New phosphane based on a β-cyclodextrin, exhibiting a solvent-tunable conformation, and its catalytic properties

Article abstract of DOI:10.1002/chem.201000379

A new diphenylphosphane based on a β-cyclodextrin skeleton that exhibits a dual solubility in water and in organic solvent was synthesised. Interestingly, a solvent-dependent conformation change was evidenced by NMR spectroscopy studies; the self-inclusion of a phenyl group of the phosphane moiety into cyclodextrin cavity observed in water disappeared in organic solvents due to a change in conformation. Hydrogenation or hydroformylation reactions performed in water and in organic solvents showed that this ligand was able to stabilise catalytically active rhodium species in solution. In the case of the hydroformylation reaction, it was demonstrated that regioselectivity was influenced by the solventdependent conformation of the ligand.

Full text of DOI:10.1002/chem.201000379

FULL PAPER
DOI: 10.1002/chem.201000379
New Phosphane Based on a b-Cyclodextrin, Exhibiting a Solvent-Tunable
Conformation, and its Catalytic Properties
Cꢀcile Machut-Binkowski,[a, b, d] FranÅois-Xavier Legrand,[a, b, d] Nathalie Azaroual,[a, c]
Sꢀbastien Tilloy,*[a, b, d] and Eric Monflier[a, b, d]
Abstract: A new diphenylphosphane
based on a b-cyclodextrin skeleton that
exhibits a dual solubility in water and
in organic solvent was synthesised. In-
terestingly, a solvent-dependent confor-
mation change was evidenced by NMR
spectroscopy studies; the self-inclusion
of a phenyl group of the phosphane
moiety into cyclodextrin cavity ob-
served in water disappeared in organic
solvents due to a change in conforma-
tion. Hydrogenation or hydroformyla-
tion reactions performed in water and
in organic solvents showed that this
ligand was able to stabilise catalytically
active rhodium species in solution. In
the case of the hydroformylation reac-
tion, it was demonstrated that regiose-
lectivity was influenced by the solvent-
dependent conformation of the ligand.
Keywords: cyclodextrins · hydrofor-
mylation · hydrogenation · self-in-
clusion · water chemistry
Introduction
theless, only some of these ligands have been used during
organometallic catalytic processes either in organic media or
in aqueous media. In organic media, asymmetric hydrogena-
tion of various prochiral substrates was performed by using
a diphosphite-a-CD (dimethyl itaconate; ee 84%),[4] a di-
phosphane-b-CD (derivatives of cinnamic acid or ester,
acrylic acid or ester and itaconic acid or ester; ee 49–
92%),[24] and a tetraphoshane-a-CD (amino acid precursor,
ee 25%) as ligands.[21] A diphosphite-a-CD[4] and a mono-
phosphinite-b-CD[3] were exploited in hydroformylation re-
actions of aryl and alkyl olefins. A diphosphane-a-CD was
also used in an ethene and propene dimerisation reaction.[16]
In aqueous media, rhodium species based on diphosphane-
b-CD were active in the hydrogenation of nitroaromatics[12]
and in hydrogenation and hydroformylation of aryl and
alkyl olefins.[7,10,11] Alkyl olefins were also hydroformylated
by using diphosphane-a-CD as a ligand.[13] To the best of
our knowledge, only one diphenyl monophosphane b-CD
has been evaluated as a water-soluble ligand in the hydroge-
nation of olefins.[7] Nevertheless, hydrogenation occurred
only under forcing conditions and led to the formation of
colloidal rhodium, which then acted as a poor catalyst.
Cyclodextrins (CDs) are cyclic glucose oligomers that can
form inclusion complexes with numerous organic com-
pounds and so have found numerous applications in various
fields, such as analytical chemistry, agrochemistry, pharma-
ceuticals, foods, cosmetics and catalysis.[1] In this last field,
CDs can be used as a macrocyclic platform to build ligands
for organometallic catalytic processes.[2] Several CD-func-
tionalised phosphorous ligands and their organometallic
complexes have been described in the literature.[3–24] Never-
[a] Dr. C. Machut-Binkowski, F.-X. Legrand, Prof. Dr. N. Azaroual,
Prof. Dr. S. Tilloy, Prof. Dr. E. Monflier
UniversitꢀLille Nord de France
59000 Lille, France
[b] Dr. C. Machut-Binkowski, F.-X. Legrand, Prof. Dr. S. Tilloy,
Prof. Dr. E. Monflier
UArtois, Rue Jean Souvraz, SP 18
62307 Lens Cedex (France)
Fax : (+33)321-791-755
E-mail: sebastien.tilloy@univ-artois.fr
[c] Prof. Dr. N. Azaroual
As an original extension of these works, we report the
synthesis of a new diphenyl monophosphane b-CD that ex-
hibits a dual solubility in both organic solvents and water.
