Hydrolytic stability and hydrogen peroxide activation of zirconium-based oxoclusters

Article abstract of DOI:10.1002/ejic.201402767

The hydrolytic stability of [Zr₆(OH)₄O₄{O(O)CC(CH₃)=CH₂}₁₂] (Zr₆), and [Zr₆O₄(OH)₄{O(O)CCH₂CH=CH₂}₁₂]₂·6[CH₂=CHCH₂C(O)OH] (Zr₁₂) oxoclusters in different environments was thoroughly investigated by FTIR, Raman, and X-ray photoelectron spectroscopy (XPS). Specific information about the local structures around the Zr centers during the stability tests was achieved by in situ extended X-ray absorption fine structure (EXAFS) measurements, and the exact compositions were determined by inductively coupled plasma MS (ICP-MS) and elemental analysis. By this multidimensional spectroscopic approach, an overview on the structures formed after different treatments could be gained. The stability of the oxoclusters was then investigated in the presence of hydrogen peroxide, and the formation of peroxo-metal complexes was detected. Thus, a kinetic study was performed in acetonitrile to evaluate the performances of the oxoclusters as oxygen transfer catalysts. The oxidation of methyl p-tolyl sulfide to the corresponding sulfoxide and sulfone was chosen as a model reaction; in some cases, an interesting selectivity towards the formation of the sulfone was found over more than 4700 catalytic cycles.

Full text of DOI:10.1002/ejic.201402767

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
FULL PAPER
DOI:10.1002/ejic.201402767
Hydrolytic Stability and Hydrogen Peroxide Activation of
Zirconium-Based Oxoclusters
Francesco Faccioli,[a] Matthias Bauer,[b] Danilo Pedron,[a]
Antonio Sorarù,[a] Mauro Carraro,*[a] and Silvia Gross*[a,c]
Keywords: Cluster compounds / Zirconium / Peroxides / Oxidation / EXAFS spectroscopy
The hydrolytic stability of [Zr6(OH)4O4{O(O)CC(CH3)=
CH2}12 (Zr6), and [Zr6O4(OH)4{O(O)CCH2CH=CH2}12]2·
structures formed after different treatments could be gained.
The stability of the oxoclusters was then investigated in the
presence of hydrogen peroxide, and the formation of peroxo–
metal complexes was detected. Thus, a kinetic study was
performed in acetonitrile to evaluate the performances of the
oxoclusters as oxygen transfer catalysts. The oxidation of
methyl p-tolyl sulfide to the corresponding sulfoxide and
sulfone was chosen as a model reaction; in some cases, an
interesting selectivity towards the formation of the sulfone
was found over more than 4700 catalytic cycles.
]
6[CH2=CHCH2C(O)OH] (Zr12) oxoclusters in different envi-
ronments was thoroughly investigated by FTIR, Raman, and
X-ray photoelectron spectroscopy (XPS). Specific information
about the local structures around the Zr centers during the
stability tests was achieved by in situ extended X-ray absorp-
tion fine structure (EXAFS) measurements, and the exact
compositions were determined by inductively coupled
plasma MS (ICP-MS) and elemental analysis. By this multi-
dimensional spectroscopic approach, an overview on the
precursors for the controlled nucleation of nanostructured
oxides,[4] and as secondary building units (SBU) for metal–
organic frameworks (MOFs).[5] Among the first reported
examples of these oxoclusters, one of the earliest to be
structurally characterized was [Ti6O4(OR)8{O(O)CMe}8],
which was obtained by the reaction of Ti(OR)4 (R = alkyl
group) with acetic acid.[6] In recent years, the synthesis of
different early transition and main group metal polynuclear
complexes (with O–M–O moieties) has been extensively ex-
plored, and oxoclusters based on Zr, Hf, Ti, Ti–Zr, Ag–Zr,
Y, T–-Hf, Zr–Ti–Hf, Ba, Ba–Ti, Ti–Y, Nb, and Sn have
been obtained.[1,3] Since the 1990s, Hubert-Pfalzgraf et
al.[7,8] have pioneered this development by exploring origi-
nal routes for the synthesis of a plethora of mono- and po-
lymetallic (e.g., Pb–Zr, Pb–Ti, Cu–Y, Y, lanthanides, La–
Zn, Ba–Ce, Bi–Ba, Zn–Ta, Y–Pr, Sm–Ti, and others) mixed
alkoxides and oxoclusters as potential single-source precur-
sors for the corresponding mixed functional oxides (e.g.,
perovskites). Further examples of metal oxide clusters
based on other transition elements (Fe, Cr) that feature a
triangular M3O core are also known.[9] The chemical prop-
erties, the tailored synthesis and modification, and the
structural issues of these oxoclusters as well as their use as
building blocks for the preparation of hybrid materials have
been thoroughly described in some reviews and research
papers.[1–3]
Introduction
Oxoclusters of the early transition metals are a class of
polynuclear compounds built from a limited number of
group 3–5 metal atoms, typically in their highest oxidation
state, linked by oxygen bridges and coordinated by biden-
tate organic ligands. Some examples also involve a second
alkaline earth metal (e.g., Ba, Mg).[1–3] They display dif-
ferent nuclearities (n = 3–12), coordination numbers of the
metal atoms (varying between six and nine), and connecti-
vity modes (corner, edge, or face sharing). These com-
pounds are globally neutral and discrete species with the
general formula [MyOx(OH)w{O(O)CR}z] (R = organic
group). During the synthesis (a controlled hydrolysis and
condensation of a metal alkoxide), carboxylic acids are in-
troduced to stabilize the structure of the inorganic core.[1–3]
Oxoclusters have been extensively used as building blocks
for inorganic–organic hybrid materials,[2,3] as molecular
[a] Dipartimento di Scienze Chimiche, Università degli Studi di
Padova,
via Marzolo 1, 35131 Padova, Italy
E-mail: mauro.carraro@unipd.it
www.chimica.unipd.it/mauro.carraro
[b] Department Chemie, Universität Paderborn,
Warburger Str. 100, 33106 Paderborn, Germany
[c] Istituto per l’Energetica e le Interfasi, IENI-CNR and INSTM,
UdR Padova,
Typically, these oxoclusters are characterized by the pres-
ence of d0 metal ions such as TiIV, ZrIV, HfIV, or NbV. It
has been widely reported that complexes containing high-
oxidation-state early transition metals display unique cata-
via Marzolo 1, 35131 Padova, Italy
E-mail: silvia.gross@unipd.it
www.chimica.unipd.it/silvia.gross
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201402767.
Eur. J. Inorg. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
lytic properties.[10] In several cases, they can activate benzene and tetrahydrofuran (THF), two poorly coordinat-
hydrogen peroxide and organic peroxides to promote the ing solvents, has already been reported.[16] In these studies,
oxidation of different substrates (olefins, sulfides and sulf- after the complete dissolution of the two compounds, the
oxides, and alcohols)[11] and can also enable enantioselec- precipitation of white powders was observed. Both the fil-
tive transformations.[12] However, although peroxo species tered solutions and the dried precipitates were analyzed by
involving zirconium atoms have been widely described[13] extended X-ray absorption fine structure (EXAFS) spec-
and interesting reactivities were observed for Zr-substituted troscopy at the ZrK-edge and by Raman spectroscopy.[17]
polyoxometalates,[14] there are still no reports of the use of Although the structures of the oxoclusters remained un-
Zr oxoclusters as oxidation catalysts in the presence of changed in solution, the X-ray absorption spectra of the
hydrogen peroxide. Hence, this works aims to shed light on powders revealed several differences with respect to those
the catalytic potential of such compounds. In addition, as of the starting compounds. It was supposed that a portion
one of the difficulties in using these Zr oxoclusters for cata- of the starting compounds degraded into dimeric or tri-
lytic purposes is their reactivity towards water or moisture, meric species, which precipitated because of their lower sol-
which leads to uncontrolled hydrolysis of the highly reactive ubility. Furthermore, temperature-dependent NMR spec-
M–O–M bonds, their stability under operative conditions troscopy studies on Zr4 and [Zr6(OH)4O4{O(O)CR}12] in
has been assessed.
CD2Cl2, as well as ab initio molecular dynamics simulations
Herein, the attention has been focused on the use of two on the latter oxoclusters, allowed the postulation of the
well-known zirconium oxoclusters, [Zr6(OH)4O4{O(O)CC- existence of rapid intramolecular exchange phenomena of
(CH3)=CH2}12] (Zr6),[1g,1h] and [Zr6O4(OH)4{O(O)CCH2- the bidentate ligands (“carboxylate shift”) on the surface of
CH=CH2}12]2·6CH2=CHCH2C(O)OH (Zr12),[1i] for cata- the oxocluster.[17b] The solvent appears to play a major role
lytic H2O2 activation. The two oxoclusters have been chosen in the structural modifications of oxoclusters, and its contri-
as model systems characterized by the same metal (Zr) but bution depends on its polarity and acidic features.
by different nuclearities (i.e., the number of metal atoms)
and structures. Zr12 can be regarded as a dimeric form of
the Zr6 species in which the two subunits are connected by
four carboxylate bridges (the structures are reported in the
Supporting Information, Figure S1). This latter feature, to-
gether with the different nuclearities and different func-
tional moieties on the surface (i.e, methacrylate vs. vinyl-
acetate groups), can account for the different hydrolytic be-
havior shown by the two species (vide infra). Formally, as
highlighted in ref.,[1i] the Zr12 clusters can be constructed
by opening two edge-bridging μ2-carboxylate ligands of Zr6
and letting two such units dimerize. Nevertheless, they are
structurally well defined and distinct polynuclear complexes
that cannot be interconverted, and their chemical behavior
has been the topic of a thorough investigation by Schubert
et al.[1i] However, the hydrolytic stability and possible use
of these polynuclear complexes in catalysis has not been
investigated. Our present investigation encompasses the as-
sessment of (1) their hydrolytic stability and (2) their ac-
tivity as sulfoxidation catalysts. As mentioned above, the
former point also entails a broader interest owing to the
use of these oxoclusters as effective versatile building blocks
for a plethora of different inorganic,[4] polymer-based,[1–3]
and porous hybrid materials.[5]
In the present work, we have focused on the effect of a
protic and polar solvent such as water on the structural
modification of Zr-based oxoclusters and, therefore, on
their stability. To accomplish this task, Zr6 and Zr12 were
(1) treated with aqueous solutions at different pH values,
(2) treated with a diluted hydrogen peroxide aqueous solu-
tion (30%), and (3) exposed to atmospheric air (Table 1).
The resulting compounds were analyzed by attenuated total
reflectance (ATR) FTIR, Raman, and EXAFS spec-
troscopy.[19–21] The possible changes in their morphologies
and structure after the treatments were investigated by
SEM, and their chemical compositions were determined by
inductively coupled plasma MS (ICP-MS) and CHNS ele-
mental analysis. The catalytic activity of the zirconium-
based oxoclusters towards H2O2 activation was finally
tested in the oxidation of methyl p-tolyl sulfide to investi-
gate the formation of metalloperoxides and catalyst sta-
bility. In this scenario, EXAFS and FTIR spectroscopy be-
fore and after catalysis allowed the assessment of the behav-
ior of the compounds under turnover conditions.
