Cyclodextrin/amphiphilic phosphane mixed systems and their applications in aqueous organometallic catalysis

Article abstract of DOI:10.1002/adsc.201100837

When mixed with a water-soluble phosphane capable of self-assembling into micelles, native or modified β-cyclodextrins (β-CDs) show very contrasting behavior depending on their neutral or ionic nature. In the post-micellar region, neutral β-CDs led to a micelle destructuring. Conversely, micelles remained stable over a well-defined range of ionic β-CD concentrations. In that case, the micelle destruction was only observed when using a large excess of ionic β-CDs. The catalytic performances of these micellar systems have been evaluated in a rhodium-catalyzed hydroformylation reaction of 1-decene. We showed that, using ionic β-CDs, the catalytic activity could be improved without a detrimental impact upon the regioselectivity. A linear/branched aldehyde ratio as high as 8.6 could be achieved. The best results were obtained with stoichiometric quantities of ionic randomly methylated β-CDs with respect to the phosphane with a beneficial effect on the decantation at the end of the reaction. Copyright

Full text of DOI:10.1002/adsc.201100837

FULL PAPERS
DOI: 10.1002/adsc.201100837
Cyclodextrin/Amphiphilic Phosphane Mixed Systems and their
Applications in Aqueous Organometallic Catalysis
Michel Ferreira,a Hervꢀ Bricout,a Nathalie Azaroual,b David Landy,c
Sꢀbastien Tilloy,a Frꢀdꢀric Hapiot,a,* and Eric Monfliera
a
Universitꢀ Lille Nord de France, CNRS UMR 8181, Unitꢀ de Catalyse et de Chimie du Solide, UCCS UArtois, Facultꢀ
Jean Perrin, rue Jean Souvraz, SP18, F-62307 Lens, France
Fax : (+33)-(0)3-2179-1755; phone: (+33)-(0)3-2179-1773; e-mail: frederic.hapiot@univ-artois.fr
Universitꢀ Lille Nord de France, UDSL, EA 4481, Facultꢀ de Pharmacie, Universitꢀ de Lille 2, BP 83, F-59006 Lille,
b
France
c
Universitꢀ Lille Nord de France, UCEIV, Universitꢀ du Littoral, 145, Avenue Maurice Schumann, MREI 1, F-59140
Dunkerque, France
Received: October 28, 2011; Revised: February 7, 2012; Published online: April 26, 2012
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201100837.
Abstract: When mixed with a water-soluble phos- decene. We showed that, using ionic b-CDs, the cata-
phane capable of self-assembling into micelles, lytic activity could be improved without a detrimental
native or modified b-cyclodextrins (b-CDs) show impact upon the regioselectivity. A linear/branched
very contrasting behavior depending on their neutral aldehyde ratio as high as 8.6 could be achieved. The
or ionic nature. In the post-micellar region, neutral best results were obtained with stoichiometric quan-
b-CDs led to a micelle destructuring. Conversely, mi- tities of ionic randomly methylated b-CDs with re-
celles remained stable over a well-defined range of spect to the phosphane with a beneficial effect on
ionic b-CD concentrations. In that case, the micelle the decantation at the end of the reaction.
destruction was only observed when using a large
excess of ionic b-CDs. The catalytic performances of
these micellar systems have been evaluated in a rho- Keywords: cyclodextrins; hydroformylation; mi-
dium-catalyzed hydroformylation reaction of 1- celles; phosphanes; supramolecular chemistry
Introduction
tained at the end of the reaction,[29,30,33] thus rendering
the catalyst recovery impossible. In this context, we
Since the mid-1970s and the pioneering works of E. envisaged the use of cyclodextrins (CDs) to avoid the
Kuntz,[1,2] the use of water in organometallic catalysis emulsion problem. It has already been shown that,
has been tackled but its application was limited to depending on the nature of both the CD and the sur-
water-soluble substrates. For hydrophobic reactants, factant, CDs could interact with amphiphilic com-
the problem has since partially been solved with the pounds.[37–47] In fact, three different behaviors could
use of co-solvents,[3,4] surfactants,[5–14] amphiphilic be identified for CD/surfactant couples. In most cases,
phosphanes,[15–23] molecular receptors,[24–26] polymers[27] b-CDs were shown to have a deleterious effect on sur-
or dispersed particles,[28] all capable of improving the factant aggregates. Below the critical micellar concen-
mass transfer between the substrate-containing organ- tration (cmc), a CD/surfactant supramolecular com-
ic phase and the catalyst. Among these strategies, the plex was formed, resulting in a delay in the micelliza-
use of amphiphilic phosphanes appeared promising as tion point. When used in excess in the post-micellar
the surface-active property and the coordination abili- region, b-CDs led to a micelle destructuring. Howev-
ty were incorporated into the same entity. Phosphane er, Milioto et al. clearly demonstrated that HP-a-CD
self-assembling led to micellar aggregates that are or HP-g-CD interacted with sodium alkanoates-based
well-known to positively affect the catalytic perform- micelles.[41,42,43] In that case, CDs adsorbed on the mi-
ances of organometallic systems, increasing both ac- celle surface because of hydrogen bonding between
tivities and selectivities.[29–37] However, their interest the carboxylate groups of the micelle components and
was very relative as stable emulsions were often ob- the CD hydroxyle functions. Garcꢁa-Rꢁo et al., for
Adv. Synth. Catal. 2012, 354, 1337 – 1346
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1337

Michel Ferreira et al.
