Rhodium catalyzed hydroformylation assisted by cyclodextrins in biphasic medium: Can sulfonated naphthylphosphanes lead to active, selective and recyclable catalytic species?

Article abstract of DOI:10.1016/j.cattod.2014.06.002

New sulfonated naphthylphosphanes having an average sulfonation degree around 2 have been synthetized and tested as ligand in the aqueous biphasic Rh-catalyzed hydroformylation of 1-decene assisted by randomly methylated β-cyclodextrins. All these water-soluble phosphanes associated to a rhodium precursor were able to perform aqueous hydroformylation of 1-decene. The best results in terms of catalyst recovery and recycling were obtained with sulfonated 2-naphthylphosphanes. With sulfonated 1-naphthylphosphanes, formation of low-coordinated catalytic species leading to a catalyst leaching in the organic phase was postulated. These results were rationalized by considering that sulfonated 1-naphthylphosphanes are bulkier ligands than sulfonated 2-naphthylphosphanes.

Full text of DOI:10.1016/j.cattod.2014.06.002

Catalysis Today 247 (2015) 47–54
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Rhodium catalyzed hydroformylation assisted by cyclodextrins in
biphasic medium: Can sulfonated naphthylphosphanes lead to active,
selective and recyclable catalytic species?
a
a
a
a
b
Maxime Elard , Julien Denis , Michel Ferreira , Hervé Bricout , David Landy ,
a
a,∗
Sébastien Tilloy , Eric Monflier
Unité de Catalyse et de Chimie du Solide (UCCS), CNRS, UMR 8181, Université d’Artois, Rue Jean Souvraz, SP 18, F-62307 Lens, France
a
b
Unité de Chimie Environnementale et Interactions sur le Vivant (UCEIV)—EA 4492, Université du Littoral Côte d’Opâle 145, Avenue Maurice Schumann,
MREI 1, F-59140 Dunkerque, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 April 2014
Received in revised form 21 May 2014
Accepted 5 June 2014
Available online 3 July 2014
New sulfonated naphthylphosphanes having an average sulfonation degree around 2 have been syn-
thetized and tested as ligand in the aqueous biphasic Rh-catalyzed hydroformylation of 1-decene assisted
by randomly methylated ␤-cyclodextrins. All these water-soluble phosphanes associated to a rhodium
precursor were able to perform aqueous hydroformylation of 1-decene. The best results in terms of
catalyst recovery and recycling were obtained with sulfonated 2-naphthylphosphanes. With sulfonated
1
-naphthylphosphanes, formation of low-coordinated catalytic species leading to a catalyst leaching
in the organic phase was postulated. These results were rationalized by considering that sulfonated
-naphthylphosphanes are bulkier ligands than sulfonated 2-naphthylphosphanes.
Keywords:
Sulfonated phosphane
Rhodium
Hydroformylation
Biphasic catalysis
Cyclodextrin
1
©
2014 Elsevier B.V. All rights reserved.
Supramolecular chemistry
1
. Introduction
Biphasic aqueous organometallic catalysis has been
back into a biphasic inactive system [17–20]. Immobilization of
the catalyst on a support such as silica [21,22], activated carbon
[23,24], or soluble polymers [25] has also been reported in the
literature.
growing in importance since the synthesis of tris(m-
sulfonatophenyl)phosphane trisodium salt (TPPTS) in the early
Among the different approaches proposed to increase the
solubility of long-chain olefins, the use of chemically modi-
fied ␤-cyclodextrins (CDs) preserves some economical viability.
Indeed, chemically modified CDs are cheap, non-toxic and bulk
commercially available compounds that allowed to achieve the
hydroformylation of long-chain olefins with high reaction rates,
while avoiding the formation of an emulsion and the partition of
the rhodium catalyst between the organic and aqueous phases.
This outstanding result was attributed to the formation of inclusion
complexes at the aqueous/organic interface (Fig. 1) [26].
1
970s [1]. The possibility to recover and to recycle the catalytic sys-
tem by a simple decantation at the end of the reaction is the main
advantage of such strategy. Furthermore, the concept of aqueous
two-phase catalysis was successfully realized in Ruhrchemie-
Rhône Poulenc process in 1984 for the propene hydroformylation.
Nevertheless, this process can only be efficiently implemented
with short-chain olefins due to the low water solubility of long-
chain olefins [2]. To overcome this drawback, the use of cosolvent
[
3,4], surfactants [5–10], lattices [11], cyclodextrins [12] or surface
active ligands [13–16] has been proposed. Another approach that
has been explored is the use of switchable systems that can be
triggered to turn into a homogeneous active system during the
reaction and, upon completion of the reaction, it can be turned
Among the different CDs used in the aqueous hydroformyla-
tion reactions catalyzed by the classical rhodium/TPPTS system, the
randomly methylated ␤-CD (RAME-␤-CD) appeared to be the most
efficient [27,28]. Nevertheless, this cyclodextrin cannot be consid-
ered as a simple mass transfer additive since this latter forms an
inclusion complex with TPPTS [29,30]. This interaction induces not
only a decrease in activity [31] due to the poisoning of the cyclodex-
trin cavity but also the formation of phosphane low-coordinated
rhodium species leading to a lower regioselectivity [32].
∗ Corresponding author. Tel.: +33 3 21 79 17 72; fax: +33 3 21 79 17 55.
E-mail address: eric.monflier@univ-artois.fr (E. Monflier).
http://dx.doi.org/10.1016/j.cattod.2014.06.002
0
920-5861/© 2014 Elsevier B.V. All rights reserved.

4
8
M. Elard et al. / Catalysis Today 247 (2015) 47–54
Fig. 1. Aqueous biphasic organometallic catalysis assisted by cyclodextrins.