Table 1. Sample labeling and experimental parameters used for the
Zr6 and Zr12 hydrolytic treatments.
The hydrolytic stability of polynuclear metal complexes
is attracting growing interest because of the implications
that this has on their functional behavior, for example, in Sample Precursor Treatment
Appearance
the field of catalysis. For instance, the hydrolytic stability of
Zr-based polynuclear systems, in this case Zr-monosubsti-
tuted monomeric and dimeric polyoxometalates in aqueous
solutions at different pH values, has been the topic of a
recent publication by Carbò et al.[14h]
Zr6_ac Zr6_pre
Zr6_ne Zr6_pre
Zr6_ba Zr6_pre
Zr6_ox Zr6_pre
Zr6_air Zr6_pre
Zr12_ac Zr12_pre aq. solution at pH 2
Zr12_ne Zr12_pre aq. solution at pH 7
Zr12_ba Zr12_pre aq. solution at pH 13
aq. solution at pH 2
aq. solution at pH 7
aq. solution at pH 13
30% aq. solution of H2O2 white precipitate
atmospheric air
colorless solution
white suspension
white precipitate
same as Zr6_pre
colorless solution
white suspension
white precipitate
As reactions such as polymerizations with co-monomers
to obtain class II organic–inorganic hybrid materials are
Zr12_ox Zr12_pre 30% aq. solution of H2O2 pale yellow precipitate
Zr12_air Zr12_pre atmospheric air same as Zr12_pre
usually performed in organic solvents, the stability of
[1g,1h]
[Zr4O2{O(O)CC(CH3)=CH2}12] (Zr4)[15] and Zr6
in
Eur. J. Inorg. Chem. 0000, 0–0
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
tion, which is characteristic of the residual free acid mol-
ecules randomly dispersed in the voids of the crystal lat-
tice.[1g,1h] The stretching of the carbonyl group of these non-
coordinated methacrylic acid molecules is responsible for
Results and Discussion
1. Hydrolytic Stability of the Zr6 and Zr12 Oxoclusters
the strong absorption at ν = 1701 cm–1 (Figure 1, a), and
˜
Zr6 and Zr12 were synthesized according to literature
procedures, and their identities were confirmed by FTIR
and Raman spectroscopy. Although the Raman spec-
troscopy characterization of Zr6 had already been reported
by Kickelbick et al.,[17] the Raman spectrum of Zr12 is re-
ported for the first time in this work. The FTIR and Raman
spectra of Zr6 are shown in Figure 1 (a and b). As reported
previously,[16,17,19–21] the stretching modes of the C–H
bonds of the methyl and olefinic groups of the methacrylate
the peak at ν = 795 cm–1 in the Raman spectrum (Figure 1,
˜
b) can also be attributed to the vibrations of this com-
pound.[16,17] The C=C stretching of the ligands occurs at ν
˜
= 1645 cm–1. At ν = 1574 and 1423 cm–1, the asymmetric
˜
and symmetric stretching modes of the COO– group of the
methacrylate anion coordinated to the inorganic core of the
oxocluster can be observed, even if they are partially over-
laid with the bending frequency of the methyl groups. The
absorption at ν = 937 cm–1 can be ascribed to the out-of-
ligands can be observed between ν = 3099 and 2927 cm–1.
The whole area shows an overlaid O–H stretching absorp-
˜
˜
plane bending mode of the two C–H bonds of the olefinic
Figure 1. (a) FTIR and (b) Raman spectra of Zr6.
Eur. J. Inorg. Chem. 0000, 0–0
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
moiety of the methacrylate ligands.[15] The region of the Zr6 of the methyl groups are replaced by those characteristic of
Raman spectrum below ν = 400 cm–1 (Figure 1, b) is of methylene groups in the range between ν = 3080 and
˜
˜
high concern for the issues treated in the present work, as 3022 cm–1, together with those of the olefinic C–H bonds.
it can be considered as the “fingerprint” of oxocluster-like
The broad absorption between ν = 3000 and 3500 cm–1
˜
compounds. The Zr–O–Zr vibration modes are found in can be ascribed to the O–H stretching mode of the six
this region and can be related to the two broad peaks cen- crystallization molecules of vinylacetic acid per formula
tered at ν = 250 and 189 cm–1,[16,17] as found in the vi- unit.[1i] The stretching of the C=O group of these molecules
˜
brational spectra of crystalline zirconia[22] and amorphous is responsible for the sharp absorption at ν = 1714 cm–1,
˜
zirconium oxohydroxide.[23–25] The IR shoulder at ν ≈ and the vibrational mode of the carbon–carbon double
˜
790 cm–1 may be an additional spectral feature of the Zr– bond can be found at ν = 1643 cm–1. The remaining part
˜
O–Zr bonds.[14a,26]
of the spectrum is characterized by the stretching modes of
The FTIR and Raman spectra of Zr12 (Figure 2, a and b) the COO– group and by the bending modes of the methyl-
maintain the main features observed for the Zr6 oxocluster, ene and olefinic moieties.
although the different acid used for the synthesis (vinyl-
In analogy with the Zr6 oxocluster, we can assume that
acetic acid) results in some key differences in the spectra. the strong scattering tail observed in the Raman spectrum
Firstly, the absorptions attributed to the stretching modes below ν = 300 cm–1 could be related to the Zr–O–Zr vi-
˜
Figure 2. (a) FTIR and (b) Raman spectra of Zr12.
Eur. J. Inorg. Chem. 0000, 0–0
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
brational modes. The latter bridging bonds may also be clusters, the total Zr–O coordination number remains un-
responsible for the IR band at ν = 720 cm–1.[14a,26]
changed, but the distribution of the eight oxygen atoms
˜
To test the hydrolytic stability of the title compounds, within the first two Zr coordination spheres is modified.
each zirconium oxocluster was left for a determined time in Unlike the precursor, for which the oxygen distribution
(1) water at three distinct pH values of 2, 7 and 13 (see around the Zr absorbing atom results in a 2+6 motif, a 4+4
Experimental Section), (2) a hydrogen peroxide aqueous motif was observed, and the distances of the first two shells
solution (30% w/w), and (3) atmospheric air for compari- of backscatters increased slightly (Table 2, Entry 2).
son (Table 1). After the treatments, the compounds were
collected and analyzed by ATR FTIR, Raman, and shell is slightly decreased and their distance remaining un-
EXAFS spectroscopy. changed within error. The average number of backscatter-
The coordination number of the Zr atoms in the third
The resulting spectra were then compared with those of ing oxygen atoms (N = 4 + 4) at 2.14–2.15 and 2.31 Å and
the reference compounds stored under a controlled argon zirconium atoms (N = 3) at a distance of 3.54 Å from the
atmosphere (Zr6_pre and Zr12_pre).
central absorbing atom seems to be compatible with the
cited tetramer [Zr4(OH)8(H2O)16]8+ [28]
,
but the presence of
1.1 Treatment of Zr6 and Zr12 under Acidic Conditions
more-complex oligomeric species likely involving the meth-
acrylate ligands cannot be excluded.
The aqueous solution chemistry of ZrIV is quite com-
plex.[24,25] The presence of mononuclear Zr4+ cations in
solution is improbable, even at extremely low pH values. As
the ion is relatively large and highly charged, slow equilibria
between polymeric partially hydrolyzed species occur, and
their nuclearity depends on the strength of the acid em-
ployed.[25] In 2.8 m HCl solution, the most-stable ion ap-
pears to be [Zr3(OH)6Cl3]3+, whereas the main species in
1–2 m perchloric acid are thought to be [Zr3(OH)4]3+ and
1.2 Treatment of Zr6 in Atmospheric Air and under
Neutral, Basic, and Oxidative Conditions
For Zr6, the ATR FTIR and Raman spectra (Figure 3)
show no appreciable degradation of the oxocluster samples
exposed to atmospheric air or to neutral aqueous solution.
The Raman spectra of Zr6_pre, Zr6_air, and Zr6_ne in the
fingerprint zone below ν = 500 cm–1 are definitely the same
˜
(Figure 3, aЈ, bЈ, and cЈ), and the absorption and scattering
bands related to the C–H and C=C bonds and the COO–
group of the methacrylate ligands in Zr6_air and Zr6_ne
also remain unchanged compared to those of Zr6_pre (Fig-
ure 3, b, bЈ, c, and cЈ). The broadening of the two bands
belonging to the two different asymmetric stretching modes
[Zr4(OH)8]8+ [25]
ion is widely reported as the “zirconyl cation” [ZrO]2+ [24–28]
.
Although the stable form of the hydrolyzed
,
no crystallographic evidence of such a group has been
found in any of its salts.[24,25] For example, hydrochloride
solutions of zirconyl oxychloride (ZrOCl2·8H2O) contain
the tetrameric species [Zr4(OH)8(H2O)16]8+ [24,25]
as has
,
of the COO– group results in a band centered at ν =
˜
been confirmed by X-ray diffraction and absorption mea-
surements.[29–34] In our experiments, no solid could be re-
trieved from the acidic solution (pH 2), and the Raman
spectra of Zr6_ac and of Zr12_ac were of little use owing to
the strong scattering of the solvent (Figure S2). Conse-
quently, the most valuable information came from the
EXAFS analyses of the liquids (Table 2).[34] For both oxo-
1549 cm–1 in the ATR FTIR spectra of Zr6_air and Zr6_ne
(Figure 3, b and c); this can be explained by a slight modifi-
cation of the coordination mode of the carboxylate ligands
to the central inorganic cluster core.
An effect of the treatments is the weakening of the bands
of the free acid molecules randomly dispersed into the voids
of the lattice at ν = 1701 and 795 cm–1,[1g,1h] which are pres-
˜
ent in both the ATR FTIR and Raman spectra.
The solvating power of a protic and polar solvent such
as water (or atmospheric moisture) could have promoted
the dissolution of these adsorbed molecules without in-
fluencing the main oxocluster structure.
Table 2. Structural parameters obtained by fitting the experimental
EXAFS data of Zr6_ac and Zr12_ac with theoretical models.
