Rigid, non-porous and tunable hybrid p-aminobenzoate/TiO2 materials: Toward a fine structural determination of the immobilized RhCl(Ph3)3 complex

Article abstract of DOI:10.1016/j.jorganchem.2014.12.032

Abstract By exchange of ligands, Wilkinson complex RhCl(PPh₃)₃ are immobilized on p-aminobenzoate/TiO₂ with different organic loading (6, 11 and 16%). This new hybrid material exhibit a linear correlation between the ligand content of the starting TiO₂ and the rhodium loading, showing the accessibility of all surfaces amines fonctions on the non-porous parent materials. ¹H, ¹³C, and ¹D, ²D INAQUEDATE refocused and J-resolved ³¹P solid-state NMR confirm the well-defined structure [(≡TiO)₂(η²-O₂C-C₆H₄-NH₂)RhCl-cis-(PPh₃)₂]. New immobilized catalysts show interesting activity in cyclohexene hydroformylation.

Full text of DOI:10.1016/j.jorganchem.2014.12.032

Journal of Organometallic Chemistry 784 (2015) 103e108
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Rigid, non-porous and tunable hybrid p-aminobenzoate/TiO2
materials: Toward a fine structural determination of the immobilized
RhCl(Ph ) complex
3
3
1
1
*
Jeff Espinas , Raed Rahal , Edy Abou-Hamad, Mohamad El Eter, Jean-Marie Basset
KAUST Catalysis Center, 4700 King Abdullah University of Science and Technology, Thuwal 23955-6900, Saudi Arabia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 October 2014
Received in revised form
By exchange of ligands, Wilkinson complex RhCl(PPh3)3 are immobilized on p-aminobenzoate/TiO2 with
different organic loading (6, 11 and 16%). This new hybrid material exhibit a linear correlation between
2
the ligand content of the starting TiO and the rhodium loading, showing the accessibility of all surfaces
20 December 2014
amines fonctions on the non-porous parent materials. H, C, and 1D, D INAQUEDATE refocused and
1
13
2
Accepted 22 December 2014
Available online 6 January 2015
J-resolved 31P solid-state NMR confirm the well-defined structure [(≡TiO)
(ɳ -O
2
2
2 6
CeC H
4
eNH
2
)RhCl-cis-
(
PPh
3
)
2
]. New immobilized catalysts show interesting activity in cyclohexene hydroformylation.
©
2015 Published by Elsevier B.V.
Keywords:
Wilkinson complex
Rhodium
Solid-state phosphorous NMR spectroscopy
Surface organometallic chemistry
Titanium dioxide
Well-defined immobilized complex
Introduction
homogeneously (inaccessible) distributed functions which do not
warranty a homogeneous grafting of any metal, the flexibility of the
alkyl linkers, and also the presence of unreacted siloxanes moieties
[5], limiting the complete structural determination of the metal
center. Therefore, the development of well-defined aminated sup-
ports is a necessary requirement for a full investigation of the final
material for “catalysis by design” to achieve structure/activity
relationship and go beyond the “making sure that the immobili-
zation worked” suggested by Blümel et al. [6].
Among the conventional processes for the development of
hybrid support, Solegel chemistry could be an alternative strategy
[7]. Through the hydrolysis of organically modified titanium iso-
propoxide, Rahal et al. published a one pot synthesis at low tem-
Ample consideration has been made for developing the immo-
bilization of catalysts on various supports to combine the advan-
tages of both homogeneous and heterogeneous catalysis [1]. While
metal complexes can be grafted directly on the carrier and result in
true “Surface OrganoMetallic Complexes”, some will require an
anchor often provided by a derivatized support for both the teth-
ering and tailoring of the metal center and provide “Supported
Homogeneous Complexes” [2]. These so-called “post-functional-
ized” hybrid supports are generally prepared by reacting the sur-
face hydroxyl groups of an inorganic oxide with an organic spacer
(
frequently an organosilane) presenting a coordinating function
such as phosphanes, thiols or amines [3]. The latter proved to
generate stability and catalytic activity of late transition metals. For
this purpose, diverse sophisticated aminated materials were re-
ported, e.g. polyamidoamine (PAMAM) dendrimers supported on
silica [4]. However, such systems generally suffer from non-
perature of a series of non-porous p-aminobenzoate/TiO
2
hybrid
2
materials, denoted [(≡TiO) (ɳ -O CeC H eNH )], presenting rigid
2 2 6 4 2
aromatic tethers and controlled amine content [8]. Here we report
the use of these tunable hybrid aminated supports as a mean to
overcome all the above mentioned disadvantages of classical
complexes heterogenization. The choice of the Wilkinson complex
3 3
RhCl(PPh ) has just been the result of a simple proof of concept.