Different strategies have been previously applied to avoid the
formation of such complexes including the use of ␣-cyclodextrins
33] or anionic ␤-cyclodextrins [34,35] but none of them give
precise speed measurement by tachometer display. The absence
of rhodium traces in the organic phase has been checked using an
ICP-AES Activa Horiba Jobin Yvon spectrometer.
[
full satisfaction. One promising solution to avoid the interaction
between RAME-␤-CD and water soluble ligands could be to replace
TPPTS by a more bulky water-soluble phosphane which would be
sufficiently crowded to hamper the inclusion in the host cavity [36].
With such configuration, each compound can fully play its own role
without mutual interaction: the water-soluble bulky phosphane
as a ligand for the metal and the CD as an efficient mass transfer
promoter.
In this context, we report herein the synthesis of new sterically
encumbered sulfonated triarylphosphanes and their application as
ligand in the aqueous biphasic Rh-catalyzed hydroformylation of 1-
decene assisted by RAME-␤-CD. More precisely, sulfonated P[(1 or
2.2. Surface tension measurements
Surface tensions of pure aqueous solutions of phosphanes were
determined by successive dilutions using an automatic Sigma
701 tensiometer (Attension, KSV Instruments Ltd.). Measurements
◦
were performed at 25 C with the Wilhelmy plate technique. Aver-
age values of surface tension were obtained from three consecutive
measurements. Before each series of measurements, samples were
carefully filtered through a Millipore filter with a 0.2 ␮m pore size
to remove any impurities. The surface tension of double distilled
water was measured before each series of measurements until a
−
1
2
-naphthyl)x(phenyl)3 ] phosphanes (x = 1, 2 or 3) with an average
value of 72 mN m was reached.
−x
sulfonation degree around 2 have been prepared (Scheme 1). The
potential interactions between these new phosphanes and ␤-CD
2.3. Formation constant measurements by UV–visible
spectroscopy
(
native and RAME-␤-CD) were checked by NMR analysis and UV
titration.
The formation constants of RAME-␤-CD/sulfonated naph-
thylphosphanes complexes were obtained by UV–visible spec-
troscopy, using a competitive method with methyl-orange (MO)
2
. Experimental
[
5
38]. All spectra were recorded three times between 350 and
50 nm, at 25 C, on a Perkin Elmer Lambda 2S spectrometer. The
2
.1. Materials
◦
The 1H, 13C { H} and P { H} NMR spectra were recorded
1
31
1
study was realized in two steps. First of all, the inclusion com-
pound between RAME-␤-CD and MO was characterized by the
well-known titration method. A fixed concentration of 0.033 mM
was used for methyl orange, while RAME-␤-CD concentration
was varied between 0 and 0.1 mM. Resulting values of the first
derivative of absorbance were treated according to a nonlinear
fit between experimental and theoretical data, leading to a for-
on a Bruker Avance 300 DPX instrument at 300.13, 75.5 and
1
21.5 MHz, respectively. Gas chromatographic analyses were car-
ried out on a Varian 3900 GC gas chromatograph equipped
with a polydimethylsiloxane capillary column (15 m × 0.25 mm)
and a flame ionisation detector (GC:FID). Dicarbonylacetylace-
tonato rhodium (I) was purchased from Strem Chemicals and
Randomly methylated ␤-cyclodextrin (RAME-␤-CD) from Wacker
Chemie. This cyclodextrin is partially methylated; statistically
−
1
mation constant equal to 16,500 M
for RAME-␤-CD/MO. In a
second time, the various sulfonated naphthylphosphanes mix-
tures were added to solutions containing a fixed composition of
RAME-␤-CD (0.1 mM) and MO (0.033 mM). An apparent phosphane
concentration of 0.1 mM was used for each mixture, which induced
weak dissociation of RAME-␤-CD/MO complexes and thus modified
the absorbance values. In order to obtain formation constants of
RAME-␤-CD/sulfonated naphthylphosphanes complexes, recorded
spectral values were then treated by a dedicated algorithmic pro-
cedure, as described in Ref. [38], considering a 1:1 stoichiometry.
Standard deviations were close to 10% for each stability. As phos-
phane mixtures were used, the obtained stability values correspond
to apparent formation constants; thanks to the competition princi-
ple, these values have a real physical meaning, since they are based
on the quantity of MO displaced from RAME-␤-CD cavity by the
phosphane mixtures.
1
.8 OH groups per glucopyranose unit were modified. The
tris(1-naphthyl)phosphane (T1NP) was purchased from Strem
Chemicals and tris(3-sodium sulfonatophenyl)phosphine (TPPTS:
P(m-C H SO Na) ) was synthesized as reported by Gärtner et al.
6
4
3
3
[
37]. Carbon monoxide/hydrogen mixture (1:1) was used directly
from cylinders (>99.9% pure; Air Liquide). Distilled deionized water
was used in all experiments. All catalytic reactions were performed
under nitrogen using standard Schlenk techniques. 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.
All the high pressure hydroformylation experiments were carried
out in a 25 mL stainless steel microclave supplied by Parr. The reac-
tors were fitted with arrangements for liquid sampling, automatic
temperature control, pressure gauge and variable stirring with

M. Elard et al. / Catalysis Today 247 (2015) 47–54
49
x = 1 : sulfonated mono 1-naphthyldiphenylphosphane (M1NPS) x = 1 : sulfonated mono 2-naphthyldiphenylphosphane (M2NPS)
x = 2 : sulfonated di 1-naphthylphenylphosphane (D1NPS)
x = 3 : sulfonated tri 1-naphthylphosphane (T1NPS)
x = 2 : sulfonated di 2-naphthylphenylphosphane (D2NPS)
x = 3 : sulfonated tri 2-naphthylphosphane (T2NPS)
Scheme 1. Structure and abbreviation of the water-soluble phosphanes.
