Substrate-selective aqueous organometallic catalysis. How size and chemical modification of cyclodextrin influence the substrate selectivity

Article abstract of DOI:10.1016/j.tet.2004.06.028

Randomly hydroxypropylated and methylated cyclodextrins with different cavity size have been used as inverse phase transfer catalysts in a palladium catalyzed Tsuji-Trost reaction with water-insoluble alkylallylcarbonates and alkylallylurethanes as substrate. It has been shown that the molecular recognition ability of both α-CD and β-CD derivatives towards these substrates was responsible for an increase in the reaction rates and remarkable substrate selectivities between a linear and a branched structure. By contrast, the too wide cavity of γ-CD derivatives did not allow these carriers to be efficient in terms of substrate selectivity. Thus, the performances of a cyclodextrin carrier in this cleavage reaction strongly depended on the size of the cavity in which the substrate had to fill in as close as possible the available space.

Full text of DOI:10.1016/j.tet.2004.06.028

Tetrahedron 60 (2004) 6487–6493
Substrate-selective aqueous organometallic catalysis.
How size and chemical modification of cyclodextrin influence
the substrate selectivity
Christophe Torque, Herv e´ Bricout, Fr e´ d e´ ric Hapiot and Eric Monflier
*
Facult e´ des Sciences Jean Perrin, Universit e´ d’Artois, Rue Jean Souvraz, SP 18-62307 Lens C e´ dex, France
Received 24 March 2004; revised 28 May 2004; accepted 7 June 2004
Abstract—Randomly hydroxypropylated and methylated cyclodextrins with different cavity size have been used as inverse phase transfer
catalysts in a palladium catalyzed Tsuji-Trost reaction with water-insoluble alkylallylcarbonates and alkylallylurethanes as substrate. It has
been shown that the molecular recognition ability of both a-CD and b-CD derivatives towards these substrates was responsible for an
increase in the reaction rates and remarkable substrate selectivities between a linear and a branched structure. By contrast, the too wide cavity
of g-CD derivatives did not allow these carriers to be efficient in terms of substrate selectivity. Thus, the performances of a cyclodextrin
carrier in this cleavage reaction strongly depended on the size of the cavity in which the substrate had to fill in as close as possible the
available space.
q 2004 Elsevier Ltd. All rights reserved.
1
. Introduction
water soluble CD derivatives is also an original way to
obtain substrate selective reactions. In fact, when the
organic phase contains an isomers mixture, the CD
preferentially transfers into the aqueous phase the isomer
that interacts with the CD cavity. The substrate/CD complex
can then reacts with the aqueous organometallic catalyst to
give the product that is released into the organic phase,
according to Figure 1.
Cyclodextrins (CDs) are a-(1-4)-linked cyclic oligo-
saccharides usually consisting of 6(a), 7(b) or 8(g)
glucopyranose units. They exhibit a torus-shaped structure
with a hydrophobic cavity into which a guest molecule of
1
appropriate size and shape can be incorporated. The use
of the host cavity as microreactor to perform chemical
reactions has attracted the interest of numerous organic
chemists since the 1960s. Indeed, the microenvironment
around the reactant in the CD cavity is different from that
in the reaction media and a change of the course of the
reaction can be observed. For example, there are numerous
studies in which binding of the substrate into the host
cavity alters the regio- or stereoselectivity of the attack by
The feasibility of this approach was clearly demonstrated
with randomly methylated b-CD (Me-b-CD) in a palladium
catalyzed Tsuji-Trost reaction with water-insoluble alkyl-
allylcarbonates and alkylallylurethanes as substrates.
Indeed, by using different pairs of structural isomers, it
was found that the size-fit concept which postulates the
highest reactivity for the best size-matched host–guest pair
was a very useful tool for predicting the values of substrate
2
3
an external reagent. These changes were attributed to a
control of the substrate conformation or to the orientation
of the substrate included in the cavity towards the reagent
6
selectivity.