The morphology of these materials and the fine chemical design of
*
Corresponding author. Tel.: þ966 (2) 808 02995.
E-mail address: jeanmarie.basset@kaust.edu.sa (J.-M. Basset).
URL: http://kcc.kaust.edu.sa/Pages/Home.aspx
The authors contributed equally to this work.
the supported organometallic moiety have been characterized by
1
BET, TEM, XRD, TGA, mass balance analysis, IR, solid state NMR ( H
13
31
1
MAS, C CP/MAS and P MAS, CP/MAS, 2D refocused INADEQUATE
http://dx.doi.org/10.1016/j.jorganchem.2014.12.032
022-328X/© 2015 Published by Elsevier B.V.
0

104
J. Espinas et al. / Journal of Organometallic Chemistry 784 (2015) 103e108
and J-resolved). The accessibility and the loading effect of the amine
functions on the nature and the surface coverage of the tethered
complex were also studied as well as their catalytic performance in
cyclohexene hydroformylation.
Results and discussion
The p-aminobenzoate/TiO
2
with an organic loading of 6, 11 and
16% (I, II and III), were synthetized as reported previously (Scheme
S1 in Supporting information) [8a]. The corresponding (I-III)-Rh
complexes were obtained by refluxing a toluene solution of the
Wilkinson complex RhCl(PPh
section, Fig. S2). The N adsorption desorption isotherms of
IeIII)-R)-Rh complexes correspond to a mesoporous structure,
3 3
) with IeIII (see Experimental
2
(
similar to that of the starting materials (IeIII) (see Fig. S3) [8a]. I-Rh
gives a type-IV isotherm with capillary hysteresis loops of type-H2.
II-Rh and III-Rh present the same type-IV isotherm and hysteresis
loops of type-H4. Upon increasing the organic loading, the hyster-
esis which is assumed to be an intrinsic property of the pores, varies
from type-H2 to type-H4. This can be explained by the disorder
occurring in the pore network due to the increase of their aggre-
gation state [9].
Fig. 1. TEM images of III-Rh particles (A), FFT pattern (B) and HRTEM images of III-Rh
particles and corresponding FFT pattern (C).
After reaction with the Wilkinson complex, the BET surface area
and the total pore volume decrease dramatically respectively from
ꢀ
1
2
ꢀ1
2
ꢀ1
3
ꢀ1
2
ꢀ1
at 1119 cm could be assigned to the due to
[11].
n(PePh) bond vibration
2
(
20, 58 and 16 m
g
to 153 m
g
(0.22 cm
g
), 58 m
g
0.09 cm3
g
ꢀ1
) and 16 m
2
g
ꢀ1
(0.05 cm
3
g
ꢀ1
) for (IeIII)-Rh,
Thermogravimetric analysis (TGA) of (IeIII) presents an average
organic loss of 6, 11 and 16%, respectively, while they are of 18, 27
and 40% for (IeIII)-Rh (Fig. S6, the reported weight losses are those
taking place above ca.125 C). The increase of organic content of the
final materials is a clear indication of the successful tethering of the
Wilkinson complex on the surface.
The fine chemical composition of (IeIII)-Rh was determined by
elemental analysis (Table 2). The rhodium content is respectively
equal to 2.58 (0.251 mmol g ), 4.62 (0.449 mmol g ) and 7.00%
respectively (Table 1). This phenomenon is likely related to the fact
that the numerous phenyl groups coming from the Wilkinson
complex reinforce inter-granular interactions and aggregation.