1H NMR (300 MHz, CDCl3, 25 C): ı = 7.04 (t, 1H, H-6,
◦
2.4. General procedure for hydroformylation reaction
3
3
JP–H = JH–H = 6.44 Hz), 7.36 (m, 10H, H-2, H-3, H-4), 7.43–7.54 (m,
In all experiments, the stirring speed and the CO/H2 pressure
3H, H-7, H-11, H-12), 7.86–7.90 (m, 2H, H-8, H-10), 8.43 (dd, 1H,
H-13, 4
J
P–H
= 3.22 Hz, 3JH–H = 7.08 Hz); C { H} NMR (75.5 MHz,
13
1
were fixed to 1500 rpm and 50 bar, respectively. In a typical exper-
iment, Rh(acac)(CO)2 (0.021 mmol), phosphane (0.107 mmol) and
RAME-␤-CD (0.21 mmol) were dissolved in 6 mL of water. The
resulting aqueous phase and an organic phase composed of 1-
◦
3
CDCl , 25 C): ı = 125.61 (d, C-7, JP–H = 1.5 Hz), 126.02 (s, C-11
3
3
or C-12), 126.12 (d, C-13, JC–P = 25.5 Hz), 126.67 (s, C-11 or
C-12), 128.62 (d, C-3, 3JC–P = 6.26 Hz), 128.64 (s, C-10 or C-8),
128.95 (s, C-4), 129.57 (s, C-10 or C-8), 132.11 (d, C-6), 133.46
decene (10.5 mmol) were charged under an atmosphere of N into
2
◦
(d, C-14, JC–P = 5.82 Hz), 134.60 (d, C-9, 3JC–P = 18.00 Hz), 134.25
2
the 25 mL reactor which was heated at 80 C. The autoclave was
(d, C-2, 2JC–P = 21.84 Hz), 135.25 (d, C-5, JC–P = 21.90 Hz), 136.10
1
then pressurized with 50 bar of CO/H2 (1/1) and mechanical stir-
ring equipped with a multipaddle unit was started. The pressure
was kept constant over the reaction time of 6 h by feeding syngas
via a pressure controller. At the reaction’s end, the autoclave was
cooled down to room temperature and depressurized. The medium
is then decanted and the organic phase analyzed by GC. For kinetic
measurements, the time corresponding to the addition of CO/H2
was considered as the beginning of the reaction.
(d, C-1, 1JC–P = 9.47 Hz); P{ H} NMR (121.5 MHz, CDCl , 25 C):
31
1
◦
3
ı = −14.4 ppm.
The same procedure was used to synthesize the different naph-
thylphosphanes, with varying the initial quantity and the nature
of the bromonaphthalene derivative (1 or 2-bromonaphthalene)
as well as the chlorophosphane used (chlorodiphenylphosphane,
dichlorophenylphosphane or trichlorophosphane) depending on
the ligand structure.
2.5. Water-soluble phosphane synthesis
2
.6.2. D1NP
3
The various water-soluble phosphanes were synthesized
4
2
1
according to a procedure described in the literature [36]. Briefly,
a halogen derivative in dry THF was vigorously mixed with
magnesium turnings to form the Grignard reactant which was
subsequently condensed on the corresponding chlorophosphane
to form the triarylphosphane. This phosphane was subsequently
dissolved in a sulfuric acid/oleum mixture for sulfonation. After
P
13
5
1
4
1
2
6
7
11
9
10
8
hydrolysis of SO , trioctylamine was added to extract the reaction
product. Finally, the sodium salt was obtained by addition of NaOH
3
Yield: 75%.
1
◦
NMR (300 MHz, CDCl3, 25 C): ı = 7.08 (t, 1H, H-6,
H
(
as an example, synthesis of M1NPS is represented Scheme 2).
3
3
JP–H = JH–H = 6.44 Hz), 7.41 (m, 10H, H-2, H-3, H-4), 7.43–7.54 (m,
H, H-7, H-11, H-12), 7.86–7.90 (m, 2H, H-8, H-10), 8.47 (dd, 1H,
3
2.6. Triarylphosphane synthesis
4
3
13
1
H-13, JP–H = 3.22 Hz, JH–H = 7.08 Hz); C { H} NMR (75.5 MHz,
◦
3
CDCl , 25 C): ı = 125.61 (d, C-7, JP–H = 1.5 Hz), 126.02 (s, C-11 or
3
2.6.1. M1NP
C-12), 126.67 (s, C-11 or C-12), 128.64 (s, C-10 or C-8), 128.67 (d,
To a suspension of magnesium (0.47 g, 19.3 mmol, 1.2 equiv.)
3
C-3, JC–P = 6.40 Hz), 128.95 (s, C-4), 129.57 (s, C-10 or C-8), 132.11
in 7 mL of anhydrous THF was introduced under nitrogen 1-
bromonaphthalene (1.10 g, 0.33 equiv.) diluted in 9 mL of THF.