An interesting property of solvent-dependent conformation
change was evidenced by NMR spectroscopic studies. Hy-
drogenation or hydroformylation reactions performed in
water or in organic solvents showed that this ligand was able
UDSL, Facultꢀ de Pharmacie
EA GRIIOT, 59006 Lille (France)
[d] Dr. C. Machut-Binkowski, F.-X. Legrand, Prof. Dr. S. Tilloy,
Prof. Dr. E. Monflier
CNRS, UMR 8181
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000379.
Chem. Eur. J. 2010, 16, 10195 – 10201
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10195

to stabilise catalytically active rhodium species in solution.
In the case of the hydroformylation reaction, it was demon-
strated that regioselectivity was influenced by the solvent-
dependent conformation of the ligand.
Results and Discussion
The
monophosphane
6A-deoxy-6A-diphenylphosphinyl-
2A,2B,2C,2D,2E,2F,2G,3A,3B,3C,3D,3E,3F,3G,6B,6C,6D,6E,6F,6G-eicosa-
O-methyl-b-cyclodextrin (permethylated b-cyclodextrin
monodiphenyl phosphane: PM-b-CD-PPh2) was prepared in
three steps from b-CD (Scheme 1). b-CD was treated with
tosyl chloride in basic medium to give b-CD-OTs (yield
Figure 1. 31P{1H} NMR spectrum (293.15 K; 121.49 MHz; 1 mm in D2O)
of PM-b-CD-PPh2. H3PO4 was used as external phosphorous reference.
Hb’, Hc’ and Hd’ can be distin-
guished). This phenomenon was
explained by T-ROESY experi-
ments. Indeed, the T-ROESY
NMR spectra clearly revealed
dipolar contacts between the
CD and phosphane moieties
(Figure 3). More precisely, the
strongest contacts were ob-
served between the protons of
one aromatic ring (Hb and Hc)
and the inner CD protons (H3
and H5). This experiment
showed that a phenyl group of
the phosphane moiety is includ-
ed into the CD cavity. Addi-
Scheme 1. Synthetic pathway of the cyclodextrin-based phosphane: PM-b-CD-PPh2. Reagents and conditions:
i) TsCl, NaOH, H2O, 2 h at RT then HCl (6n), 24 h at 48C, 19%; ii) CH3I, NaH, anhyd DMF, 6 h at 08C then
12 h at RT, 50%; iii) KPPh2, anhyd DMF/THF, 18 h at 658C, 85%.
tional evidence was found by a
1H NMR spectroscopic analysis
performed in [D6]DMSO The
spectral dispersion of aromatic
protons observed in water col-
19%),[25] which was permethylated in the presence of
methyl iodide and NaH to give PM-b-CD-OTs (yield
50%).[26] PM-b-CD-OTs was then treated with KPPh2
(2 equiv) in a DMF/THF mixture. After purification by
column chromatography under N2, PM-b-CD-PPh2 was ob-
tained as a white amorphous solid (yield 85%).
lapses and only one large peak is observed in DMSO
(Figure 4). The strongly solvating power of DMSO for the
CD cavity is known to preclude the formation of inclusion
complexes.[27,28] These two last experiments undoubtedly
show that an inter- or intramolecular complexation phenom-
enon occurs leading to a distinction between the two phenyl
groups.
PM-b-CD-PPh2 was first characterised by matrix-assisted
laser desorption/ionisation time-of-flight mass spectrometry
(MALDI-TOF MS). The MS spectrum showed the expected
molecular peaks (see the Supporting Information). The first
NMR spectroscopic characterisations were performed in
pure D2O. The 31P NMR spectrum of PM-b-CD-PPh2
showed only one peak at d=ꢀ24.3 ppm, which indicated
that no phosphane oxide was present (Figure 1).
To provide a distinction between inter- or intramolecular
complexes, NMR spectroscopic analyses were performed in
aqueous medium at various concentrations (in the range of
1
0.5 to 3 mm). No displacement of H or 31P NMR signals was
observed during these dilution experiments (see the Sup-
porting Information). In addition, temperature increase (up
to 508C) did not substantially change the spectral character-
1
1
The H and 13C NMR spectroscopic assignments were per-
istics of H or 31P NMR signals (see the Supporting Informa-
formed by using one and two-dimensional high-resolution
experiments (see Figure 2 and the Supporting Information).
tion). All these observations are in line with the formation
of a self-inclusion phenomenon. It is important to underline
that a such phenomenon has already been observed for CD
possessing a pendant substituent.[8,27–29]
1
Surprisingly, the H and 13C NMR spectra showed that the
two phenyl groups are non-equivalent (protons Hb, Hc, Hd,
10196
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10195 – 10201

Catalytic Phosphanes
FULL PAPER
termined by NMR spectroscop-
ic studies in aqueous medium
and for comparison it was also
determined with Trime-b-CD (a
permethylated b-CD). Indeed,
ACNa is included in the Trime-
b-CD cavity by the secondary
face (wide side). Because the
secondary face of PM-b-CD-
PPh2 is similar to that of Trime-
b-CD, this latter is a good con-
trol. The values of association
constants with ACNa are equal
to 10700mꢀ1 in the case of
Trime-b-CD and 265mꢀ1 in the
case of PM-b-CD-PPh2. This
lower value signifies that the
self-inclusion complex is very
strongly stabilised. So, the occu-
pation of the cavity by this aro-
matic moiety will prevent the
formation of stable inclusion
Figure 2. 1H NMR spectrum (295.15 K; 500.13 MHz; 1 mm in D2O) of PM-b-CD-PPh2 showing the dispersion
of the aromatic and anomeric protons.