Transmission electron microscopy (TEM) indicates spherical
nanoparticles with an average size of 5e6 nm (Fig. 1(AeC) and
Fig. S4). Interestingly, the size of the particles remains stable before
and after tethering the complex, only their state of aggregation
increases, as discussed previously, with the organic contents.
Fig. 1(B and C) displays Fast Fourier Transform (FFT) patterns and
ꢁ
ꢀ
1
ꢀ1
ꢀ
1
(
0.680 mmol g ). This latter is linearly correlated to the organic
2
the HRTEM of TiO nanoparticles. It shows sets of lattice fringes
loading of the corresponding IeIII, even at high organic loading
especially for III (Graph 1). This is not the case with previously
reported PAMAM dendrimers materials for which the rhodium
contents, less than 2% (0.2 mmol g ), decrease while increasing
the amine moiety [4a]. In the present case, approximately one
rhodium per nitrogen atom and similar ratios of 1 ± 0.1 Cl/Rh and
(
Interplanar spacing d101 ¼ 0.35 nm) showing the presence of a
highly crystalline anatase phase structure.
The XRD patterns of (IeIII)-Rh are assigned to a pure tetragonal
anatase phase with a high crystallization degree (JCPDS No. 21-
ꢀ1
1272). The average crystallite size (5e6 nm) estimated by the
Scherrer equation is in good agreement with the TEM data (Fig. S5).
It is noteworthy that the tethering of the Wilkinson complex does
not influence the size and the crystalline phase of IeIII. The IR
spectra of (IeIII)-Rh (Fig. S5) shows the presence of all bands
characteristic of the parent samples (IeIII) [8a], indicating that the
p-aminobenzoate moiety was basically maintained throughout the
2
± 0.1 P/Rh are observed. This composition is derived from a
bis(triphenylphosphino)chlororhodium moiety, given that the
Wilkinson complex usually coordinates to an amine function by
substitution of a triphenylphosphine group [3e,5]. These results
confirmed the presence of exclusively surface accessible amine
2
moieties on the TiO particles. The only result which is not in line
grafting procedure on the surface of TiO
2
nanoparticles. The broad
ꢀ
1
with the surface complex structure is the C/Rh ratios which are
ranging from 48.1 to 52.8 and higher than the expected value (43).
This is likely due to residual physisorbed toluene molecules
confirmed by the C CP/MAS spectroscopy (vide infra).
Each solid-state NMR spectrum of both (IeIII) materials and
IeIII)-Rh complexes collections present similarities and the cases
bands above 2500 cm are assigned to the stretching bands of NH
2
ꢀ
1
and OH groups. Bands below 1000 cm , could be assigned to the
lattice vibrations of (TieO) [10]. New bands assigned to the phenyl
fragments of the Wilkinson complex also appear at 3046 cm
n
13
ꢀ1
ꢀ1
(n
(Csp2-H)). Simultaneously, the band at 1431 cm attributed to the
(
aromatic C) vibration of the p-aminobenzoate considerably
(C
n ]
increased in intensity after coordination of the complex. The band
Table 2
Table 1
Elemental content of the supported complexes (IeIII)-Rh.
Surface area and total pore volume of (IeIII)-Rh (and IeIII).
Rh
Cl
P
N
C
Rh/N
P/Rh
Cl/Rh
C/Rh
wt % wt % wt % wt % wt % (th. 1)b (th. 2)b (th. 1)b (th. 43)b
a
a
a
a
a
Sample
Surface area
Total pore volume
(cm
Pore size diameter
(nm)
2
ꢀ1
3
ꢀ1
g )
(
m
g
)
I-Rh 2.58 0.90 1.36 0.35 15.88 1.0
II-Rh 4.62 1.69 2.71 0.80 27.47 0.8
III-Rh 7.00 2.58 4.24 0.84 39.28 1.1
1.8
1.8
2.0
1.0
1.1
1.1
52.8
51.0
48.1
I-Rh (I)
II-Rh (II)
III-Rh (III)
153 (220)
58 (216)
16 (153)
0.22 (0.26)
0.09 (0.19)
0.05 (0.2)
4 (3.7)
4.2 (3.8)
8.8 (4.9)
a
0
±
.05 and 0.1% error are estimated for values ranging from 0 to 1%.
b
0.1.