FULL PAPERS
their part, showed that above the cmc, native or sulfo- reflected the association constant relative to the CD/
butyl ether b-CD and sodium dodecyl sulfate co-exist- phosphane supramolecular complex. A 1/1 stoichiome-
ed in water without interacting.[46,47] Given the consid- try was found using ITC whatever the b-CD. Very dif-
erable haziness regarding the CD/surfactant interac- ferent values were measured for Kass, depending upon
tions,[48] we were interested in evaluating the effects the CD shape, from 4,500MÀ1 for DMEA-b-CD+, ClÀ
of neutral and ionic b-CDs in micellar organometallic to 326,000MÀ1 for RAME-b-CD. The lowest associa-
catalysis.
tion constants measured for DMEA-b-CD+, ClÀ and
Herein, we show that the appropriateness of a sur- DMOA-b-CD+, ClÀ probably resulted from a partial
face active sulfonated phosphane with well-defined inclusion of the substituent hydrophobic part into the
ionic b-CDs could significantly improve the perform- CD cavity which competed with the inclusion of 1.
ances of the Rh-catalyzed hydroformylation of 1- SBE-b-CDÀ, Na+, for its part, displayed a low Kass
decene in micellar conditions. Moreover, for some of (8,070MÀ1) due to electrostatic repulsions between
the studied b-CDs, both the organic and aqueous 1 and the CDs.
phases could be separately recovered by simple de-
cantation when the reaction was over, giving our
system a gain over other catalytic micellar systems.
NMR Investigations
The existence of supramolecular complexes at high
concentration (10 mM) of neutral or ionic CDs and
1 was revealed using NMR spectroscopy. The geome-
Results and Discussion
The recently described phosphane 1 (Figure 1)[21] has try of the CD/1 complexes could be determined using
been used as both micelle-constitutive building block a 2D NMR ROESY sequence which is sensitive to di-
and water soluble rhodium-stabilizing ligand. In the polar contacts between the CD host and the phos-
studied catalytic system, phosphane 1 was always used phane guest. The 2D ROESY NMR spectra of mix-
in excess regarding the rhodium precursor (5 equiv.). tures of 1 and neutral native b-CD or cationic
Hence, non-coordinated 1 was always available to HPTMA-b-CD+, ClÀ are depicted in Figure 2 and
form micelles. All the studied neutral and ionic b- Figure 3, respectively. The spectra clearly revealed di-
CDs are listed in Table 1. In the first part of the polar contacts between components, indicative of in-
paper we focus our attention on the interactions exist- clusion of the sulfonated biphenyl and/or the phenyl
ing between the neutral or ionic b-CDs and 1. In the moieties of 1 into the CD cavity. Both the phenyl and
second part, the catalytic performances of a rhodium the biphenyl groups could not be included within the
precursor in the presence of 1 and neutral or ionic CD cavities at the same time, in line with the above
CDs are discussed through the hydroformylation of 1- 1/1 stoichiometry. Moreover, the recognition process
decene.
depended upon the nature of the CD. While the
native b-CD could interact with both the phenyl and
the biphenyl groups (cross-peaks between H3 and H5
with all the aromatic protons, Figure 2), HPTMA-b-
CD+, ClÀ had a strong preference for phenyl groups
Isothermal Titration Calorimetry
The inclusion complex formation between neutral or (cross-peaks between H3 and H5 mainly with the
ionic b-CDs and 1 was first studied using isothermal phenyl protons, Figure 3). Cross-correlations were
titration calorimetry (ITC). The stoichiometry and as- also detected in the 2D NMR ROESY spectrum of
sociation constant (Kass, Table 1) of each CD/1 couple the SBE-b-CDÀ, Na+/1 mixture with a marked prefer-
were determined at a low concentration (0.067 mM) ence for a CD/phenyl interaction (see the Supporting
to avoid heat variations resulting from the presence Information).
of micelles at higher concentration [cmc (1)=
0.6 mM]. Thus, at low concentration, the values only
Tensiometry and CryoTEM Imaging
The intrinsic surface activity of neutral and ionic CDs
has first been measured in pure water at 258C. While
some of the CDs greatly impacted the surface tension
g (especially cationic CDs such as DMEA-b-CD+, ClÀ
and MPTMA-RAME-b-CD+, ClÀ, see Figure 4),
others had only a slight effect (for example, HPTMA-
b-CD+, ClÀ, native b-CD and SBE-b-CDÀ, Na+;
Figure 4). No clear correlation could be established
Figure 1. Water-soluble phosphanes 1 and 2.
between the neutral or ionic CD character and the
1338
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 1337 – 1346

Cyclodextrin/Amphiphilic Phosphane Mixed Systems and their Applications in Aqueous Organometallic Catalysis
Table 1. Structure of native and modified b-CDs and their association constant (Kass) with phosphane 1.