complexes
guests.
with
numerous
PM-b-CD-PPh2 is also solu-
ble in various solvents (MeOH,
EtOH, CHCl3, CH2Cl2, THF,
DMF, cyclohexane and hep-
tane). Because the main driving
forces of self-inclusion phenom-
ena are due to hydrophobic ef-
fects, it seemed interesting to
study this potential self-inclu-
sion in organic medium. A T-
ROESY experiment was per-
formed on PM-b-CD-PPh2 in
C6D12 and no dipolar contact
was observed (see the Support-
ing Information). This result
suggests that the two phenyl
groups are outside the CD
cavity. The self-inclusion phe-
nomenon observed in water dis-
appears in organic solvents, and
so the conformation of PM-b-
CD-PPh2 depends on the sol-
vent. To illustrate the PM-b-
CD-PPh2 arrangement in each
solvent, a schematic representa-
tion is depicted in Figure 5.
Figure 3. Partial contour plot of a T-ROESY experiment (295.15 K; 500.13 MHz; 1 mm in D2O; spin lock time
500 ms) performed on PM-b-CD-PPh2. Vertical scale: aromatic area, horizontal scale: CD area. See Figure 2
for the atom labelling system.
To estimate the strength of the self-inclusion complex, at-
tempts were made to include an external host in the PM-b-
CD-PPh2 cavity. 1-Adamantanecarboxylate sodium salt
(ACNa) was selected because of its high affinity for the b-
CD cavity. Indeed, the value of association constant for the
b-CD/ACNa complex was equal to 30000mꢀ1.[30] The associ-
ation constant between ACNa and PM-b-CD-PPh2 was de-
This interesting property of dual solubility could allow or-
ganometallic catalytic reactions to be performed in aqueous,
organic, or aqueous/organic biphasic media. Because PM-b-
CD-PPh2 is soluble in water as much as in heptane (3 mm)
and because heptane is often used as the organic phase
during aqueous biphasic organometallic catalytic processes,
the distribution of PM-b-CD-PPh2 in a mixture of heptane/
Chem. Eur. J. 2010, 16, 10195 – 10201
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10197

S. Tilloy et al.
mation). It is important to un-
derline that this substrate does
not form an inclusion complex
with PM-b-CD-PPh2, which in-
dicates that this latter behaves
only as a ligand for the transi-
tion metal.
For the catalytic system [Rh-
AHCTUNGTREGUN(NN cod)2]BF4/PM-b-CD-PPh2, the
conversion was low at 208C and
it was necessary to increase the
temperature up to 608C to
reach high conversion after 4 h
(Table 1, entries 1 and 2). When
[RhACHTNUTGRNEU(GN acac)(CO)2] was used as a
Figure 4. 1H NMR spectra (293.15 K; 300.13 MHz; 1 mm) showing the aromatic and anomeric regions of PM-
b-CD-PPh2 in water (top) and DMSO (bottom).
catalytic precursor, the conver-
sion was four times lower
(Table 1, entry 3). The increase
in PM-b-CD-PPh2/Rh ratio de-
creased the conversion; indeed the catalytic species are the
most stable and, therefore, the least active (Table 1,
entry 4). In each case, it is important to underline that the
aqueous catalytic phase stayed yellow and that no colloidal
rhodium particles were observed during or at the end of the
reaction. These observations indirectly prove that PM-b-
CD-PPh2 is able to stabilise rhodium species in aqueous so-
lution.
Figure 5. Schematic representation of the PM-b-CD-PPh2 conformation
as a function of solvent.
We were then interested in the performance of PM-b-CD-
PPh2 in the hydroformylation of methyl 4-pentenoate in
pure water (Table 2). The hydroformylation reaction was
chosen as a model reaction because it is able to provide in-
teresting information concerning activity, chemoselectivity
and regioselectivity. By using NMR spectroscopy studies, it
was demonstrated that methyl 4-pentenoate is not able to
form an inclusion complex with PM-b-CD-PPh2 (see the
water was determined. The partition coefficient (Pow) is
equal to 0.48, that is, 68% of PM-b-CD-PPh2 stays in water
and 32% moves to the organic layer. These values are not
in accordance with the specifications required for an appli-
cation in biphasic aqueous organometallic catalytic process-
es because the water-soluble catalyst does not remain in the
aqueous layer, which prevents recycling. So, PM-b-CD-PPh2
can be used as a ligand in pure water or in pure organic sol-
vent, but not in a mixture of the two.