After a few minutes, the reaction began and 2.21 g (10.7 mmol,
2
(
(
d, C-6), 132.98 (s, C-13), 133.46 (d, C-14, JC–P = 5.82 Hz), 134.60
3
2
d, C-9, J
= 17.50 Hz), 134.25 (d, C-2, J
= 21.84 Hz), 135.25 (d,
C–P
C–P
1
1
31
1
C-5, JC–P: 21.84 Hz), 136.10 (d, C-1, JC–P = 9.47 Hz); P{ H} NMR
0.66 equiv.) of the bromide derivative in THF (18 mL) were added
◦
(
121.5 MHz, CDCl , 25 C): ı = −23 ppm.
3
dropwise. The reaction mixture was then refluxed for 1 h. After
cooling, chlorodiphenylphosphane (3.53 g, 16 mmol, 1 equiv.) in
THF (7 mL) was added dropwise and then refluxed for 1 h. Once the
reaction was complete, the mixture was poured into an ice–water
mixture (10.6 g). The organic phase was dried over anhydrous
2
.6.3. M2NP
3
4
2
1
P
MgSO filtered and concentrated by rotary evaporation. The result-
4
5
8
14
ingproduct wasrecrystallized frommethanolto give white crystals.
6
7
13
Yield: 57%.
3
12
4
2
1
11
9
10
P
13
Yield: 62%.
5
14
1
2
6
7
1
◦
H NMR (300 MHz, CDCl , 25 C): ı = 7.44 (m, 10H, H-2, H-3, H-
3
1
1
4), 7.55-7.50 (m, 1H, H-14), 7.54–7.68 (m, 2H, H-9, H-10), 7.79–7.81
m, 1H, H-12), 7.75–7.88 (m, 1H, H-7), 7.81–7.95 (m, 2H, H-6, H-11);
9
10
8
(

5
0
M. Elard et al. / Catalysis Today 247 (2015) 47–54
Scheme 2. M1NPS synthesis.
1
3
1
◦
C{ H} NMR (75.5 MHz, CDCl , 25 C): ı = 126.20 (s, C-12 or C-9),
oleum depending on the structure of the ligand (2.5 equiv. of SO3
per aromatic ring are used, so 15 equiv. of SO3 per ligand are used
to carry out the sulfonation of T1NP and T2NP). As an example, the
3
1
1
26.80 (s, C-12 or C-9), 126.90 (s, C-11 or C-10), 127.80 (s, C-10 or C-
1), 128.08 (d, C-7, JC–P = 8.95 Hz), 128.74 (d, C-3, JC–P = 8.95 Hz),
28.93 (s, C-4), 130.10 (d, C-14, JC–P = 17.90 Hz), 133.31 (d, C-13,
JC–P = 7.67), 133.41 (s, C-8), 133.45 (d, C-6, J
3
3
2
1
1
protocol used to do the M1NP sulfonation is given below. The
H
2
2
31
1
= 20.50 Hz), 134.45
and P{ H} NMR data for all sulfonated ligands can be found in the
electronic supplementary material. For each sulfonated phosphane,
the average number of sulfonate groups per phosphane (nS/P) was
determined from elemental analysis. 31P{ H} NMR analysis indi-
cated that less than 5% of phosphine oxide was present in the
sulfonated phosphanes.
C–P
2
1
(
d, C-2, JC–P = 21.73), 134.58 (d, C-5, JC–P = 11.51 Hz), 137.00 (d,
1
31
1
◦
C-1, JC–P = 11.93 Hz);
ı = −5.2 ppm.
P{ H} NMR (121.5 MHz, CDCl , 25 C):
3
1
2
.6.4. D2NP
3
M1NP (1.5 g) was dissolved in 10.9 mL of 100% concentrated
sulfuric acid obtained by mixture of 8.7 mL of concentrated sul-
4
2
1
◦
furic acid (96%) and 2.2 mL of oleum (65%). After cooling to 5 C, the
P
oleum (65%, 1.9 mL, 2.5 equiv. of SO3 per aromatic ring) was slowly
added under vigorous stirring and keeping the temperature below
5
8
1
4
6
7
13
◦
10 C. The reaction mixture was then kept at room temperature for
12
4
days under a nitrogen atmosphere. Excess of SO was transformed
3
11
9
to H SO4 by addition of 12 mL of degassed water. The mixture was
2
10
poured into a mixture of water and ice (28 mL/28 g), and triocty-
lamine (6.8 g, 19 mmol) was then added. The ammonium salt of
the sulfonated phosphane was recovered from the acidic aqueous
layer by addition of chloroform (23 mL) and the organic layer was
washed with water up to neutral pH. The sulfonated salt was recov-
ered by a succession of extractions with NaOH solution (2N). Each
Yield: 70%.
1
◦
H NMR (300 MHz, CDCl , 25 C): ı = 7.43 (m, 10H, H-2, H-3, H-
), 7.48–7.51 (m, 1H, H-14), 7.54–7.58 (m, 2H, H-9, H-10), 7.79–7.81
3
4
(
m, 1H, H-12), 7.86–7.88 (m, 1H, H-7), 7.86–7.94 (m, 2H, H-6, H-11);
1
3
1
◦
C{ H} NMR (75.5 MHz, CDCl , 25 C): ı = 126.00 (s, C-12 or C-9),
3
1
1
26.65 (s, C-12 or C-9), 126.80 (s, C-11 or C-10), 127.50 (s, C-10 or C-
31
1
fraction was analyzed by P{ H} NMR spectroscopy and the purest
3
3
1), 128.00 (d, C-7, JC–P = 8.95 Hz), 128.70 (d, C-3, JC–P = 8.95 Hz),
28.73 (s, C-4), 129.90 (d, C-14, JC–P = 17.90 Hz), 133.25 (d, C-13,
fractions were concentrated under vacuum. Yield: 41%.
2
1
Elemental analysis (%) found:
C 45.30, H 3.31, S 11.82
2
2
JC–P = 7.67 Hz), 133.32 (s, C-8), 133.30 (d, C-6, JC–P = 20.50 Hz),
(nS/P = 2.15); calculated for C22H15Na O PS ·2H O (552.46): C
2
6
2
2
34.45 (d, C-2, JC–P = 21.73 Hz), 134.56 (d, C-5, 1JC–P = 11.51 Hz),
2
1
1
2
47.83, H 3.47, S 11.61.