Entry Sample Abs–Bs[a] N(Bs)[b] R(Abs–Bs) [Å][c] σ [Å2][d]
Ef [eV]
(R)[e]
A very different picture was obtained after treatment in
basic conditions. Indeed, the ATR FTIR and Raman spec-
tra of Zr6_ba are completely different from those of Zr6_pre
(Figure 3, d and dЈ). In this case, the degree of hydrolysis is
marked, as can be seen by the broad absorptions centered
Zr6_pre
Zr6_pre
Zr6_pre
Zr6_ac
Zr6_ac
Zr6_ac
Zr12_pre
Zr12_pre
Zr12_pre Zr–Zr
Zr12_ac
Zr12_ac
Zr12_ac
Zr–O
Zr–O
Zr–Zr
Zr–O
Zr–O
Zr–Zr
Zr–O
Zr–O
2
6
4
2.11Ϯ0.02
2.27Ϯ0.02
3.52Ϯ0.04
2.14Ϯ0.02
2.31Ϯ0.02
3.54Ϯ0.04
2.13Ϯ0.02
2.28Ϯ0.02
3.53Ϯ0.04
2.15Ϯ0.02
2.31Ϯ0.02
3.54Ϯ0.04
0.004
0.013
0.012
0.002
0.002
0.012
0.004
0.013
0.014
0.006
0.005
0.012
1
2
3
4
5.2 (27.8)
3.9 (22.6)
5.6 (29.9)
5.3 (23.9)
4.1Ϯ0.4
4.1Ϯ0.4
3.0Ϯ0.6
2
at ν = 3300 and 1600 cm–1 in the ATR FTIR spectrum (Fig-
˜
ure 3, d). According to a previous report, the former can be
assigned to the O–H asymmetric and symmetric stretching
modes of lattice water molecules, and the latter can be as-
cribed to the corresponding H–O–H bending mode.[35] The
O–H bending mode of the hydroxy groups bonded to the
6
4
Zr–O
Zr–O
Zr–Zr
4.4Ϯ0.4
4.4Ϯ0.4
3.0Ϯ0.6
Zr atoms can be clearly seen at ν = 1060 cm–1 in the Raman
˜
[a] Abs = X-ray absorbing atom, Bs = backscatterer. [b] Coordina-
tion number. [c] Interatomic distance Abs–Bs. [d] Debye–Waller-
like factor. [e] Ef accounts for a shift between the experimental and
spectrum of Zr6_ba (Figure 3, dЈ), in agreement with the
literature value.[35] The two broad peaks at ν = 414 and
˜
calculated EXAFS function, R is the quality of fit (deviation in %). 523 cm–1 are probably ascribed to the Zr–O vibrations and
Eur. J. Inorg. Chem. 0000, 0–0
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
Figure 3. ATR FTIR (top) and Raman spectra (bottom) of Zr6 after the hydrolytic treatments. (a, aЈ) Zr6_pre, (b, bЈ) Zr6_air, (c, cЈ)
Zr6_ne, (d, dЈ) Zr6_ba, and (e, eЈ) Zr6_ox.
Zr–O–Zr bending modes.[23–25,33] The FTIR and Raman ATR FTIR and Raman spectra (Figure 3, d and dЈ), such
spectra of Zr6_ba both show analogies to those of oxohyd- a polymer is characterized by the presence of bridging or
roxo zirconium species obtained through the basic precipi- terminal hydroxy groups, organic methacrylate ligands still
tation of gels from acidic solutions of zirconyl chloride.[33] coordinated to the inorganic core, and bonded or adsorbed
Accordingly, the precipitate analyzed displays several fea- water molecules. The suggested polymerization mechanism
tures in common with zirconium oxohydroxides, ZrOx(OH) agrees with that proposed by Clearfield[36,37] to explain the
4–2x·n(H2O), although it contains some residual methacry- different properties of amorphous hydrous zirconia precipi-
late ligands, as revealed by the weak Raman resonances in tated from zirconyl chloride solutions at different pH val-
the methyl stretching and bending region (Figure 3, dЈ). The ues. Indeed, these aqueous systems contain the stable tetra-
precipitation process could involve a nucleophilic attack of meric cation [Zr4(OH)8(H2O)16]8+ [28]
in which the Zr–OH2
,
OH– ions (in a large excess in solution) to the Zr coordina- bonds undergo hydrolysis and polymerization at a rate
tion sphere; this would change the hapticity of the meth- strongly dependent on the availability of OH– ions.[32]
acrylate ligands and, eventually, they would be released into
The ATR FTIR and Raman spectra of Zr6_ox are shown
solution. Further condensation of the hydroxide-modified in Figure 3, e and eЈ, respectively. As in the previous case
oxocluster could lead to polymerization. As shown by the (Zr6_ba), an extended hydrolysis of the precursor Zr6_pre
Eur. J. Inorg. Chem. 0000, 0–0
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
(Figure 3, a and aЈ) is evident, as shown by the strong and Table 4. EXAFS data of Zr6 samples after hydrolytic treatment (see
broad absorption bands centered at ν = 3300 and 1600 cm–1
Table 2 for acronyms used).
˜
in the ATR FTIR spectrum (Figure 3, e).[35] The presence
of water molecules in the precipitate is in agreement with
the peroxidic hydrolysis of ZrIV salts with an excess of H2O2
reported by Connor et al.[13b] More detailed information
can be gained from the Raman spectrum, in which the vi-
brations of the residual C–H groups of the methacrylic acid
Entry Sample Abs–Bs
N(Bs)
R(Abs–Bs)
σ
Ef [eV]
(R)
[Å]
[Å2]
1
2
3
4
5
Zr6_pre Zr–O
Zr6_pre Zr–O
Zr6_pre Zr–Zr
2
6
4
2.11Ϯ0.02
2.27Ϯ0.02
3.52Ϯ0.04
2.11Ϯ0.02
2.27Ϯ0.02
3.53Ϯ0.04
2.12Ϯ0.02
2.27Ϯ0.02
3.54Ϯ0.04
2.10Ϯ0.02
2.23Ϯ0.02
3.45Ϯ0.03
2.18Ϯ0.02
3.36Ϯ0.03
0.004
0.013
0.012
0.004
0.013
0.012
0.004
0.013
0.012
0.004
0.013
0.012
0.016
0.016
5.2 (27.8)
4.8 (26.1)
4.9 (25.8)
Zr6_air
Zr6_air
Zr–O 2. 5Ϯ0.3
Zr–O 6.5Ϯ0.7
can be observed between ν = 2879 and 2986 cm–1 and at ν
˜
˜
Zr6_air Zr–Zr 3.5Ϯ0.7
≈ 1400 cm–1. Their presence in the precipitate can be ex-
plained on the basis of the hydrolysis mechanism of the
oxocluster proposed before, owing to the acidity of H2O2
solutions (the pH of a 30% aqueous solution of H2O2 is as
Zr6_ne
Zr6_ne
Zr6_ne
Zr6_ba
Zr6_ba
Zr–O
Zr–O
Zr–Zr 3.2Ϯ0.6
Zr–O
Zr–O
2.6Ϯ0.3
6.5Ϯ0.7
2.8Ϯ0.3
5.2Ϯ0.5
5.9 (30.6)
7.8 (22.9)
low as 2). A quite intense band can be observed at ν =
˜
Zr6_ba Zr–Zr 0.8Ϯ0.1
839 cm–1 in the Raman spectrum of Zr6_ox, followed by a
Zr6_ox
Zr6_ox
Zr–O
8.7Ϯ0.9
broader one at ν = 436 cm–1 with a shoulder at 517 cm–1
Zr–Zr 0.9Ϯ0.1
˜
(Figure 3, eЈ). According to previous work on bulk zirconia
treated with hydrogen peroxide solution, the former is as-
signed to peroxo ν(O–O) vibrations, and the second is at-
tributed to the superimposed Zr(O2) asymmetric and sym-
metric stretching modes,[35,38] although an overlap with the
scattering bands related to the Zr–O–Zr vibrational modes
has to be considered. Even if the attribution of the hapticity
of the peroxo ligand in metal complexes (bidentate, bridg-
ing rather than terminal) through spectroscopic methods is
quite difficult,[13b,13c] a comparison with the Raman and
FTIR characterization of previously reported Zr complexes
The treatments in neutral water and in atmospheric air
(Zr6_ne and Zr6_air) yield identical structural parameters
to those of Zr6_pre within error.
On the contrary, basic conditions induce significant
changes; the most striking changes are for the higher Zr–O
and Zr–Zr shells; the Zr–Zr coordination number in par-
ticular suggests that the initial Zr core is reduced to a di-
meric motif,[17] and the distance parameter changes from
3.52 to 3.45 Å.
For Zr6_ox, a Zr–O distance of 2.18 Å is also characteris-
tic of Zr–peroxo groups.[13p,14] However, as only one average
Zr–O shell could be fitted, no further information about
possible peroxo groups can be extracted from the EXAFS
measurements.
Additionally, only one Zr–Zr shell is found at 3.36 Å.
This picture corroborates the presence of peroxo ligands co-
ordinated to the zirconium atoms in Zr6_ox, as already
indicated by the analysis of its Raman spectrum.
Finally, a significant structural alteration is found in
Zr6_ox, as only one Zr–O shell with 8.7Ϯ0.9 atoms can be
found. The corresponding distance of 2.18 Å lies between
the original Zr–O distances of the Zr6_pre sample.
such as K2[Zr(O2)(ox)2]·2H2O (ox = oxalate), K2[Zr(O2)2-
[13h]
(cit)]·2H2O (cit = citrate),[13j] and (NH4)3Zr(O2)F5
(all
bearing η2-peroxo ligands) strengthen the hypothesis that
the real coordination mode of the O–O ligand to zirconium
is bidentate, especially in light of the position of the O–O
stretching band at ν = 839 cm–1 in the Raman spectrum of
˜
Zr6_ox (see Table 3).
Table 3. Comparison of the vibrational frequencies of the O–O and
Zr(O2) bonds in Zr6_ox, Zr12_ox, and some previously reported
compounds.
ν(O–O) [cm–1
]
ν[Zr(O2)] [cm–1
]
Reference
this work
Zr6_ox
Zr12_ox
839
844
848
853
840
850
837
877
517, 436
517, 446
565
this work
[13j]
K2[Zr(O2)(ox)2]·2H2O
K2[Zr(O2)2(cit)]·H2O
Zr(O2)(H2EDTA)
(NH4)2[ZrO(O2)F2]
(NH4)3[Zr(O2)F5]
H2O2
[13j]
1.3 Treatment of Zr12 in Atmospheric Air and under
Neutral, Basic, and Oxidative Conditions
550
[13k,35]
[13l,35]
[13h,35]
[13c]
650, 600
640, 585
550, 471
–
The ATR FTIR and Raman spectra of the Zr12 samples
treated in the same hydrolytic environments as Zr6 are dis-
played in Figure 4. Some analogies in the behavior of the
two oxoclusters can be pointed out: Zr12 is also quite resist-
As this value is significantly different from that of free ant towards hydrolysis from the moisture in atmospheric
hydrogen peroxide (the last entry in Table 3), the presence air. In the Raman spectrum of Zr12_air, no substantial
of hydrogen peroxide molecules of crystallization or ad- modifications can be observed with respect to that of
sorbed in the sample can be excluded. The existence of co- Zr12_pre (Figure 4, bЈ).
ordinated peroxo groups in the isolated compound has been
unequivocally verified by a qualitative iodometric assay[27] two different asymmetric stretching modes of the COO–
on Zr6_ox previously washed with deionized water. group results in the appearance of one broad band centered
The results of the EXAFS analysis (Table 4) confirm the at ν = 1548 cm–1 in the ATR FTIR spectrum of Zr12_air
As for Zr6_air, the broadening of the two bands of the
˜
findings of the vibrational spectroscopy characterizations.