Table 2. Rhodium-catalysed hydroformylation of methyl 4-pentenoate in
We first studied the behaviour of PM-b-CD-PPh2 during
the hydrogenation reaction of 2-methyl-3-buten-2-ol in pure
water (Table 1). The potential interactions between 2-
methyl-3-buten-2-ol and PM-b-CD-PPh2 were studied by
using NMR spectroscopy in water (see the Supporting Infor-
the presence of PM-b-CD-PPh2 as the ligand.[a]
Table 1. Rhodium-catalysed hydrogenation of 2-methyl-3-buten-2-ol in
the presence of PM-b-CD-PPh2 as ligand.[a]
Solvent
L/Rh
PACTHNGUTERNNUG
(CO/H2) Conversion[b] Selectivity[c] l/b
[bar]
[%]
[%]
ratio[d]
1
2
3
4
5
6
water
water
water
heptane
heptane
heptane
4
8
4
4
8
4
50
50
25
50
50
25
96
97
89
83
82
88
99
99
99
99
98
98
1.8
1.8
1.9
1.2
1.2
1.2
Catalytic precursor
[Rh(cod)2]BF4
[Rh(cod)2]BF4
[Rh(acac)(CO)2]
[Rh(cod)2]BF4
L/Rh
T [8C]
Conversion[b] [%]
1
2
3
4
N
2
2
2
4
20
60
60
60
6
97
18
63
[a] Experimental conditions: [RhACHTUNGTRNEUNG
(acac)(CO)2] (1.94ꢂ10ꢀ3 mmol); L: PM-
b-CD-PPh2; solvent (6 mL); methyl 4-pentenoate (0.97 mmol); 1500 rpm;
T=808C; time: 2 h. [b] Olefin conversion (calculated with respect to the
starting olefin). [c] Aldehyde selectivity, i.e., (mol of aldehydes)/(mol of
converted olefins)ꢂ100. The side products were mainly isomeric olefins.
[d] Ratio of linear to branched aldehyde product.
[a] Experimental conditions: [RhACTHNUTRGENNUG(cod)2]BF4 or [RhACHTUNGTERN(NUGN acac)(CO)2] (7.55ꢂ
10ꢀ3 mmol), L: PM-b-CD-PPh2; water (12 mL), 2-methyl-3-buten-2-ol
(1.89 mmol), 1250 rpm, P(H2)=1 atm; t=4 h. [b] Olefin conversion (cal-
culated with respect to the starting olefin).
10198
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10195 – 10201

Catalytic Phosphanes
FULL PAPER
identified by using UV fluorescence and/or staining with a solution of
phosphomolybdic acid in aqueous sulfuric acid and EtOH. Characterisa-
tion and structure determinations were achieved by NMR spectroscopy
experiments by using a Bruker Avance DRX300 spectrometer operating
at 300.13 MHz for 1H nuclei, 75.47 MHz for 13C nuclei and 121.49 MHz
for 31P nuclei, or by using a Bruker Avance 500 spectrometer operating
at 500.13 MHz for 1H nuclei and 125.76 MHz for 13C nuclei. 1D and 2D
NMR spectroscopy experiments were obtained by using the pulse pro-
grams available from the Bruker library. Details concerning experimental
conditions are given in the Figure captions. All NMR measurements
were performed under careful temperature regulation by using a Bruker
BVT variable temperature unit. Chemical shifts are given in parts per
million (ppm) relative to an external reference by using internal capillary
(sodium salt of 3-(trimethylsilyl)-2,2,3,3-tetradeuteropropionic acid (98%
atom D) in D2O for 1H and 13C NMR and H3PO4 in H2O for 31P NMR)
and calibration was performed by using the signal of the residual signals
of the solvent as a secondary reference while taking into account temper-
Supporting Information). When PM-b-CD-PPh2 was used as
a ligand (Table 2, entry 1), high conversion was reached
after 2 h. The aldehyde selectivity was 99%, which shows
that no isomerisation or hydrogenation of the double bound
occurred. The ratio of linear to branched aldehydes (l/b)
was 1.8. When the PM-b-CD-PPh2/Rh ratio was increased
from 4 to 8 (Table 2, entries 1 and 2) or the CO/H2 pressure
was decreased from 50 to 25 bar (Table 2, entries 1 and 3),
no remarkable change was observed. It was interesting to
evaluate the catalytic properties of Rh/PM-b-CD-PPh2 in or-
ganic medium because PM-b-CD-PPh2 possesses a dual sol-
ubility. Heptane was chosen as the reaction solvent and
methyl 4-pentenoate as the substrate because it is miscible
with heptane, and experiments similar to those described
above were performed in heptane (compare Table 2, en-
tries 1–3 and 4–6). The conversions were slightly lower in
heptane, but the chemoselectivities were almost unchanged.