4
in solution. The selective binding of the substrate by
the CD cavity can also lead to substrate selective
reaction. Indeed, when the cyclodextrin itself catalyzes the
reaction, only the substrate that penetrates into the host
In this paper, experiments with different chemically
modified CDs have been performed to systematically
determine effect of the size cavity and the chemical
modification on the substrate selectivity. Six CD derivatives
have been evaluated in the palladium catalyzed cleavage of
isomers of biphenylmethylallylcarbonate or N-alkyl-O-
allylurethanes: the hydroxypropylated a-CD (HP-a-CD),
the hydroxypropylated b-CD (HP-b-CD), the hydroxypro-
pylated g-CD (HP-g-CD), the methylated a-CD (Me-a-
CD), the methylated b-CD (Me-b-CD), and the methylated
g-CD (Me-g-CD) (Fig. 2).
5
cavity reacts.
The combination of an aqueous organometallic catalyst and
Keywords: Palladium; Water-soluble phosphines; Cyclodextrins;
Molecular recognition; Carbonate and urethane cleavage.
*
Corresponding author. Tel.: þ33-321791772; fax: þ33-321791736;
e-mail address: monflier@univ-artois.fr
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.06.028

6488
C. Torque et al. / Tetrahedron 60 (2004) 6487–6493
Figure 1. Concept of the substrate-selective aqueous organometallic catalysis mediated by cyclodextrin derivatives. Only the substrate S1 fits properly in the
host cavity of the cyclodextrin and, consequently reacts with the water soluble catalyst to give the product P.
Figure 2. Chemically modified cyclodextrins used in this work.

C. Torque et al. / Tetrahedron 60 (2004) 6487–6493
6489
Scheme 1.
The cleavage of these substrates was catalyzed by a
palladium/trisulfonated triphenylphosphine combination in
the presence of diethylamine as allyl scavenger (Scheme 1).
that ortho-biphenylmethylallylcarbonate and para-biphenyl-
methylallylcarbonate form a relevant pair of isomers to
evaluate the discriminating power of the CD derivatives.
7
This reaction can find numerous applications in the field
of protection/deprotection chemistry by using the allyloxy-
carbonyl group (Alloc) as protecting group.
The results obtained without addition of any CD (reference
experiment) and in the presence of the six CD derivatives
are gathered in Figure 3.
8
For both substrates, the relative reaction rate (ratio between
the initial catalytic activity in the presence of CD and the
initial catalytic activity without CD) has been measured and
the ratio of that obtained with the linear isomer was divided
by that of the bent isomer to give the substrate selectivity.
As expected, the reaction rates without CD were very low
(initial catalytic activity: 0.3 h2 ) and identical. This last
result indicates clearly that the two isomers had the same
water solubility and exhibited the same reactivity towards
palladium species. When a modified CD was added to the
system, the reaction rates increased whatever the CD and a
discriminating process appeared with a- and b-CD
derivatives.
2
. Results and discussion
The first experiments that have been carried out in the
following study were performed with two alkylallylcarbo-
nate isomers: the linear para-biphenylmethylallylcarbonate
and the bent ortho-biphenylmethylallylcarbonate. These
two molecules strongly differed from their shape and
consequently from the space they occupied in the CD cavity.
1
First of all, both isomers have been cleaved in a biphasic
medium in which a classical co-solvent (acetonitrile) was
used as mass-transfer promoter to reduce mass transfer
limitation (see Section 4.3). This control experiment was
carried out to make sure that substrate selectivity was not
due to a different reactivity of the palladium species towards
the substrate or nucleophile. Thus, the deprotection of a
More precisely, an analysis of the results obtained with
HP-a-CD showed that the cleavage reaction of the linear
para-biphenylmethylallylcarbonate occurred at a relative
reaction rate of 53, whereas it was only 14 with the bent
ortho-biphenylmethylallylcarbonate. The ratio between
these two initial catalytic activities gave a substrate
selectivity of 3.8, which means that the linear isomer was
converted 3.8 times faster than the bent one. When using
HP-b-CD as a carrier, relative reaction rates were higher
than those obtained with HP-a-CD (88 and 37 for the para
5
0/50 mixture of ortho-biphenylmethylallylcarbonate and
para-biphenylmethylallylcarbonate was carried out in the
presence of 0.4 g of acetonitrile (20% by weight of
acetonitrile in the aqueous phase). In these experimental
conditions, the two isomers were cleaved at the same rate
(initial activity of 1.2 h ) and, consequently, no substrate
selectivity was observed. Thus, this experiment confirms
2
1
Figure 3. Reaction rate and substrate selectivity observed with para-biphenylmethylallyl carbonate and ortho-biphenylmethylallylcarbonate. (a) The relative
reaction rate was defined as the ratio between the initial catalytic activity in the presence of cyclodextrin and the initial catalytic activity without cyclodextrin.