Cyclodextrin
Substituent (R)
none
Carbon bearing the Number of R
Kass (CD/1) (MÀ1) in
water at 258C
R group
groups by CD
Native b-CD
–
–
50,000
HP-b-CD
2, 3, 6
2, 3, 6
5.6
210,000
326,000
12.6 (average
number)
RAME-b-CD
GUA-b-CD+, ClÀ
6
6
1
1
161,500
4,500
DMEA-b-CD+, ClÀ
DMOA-b-CD+, ClÀ
HPTMA-b-CD+, ClÀ
6
2
1
1
5,280
178,500
MPTMA-RAME-b-
2
1
14
53,500
8,070
CD+, ClÀ
2, 3, 6
SBE-b-CDÀ, Na+
2
1
SO3CH2Trz-RAME-b-
6
1
12
48,900
CDÀ, Na+
2, 3, 6
CD’s intrinsic surface activity. On the other side, ten- effect was particularly pronounced with excess native
siometric measurements carried out into aqueous sol- b-CD that rapidly led to a total destruction of mi-
utions containing 1 and various CD concentrations re- celles as the surface tension of pure water was recov-
vealed a strong difference between neutral and ionic ered (g=72 mN·mÀ1, Figure 5). In that case, the
CDs (Figure 5).[49]
3.0 mM/0.6 mM aqueous solution of b-CD/1 only con-
Addition of neutral CDs into aqueous solutions of tained inclusion complexes having no affinity for the
1 at the critical micellar concentration resulted in aqueous/organic interface (Figure 6).
a significant increase in the surface tension indicative
Conversely, we showed that 1-based micelles be-
of an alteration of the micellar system. As neutral b- haved very differently in the presence of ionic CDs.
CDs had very little impact upon the surface tension While an increase in the surface tension was observed
when dissolved in pure water (Figure 4), the increase upon addition of neutral b-CDs, addition of ionic b-
in g observed in the presence of 1 was necessarily the CDs into aqueous solutions of 1 at the cmc induced
consequence of the existence of neutral b-CD/1 com- a more or less marked decrease in the surface tension,
plexes and adequately corroborated previous results indicative of new surface active entities (Figure 5).
about the deleterious effect of neutral b-CDs on mi- For example, g dropped from 48 mN·mÀ1 without CD
celles.[50] As soon as the CD concentration increased, to 29 mN·mÀ1 with DMEA-b-CD+, ClÀ at an approx.
the micellar system was significantly altered.[51] The 0.8 mM concentration. Note that the drop in surface
Adv. Synth. Catal. 2012, 354, 1337 – 1346
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1339

Michel Ferreira et al.
FULL PAPERS
Figure 2. 2D ROESY spectrum of a stoichiometric mixture of native b-CD and 1 (10 mM/10 mM) at 258C in D2O.
Figure 3. 2D ROESY spectrum of a stoichiometric mixture of HPTMA-b-CD+, ClÀ and 1 (10 mM/10 mM) at 258C in D2O.
1340
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 1337 – 1346

Cyclodextrin/Amphiphilic Phosphane Mixed Systems and their Applications in Aqueous Organometallic Catalysis
Figure 4. Effect of increasing concentrations of CDs on the surface tension of water (258C).
Figure 5. Effect of increasing concentrations of CDs on the surface tension of a 0.6 mM aqueous solution of 1 (258C).
tension did not exclusively result from the amphiphil- 42 mN·mÀ1 when increasing the HPTMA-b-CD+, ClÀ
ic character of the ionic CDs. Indeed, although concentration from 0 to 1.8 mM, Figure 5).
HPTMA-b-CD+, ClÀ had no effect on the surface ten-
Cryo-TEM experiments confirmed the presence of
sion when dissolved in pure water (72 mN·mÀ1 what- micelles upon addition of ionic CDs into a solution of
ever the concentration, Figure 4), its impact upon the 1. As examples, cryo-TEM images obtained for stoi-
surface tension was not negligible when mixed with chiometric MPTMA-RAME-b-CD+, ClÀ/1 and
a micellar solution of 1 (g dropped from 48 to SO3CH2Trz-RAME-b-CDÀ, Na+/1 mixtures are dis-
Adv. Synth. Catal. 2012, 354, 1337 – 1346
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1341

Michel Ferreira et al.
FULL PAPERS
Figure 6. Micelle destruction in the presence of excess neu-
tral b-CDs.
played in Figure 7 and Figure 8. The cryo-TEM
images revealed spherical entities in a Gaussian distri-
bution with an average diameter of 0.8 nm and a ho-
Figure 9. CryoTEM image of an HPTMA-b-CD+, ClÀ/1 mix-
ture (4.5 mM/4.5 mM).
mogeneous dispersion. Of interest, the nature of the
micellar aggregates significantly changed when using
HPTMA-b-CD+, ClÀ as an additive. Vesicule-like ag-
gregates were then observed (Figure 9). Thus, our mi-
cellar system supports the view that an interaction
exists between the amphiphilic phosphane and ionic
b-CDs. The decrease in g suggests the formation of
new amphiphilic supramolecular objects with more
pronounced surface-active ability over a well-defined
concentration range. However, it is not clear at this
stage how these new entities were formed. Two hy-
potheses could be given (Figure 10). First, as previ-
ously described in the literature, ionic b-CDs could in-
teract with the micelles outer sphere (electrostatic
ammonium/sulfonate interaction, for example) and/or
could partially penetrate into the micelle through
a recognition process between the CD cavity and the
sulfobiphenyl group of 1. Second, “free” 1 molecules
(in equilibrium with micelles) were captured by ionic
CD cavities to furnish the amphiphilic CD/1 supra-
molecular complexes responsible for the decrease in
g. Whatever the mechanism and contrary to what was
observed with neutral b-CDs, the integrity of the mi-
Figure 7. CryoTEM image of an MPTMA-RAME-b-CD+,
ClÀ/1 mixture (4.5 mM/4.5 mM).