However, the l/b ratio decreased from 1.8 in water to 1.2 in
heptane. To understand this phenomenon, experiments were
performed in heptane with triphenylphosphane (TPP) as the
ligand, and in water with trisulfonated triphenylphosphane
(TPPTS) as the ligand. After a reaction time of 2 h, high
conversions (>98%) and selectivities (>99%) were
reached with a l/b ratio of 1.8 in the presence of each phos-
phane. These results suggest that the l/b ratio decrease was
not due to a solvent effect. One hypothesis could be a modi-
fication of the conformation of PM-b-CD-PPh2. Indeed, the
self-inclusion phenomenon observed in water disappears in
organic solvents and, therefore, the conformation of PM-b-
CD-PPh2 is different in each solvent. PM-b-CD-PPh2 is rigid
in water, whereas it is more flexible in heptane because the
two phenyl groups are outside the CD cavity.
ature effects. The MALDI-TOF mass spectra were recorded on
a
MALDI-TOF-TOF Bruker Daltonics Ultraflex II spectrometer in posi-
tive reflectron mode by using 2,5-dihydroxybenzoic acid as a matrix and
external peptide calibration standard kit (Bruker Daltonics) within a
mass range of 750–4200. The acceleration voltage was fixed at 25 keV,
the delayed extraction time at 10 ns and the number of laser shots at 200.
The samples were dissolved either in H2O, acetone or MeOH and equally
mixed with the matrix solution (10 mgmLꢀ1 of 2,5-dihydroxybenzoic acid
(2,5-DHB) in H2O/0.1% TFA/MeCN, 70:30 (v/v)) and spotted onto a
ground-style MALDI target according to the dried droplet method. The
UV/Vis experiments were performed by using
a Perkin–Elmer
Lambda 19 spectrophotometer at ambient temperature (293.15 K) with a
10 mm quartz cell. Gas chromatographic analyses were carried out on a
Shimadzu GC-17 A gas chromatograph equipped with a methyl silicone
capillary column (30 mꢂ0.32 mm) and a flame ionisation detector.
Synthesis of 6A-deoxy-6A-diphenylphosphinyl-2A,2B,2C,2D,2E,2F,2G,3A,3B,-
3C,3D,3E,3F,3G,6B,6C,6D,6E,6F,6G-eicosa-O-methyl-b-cyclodextrin: 6A-O-(p-
tolylsulfonyl)-2A,2B,2C,2D,2E,2F,2G,3A,3B,3C,3D,3E,3F,3G,6B,6C,6D,6E,6F,6G-eicosa-
O-methyl-b-cyclodextrin was obtained in two steps from the native b-cy-
clodextrin, as previously described in the literature.[25,26] A solution of po-
tassium phosphide (KPPh2) in THF (0.5m, 6 mL, 3 mmol) was added
under an inert atmosphere to a stirred solution of dried 6A-O-(p-tolylsul-
fonyl)-2A,2B,2C,2D,2E,2F,2G,3A,3B,3C,3D,3E,3F,3G,6B,6C,6D,6E,6F,6G-eicosa-O-
methyl-b-cyclodextrin (2.3 g, 1.47 mmol) in dry DMF (15 mL) and freshly
distilled THF (20 mL). The reaction mixture was stirred for 18 h at 658C
under N2, then degassed, deionised (30 mL) H2O was added. The product
was extracted under N2 with degassed EtOAc (4ꢂ30 mL). After evapora-
tion of most of the solvent, the residual syrup was then purified by
column chromatography on silica gel under N2 by using EtOAc as the
eluent to give the title product as a white amorphous solid (2.0 g, 85%).
Conclusion
PM-b-CD-PPh2 is a valuable ligand for organometallic cata-
lytic processes performed in water or organic solvents. In
addition, the conformation of this ligand is solvent-tunable,
and this interesting property opens the way to specific selec-
tivity during organometallic processes.
1H NMR (500.13 MHz, D2O, 295.15 K, TMSP): d=7.71 (t, 3J
7.3 Hz, 1H; Hd), 7.69–7.61 (m, 3H; Hc’, Hd’), 7.61–7.54 (m, 4H; Hb’, Hc),
ACHTUNGTRENNUNG(Hc,Hd)=
3
7.35 (t, 3J
1H; H1G), 5.31–5.28 (m, 2H; H1B, H1C), 5.20 (d, 3J
H1F), 5.17 (d, 3J(H1,H2)=3.2 Hz, 1H; H1A), 5.11 (d, 3J
1H; H1E), 5.02 (d, 3J(H1,H2)=3.5 Hz, 1H; H1D), 4.38 (dd, 3J
4 Hz, 2J(H6,H6’)=12 Hz, 1H; H6’B), 4.31–4.23 (m, 1H; H5B), 4.08 (dd, 3J-
G
R
ACHTUGNTRNEN(UGN H1,H2)=4 Hz,
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
E
ACHTUNGTRENNUNG
Experimental Section
AHCTUNGTRENNUNG
ACHUTNGRENNUG CAHTUNGTRENNUGN
(H5,H6’)=3.5 Hz, 2J(H6,H6’)=11 Hz, 1H; H6’C), 3.98–3.15 (m, 93H;
H2AꢀG, H3AꢀF, H4AꢀG, H5A, H5C, H5E–G, H6B–D, H6F–G, H6’D, H6’F–G, Me2
,
AꢀG
General: The starting b-cyclodextrin was a generous gift from Roquette
Frꢃres (Lestrem, France). Most of the chemical products, reagents and
solvents used in this study were purchased from Acros Organics and
Sigma–Aldrich in their highest purity and used without further purifica-
tion. Catalytic precursors, heptakis(2,3,6-tri-O-methyl)-b-cyclodextrin and
deuterated solvents were purchased respectively from Strem Chemicals
(Bischheim, France), Cyclolab (Budapest, Hungary) and Euriso-Top (Gif
sur Yvette, France) in their highest purity and used without further pu-
rification. Distilled water was used in all experiments. The sodium salt of
TPPTS was synthesised as previously reported.[31] CO/H2 mixture (1:1)
and H2 were used directly from cylinders (>99.9% pure; Air Liquide).