D1NPS, sulfonation yield: 46%, elemental analysis (%)
found: 52.3, 3.60, 11.01 (nS/P = 2.05); calculated for
C26H17Na O PS ·2H O (602.52): C 51.83, H 3.51, S 10.64.
37.00 (d, C-1, 1JC–P = 11.93 Hz); P{ H} NMR (121.5 MHz, CDCl ,
31
1
3
◦
5 C): ı = −4.4 ppm.
C
H
S
2
6
2
2
2
.6.5. T2NP
T1NPS, sulfonation yield: 49%, elemental analysis (%)
found: 57.01, 3.32, 9.72 (nS/P = 1.92); calculated for
C
H
S
C30H19Na O PS ·2H O (652.58): C 55.22, H 3.55, S 9.83.
2
6
2
2
M2NPS, sulfonation yield: 42%, elemental analysis (%)
found: C 45.77, H 3.52, S 10.03 (nS/P = 1.81); calculated for
C22H15Na O PS ·2H O (552.46): C 47.83, H 3.47, S 11.61.
P 1
2
6
2
2
1
0
2
3
D2NPS, sulfonation yield: 33%, elemental analysis (%)
found: C 50.67, H 3.44, S 11.04 (nS/P = 2.12); calculated for
C26H17Na O PS ·2H O (602.52): C 51.83, H 3.51, S 10.64.
9
8
7
4
2
6
2
2
5
T2NPS, sulfonation yield: 44%, elemental analysis (%)
6
found:
C 55.05, H 3.67, S 9.52 (nS/P = 1.94); calculated for
Yield: 68%.
1
◦
C30H19Na2O6PS2·2H2O (652.58): C 55.22, H 3.55, S 9.83.
H NMR (300 MHz, CDCl , 25 C): ı = 7.85 (m, 1H, H-10), 7.56
3
(
2
(
1
m, 2H, H-5, H-6), 7.79 (m, 1H, H-8), 7.86 (m, 1H, H-3), 7.88 (m,
1
3
1
◦
H, H-2, H-7); C { H} NMR (75.5 MHz, CDCl , 25 C): ı = 126.20
3
3. Results and discussion
s, C-8 or C-5), 126.80 (s, C-8 or C-5), 126.90 (s, C-6 or C-7),
3
27.80 (s, C-6 or C-7), 128.08 (d, C-3, JC–P = 7.67 Hz), 130.10 (d,
3.1. Synthesis and properties of the phosphanes
2
3
C-10, JC–P = 20.5 Hz), 133.31 (d, C-9, JC–P = 4.47 Hz), 133.41 (s, C-
2
1
4
), 133.45 (d, C-2, JC–P = 14.9 Hz), 134.57 (d, C-1, JC–P = 11.51 Hz);
Phosphanes 31P{ H} NMR spectra show different resonances
indicating that the sulfonation is not regioselective when naph-
thyl groups are present (ESI). Moreover, the sulfonation degree of
the different sulfonated naphthylphosphanes is surprisingly not as
high as what is usually observed for typical triphenylphosphane
sulfonation. Indeed, elemental analysis is consistent with an aver-
age sulfonation degree around 2 for the different naphthyl-ligands
1
3
1
1
◦
P{ H} NMR (121.5 MHz, CDCl , 25 C): ı = −4.1 ppm.
3
2
.7. Ligand sulfonation
The same procedure was used to synthesize the different sul-
fonated naphthylphosphanes, with varying the initial quantity of

M. Elard et al. / Catalysis Today 247 (2015) 47–54
51
7
7
6
6
5
5
4
4
3
5
0
5
0
5
0
5
0
5
TPPTS
M1NPS
D1NPS
T1NPS
M2NPS
D2NPS
T2NPS
-1
0
.00001
0.0001
0.001
0.01
0.1 (mol.L )
Fig. 2. Surface tension curves of aqueous solutions of synthesized naphthylphos-
◦
phanes at 25 C.
while sulfonation of triphenylphoshane (TPP) is quantitative
(
sulfonation degree: 3). Even with more drastic experimental con-
dition, i.e. higher SO3 concentration (increased from 2.5 to 5 equiv.
per aromatic ring), higher reaction time (7 days instead of 4)
◦
or higher reaction temperature (40 C instead of room temper-
ature), the formation of sulfonated isomers mixtures with an
average sulfonation degree close to 2 is always observed. These
unexpected results let us believe that steric parameters are respon-
sible for such low sulfonation degree. Indeed, given that naphthyl
rings are more reactive toward electrophilic attack compared to
phenyl rings [39], steric hindrance of the sulfonated naphthyl ring
would hamper a higher sulfonation degree. Such observations have
Fig. 3. Partial 1H NMR spectra of ␤-CD/phosphane mixtures (6 mM/6 mM) in D2O
at 293 K.
ꢀ
already been described when the well-known BINAP ligand (2,2 -
ꢀ
bis(diphenylphosphino)-1,1 -binaphthyl) was sulfonated in similar
conditions. Indeed, a mixture was obtained with a major product
which is told to be tetrasulfonated while 8 different phenyl groups
are reactive toward sulfonation [40,41].
Surface tension measurements were realized in order to evalu-
ate the surface-active properties of the different phosphanes. The
evolution of the surface tension (ꢀ) versus the phosphane concen-
tration in water is presented in Fig. 2.
Although these ligands lead to a gradual decrease in the water
surface tension when their concentration is increased, their curve
profile is different from typical surfactant. Indeed, no break in
the slope of the curves and no plateau were observed. Hence,
although they have hydrotropic properties due to the fact that
their sulfonation degree is lower than that of TPPTS, naphthyl lig-
ands self-aggregation does not occur spontaneously in the aqueous
phase.