J. Espinas et al. / Journal of Organometallic Chemistry 784 (2015) 103e108
105
Graph 1. Correlation between the organic content of (IeIII) determined by TGA (wt %)
and the rhodium content of (IeIII)-Rh determined by elemental analysis (wt %).
Fig. 3. 31P MAS and CP/MAS NMR spectra of III-Rh. See Figs. S21 and S22 for spectra of
I-Rh and II-Rh, respectively.
of III and III-Rh will be discussed thereafter as an illustration (other
[
Si]eNEt
gives two peaks at 51.1 and 58.3 ppm [14]. The absence of any an-
chor in [Si]-NEt RhCl(Ph gives a rigid attached complex, more
2
RhCl(Ph
3
)
2
([Si] ¼ ≡SieOe)), reported by Petrucci et al.,
1
H, C and P solid-state NMR spectra in Figs. S10eS22). The 13
13
31
C
CP/MAS spectrum of III-Rh shows small peaks at 114, 151 and
76 ppm respectively, attributed to m-CAr, CAr-NH and COO of the
2
3 2
)
1
2
likely with a single-site character. In the present case, the amino-
benzoate could be considered rigid enough to limit the sites
interaction. The two peaks of III-Rh can be resolved into two
doublets using a P CP/MAS sequence (Fig. 3), with J(Rh, P)
coupling constants of 200 Hz for 49.5 ppm and 165 Hz for 56.7 ppm,
in the same range as that of the molecular complex [15].
p-aminobenzoate linker (Fig. 2B) [8a,12]. Its ipso and ortho aromatic
carbons signals and these of triphenylphosphine fragments range
from 128 to 141 ppm. Some residual toluene is present as judged
from its methyl signal at 21 ppm (confirmed by elemental analysis)
and contributes to the change of the relative intensities observed
for the aromatic carbon signals.
31
1
The mutual position of the two phosphorus atoms can be
The 31P MAS NMR spectrum displays two signals at 49.5 and
31
deduced from two dimensional P NMR refocused INADEQUATE
5
6.7 ppm and no peak in the e 5 ppm region from eventual un-
pulse sequences [16]. This method uses the homonuclear indirect
coordinated phosphane. The absence of phosphine oxide species
31
31
Pe P dipoleedipole J interaction between phosphanes species
31
was confirmed when the P MAS spectrum of III-Rh was in contact
with air showing the appearance of one broad peak at 27.9 ppm
flanked to a smaller one at 39.0 ppm. This is reasonably attributed
to phosphine oxide signals (Fig. S23) [13]. Given that the 3 P MAS is
quantitative, the integration of the two peaks from the fresh sample
linked via Rh bridges for creating double quantum coherences. The
31
P through bound correlation spectrum obtained, using the refo-
cused INADEQUATE pulse sequences is shown in Fig. 4. The reso-
nance of two coupled phosphorous atoms occurs at the common
double quantum frequency in the
with the single quantum frequency of each peak in the
sion. The P eRheP (P s P ) connectivity is clearly indicated by
the two intense correlations. The resonances among inter-
connected phosphanes species can be correlated via the existence
of a J-coupling that is estimated to be less than 60 Hz, a J(P, P) value
attributed to phosphorous atoms in cis position ( J(P, P)trans would
1
u
1
indirect dimension correlating
(
1:1) supports the presence of two non-equivalent phosphorous
u
2
dimen-
atoms (Fig. 3). Interestingly, some groups mention the presence of
one large signal centered at 29 or 32 ppm [3e,5] for the Wilkinson
complex anchored through aminopropyl tethers. Their relative
flexibility favors the formation of surface species in different en-
vironments which probably explains the broadening of the spectral
1
2
1
2
2
2
31
line width. Conversely, the P CP/MAS spectrum of the system
have been of 365 Hz) [15].
In order to obtain an accurate value for 2J(P,P)cis, a 31P 2D
J-resolved CP/MAS experiment was carried out [15]. The isotropic
Fig. 2. 13C CP/MAS NMR of III (A) and III-Rh (B); (*: rotation bands).
Fig. 4. 31P 2D refocused INADEQUATE NMR of III-Rh.