2
1
The initial catalytic activity without cyclodextrin is 0.3 h for each isomer (reference). (b) The substrate selectivity was defined as the ratio between the
relative reaction rates observed between the two isomers.

6490
C. Torque et al. / Tetrahedron 60 (2004) 6487–6493
and ortho isomers respectively) but the substrate selectivity
was lower (2.4). Although a rate enhancement was observed
relative to the reference experiments, the relative reaction
rates were lower with HP-g-CD than with HP-b-CD (29 vs.
raphy, was used as a measure for substrate selectivity. As
already mentioned, no substrate selectivity was observed in
a control experiment in which acetonitrile was used as a
mass transfer promoter. By contrast, the use of Me-a-CD or
Me-b-CD led to substantial substrate selectivity. Indeed,
substrate selectivities of 7.2 and 3.6 were observed in the
initial stages of the reaction with the Me-a-CD or Me-b-
CD, respectively.
8
8 respectively for the para isomer and 28 vs. 37
respectively for the ortho isomer) and no substrate
selectivity was observed with the HP-g-CD.
These first results might be interpreted in terms of
recognition ability of the supramolecular carrier towards
the shape of the substrate. The HP-a-CD constitutes a better
discriminant carrier than the HP-b-CD because of its
smaller cavity. The para-biphenylmethyl group of the
linear isomer is able to deeply penetrate inside the HP-a-
CD cavity since no interaction hindered its recognition. By
contrast, the bent isomer can only be slightly recognised by
the cavity of HP-a-CD because the approach of both
molecules is hindered by steric repulsions. Actually, if one
considered that the biphenyl group comes into the cavity,
the allylcarbonate moiety becomes very close to the edge of
the CD and the inclusion is consequently not so deep than it
is with the linear isomer. When the cavity size is increased,
the penetration of both isomers in the cavity is easier.
However, the benefit effect is more marked for the bent
isomer that fits poorly in the HP-a-CD cavity than for the
linear isomer that fits well in the cavity. This difference
behaviour induces logically a decrease in the substrate
selectivity (3.8 with HP-a-CD vs. 2.4 with HP-b-CD). In
the case of HP-g-CD, the cavity size is clearly too wide to
efficiently recognise the linear isomer, resulting in a drastic
decrease in the reaction rate. In fact, it appears that both
isomers are transferred with the same rate in the aqueous
phase leading to no substrate selectivity.
According to us, two factors can account for the higher
efficiency of methylated CDs. The first is the surface-active
behavior of these CDs. Indeed, surface tension measure-
9
ments indicated that all methylated CDs are much more
surface active than hydroxypropylated CDs. So, by
decreasing the interfacial tension between the aqueous and
the organic layers, these CDs greatly contribute to increase
the mass transfer between the two phases. The second
factor, and the most important, is the presence of a well-
defined extended hydrophobic host cavity. Indeed, attach-
ment of 12.8 methyl groups to the b-CD extends its cavity.
Thus, these methylated CDs have much more important
cavity volumes approximately 10–20% larger than HP-CDs
1
0,11
due to the increase in the height of the CD torus.
This
gives rise to CDs that accommodate more easily highly
hydrophobic substrates and, consequently transfer more
1
1
efficiently the substrate into the aqueous phase. Moreover,
the benefit effect of an extended cavity is clearly much more
important for a linear substrate than for a bent substrate.