Figure 8. CryoTEM image of an SO3CH2Trz-RAME-b-CDÀ,
Na+/1 mixture (4.5 mM/4.5 mM).
Figure 10. Possible interactions between ionic b-CDs and
1 (X=ionic substituent).
1342
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 1337 – 1346

Cyclodextrin/Amphiphilic Phosphane Mixed Systems and their Applications in Aqueous Organometallic Catalysis
Table 2. Rhodium-catalyzed hydroformylation of 1-decene using neutral or ionic b-CDs as additives.[a]
Entry
Additive
Additive/1
Conversion [%][b]
Chemoselectivity[c]
l/b ratio[d]
1
2
3
4
5
6
7
8
–
0
1
1
1
2
4
7
1
1
1
1
1
1
1
4
1
1
15
9
9
96
97
96
93
93
93
96
95
95
96
93
95
96
93
96
96
96
9.1
4.1
4.6
3.6
2.9
2.6
9.0
7.5
7.1
8.3
6.3
8.6
8.6
8.4
3.0
5.2
7.3
b-CD
HP-b-CD
RAME-b-CD
RAME-b-CD
15
32
67
10
65
41
59
33
41
40
42
52
33
71
RAME-b-CD
a-d-methylglucopyranose
GUA-b-CD+, ClÀ
DMEA-b-CD+, ClÀ
DMOA-b-CD+, ClÀ
HPTMA-b-CD+, ClÀ
MPTMA-RAME-b-CD+, ClÀ
MPTMA-RAME-b-CD+, ClÀ
MPTMA-RAME-b-CD+, ClÀ
MPTMA-RAME-b-CD+, ClÀ
SBE-b-CDÀ, Na+
9
10
11
12
13e)
14f)
15
16
17
SO3CH2Trz-RAME-b-CDÀ, Na+
[a]
Experimental conditions: RhACHTNUGRTNEUNG
(acac)(CO)2 (4.07ꢃ10À2 mmol), phosphane 1 (0.21 mmol), cyclodextrin (0.21 mmol) or a-d-
methylglucopyranose (1.47 mmol), H2O (11.5 mL), 1-decene (10.6 mmol), undecane (internal standard): 1 mmol (0.167 g),
1500 rpm, CO/H2 (1/1): 20 bar, 808C, 6 h.
Calculated with respect to the starting olefin.
(mol of aldehydes)/(mol of converted olefins)ꢃ100. The side products were mainly isomeric olefins.
Ratio of linear to branched aldehyde product.
Recycling of the aqueous phase of entry 12.
[b]
[c]
[d]
[e]
[f]
Recycling of the aqueous phase of entry 13.
celles was not altered over a wide concentration b=9.1).[53] This result was in agreement with the
range. Whereas the micelle destruction occurred at result obtained with 2 (Figure 1), a phosphane unable
low neutral b-CDs concentration, large quantities of to self-assemble into micelles and for which a low l/
ionic b-CDs were necessary to totally destroy the mi- b ratio of 2.8 was obtained.
celles, the equilibrium being then totally displaced to-
Addition of stoichiometric amounts of neutral b-
wards the formation of CD/1 supramolecular com- CDs into a micellar solution of 1 had a negative
plexes (Figure 10).
impact on the catalytic activity. The conversions were
at most equivalent (15% conversion after 6 h with
RAME-b-CD) or lower (9% conversion after 6 h with
the native b-CD or HP-b-CD) than that measured
without CD, in line with the rise in surface tension
Catalysis
Once the behavior of neutral and ionic CDs in the observed above (Figure 5). The decrease in regiose-
presence of 1 had been clarified, the catalytic per- lectivity (up to l/b=3.6 with RAME-b-CD) also sug-
formances of each system was evaluated in the Rh- gested a poor contribution of micelles in the catalytic
catalyzed hydroformylation of terminal alkenes. This process. Increasing the amount of RAME-b-CD led
model reaction has been chosen as it is of industrial to a contrasting result. For example, with four equiva-
importance, the resulting products being key inter- lents of RAME-b-CD referred to 1, the conversion
mediates in the synthesis of plasticizers, lubricants or underwent a 4.5-fold increase (entries 4 and 6) where-
surfactants.[52] Moreover, no less than three parame- as the regioselectivity decreased from 3.6 to 2.6,
ters (conversion, chemo- and regioselectivities) are in- a value closer to what was obtained with neutral CD-
dicative of the efficiency of the catalytic system. based systems in non-micellar conditions (1.8<l/b<
Table 2 collects the catalytic results obtained in the 2.8).[24]
hydroformylation of 1-decene at 808C under 20 bar
The conversion increase resulted from the presence
CO/H2. Without CD or additive (entry 1), the catalyt- of “free” CDs (in excess with respect to the phos-
ic activity was 5-fold that obtained with the well- phane) that allowed for a better substrate transfer be-
known TPPTS (15% vs. 3%, respectively),[24] indica- tween the organic and aqueous phases. The l/b drop,
tive of the surface active character of 1. Concurrently, for its part, was undoubtedly a consequence of both
the linear to branched aldehydes ratio (l/b) was in the the masking of the surface activity of 1 and the result-
range of what was usually observed with micelles (l/ ing micelle destructuring. Indeed, as shown above,
Adv. Synth. Catal. 2012, 354, 1337 – 1346
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1343

Michel Ferreira et al.
