DENDRI-POM hybrids based on the Keggin, Dawson, Preyssler and Venturello polyanions and their catalytic evaluation in oxidation reactions

Article abstract of DOI:10.1016/j.poly.2013.04.013

A series of four dendritic polyoxometalate hybrids (DENDRI-POMs) have been synthesized by electrostatic coupling of diallyl carbinol benzyl ammonium dendron with the Keggin [PW₁₂O₄₀]^3-, Dawson [P₂W₁₈O₆₂]^6-, Preyssler [NaP ₅W₃₀O₁₁₀]^14- and Venturello [PO ₄(WO(O₂)₂)₄]^3- polyanions, respectively. The structures of these DENDRI-POMs were examined in solution by standard physicochemical techniques such as IR and NMR spectroscopy as well as elemental analysis. Particular focus was devoted to ³¹P and ¹⁸³W NMR spectroscopy of these hybrids, as these techniques offer evidence for the presence of the intact polyanion in the corresponding hybrids. The ¹⁸³W NMR spectrum of the Keggin-based DENDRI-POM showed the expected signal, whereas no signal was observed for the other hybrids. On the other hand, the expected ³¹P NMR signal was observed for all four DENDRI-POMs. These POM-based dendrimers oxidized methyl phenyl sulfide to the corresponding sulfoxide and sulfone, with the Venturello-based DENDRI-POM being the most efficient catalyst. In addition, the latter catalyzed efficiently the oxidation of cyclooctene to the corresponding epoxide, whereas hybrids based on the other three polyanions were inert. Catalyst recovery experiments showed that, after two cycles of oxidation reactions, the activity of the Keggin- and Venturello-based DENDRI-POMs were not changed, whereas that of the Dawson-based hybrid decreased dramatically after the first cycle.

Full text of DOI:10.1016/j.poly.2013.04.013

Polyhedron 57 (2013) 57–63
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
DENDRI-POM hybrids based on the Keggin, Dawson, Preyssler and Venturello
polyanions and their catalytic evaluation in oxidation reactions
a,
Claire Jahier a, Sib Sankar Mal b, Rami Al-Oweini b, Ulrich Kortz b, , Sylvain Nlate
⇑
⇑
a Institut Européen de Chimie et Biologies, IECB-CBMN, UMR 5248 CNRS, Université Bordeaux 1, 2 Rue Robert Escarpit, 33607 Pessac Cedex, France
b School of Engineering and Science, Jacobs University, P.O. Box 750 561, 28725 Bremen, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of four dendritic polyoxometalate hybrids (DENDRI-POMs) have been synthesized by electro-
static coupling of diallyl carbinol benzyl ammonium dendron with the Keggin [PW12O40]
3À, Dawson
Received 13 February 2013
Accepted 4 April 2013
Available online 16 April 2013
14À
[P2W18O62
]
6À, Preyssler [NaP5W30O110
]
and Venturello [PO4(WO(O2)2)4]3À polyanions, respectively.
The structures of these DENDRI-POMs were examined in solution by standard physicochemical tech-
niques such as IR and NMR spectroscopy as well as elemental analysis. Particular focus was devoted to
31P and 183W NMR spectroscopy of these hybrids, as these techniques offer evidence for the presence
of the intact polyanion in the corresponding hybrids. The 183W NMR spectrum of the Keggin-based DEN-
DRI-POM showed the expected signal, whereas no signal was observed for the other hybrids. On the other
hand, the expected 31P NMR signal was observed for all four DENDRI-POMs. These POM-based
dendrimers oxidized methyl phenyl sulfide to the corresponding sulfoxide and sulfone, with the
Venturello-based DENDRI-POM being the most efficient catalyst. In addition, the latter catalyzed
efficiently the oxidation of cyclooctene to the corresponding epoxide, whereas hybrids based on the other
three polyanions were inert. Catalyst recovery experiments showed that, after two cycles of oxidation
reactions, the activity of the Keggin- and Venturello-based DENDRI-POMs were not changed, whereas
that of the Dawson-based hybrid decreased dramatically after the first cycle.
Keywords:
Polyoxometalate
Dendrimer
183W NMR spectroscopy
Oxidation-catalysis
Catalyst recovery
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
solubility, catalytic efficiency, stability and recovery capacity [6].
Thus, by using an appropriate organic dendritic countercation, a
Polyoxometalates (POMs) [1] are discrete, anionic metal-oxide
clusters with a diverse range of properties that render them attrac-
tive candidates for potential applications in a variety of fields such
as catalysis, medicine, analytical chemistry and material science
[2]. Owing to their unique metal-oxo cage structure and reversible
electron transformation abilities [3], POMs have been extensively
investigated as catalysts in a variety of oxidation reactions, such
as epoxidation and desulfurization [4].
However, in order to study the activity of POMs as homogenous
catalysts in common organic solvents and due to the poor solubil-
ity of most alkali-POM salts in organic media, organic salts of these
POM catalysts need to be prepared. This can usually be accom-
plished by cation exchange. Hence, to overcome the poor solubility
problem of alkali-POM salts in organic solvents, a variety of cat-
ionic organic moieties have been used as countercations of polyan-
ions [5], resulting in a wide range of hybrids soluble in organic
media. Amongst these organic cations used with polyanions, den-
dritic cations have been recently reported. It has been shown that
dendritic cations modify some properties of polyanions, such as
variety of inorganic–organic hybrids with specific properties can
be prepared.
For several years, we have been working on the preparation and
characterization of POM-based dendritic hybrids, in order to eval-
uate the influence of the dendritic wedges on the properties of the
polyanions. Accordingly, we have prepared various DENDRI-POM
salts, based on electrostatic bonding between various POM polyan-
ions and dendritic cations. These DENDRI-POMs have been used as
recoverable catalysts in the oxidation of alkenes, alcohols and sul-
fides [6d–n]. These studies have clearly shown that a close rela-
tionship exists between the structure of the dendritic
countercation and the properties of the hybrid polyanion. More re-
cently, we have reported optically active DENDRI-POMs, prepared
by electrostatic coupling of achiral polyanions with enantiopure
dendritic cations [6j–l]. These optically active DENDRI-POMs cata-
lyzed oxidation reactions with enantioselectivity, highlighting a
chirality transfer from the dendritic wedges to the catalytically ac-
tive polyanion. Recently, Musumeci et al. have also shown that the
nature of the countercation can affect the self-assembly processes
of hybrid polyoxometalates on surfaces [7]. Thus, due to the inter-
est in dendritic materials on the one hand and POM chemistry on
the other hand, exploring new DENDRI-POM hybrid materials
remains an interesting and challenging topic for chemists. The
⇑
Corresponding authors. Fax: +33 5 4000 2215 (S. Nlate).