Analytical thin-layer chromatography plates (TLC silica gel 60 F254 alumi-
nium) and silica (Geduranꢄ Si 60 (0.063–0.200 mm)) for preparative
column chromatography were purchased from Merck. Compounds were
AꢀG
Me3
A
,
Me6B–G), 3.12 (d, 2J
(H6,H6’)=11 Hz, 1H; H6’E), 2.95 (t, 3J-
(H2,H3)ꢁ J
(H3,H4)=9.5 Hz, 1H; H3G), 2.92–2.86 (m, 1H; H5D), 2.65 (d,
ACHTUNGTRENNUNG
3
2J(H6,H6’)=11 Hz, 1H; H6E), 2.51 (dd, 2J(P,H6’)=5.7 Hz, 2J
ACHTUGNTERNNNUG AHCTNUTREGNNUGN ACHTUNGTRENNUNG(H6,H6’)=
13.5 Hz, 1H; H6’A), 2.22–2.10 ppm (m, 1H, H6A); 13C NMR (125.76 MHz,
D2O, 295.15 K, TMSP): d=141.2 (Ca), 139.0 (Ca’), 135.2 (Cb), 134.3 (Cb’),
132.6 (Cd), 132.0 (Cd’), 131.7 (Cc’), 131.3 (Cc), 100.8 (C1E), 100.7 (C1D),
100.6 (C1B, C1C, C1F), 99.5 (C1G), 98.5 (C1A), 84.4 (C3A), 83.7 (C3C, C3G),
83.6 (C3D), 83.2 (C2A), 83.1 (C2 or C2E, C3F), 83.0 (C4G), 82.9 (C3B), 82.8
D
(C3E), 82.7 (C2C, C4C), 82.4 (C2 or C2E), 82.3 (C2F), 82.1 (C2G), 82.0 (C4F),
D
81.8 (C2B), 81.2 (C4B), 80.9 (C4E), 80.5 (C4A), 78.8 (C4D), 74.0 (C5G), 73.6
(C5C), 73.5 (C5B), 73.2 (C5F), 73.0 (C6B, C6D), 72.8 (C6G), 72.6 (C5E), 71.9
(C5D), 72.7 (C6C), 71.5 (C6E), 71.2 (C6F), 69.3 (C5A), 63.9 (Me3G), 63.5
(Me3C), 63.3 (Me3B), 63.2 (Me3F), 62.4 (Me3E), 62.0 (Me2G, Me3D), 61.4
Chem. Eur. J. 2010, 16, 10195 – 10201
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10199

S. Tilloy et al.
D
(Me2C), 61.1 (Me6B), 61.0 (Me3A), 60.9 (Me6G), 60.7 (Me6C–F), 60.4 (Me2
reactor through a high-pressure regulator valve, which allowed the pres-
sure in the reactor to be kept constant throughout the whole reaction.
The reaction medium was sampled after 2 h for GC analyses of the or-
ganic phase after decantation. For kinetic measurements, the time corre-
sponding to the addition of CO/H2 was considered to be the beginning of
the reaction.
or Me2E), 60.0 (Me2A), 59.9 (Me2B), 59.8 (Me2 or Me2E, Me2F), 33.5 ppm
(C6A); 31P{1H} NMR (121.49 MHz, D2O, 293.15 K, TMSP): d=ꢀ24.3 ppm
(s); UV/Vis (H2O, 293.15 K): lmax (loge)=247 nm (3.79); MALDI-TOF-
MS: m/z calcd for [C74H119O34P+H]+: 1583.74; found: 1583.64; m/z calcd
for [C74H119O34P+Na]+: 1605.72; found: 1605.50; m/z calcd for
[C74H119O34P+K]+: 1621.70; found: 1621.46.
D
Determination of the partition coefficient: The determination of partition
coefficient of the cyclodextrin-based phosphane in a heptane/water mix-
ture was determined according to the shake-flask method. Both solvents
were mutually saturated before the measurement, which was carried out
at a concentration of 0.1m and for a heptane/water ratio equal to 1:1.