106
J. Espinas et al. / Journal of Organometallic Chemistry 784 (2015) 103e108
Fig. 6. TON (black) and cyclohexane carboxyaldehyde selectivity (red) of (IeIII)-Rh
materials. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)
2
GC/MS. A blank test with TiO in the same condition did not present
any activity. The cyclohexene conversions start from 61% and
slightly increase (61, 66 and 68% for (IeIII)-Rh, respectively),
however not proportionally, with the rhodium loading. Thus, the
turnover numbers (TON) calculated from the number of moles of
cyclohexene converted are respectively of 1436, 864 and
Fig. 5. 31P J-Resolved NMR of III-Rh.
ꢀ
1
5
92 mol molRh while the TON of the molecular Wilkinson catalyst is
ꢀ
1
of 907 mol molRh (Fig. 6). Although the TON of I-Rh is in the same
range than the ones reported for supported Wilkinson catalysts in
similar catalytic conditions [3d], the decrease of the II-Rh and III-Rh
ones can be explained by surface interactions between metal cen-
ters. Blümel et al. studied the surface coverage effect on the cata-
lytic performance in olefin hydrogenation of supported Wilkinson
complex on silica and observed that reducing the Rh concentration
leads to a better catalytic activity and maximal conversion can be
reached faster [3b]. This can be related to a greater distance be-
tween the metal center avoiding dimerization or agglomeration
deactivating the catalyst. In the present case, both TEM microscopy
and BET analysis has revealed agglomerated materials when PABA
loading is increased. Also, the organic and thus the metal surface
coverage is sufficiently high for II-Rh and III-Rh and their surface
areas sorely drop after grafting of the Wilkinson complex (vide
supra BET study). Inter-granular and thus metal center interactions,
as well as diffusion resistance of the substrates, seem to be favored
and also support the lower TON.
2
Scheme 1. Structure [(≡TiO)
2
(ɳ -O
2
CeC
6 4 2
H eNH )RhCl-cis-(PPh
3
)
2
] proposed for the
three surface complexes (IeIII)eRh.
part of the 2D J-resolved spectrum is shown in Fig. 5. In the 31P
homonuclear J-resolved 2D spectrum, the heteronuclear J coupling
due to J(Rh,P) remains along the F2 dimension while the homo-
nuclear J coupling between P appears along the F1 dimension.
The splitting due to J(P,P)cis is clearly resolved. The value of
1
31
2
2
J(P,P)cis is 60 ± 5 Hz from the 2D spectrum, which is similar to that
Conclusion
observed for RhCl(PPh
3 3
) [17].
The discrete organometallic complex structure for (IeIII)-Rh
denoted [(≡TiO) (ɳ -O CeC H eNH )RhCl-cis-(PPh ) ] is highly
2 2 6 4 2 3 2
advanced (Scheme 1).
2
In summary, the Wilkinson complex RhCl(PPh
anchored for the first time on p-aminobenzoate/TiO
3
)
3
has been
aminated
2
hybrid materials, by exchange of one phosphane ligand by the
amine function. The crystalline phase and the morphology of the
nanoparticles remain stable after the incorporation of the complex,
as judged by TEM and XRD microscopies. The rhodium loading is
proportional to the organic content and demonstrates the acces-
sibility of all the surface amine functions of the non-porous parent
materials. The rigidity of the p-aminobenzoate tether promotes the
Catalytic test
(
IeIII)-Rh complexes show a catalytic activity in the hydro-
formylation of cyclohexene into cyclohexane carboxyaldehyde and
the results are summarized in Table 3. No alcohol was observed in
Table 3
a
Hydroformylation of cyclohexene catalyzed by (I-III)-Rh.
Supported complex
Rh before reaction
wt %)
Rh after reaction
(wt %)
Cyclohexene
conversion (%)
TONb
Selectivity
(
Cyclohexane carboxy-aldehyde
Cycloh-exane
c
RhCl(PPh
3
)
3
e
e
70
61
66
68
907
1436
864
100
81
98
0
19
2
TiO
TiO
TiO
2
2
2
-PABA-Rh-I
-PABA-Rh-II
-PABA-Rh-III (1st cycle)
2.58
4.62
7.00
0.43
0.47
1.82
592
100
0
a
ꢁ
0
.100 (IeIII)-Rh, 50 ml THF, 6 ml cyclohexene, 28 bar CO/H
2
(1:1), 100 C, 20 h.
b
TON (turnover number) ¼ converted mol of cyclohexene/mol Rh.
c
m
RhCl(PPh3)3 ¼ 0.042, nRh ¼ 0.0454 mmol.