This last point is crucial to understand why the discrimi-
nation is higher with Me-CDs than with HP-CDs (substrate
selectivity: 8 vs. 3.8 for a-CDs and 4.0 vs. 2.4 for b-CDs).
Finally, it must be pointed out that the decrease in the
substrate selectivity when the cavity was wider could be
rationalized by assuming that a destabilization of the para-
isomer/CD complex occurred because of a growing
mobility of the guest in the host cavity and, by contrast,
that the ortho-isomer/CD complex was more stabilized due
to a deeper penetration of the substrate in the cavity.
Experiments conducted with the methylated CDs instead of
hydroxypropylated CDs gave a more clear view of
processes leading to rate enhancement and substrate
selectivity. What striked when looking at the values
obtained with Me-a-CD, Me-b-CD and Me-g-CD was
their greater ability to improve the performances of the
system whether in terms of reaction rate or substrate
selectivity. The effect of methylation on the reaction rate
was particularly visible with Me-a-CD for which a relative
reaction rate of 283 was measured with the para-
biphenylmethylallylcarbonate (it was only 53 with HP-a-
CD). Similarly, the relative reaction rate observed with the
ortho isomer was higher with Me-a-CD than with HP-a-CD
In a second part of this study, we wanted to enlarge our
conclusions to other substrates. Tsuji-Trost reactions have
then been carried out in the same experimental conditions
than those described above with two alkylallylurethane
isomers: the linear N-dodecyl-O-allylurethane and the
branched N,N-dihexyl-O-allylurethane. Due to the presence
of alkyl chains, these substrates are expected to be more
flexible than the biphenylmethylallylcarbonate isomers. As
above mentioned for the carbonates, it was also carefully
checked that these two isomers exhibited the same reactivity
in a biphasic medium in which acetonitrile was used as a
mass-transfer promoter (see Section 4.3). In these experi-
ments, the deprotection of a 50/50 mixture of N,N-dihexyl-
O-allylurethane and N-dodecyl-O-allylurethane was carried
out in the presence of acetonitrile (2 g; 100% by weight of
acetonitrile in the aqueous phase). The ratio of the products
N-dodecylamine and N,N-dihexylamine was used as a
measure for substrate selectivity. As expected, no substrate
selectivity was observed in this control experiment. Indeed,
the two substrates were converted at the same rate
(0.25 h2 ).
(
35 vs. 14 respectively). Me-b-CD and Me-g-CD led to the
same conclusions when compared respectively to HP-b-CD
and HP-g-CD but in a less extend. More interesting is the
fact that the widening of the diameter of the Me-CDs led to a
decrease in cleavage rate for the linear isomer in parallel to
an increase in the cleavage rate for the bent isomer, inducing
a drop of the substrate selectivity. Indeed, the substrate
selectivity was 8.0, 4.0 and 1.1 with the Me-a-CD, Me-b-
CD and Me-g-CD, respectively. Interestingly, it is worth
mentioning that the high substrate selectivities observed
with the Me-a-CD and the Me-b-CD were also confirmed
with competitive experiments. In these experiments, the
deprotection of a 50/50 mixture of ortho- and para-
biphenylmethylallyl carbonate was carried out with meth-
ylated CDs. The ratio of the products p-biphenylmethanol
and o-biphenylmethanol, as determined by chromatog-
1
The results obtained without CD and in the presence of the
CD derivatives are summarized in Figure 4.

C. Torque et al. / Tetrahedron 60 (2004) 6487–6493
6491
Figure 4. Reaction rate and substrate selectivity observed with N-dodecyl-O-allylurethane and N,N-dihexyl-O-allylurethane. (a) the relative reaction rate was
defined as the ratio between the initial catalytic activity in the presence of cyclodextrin and the initial catalytic activity without cyclodextrin. The initial
21
catalytic activity without cyclodextrin is 0.03 h for each isomer (reference). (b) The substrate selectivity was defined as the ratio between the relative reaction
rates observed between the two isomers.
As for alkylallylcarbonates, the solubility of both isomers
appeared to be the same since they were cleaved at the same
rate when no cyclodextrin was added to the catalytic system.