E-mail addresses: u.kortz@jacobs-university.de (U. Kortz), s.nlate@iecb.u-bor-
deaux.fr (S. Nlate).
0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.04.013

58
C. Jahier et al. / Polyhedron 57 (2013) 57–63
chemistry of DENDRI-POMs is not yet well understood, due to the
difficulty of determining their real structure in solution. To date,
the determination of the structure of DENDRI-POMs is usually
based on solution NMR spectra (13C, 1H, 31P), IR spectroscopy and
elemental analysis. It would be desirable to also use 183W NMR
spectroscopy, which is a widely applied technique to investigate
the solution stability of POMs [8], as well as X-ray diffraction to
gain more information about the structure of DENDRI-POMs. A
large number of polyoxotungstates have been characterized by
solution 183W NMR, whereas attempts to identify the structure of
polytungstates in DENDRI-POM assemblies by solution 183W
NMR were unsuccessful. In 183W NMR spectroscopy, the chemical
shifts of the tungsten atoms are very sensitive to their surrounding
environment, including electric charge, type of adjacent elements,
counterions and solvent, which is reflected by a very large chemi-
cal shift range. Hence we decided to study the influence of the den-
dritic structure on the 183W NMR chemical shifts of four
structurally different tungstophosphates, compared to the free
polyanions. Thus, using the diallyl carbinol benzyl ammonium
moval of the solvent under vacuum, the residue was washed with
petroleum ether (3 Â 30 mL) and dried under vacuum, yielding the
ammonium salt 2 with 95% yield (3.40 g). 1H NMR (250.13 MHz,
CDCl3): d = 7.80 (d, 2H, Ar), 7.52 (d, 2H, Ar), 4.89 (s, 2H, CH2–N),
3.72 (s, 3H, O–CH3), 3.14 (broad, 6H, CH2–N), 1.60 (broad, 6H, CH2),
1.09 (broad, 18H, CH2), 0.59 (broad, 9H, CH3). 13C NMR
(62.91 MHz, CDCl3): d = 165.8 (C@O), 132.8 (CH, Ar), 132.4 (Cq, Ar),
131.8 (Cq, Ar), 129.9 (CH, Ar), 62.3 (CH2–N), 59.1 (CH2–N), 52.3 (O–
CH3), 31.0 (CH2), 25.9 (CH2), 22.6 (CH2), 22.3 (CH2), 13.7 (CH3). Anal.
Calc. for C27H48BrNO2: C, 65.04; H, 9.70. Found: C, 64.94; H, 9.76%.
2.2.2. Synthesis of diallyl carbinol benzyl ammonium 3
In a Schlenk tube, allyl bromide (1.74 mL, 2.43 g, 20.06 mmol)
was slowly added to a mixture of magnesium (0.54 g, 22.06 mmol)
in anhydrous diethyl ether (15 mL) cooled to 0 °C. Then, a solution
of trihexylammonium benzoate 2 (2.00 g, 4.01 mmol) in anhydrous
diethyl ether (10 mL) was added to the reaction mixture at 0 °C.
The mixture was stirred for 2 h at room temperature. Then solvent
was removed under vacuum and the product was extracted with
dichloromethane (3 Â 20 mL) and dried over sodium sulfate. The
solvent was removed under vacuum and the product 3 was ob-
tained as colorless oil in 93% yield (2.05 g). 1H NMR (250.13 MHz,
CDCl3, TMS): d = 7.46 (s, 4H, Ar), 5.59–5.46 (m, 2H, CH@CH2),
5.08–5.02 (m, 4H, CH@CH2), 4.83 (s, 2H, CH2–N), 3.27 (broad, 6H,
CH2–N), 2.60 (dm, 4H, CH2–CH@CH2), 2.44 (s, 1H, OH), 1.73 (broad,
6H, CH2), 1.29 (broad, 18H, CH2), 0.85 (broad, 9H, CH3). 13C NMR
(62.91 MHz, CDCl3, TMS): d = 149.0 (Cq, Ar), 132.8 (CH, Ar), 132.2
(CH@CH2), 126.6 (CH, Ar), 125.6 (Cq, Ar), 119.7 (CH@CH2), 75.0
(Cq, C–OH), 62.9 (CH2–N), 58.8 (CH2–N), 46.6 (CH2–CH@CH2),
31.2 (CH2), 26.1 (CH2), 22.7 (CH2), 22.4 (CH2), 13.9 (CH3). MS
(ESI): Calc. for C32H56NO MÀBr+ = 470.80, found 470.43. Anal. Calc.
for C32H56BrNO: C, 69.79; H, 10.25. Found: C, 69.19; H, 10.23%.
dendron as the model countercation, we have prepared four DEN-
3À
DRI-POM hybrids based on the Keggin [PW12O40
]
[9], Dawson
6À
14À
[P2W18O62
]
[10], Preyssler [NaP5W30O110
]
[11], and Ventu-
rello [PO4(WO(O2)2)4]3À [12] polyanions, respectively. We studied
the solution 183W NMR of these DENDRI-POMs, in order to demon-
strate the presence of the intact polyanion. We also investigated
the catalytic efficiency of these DENDRI-POMs in the oxidation of
methyl phenyl sulfide and cyclohexene with H2O2, as well as their
recovery capacities.