The samples were hand-shaken and stirred with a magnetic stirrer for
15 min, then left to equilibrate for 24 h at RT. The amount of PM-b-CD-
PPh2 in each phase was determined by UV/Vis spectroscopy. The parti-
tion coefficients are mean values for the different experiments. The stan-
dard deviation of the partition coefficient was usually around 2% and
always less than 5%.
Determination of the stoichiometry of inclusion complexes by the contin-
uous variation method: The stoichiometry of CD/ACNa complexes was
provided by the continuous variation method (Jobꢅs method) by using
NMR spectroscopy. A series of samples containing ratios of CD and
ACNa that varied from 0 to 1 was prepared, keeping the total concentra-
tion of interacting species constant (1 mm). Chemical shift variation of
the CD or ACNa was measured for a given molar ratio. The product
Ddꢂc (Dd is the chemical shift variation observed and c is the concentra-
tion of the observed compound, that is, CD or ACNa) was plotted as a
function of the molar ratio, r.
Determination of the association constant of inclusion complexes: The
evaluation of the host inclusion ability towards the sodium salt of 1-ada-
mantane carboxylic acid (ACNa) was determined by NMR spectroscopy
combined with the direct titration method. This method was applied for
a
fixed concentration of CD (1 mm) and varying concentrations of
Acknowledgements
ACNa.
Assuming a 1:1 stoichiometry, the calculation of the formation constant
This work was supported by the Centre National de la Recherche Scienti-
fique (CNRS). F.X. Legrand is grateful to the Ministꢃre de l’Education et
de la Recherche (2007–2010) and the Rꢀseau de Recherche (RDR2,
Chimie ꢀco-compatible, 2008) for financial support. Roquette Frꢃres
(Lestrem, France) is gratefully acknowledged for generous gifts of cyclo-
dextrins. The authors are grateful to G. Crowyn for NMR analysis. The
500 MHz NMR facilities were funded by the Rꢀgion Nord-Pas de Calais
(France), the Ministꢃre de la Jeunesse, de l’Education Nationale et de la
Recherche (MJENR) and the Fonds Europꢀens de Dꢀveloppement Rꢀ-
gional (FEDER).
Kf was developed as follows:
½CD=ACNaꢂ
½CD=ACNaꢂ
Kf ¼
¼
½CDꢂ ꢃ ½ACNaꢂ ð½CDꢂ0ꢀ½CD=ACNaꢂÞ ꢃ ð½Guestꢂ0ꢀ½CD=ACNaꢂÞ
and [CD/ACNa] can be estimated by:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ꢀ
ꢁ
1
2
1
Kf
½CD=ACNaꢂ ¼ꢀ
ꢃ
ꢃ
þ ½CDꢂ0þ½ACNaꢂ0 2ꢀ4 ꢃ ½CDꢂ0ꢃ½ACNaꢂ0
ꢀ
ꢁ
1
2
1
Kf
þ
þ ½CDꢂ0þ½ACNaꢂ0
[1] a) J. Szejtli, Chem. Rev. 1998, 98, 1743–1753; b) A. R. Hedges,
Chem. Rev. 1998, 98, 2035–2044.
[2] F. Hapiot, S. Tilloy, E. Monflier, Chem. Rev. 2006, 106, 767–781.
[3] R. M. Deshpande, A. Fukuoka, M. Ichikawa, Chem. Lett. 1999, 13–
14.
[4] E. Engeldinger, D. Armspach, D. Matt, L. Toupet, M. Wesolek, C.
R. Chim. 2002, 5, 359–372.
[5] M. Sawamura, K. Kitayama, Y. Ito, Tetrahedron: Asymmetry 1993,
4, 1829–1832.
For a given value of Kf, the [CD/ACNa] concentration was known and
thus the chemical shift of the complex can be calculated. An algorithm
treatment was then applied to minimise the difference between the ex-
perimental and the theoretical values of the spectral characteristics for
each concentration of CD. The association constant was determined by
computer fitting of the experimental titration curve.
General procedure for the hydrogenation reaction: The first step was the
preparation of the catalytic solutions under N2 by using standard Schlenk
techniques. In a typical experiment, [RhACTHUNTGRNNEUG(cod)2]BF4 or [RhACHTUTGNREN(NGUN acac)(CO)2]
[6] M. T. Reetz, J. Rudolph, Tetrahedron: Asymmetry 1993, 4, 2405–
2406.
(7.55ꢂ10ꢀ3 mmol) and phosphane (1.51ꢂ10ꢀ2 mmol) were dissolved in
deoxygenated H2O (12 mL) over 15 min. Then the catalytic aqueous solu-
tion was preincubated overnight under a H2 atmosphere in a glass reactor
to saturate the solution and to generate the active catalytic species. After
this incubation step, 2-methyl-3-buten-2-ol (1.89 mmol) was added to the
catalytic solution. The glass reactor was then heated at 608C and stirred
with a magnetic stirrer, and the link between the glass reactor and the H2
gas burette was established. For kinetic measurements the time corre-
sponding to the desired temperature was considered to be the beginning
of the reaction. The reaction medium was sampled during the reaction by
GC analyses and H2 consumption.