FULL PAPERS
neutral b-CDs strongly interacted with both the
Concerning anionic b-CDs (SBE-b-CDÀ, Na+ and
phenyl and biphenyl groups of 1. Once 1 was included SO3CH2Trz-RAME-b-CDÀ, Na+), the conversion in-
in the CD cavity, its amphiphilic character was crease suggested that anionic CDs also favored the
masked resulting in micelle destruction and a subse- exchange between the hydrophobic substrate-contain-
quent decrease in the l/b ratio (Figure 6). The involve- ing micelle core and the catalyst-containing aqueous
ment of the neutral b-CD/1 complexes in the micelle solution (Table 2, entries 16 and 17). The l/b values
destabilization was also supported by a catalytic test were indicative of micelles in the medium (as re-
carried out in the presence of the CD building block, vealed by cryo-TEM experiments), even if the regio-
namely a-d-methylglucopyranose, in which 1 could selectivity measured with SBE-b-CDÀ, Na+ reflected
not be included. In that case, the catalytic results a lower contribution of the micelle in the catalytic
were comparable to those obtained without CD (com- system (Table 2, entry 16). SO3CH2Trz-RAME-b-
pare entries 1 and 7) which indirectly confirmed the CDÀ, Na+, for its part, was the most efficient ionic
implication of the CD cavity in the micelle destructur- CD in terms of activity (Table 2, entry 17). Moreover,
ing process. Thus, the conversion and the regioselec- as underlined above for raandomly methylated CDs,
tivity could not be improved at the same time in the a rapid decantation was observed once the reaction
presence of neutral b-CDs.
was complete when using these CDs. Note that better
In sharp contrast to neutral b-CDs, the use of a stoi- conversions and regioselectivities were measured with
chiometric amount of cationic b-CDs regarding 1 great- ionic b-CDs (when compared to neutral b-CDs) what-
ly improved the conversion of 1-decene without any ever the association constant between the CD and 1.
significant loss in either chemoselectivity or regioselec- Increasing the ionic b-CDs amounts led to an increase
tivity. For example, the conversion measured using in the conversion and a concomitant decrease in the l/
GUA-b-CD+, ClÀ as an additive reached 65% after 6 h b ratio resulting from a total disorganization of the
reaction time with a 95% aldehyde chemoselectivity micelle building blocks. For example, with four equiv-
and a 7.5 l/b ratio (Table 2, entry 8). Even HPTMA-b- alents of MPTMA-RAME-b-CD+, ClÀ with respect to
CD+, ClÀ (that did not adsorb at the aqueous/air inter- 1, the conversion slightly increased from 41 to 52%
face) gave a better conversion than that obtained with- whereas the l/b ratio significantly decreased from 8.6
out CD. The high l/b ratios confirmed the presence of to 3.0 (entries 12 and 15). Thus, as observed above for
micelles in the catalytic solution and corroborated neutral CDs, excess ionic CDs led to a total destruc-
cryo-TEM experiments. Lower l/b values have been tion of 1-based micelles. The difference between neu-
obtained for micelle-free systems. For example, a cata- tral and ionic b-CDs lies in the fact that the destruc-
lytic test carried out under the same catalytic condi- tion process was delayed for ionic b-CDs due to the
tions with RAME-b-CD as additive and TPPTS as existence of stable amphiphilic entities over a well-de-
ligand led to a 100% conversion within 6 h, a 95% al- fined concentration range.
dehyde selectivity but a low l/b ratio (1.8),[54] thus high-
lighting the necessity of a well-designed phosphane to
reach a high regioselectivity.
Conclusions
Of interest, note that a partial methylation of the b-
CDs had a beneficial impact upon the decantation of To sum up, we have found that the addition of ionic
the micellar system once the reaction was complete. b-CDs in stoichiometric proportions into an amphi-
Indeed, the phase separation occurred within a few philic phosphane 1-containing micellar solution unex-
minutes with the partially methylated RAME-b-CD pectedly improved the performances of the Rh-cata-
when several hours were necessary with non-methy- lyzed hydroformylation of 1-decene in which comple-
lated CDs. Once the aqueous phase had been recov- mentary properties of the ligand and the surfactant
ered, the reusability of the catalytic system was clear- were expressed in a single material. An average 4-
ly established using MPTMA-RAME-b-CD+, ClÀ as fold increase in the catalytic activity was obtained
additive (Table 2, entries 13 and 14). No trace of rho- without any detrimental effect on the regioselectivity.
dium could be detected in the organic phase by ICP- Additionally, the main advantage of the present
AES, i.e., below the detection limit (0.2 ppm). To un- system was the easy recovery of both the aqueous and
ambiguously prove the absence of rhodium in the or- organic phases at the end of the reaction when ran-
ganic phase, that of entry 14 was recovered after reac- domly methylated b-CDs were used as additives. The
tion and submitted to another catalytic experiment as results of this structure-activity investigation have
follows: 1-decene (10.6 mmol) was added to the or- provided compelling evidence that the CD/phosphane
ganic phase and the resulting solution was mixed with association could be of great interest in the hydrofor-
an MPTMA-RAME-b-CD+, ClÀ-containing solution mylation of terminal alkenes. However, we believe
that did not contain any Rh
under 50 bar CO/H2, no conversion could be detected tion reaction and could be extended to other catalytic
even after 24 h reaction time. systems in aqueous media.