2. Experimental
2.1. General remarks
2.3. Synthesis of DENDRI-POMs
Reagent-grade tetrahydrofuran (THF) and diethyl ether were
predried over Na foil and distilled from sodium-benzophenone
under argon atmosphere immediately before use. Acetonitrile
(CH3CN) was stirred under argon atmosphere overnight over phos-
phorus pentoxide, distilled from sodium carbonate, and stored un-
der argon atmosphere. Methylene chloride (CH2Cl2) was distilled
from calcium hydride just before use. All other chemicals were
used as received. The 1H, 13C, 31P NMR spectra were recorded at
25 °C with a Bruker AC 250 FT spectrometer (1H: 250.13, 13C:
62.91 MHz) and a Bruker AC 200 FT spectrometer (1H: 200.16,
13C: 50.33, 31P: 81.02 MHz) at CESAMO (Bordeaux, France). The
183W NMR spectra were recorded on a 400 MHz JEOL ECX instru-
ment and performed in 10 mm tubes (183W: 16.67 MHz) at Jacobs
University Bremen. Mass spectra were performed by the CESAMO
on a QStar Elite mass spectrometer (Applied Biosystems). The
instrument is required with an ESI source and spectra were re-
corded in the positive mode. The electrospray needle was main-
tained at 4500 V and operated at room temperature. Samples
were introduced by injection through a 10 L sample loop into a
200 L/min flow of methanol from the LC pump. Elemental analyses
were performed at the Vernaison CNRS center at Lyon-Villeurb-
anne (France). The infrared spectra were recorded in KBr pellets
on a FT-IR Paragon 1000 Perkin–Elmer spectrometer, unless other-
wise indicated. Organic oxidation products were identified by cor-
relation to authentic samples.
2.3.1. Synthesis of Keggin-based DENDRI-POM 4
To an aqueous solution (2 mL) of Na3[PW12O40
] (0.429 g,
0.140 mmol) was added a solution of trihexylammonium diallyl
carbinol salt 3 (0.253 g, 0.462 mmol) in dichloromethane (5 mL).
The mixture was vigorously stirred at room temperature for 2 h.
The dichloromethane layer was dried over sodium sulfate, filtered
and evaporated under vacuum. The residue was washed three times
with diethyl ether to provide the DENDRI-POM 4 as a light yellow
solid (0.564 g, 94%). 1H NMR (250.13 MHz, [D6]acetone, TMS):
d = 7.57 (m, 12H, Har), 5.65 (m, 6H, CH@CH2), 4.95 (m, 12H,
CH@CH2), 4.67 (s, 6H, CH2–N), 3.31 (m, 18H, CH2–N), 2.60 (dd,
12H, CH2–CH@CH2), 1.95 (m, 18H, CH2), 1.38 (m, 54H, CH2), 0.88
(t, 27H, CH3). 13C NMR (62.91 MHz, [D6]acetone, TMS): d = 149.3
(Cq, Ar), 133.8 (CH@CH2), 131.8 (CH, Ar), 126.6 (CH, Ar), 125.3
(Cq, Ar), 117.5 (CH@CH2), 75.1 (Cq, C–OH), 58.3 (CH2–N), 53.4
(CH2–N), 46.6 (CH2–CH@CH2), 31.0 (CH2), 25.7 (CH2), 23.7 (CH2),
22.2 (CH2), 21.7 (CH2), 13.4 (CH3). 31P NMR (81.02 MHz, [D6]
acetone): d = À15.1 (PO4). 183W NMR (16.67 MHz, [D6]acetone):
d = À88.6; FTIR (KBr pellets),
m
(cmÀ1) = 1079 (P–O)as, 978(W–O)as,
894 (W–O–W)as, 815 (W–O–W)as. Anal. Calc. for C96H168O43N3
PW12: C, 26.88; H, 3.95; N, 0.98; P, 0.72; W, 51.43. Found: C,
26.22; H, 3.87; N, 1.07; P, 0.92; W, 52.23%.
2.3.2. Synthesis of Dawson-based DENDRI-POM 5
To an aqueous solution (2 mL) of K6[P2W18O62
] (0.450 g,
2.2. Synthesis of diallyl carbinol benzyl ammonium dendron 3
0.098 mmol) was added a solution of trihexylammonium diallyl
carbinol salt 3 (0.354 g, 0.647 mmol) in dichloromethane (5 mL).
The mixture was vigorously stirred at room temperature for 2 h.
The dichloromethane layer was dried over sodium sulfate, filtered
and evaporated under vacuum. The residue was washed three
times with diethyl ether to provide DENDRI-POM 5 as a light
2.2.1. Synthesis of trihexylammonium benzoate 2
In a Schlenk tube, a mixture of 4-(bromomethyl)benzoate 1
(1.67 g, 7.29 mmol) and 12 mL of tri-n-hexylamine (12 mL, 9.80 g,
36.40 mmol) in 3 mL of CH3CN was stirred for 16 h at 80 °C. After re-

C. Jahier et al. / Polyhedron 57 (2013) 57–63
59
yellow solid (0.634 g, 92%). 1H NMR (250.13 MHz, [D6]acetone,
TMS): d = 7.55 (broad, 24H, Har), 5.65 (m, 12H, CH@CH2), 4.95
(m, 24H, CH@CH2), 4.83 (s, 12H, CH2–N), 3.37 (broad, 36H, CH2–
N), 2.58 (broad, 24H, CH2–CH@CH2), 1.93 (broad, 36H, CH2),
1.46–1.25 (broad, 108H, CH2), 0.86 (broad, 54H, CH3). 13C NMR
(62.91 MHz, [D6]acetone, TMS): d = 148.9 (Cq, Ar), 134.07
(CH@CH2), 132.1 (CH, Ar), 126.3 (CH, Ar), 126.0 (Cq, Ar), 117.3
(CH@CH2), 75.0 (Cq, C–OH), 59.5 (CH2–N), 53.2 (CH2–N), 46.3
(CH2–CH@CH2), 31.0 (CH2), 25.7 (CH2), 22.6 (CH2), 22.2 (CH2),
21.8 (CH2), 13.5 (CH3). 31P NMR (81.02 MHz, [D6]acetone):
H2O2 (35% in water) was added to the reaction mixture at 35 °C.