J. Espinas et al. / Journal of Organometallic Chemistry 784 (2015) 103e108
107
isolation of the rhodium complex as confirmed by H, 13C, and 1D,
1
speed of 14 KHz, 90 pulse of 2.5
ꢁ
m
s length. 32 data sets with 2056
31
2
D INADEQUATE refocused and J-resolved P solid-state NMR
suggesting the well-defined discrete structure [(≡TiO)
CeC eNH )RhCl-cis-(PPh ]. The three (IeIII)-Rh complexes
scan each with delay of 1 s the t1 increment was synchronized with
the rotor t1 ¼ 2n TR. Again, all spinning sidebands become coin-
cident in the F1 dimension, enhancing the 2D signals.
2
2
(ɳ -
O
2
6
H
4
2
3 2
)
present good catalytic performances in the hydroformylation of
cyclohexene. Interestingly, the activity decreases with the rhodium
loading, related to local metal centers interaction but also
agglomeration of the material. This unprecedented category of
materials will serve as a new insight for catalysis by design.
Preparation of (IeIII)-Rh complexes
The preparation of I-Rh is given as an example. Typically,
3 3
RhCl(PPh ) (0.359 g, 0.388 mmol, 1.1 eq./nitrogen loading) in
toluene (20 ml) was stirred with I (0.800 g, 0.353 mmolN, previ-
Experimental section
ꢁ
ꢀ5
ously dehydrated at 100 C for 4 h under vacuum (10 Torr)), at
ꢁ
1
10 C for 16 h. The solid was filtrated, thoroughly washed with
All manipulations were carried out under a dry argon atmo-
sphere using glove-box and standard Schlenk techniques. Graftings
were performed following the double-Schlenk procedure. Toluene
was dried over Na/benzophenone, degased and freshly distilled
toluene (5 ꢂ 20 ml) to remove unreacted complex and dried under
ꢀ5
vacuum (10 Torr) over 5 h to provide a brick-color powder.
Catalytic tests
2
prior to use. N adsorption desorption isotherm were recorded on a
Micromeritics ASAP 2420 apparatus. High-resolution transmission
electron microscope (HR-TEM) and fast Fourier transform (FFT)
pattern were obtained using FEI Titan Cryo Twin operating at
Typically, 0.100 g of (IeIII)-Rh catalyst was charged in a Parr
autoclave in a glove-box followed by the addition of 50 ml THF and
6
H
ml cyclohexene. The reactor was purged with the syngas (1:1 CO/
3
00 kV. Powder X-ray Diffractions (XRD) were recorded on a Bruker
ꢁ
ꢁ
2
) at room temperature and then allowed to reach 100 C. 28 bar
D8 Advance diffractometer with a range of 10e90 2
q
using CuK
a
of the syngas was charged and the reaction media stirred over 20 h
radiation as X-ray source. Thermogravimetric analysis (TGA) ex-
periments were carried out in air within the temperature range of
ꢁ
under constant pressure. After cooling down the apparatus to 0 C
in an ice-bath, the pressure was released and an aliquot of the
brown colored liquid phase was kept aside for GC/MS analysis. The
brownish catalyst was recovered by filtration and used in the same
condition for leaching study.
ꢁ ꢁ
7e800 C at a temperature ramp of 10 C using a Mettler-Toledo
2
TGA-DSC/1. Elemental analyses (EA) were performed at the Mik-
roanalytisches Labor Pascher of Remagen (Germany). Fourier
transform infrared (FT-IR) spectra were recorded on a Nicolet 6700
1
13
31
FT-IR instrument. One dimensional H MAS, C CP-MAS and P CP-
MAS solid state NMR spectra were recorded on a Bruker AVANCE III
spectrometer operating at 400, 100 and 162 MHz resonance fre-
Appendix A. Supplementary data
quencies for 1H, C and P respectively, with a conventional
double resonance 4 mm CPMAS probe. The samples were intro-
duced under argon into zirconia rotors, which were then tightly
13
31
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2014.12.032.