ACHTUNGTRENN(NUG acac)(CO)2. At 808C that this work is not restricted to the hydroformyla-
1344
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 1337 – 1346

Cyclodextrin/Amphiphilic Phosphane Mixed Systems and their Applications in Aqueous Organometallic Catalysis
Experimental Section
Cryo-TEM
Specimens for cryo-TEM observation were prepared using
a cryoplunge cryo-fixation device (Gatan, USA) in which
a drop of the aqueous suspension was deposited on to glow-
discharged holey-type carbon-coated grids (Ted Pella Inc.,
USA). The TEM grid was then prepared by blotting the
drop containing the specimen to a thin liquid layer that re-
mained across the holes in the support carbon film. The
liquid film was vitrified by rapidly plunging the grid into
liquid ethane cooled by liquid nitrogen. The vitrified speci-
mens were mounted in a Gatan 910 specimen holder
(Gatan, USA) that was inserted in the microscope using
a CT-3500-cryotransfer system (Gatan, USA) and cooled
with liquid nitrogen. TEM images were then obtained from
specimens preserved in vitreous ice and suspended across
a hole in the supporting carbon substrate.
Materials
All chemicals were purchased from Acros and Aldrich
Chemicals in their highest purity. All solvents were used as
supplied without further purification. Distilled water was
used in all experiments. NMR spectra were recorded on
a Bruker DRX300 spectrometer operating at 300 MHz for
1H nuclei, 121 MHz for 31P nuclei and 75 MHz for 13C
nuclei. NMR ROESY experiments were performed on
1
a Bruker Avance spectrometer operating at 500 MHz for H
with a TXI probe at 295 K with a mixing time of 500 ms.
CDCl3 (99.50% isotopic purity) and D2O (99.92% isotopic
purity) were purchased from Euriso-Top. The correct assign-
ments of the chemical shifts were confirmed when necessary
by two-dimensional correlation measurements attained by
1H-1H COSY and H-13C HSQC experiments. Mass spectra
1
Microscopy
were recorded on a MALDI-TOF/TOF Bruker Daltonics
Ultraflex II in the positive reflectron mode with 2,5-DHB as
matrix. The absence of rhodium traces in the organic phase
has been checked using an ICP-AES Activa Horiba Jobin
Yvon spectrometer.
The samples were observed under low dose conditions
(<10 eÀ/A2), at À1788C, using a JEM 1230 ꢄCryoꢅ micro-
scope (Jeol, Japan) operated at 80 kV and equipped with an
LaB6 filament. All the micrographs were recorded on
a Gatan 1.35 Kꢃ1.04 Kꢃ12 bit ES500W CCD camera.
Surface Tension
Catalytic Experiments
Interfacial tension measurements were performed using
a KSV Instruments digital tensiometer (Sigma 70) with
a platinum plate. The precision of the force transducer of
the surface tension apparatus was 0.1 mN·mÀ1. The experi-
ments were performed at 293Æ0.5 K controlled with a ther-
mostatted bath Lauda (RC6 CS).
All catalytic reactions were performed under nitrogen using
standard Schlenk techniques. In
a typical experiment,
Rh
AHCTUNGTRENNUNG
and the additive (0.21 mmol) were dissolved in 11.5 mL of
water. The resulting aqueous phase and the olefin
(10.6 mmol) were charged under nitrogen into a 50-mL reac-
tor heated at the desired temperature. The mixture was me-
chanically stirred using a multipaddle unit (1500 rpm) and
the autoclave was pressurized under 20 atm CO/H2 (1:1)
using a gas reservoir connected to the reactor through
a high-pressure regulator valve allowing us to keep the pres-
sure constant in the reactor during the reaction. Once the
reaction was complete, the organic phase was analyzed by
gas chromatography on a Shimadzu GC-17A gas chromato-
graph equipped with a methylsilicone capillary column
(30 mꢃ0.32 mm) and a flame ionization detector (GC:FID).