The latter was stirred and monitored by 1H NMR. Upon reaction
completion, the organic layer was separated and concentrated un-
der vacuum to about 0.2 mL. The catalyst was precipitated by addi-
tion of Et2O (5 mL). The solid was filtered, washed with diethyl
ether (3 Â 1 mL) and dried under vacuum to yield the DENDRI-
POM catalyst, which was analyzed by 1H and 31P NMR spectros-
copy, before a new catalytic experiment was performed. The
diethyl ether solution was evaporated under vacuum and the oxi-
dized product was purified by chromatography on a silica gel col-
umn (petroleum ether/diethyl ether 1:9; v/v. The catalyst was
recovered following the typical procedure and conditions de-
scribed above for the first cycle, the organic solvent and the reac-
tants being adjusted to the amount of catalyst used.
d = À12.9 (PO4). FTIR (KBr pellets),
m
(cmÀ1) = 1089 (P–O)as,
955(W–O)as, 908 (W–O–W)as, 792 (W–O–W)as. Anal. Calc. for C192-
H336O68N6P2W18: C, 32.08; H, 4.71; N, 1.17; P, 0.86; W, 46.04.
Found: C, 32.20; H, 4.55; N, 1.22; P, 1.09; W, 45.12%.
2.3.3. Synthesis of Preyssler-based DENDRI-POM 6
3. Results and discussion
To an aqueous solution (2 mL) of (NH4)14[NaP5W30O110
]
(0.332 g, 0.040 mmol) was added a solution of trihexylammonium
diallyl carbinol salt 3 (0.300 g, 0.548 mmol) in dichloromethane
(5 mL). The mixture was vigorously stirred at room temperature
for 4 h. The dichloromethane layer was dried over sodium sulfate,
filtered and evaporated under vacuum. The residue was washed
three times with diethyl ether to provide the DENDRI-POM 6 as a
light yellow solid (0.337 g, 60%). 1H NMR (250.13 MHz, [D6]ace-
tone, TMS): d = 7.38 (broad, 56H, Har), 5.53 (m, 28H, CH@CH2),
4.95 (m, 56H, CH@CH2), 4.83 (s, 28H, CH2–N), 3.20 (broad, 384H,
CH2–N), 2.50 (broad, 56H, CH2–CH@CH2), 1.70 (broad, 84H, CH2),
1.22 (broad, 252H, CH2), 0.79 (broad, 126H, CH3). 13C NMR
(62.91 MHz, CDCl3, TMS): d = 142.9 (Cq, Ar), 128.0 (CH@CH2),
127.2 (CH, Ar), 121.1 (CH, Ar), 120.8 (Cq, Ar), 113.6 (CH@CH2),
69.6 (Cq, C–OH), 53.3 (CH2–N), 40.9 (CH2–N), 26.1 (CH2–CH@CH2),
25.6 (CH2), 25.2 (CH2), 24.2 (CH2), 17.2 (CH2), 8.7 (CH3). 31P NMR
3.1. Synthesis and characterization of DENDRI-POM hybrids 4–7
3.1.1. Synthesis of 4–7
DENDRI-POM salts of the various polyanions were obtained by
electrostatic coupling of diallyl carbinol benzyl ammonium den-
dron 3, prepared in two steps as summarized in Scheme 1, with
the Keggin [PW12O40
] ]
3À, Dawson [P2W18O62 6À, Preyssler [NaP5
W30O110
]
14À, and Venturello [PO4(WO(O2)2)4]3À ions respectively.
The first step involves benzylation of trihexylamine by 4-(bro-
momethyl)-benzoate 1 to give the corresponding trihexyl-ammo-
nium benzoate 2. The allylation of
2 with allylmagnesium
bromide gives the diallyl carbinol benzyl ammonium dendron 3
in 93% yield. NMR and mass spectrometry as well as elemental
analysis of compound 3 are consistent with the proposed structure.
The reaction of 3 with the sodium salt of the Keggin ion Na3
[PW12O40], the potassium salt of the Dawson ion K6[P2W18O62
and the ammonium salt of the Preyssler ion (NH4)14[NaP5W30O110
(81.02 MHz, [D6]acetone): d = À9.6 (PO4); FTIR (KBr pellets),
m
(cmÀ1) = 1162 (P–O)as, 1076 (W–O)as, 910 (W–O–W)as, 733 (W–
O–W)as, 571, 537. Anal. Calc. for C448H784O124N14P5NaW30: C,
38.31; H, 5.63; N, 1.40; P, 1.10; W, 39.27. Found: C, 37.99; H,
5.05; N, 1.32; P, 1.02; W, 38.12%.
]
]
in a biphasic medium of water and CH2Cl2 led to the corresponding
polyallyl DENDRI-POM hybrids 4, 5 and 6, respectively (Scheme 2).
These DENDRI-POMs were isolated from the organic layer and ob-
tained in good to excellent yields. In the 1H NMR spectra of hybrids
4, 5 and 6, the signals attributed to the CH2N groups of diallyl cat-
ion were shifted from 4.83 to 3.27 ppm for the free cation (see the
Supporting information, Fig. S3) to 4.67 and 3.31, 4.83 and 3.37,
4.83 and 3.20 ppm, for hybrids 4, 5 and 6, respectively (see the
Supporting information, Figs. S5, S10 and S14 respectively). In
addition to the spectroscopy data and elemental analysis of hy-
brids 4, 5 and 6 reported in the Section 2, this 1H NMR modification
further supports the existence of an interaction between the cation
and the POM species. DENDRI-POM 7 based on the Venturello ion
[PO4(WO(O2)2)4]3À was prepared from the heteropolyacid
H3PW12O40, in a biphasic mixture of water and dichloromethane,
in the presence of H2O2 [13,6d–6m]. As shown in Scheme 2, all
the peripheral allyl groups of the dendrons were oxidized giving
the corresponding 6-epoxy DENDRI-POM 7. 1H NMR spectra of 7
2.3.4. Synthesis of Venturello-based DENDRI-POM 7
H2O2 (5.2 mL, 35% in water) was added to a solution of the com-
mercial heteropolyacid H3PW12O40 (0.303 g, 0.105 mmol) in water.