1
closed. The spinning frequency was set to 17, 10, and 15 KHz for H,
References
1
3
31
C and P spectra, respectively. 1H and 13C NMR chemical shifts
31
[
1] a) R. Anwander, in: Handbook of Heterogeneous Catalysis, Wiley-VCH Verlag
are reported with respect to TMS as an external reference.
P
GmbH & Co. KGaA, 2008;
b) V. Dal Santo, F. Liguori, C. Pirovano, M. Guidotti, Molecules 15 (2010)
3829e3856;
c) M. Lenarda, L. Storaro, R. Ganzerla, J. Mol. Catal. A Chem. 111 (1996)
203e237;
d) J. Guzman, B.C. Gates, Dalton Trans. (2003) 3303e3318;
e) J.-M. Basset, R. Psaro, D. Roberto, R. Ugo, Modern Surface Organometallic
Chemistry, Wiley Ed., Weinheim, 2009.
chemical shifts are reported relative to H
3
PO
4
. For CP/MAS, the
ꢁ
following sequence was used: 90 pulse on the proton (pulse length
2
.4
m
s for 13C and 1.8
m
s for P), then a cross-polarization step with
31
13
a contact time typically 2 ms, and finally acquisition of the C and
3
1
P under high power proton decoupling. The delay between the
13
31
scan was set to 5 s for C and 10 s for P, to allow the complete
relaxation of the 1H nuclei and the number of scans was between
[
2] F. Averseng, M. Vennat, M. Che, in: Handbook of Heterogeneous Catalysis,
Wiley-VCH Verlag GmbH & Co. KGaA, 2008.
3] a) C. Merckle, J. Blümel, Chem. Mater. 13 (2001) 3617e3623;
b) C. Merckle, J. Blümel, Top. Catal. 34 (2005) 5e15;
3
000e5000 for carbon, 2000 scan for phosphorous and 32 for
[
proton. An apodization function (exponential) corresponding to a
line broadening of 80 Hz was applied prior to Fourier trans-
c) C. Merckle, S. Haubrich, J. Blümel, J. Organomet. Chem. 627 (2001) 44e54;
d) L. Huang, Y. He, S. Kawi, Appl. Catal. A General 265 (2004) 247e257;
e) L. Huang, S. Kawi, J. Mol. Catal. A Chem. 211 (2004) 23e33;
f) L. Huang, S. Kawi, Catal. Lett. 92 (2004) 57e62;
g) L. Huang, J. Wu, S. Kawi, React. Kinet. Catal. Lett. 82 (2004) 65e71;
h) L. Huang, J.C. Wu, S. Kawi, J. Mol. Catal. A Chem. 206 (2003) 371e387;
i) M. Marchetti, S. Paganelli, E. Viel, J. Mol. Catal. A Chem. 222 (2004) 143e151;
j) W. Zhou, D. He, Catal. Lett. 127 (2009) 437e443;
31
31
formation.
Pe P INADEQUATE were recorded on a Bruker
AVANCE III spectrometer operating at 162 MHz with a conventional
double resonance 4 mm CPMAS probe according to the following
general scheme. Following a short evolution time t , during which
1
they evolve under their combined resonance offsets, the DQ co-
herences are completely eliminated by phase cycling. For creating a
full PeRheP connectivity map between the various phosphate
k) H. Gao, R.J. Angelici, Organometallics 17 (1998) 3063e3069.
4] a) P. Li, S. Kawi, Catal. Today 131 (2008) 61e69;
[
[
b) S.C. Bourque, F. Maltais, W.-J. Xiao, O. Tardif, H. Alper, P. Arya, L.E. Manzer,
J. Am. Chem. Soc. 121 (1999) 3035e3038;
c) R. Abu-Reziq, H. Alper, D. Wang, M.L. Post, J. Am. Chem. Soc. 128 (2006)
units, 2D INADEAQUATE spectra were recorded using a spinning
ꢁ
speed of 14 KHz, 90 pulse of 2.5
ms length. 32 data sets with 2056
5279e5282.
scan each with delay of 1 s, under rotor-synchronized condition at
t1 increments of 71.43 s. A double coherence buildup time of
5] A. Grünberg, X. Yeping, H. Breitzke, G. Buntkowsky, Chem. e Eur. J. 16 (2010)
6993e6998.