Isothermal Titration Calorimetry
An isothermal calorimeter (ITC200, MicroCal Inc., USA)
was used for determining the formation constant of the stud-
ied complexes. The following titration protocol was used:
a degassed aqueous solution of phosphane 1 (204. 5 mL;
phosphate buffer: pH 6.5) was titrated at 298 K with a solu-
tion of cyclodextrin (same buffer) in a 40 mL syringe. The
concentrations of phosphane 1 and cyclodextrins in stock
solutions were 0.067 and 2.5 mM, respectively. After addi-
tion of an initial aliquot of 0.4 mL, 7 aliquots of 5.4 mL of the
cyclodextrin solution were delivered over 10.8 s. The corre-
sponding heat flow was recorded as a function of time. The
time interval between two consecutive injections was 100 s
and agitation speed was 1000 rpm for all experiments. In ad-
dition, the heat effects due to dilution of cyclodextrin were
corrected for by performing blank titrations. The areas
under the peak following each injection (obtained by inte-
gration of the raw signal) were then expressed as the heat
effect per mole of added cyclodextrins. Binding constants
were finally determined by non-linear regression analysis of
the binding isotherms using built-in binding models within
MicroCal Origin 7.0 software package (MicroCal, North-
ampton, MA). Each titration experiment was performed
three times to ensure reproducibility of the results. Note
that at pH 6.5, no phosphonium species existed in the buffer
solution as confirmed by 31P NMR measurements. Thus, ITC
measurements were only related to CDs/1 supramolecular
complexes.
Acknowledgements
This work was supported by the Centre National de la Re-
cherche Scientifique (CNRS). MF is grateful to the Ministꢀre
de l’Enseignement Supꢁrieur et de la Recherche for financial
support (2007–2009). Dr. Cꢁdric Gaillard is gratefully ac-
knowledged for the electronic microscopy experiments and
for fruitful discussions. Roquette Frꢀres (Lestrem, France) is
gratefully acknowledged for generous gifts of cyclodextrins.
The 500 MHz NMR facilities were funded by the Rꢁgion
Nord-Pas de Calais (France), the Ministꢀre de la Jeunesse, de
l’Education Nationale et de la Recherche (MJENR) and the
Fonds Europꢁens de Dꢁveloppement Rꢁgional (FEDER).
Adv. Synth. Catal. 2012, 354, 1337 – 1346
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1345

Michel Ferreira et al.
FULL PAPERS
References
[30] A. Riisager, B. E. Hanson, J. Mol. Catal. A: Chem.
2002, 189, 195–202.
[1] E. G. Kuntz, CHEMTECH 1987, 570–575.
[2] E. G. Kuntz, (Rhꢆne–Poulenc), French Patent
2,366,237, 1976.
[3] P. Kalck, M. Dessoudeix, Coord. Chem. Rev. 1999, 190–
192, 1185–1198.
[4] P. Purwanto, H. Delmas, Catal. Today 1995, 24, 135–
140.
[5] M. Gimenez-Pedros, A. Aghmiz, C. Claver, A. M.
Masdeu-Bultꢇ, D. Sinou, J. Mol. Catal. A: Chem. 2003,
200, 157–163.
[6] M. Li, Y. Z. Li, H. Chen, Y.-e. He, X. J. Li, J. Mol.
Catal. A: Chem. 2003, 194, 13–17.
[7] A. Riisager, B. E. Hanson, J. Mol. Catal. A: Chem.
2002, 189, 195–202.
[8] L. Wang, H. Chen, Y.-e He, Y. Li, M. Li, X. Li, Appl.
Catal. A 2003, 242, 85–88.
[31] M. Li, Y. Li, H. Chen, Y.-e He, X. Li, J. Mol. Catal. A:
Chem. 2003, 194, 13–17.
[32] H. Chen, Y. Li, R. Li, P. Cheng, X. Li, J. Mol. Catal. A:
Chem. 2003, 198, 1–7.
[33] M. Gimꢀnez-Pedrꢇs, A. Aghmiz, C. Claver, A. M.
Masdeu-Bultꢇ, D. Sinou, J. Mol. Catal. A: Chem. 2003,
200, 157–163.
[34] M. Yuan, H. Chen, R. Li, Y. Li, X. Li, Catal. Lett. 2004,
94, 15–16.
[35] M. Li, H. Fu, M. Yang, H. Zheng, Y.-e He, H. Chen, X.
Li, J. Mol. Catal. A: Chem. 2005, 235, 130–136.
[36] H. Fu, M. Li, H. Chen, X. Li, J. Mol. Catal. A: Chem.
2006, 259, 156–160.
[37] R. Lu, J. Hao, H. Wang, L. Tong, J. Colloid Interface
Sci. 1997, 192, 37–42.
[38] H.-J. Buschmann, E. Cleve, E. Schollmeyer, J. Inclusion
Phenom. Macrocyclic Chem. 1999, 33, 233–241.
[39] R. De Lisi, S. Milioto, A. De Giacomo, A. Inglese,
Langmuir 1999, 15, 5014–5022.
[40] R. De Lisi, S. Milioto, N. Muratore, Langmuir 2000, 16,
4441–4446.
[41] R. De Lisi, S. Milioto, N. Muratore, J. Phys. Chem. B
2002, 106, 8944–8953.
[42] R. De Lisi, G. Lazzara, S. Milioto, N. Muratore, J. Phys.
[9] C. Yang, X. Bi, Z.-S Mao, J. Mol. Catal. A: Chem.
2002, 187, 35–46.
[10] H. J. V. Barros, B. E. Hanson, E. V. Gusevskaya, E. N.
Dos Santos, Appl. Catal. A 2004, 278, 57–63.
[11] H. Fu, M. Li, H. Chen, X. Li, J. Mol. Catal. A: Chem.