Nevertheless, the intrinsic solubility of urethanes was lower
than that of the carbonates since the latter were cleaved 10
times faster than the former (initial catalytic activity:
inverse phase transfer catalyst than HP-a-CD and HP-b-
CD, as already observed for biphenylmethylallylcarbonates.
A last series of experiments concerned the randomly
methylated cyclodextrins. The cleavage rates of N-dodecyl
and N,N-dihexyl-O-allylurethanes were compared using
Me-a-CD, Me-b-CD and Me-g-CD successively (Fig. 4).
Once again, the relative reaction rates were lower than those
measured for biphenylmethylallylcarbonates in the same
conditions, but the higher affinity of the cavity of Me-CDs
for lipophilic substrates still appeared clearly through the
comparison with the relative reaction rates obtained with
HP-CDs (for example, 150 with Me-a-CD vs. 23 with
HP-a-CD for the N-dodecyl isomer and 8 with Me-a-CD vs.
3.3 with HP-a-CD for the N,N-dihexyl isomer). Thus, the
hypothesis according which the surface active properties
and the presence of an extended hydrophobic cavity lead to
better rate reaction appeared valid. In particular, an
extended cavity allows to wrap more efficiently the linear
isomer than the bent isomer, inducing a rate increase more
2
1
21
0
.3 h vs. 0.03 h respectively).
When HP-CDs were used to transport the substrate from the
organic phase to the aqueous phase, the relative reaction
rates were all lower than those obtained with alkylallyl-
carbonates. Consequently, the CDs seemed to be more
appropriate to recognise biphenylmethylallylcarbonates
than N-alkyl-O-allylurethanes. In each case, the relative
reaction rates were always higher for N-dodecyl-O-allyl-
urethane than for N,N-dihexyl-O-allylurethane. As for the
previous bent structure of the ortho-biphenylmethylcarbo-
nate, the branched N,N-dihexyl-O-allylurethane could not
penetrate as deeply in the cavity of the CDs than the linear
N-dodecyl isomer. Both the length of the alkyl chains and
the steric hindrance of the non-included moiety of the
substrate were suspected to be responsible for this difference
of cleavage rates. It can also be noticed that in that case, the
discriminating power of HP-a-CD and HP-b-CD was the
same (substrate selectivities¼6.9) even if HP-b-CD led to
better relative reaction rates than HP-a-CD (30 vs. 23
respectively for the N-dodecyl isomer and 4.3 vs. 3.3
respectively for the N,N-dihexyl isomer). This surprising
result suggests that the flexible alkyl chains allow to
urethanes to easily adjust to the size of the cavity. This
assumption was partially confirmed by the result obtained
with the HP-g-CD. Indeed, the wider cavity of HP-g-CD is
always discriminant since a substrate selectivity of 5.3
was measured. In fact, the branched structure of the
N,N-dihexyl-O-allylurethane was not recognised at all by
the cavity of HP-g-CD (relative reaction rate¼1) whereas
the long flexible N-dodecyl chain was always able to
sufficiently interact with HP-g-CD to allow the substrate to
be carried to the aqueous phase. Whatever its potentialities
in terms of substrate selectivity, HP-g-CD remained a worse
Figure 5. Conversion of N-dodecyl-O-allylurethane and N,N-dihexyl-O-
allylurethane in function of time during a competition experiment in which
both substrates are present. Experimental conditions: Pd(OAc)
0.045 mmol), TPPTS (0.40 mmol), Me-b-CD (0.31 mmol), water (2 g),
2
(
N-dodecyl-O-allylurethane (0.56 mmol), N,N-dihexyl-O-allylurethane
(0.56 mmol), diethylamine (2.22 mmol, 160 mg) and toluene (2 g).

6492
C. Torque et al. / Tetrahedron 60 (2004) 6487–6493
pronounced for the linear isomer than for the bent isomer
and consequently explaining the higher discriminating
power of the methylated CDs relative to the hydroxy-
propylated CDs. Finally, this high discrimination power of
Me-b-CD was also confirmed by a competitive experiment.