The mixture was stirred at room temperature for 30 min. A solu-
tion of trihexylammonium diallyl carbinol salt
3 (0.150 g,
0.274 mmol) in CH2Cl2 (3 mL) was added, and the mixture was stir-
red for 90 min. The CH2Cl2 layer was washed with water and dried
over sodium sulfate. The product was obtained by removing the
solvent under vacuum to provide DENDRI-POM 7 as a white solid
(0.245 g, 91%). 1H NMR (250.163 MHz, CDCl3, TMS): d = 7.48 (br,
12H, Har), 4.59 (br, 6H, CH2–N), 3.08–2.92 (mbr, 30H, CH2–N,
CH2–O), 2.63 (br, 8H, CH–O, CH2), 1.73 (br, 18H, CH2), 1.31 (br,
54H, CH2), 0.88 (br, 27H, CH3). 13C NMR (62.91 MHz, CDCl3,
TMS): d = 149.0 (Cq, Ar), 131.4 (CH, Ar), 126.7 (Cq, Ar), 126.2 (CH,
Ar), 74.4 (Cq–OH), 74.1 (CH–N), 60.4 (CH2–N), 54.9 (CH2–N), 45.2
(CH–O), 45.1 (CH2–O), 38.2 (CH2), 37.4 (CH2), 26.4 (CH2), 23.6
(CH2), 16.6 (CH2), 14.2 (CH3). 31P NMR (121.49 MHz, CDCl3):
Br
m
(cmÀ1) = 3414.1 (s), 2955.6 (s),
Br
N(Hex)3
O
d = 2.72 (PO4). FTIR (KBr pellets),
(Hex)3N
O
CH3CN
1467.0 (s), 1086.9 (m, P–O), 1053.5 (m, P–O), 968.3 (s, W@O),
845.5 (m, O–O), 729 (m), 649 (m), 573 (w, W(O2))s,as, 549 (w,
W(O2))s,as. Anal. Calc. for C96H168N3O33PW4: C, 43.37; H, 6.37; P,
1.16; W, 27.66. Found: C, 43.22; H, 6.26; P, 1.10; W, 26.71%.
OCH3
12h, 80°C
OCH3
1
2
MgBr
Et2O, 0°C to r.t.
5 min. 93%
Br
2.4. General procedure for the catalytic oxidation reactions with
DENDRI-POMs and catalyst recovery experiments
(Hex)3N
OH
3
The DENDRI-POM catalyst and the substrate (250 equiv.) were
dissolved in 1 mL of CDCl3. 800 equiv of an aqueous solution of
Scheme 1. Synthesis of diallyl carbinol benzyl ammonium dendron 3.

60
C. Jahier et al. / Polyhedron 57 (2013) 57–63
OH
(Hex)3N
(Hex)3N
+
+
OH
6 -
N
+
HO
3 -
HO
+
N(Hex)3
OH
(Hex)3N
OH
Na3PW12O40
K6[P2W18O62]
+
N
+
CH2Cl2 / H2O
r.t. 2 h, 94%
CH2Cl2 / H2O
r.t. 2 h, 92%
HO
N(Hex)3
+
+
N
N(Hex)
3 HO
+
OH
-
Br
+
N(Hex)3
5
4
HO
HO
HO
O
HO
(Hex)3N
(Hex)3N
O
HO
HO
(Hex)3N
N
+
HO
3 -
N(Hex)3
O
W
O
O
(Hex)3N
14
3
HO
O
HO
O
O
W
O
O
N(Hex)3
O
O
O
O
W
O
O
(Hex)3N
OH
P
O
O
O
N
+
N(Hex)3
N(Hex)3
O
HO
O
OO
W
O
O
[NH4]14NaP5W30O110
H3PW12O40
O
(Hex)3N
O
OH
O
+
CH2Cl2 / H2O
r.t. 4 h, 60%
N
H2O2, H2O CH2Cl2,
r.t. 2 h 91%
(Hex)3N
N(Hex)3
OH
OH
N(Hex)3
O
OH
N(Hex)3
O
7
OH
OH
OH
6
Scheme 2. Synthesis of DENDRI-POMs 4–7.
(see the Supporting information, Fig. S17) showed the complete
disappearance of the signals at 5.53, 5.06 and 2.60 ppm attributed
to CH2CH@CH2 groups of compound 3 (Supporting information,
Fig. S3), and the appearance of broad signals at 2.92–2.38 ppm as-
signed to the terminal epoxide groups. In addition, the signals
attributed to the CH2N groups of diallyl cation were shifted from
4.83 to 3.27 ppm for the free cation to 4.59 and 3.06 ppm for the
POM hybrid 7.
observed by Massart et al. [14], when they studied the 31P NMR
spectroscopy of the Keggin anion [PW12O40
3À by varying parame-
]
ters such as countercations of the POM salt and the NMR solvent.
They did not observe any significant variation of the 31P NMR
chemical shifts of POMs by varying the cation or the NMR solvent.
In the case of the Venturello anion, one signal was obtained at
d = 2.7 ppm in chloroform for hybrid 7. This value is similar to
those obtained for [PO4(WO(O2)2)4]3À based hybrids and different
to that of the Keggin anion precursor. The 31P NMR spectroscopy
of all the DENDRI-POMs studied confirms the presence of the cor-
responding polyanion in their structure.
3.1.2. Characterization of 4–7
3.1.2.1. 31P NMR spectroscopy on 4–7. The 31P NMR spectra of DEN-
DRI-POMs 4–7 showed a single sharp peak for each hybrid at
d = À15.1 (Keggin), À12.9 (Dawson), À9.6 (Preyssler), and 2.7 (Ven-
turello) ppm, respectively, which can be assigned to the phosphate
groups of the respective polyanion. The chemical shifts of these
singlets are comparable to those of the free POM precursors
3.1.2.2. 183W NMR spectroscopy on 4. 183W NMR spectroscopy was
used to identify the polyanions in the DENDRI-POM hybrids, and
to obtain information about the interaction between the dendritic
structure and the polyanion. The 183W NMR spectrum of the Keg-
gin-based DENDRI-POM 4 consists of one peak at d = À88.6 ppm
3À
6À
([PW12O40
]
at À14.9 ppm, [P2W18O62
]
at À12.7 ppm, and
14À
[NaP5W30O110
]
at À9.9 ppm) [14] and the ammonium salts of
the Venturello anion [6d,l–m,12,13] already reported by our group.
These results imply that the starting polyanions are present in the
respective DENDRI-POMs, and that they are magnetically and
hence structurally equivalent. As expected, substitution of sodium
or ammonium countercations by dendritic cations did not show
any significant change in the 31P NMR chemical shifts of the Keg-
gin, Dawson and Preyssler polyanion-based DENDRI-POMs
(Table 1).