2006, 259, 156–160.
[12] S. L. Desset, D. J. Cole-Hamilton, D. F. Foster, Chem.
Commun. 2007, 1933–1935.
[13] C. C. Miyagawa, J. Kupka, A. Schumpe, J. Mol. Catal.
A: Chem. 2005, 234, 9–17.
Chem. B 2003, 107, 13150–13157.
[43] R. De Lisi, G. Lazzara, S. Milioto, N. Muratore, I. V.
Terekhova, Langmuir 2003, 19, 7188–7195.
[44] L. Garcꢁa-Rꢁo, J. R. Leis, J. C. Mejuto, A. Navarro-
Vꢉzquez, J. Pꢀrez-Juste, P. Rodriguez-Dafonte, Lang-
muir 2004, 20, 606–613.
[45] G. M. Nicolle, A. E. Merbach, Chem. Commun. 2004,
854–855.
[46] L. Garcꢁa-Rꢁo, M. Mꢀndez, M. R. Paleo, F. J. Sardina, J.
[14] M. Gottardo, A. Scarso, S. Paganelli, G. Strukul, Adv.
Synth. Catal. 2010, 352, 2251–2262.
[15] B. Fell, G. Papadogianakis, J. Mol. Catal. 1991, 66, 143–
154.
[16] H. Ding, B. E. Hanson, T. Bartik, B. Bartik, Organome-
tallics 1994, 13, 3761–3763.
[17] B. E. Hanson, H. Ding, C. W. Kohlpaintner, Catal.
Today 1998, 42, 421–429.
[18] M. S. Goedheijt, B. E. Hanson, J. N. H. Reek, P. C. J.
Kamer, P. W. N. M. van Leeuwen, J. Am. Chem. Soc.
2000, 122, 1650–1657.
Phys. Chem. B 2007, 111, 12756–12764.
[47] M. Cepeda, R. DaviÇa, L. Garcꢁa-Rꢁo, M. Parajꢇ,
Chem. Phys. Lett. 2010, 499, 70–74.
[19] S. Bischoff, M. Kant, Catal. Today 2001, 66, 183–189.
[20] Q. Peng, Y. Yang, C. Wang, X. Liao, Y. Yuan, Catal.
Lett. 2003, 88, 219–225.
[21] M. Ferreira, H. Bricout, N. Azaroual, C. Gaillard, D.
Landy, S. Tilloy, E. Monflier, Adv. Synth. Catal. 2010,
352, 1193–1203.
[48] L. Jiang, Y. Yan, J. Huang, Adv. Colloid Interface Sci.
2011, in press; doi:10.1016/j.cis.2011.07.002.
[49] Tensiometric measurements have been carried out at
the cmc of phosphane 1 (0.6 mM, g=48 mN·mÀ1). The
air/water interface is then saturated by 1. The effect of
CDs on the interface at the cmc illustrates the behavior
of CDs towards 1-based micelles at higher concentra-
tions of 1 in water.
[50] S. K. Mehta, K. K. Bhasin, Shilpee Dham, M. L. Singla,
J. Colloid Interface Sci. 2008, 321, 442–451.
[51] CryoTEM experiments unambiguously confirmed that
micelles still existed in solution upon addition of neu-
tral CDs but only in a very narrow concentration range
(just over the phosphane cmc) (Supporting Informa-
tion). The system could then be described as a mixture
of 1-based micelles and neutral CD/1 complexes.
[52] B. Cornils, Org. Process Res. Dev. 1998, 2, 121–127.
[53] H. Chen, Y. Li, J. Chen, P. Cheng, Y. He, X. Li, J. Mol.
Catal. A 1999, 149, 1–6.
[22] H. Bricout, E. Banaszak, F. Hapiot, C. Len, E. Monfli-
er, Chem. Commun. 2010, 46, 7813–7815.
[23] L. Caron, M. Canipelle, S. Tilloy, H. Bricout, E. Monfli-
er, Tetrahedron Lett. 2001, 42, 8837–8840.
[24] F. Hapiot, L. Leclercq, N. Azaroual, S. Fourmentin, S.
Tilloy, E. Monflier, Curr. Org. Synth. 2008, 5, 162–172.
[25] L. Monnereau, D. Sꢀmeril, D. Matt, L Toupet, Adv.
Synth. Catal. 2009, 351, 1629–1636.
[26] H. Bricout, F. Hapiot, A. Ponchel, S. Tilloy, E. Monfli-
er, Curr. Org. Chem. 2010, 14, 1296–1307.
[27] M. Bortenschlager, N. Schçllhorn, A. Wittmann, R.
Weberskirch, Chem. Eur. J. 2007, 13, 520–528.
[28] K. Kunna, C. Mꢈller, J. Loos, D. Vogt, Angew. Chem.
2006, 118, 7447–7450; Angew. Chem. Int. Ed. 2006, 45,
7289–7292.
[54] E. Monflier, H. Bricout, F. Hapiot, S. Tilloy, A.
Aghmiz, A. M. Masdeu-Bultꢇ, Adv. Synth. Catal. 2004,
346, 425–431.
[29] F. Van Vyve, A. Renken, Catal. Today 1999, 48, 237–
243.
1346
asc.wiley-vch.de
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2012, 354, 1337 – 1346