In this experiment, the deprotection of a 50/50 mixture of
N,N-dihexyl-O-allylurethane and N-dodecyl-O-allyl-
urethane was carried out in the presence of Me-b-CD
indicated that the product was a mixture of TPPTS (ca.
98%) and its oxide (ca. 2%). Distilled deionized water was
used in all experiments. All catalytic reactions were
performed under nitrogen using standard Schlenk tech-
niques. All solvents and liquid reagents were degassed by
bubbling nitrogen for 15 min before each use or by two
freeze-pump-thaw cycles before use.
(
Fig. 5).
4.2. Catalytic experiments
As displayed in Figure 5, the N-dodecyl-O-allylurethane
was undoubtedly converted much faster than the N,N-di-
hexyl-O-allylurethane. In fact, a 96/4 product ratio was
observed in the initial stages of the reaction, confirming the
substrate selectivity obtained in the experiments where
urethanes were tested alone (24 vs. 20). Interestingly, the
curves profiles also indicated that no substrate or product
inhibition occured.
In a typical experiment, Pd(OAc) (0.045 mmol, 10 mg),
2
TPPTS (0.40 mmol, 227 mg), cyclodextrin (0.31 mmol) and
water (2 g) were introduced under nitrogen atmosphere into
a Schlenk tube. After stirring with a magnetic bar for 1 h,
the yellow solution was transferred into a mixture of the
substrate (1.12 mmol), diethylamine (2.22 mmol, 160 mg),
toluene (2 g) and dodecane as internal standard (0.10 g,
0.588 mmol). The medium was stirred at 1000 rpm at room
temperature and the reaction was monitored by quantitative
gas chromatographic analysis of the organic layer.
3. Conclusion
4.3. Competitive and control experiments
This study has clearly demonstrated that the reaction rate
and substrate selectivity strongly depended on the size and
chemical modification of the CD. The methylated CDs are
the most efficient CDs to perform substrate selective
reactions due to their surface active behaviour and to the
presence of a well defined hydrophobic cavity. Furthermore,
it appears that cavity size is a crucial factor to control the
substrate selectivity. For the tested substrates, the smaller
the cavity, the higher the substrate selectivity. This
conclusion is particularly true for rigid substrates. Thus,
a-CDs were more discriminant than b-CDs and much more
discriminant than g-CDs for biphenylmethylallylcarbonate
isomers. With urethanes containing flexible alkyl chains, the
discriminating power of a-CDs and b-CDs was found
equivalent and significantly higher than that of g-CDs.
Experiments are currently under way to examine the ability
of these methylated CDs to perform enantioselective
reactions.
The competitive experiments with a 1:1 mixture of ortho-
biphenylmethylallylcarbonate and para-biphenylmethyl-
allylcarbonate in the presence of Me-a-CD or Me-b-CD
were conducted as described in Section 4.2 except
that a mixture of ortho-biphenylmethylallylcarbonate
(0.56 mmol, 0.15 g) and para-biphenylmethylallylcarbo-
nate (0.56 mmol, 0.15 g) was used. In the control experi-
ment, 0.4 g of acetonitrile was added to the reaction medium
in the place of Me-a-CD or Me-b-CD.
The competitive experiments with a 1:1 mixture of
N-dodecyl-O-allylurethane and N,N-dihexyl-O-allyl-
urethane was conducted as described in the in Section 4.2
except that a mixture of N-dodecyl O-allylurethane
(0.56 mmol, 0.15 g) and N,N-dihexyl-O-allylurethane
(0.56 mmol, 0.15 g) was used. In the control experiment,
2 g of acetonitrile was added to the reaction medium in the
place of Me-b-CD.
4. Experimental
4
.1. Materials
References and notes
HP-a-CD and HP-b-CD were obtained from Aldrich
Chemical Co. and was used as received without further
purification. HP-g-CD was a generous gift of Wacker
Chemie Co. These HP-CDs were native CDs partially
O-2-hydroxypropylated with statistically 0.8 OH groups
modified per glucopyranose unit. Me-b-CD was purchased
from Aldrich Chemical Co. The Me-a-CD and Me-g-CD
were prepared by adapting a procedure reported by
1. For reviews see: (a) Wenz, G. Angew. Chem., Int. Ed. Engl.
1994, 33, 803. (b) Saenger, W. Angew. Chem., Int. Ed. Engl.