␤-CD protons chemical shifts are observed in the case of T2NPS
suggesting a deeper inclusion of the 2-naphthyl group in the ␤-
CD cavity compared to the more crowded 1-naphthyl group. This
assumption was supported by T-ROESY NMR experiments (Fig. 4).
Indeed, the strongest interactions are observed between the inter-
nal protons of the ␤-CD (H-3 and to a lesser extend H-5) and the
aromatic protons of T2NPS.
Finally, 1H MNR studies were also performed on various mix-
tures of RAME-␤-CD and sulfonated naphthylphosphanes. The
NMR spectra clearly demonstrate that sulfonated naphthylphos-
phanes can also interact with the methylated CDs (ESI). UV–visible
experiments were performed to evaluate the affinity of these
phosphanes for RAME-␤-CD. By considering a 1:1 stoichiome-
try for the RAME-␤-CD/ligand inclusion complexes, it was found
that the association constants were low for all the phosphanes
−
1
(
<500 M ). It must be pointed out that these association constant
3.2. Studies of ˇ-CD/phosphane interaction
values are lower than that found for the RAME-␤-CD/TPPTS com-
−
1
plex (840 M ), confirming that sulfonated naphthylphosphanes
are bulkier that the TPPTS [36].
NMR spectroscopic studies have been performed to provide
information concerning the potential interactions between these
water-soluble naphthylphosphanes and ␤-CD in aqueous medium.
The NMR spectra of the different samples containing mixtures of
3.3. Catalytic experiments
␤
-CD and phosphanes are presented in Fig. 3.
Each spectrum denotes chemical shift variations for the host
The 1-decene biphasic hydroformylation reaction results are
summarized in Table 1. Although isomers mixture with an aver-
age of 2 sulfonate groups by phosphane is used, it is very important
to emphasize that similar activities and selectivities are obtained
with different catalysts prepared from several batches of ligands.
By way of comparison, results obtained with TPPTS are reminded
(entries 1 and 2). Two groups of results can be highlighted.
The first group involves the experiments conducted with
sulfonated 1-naphthyl ligands (entries 3–8) and the second
group is related to the use of sulfonated 2-naphthyl ligands
(entries 9–14). When 1-naphthyl ligands are used without CD,
protons. The results obtained with M1NPS, D1NPS, M2NPS
and D2NPS are not surprising and clearly indicate an inter-
action between the two compounds. Indeed, these 4 modified
naphthylphosphanes possess a m-sulfonatophenyl group or a non-
sulfonated phenyl group which are known to form an inclusion
complex with ␤-CD and RAME-␤-CD [30,42]. In the same way,
inclusion via a sulfonated or non-sulfonated 1-naphthyl or 2-
naphthyl group is not excluded as T1NPS and T2NPS, which do
not possess phenyl group, also lead to spectral modifications. Nev-
ertheless, for these two compounds, the largest differences in the

5
2
M. Elard et al. / Catalysis Today 247 (2015) 47–54
Fig. 4. Partial contour plot of the T-ROESY spectrum of a solution containing ␤-CD (6 mM) and a) T1NPS or b) T2NPS (6 mM) in D2O at 298 K with a 300 ms mixing time.
conversions are unexpectedly high compared to conversion
obtained with TPPTS. Moreover, a gradual increase in the conver-
sion is observed when the number of 1-naphthyl substituents on
the phosphorus atom increased (from M1NPS to T1NPS). These
results could be explained by considering the equilibria between
the different catalytic species. During the biphasic aqueous hydro-
formylation, these equilibria depend on ligand exchange between
phosphane and carbon monoxide (equilibria 1—Scheme 3).
With sterically demanding ligands, it is well known that
the equilibria 1 are shifted toward phosphine low coordinated
species (II) which are more reactive. Finally, it is worth men-
tioning that the organic phases at the end of the reaction
are brown, suggesting that a part of the catalyst is leached
into these phases. This leaching is also consistent with the
formation of low-coordinated catalytic species under reaction
conditions and confirms the poor coordinating properties of the
sulfonated 1-naphthylphosphanes. In the presence of RAME-␤-CD,
the phenomena observed without CD are amplified. A complete
conversion and a catalyst leaching are observed in all cases (entries
4, 6 and 7). These results could be attributed to the formation
of inclusion complexes between RAME-␤-CD and sulfonated 1-
naphthylphosphanes (equilibrium 2 in Scheme 3). In fact, we
postulate that the shift of equilibria 1 toward the formation of
low-coordinated catalytic species is amplified by the trapping
of the sulfonated 1-naphthylphosphanes into the RAME-␤-CD
cavity.
Table 1
Cyclodextrin-based biphasic 1-decene hydroformylation reaction.
CHO
Rh-catalyst
CHO
n-C8H17
+ CO + H2
n-C8H17
+
n-C8H17
Cyclodextrin
linear (l)
branched (b)
Entry
Ligand
CD
Conversion
Aldehydes selectivitya
l/b ratio
Organic layer aspect after reaction
1
2
3
4
5
6
7
8
9
1
1
1
1
1
TPPTS
TPPTS
–
3
97
50
100
83
100
100
100
5
40
4
39
3
59
98
65
60
66
62
63
63
96
96
91
90
94
95
2.8
1.8
2.1
1.9
1.4
1.9
1.5
1.5
2.9
2.1
2.9
2.1
2.6
2
Colorless
Colorless
RAME-␤-CD
–
RAME-␤-CD
–
RAME-␤-CD
–
RAME-␤-CD
–
RAME-␤-CD
–
RAME-␤-CD
–
RAME-␤-CD
M1NPS
M1NPS
D1NPS
D1NPS
T1NPS
T1NPS
M2NPS
M2NPS
D2NPS
D2NPS
T2NPS
T2NPS
Dark brown
Dark brown
Dark brown
Dark brown
Dark brown
Dark brown
Colorless
Colorless
Colorless
Colorless
Colorless
0
1
2
3
4
41
Colorless
Experimental conditions: Rh(acac)(CO)2 = 21 ␮mol (1 equiv.), ligand = 107 ␮mol (5 equiv.), RAME-␤-CD = 210 ␮mol (10 equiv.) when used, 1-decene = 10.5 mmol (500 equiv.),
◦
water = 6 mL, 80 C, 50 bar CO/H2 (1/1), 1500 rpm, reaction time = 6 h.