Table 1
31P NMR chemical shifts of DENDRI-POM hybrids and their polyanion precursors
[8,13,6d–m].
Hybrid
Solvent
(d)
POM precursor
Na3[PW12O40
Na6[P2W18O62
(NH4)14[NaP5W30O110
(cation)3[PW4O24
(d)a
4
5
6
7
[D6]acetone
[D6]acetone
[D6]acetone
CDCl3
À15.1
À12.9
À9.6
2.7
]
À14.9
À12.7
À9.9
]
]
]
2.08–3.2b
These results can be explained by the fact that the central PO4
group is surrounded and protected by WO6 subunits, and thus can-
not interact directly with peripheral cations. This feature was also
a
31P NMR performed in D2O at room temperature.
31P NMR performed in CDCl3 at room temperature.
b

C. Jahier et al. / Polyhedron 57 (2013) 57–63
61
in [D6]-acetone, indicating that all tungsten atoms are equivalent
(Fig. 1).
spectra of DENDRI-POMs 4–6 are very similar to those of their
respective POM precursor salts, suggesting that the polyanions re-
tain their structures in the DENDRI-POM hybrids.
This chemical shift is comparable to that obtained for the so-
dium salt of the Keggin ion Na3[PW12O40] (d = À99.4 ppm) in D2O
[15]. Our result represents the first successful 183W NMR spectrum
of a DENDRI-POM hybrid, showing a clear-cut signal. The small
chemical shift variation in the spectrum of 4 and that of the Na3
[PW12O40] sodium precursor indicates no structural change of the
polyanion, but rather reflects substitution of the sodium cations
by a dendritic ammonium structure, as well as solvent effects.
Analogous studies were also carried out with the DENDRI-POMs
5–7, containing the Dawson, Preyssler and Venturello ions, respec-
tively. We obtained 183W NMR spectra for the Dawson precursor
salt Na6[P2W18O62] (two singlets at d = À169.9 and À124.5 ppm)
and the Preyssler salt (NH4)14[ NaP5W30O110] (four singlets at
d = À287.8, À275.5, À209.7 and À207.6 ppm), but the solution
183W NMR measurements on the DENDRI-POMs 5–7 were unsuc-
cessful. The 183W NMR spectra of the n-tetrabutyl ammonium
and the Arquad (n-C18H37(75%) + [n-C16H33(25%)]2N(CH3)2) salts
of the Venturello anion were reported by Brégeault and co-workers
[16]. In the case of n-tetrabutyl ammonium salt, a well-resolved
doublet at À593 ppm was obtained, whereas a broad resonance
was obtained with a solution of the Arquad salt at À586 ppm. This
result indicates that the 183W NMR measurement of the Venturello
anion becomes more difficult with the bulky Arquad versus n-tet-
rabutyl cation. Furthermore, in our earlier studies, all attempts to
obtain 183W NMR spectra of a series of Venturello-based DEN-
DRI-POMs ion were unsuccessful.
The spectra of 4–6 exhibit prominent bands from 400 to
1200 cmÀ1 for the polyanions and from 2000 to 3400 cmÀ1 for
the dendrimer. A comparison with the POM precursor indicates a
slight difference in the stretching vibrations of the W–O–W and
the W@O bonds, whereas the P–O bands are essentially un-
changed. This is not unexpected, as the phosphate groups are lo-
cated inside the polyanions, and hence not very sensitive to a
change of the countercations. On the other hand, the differences
in the tungsten-oxo vibrations indicate an interaction between
the polyanions and the dendrimer.
In addition to IR and NMR, we also obtained good elemental
analysis data for the DENDRI-POM hybrids (see Section 2). How-
ever, all attempts to crystallize the DENDRI-POMs in a variety of
organic solvents failed. In general, amorphous materials were ob-
tained in pure or mixed solvents. Crystallization of 4 by slow vapor
diffusion resulted in small crystals after about three weeks, but
they were not suitable for single-crystal X-ray analysis.
In order to evaluate the stability of 4, 5 and 7 under catalytic
conditions, and their catalytic efficiency and recovery potential,
we decided to study the oxidation of methyl phenyl sulfide and
cyclooctene in the presence of H2O2.
3.2. Catalytic oxidation of methyl phenyl sulfide and cyclooctene with
DENDRI-POMs 4, 5 and 7 using H2O2
We tested the catalytic performance of DENDRI-POMs 4, 5 and 7
in the oxidation of methyl phenyl sulfide and cyclooctene in an
aqueous/CDCl3 biphasic mixture containing 0.4 mol-% of the selec-
tive DENDRI-POM, 250 equiv of the appropriate substrate and 800
equiv of aqueous hydrogen peroxide (35%). The reaction was
accomplished by vigorous stirring at 35 °C. The reaction kinetics
3.1.2.3. IR spectroscopy on 4–6. Infrared spectroscopy was used to
obtain information on the interaction between the dendritic struc-
tures and the polyanions. Thus, the IR spectra of DENDRI-POMs
4–6 were acquired and compared to those of their POM salt precur-
sors (Table 2). The data summarized in Table 2 indicate that the IR
Fig. 1. 183W NMR spectrum of 4.

62
C. Jahier et al. / Polyhedron 57 (2013) 57–63
Table 2
the selectivity for sulfoxide from 56% to 75% after 30 min reaction,
without affecting the reaction kinetics.
Infrared data of DENDRI-POMs 4–6 and their POM precursors.a
The catalytic efficiency of DENDRI-POMs 4, 5 and 7 was also
evaluated in the oxidation of cyclooctene using similar catalytic
conditions. Amongst these catalysts, only the hybrid 7 was active
in the oxidation of cyclooctene to the corresponding epoxide, with
100% conversion after 4 h (Table 3). The DENDRI-POMs 4 and 5 do
not show any reactivity towards cyclooctene, under similar condi-
tions. Interestingly, when methyl phenyl sulfide (8) was treated
with H2O2, in similar conditions without a POM catalyst, only 4%
of sulfoxide was obtained after 6 days (Table 3, entry 4). In the case
of cyclooctene (9), no reaction was observed.