(Figure 4, b); this can be explained by a slight modification
Eur. J. Inorg. Chem. 0000, 0–0
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
Figure 4. ATR FTIR (top) and Raman spectra (bottom) of Zr12 after the hydrolytic treatments. (a, aЈ) Zr12_pre, (b, bЈ) Zr12_air, (c, cЈ)
Zr12_ne, (d, dЈ) Zr12_ba, and (e, eЈ) Zr12_ox.
of the coordination mode of the vinylacetate ligands to each oxocluster (Figure 4, d and dЈ). The signs of an advanced
of the two inorganic subunits from which Zr12 is built up.[1i] hydrolysis are evidenced by the broad absorptions centered
As was already observed for Zr6, the treatment with a at ν = 3300 and 1600 cm–1 in the ATR FTIR spectrum of
˜
neutral aqueous solution seems to cause only some slight Zr12_ba (Figure 4, d). However, the presence of weak peaks
modifications to the pristine oxocluster structure, as is evi- at ν ≈ 1400 and 2900 cm–1 in the Raman spectrum (Fig-
˜
dent from the fingerprint region of the Zr12_ne spectrum in ure 4dЈ) suggests that traces of vinylacetate ligands are still
Figure 4 (cЈ). The feature common to the spectra of present in the precipitate.
Zr12_air and Zr12_ne is the decreased intensity of the band
Moreover, the striking analogies between both the ATR
at ν = 1714 cm–1 (Figure 4, bЈ and cЈ) for crystallization FTIR and Raman spectra of Zr12_ba and Zr6_ba (Figure 5)
˜
molecules of vinylacetic acid, which could be replaced dur- makes us confident that the precipitate obtained in the two
ing the transition of the compound from a crystalline to an cases is the same, at least as far as the inorganic moiety is
amorphous form.
concerned. The Raman spectrum of Zr12_ba presents the
In analogy with results for Zr6_ba, the treatment of Zr12 same bands at ν = 1060, 414, and 523 cm–1 that charac-
˜
with a basic solution radically changes the structure of the terized Zr6_ba (Figure 5, aЈ and bЈ).
Eur. J. Inorg. Chem. 0000, 0–0
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
Figure 5. ATR FTIR (top) and Raman spectra (bottom) of Zr6_ba (a and aЈ) and Zr12_ba (b and bЈ). The Raman spectra are normalized
with respect to the peak at ν = 1060 cm–1.
˜
As already discussed, these bands can be assigned to O– (Table 3), and the band at ν = 446 and the shoulder at
˜
H bending,[35] Zr–O vibrations, and Zr–O–Zr bending 517 cm–1 can be ascribed to superimposed Zr(O2) asym-
modes, respectively;[23,33] therefore, it can be supposed that metric and symmetric stretching modes,[35,38] which are
the precipitate in this case is also built up of a zirconium probably overlapped with the scattering bands of the Zr–
oxohydroxide species of the general formula ZrOx(OH)4–2x
·
O–Zr vibrational modes.[23]
n(H2O) with vinylacetate residues.
As for Zr6_ox, the presence of peroxo groups in the com-
The ATR FTIR and Raman spectra of Zr12_ox are pound has been further confirmed by a qualitative iodo-
shown in Figure 4, e and eЈ, respectively. Their features metric assay.[27] Nonetheless, as was pointed out for Zr12_ba
mostly resemble those of Zr6_ox, as can be seen from the and in analogy with the Zr6_ox case, the presence of an
comparison in Figure 6. On the basis of the same consider- organic moiety in Zr12_ox is evident. The broad peaks cen-
ations made for Zr6_ox, the peak in the Raman spectrum tered at ν = 1427 and 2937 cm–1 resemble those of the C–
˜
of Zr _ox at ν = 844 cm–1 (Figure 4, eЈ) can be assigned to H bending and stretching modes of the vinylacetate ligands
˜
12
the O–O stretching mode of a bidentate peroxide ligand in Zr12_pre (Figure 4, aЈ). Hence, a partial hydrolysis of the
Eur. J. Inorg. Chem. 0000, 0–0
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
Figure 6. ATR FTIR (top) and Raman spectra (bottom) of Zr6_ox (a and aЈ) and Zr12_ox (b and bЈ). The Raman spectra are normalized
with respect to the peaks at ν = 839 and 844 cm–1.
˜
oxocluster can also be inferred in this case, together with units of Zr12,[1i] structural rearrangements involving the de-
the formation of peroxo species on the surface of the oxide composition of the Zr12 cluster into separate Zr6 units can-
network.
not be excluded.[39,40] Nonetheless, it can be inferred from
The conclusions drawn from the ATR FTIR and Raman the vibrational spectra of Zr12_air (Figure 4, b and bЈ) that
spectroscopy characterizations seem to be confirmed by the the structure of Zr12 is not modified upon exposure to at-
EXAFS data (Table 5).
mospheric moisture, although the same cannot be assumed
It may be pointed out that the treatments with atmo- from the fingerprint region in the Raman spectrum of
spheric air and water at pH 7 (Zr12_air and Zr12_ne) do not Zr12_ne (Figure 4, cЈ). The same trend observed throughout
have any major impact on the formed structures, as was the the series of the Zr6 treatments is repeated for Zr12_ba and
case for Zr6_air and Zr6_ne, because the number of back- Zr12_ox. The changes to their structures are highlighted
scatterers and the distance remain unchanged. However, as mainly by the increment of the oxygen atoms in the Zr–O
EXAFS probes only the local environment of the absorbing shell and the decrement of the Zr atoms in the Zr–Zr neigh-
atoms and does not provide information on the distances bors, as well as their distance parameters. Once more, the
between the zirconium atoms located in two different sub- structural parameters (coordination numbers and shell ra-
Eur. J. Inorg. Chem. 0000, 0–0
10
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
Table 5. EXAFS data of Zr12 samples after hydrolytic treatments (TOF, calculated as mol of oxidized substrate per mol of
(see Table 2 for acronyms used).
catalyst per time unit), and selectivity ratio S = k1/k2. The
results obtained with both oxoclusters after 3 h (when the
substrate is completely consumed in most cases) are col-
lected in Table 6.
Entry Sample Abs–Bs
N(Bs)
R(Abs–Bs)
σ
Ef [eV]
(R)
[Å]
[Å2]
1
2
3
4
5
Zr12_pre Zr–O
Zr12_pre Zr–O
Zr12_pre Zr–Zr
2
6
4
2.13Ϯ0.02
2.28Ϯ0.02
3.53Ϯ0.04
2.13Ϯ0.02
2.27Ϯ0.02
3.54Ϯ0.04
2.13Ϯ0.02
2.28Ϯ0.02
3.54Ϯ0.04
2.11Ϯ0.02
2.25Ϯ0.02
3.46Ϯ0.03
2.20Ϯ0.02
3.36Ϯ0.03
0.004
0.013 5.6 (29.9)
0.014
0.004
0.013 5.4 (29.9)
0.014
0.004
0.013 5.6 (27.3)
0.012
0.004
0.013 6.2 (29.0)
0.014
Table 6. Oxidation of methyl p-tolyl sulfide with metal oxoclus-
Zr12_air
Zr12_air
Zr–O
Zr–O
2.0Ϯ0.2
6.0Ϯ0.6
4.0Ϯ0.8
2.4Ϯ0.3
6.7Ϯ0.7
3.8Ϯ0.7
3.1Ϯ0.3
5.8Ϯ0.6
1.0Ϯ0.2
8.7Ϯ0.9
0.4Ϯ0.1
ters.[a]
Zr12_air Zr–Zr
Entry Oxocluster Catalyst Conversion TON[c] TOF[d] SO/
S
Zr12_ne
Zr12_ne
Zr–O
Zr–O
[c]
[mol-%]
[%][b,c]
[min–1] SO2 (k1/k2)
1[e]
2[e]
3[f]
4[f]
5[f]
6[f]
7[f]
8[f]
9[f]
10[f]
Zr12
Zr6
Zr12
Zr6
Zr12
Zr6
Zr12
Zr6
0.08
0.08
0.08
0.08
0.16
0.16
0.24
0.24
0.32
0.32
90
90
Ͼ99
96
Ͼ99
97
Ͼ99
99
1850
1225
2350
1700
1225
850
817
708
613
513
27
18
81
47
67
25
74
45
81
31
37:63 1.4
91:9 16
12:88 1.1
57:43 8.7
Zr12_ne Zr–Zr
Zr12_ba
Zr12_ba
Zr–O
Zr–O
Zr12_ba Zr–Zr
Zr12_ox Zr–O
Zr12_ox Zr–Zr
0.018 8.3 (24.2)
0.012
4:96
61:39 2.2
4:96 0.6
28:72 2.0
3:97 0.6
35:65 1.8
0.6
Zr12
Zr6
Ͼ99
99
dius) of Zr12_ox could be compatible with the coordination
of peroxo ligands to zirconium atoms in the final com-
pound, also in view of the results of the Raman spec-
troscopy characterization.
[a] Reaction conditions: the specified amount of oxocluster was
added to acetonitrile (1.2 mL) containing methyl p-tolyl sulfide
(1 mmol) and H2O2 (2.5 mmol, added as 30% aqueous solution).
[b] Methyl p-tolyl sulfide conversion. A blank reaction at 50 °C
without oxoclusters led to sulfide consumption in 24 h with no sulf-
one formation. [c] Values obtained at t = 3 h. [d] Number of cata-
lytic cycles per minute, measured at ca. 20% substrate conversion.
[e] T = 30 °C. [f] T = 50 °C.