a
Aldehydes selectivity: (mol. of aldehydes)/(mol. of converted olefins) × 100; the side products are mainly isomeric olefins; hydroformylation of the internal olefins was
not observed.

M. Elard et al. / Catalysis Today 247 (2015) 47–54
53
4. Conclusion
This work shows that the sulfonation of naphthylphos-
phanes is possible and that the average sulfonation degree is
unexpectedly around 2 whatever the reaction conditions. All
the synthesized sulfonated naphthylphosphanes associated to a
rhodium precursor are able to perform aqueous hydroformy-
lation of 1-decene with or without cyclodextrin but only the
sulfonated 2-naphthylphosphanes lead to stable catalytic sys-
tem. These results are rationalized by considering that sulfonated
1
-naphthylphosphanes are bulkier ligands than sulfonated 2-
naphthylphosphanes. Works are currently under progress to
evaluate the behavior of these new ligands in other reac-
tions requiring sterically encumbered phosphanes such as the
Suzuki–Miyaura coupling of challenging aryl chlorides.
Acknowledgements
This work was supported by the Centre National de la Recherche
Scientifique (CNRS) which is gratefully acknowledged. Similarly,
Roquette Frères (Lestrem, France) is gratefully acknowledged for
financial support.
Appendix A. Supplementary data
Supplementary material related to this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.cattod.
2
014.06.002.
References
Scheme 3. Equilibria between the different catalytic species during aqueous
biphasic hydroformylation assisted by RAME-␤-CD.
[
[
[
[
[
1] K.H. Shaughnessy, Chem. Rev. 109 (2009) 643–710.
2] L. Obrecht, P.C.J. Kamer, W. Laan, Catal. Sci. Technol. 3 (2013) 541–551.
3] P. Purwanto, H. Delmas, Catal. Today 24 (1995) 135–140.
4] P. Kalck, L. Miquel, M. Dessoudeix, Catal. Today 42 (1998) 431–440.
5] H. Fu, M. Li, H. Mao, Q. Lin, M. Yuan, X. Li, H. Chen, Catal. Commun. 9 (2008)
The results obtained with the sulfonated 2-naphthyl ligands
notably differ from those obtained with sulfonated 1-naphthyl
ligands (entries 9, 11 and 13). In the absence of RAME-␤-CD,
the conversions are low (<5%) and the organic phases are col-
orless, suggesting that the catalyst is totally immobilized into
the aqueous phase and that the reaction is limited by mass
transfer. In the presence of RAME-␤-CD, an increase in the
conversion is clearly observed for the three ligands (entries
1
539–1544.
[6] T. Hamerla, A. Rost, Y. Kasaka, R. Schomäcker, ChemCatChem
854–1862.
[
[
5
(2013)
1
7] A.T. Straub, M. Otto, I. Usui, B. Breit, Adv. Synth. Catal. 355 (2013) 2071–2075.
8] S.L. Desset, S.W. Reader, D.J. Cole-Hamilton, Green Chem. 11 (2009) 630–637.
[9] H. Nowothnick, A. Rost, T. Hamerla, R. Schomäcker, C. Müller, D. Vogt, Catal.
Sci. Technol. 3 (2013) 600–605.
10] H. Chen, Y. Li, R. Li, P. Cheng, X. Li, J. Mol. Catal. A: Chem. 198 (2003) 1–7.
11] K. Kunna, C. Müller, J. Loos, D. Vogt, Angew. Chem. Int. Ed. Engl. 45 (2006)
7289–7292.
12] F. Hapiot, L. Leclercq, N. Azaroual, S. Fourmentin, S. Tilloy, E. Monflier, Curr. Org.
Synth. 5 (2008) 162–172.
[
[
[
[
1
0, 12 and 14) and the organic phases are always colorless.
Interestingly, the reusability of the RAME-␤-CD/ sulfonated 2-
naphthylphosphanes systems could be demonstrated as 2 catalytic
cycles were performed without loss of activity and selectivity
with the T2NPS ligand. To unambiguously prove the reusability
of the RAME-␤-CD/ sulfonated 2-naphthylphosphanes systems,
the organic phase of entry 14 was recovered after reaction and
ICP-AES analyses were performed. No trace of rhodium could
be detected, i.e., below the detection limit (0.2 ppm). Finally, it
is also worth mentioning that the formation of emulsion was
never observed at the end of the reaction and that the phase
separation is very fast (<1 min). From a regioselectivity point
of view, a small decrease is always observed with a l/b ratio
around 2.1. All of these results demonstrate that the RAME-␤-
CD/sulfonated 2-naphthyl ligands combinations allow to achieve
the 1-decene hydroformylation in an aqueous biphasic system.
Unfortunately, no notable increase in the regio- or chemoselectiv-
ity could be observed relative to RAME-␤-CD/TPPTS combination.