[6] J. Blümel, Coord. Chem. Rev. 252 (2008) 2410e2423.
m
31
2
t
¼ 6 ms was used. P J-resolved solid state NMR spectroscopy
[
7] a) C. Sanchez, G.J.d.A.A. Soler-Illia, F. Ribot, T. Lalot, C.R. Mayer, V. Cabuil,
experiments were conducted on a Bruker AVANCE III spectrometer
using a 4 mm MAS probe. The experiments were performed ac-
cording to the following scheme pulse sequence. The first 900 pulse
was replaced by the cross-polarization segment and high-power
proton decoupling was performed during both the evolution (t1)
Chem. Mater. 13 (2001) 3061e3083;
b) C. Sanchez, L. Rozes, F. Ribot, C. Laberty-Robert, D. Grosso, C. Sassoye,
C. Boissiere, L. Nicole, C. R. Chim. 13 (2010) 3e39;
c) C. Sanchez, B. Julian, P. Belleville, M. Popall, J. Mater. Chem. 15 (2005)
3559e3592;
d) K.J. Shea, J. Moreau, D. A. Loy, R. J. P. Corriu, B. Boury, Wiley-VCH Verlag
and the detection (t
2
). The spectra were recorded using a spinning
GmbH & Co. KGaA, 2004, pp. 50e85 e) D. Avnir, T. Coradin, O. Lev, J. Livage,

108
J. Espinas et al. / Journal of Organometallic Chemistry 784 (2015) 103e108
J. Mater. Chem. 16 (2006) 1013e1030;
[14] M.G.L. Petrucci, A.K. Kakkar, Organometallics 17 (1998) 1798e1811.
[15] a) G. Wu, R.E. Wasylishen, Organometallics 11 (1992) 3242e3248;
b) G. Wu, R.E. Wasylishen, Inorg. Chem. 31 (1992) 145e148.
[16] a) A. Lesage, M. Bardet, L. Emsley, J. Am. Chem. Soc. 121 (1999) 10987e10993;
b) S.H. Santagneli, J. Ren, M.T. Rinke, S.J.L. Ribeiro, Y. Messaddeq, H. Eckert,
J. Non-Cryst. Solids 358 (2012) 985e992;
f) U. Schubert, N. Huesing, A. Lorenz, Chem. Mater. 7 (1995) 2010e2027.
[
[
8] a) R. Rahal, S. Daniele, L.G. Hubert-Pfalzgraf, V. Guyot-Ferr eꢀ ol, J.-F. Tranchant,
Eur. J. Inorg. Chem. (2008) 980e987;
b) R. Rahal, S. Daniele, H. Jobic, Chem. Phys. Lett. 472 (2009) 65e68.
9] G. Mason, Proc. R. Soc. Lond. Ser. A Math. Phys. Sci. 390 (1983) 47e72.
[
[
10] T. Lopez, E. Sanchez, P. Bosch, Y. Meas, R. Gomez, Mater. Chem. Phys. 32
1992) 141e152.
11] J. Kiefer, K. Obert, S. Himmler, P.S. Schulz, P. Wasserscheid, A. Leipertz,
ChemPhysChem 9 (2008) 2207e2213.
c) F. Fayon, G. Le Saout, L. Emsley, D. Massiot, Chem. Commun. (2002)
1702e1703;
d) S. Cadars, J. Sein, L. Duma, A. Lesage, T.N. Pham, J.H. Baltisberger, S.P. Brown,
L. Emsley, J. Magn. Reson. 188 (2007) 24e34.
(
[
[
12] S. Gould, D.A. Laufer, J. Magnetic Reson. (1969) 34 (1979) 37e55.
13] A.J. Naaktgeboren, R.J.M. Nolte, W. Drenth, J. Am. Chem. Soc. 102 (1980)
[17] C.M.H.J.M. De, J.W.E. Coenen, J. Mol. Catal. 11 (1981) 301e311.
3350e3354.