2. Catalytic Tests
Several mechanistic studies of oxygen transfer reactions
with mononuclear d0 peroxometal complexes of the general
At 30 °C (Table 6, Entries 1 and 2), the two oxoclusters
formula [LnM(O2)m] (M = TiIV, VV, MoVI, WVI) have been display a different reactivity, and Zr12 is more efficient than
reported.[11a,12a] Accordingly, we have addressed the oxi- Zr6. The statistical factor, that is, the increased number of
dation of aromatic sulfides by H2O2 in the presence of a metal centers in Zr12, is not responsible for this higher ac-
catalytic amount of oxocluster. The oxidation of sulfur-con- tivity, as the turnover frequency per metal ion (TOF/Zr) is
taining compounds is an important synthetic process[41] for 2.3 and 3 for Zr12 and Zr6, respectively (see also Table 6,
the production of sulfoxides and sulfones and is also widely Entries 3 and 6 with the same total concentration of Zr
explored as a method for the oxidative desulfurization of ions). Thus, the hydrolysis of the dimeric structure to two
fuels.[42] Under such conditions, methyl p-tolyl sulfide, cho- hexanuclear clusters with analogous reactivity can be ex-
sen as a model substrate, is oxidized to its corresponding cluded.
sulfoxide and then to the sulfone by consecutive bimolecu-
The different reactivity of the two oxoclusters is also
highlighted by the different product distribution and selec-
lar, mono-oxygenation steps (Scheme 1).[43]
The reactions were performed in acetonitrile at different tivity ratio S, which is tenfold lower for Zr12. Indeed, al-
temperatures (T = 30 and 50 °C) with Zr6 and Zr12 as cata- though k1 for the dimeric oxocluster is double the corre-
lysts and an excess of H2O2 (2.5 equiv.) to promote the oxi- sponding value for Zr6, k2 is more than twenty times bigger
dation to sulfone. With k1 and k2 as the second-order (see Supporting Information, Table S1).
kinetic constants of the first (from sulfide to sulfoxide, k1)
As demonstrated by the higher reaction yield and the
and second (from sulfoxide to sulfone, k2) oxidation steps, increased sulfone production, the reactions are faster at
the reactions have been compared in terms of sulfide con- 50 °C (Table 6, Entries 3–4 and Table S1), and a TON of
version, turnover number (TON, calculated as mol of oxid- 2350 was obtained after 3 h with Zr12. The kinetic behavior
ized substrate per mol of catalyst), turnover frequency of the reactions with the two catalysts under such condi-
Scheme 1. Oxidation of methyl p-tolyl sulfide to methyl p-tolyl sulfoxide and methyl p-tolyl sulfone by H2O2.
Eur. J. Inorg. Chem. 0000, 0–0
11
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
tions is shown in Figure 7 and highlights the different con- trations is in agreement with first-order kinetics with re-
centration profile of the sulfoxide intermediate, which is spect to the concentration of both oxoclusters and excludes
readily oxidized only in the presence of Zr12.[44]
the dissociation of the dimeric structure into smaller active
species (Figure S12).
Concerning the electronic character of the oxidant spe-
cies, previous kinetics studies showed that mononuclear d0
metal peroxide complexes usually behave like electrophilic
oxidants,[43a,43b] and the S ratio is generally much higher
than 1, as the rate of sulfide oxidation is up to two orders
of magnitude faster than that of the following step.[43a,43d,45]
In our case, lower S values were generally obtained, and a
reversal of reactivity was observed at 50 °C for Zr12, for
which S became less than 1. For this latter system, the high
reactivity towards the less-activated sulfoxide, as a result of
an interplay of orbital and electrostatic interactions, may
reveal an electron-rich character of the active species.[46]
A comparison of the elemental analyses before and after
the reaction showed a major recovery of the Zr content for
both catalysts: values of 29.9 (calcd. 22.8%) and 27.6%
(calcd. 28.0%) were found for Z6 and Zr12, respectively.[47]
However, major changes where found for the composition
of the organic domain, and a larger decrease of the organic
components was observed for Zr6 (see Table S2). In ad-
dition, the presence of a small amount of sulfur (up to 3.5%
for Zr12) suggests a significant adsorption of the products
on the spent catalysts. To check for any poisoning or deacti-
vation, the reaction mixtures for Table 6, Entries 3 and 4
were recharged with further substrate and oxidant: thus, the
initial concentration of the reactant and H2O2 were re-
stored, and the reaction started again. An analogous effi-
ciency (98% conversion) and product distribution (SO/SO2
71:29) were obtained, in 3 h, in the presence of Zr6 (total
TONϾ 3000), whereas a lower efficiency with respect to
the first run was observed for Zr12, which provided a 99%
conversion and a SO/SO2 ratio of 47:53 in 3 h (total
TONϾ4700).
Figure 7. Kinetic traces for the oxidation of methyl p-tolyl sulfide
(squares) to its sulfoxide (circles) and sulfone (triangles) by
hydrogen peroxide in the presence of 0.08 mol-% of Zr6 (top) or
Zr12 (bottom) at 50 °C.
The decreased performance of the catalysts seems to be
connected with the appearance of a white precipitate in the
reaction vessels. Thus, some useful insights were obtained
A different oxidant utilization efficiency has also been through the analysis of the precipitates recovered from reac-
highlighted for the two oxoclusters. With a stoichiometric tion mixtures, namely, Zr6_cat and Zr12_cat. Firstly, the
amount of H2O2 (2 equiv.), we observed only minor perox- FTIR spectra of the retrieved solids were recorded after the
ide decomposition in the presence of Zr12, which gave 98% solid was washed with isopropanol. The analogies with the
yield in 8 h and 4:96 SO/SO2 selectivity, whereas the H2O2 spectrum of the oxocluster before the reaction are remark-
was completely consumed after 24 h to afford 88% yield able for Zr6_cat, for which the COO– stretching modes and
and 23:77 selectivity with Zr6.
(=)C–H bending absorbances are maintained, whereas
An increased amount of Z12 (more than 0.16 mol-%) al- minor differences in the spectral features can be observed
lowed the complete conversion of the substrate to sulfone for Zr12_cat (Figure S13), for which the C=C and C=O
in ca. 3 h (with respect to Entries 1 and 2, Table 6). A TOF stretching bands appear at lower wavenumbers and a few
of up to 81 min–1 was found under such temperature condi- signals of methyl p-tolyl sulfone can also be recognized.
tions. The general decreasing trend of the S values for both
A further insight into the stability of the oxoclusters un-
oxoclusters (Table 6, Entries 1 and 2) suggests the occur- der catalytic conditions was obtained from the EXAFS
rence of a temperature-dependent evolution of the active data. The ZrK-edge EXAFS spectra of Zr6_cat and
species. However, as smooth kinetic traces with no induc- Zr12_cat show that the distance parameters do not vary re-
tion time were recorded in all cases (Figures 7 and S4–S11), markably with respect to those of the precursors (Table 7).
the evolution to the active species is likely to occur just after For Zr6_cat, the first Zr–O coordination number is slightly
H2O2 addition. In addition, the linear behavior observed increased, and the second oxygen contribution remains un-
for the initial reaction rate (R0) at different catalyst concen- altered within error. As previously discussed (see Table 1),
Eur. J. Inorg. Chem. 0000, 0–0
12
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
spectral changes are likely to result from modifications of EXAFS spectroscopy investigations. The oxoclusters are
the coordination modes or displacement of the carboxylate stable in air and neutral aqueous solution, whereas they
ligands.
evolve to lower nuclearity complexes or to polymeric struc-
tures under acidic and alkaline conditions, respectively. As
indicated by the Raman and EXAFS spectroscopy results,
the zirconium oxoclusters can coordinate hydrogen perox-
ide. The applicability of the resulting peroxo complexes as
oxidation catalysts was demonstrated by their ability to pro-
mote oxygen transfer to a model aromatic sulfide in aceto-
nitrile solution. Indeed, methyl p-tolyl sulfide was smoothly
oxidized in 100–500 min to yield a mixture of the corre-
sponding sulfoxide and sulfone. Under multi-turnover con-
ditions, the two oxoclusters exhibit different stabilities and
electronic characters. Zr6 is generally less reactive (58%
yield of oxygenated products in 3 h with 0.08 mol-% cata-
lyst, whereas Zr12 afforded 90% yield) with decreased ac-
tivity towards sulfoxide oxidation and lower H2O2 utiliza-
tion efficiency. However, the monomeric oxocluster seems
to be more stable under operative catalysis conditions, as
the recovered catalyst maintains the spectral features of the
starting material, and its reactivity can be restored upon the
addition of further quantities of reagent. Despite the re-
duced hydrolytic stability of Zr12, this dimeric oxocluster
displays a higher reactivity (TOFϾ80 min–1), coupled with
a higher selectivity toward the oxidation of the sulfoxide
(up to 97% sulfone in 3 h).
Table 7. EXAFS data of Zr6 and Zr12 samples after the catalytic
oxidation of methyl p-tolyl sulfide (see Table 2 for acronyms used).
Sample
Zr6_pre
Abs–Bs
N(Bs)
R(Abs–Bs)
σ
Ef [eV]
(R)
[Å]
[Å2]
Zr–O
Zr–O
Zr–Zr
Zr–O
Zr–O
Zr–Zr
Zr–O
Zr–O
Zr–Zr
Zr–O
Zr–O
Zr–Zr
2
6
4
2.11Ϯ0.02
2.27Ϯ0.02
3.52Ϯ0.04
2.13Ϯ0.02
2.27Ϯ0.02
3.55Ϯ0.04
2.13Ϯ0.02
2.28Ϯ0.02
3.53Ϯ0.04
2.13Ϯ0.02
2.27Ϯ0.02
3.55Ϯ0.04
0.004
0.013 5.2 (27.8)
0.012
0.004
0.013 4.9 (25.8)
0.012
0.004
0.013 5.6 (29.9)
0.014
0.004
0.013 4.5 (26.4)
0.014
Zr6_cat
Zr12_pre
Zr12_cat
2.9Ϯ0.3
6.3Ϯ0.6
2.4Ϯ0.5
2
6
4
2.1Ϯ0.2
4.5Ϯ0.5
2.0Ϯ0.4
However, the reduction of the Zr–Zr coordination
number to almost 50% of the precursor value indicates a
distortion in the Zr–Zr shell.
Compared to the previous sample, the structural changes
in Zr12_cat are more significant and affect both the second
Zr–O shell and the Zr–Zr shell, for which a reduction of
the coordination numbers is found.
The overall increase of stability in aqueous acetonitrile
suggests that the activity and stability of the oxoclusters
could be further improved by using water-free peroxo-based
oxidants or biphasic reaction mixtures. The possibility to
embed the oxoclusters within hydrophobic matrixes will
also be explored. Owing to this interesting reactivity, appli-
cations in the field of oxidative desulfurization are under
investigation in our laboratories.
This trend resembles those observed for Zr6_ba, Zr12_ba,
Zr6_ox, and Zr12_ox, for which the modification of the par-
ent structure of the oxoclusters has been explained with an
extended hydrolysis of the precursor compounds. However,
the same cannot be said about Zr6_cat and Zr12_cat, as
their infrared spectra still present many features of the pre-
cursors and the oxidation of methyl p-tolyl sulfide was per-
formed in an organic solvent with a low amount of water.
Nonetheless, the precipitation of the analyzed samples from
the reaction mixture should be explained by the rearrange-
ment of the parent oxoclusters into oligomeric species with
a lower solubility and, for Zr12, a lower catalytic perform-
ance than the precursor.