It appears also that the affinity of sulfonated 2-naphthyl ligands for
RAME-␤-CD is yet too high to avoid its influence on the equilib-
ria between the different catalytic species. Indeed, a l/b ratio
decrease is observed when RAME-␤-CD is added into the reaction
medium.
13] H. Ding, B.E. Hanson, T. Bartik, B. Bartik, Organometallics 13 (1994) 3761–3763.
[14] H. Fu, M. Li, J. Chen, R. Zhang, W. Jiang, M. Yuan, H. Chen, X. Li, J. Mol. Catal. A:
Chem. 292 (2008) 21–27.
[
15] M. Ferreira, H. Bricout, N. Azaroual, C. Gaillard, D. Landy, S. Tilloy, E. Monflier,
Adv. Synth. Catal. 352 (2010) 1193–1203.
[16] M. Ferreira, H. Bricout, N. Azaroual, D. Landy, S. Tilloy, F. Hapiot, E. Monflier,
Adv. Synth. Catal. 354 (2012) 1337–1346.
[
[
[
[
17] S.M. Mercer, T. Robert, D.V. Dixon, P.G. Jessop, Catal. Sci. Technol. 2 (2012)
1315–1318.
18] A. Buhling, P.C.J.J. Kamer, P.W.N.M. van Leeuwen, J.W. Elgersma, P.W.N.M. Van
Leeuwen, J. Mol. Catal. A: Chem. 116 (1997) 297–308.
19] S.L. Desset, D.J. Cole-Hamilton, Angew. Chem. Int. Ed. Engl. 48 (2009)
1472–1474.
20] C. Liu, J. Jiang, Y. Wang, F. Cheng, Z. Jin, J. Mol. Catal. A: Chem. 198 (2003) 23–27.
[21] A.J. Sandee, V.F. Slagt, J.N.H. Reek, P.C.J. Kamer, P.W.N.M. van Leeuwen, Chem.
Commun. (1999) 1633–1634.
[
[
22] I. Tóth, I. Guo, B.E. Hanson, T. Imre, J. Mol. Catal. A: Chem. 116 (1997) 217–229.
23] C. Disser, C. Muennich, G. Luft, Appl. Catal. A: Gen. 296 (2005) 201–208.
[24] J. Boulanger, A. Ponchel, H. Bricout, F. Hapiot, E. Monflier, Eur. J. Lipid Sci.
Technol. 114 (12) (2012) 1439–1446.
[
[
25] D.E. Bergbreiter, J. Tian, C. Hongfa, Chem. Rev. 109 (2009) 530–582.
26] F. Hapiot, H. Bricout, S. Tilloy, E. Monflier, Top. Curr. Chem. 342 (2013) 49–78.
[27] F.X. Legrand, M. Sauthier, C. Flahaut, J. Hachani, C. Elfakir, S. Fourmentin, S.
Tilloy, E. Monflier, J. Mol. Catal. A: Chem. 303 (2009) 72–77.
[
28] M. Ferreira, F.X. Legrand, C. Machut, H. Bricout, S. Tilloy, E. Monflier, Dalton
Trans. 41 (2012) 8643–8647.
[29] E. Monflier, S. Tilloy, C. Méliet, A. Mortreux, S. Fourmentin, D. Landy, G. Sur-
pateanu, New J. Chem. 23 (1999) 469–472.

5
4
M. Elard et al. / Catalysis Today 247 (2015) 47–54
[
[
[
[
[
[
30] L. Caron, S. Tilloy, E. Monflier, J.-M. Wieruszeski, G. Lippens, D. Landy, S. Four-
mentin, G. Surpateanu, J. Incl. Phenom. Macrocycl. Chem. 38 (2000) 361–379.
31] C. Binkowski, J. Cabou, H. Bricout, F. Hapiot, E. Monflier, J. Mol. Catal. A: Chem.
[36] M. Ferreira, H. Bricout, A. Sayede, A. Ponchel, S. Fourmentin, S. Tilloy, E. Monflier,
Adv. Synth. Catal. 350 (2008) 609–618.
[37] R. Gärtner, B. Cornils, H. Springer, P. Lappe, German Patent DE 3235030, 1982.
[38] D. Landy, S. Fourmentin, M. Salome, G. Surpateanu, J. Incl. Phenom. Macrocycl.
Chem. 38 (2000) 187–198.
[39] A.R. Katritzky, M.S. Kim, D. Fedoseyenko, K. Widyan, M. Siskin, M. Francisco,
Tetrahedron 65 (2009) 1111–1114.
[40] K. Wan, M.E. Davis, J. Chem. Soc. Chem. Commun. (1993) 1262–1264.
[41] M. Berthod, G. Mignani, G. Woodward, M. Lemaire, Chem. Rev. 105 (2005)
1801–1836.
2
15 (2004) 23–32.
32] E. Monflier, H. Bricout, F. Hapiot, S. Tilloy, A. Aghmiz, A.M. Masdeu-bultó, Adv.
Synth. Catal. 346 (2004) 425–431.
33] L. Leclercq, M. Sauthier, Y. Castanet, A. Mortreux, H. Bricout, E. Monflier, Adv.
Synth. Catal. 347 (2005) 55–59.
34] P. Blach, D. Landy, S. Fourmentin, G. Surpateanu, H. Bricout, A. Ponchel, F. Hapiot,
E. Monflier, Adv. Synth. Catal. 347 (2005) 1301–1307.
35] D. Kirschner, T. Green, F. Hapiot, S. Tilloy, L. Leclercq, H. Bricout, E. Monflier,
Adv. Synth. Catal. 348 (2006) 379–386.
[42] M. Canipelle, L. Caron, C. Christine, S. Tilloy, E. Monflier, Carbohydr. Res. 337
(2002) 281–287.