Hybrid
(P–O)as
1079
(W@O)as
(W–O–W)as
DENDRI-POM 4
978
995–982
958
895–814
900–805
910–792
913–788
918–783
938–784
Na3[PW12O40
DENDRI-POM 5
K6[P2W18O62
DENDRI-POM 6
(NH4)14[NaP5W30O110
]
1081
1080
1081
]
983
–
960
1183–1081
1184–1078
]
a
IR measurements performed in KBr pellets,
m
(cmÀ1).
was monitored over time by plotting the ratio between the inten-
sity of the disappearing 1H NMR signals of the substrate and the
new peaks of the product. The results summarized in Table 3 show
that the hybrids 4, 5 and 7 catalyzed the oxidation of methyl phe-
nyl sulfide to the corresponding methyl phenyl sulfoxide and
methyl phenyl sulfone, with 7 being the most active DENDRI-
POM catalyst. As shown in Table 3, methyl phenyl sulfide is oxi-
dized by DENDRI-POM 4 with 74% conversion after 24 h (Table 3,
entry 1), giving 66% and 8% of the corresponding sulfoxide and sul-
fone, respectively. A similar conversion rate (77%) and product dis-
tribution (sulfoxide 69% and sulfone 8%) were obtained after 72 h,
indicating a deactivation of the catalyst. In the case of the Dawson-
based DENDRI-POM 5, a conversion of 77% was obtained after 24 h
with 69% sulfoxide and 8% sulfone (Table 3, entry 2). In this case,
the reaction kinetics is very slow, compare to that of the Keggin-
based DENDRI-POM 4. In contrast to DENDRI-POMs 4 and 5, the
DENDRI-POM 7 bearing the Venturello ion is more active. It cata-
lyzes the oxidation of sulfide with 98% conversion within 30 min,
whereas 24 h were needed in the case of 4 and 5 to reach 74%
and 77%, respectively. However, DENDRI-POM 7 was less selective
than 4 and 5 under similar reaction conditions.
Studies on the amount of oxidant used in the oxidation of
methyl phenyl sulfide with the DENDRI-POMs showed a significant
change in both the reaction kinetics and the selectivity in the case
of hybrids 4 and 5. We discovered that decreasing the amount of
H2O2 with DENDRI-POM hybrids 4 and 5 decreases the reaction
kinetics. The use of 2 equiv of H2O2 versus 3.2 equiv at the initial
conditions with hybrids 4 and 5 led to 8% versus 30% and 6% versus
23% conversion, respectively, after 2 h. At such conditions, no trace
of sulfone was observed. For the Venturello-based DENDRI-POM 7,
decreasing the concentration of H2O2 from 3.2 to 2 equiv increased
3.3. Catalyst recovery and recycling
In order to check the stability of the DENDRI-POMs at catalytic
conditions, two catalytic cycles were performed with hybrids 4, 5
and 7 in the oxidation of sulfide and cyclooctene. The catalyst
was recovered by precipitation after each catalytic cycle and
checked by 1H and 31P NMR before the second catalytic experi-
ment. The Keggin- and Venturello-based DENDRI-POMs 4 and 7
oxidized methyl phenyl sulfide without any obvious loss in activity
and selectivity over two cycles. In contrast, the activity and iso-
lated yield of the Dawson-based dendritic hybrid 5 in the oxidation
of sulfide decreased significantly after the first cycle. This was ob-
served in the drop in conversion rate from 74% to 54% as well as the
isolated yield from 72% to 43% after the first cycle. The deactivation
of catalyst 5 is probably due to the decomposition of the POM
structure in acidic medium. This feature could be attributed to
the size of the ammonium dendron 3 used for these studies, which
is probably not sufficiently dendritic to protect the Dawson ion
against external acidic attack under these reaction conditions, in
contrast to the smaller Keggin and Venturello ions, which are easily
stabilized. According to our previous studies [6], it has clearly been
established that the dendritic structures around the POM increase
their stability in solution, allowing the facile recovery and reuse of
the catalyst.
4. Conclusions
We have synthesized four DENDRI-POMs based on the diallyl
carbinol benzyl ammonium dendron 3 and tungstophosphates
Table 3
Catalytic oxidation of methyl phenyl sulfide (8) and cyclooctene (9) with DENDRI-POMs 4, 5 and 7, using H2O2.a
Entry
Catalyst
Substrate
Time (h)b
Conversion (%)c
Products (yield, %)
1
2
3
4
5
6
7
4
5
7
–
4
5
7
8
24
24
0.5
144
72
74
77
98
4
0
0
10(66)/11(8)
10(69)/11(8)
10(54)/11(44)
10
12
12
12
9
72
4
100
a
b
c
Reaction conditions: catalyst (0.4 mol-%), substrate (250 equiv), 35% H2O2 (800 equiv), CDCl3 (1 mL).
Reactions were monitored by 1H NMR.
Conversions were determined by integration of the 1H NMR signals of substrate and product.

C. Jahier et al. / Polyhedron 57 (2013) 57–63
63
[3] (a) D.A. Judd, J.H. Nettles, N. Nevins, J.P. Snyder, D.C. Liotta, J. Tang, J. Ermolieff,
R.F. Schinazi, C.L. Hill, J. Am. Chem. Soc. 123 (2001) 886;
(b) S. Shigeta, S. Mori, E. Kodama, J. Kodama, K. Takahashi, T. Yamase, Antiviral
Res. 58 (2003) 265;
(c) X. Wang, J. Liu, M.T. Pope, Dalton Trans. 5 (2003) 957;
(d) Y. Tajima, Microbiol. Immunol. 47 (2003) 207.
[4] (a) M. Sadakane, E. Steckhan, Chem. Rev. 98 (1998) 219;
(b) C. Fleming, D.-L. Long, N. McMillan, J. Johnston, N. Bovet, V. Dhanak, N.
Gadegaard, P. Kögerler, L. Cronin, M. Kadodwala, Nat. Nanotechnol. 3 (2008)
229.
[5] (a) N. Mizuno, K. Yamaguchi, K. Kamata, Coord. Chem. Rev. 249 (2005) 1944;
(b) W. Zhu, H. Li, X. Jiang, Y. Yan, J. Lu, L. He, J. Xia, Green Chem. 10 (2008) 641.
[6] (a) H. Zeng, G.R. Newkome, C.L. Hill, Angew. Chem., Int. Ed. 39 (2000) 1771;
(b) D. Volkmer, B. Bredenkötter, J. Tellenbröker, P. Kögerler, D.G. Kurth, P.
Lehmann, H. Schnablegger, D. Schwahn, M. Piepenbrink, B. Krebs, J. Am. Chem.
Soc. 124 (2002) 10489;
with the Keggin, Dawson, Preyssler and Venturello structure. The
DENDRI-POMs 4–7 were fully characterized by IR and multinuclear
(1H, 13C, 31P) NMR spectroscopy as well as elemental analysis. We
also performed catalytic studies on these DENDRI-POMs for the
homogeneous oxidation of methyl phenyl sulfide and cyclooctene
using H2O2 as oxidant. Our work has demonstrated that DENDRI-
POMs are highly attractive catalysts, which can be recovered and
reused. We also report an example of a DENDRI-POM characterized
by solution 183W NMR.
Acknowledgments
(c) M.C. Rogers, B. Adisa, D.A. Bruce, Catal. Lett. 98 (2004) 29;
(d) L. Plault, A. Hauseler, S. Nlate, D. Astruc, J. Ruiz, S. Gatard, R. Neumann,
Angew. Chem., Int. Ed. 43 (2004) 2924;
S. N. and C. J. thank the Agence National de la Recherche (grant
ANR-06-BLAN-0215), the University Bordeaux 1 and the CNRS for
financial support. C. J. thanks ESF COST D40 action for a STSM
grant, allowing for a research visit at Jacobs University. U. K. thanks
Jacobs University and the Fonds der Chemischen Industrie for re-
search support.
(e) M.V. Vasylyev, D. Astruc, R. Neumann, Adv. Synth. Catal. 347 (2005) 39;
(f) S. Nlate, D. Astruc, R. Neumann, R. Adv, Synth. Catal. 346 (2004) 1445;
(g) S. Nlate, L. Plault, D. Astruc, Chem. Eur. J. 12 (2006) 903;
(h) S. Nlate, L. Plault, D. Astruc, New J. Chem. 31 (2007) 1264;
(i) C. Jahier, L. Plault, S. Nlate, Isr. J. Chem. 49 (2009) 109;
(j) C. Jahier, M. Cantuel, N.D. McClenaghan, T. Buffeteau, D. Cavagnat, F.
Agbossou, M. Carraro, M. Bonchio, S. Nlate, Chem. Eur. J. 15 (2009) 8703;
(k) C. Jahier, S.S. Mal, U. Kortz, S. Nlate, Eur. J. Inorg. Chem. (2010) 1559;
(l) C. Jahier, M.-F. Coustou, M. Cantuel, N.D. McClenaghan, T. Buffeteau, D.
Cavagnat, M. Carraro, S. Nlate, Eur. J. Inorg. Chem. (2011) 727;
(m) C. Jahier, S. Nlate, Eur. J. Inorg. Chem. (2012) 833;
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.poly.2013.04.013.
(n) S. Nlate, C. Jahier, Eur. J. Inorg. Chem. (2013) 1606.
[7] C. Musumeci, M.H. Rosnes, F. Giannazzo, M.D. Symes, L. Cronin, B. Pignataro,
ACS Nano 5 (2011) 9992.
[8] Y.-G. Chen, J. Gong, L.-Y. Qu, Coord. Chem. Rev. 248 (2004) 245.
[9] (a) J.F. Keggin, Nature 13 (1933) 908;
References
(b) J.W. Illingworth, J.F. Keggin, J. Chem. Soc. (1935) 575.
[10] (a) B. Dawson, Acta Crystallogr. 6 (1953) 113;
(b) I.-M. Mbomekalle, Y.W. Lu, B. Keita, L. Nadjo, Inorg. Chem. Commun. 7
(2004) 86.
[11] (a) C. Preyssler, Bull. Soc. Chim. Fr. (1970) 30;
[1] (a) M.T. Pope, A. Müller, Angew. Chem., Int. Ed. Engl. 30 (1991) 34;
(b) J.T. Rhule, C.L. Hill, D.A. Rud, R.F. Schinazi, Chem. Rev. 98 (1998) 327;
(c) Dimitris E. Katsoulis, Chem. Rev. 98 (1998) 359;
(d) M.T. Pope, A. Müller (Eds.), Polyoxometalate Chemistry: From Topology via
Self-Assembly to Applications, Kluwer Academic, Dordrecht, 2001.
[2] (a) Chem. Rev. Special Thematic Issues, Polyoxometalates. 98 (1998) pp. 1–
389.;
(b) M.H. Alizadeh, S.P. Harmalker, Y. Jeannin, J. Martin-Frère, M.T. Pope, J. Am.
Chem. Soc. 107 (1985) 2662.
[12] C. Venturello, R. D’Aloisio, J.C.J. Bart, M. Ricci, J. Mol. Catal. 32 (1985) 107.
[13] D.C. Duncan, R.C. Chambers, E. Hecht, C. Hill, J. Am. Chem. Soc. 117 (1995)
1681.
[14] R. Massart, R. Constant, J.-M. Fruchart, J.-P. Ciabrini, M. Fournier, Inorg. Chem.
16 (1977) 2916.
(b) A. Proust, R. Thouvenot, P. Gouzerh, Chem. Commun. 16 (2008) 1837–
1852;
(c) D.L. Long, E. Burkholder, L. Cronin, Chem. Soc. Rev. 36 (2007) 105–121;
(d) D. Long, R. Tsunashima, L. Cronin, Angew. Chem. Int. Ed. 49 (2010) 1736–
1758;
[15] R. Acerete, C.F. Hammer, L.C.W. Baker, J. Am. Chem. Soc. 104 (1982) 5384.
[16] C. Aubry, G. Chottard, N. Platzer, J.-M. Brégeault, R. Thouvenot, F. Chauveau, C.
Huet, H. Ledon, Inorg. Chem. 30 (1991) 4409.
(e) P. Mialane, A. Dolbecq, F. Sécheresse, Chem. Commun. (2006) 3477–3485;
(f) A. Dolbecq, E. Dumas, C.R. Mayer, P. Mialane, Chem. Rev. 110 (2010) 6009–
6048.