Experimental Section
[Zr6(OH)4O4{O(O)CC(CH3)=CH2}12] (Zr6): Zr6 was synthesized
according to standard literature procedures.[1g,1h] FTIR (KBr): ν =
˜
To further assess the changes in the structures of the
3253 (m br), 3099 [w, ν(C–H)], 3018 (w), 2991 (sh), 2975 [m,
oxoclusters under oxidative conditions, the samples before νas(CH3)], 2960 [m, νas(CH3)], 2927 [m νs(CH3)], 2858 (w), 1701
[m, ν(O=COH)], 1645 [m, ν(C=C)], 1574 [s, νas(COO–)], 1549 [s,
and after the catalytic process were analyzed by SEM to
provide information on their morphologies. The collected
micrographs are shown in Figure S14. From a comparison
of the images, it can be appreciated that the morphology of
νas(COO–)], 1498 (w), 1458 [s, νs(COO–)], 1423 [s, δ(CH3)], 1400
(sh), 1371 [m, δ(CH3)], 1288 (w), 1246 [s, δ(CH3)], 1201 (m), 1174
(m), 1095 (w), 1007 [m δ(CH3)], 937 [m, δoop(C=CH2)], 883 (w),
850 (w), 825 [m, δ(CH3)], 796 (w), 665 (m br), 636 (m), 617 [s,
Zr6 is not dramatically affected by the catalysis process, but
δ(OCO)], 565 (w), 503 (m), 467 (m), 424 (m) cm–1. Raman: ν =
˜
the morphology of Zr12 is slightly modified after catalysis;
the facets of the material are less sharp and more rounded
as the catalysis process could result in a slight “etching” of
the material. These data confirm the enhanced stability of
Zr6 with respect to the dimeric Zr12.
3267 [w, ν(O–H)], 3104 [m, ν(C–H)], 3041(w), 3016 (m), 2995 [s,
νas(CH3)], 2978 (sh), 2958 (m), 2925 [s, νs(CH3)], 2874 (sh), 2728
(w), 1695 [m, ν(O=COH)], 1645 [s, ν(C=C)], 1579 (w), 1494 (sh),
1460 (sh), 1434 (s), 1420 (s), 1399 (s), 1373 (m), 1300 (w), 1242 (w),
1116 (w), 1044 (w), 1007 (m), 938 (m), 884 (m), 853 (s), 826 (w),
795 (m), 659 (w), 616 (w), 585 (m), 542 (w), 508 (w), 426 (w), 395
(w), 380 (w), 361 (w), 250 (w, Zr–O–Zr), 236 (w), 189 (m, Zr–O–
Zr) cm–1.
Conclusions
[Zr6O4(OH)4{O(O)CCH2CH=CH2}12]2·6CH2=CHCH2C(O)OH
(Zr12):[1i] In a previously dried Schlenk tube, vinylacetic acid
(3.71 mL, 40.94 mmol) was added dropwise under an Ar atmo-
The hydrolytic stability and the catalytic properties of
Zr6 and Zr12 oxoclusters were thoroughly investigated by a
multiple-technique approach involving FTIR, Raman, and sphere to a 70% (w/w) solution of Zr(nPrO)4 (2.53 g, 5.41 mmol)
Eur. J. Inorg. Chem. 0000, 0–0
13
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
in propanol to achieve a 1:7.4 zirconium/acid molar ratio. After
F(methyl p-tolyl sulfide) = 0.58, F(methyl p-tolyl sulfoxide) = 0.48,
mixing at room temperature, the solution was allowed to stand at F(methyl p-tolyl sulfone) = 0.57; F = (AS/MS)/(AX/MX).
room temperature for 3 d under an inert atmosphere; this resulted
The second-order kinetics constants were determined by fitting the
in the formation of cubic colorless crystals, which were decanted
experimental values with the Scientist Micromath® software for the
and dried under low vacuum for 2 h (yield: 96%). FTIR (KBr): ν
˜
kinetics scheme (Scheme S1). A minor contribution of H2O2 de-
= 3398 [m, ν(O–H)], 3255 [m br, ν(O–H)], 3080 [m, ν(C–H)], 3022
[w, ν(C–H)], 2983 [w, νas(CH2)], 2964 [w, νs(CH2)], 2908 (sh), 1714
[s, ν(O=COH)], 1643 [m, ν(C=C)], 1605 [s, νas(COO–)], 1554 [s,
νas(COO–)], 1489 (w), 1439 [s, νs(COO–)], 1419 [s, δ(CH2)], 1396 [s,
δ(CH2)], 1313 (w), 1296 (w), 1265 [s, δ(CH2)], 1200 (m), 1099 [br
m, δ(CH2)], 1016 (w), 995 (m), 914 [s, δoop(C=CH2)], 812 (m br),
composition (kՅ 4ϫ10–5 –1 s–1) was also considered.
m
The stabilities of the oxoclusters were assessed by comparison of
the FTIR and EXAFS spectra before and after the reaction; the
recovered precipitate was washed with isopropanol. To obtain
enough oxoclusters to analyze after the oxidation process, the reac-
tions were performed with the Zr6 and Zr12 molar ratios increased
to 1% with respect to sulfide.
729 (m), 648 (s), 507 (m), 469 (m), 418 (m) cm–1. Raman: ν = 3395
˜
[w, ν(O–H)], 3271 [w, ν(O–H)], 3084 [m, ν(C–H)], 3021 [s,
νas(CH2)], 2984 [s, νas(CH2)], 2912 [s br, νs(CH2)], 1716 [w,
ν(O=COH)], 1642 [s, ν(C=C)], 1602 (w), 1584 (w), 1558 (w), 1512
(w), 1456 (w), 1420 (m), 1399 (m), 1296 (s), 1273 (w), 1291 (w),
1125 (w), 996 (w), 975 (m), 936 (sh), 913 (m), 890 (m), 765 (w br),
728 (w), 704 (w), 626 (w), 589 (w), 517 (m), 474 (w), 457 (w), 414
(w), 398 (w), 376 (w), 351 (w), 322 (w), 298 (w), 242 (m), 226 (m),
203 (m), 169 (m) cm–1.
EXAFS Measurements: The EXAFS and X-ray absorption near
edge structure (XANES) measurements were performed at
beamline C1 at HASYLAB (Hamburg). A Si(311) double-crystal
monochromator was used for measurements at the ZrK-edge
(17.998 keV). The second monochromator crystal was tilted for op-
timal harmonic rejection. The spectra were recorded in trans-
mission mode by using ionization chambers and in fluorescence
mode by using a passivated implanted planar silicon (PIPS) detec-
tor. For the Zr6_cat and Zr12_cat samples (see main text), only the
fluorescence data were useful owing to the low Zr concentration.
Energy calibration was performed with Zr metal foil; however, a
drift in the spectra could occur as it was not possible to measure
the Zr foil in parallel to the sample measurements. Solid samples
were pressed into self-supporting wafers with cellulose as a binder.
The solutions were measured in a sealed sample cell that can be
filled under an inert atmosphere.[19,20]
Stability Tests: All aqueous solutions were prepared with MilliQ
water at pH 7. The acidic solutions (pH 2) were prepared from a
37% (w/w) solution of HCl, and the basic solutions (pH 13) were
prepared with solid NaOH. Hydrolysis tests in an oxidative envi-
ronment were performed with a 30% (w/w) aqueous solution of
H2O2. The samples were prepared by grinding Zr6 or Zr12 crystals.
The cluster compounds (100 mg of each) were added to the respec-
tive solutions (5 mL). All of the treatments lasted 24 d, after which
the precipitates were collected by filtration through a G4 Gooch
filter, carefully washed with deionized water, and finally dried. The
hydrolysis effect of the atmospheric moisture was tested by expos-
ing the ground oxoclusters (100 mg) to atmospheric air for 24 d.
Data evaluation started with the removal of background absorp-
tion from the experimental absorption spectrum by subtracting a
Victoreen-type polynomial. As there were several turning points in
the absorption edge, the threshold energy E0 was determined by
taking the energy at half of the edge jump.[20] To determine the
smooth part of the spectrum, corrected for pre-edge absorption, a
piecewise polynomial was used. It was adjusted in such a way that
the low-R components of the resulting Fourier transform were
minimal (Figure S3). After division of the background-subtracted
spectrum by its smooth part, the photon energy was converted to
photoelectron wavenumbers k. The resulting χ(k) function was
weighted with k3 and a Fourier transform was applied with a
Hanning window function. Data analysis was performed in k-space
on unfiltered data. The adjustment of the common theoretical
EXAFS expression
Catalytic Tests: The oxidation reactions were run at a fixed tem-
perature (30 and 50 °C) in a closed vial. In a typical oxidation
reaction, Zr6 or Zr12 (molar percent with respect to sulfide 0.08–
0.32%) was added to acetonitrile (1.2 mL). Then, methyl p-tolyl
sulfide (1 mmol) and H2O2 (2.5 mmol, from a 30% aqueous solu-
tion) were added under magnetic stirring. Samples (50 μL) were
withdrawn at fixed time intervals, diluted in a 5 mm solution
(500 μL) of n-undecane (standard) in acetonitrile, and treated with
triphenylphosphine [as a solid or supported on poly(styrene/divin-
ylbenzene) (PS/DVB) resin] as an oxidant quencher. The occur-
rence of unproductive decomposition of the oxidant was evaluated
in a reaction with 2 equiv. of H2O2 (T = 50 °C). A blank reaction
was performed in the absence of Zr oxoclusters at 50 °C, and it
showed a quantitative substrate conversion after 24 h with negligi-
ble sulfone production.
Nj
kri2
–2k2σj2
2
j
χ(k) =
S0 (k)Fj(k)e
e
–2r /λsin[2krj + δj(k)]
͚
i
was made according to the curved-wave formalism of the
EXCURV98 program with XALPHA phase and amplitude func-
tions.[21] The mean free path of the scattered electrons was calcu-
lated from the imaginary part of the potential (VPI, set to
–4.00 eV). An inner potential correction Ef was introduced when
the experimental data was fitted with theoretical models that ac-
count for an overall phase shift between the experimental and cal-
culated spectra.
In the “recharge” reaction, methyl p-tolyl sulfide (1 mmol) was
added, followed by H2O2 (2 mmol).
GC analyses were performed with a Shimadzu GC2010 instrument
equipped with an ionization flame detector and a Equity-5
(15 mϫ0.1 mm) capillary column of poly(5% diphenyl/95% di-
methylsiloxane) with 0.1 μm film thickness. Conditions: Tinj
=
270 °C, Tdet = 280 °C, carrier gas: He, Tinitial = 90 °C for 1 min,
rate = 90 °C/min; Tfinal = 260 °C for 5 min. Retention times: n-
undecane: 1.9 min, methyl p-tolyl sulfide: 2.28 min, methyl p-tolyl
Raman Measurements: The Raman spectra were collected with a
homemade micro-Raman system with a single 320 mm focal length
sulfoxide: 2.83 min, methyl p-tolyl sulfone: 2.96 min. Calibration imaging spectrograph (Triax-320 Jobin Yvon) equipped with a
was performed by integration of the chromatographic peak areas
(A) of the standard (S) and of the analytes (X). The concentrations
(M) of methyl p-tolyl sulfide and its oxidation products were calcu-
lated by using the following response factors: F(undecane) = 1,
holographic 1800 g/mm grating and a liquid-nitrogen-cooled CCD
detector. The excitation source was a Spectra Physics Ar+ ion laser
(Stabilite 2017) operating at 514.5 nm, and an appropriate long-
pass edge filter (Semrock Filters) was used to reduce the stray-light
Eur. J. Inorg. Chem. 0000, 0–0
14
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
[4]
[5]
M. A. Sliem, D. A. Schmidt, A. Bétard, S. B. Kalidindi, S.
Gross, M. Havenith-Newen, A. Devi, R. A. Fischer, Chem.
