
known content of 1, was treated with 4 mL of a given solvent
(dichloromethane, THF, acetone, ethanol or hexane). After 5 min of
stirring, the upper phase was separated and investigated, as above, to
measure the leaching of copper. (c) Styrene cyclopropanation: A solution of
1 in FOMBLIN1 (2.4 6 1023 M, 4 mL of FOMBLIN1) was charged
with 0.5 mmol (57 mg) of ethyl diazoacetate and 2 mmol (208 mg) of
styrene. The mixture was stirred for 1.5 h, and then 4 mL of hexane added.
After 5 min of stirring, the organic phase was separated and investigated by
GC, as previously reported.6 In addition, the amount of copper leached
was also determined by AAS.
Fig. 4 Nitrogen evolution in the reaction of EDA and styrene, catalyzed
by 1 in FOMBLIN1 at room temperature. Run #3 has been omitted for
clarity.
reaction medium has not affected its catalytic properties. Overall,
the use of this copper–FOMBLIN1 system achieves the ultimate
goal for when moving from a homogeneous catalyst towards
heterogeneous conditions: to easily separate the catalyst and
products while operating with the same activities and selectivities
as those of purely homogeneous conditions. There are only a few
examples of fluorous catalysis for the carbene transfer reaction
from diazo compounds to organic substrates, these being based
exclusively on rhodium10 or copper.11 In all cases, a fluorous
ponytail-containing catalyst and a fluorous reaction medium were
employed, with differing results to those presented herein, and
using a fluorous medium with a non-fluorous catalyst. Another
singularity in our case is that the already-reported copper-based
systems for styrene cyclopropanation, under fluorous conditions,
displayed a substantial decrease in yields when the catalyst was
reused in successive cycles. This is not the case for this system,
where yields and selectivities are maintained within a narrow
interval. However, we wish to point out that it is possible that the
catalytic reaction might occur in a homogeneous phase formed by
styrene and EDA, and that the addition of hexane could force the
catalyst to return to the fluorous phase. In any case, this would not
affect the catalytic capabilities of this system in terms of recovery
and reuse.
{ The rate determining step in the metal-catalyzed diazo compound
decomposition corresponds to the formation of a transient metallacarbene
intermediate from the interaction of the catalyst and the diazo compound,
and the concomitant formation of a molecule of N2. See ref. 9.
1 T. Horva´th and J. Ra´bai, Science, 1994, 266, 72.
2 A. P. Dobbs and M. R. Kimberley, J. Fluorine Chem., 2002, 118, 3.
3 J. A. Gladysz and D. P. Curran, Tetrahedron, 2002, 58, 3823.
4 (a) E. de Wolf, G. van Koten and B.-J. Deelman, Chem. Soc. Rev., 1999,
28, 37; (b) I. T. Horva´th, Pure Appl. Chem., 1998, 31, 641.
5 S. Trofimenko, Scorpionates, The Coordination Chemistry of
Polypyrazolylborate Ligands, Imperial College Press, London, 1999.
6 (a) Olefin cyclopropanation: M. M. D´ıaz-Requejo, T. R. Belderrain,
S. Trofimenko and P. J. Pe´rez, J. Am. Chem. Soc., 2001, 123, 3167;
M. M. D´ıaz-Requejo, A. Caballero, T. R. Belderrain, M. C. Nicasio,
S. Trofimenko and P. J. Pe´rez, J. Am. Chem. Soc., 2002, 124, 978; (b)
Alkyne cyclopropenation: M. M. D´ıaz-Requejo, M. A. Mairena,
T. R. Belderrain, M. C. Nicasio, S. Trofimenko and P. J. Pe´rez, Chem.
Commun., 2001, 1804; (c) N–H insertion: M. E. Morilla, M. M. D´ıaz-
Requejo, T. R. Belderrain, M. C. Nicasio, S. Trofimenko and P. J. Pe´rez,
Chem. Commun., 2002, 2998; (d) O–H insertion: M. E. Morilla,
M. J. Molina, M. M. D´ıaz-Requejo, T. R. Belderrain, M. C. Nicasio,
S. Trofimenko and P. J. Pe´rez, Organometallics, 2004, 23, 2914; (e)
Addition to aromatic rings: M. E. Morilla, M. M. D´ıaz-Requejo,
T. R. Belderrain, M. C. Nicasio, S. Trofimenko and P. J. Pe´rez,
Organometallics, 2004, 23, 293; (f) Addition to furans: A. Caballero,
M. M. D´ıaz-Requejo, M. C. Nicasio, S. Trofimenko, T. R. Belderrain
and P. J. Pe´rez, J. Org. Chem., 2005, 70, 6101.
In conclusion, we have discovered that complex
TpBr3Cu(NCMe) (1) can be absorbed in perfluoropolyether
CF3O[–CF(CF3)CF2O–]x(–CF2O–)yCF3, FOMBLIN1 (60 cSt),
in spite of the lack of any fluorine atoms in the structure of the
copper complex. The absorbed amounts of 1 can be employed to
catalyze the styrene cyclopropanation reaction with ethyl diazoa-
cetate, in a process that provides the expected cyclopropanes with
constant yields and diastereoselectivities along five cycles of
recovery and reuse of the catalyst-containing fluorous phase. We
believe that this finding could open up a new area in designing
non-fluorine-containing catalysts for use in certain fluorous
reaction media.
7 (a) M. M. D´ıaz-Requejo, T. R. Belderrain, M. C. Nicasio,
S. Trofimenko and P. J. Pe´rez, J. Am. Chem. Soc., 2002, 124, 896; (b)
A. Caballero, M. M. D´ıaz-Requejo, T. R. Belderrain, M. C. Nicasio,
S. Trofimenko and P. J. Pe´rez, J. Am. Chem. Soc., 2003, 125, 1446; (c)
A. Caballero, M. M. D´ıaz-Requejo, T. R. Belderrain, M. C. Nicasio,
S. Trofimenko and P. J. Pe´rez, Organometallics, 2003, 22, 4145.
8 M. M. Dı´az-Requejo, P. Wehrmann, M. D. Leatherman,
S. Trofimenko, S. Mecking, M. Brookhart and P. J. Pe´rez,
Macromolecules, 2005, 38, 4966.
9 M. P. Doyle, M. A. McKervey and T. Ye, Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds, John Wiley & Sons, New
York, 1998.
10 (a) A. Biffis, M. Braga, S. Cadamuro, C. Tubaro and M. Basato, Org.
Lett., 2005, 7, 1841; (b) A. Endres and G. Maas, J. Organomet. Chem.,
2002, 643–644, 174; (c) A. Endres and G. Maas, Tetrahedron, 2002, 58,
3999; (d) A. Endres and G. Maas, Tetrahedron Lett., 1999, 40, 6365.
11 (a) J. Bayardon, D. Sinou, O. Holczknecht, L. Mercs and G. Pozzi,
Tetrahedron: Asymmetry, 2005, 16, 2319; (b) I. Shepperson, S. Quici,
G. Pozzi, M. Nicoletti and D. O’Hagan, Eur. J. Org. Chem., 2004, 4545;
(c) R. Annunziatta, M. Benaglia, M. Cinquini, F. Cozzi and G. Pozzi,
Eur. J. Org. Chem., 2003, 1191.
We thank the MEC (Proyecto BQU2002-01114) for financial
support and the Ramo´n y Cajal Program for a Research
Fellowship (M. M. D. R.). We also thank the Universidad
de Huelva for a Research Studentship (J. U.) and the NMR
Service.
1002 | Chem. Commun., 2006, 1000–1002
This journal is ß The Royal Society of Chemistry 2006