[7] M. T. Reetz, Catal. Today 1998, 42, 399–411.
[8] M. T. Reetz, J. Rudolph, R. Goddard, Can. J. Chem. 2001, 79, 1806–
1811.
[9] M. T. Reetz, I. D. Kostas, S. R. Waldvogel, Inorg. Chem. Commun.
2002, 5, 252–254.
[10] M. T. Reetz, S. R. Waldvogel, Angew. Chem. 1997, 109, 870–873;
Angew. Chem. Int. Ed. Engl. 1997, 36, 865–867.
[11] M. T. Reetz, J. Heterocycl. Chem. 1998, 35, 1065–1073.
[12] M. T. Reetz, C. Frçmbgen, Synthesis 1999, 1555–1557.
[13] D. Armspach, D. Matt, Chem. Commun. 1999, 1073–1074.
[14] E. Engeldinger, D. Armspach, D. Matt, Angew. Chem. 2001, 113,
2594–2597; Angew. Chem. Int. Ed. 2001, 40, 2526–2529.
[15] E. Engeldinger, D. Armspach, D. Matt, P. G. Jones, R. Welter,
Angew. Chem. 2002, 114, 2705–2708; Angew. Chem. Int. Ed. 2002,
41, 2593–2596.
[16] E. Engeldinger, D. Armspach, D. Matt, P. G. Jones, Chem. Eur. J.
2003, 9, 3091–3105.
[17] E. Engeldinger, L. Poorters, D. Armspach, D. Matt, L. Toupet,
Chem. Commun. 2004, 634–635.
[18] L. Poorters, D. Armspach, D. Matt, L. Toupet, P. G. Jones, Angew.
Chem. 2007, 119, 2717–2719; Angew. Chem. Int. Ed. 2007, 46, 2663–
2665.
General procedure for the hydroformylation reaction: All catalytic reac-
tions were performed under N2 by using standard Schlenk techniques. All
solvents and liquid reagents were degassed with bubbling N2 for 15 min
before each use or by performing two freeze–pump–thaw cycles before
use. In a typical experiment, [Rh
N
phosphane (7.76ꢂ10ꢀ3 mmol) were dissolved in degassed solvent (6 mL).
The resulting aqueous phase and the olefinic-type substrate (0.97 mmol)
were added to the 25 mL stainless steel autoclave (Parr) under an atmos-
phere of N2, then heated at 808C. The mechanical stirrer equipped with a
multipaddle unit was then started (1500 rpm) and the autoclave was pres-
surised with CO/H2 (50 bar; 1:1) from a gas reservoir connected to the
10200
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10195 – 10201

Catalytic Phosphanes
FULL PAPER
[19] L. Poorters, D. Armspach, D. Matt, L. Toupet, S. Choua, P. Turek,
Chem. Eur. J. 2007, 13, 9448–9461.
[26] J. A. Faiz, N. Spencer, Z. Pikramenou, Org. Biomol. Chem. 2005, 3,
4239–4245.
[20] L. Poorters, D. Armspach, D. Matt, Eur. J. Org. Chem. 2003, 1377–
1381.
[21] L. Poorters, D. Armspach, D. Matt, L. Toupet, Dalton Trans. 2007,
3195–3202.
[27] F. Djedaꢆni-Pilard, N. Azaroual-Bellanger, M. Gosnat, D. Vernet, B.
Perly, J. Chem. Soc. Perkin Trans. 2 1995, 723–730.
[28] P. Berthault, N. Birlirakis, Tetrahedron: Asymmetry 2000, 11, 2463–
2469.
[22] C. Yang, Y. K. Cheung, J. Yao, Y. T. Wong, G. Jia, Organometallics
2001, 20, 424–429.
[23] C. Yang, Y. T. Wong, Z. Li, J. J. Krepinsky, G. Jia, Organometallics
2001, 20, 5220–5224.
[24] Y. T. Wong, C. Yang, K. C. Ying, G. Jia, Organometallics 2002, 21,
1782–1787.
[29] N. Bellanger, B. Perly, J. Mol. Struct. 1992, 273, 215–226.
[30] M. V. Rekharsky, Y. Inoue, Chem. Rev. 1998, 98, 1875–1917.
[31] R. Gꢇrtner, B. Cornils, H. Springer, P. Lappe, 1982, DE 3235030.
[32] D. Landy, S. Fourmentin, M. Salome, G. Surpateanu, J. Inclusion
Phenom. Mol. Recognit. Chem. 2000, 38, 187–198.
[25] R. C. Petter, J. S. Salek, C. T. Sikorski, G. Kumaravel, F. T. Lin, J.
Am. Chem. Soc. 1990, 112, 3860–3868.
Received: February 11, 2010
Published online: June 30, 2010
Chem. Eur. J. 2010, 16, 10195 – 10201
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10201