Mater. 2012, 24, 4274–4282.
a) V. Guillerm, S. Gross, C. Serre, T. Devic, M. Bauer, G. Férey,
Chem. Commun. 2010, 46, 767–769; b) V. Guillerm, F. Ragon,
M. Dan-Hardi, T. Devic, M. Vishnuvarthan, B. Campo, A. Vi-
mont, G. Clet, Q. Yang, G. Maurin, G. Férey, A. Vittadini, S.
Gross, C.Serre, Angew. Chem. Int. Ed. 2012, 51, 9267–9271;
Angew. Chem. 2012, 124, 9401–9405.
a) I. Gautier-Luneau, A. Mosset, J. Z. Galy, Z. Kristallogr.
1987, 180, 83–95; b) S. Doeuff, Y. Dromzee, F. Taulelle, C.
Sanchez, Inorg. Chem. 1989, 28, 4439–4445.
a) L. G. Hubert-Pfalzgraf, G. Liliane, J. Mater. Chem. 2004,
14, 3113–3123; b) L. G. Hubert-Pfalzgraf, Coord. Chem. Rev.
1998, 178–180, 967–997; c) L. G. Hubert-Pfalzgraf, New J.
Chem. 1995, 19, 727–750.
a) A. Brethon, L. G. Hubert-Pfalzgraf, J. Sol-Gel Sci. Technol.
2006, 39, 159–167; b) A. Brethon, L. G. Hubert-Pfalzgraf, J. C.
Daran, Dalton Trans. 2006, 250–257; c) d) L. G. Hubert-Pfalz-
graf, S. Daniele, C. R. Chim. 2004, 7, 521–527; e) S. Daniele,
L. G. Hubert-Pfalzgraf, P. B. Hitchcock, M. F. Lappert, Inorg.
Chem. Commun. 2000, 3, 218–220; f) H. Guillon, L. G. Hubert-
Pfalzgraf, J. Vaissermann, Eur. J. Inorg. Chem. 2000, 1243–
1252; g) S. Daniele, L. G. Hubert-Pfalzgraf, J. Vaissermann,
Polyhedron 1998, 17, 4249–4256; h) L. G. Hubert-Pfalzgraf, C.
Sirio, C. Bois, Polyhedron 1998, 17, 821–830; i) S. Parola, R.
Papiernik, L. G. Hubert-Pfalzgraf, C. Bois, J. Chem. Soc., Dal-
ton Trans. 1998, 5, 737–739; j) S. Boulmaaz, L. G. Hubert-
Pfalzgraf, S. Halut, J. C. Daran, J. Chem. Soc., Chem. Com-
mun. 1994, 601–602; k) L. G. Hubert-Pfalzgraf, S. Daniele, A.
Bennaceur, J. C. Daran, J. Vaissermann, Polyhedron 1997, 16,
1223–1234; l) S. Daniele, L. G. Hubert-Pfalzgraf, J. C. Daran,
Polyhedron 1996, 15, 1063–70; m) S. Daniele, L. G. Hubert-
Pfalzgraf, J. C. Daran, S. Halut, Polyhedron 1994, 13, 927–32.
a) G. Losada, M. A. Mendiola, M. T. Sevilla, Inorg. Chim.
Acta 1997, 255, 125–131; b) K. L. Zhang, Y. J. Shi, X. Z. You,
K. B. Yu, J. Mol. Struct. 2005, 743, 73–77; c) D. L. Long, P.
Kögerler, L. J. Farrugia, L. Cronin, Chem. Asian J. 2006, 1,
352–357.
level. An optical microscope (Olympus BX 40), equipped with three
objectives (20ϫ/0.35, 50ϫ/0.75, and 100ϫ/0.90) was optically cou-
pled to the spectrograph and used to collect the Raman spectra in
backscattering micro configuration. In the present work, the Ra-
man spectra were recorded between 100 and 4000 cm–1 with an
instrumental resolution of ca. 2 cm–1 by using the long working
distance 20ϫ/0.35 objective lens. The irradiation power on the sam-
ple was typically 40–80 mW, and no damage to the solid samples
was observed.
[6]
[7]
SEM Measurements: Scanning electron microscopy measurements
were performed with a field-emission SEM (Zeiss SUPRA 40VP)
equipped with an energy-dispersive X-ray spectroscopy (EDXS)
system (Oxford INCA).
FTIR Measurements: The infrared spectra were collected with a
Thermo Quest Nicolet 5700 instrument. The oxocluster or spent
catalyst (a few milligrams) was dispersed into KBr pellets.
[8]
Elemental Analyses: Inductively coupled plasma mass spectrometry
(ICP-MS) was performed with an Agilent Technologies 7700x ICP-
MS System to establish the Zr content of the oxoclusters. The sam-
ples were mineralized upon heating at 220 °C by using a mixture
of H2SO4, HNO3, and HF in ratio of 4:8:2. CHNS analyses were
performed with a Fison EA1108 CHNS analyzer.
Supporting Information (see footnote on the first page of this arti-
cle): Structures of Zr6 and Zr12; FTIR, Raman, and EXAFS spec-
tra; kinetic traces; SEM micrographs.
Acknowledgments
[9]
The Italian Foreign Affairs Ministry (Ministero degli Affari Esteri,
MAE, Italy) and the National Research Council (CNR, Italy) are
gratefully acknowledged by S. G. for the financial support of this
work in the framework of the “Progetto Grande Rilevanza Italia–
Slovenia 2013”. The University of Padova and the Italian Consor-
tium INSTM are acknowledged for providing money and equip-
ment. The authors would like to thank the Deutsches Elektronen
Synchrothron (DESY) and the European Union Calipso Pro-
gramme for financial support in activities at Hasylab, DESY. The
research leading to these results has received funding from the
European Union Seventh Framework Program (FP7/2007-2013)
under grant agreement number 226716. M. B. acknowledges finan-
cial support by the Carl-Zeiss Foundation. A. S and M. C. acknow-
ledge the financial support from the Italian Fondazione Cariparo
(Progetto Dottorati di Ricerca, PARO118894/11).
[10]
[11]
a) R. H. Holm, Chem. Rev. 1987, 87, 1401–1449; b) M. H.
Dickman, M. T. Pope, Chem. Rev. 1994, 94, 569–584; c) I. A.
Weinstock, Chem. Rev. 1998, 98, 113–170.
a) V. Conte, F. Di Furia, G. Licini, Appl. Catal. A 1997, 157,
335–361; b) B. S. Lane, K. Burgess, Chem. Rev. 2003, 103,
2457–2473; c) R. Noyori, M. Aokib, K. Sato, Chem. Commun.
2003, 1977–1986; d) N. V. Maksimchuk, M. S. Melgunov, J.
Mrowiec-Białon´, A. B. Jarze˛bski, O. A. Kholdeeva, J. Catal.
2005, 235, 175–183; e) N. S. Antonova, J. J. Carbó, U. Kortz,
O. A. Kholdeeva, J. M. Poblet, J. Am. Chem. Soc. 2010, 132,
7488–7497; f) I. W. C. E. Arends, V. Conte, G. Licini, in: Innov-
ative Catalysis in Organic Synthesis: Oxidation, Hydrogenation,
and C–X Bond Forming Reactions (Ed.: P. G. Andersson),
Wiley-VCH, Weinheim, Germany, 2012, p. 77–102.
a) M. Bonchio, G. Licini, S. Mantovani, G. Modena, W. A.
Nugent, J. Org. Chem. 1999, 64, 1326–1330; b) B. Saito, T.
Katsuki, Tetrahedron Lett. 2001, 42, 3873–3876; c) K. P. Bryl-
iakov, E. P. Talsi, J. Mol. Catal. A 2007, 264, 280–287; d) H.
Srour, P. Le Maux, S. Chevance, G. Simonneaux, Coord. Chem.
Rev. 2013, 257, 3030–3050.
a) R. Schwartz, H. Z. Giese, Z. Anorg. Allg. Chem. 1928, 176,
209–232; b) J. A. Connor, E. A. V. Ebsworth, Adv. Inorg. Chem.
Radiochem. 1964, 6, 279–381; c) W. P. Griffith, T. D. Wickins,
J. Chem. Soc. A 1968, 176, 397–400; d) G. D. Gupta, G. V. Jere,
Indian J. Chem. 1968, 6, 54–59; e) G. V. Jere, G. D. Gupta, J.
Inorg. Nucl. Chem. 1970, 32, 537–542; f) G. D. Gupta, G. V.
Jere, Indian J. Chem. 1972, 10, 102; g) G. V. Jere, M. T. San-
thamma, Inorg. Chim. Acta 1977, 24, 57–61; h) R. Schmidt, G.
Pausewang, W. Massa, Z. Anorg. Allg. Chem. 1986, 535, 135–
142; i) M. T. H. Tarafder, M. A. L. Miah, Inorg. Chem. 1986,
25, 2265–2268; j) A. C. Dengel, W. P. Griffith, Polyhedron 1989,
[1] a) U. Schubert, Macromol. Symp. 2008, 267, 1–8 and references
cited therein; b) U. Schubert, Acc. Chem. Res. 2007, 40, 730–
737 and references cited therein; c) U. Schubert, in: Macromol-
ecules Containing Metal and Metal-like Elements, vol. 7 (Eds.:
A. Abd-El Aziz, C. Carraher, C. Pittman, M. Zeldin), Wiley,
New York, 2006, p. 55; d) U. Schubert, J. Mater. Chem. 2005,
15, 3701–3715, and references cited therein; e) U. Schubert, J.
Sol-Gel Sci. Technol. 2004, 31, 19–24; f) U. Schubert, Chem.
Mater. 2001, 13, 3487–3494; g) G. Kickelbick, U. Schubert,
Chem. Ber./Recueil 1997, 130, 473–477; h) G. Kickelbick, P.
Wiede, U. Schubert, Inorg. Chim. Acta 1999, 284, 1–7; i) M.
Puchberger, F. R. Kogler, M. Jupa, S. Gross, H. Fric, G. Kick-
elbick, U. Schubert, Eur. J. Inorg. Chem. 2006, 3283–3293.
[2] a) U. Schubert, Chem. Soc. Rev. 2011, 40, 575–582, and refer-
ences cited therein; b) M. Carraro, S. Gross, Materials 2014, 7,
3956–3989, and references cited therein.