1980, 19, 344. (c) Szejtli, J. J. Mater. Chem. 1997, 7, 575.
(d) Szejtli, J. Chem. Rev. 1998, 98, 1743.
2. For reviews see: (a) Takahashi, K. Chem. Rev. 1998, 98, 2013.
(b) Osa, T.; Suzuki, I. In Reactivity of Included Guests.
Comprehensive Supramolecular Chemistry; Atwood, J. L.,
Davies, J. E. D., MacNicol, D. D., V o¨ gtle, F., Eds.; Pergamon:
Oxford, 1996; Vol. 3, pp 367–400. (c) Breslow, R. Acc. Chem.
Res. 1991, 24, 317. (d) Takahashi, K.; Hattori, K. J. Incl.
Phenom. 1994, 17, 1.
1
2
Y. Kenichi et al. These CDs were partially methylated;
statistically 1.8 OH groups per glucopyranose unit were
modified. Palladium acetate and organic compounds were
purchased from Strem Chemicals, Aldrich Chemical Co.
and Acros Organics in their highest purity and used without
further purification. Trisodium tris(m-sulfonatophenyl)-
phosphine ((P(C H SO Na) ; TPPTS)) was synthesized as
3. See for examples: (a) Fornasier, R.; Reniero, F.; Scrimin, P.;
Tonellato, U. J. Org. Chem. 1985, 50, 3209. (b) Schneider,
H. J.; Sangwan, N. K. Angew. Chem., Int. Ed. 1987, 26, 896.
(c) Hattori, K.; Takahashi, K.; Sakai, N. Bull. Chem. Soc. Jpn
1992, 65, 2690. (d) Reddy, M. A.; Surenda, K.; Bhanumathi,
N.; Rao, K. R. Tetrahedron 2002, 58, 6003. (e) Pattekhan,
6
4
3
3
3
1
reported by G a¨ rtner et al. The purity of the TPPTS was
1
carefully controlled. In particular, P{ H} solution NMR
3
1

C. Torque et al. / Tetrahedron 60 (2004) 6487–6493
H. H.; Divakar, S. J. Mol. Catal. A: Chem. 2002, 184, 79.
6493
J. P.; Blart, E.; Savignac, M.; Lemeune, S.; Lemaire-Audoire,
S.; Paris, J. M.; Bernard, J. M. Tetrahedron 1994, 50, 497.
(d) Lemaire-Audoire, S.; Savignac, M.; Pourcelot, G.; Gen eˆ t,
J. P.; Bernard, J. M. J. Mol. Catal. 1997, 116, 247.
8. (a) Gen eˆ t, J. P.; Savignac, M.; Lemaire-Audoire, S. Chemistry
for the 21st century. IUPAC Monographs. In Transition Metal
Catalyzed Reactions; Murahashi, S. I., Davies, S. G., Eds.;
Academic: New York, 1999; pp 55–79. (b) Gen eˆ t, J. P.;
Savignac, M. J. Organomet. Chem. 1999, 576, 305. (c) Sinou,
D. Aqueous-Phase Organometallic Catalysis. In Allylic
Substitution; Cornils, B., Hermann, W. A., Eds.; Wiley–
VCH: Weinheim, Germany, 1998; pp 401–407.
(
f) Veglia, A. V.; deRossi, R. H. J. Org. Chem. 1988, 53, 5281.
g) Guy, A.; Doussot, J.; Dallamaggiore, A.; Garreau, R. Can.
(
J. Chem. 1995, 73, 599–603.
4
. See for examples: (a) Shiraishi, Y.; Kojima, S.; Tomita, H.;
Ohsuka, H.; Kawamura, T.; Toshima, N.; Hirai, H. Polym. J.
1
996, 28, 619. (b) Ravi, P.; Divakar, S. J. Incl. Phenom. 2001,
9, 27. (c) Shiraishi, Y.; Toshima, N.; Kawamura, T.; Mihori,
3
H.; Shirai, H.; Hirai, H. J. Mol. Catal. 1999, 139, 149.
d) Reddy, L. R.; Reddy, M. A.; Bhanumathi, N.; Rao, K. R.
(
Synlett 2000, 339. (e) Fornasier, R.; Marcuzzi, F.; Parmagnani,
M.; Tonellato, U. Carbohydr. Res. 1991, 217, 245.
(
f) Fornasier, R.; Marcuzzi, F.; Tonellato, U. J. Incl. Phenom.
994, 18, 81. (g) Wang, J. T.; Feng, X.; Tang, L. F.; Li, Y. M.
9. (a) Hapiot, F.; Lyskawa, J.; Bricout, H.; Tilloy, S.; Monflier, E.
Adv. Synth. Catal. 2004, 346, 8. (b) Uekama, K.; Irie, T.
Cyclodextrins and their industrial use. In Pharmaceutical
Applications of Methylated Cyclodextrin derivatives;
Duch eˆ ne, D., Ed.; Editions de la Sant e´ : Paris, 1987; pp
395–439.
1
Polyhedron 1996, 15, 2997. (h) Tee, O. S.; Javed, B. C.
J. Chem. Soc., Perkin Trans. 2 1994, 23. (i) Chung, W. S.;
Turro, N. J.; Silver, J.; le Noble, W. J. J. Am. Chem. Soc 1990,
1
12, 1202.
. For reviews see: (a) Brelow, R.; Dong, S. Chem. Rev. 1998,
997. (b) Komiyama, M. In Cyclodextrins as Enzyme Models.
5
10. (a) Tanaka, N.; Yamaguchi, A.; Araki, Y.; Araki, M. Chem.
Lett. 1987, 715. (b) Czugler, M.; Ecke, E.; Stzzowski, J. J.
J. Chem. Soc., Chem. Commun. 1981, 1294.
1
Comprehensive Supramolecular Chemistry; Atwood, J. L.,
Davies, J. E. D., MacNicol, D. D., V o¨ gtle, F., Eds.; Pergamon:
Oxford, 1996; Vol. 3, pp 401–422. (c) Breslow, R. Acc. Chem.
Res. 1995, 28, 146.
11. (a) Lipkowitz, K. B. Chem. Rev. 1998, 98, 1829.
(b) Rekharsky, M. V.; Inoue, Y. Chem. Rev. 1998, 98, 1875.
(c) Fenyvesi, E.; Szente, L.; Russell, N. R.; McNamara, M. In
Specific guest types; Atwood. Comprehensive Supramolecular
Chemistry; Atwood, J. L., Davies, J. E. D., MacNicol, D. D.,
V o¨ gtle, F., Eds.; Pergamon: Oxford, 1996; Vol. 3, pp
305–366.
6
. (a) Lacroix, T.; Bricout, H.; Tilloy, S.; Monflier, E. Eur. J. Org.
Chem. 1999, 3127. (b) Bricout, H.; Caron, L.; Bormann, D.;
Monflier, E. Catal. Today 2001, 66, 355. (c) Cabou, J.;
Bricout, H.; Hapiot, F. Catal. Commun. 2004, 5, 265–270.
. (a) Gen eˆ t, J. P.; Blart, E.; Savignac, M.; Lemeune, S.; Paris,
J. M. Tetrahedron Lett. 1993, 34, 4189. (b) Lemaire-Audoire,
S.; Savignac, M.; Blart, E.; Pourcelot, G.; Gen eˆ t, J. P.;
Bernard, J. M. Tetrahedron Lett. 1994, 35, 8783. (c) Gen eˆ t,
7
12. Kenichi, Y.; Atsushi, M.; Yukio, T.; Mitsukatsu, S.; Yoshiaki,
Y.; Tomoyuki, I. JP Patent 8333406, 1996.
13. G a¨ rtner, R.; Cornils, B.; Springer, H.; Lappe, P. DE Patent
3235030, 1982.