[12]
[13]
[3] S. Gross, J. Mater. Chem. 2011, 21, 15853–15861, and refer-
ences cited therein.
Eur. J. Inorg. Chem. 0000, 0–0
15
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
8, 1371–1377; k) M. T. H. Tarafder, A. A. M. A. Islam, Polyhe-
dron 1989, 8, 109–112; l) C. R. Bhattacharjee, M. Bhattach-
arjee, M. K. Chaudhuri, M. Choudhury, Polyhedron 1990, 9,
1653–1657; m) D. D. Agarwal, R. Jain, R. P. Bhatnagar, S. Sriv-
astava, Polyhedron 1990, 9, 1405–1409; n) M. T. H. Tarafder,
A. R. Khan, Polyhedron 1991, 10, 819–822; o) M. T. H. Tar-
afder, A. R. Khan, Polyhedron 1991, 10, 973–976; p) T. Okachi,
N. Murai, M. Onaka, Org. Lett. 2003, 5, 85–87; q) G. Wang,
H. H. Y. Sung, I. D. Williams, W. Leung, Inorg. Chem. 2012,
51, 3640–3647.
a) O. A. Kholdeeva, G. M. Maksimov, R. I. Maksimovskaya,
M. P. Vanina, T. A. Trubitsina, D. Yu. Naumov, B. A. Kolesov,
N. S. Antonova, J. J. Carbó, J. M. Poblet, Inorg. Chem. 2006,
45, 7224–7234; b) O. A. Kholdeeva, R. I. Maksimovskaya, J.
Mol. Catal. A 2007, 262, 7–24; c) S. S. Mal, N. H. Nsouli, M.
Carraro, A. Sartorel, G. Scorrano, H. Oelrich, L. Walder, M.
Bonchio, U. Kortz, Inorg. Chem. 2010, 49, 7–9; d) C. Jahier,
S. S. Mal, U. Kortz, S. Nlate, Eur. J. Inorg. Chem. 2010, 1559–
1566; e) G. Al-Kadamany, S. S. Mal, B. Milev, B. G. Donoeva,
R. I. Maksimovskaya, O. A. Kholdeeva, U. Kortz, Chem. Eur.
J. 2010, 16, 11797–11800; f) M. Carraro, N. H. Nsouli, H. Oel-
rich, A. Sartorel, A. Sorarù, S. S. Mal, G. Scorrano, L. Walder,
U. Kortz, M. Bonchio, Chem. Eur. J. 2011, 17, 8371–8378; g)
D. Li, H. Han, Y. Wang, X. Wang, Y. Li, E. Wang, Eur. J.
Inorg. Chem. 2013, 1926–1934; h) P. Jiménez-Lozano, J. J.
Carbó, A. Chaumont, J. M. Poblet, A. Rodríguez-Fortea, G.
Wipff, Inorg. Chem. 2014, 53, 778–786.
[27] A. I. Vogel, Vogel’s Textbook Of Macro and SemiMicro Quali-
tative Inorganic Analysis, 5th ed. (revised by G. Svehla), Long-
man, London, 1979.
[28] A. Clearfield, P. A. Vaughan, Acta Crystallogr. 1956, 9, 555–
558.
[29] A. Clearfield, Inorg. Chem. 1964, 3, 146–148.
[30] T. C. W. Mak, Can. J. Chem. 1968, 46, 3491–3497.
[31] M. Aberg, Acta Chem. Scand., Ser. B 1977, 31, 171–181.
[32] G. T. Mamott, P. Barnes, S. E. Tarling, S. L. Jones, C. J. Nor-
man, J. Mater. Sci. 1991, 26, 4054–4061.
[33] G. Y. Guo, Y. L. Chen, W. J. Ying, Mater. Chem. Phys. 2004,
84, 308–314.
[34] a) M. Bauer, C. Gastl, C. Köppl, G. Kickelbick, H. Bertagnolli,
Monatsh. Chem. 2006, 137, 567–581; b) M. Bauer, S. Müller,
G. Kickelbick, H. Bertagnolli, New J. Chem. 2007, 31, 1950–
1959.
[35] K. Nakamoto, Infrared and Raman Spectra of Inorganic & Co-
ordination Compound, 6th ed., Wiley Interscience, New York,
2009.
[36] A. Clearfield, Rev. Pure Appl. Chem. 1964, 14, 91–108.
[37] A. Clearfield, J. Mater. Res. 1990, 5, 161–162.
[38] M. Kantcheva, C. J. Koz, Mater. Sci. 2007, 42, 6074–6086.
[39] M. Bauer, H. Bertagnolli, J. Phys. Chem. B 2007, 111, 13756–
13764.
[40] F. Faccini, H. Fric, U. Schubert, E. Wendel, O. Tsetsgee, K.
Müller, H. Bertagnolli, A. Venzo, S. Gross, J. Mater. Chem.
2007, 17, 3297–3307.
[14]
[41] a) R. A. Sheldon, I. W. C. E. Arends, U. Hanefeld, Green
Chemistry and Catalysis, Wiley-VCH, Weinheim, Germany,
2007; b) K. Kaczorowska, Z. Kolarska, K. Mitka, P. Kowalski,
Tetrahedron 2005, 61, 8315–8327; c) K. Sato, M. Hyodo, M.
Aoki, X.-Q. Zheng, R. Noyori, Tetrahedron 2001, 57, 2469–
2476.
[42] a) K. Yazu, Y. Yamamoto, T. Furuya, K. Miki, K. Ukegawa,
Energy Fuels 2001, 15, 1535–1536; b) J. M. Campos-Martin,
M. C. Capel-Sanchez, J. L. G. Fierro, Green Chem. 2004, 6,
557–562; c) M. Carraro, G. Fiorani, L. Mognon, F. Caneva,
M. Gardan, C. Maccato, M. Bonchio, Chem. Eur. J. 2012, 18,
13195–13202.
[43] a) O. Bortolini, S. Campestrini, V. Conte, F. Di Furia, G. Mod-
ena, J. Org. Chem. 1988, 53, 5721–5724; b) W. Adam, W. Haas,
B. B. Lohray, J. Am. Chem. Soc. 1991, 113, 6202–6208; c) F. B.
Ballistreri, M. Bonchio, V. Conte, M. A. De Conciliis, F.
Di Furia, G. A. Tomaselli, R. M. Toscano, J. Org. Chem. 1995,
60, 4475–4480; d) M. Bonchio, S. Campestrini, V. Conte, F.
Di Furia, S. Moro, Tetrahedron 1995, 51, 12363–12372.
[15]
[16]
[17]
G. Trimmel, S. Gross, G. Kickelbick, U. Schubert, Appl. Or-
ganomet. Chem. 2001, 15, 401–406.
M. P. Feth, PhD Thesis, University of Stuttgart, Germany,
2003.
a) G. Kickelbick, M. P. Feth, H. Bertagnolli, M. Puchberger,
D. Holzinger, S. Gross, J. Chem. Soc., Dalton Trans. 2002,
3892–3898; b) P. Walther, F. Puchberger, F. R. Kogler, K.
Schwarz, U. Schubert, Phys. Chem. Chem. Phys. 2009, 11,
3640–3647.
F. R. Kogler, M. Jupa, M. Puchberger, U. Schubert, J. Mater.
Chem. 2004, 14, 3133–3138.
[18]
[19]
a) M. Bauer, G. Heusel, S. Mangold, H. Bertagnolli, J. Syn-
chrotron Radiat. 2010, 17, 273–279; b) M. Bauer, C. Gastl,
Phys. Chem. Chem. Phys. 2010, 12, 5575–5584.
[20] a) M. Newville, P. Livins, Y. Yacoby, J. J. Rehr, E. A. Stern,
Phys. Rev. B 1993, 47, 14126–14131; b) T. S. Ertel, H. Bertag-
nolli, S. Hückmann, U. Kolb, D. Peter, Appl. Spectrosc. 1992,
46, 690–698.
[21] S. J. Gurman, N. Binsted, I. Ross, J. Phys. C 1984, 17, 143– [44] Interestingly, Zr6 displays a selectivity similar to that reported
152.
for Zr6- and Hf6-containing polyoxotungstates, which were
used to catalyze the oxidation of methyl phenyl sulfide with
H2O2 (see ref.[14e]).
[22] C. Carlone, Phys. Rev. B 1992, 45, 2079–2082.
[23] V. G. Keramidas, W. B. White, J. Am. Ceram. Soc. 1974, 57,
22–24.
[24] a) A. F. Holleman, E. Wiberg, Lehrbuch der Anorganische
Chemie, 101st ed., De Gruyter, Berlin, 1995, p. 1411; b) F. A.
Cotton, G. Wilkinson, Advanced Inorganic Chemistry, 3rd ed.,
Interscience, New York, 1972, p. 930.
[25] N. N. Greenwood, A. Earnshaw, Chemistry of the Elements,
Butterworth-Heinemann Ltd., Oxford, UK, 1984.
[26] R. Villanneau, H. Carabineiro, X. Carrier, R. Thouvenot, P.
Herson, F. Lemos, F. Ramôa Ribeiro, M. Che, J. Phys. Chem.
B 2004, 108, 12465–12471.
[45] M. Carraro, L. Sandei, A. Sartorel, G. Scorrano, M. Bonchio,
Org. Lett. 2006, 8, 3671–3674.
[46] G. Licini, C. Zonta, Angew. Chem. Int. Ed. 2013, 52, 2911–
2914; Angew. Chem. 2013, 125, 2983.
[47] The experimental data agree with the following formulas for
the starting materials: [Zr6(OH)4O4{O(O)CC(CH3)CH2}12]·
8[CH2C(CH3)C(O)OH]8
CHCH2}12}2]·6[CH2CHCH2C(O)OH]6.
Received: August 8, 2014
and
[{Zr6O4(OH)4{O(O)CCH2-
Published Online: ᭿
Eur. J. Inorg. Chem. 0000, 0–0
16
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Job/Unit: I42767
/KAP1
Date: 05-11-14 16:23:32
Pages: 17
www.eurjic.org
FULL PAPER
Cluster Compounds
F. Faccioli, M. Bauer, D. Pedron,
A. Sorarù, M. Carraro,*
S. Gross* ........................................ 1–17
The hydrolytic stability and reactivity of
Zr6 and Zr12 oxoclusters are investigated by
FTIR, Raman, and extended X-ray ab-
sorption fine structure (EXAFS) spec-
troscopy. The clusters are stable in water
at neutral pH, and peroxo complexes are
detected in the presence of hydrogen per-
oxide. A kinetic study highlights the cata-
lytic potential of the clusters for the oxi-
dation of methyl p-tolyl sulfide.
Hydrolytic Stability and Hydrogen Per-
oxide Activation of Zirconium-Based Oxo-
clusters
Keywords: Cluster compounds
conium / Peroxides / Oxidation / EXAFS
spectroscopy
/
Zir-
Eur. J. Inorg. Chem. 0000, 0–0
17
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim