A Linear Trinuclear Oxidodiperoxido-molybdenum(VI) Complex with Single Triazole Bridges: Catalytic Activity in Epoxidation, Alcoholysis, and Acetalization Reactions

Article abstract of DOI:10.1002/cctc.201800232

The complex (Htrz)₂[Mo₃O₆(O₂)₄(trz)₂]?H₂O (1), isolated from the reaction of MoO₃ with H₂O₂ in the presence of 1,2,4-triazole (trz), displays singular structural features such as double bridges comprising one oxido and one triazole ligand, and peripheral oxidodiperoxido-molybdenum(VI) units. A comparative study was performed regarding the catalytic activity and stability of complex 1 and the hybrid material [MoO₃(trz)_0.5] (2) for the epoxidation of cis-cyclooctene (Cy8), the alcoholysis of epoxides, and the acetalization of benzaldehyde (PhCHO). Methanol or ethanol were used as solvents and reagents for the latter two reactions. For the substrates Cy8, styrene oxide, and PhCHO, the corresponding epoxide, β-alkoxy alcohol, and dialkyl acetal were obtained with 100 % selectivity at very high conversions. In general, 1 performed better than 2 for the different catalytic reactions. Under certain oxidant/co-solvent conditions used for Cy8 epoxidation, 1 was converted to 2 during the reaction.

Full text of DOI:10.1002/cctc.201800232

DOI: 10.1002/cctc.201800232
Full Papers
A Linear Trinuclear Oxidodiperoxido-molybdenum(VI)
Complex with Single Triazole Bridges: Catalytic Activity in
Epoxidation, Alcoholysis, and Acetalization Reactions
Margarida M. Antunes, Tatiana R. Amarante, Anabela A. Valente,* Filipe A. Almeida Paz,
Isabel S. GonÅalves, and Martyn Pillinger*[a]
The complex (Htrz)2[Mo3O6(O2)4(trz)2]·H2O (1), isolated from the
reaction of MoO3 with H2O2 in the presence of 1,2,4-triazole
(trz), displays singular structural features such as double
bridges comprising one oxido and one triazole ligand, and pe-
ripheral oxidodiperoxido-molybdenum(VI) units. A comparative
study was performed regarding the catalytic activity and stabil-
ity of complex 1 and the hybrid material [MoO3(trz)0.5] (2) for
the epoxidation of cis-cyclooctene (Cy8), the alcoholysis of ep-
oxides, and the acetalization of benzaldehyde (PhCHO). Metha-
nol or ethanol were used as solvents and reagents for the
latter two reactions. For the substrates Cy8, styrene oxide, and
PhCHO, the corresponding epoxide, b-alkoxy alcohol, and dia-
lkyl acetal were obtained with 100% selectivity at very high
conversions. In general, 1 performed better than 2 for the dif-
ferent catalytic reactions. Under certain oxidant/co-solvent con-
ditions used for Cy8 epoxidation, 1 was converted to 2 during
the reaction.
Introduction
Azoles are five-membered ring heterocycles containing one ni-
trogen atom and at least one other non-carbon heteroatom.[1,2]
Nitrogen-rich azoles such as imidazole,[2] pyrazole,[3] triazole,[4,5]
and tetrazole[5] display an elaborate coordination chemistry
and are found in a plethora of metal complexes and polymeric
coordination networks. Triazoles (1,2,3-triazole and 1,2,4-tria-
zole) are especially interesting as they combine in one mole-
cule the coordinative abilities of both pyrazoles and imida-
zoles.[4] The upshot of this is that mononuclear complexes con-
taining these ligands are rare. Indeed, for the parent 1,2,4-tria-
zole (abbreviated without prefix locant numbers as trz), dinu-
clear, oligonuclear, and polymeric structures are usually formed
by use of the bridging N1/N2, N2/N4 and (for deprotonated
trz) N1/N2/N4 coordination modes.[1,4,5] The structural diversity
is further amplified by use of derivatives with coordinating
substituents that can function as polytopic ligands. In a review
of crystal structures published before the end of 2008 for dis-
crete metal complexes bearing 1,2,4-triazole derivatives,
Gamez and co-workers found that the predominant nuclearity
was dinuclear (40% of compounds), followed by trinuclear
(19% of compounds).[5] This trend was attributed to the preva-
lence of the bis(monodentate) 1,2-m:h1h1 bridging mode. Most
complexes display either double [M-(N-N)2-M] or triple [M-(N-
N)3-M] bridges, whereas only a few complexes with single [M-
(N-N)-M] bridges are known.
Triazoles are effective ligands for bridging MoVI centers that
are also linked by m-O2À groups. Remarkably, a search in the lit-
erature and in the Cambridge Structural Database (CSD version
5.38;[6] 2017+3 updates) reveals that only polymeric structures
have been reported for compounds based on either simple or
derivatized 1,2,4-triazole units, which effectively establish N,N-
bridging coordination modes with neighboring metal centers.
This occurs not only for the parent 1,2,4-triazole molecule[7]
but also for other (typically long) derivatives, which include bi-
topic molecules combining two 1,2,4-triazol-4-yl sites with
extra functional (spacing) units.[8] The use of the parent 1,2,4-
triazole molecule leads to
a rare perovskite-like hybrid
[MoO3(trz)0.5] consisting of layers of corner-sharing MoO5N oc-
tahedra with the organic subunits projecting into the interlam-
ellar regions.[7] In contrast, 4-substituted triazole molecules ex-
hibit a tendency to support one-dimensional (1D) molybde-
num(VI) oxide topological motifs (e.g., chains, hydrated chains,
ribbons, and helixes), which can be transformed into 2D or 3D
frameworks by the use of 1,2,4-bitriazole ligands.[8a,e,f] We have
been studying these triazolylmolybdenum(VI) oxide hybrids as
(pre)catalysts for the oxidation of olefins (epoxidation), sulfides,
alcohols, and aldehydes.[7b,8e,f] Layered [MoO3(trz)0.5] displayed a
truly unique reaction-induced self-separating (RISS) catalytic
behavior when used with aqueous H2O2 as the oxidant.[7b]
Under RISS conditions the hybrid reacts with the oxidant to
form a soluble active species. Exhaustion of the oxidant leads
to spontaneous reassembly of the 2D structure of [MoO3(trz)0.5]
[a] Dr. M. M. Antunes, Dr. T. R. Amarante, Dr. A. A. Valente,
Dr. F. A. Almeida Paz, Dr. I. S. GonÅalves, Dr. M. Pillinger
Department of Chemistry, CICECO—Aveiro Institute of Materials
University of Aveiro
Campus Universitꢀrio de Santiago, 3810-193 Aveiro (Portugal)
E-mail: atav@ua.pt
mpillinger@ua.pt
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under https://doi.org/10.1002/
cctc.201800232
This manuscript is part of a Special Issue on the “Portuguese Conference
on Catalysis” based on the International Symposium on Synthesis and
Catalysis (ISySyCat).
ChemCatChem 2018, 10, 1 – 11
1
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
and the compound precipitates, allowing straightforward re-
covery and reuse. To the best of our knowledge, this was the
first report of a self-precipitating catalyst with an extended 1D,
2D, or 3D crystalline structure.
a significant excess of hydrogen peroxide, which, upon degra-
dation into water at high temperature, can in situ liberate the
required proton captured by the organic base.
The asymmetric unit of 1 is composed of two metal centers
with the trinuclear nature of the complex being ultimately
formed by crystallographic symmetry (Figure 1): the complex
itself is assembled by a C2 crystallographic proper rotation axis
bisecting the unit and passing through Mo2 as depicted in
Figure 1. The coordination environments of the two metal cen-
ters are strikingly distinct. The peripheral Mo1 metal center is
bound to two peroxido ligands, one terminal and one m2-
bridging oxido group, and one N,N-bridging 1,2,4-triazole, in a
simplified (based on the centers of gravity of the peroxido li-
gands) penta {MoNO6} coordination environment strongly re-
sembling a distorted trigonal bipyramid as depicted in Figure 1
(the terminal oxido and the organic linker are positioned at
the apexes of the pyramid). The heterogeneity of coordinating
units is reflected in the spread of the Mo1À(N,O) bond lengths,
which range from 1.691(3) ꢁ (for the terminal oxido ligand) to
2.351(3) ꢁ (for the Mo1ÀN connection to the organic ligand;
see Table S2 in the Supporting Information for additional geo-
metrical details). The central Mo2 atom exhibits a more tradi-
tional distorted octahedral {MoN2O4} coordination environ-
ment, in which the equatorial plane is composed of two N,N-
bridging 1,2,4-triazole ligands plus two terminal oxido units,
whereas the apical positions are occupied by the two symme-
try-related m2-bridging oxido ligands. The Mo2À(N,O) bond
lengths range from 1.724(3) ꢁ (for the terminal oxido units) to
2.348(3) ꢁ (for the Mo2ÀN bonds to the 1,2,4-triazole ligands).
For both metal centers the trans effect of the terminal oxido li-
gands is notably present, which is ultimately reflected in the
longer connections to the organic linkers.
To better understand the behavior of [MoO3(trz)0.5] as a RISS
catalyst, it is necessary to ascertain the nature of active species
formed in solution. To this end, we have studied the reaction
of MoO3 with H2O2 in the presence of 1,2,4-triazole, and herein
describe the isolation of the trinuclear oxidoperoxido complex
(Htrz)2[Mo3O6(O2)4(trz)2]·H2O (1). The crystallographic characteri-
zation of 1 reveals several standout features, most notably that
it is the first example of a discrete MoVI complex bearing trz li-
gands, and a rare case of a discrete metal–triazole complex
that contains only single [M-(N-N)-M] bridges. Compound 1
was explored for oxidation and acid catalysis, specifically, olefin
epoxidation with hydroperoxides, epoxide alcoholysis, and al-
dehyde acetalization. The hybrid material [MoO3(trz)0.5] (2) was
included in these studies to compare catalytic performances
and stabilities.
Results and Discussion
Catalyst synthesis and characterization
The reaction of MoO3 with excess H2O2 (H2O2/Moꢀ2) gives
yellow solutions containing low-condensed oxidoperoxido spe-
cies (mainly [MoO(O2)2(H2O)2] and [O{MoO(O2)2(H2O)}2]2À).[9] Ad-
dition of one or two equivalents of an organic ligand to such
solutions can result in ligand exchange and isolation of the re-
spective mononuclear or dinuclear complex.[10,11] For example,
with pyrazole (pz), the complexes [MoO(O2)2(H2O)(pz)] and
[MoO(O2)2(pz)2] have been isolated,[10] whereas with 4,4’-bipyri-
dine (4,4’-bipy) the oxido-bridged complex [O{MoO(O2)2(4,4’-
Hbipy)}2] was obtained.[11] Less predictable outcomes have
been observed in other systems owing to the existence of
complex equilibria between different peroxide complexes. For
example, with imidazole (im), the complex (Him)2-
[O{MoO(O2)2(H2O)}2] was isolated rather than [MoO(O2)2(im)2] or
[O{MoO(O2)2(im)}2], and other species such as [MoO2(O2)(im)]
and K2[MoO2(O2)2(im)] were identified.[12] In the present work,
MoO3 was reacted with a large excess of H2O2 (14 equiv) and
the ligand trz (2 equiv) was added to the resultant yellow solu-
tion. The title compound, (Htrz)2[Mo3O6(O2)4(trz)2]·H2O (1), pre-
cipitated over several days as a microcrystalline solid in very
good yield. Single crystals suitable for X-ray diffraction (XRD)
were obtained by slow evaporation at ambient temperature of
a solution of 1 in CH2Cl2. The phase purity of the bulk micro-
crystalline product was confirmed by comparison of its
powder XRD pattern with a simulated one calculated from the
single-crystal structure data (Figure S1 in the Supporting Infor-
mation).
The trinuclear anionic [Mo3O6(O2)4(trz)2]2À complex with C2
symmetry is truly unique and unprecedented. Apart from
being the first example of a discrete MoVI complex bearing trz
ligands, and a rare case of a discrete metal–triazole complex
that contains only single [M-(N-N)-M] bridges, 1 is the first
MoVI-peroxido compound of any kind that contains coordinat-
ed triazole or substituted triazole ligands. Metal bridges are en-
sured simultaneously by both m2-bridging oxido units and by
the N,N-bridging 1,2,4-triazole molecules, imposing
a
Mo1···Mo2 distance of 3.6547(2) ꢁ with a kink angle (of the m2-
O linker) of 138.73(17)8. Owing to the imposed symmetry, the
two organic ligands are located on the same side of the com-
plex with the average planes of the ligands subtending a dihe-
dral angle of approximately 658. These ligands are engaged in
a weak intramolecular CÀH···O supramolecular interaction with
a neighboring peroxido unit [dC···O =3.172(5) ꢁ with <(CHO)=
1488], which further strengthens the robustness of the anionic
trinuclear complex.
The role of the organic molecules in the overall supramolec-
ular network of the crystal structure of 1 is strikingly distinct if
the ligand is coordinated to a metal center or if it is the
charge-balancing 1,2,4-triazolium: the coordinated molecules
are solely engaged in both strong NÀH···O and weak CÀH···O
supramolecular contacts with neighboring anionic complexes
(not shown), thus establishing effective bridges between adja-
Compound 1 is a hybrid salt in which the anionic À2 charge
of the trinuclear anionic [Mo3O6(O2)4(trz)2]2À complex is bal-
anced by protonated 1,2,4-triazole molecules, which co-crystal-
lize with water molecules as revealed by the crystallographic
studies. The protonation of the organic molecule can easily be
rationalized by considering that the reaction mixture contains
&
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
2
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 1. (a) Schematic representation of the trinuclear anionic [Mo3O6(O2)4(trz)2]2À tetrakisperoxido complex present in the hybrid salt
(Htrz)2[Mo3O6(O2)4(trz)2]·H2O, plus two additional views (top and front). Non-hydrogen atoms belonging to the asymmetric unit are represented as thermal el-
lipsoids drawn at the 50% probability level. Polyhedral representation of the distorted (b) {MoNO6} trigonal bipyramidal and (c) {MoN2O4} octahedral coordina-
tion environments of Mo1 and Mo2, respectively. Selected bond lengths and angles are given in Table S2 in the Supporting Information. Symmetry transfor-
mation used to generate equivalent atoms: (i) 1Àx, y, 1/2Àz.
cent complexes [d(C,N)···O in the 2.746(4)–2.970(6) ꢁ range; inter-
action angles between 120 and 1738]; the charge-balancing
1,2,4-triazolium cation establishes instead connections be-
tween all individual units (including the water molecule of
crystallization), with internuclear supramolecular d(C,N)···O distan-
ces ranging from 2.715(6) to 3.205(7) ꢁ (with interaction angles
varying between 115 and 1778). Additionally, the 1,2,4-triazoli-
um cation further interacts with the coordinated ligand
through an offset weak ring-to-ring interaction [inter-centroid
distance of 3.558(3) ꢁ—not shown].
requirement of at least a five- or six-fold coordination sphere.
Hence, we see the formation of the trinuclear anionic
[Mo3O6(O2)4(trz)2]2À complex, in which the organic ligand
bridges adjacent metal centers, with the peroxido ligands
being solely present in the less sterically hindered peripheral
five-coordinated MoVI metal centers.
The FTIR spectra of 1 and the layered hybrid [MoO3(trz)0.5]
(2) display some similarities in the high-frequency region,
above 1000 cmÀ1, where several key ligand (trz) modes occur
(Figure S2 in the Supporting Information).
A band with
The
isolation
of
the
unusual
hybrid
salt
medium intensity at 1519 cmÀ1 for both compounds is as-
signed to a C=N stretching vibration of coordinated (bridging)
1,2,4-triazole units. As shown schematically in Figure 2, the
same type of coordination mode is present in both com-
pounds, involving single trz units that bridge oxido-bridged
(Htrz)2[Mo3O6(O2)4(trz)2]·H2O can be rationalized by considering
the main structural features of the metal centers and of the
ligand itself. Although a handful of MoVI and WVI discrete com-
plexes are known in which pyrazole derivatives are N,N-chelat-
ed to the same metal center, none of these reported com-
plexes contain inorganic peroxido ligands in the coordination
sphere. Examples of discrete complexes of the former kind
have, for example, been reported with 3,5-dimethylpyrazole,[13]
3,5-di-tert-butylpyrazole,[14]
3,5-di-tert-butyl-4-bromopyrazo-
le,[14a,e] 3,5-di-tert-butyl-4-methylpyrazole,[14a] and 3,5-diphenyl-
pyrazole.[15] A search in the CSD (version 5.38;[6] 2017+3 up-
dates) did not reveal any type of peroxido complex in which
the coordination sphere contains a single N,N-chelated pyra-
zole or triazole of any kind. For the particular case of MoVI or
WVI metal centers, this fact can be understood by taking note
of the need to balance the overall charge while fulfilling the
Figure 2. Comparison of the trz bridging coordination mode in com-
pounds 1 and 2. The subscript b indicates oxygen atoms that bridge MoVI
centers.
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
3
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
MoVI centers, and therefore the close proximity of the n(C=N)
frequency is expected. The spectrum of 1 exhibits an addition-
al n(C=N) band with medium intensity at 1566 cmÀ1. which is
attributed to triazolium cations.[16] These moieties give rise to
an intense n(CÀH) absorption at 3135 cmÀ1, whereas the
shoulder on this band at about 3127 cmÀ1 is assigned to the
n(CÀH) mode for coordinated 1,2,4-triazole. Below 1000 cmÀ1
,
the spectrum of 1 can be divided into a number of sub-re-
gions. The assignment of the multiple bands between 800 and
1000 cmÀ1 is not trivial owing to the possibility of n(Mo=O),
n(OÀO), and triazole-centered vibrations in this region. Never-
theless, by reference to published data for compounds that
possess
dioxido
[MoO2(m2-O)2(L)]
or
oxidodiperoxido
[MoO(O2)2(m2-O)(L)] fragments,[11,17] bands found above
900 cmÀ1 (931, 956 cmÀ1) may be attributed to Mo=O stretch-
ing vibrations, whereas the band at 854 cmÀ1 must be due to
n(OÀO). Two intermediate bands may arise from n(Mo=O)
(894 cmÀ1) or n(OÀO) (873 cmÀ1) modes or may instead be due
to triazole-centered vibrations. The triazole and triazolium spe-
cies in 1 are responsible for several ring torsion bands ob-
served in the 600–700 cmÀ1 region, whereas for 2 only one
band is observed at 625 cmÀ1.[7b] Towards lower wavenumbers,
bands at 525 and 571 cmÀ1 for 1 are assigned to n[Mo(O2)2] vi-
brations.
The solid-state 13C{1H} CP MAS NMR spectrum of 1 displays
two peaks of roughly equal intensity at 141.8 and 145.1 ppm,
and a more intense peak at 140.4 ppm (Figure S3 in the Sup-
porting Information). With reference to the crystal structure,
the two downfield-shifted resonances are assigned to the non-
equivalent C3 and C5 carbon atoms of the metal-coordinated
triazole molecule, whereas the more intense signal is assigned
to overlapping resonances for the carbon atoms of the proton-
ated ligand that co-crystallizes with the trinuclear complex. For
comparison, the 13C{1H} CP MAS NMR spectrum of the layered
Figure 3. cis-Cyclooctene epoxidation in the presence of 1 with H2O2/CH3CN
^
(&), TBHPA/CH3CN (ꢂ), TBHPA/TFT (+), TBHPD/CH3CN ( ), and TBHPD/TFT (*),
at 708C.
21% conversion. The results reveal faster kinetics when using
(i) TBHPD instead of TBHPA, and (ii) TFT as co-solvent instead of
CH3CN, suggesting that non-aqueous, less polar solvents are
favorable for 1-TBHP systems. On the other hand, the use of
aqueous H2O2 as oxidant instead of TBHPA (with CH3CN as co-
solvent) led to considerably improved catalytic performance
(92% conversion at 6 h, Figure 3).
Catalytic results with 1 are compared in Table 1 with those
obtained with the hybrid 2 under equivalent reaction condi-
tions. Cy8Ox selectivity was always 100% for both compounds.
With the exception of the combination TBHPD/CH3CN, which
gave much poorer results with 2 than with 1, the catalytic per-
formances for the two compounds are quite similar for a given
oxidant/co-solvent pairing.
hybrid [MoO3(trz)0.5] (2) contains
a single resonance at
145.8 ppm for the triazole C3 and C5 carbon atoms as these
are chemically (and crystallographically) equivalent as shown
schematically in Figure 2.
Table 1. Comparison of catalytic results for cis-cyclooctene epoxidation
in the presence of 1 or 2.[a]
Catalyst
Oxidant
Co-solvent
t [h]
Conv. [%]
Ref.
Catalytic epoxidation
1
2
1
2
1
2
1
2
TBHPD
TBHPD
TBHPD
TBHPD
TBHPA
TBHPA
H2O2
TFT
TFT
6/24
6/24
6/24
24
24
24
94/97
83/97
95/99
35
21
20
this work
this work
this work
[7b]
this work
[7b]
Compound 1 was explored for catalytic epoxidation at 708C
by using cis-cyclooctene (Cy8) as a model substrate, and differ-
ent hydroperoxide oxidant/co-solvent combinations. cis-Cyclo-
octene oxide (Cy8Ox) was always the only product (100% se-
lectivity). On the other hand, the reaction kinetics was strongly
dependent on the conditions used (Figure 3). For the 1/tert-bu-
tylhydroperoxide (TBHP) based systems, the initial conversions
at 30 min increased in the order TBHPA/CH3CN (<1%)<TBHPA/
TFT (2.5%)<TBHPD/CH3CN (39%)<TBHPD/TFT (52%), where
TBHPA indicates aqueous TBHP, TBHPD indicates TBHP in
decane, and TFT=a,a,a-trifluorotoluene (Figure 3). The Cy8 re-
action with TBHPD/TFT in the absence of 1 was very sluggish
(<14% conversion at 24 h), confirming the catalytic role of the
molybdenum species for olefin epoxidation. After 24 h reaction
in the presence of 1, at least 86% conversion was reached
under all conditions except for TBHPA/CH3CN, which led to
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
6/24
6/24
92/99
90/98
this work
[7b]
H2O2
[a] Reactions conditions: 708C, Mo/Cy8/oxidant molar ratio=1:100:153.
The reaction mixtures obtained after 24 h batch runs when
using 1 and different oxidant/co-solvent pairings were always
biphasic solid–liquid. To better understand the catalytic per-
formance, the catalyst stability was studied by characterizing
the recovered solid phases by attenuated total reflection (ATR)
FTIR spectroscopy (Figure 4) and powder XRD (PXRD; Figure 5).
&
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
4
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 4. ATR FTIR spectra of (a) 1, and the corresponding solids recovered
from the reaction of Cy8 at 708C; (b) 1-TBHPD/TFT; (c) 1-TBHPD/TFT-run2;
(d) contact test CT-1-TBHPD/TFT; (e) 1-TBHPD/CH3CN; (f) 1-TBHPA/TFT (3);
(g) 1-TBHPA/CH3CN, (h) 1-H2O2/CH3CN; (i) contact test CT-1-H2O2/CH3CN (3);
and (j) 2-TBHPD/TFT (which is similar to 2[7b]).
Figure 5. PXRD patterns of (a) 1, and the corresponding solids recovered
from the reaction of Cy8 at 708C; (b) 1-TBHPD/TFT; (c) 1-TBHPD/TFT-run2;
(d) contact test CT-1-TBHPD/TFT; (e) 1-TBHPD/CH3CN; (f) 1-TBHPA/TFT (3);
(g) 1-TBHPA/CH3CN; (h) 1-H2O2/CH3CN; (i) contact test CT-1-H2O2/CH3CN (3);
and (j) 2-TBHPD/TFT (which is similar to 2[7b]).
For the systems TBHPA/CH3CN and H2O2/CH3CN, a white micro-
crystalline solid was recovered with IR and PXRD data (lines (g)
and (h) in Figures 4 and 5) that matched those for the hybrid
material [MoO3(trz)0.5] (2) (line (j) in Figures 4 and 5). Hence,
under these conditions, 2 is formed from 1, and this is consis-
tent with the observation of similar catalytic performances for
1 and 2 with these two oxidant/co-solvent combinations,
namely TBHPA/CH3CN and H2O2/CH3CN (Table 1).
absent. Compound 3 is active as a homogeneous catalyst for
Cy8 epoxidation with H2O2/CH3CN, giving 69% conversion at
24 h, but was found to convert to insoluble 2 upon consump-
tion of the oxidant.[7b] Hence, depending on the presence or
absence of H2O2, 2 and 3 can interconvert. Overall, the system
1/H2O2/CH3CN seems to involve conversion of metal species
via the pathway 1 to 3 to 2. The interconversion between 2
and 3 reported previously for the system 2/H2O2/CH3CN did
not, to the best of our knowledge, involve species of the type
1.[7b] Accordingly, 1 is an alternative source of species 3. The
lower Cy8 conversion for 3 in relation to 1 and 2 with H2O2/
CH3CN, may be partly due to the fact that, as the structure of
3 is unknown, it was used in an initial mass amount equal to
that used for 2, but the initial Mo/substrate molar ratio may be
different, influencing the reaction kinetics.
Although 2 was formed from 1 with TBHPA/CH3CN, with
TBHPA/TFT a yellow microcrystalline solid (hereafter denoted as
3) was obtained with PXRD and IR data different from those of
either 1 or 2. This may account for why better catalytic per-
formance was obtained with TBHPA/TFT (86% conversion at
24 h) than with TBHPA/CH3CN (21% conversion).
The identity of 3 has not yet been unequivocally con-
firmed.[18] The same compound has, however, been isolated
under other conditions, specifically: (1) as an undissolved solid
after contacting 1 with H2O2/CH3CN at 708C for 24 h in the ab-
sence of Cy8 (CT-1-H2O2/CH3CN in Figures 4 and 5), or (2) from
the yellow solution obtained after contacting 2 with H2O2/
CH3CN at 708C for 24 h.[7b] As proposed previously, the charac-
terization data for 3 indicate that it may be a dinuclear or poly-
nuclear oxidoperoxido molybdenum species bearing coordi-
nated trz moieties.[7b,18] It is worth highlighting that the solid
phase obtained with 1/H2O2/CH3CN corresponds to the hybrid
2 when Cy8 is present, and compound 3 when the substrate is
As noted above, the catalytic performances for 1 and 2 are
strikingly different for TBHPD/CH3CN. Under these conditions, 2
gives a biphasic reaction mixture and characterization of the
solid phase showed it to be 2.[7b] For 1, although the IR spec-
trum of the yellow recovered solid 1-TBHPD/CH3CN (Figure 4e)
is quite similar to that for 1, the PXRD patterns do not match,
suggesting that the recovered solid has a different crystalline
structure. Based on these results, the presence of species of
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
5
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
the type 1 in the reaction medium enhances the kinetics of
the catalytic reaction.
bipy)2], and [MoVI8O24(di-tBu-bipy)4] (di-tBu-bipy=4,4’-di-tert-
butyl-2,2’-bipyridine) gave good results when used to promote
Lewis acid-catalyzed reactions performed under mild condi-
tions, namely the acetalization of benzaldehyde (PhCHO) and
the ethanolysis of styrene oxide (StyOx).[19] As complex 1 may
also possess intrinsic Lewis acidity, its performance in alcoholy-
sis and acetalization reactions was studied (Scheme 1).
The catalytic performance of 2 with TBHPD/CH3CN is poor
(35% conversion at 24 h), as reported in ref. [7b]. In the pres-
ent work, a much higher activity was observed when using
TFT as the co-solvent (97% conversion at 24 h), even though
under these conditions a biphasic reaction mixture was ob-
tained and characterization of the recovered solid showed it to
be 2 (line (j) in Figures 4 and 5), as also found for TBHPD/
CH3CN. Such a pronounced solvent-dependent catalytic activity
was not observed with 1-TBHPD, although the initial reaction
was faster with TFT than with CH3CN (Figure 3). A yellow mi-
crocrystalline solid was isolated from 1 with TBHPD/TFT, denot-
ed as 1-TBHPD/TFT. A careful inspection of the PXRD and IR
data for this solid indicate that it may be a mixture of phases
corresponding to compound 3 and the recovered solid ob-
tained with TBHPD/CH3CN (1-TBHPD/CH3CN). The solid 1-TBHPD/
TFT was reused for a 24 h batch run, and led to very similar
catalytic results to those for run 1 (Figure 6). The solid phases
isolated from the two batch runs (1-TBHPD/TFT and 1-TBHPD/
TFT-run2) presented similar characterization data (Figures 4
and 5). The catalytic reaction mixtures consisted of yellow
liquid and solid phases, indicating the presence of soluble and
insoluble metal species. A contact test was carried out for 1
with TBHPD/TFT (24 h, without olefin), and the resultant liquid
and solid phases were tested separately for the reaction Cy8/
TBHPD/TFT (denoted LP-CT-1 and SP-CT-1, respectively). Both
led to relatively fast Cy8 reaction, somewhat in parallel to that
observed for the normal catalytic test (Figure 6), showing that
the catalytic process with 1 occurred in the homogeneous
phase. Slight differences in catalytic results between the
normal catalytic test and the contact test may be due to differ-
ences in the initial amount of metal species.
The reaction of StyOx in methanol or ethanol (acting as
both reactant and solvent) at 358C in the presence of 1 gave
the corresponding b-alkoxy alcohol products, methoxy-2-phe-
nylethanol and ethoxy-2-phenylethanol, in quantitative yield at
30 min and 6 h reaction, respectively (Figure 7). Without cata-
lyst, the reaction is sluggish (<11% conversion at 24 h for
methanol, and <5% conversion at 24 h for ethanol).
The excellent regioselectivity towards the b-alkoxy alcohol
product may be attributed to the Lewis acid character of 1.
For acid-catalyzed reactions, the reaction mechanism may in-
Scheme 1. Acid-catalyzed reactions studied in this work (R=CH3, CH3CH2).
Figure 6. Cy8 conversion in the presence of 1 with TBHPD/TFT at 708C at 1 h
(black bar), 3 h (dark gray bar), and 6 h (light gray bar) of reaction, in run 1,
run 2, LP-CT-1 and SP-CT-1.
Acid-catalyzed reactions
~
Figure 7. Ethanolysis of StyOx (&), Cy6Ox ( ), or MeStyOx (c) in the pres-
We recently reported that the complexes [MoVIO2Cl2(di-tBu-
^
ence of 1, and methanolysis of StyOx in the presence of 1 ( ) or 2 (*). Re-
action temperature=358C. The dashed lines are visual guides.
bipy)], [MoV O2(m2-O)Cl4(di-tBu-bipy)2], [MoVI2O2(m2-O)2Cl2(di-tBu-
2
&
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
volve interaction between the Lewis acidic metal center and
the oxygen atom of the substrate, facilitating the nucleophilic
attack of the reacting solvent (methanol, ethanol) at the more
substituted carbon atom of the oxirane ring of the sub-
strate.[19,20] The different alcoholysis reaction rates observed for
methanol and ethanol (78% and 10% StyOx conversion, re-
spectively, at 10 min) may be due to steric hindrance effects
becoming more important in the nucleophilic attack of the
bulkier alcohol molecule at the more substituted (relatively
hindered) carbon atom of the oxirane ring.[21]
StyOx. The reaction was much slower for 2 (60% conversion at
6 h, 100% conversion at 24 h) than for 1 (100% conversion at
30 min) (Figure 7). An initial induction period of 30 min is ob-
served for 2 (and not for 1), which may be partly due to poor
solubility and slow dissolution of the metal species. This hy-
pothesis was supported by measuring the mass of undissolved
solid for the systems 1/methanol and 2/methanol after stirring
for 1 h at 358C, for the same initial mass basis: for 1/methanol,
approximately 48 wt% of the initial mass of solid was recov-
ered, whereas for 2/methanol, >97 wt% of the solid was re-
covered.[23] Hence, the solubility of the metal species seems to
be much lower for 2 than for 1, which may partly contribute
to the slower kinetics in the former case.
Contact tests were carried out by using 1/methanol or 1/
ethanol, at 358C, and without substrate. The use of the resul-
tant liquid phases for StyOx alcoholysis at 358C led to 100%
conversion at 30 min for the methanol phase, and 90% con-
version at 6 h for the ethanol phase. These catalytic results are
not dissimilar from those for the corresponding normal catalyt-
ic tests performed in the presence of 1. Furthermore, the reac-
tion mixtures were always biphasic solid–liquid, and a two-fold
increase in the amount of 1 did not significantly affect the re-
action kinetics with methanol or ethanol as nucleophiles.
These results are consistent with the catalytic process being
homogeneous, and the liquid phase being saturated with
active metal species. The slight differences in catalytic results
for the normal catalytic test (performed in the presence of 1)
and the respective contact test may be due to differences in
the initial amount of soluble active species.
Compound 1 promoted the acetalization of PhCHO with
methanol or ethanol, at 358C, leading to the corresponding
(dialkoxymethyl)benzene product with 100% selectivity at very
high conversions (Scheme 1). The PhCHO/methanol catalytic
reaction gave 100% acetal selectivity at 92% conversion at 1 h
(without catalyst, conversion at 4 h was 24%), and the PhCHO/
ethanol catalytic reaction gave 100% acetal selectivity at 73%
conversion at 24 h (without catalyst no reaction occurred). The
catalytic acetalization was much faster with methanol than
with ethanol (Figure 8), which may be partly due to steric ef-
fects.[24] According to Herbst et al.,[24a] the reaction mechanism
of carbonyl group acetalization in the presence of Lewis acid
catalysts may involve the coordination of the alcohol (reacting
solvent) and aldehyde substrate to Lewis acid sites, leading to
chelating intermediates, which are subsequently converted to
the acetal and water. In this process, steric hindrance effects
with bulkier nucleophiles (ethanol versus methanol, for exam-
ple) may be more significant.
Compound 1 was further explored for catalytic ethanolysis
of (1R,2R)-phenylpropylene oxide (MeStyOx) and cyclohexene
oxide (Cy6Ox) (Figure 7, Scheme 1). The reaction of MeStyOx
gave 1-ethoxy-1-phenyl-2-propanol with 100% selectivity at
100% conversion, in 24 h. The reaction of Cy6Ox gave 2-ethox-
ycyclohexanol in 92% yield at 99% conversion, in 24 h; cyclo-
hexan-1,2-diol was formed as a byproduct in 7% yield. Without
catalyst, MeStyOx conversion was <4% at 24 h reaction,
whereas the reaction of Cy6Ox did not occur to a measurable
extent.
The kinetic profile of the acetalization of PhCHO with metha-
nol tends to a plateau below 100% conversion (Figure 8),
which is likely due to thermodynamic limitations caused by
the presence of water. Water, which is formed in the acetaliza-
There is a paucity of literature data available on the use of
oxomolybdenum compounds to promote the ring-opening re-
action of epoxides with alcohols. In the study mentioned
above,[19] the highest StyOx ethanolysis conversion recorded at
6 h (for a reaction temperature of 558C) was 96% for the octa-
nuclear complex [Mo8O24(di-tBu-bipy)4]. Complex 1 seems to
be a superior catalyst for this reaction as a comparable conver-
sion of 94% was reached within 3 h and at the lower reaction
temperature of 358C. Jeyakumar and Chand[22] used MoO2Cl2
as a catalyst for the transformation of various epoxides to the
corresponding b-alkoxy alcohols at room temperature; the
yields were 95% (2 h) for methoxy-2-phenylethanol, 92% (2 h)
for ethoxy-2-phenylethanol (both from StyOx), and 91% (1 h)
for 2-methoxycyclohexanol (from Cy6Ox). Carrasco et al. report-
ed that the oxidation of cyclohexene with H2O2 in methanol in
the presence of the octamolybdate salt [C4mim]4[Mo8O26]
(C4mim=1-butyl-3-methylimidazolium) as catalyst could give
2-methoxycyclohexanol in 98% yield under certain conditions
(608C, 18 h).[21c]
Figure 8. Reaction of benzaldehyde (PhCHO) at 358C in the presence of 1,
For the sake of comparison, the hybrid material [MoO3(trz)0.5]
(2) was tested as an acid catalyst for the methanolysis of
^
when using methanol ( ) or ethanol (&), or in the presence of 2 when
using methanol (*).
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
7
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
tion reaction, may favor the reverse reaction, that is, the hy-
drolysis of the acetal to the aldehyde.[19,24,25]
anism of these reactions is not clear at present, the results indi-
cate that 1 has sufficient Lewis acidity and oxophilicity to inter-
act with (and activate) the epoxide/aldehyde substrate.
To the best of our knowledge, previous studies on the use
of organo-oxomolybdenum compounds for catalytic acetaliza-
tion of PhCHO are limited to the complex [MoO2Cl2(di-tBu-
bipy)], which led to 86% acetal yield after 10 min of reaction
of PhCHO with ethanol at 558C.[19] The catalytic performance
of 1 and 2 was compared under identical PhCHO reaction con-
ditions (Figure 8). The hybrid material 2 led to slower kinetics,
and selectivity to the acetal product was higher for 1 (100% at
92% conversion, 1 h reaction) than 2 (phenylmethanediol was
formed as a byproduct in 5% yield at 93% conversion, 6 h).
Experimental Section
Materials and methods
For synthesis, MoO3 (99.5%, Analar, BDH Chemicals), 1,2,4-triazole
(98%), diethyl ether (puriss p.a.), 30% aq. H2O2, and CH2Cl2 (puriss
p.a., Fluka) were acquired from Sigma–Aldrich unless otherwise in-
dicated. For catalysis, cis-cyclooctene (95%), styrene oxide (purum,
ꢀ97%, Fluka), (1R,2R)-phenylpropylene oxide (99%), cyclohexene
oxide (98%), benzaldehyde (99%), acetonitrile (ꢀ99%), anhydrous
a,a,a-trifluorotoluene (ꢀ99%), 30% aq. H2O2 (Reidel-de-Haen),
5.5m tert-butylhydroperoxide in decane, and 70% aq. tert-butylhy-
droperoxide were acquired from Sigma–Aldrich unless otherwise
indicated. All chemicals were used as received without further pu-
rification.
Microanalyses for C, H, and N were carried out with a Truspec
Micro CHNS 630-200-200 elemental analyzer. PXRD data were col-
lected at ambient temperature with a Philips Analytical Empyrean
(q/2q) diffractometer equipped with a PIXcel1D detector, with au-
tomatic data acquisition (X’Pert Data Collector software v4.2) using
monochromatized CuKa radiation (l=1.5406 ꢁ). Intensity data
were collected by the step counting method (step 0.02–0.038, 70 s
per step), in continuous mode, in the ca. 3.5ꢁ2qꢁ708 range.
FTIR spectra were collected by using KBr pellets (Sigma–Aldrich,
99%, FTIR grade) and a Mattson-7000 infrared spectrophotometer.
Attenuated total reflectance (ATR) FTIR spectra were measured
with the same instrument equipped with a Specac Golden Gate
Mk II ATR accessory with a diamond top plate and KRS-5 focusing
lenses. Solid-state 13C cross-polarization (CP) magic-angle spinning
(MAS) NMR spectra were recorded by using a Bruker Avance 400
spectrometer (9.4 T) at 100.62 MHz with 3.7 ms 1H 908 pulses,
2000 ms contact time, spinning rate of 12 kHz, and 5 s recycle
delays. Chemical shifts are quoted in parts per million (ppm) from
tetramethylsilane.
Conclusions
The isolation of the trinuclear complex 1 from the reaction of
MoO3 with H2O2 in the presence of 1,2,4-triazole (trz) under-
lines the structure-directing influence that organic ligands can
exert in molybdenum-peroxido aqueous systems. Complex 1 is
unprecedented from a structural perspective. It is, to the best
of our knowledge, the first example of a discrete molybde-
num(VI) complex bearing any kind of 1,2,4-triazole ligand. The
coordination motif comprising the oxo-linked chain Mo-O-Mo-
O-Mo supported by single N1,N2-triazole bridges resembles
the one-dimensional Mo/oxido/triazole chain motif found in
some molybdenum(VI) oxide hybrid materials containing 4-
substituted 1,2,4-triazole molecules. Discrete metal–triazole
complexes containing single trz bridges are extremely rare and
complex 1 may be the first example of a trinuclear complex to
exhibit this feature. The final noteworthy structural highlight of
1 is the presence of the peripheral oxidodiperoxido moieties.
Owing to these structural characteristics, complex 1 is of in-
terest as a catalyst for oxidation reactions that employ hydro-
peroxide oxidants. The catalytic performance for the epoxida-
tion of cis-cyclooctene is greatly dependent on the oxidant/co-
solvent pairing. Although high catalytic activity was observed
with H2O2/CH3CN, the results with TBHPA/CH3CN were poor,
even though in both systems complex 1 was converted in situ
to the hybrid material [MoO3(trz)0.5] (2). The catalytic per-
formance with TBHPA improved by using TFT as the co-solvent
instead of CH3CN; under these conditions 2 did not form and
instead 1 was converted to an oxidoperoxido molybdenum–trz
complex (3). With TFT as co-solvent, the best epoxidation per-
formance was obtained with TBHPD as the oxidant. Under
these conditions, the catalysis was homogeneous, and the
solid phase of the reaction mixture was a mixture of species,
one of which was 3.
A full description of the crystallographic studies is given in the
Supporting Information, including details of the data collection
and refinement (Table S1) and selected bond lengths and angles
(Table S2). CCDC 1573386 contains the supplementary crystallo-
graphic data for this paper. These data are provided free of charge
by The Cambridge Crystallographic Data Centre.
(Htrz)2[Mo3O6(O2)4(trz)2]·H2O (1)
MoO3 (2.00 g, 13.90 mmol) was treated with 30% H2O2 (14 equiv,
20 mL, 195 mmol) and stirred at 558C for 1 h, giving a yellow sus-
pension. 1,2,4-Triazole (2 equiv) was then added slowly over a
period of 25 min while the solution was stirred at 08C. The mixture
was then stirred for 4 h. The resultant yellow solution was filtered
off, slowly concentrated, and left to stand at ambient temperature
for 10 days. Compound 1 precipitated as a yellow microcrystalline
solid, which was recovered by filtration and dried (2.9 g, 78%).
Single crystals suitable for XRD were obtained by slow evaporation
at ambient temperature of a solution of 1 in CH2Cl2. Elemental
analysis calcd (%) for C8H14N12Mo3O14·H2O (808.10): C 11.89, H 1.99,
N 20.80; found: C 11.53, H 2.28, N 20.95; FTIR (KBr): n˜ =310 (m),
329 (w), 360 (w), 374 (m), 422 (m), 525 (m), 571 (s), 628 (vs), 663
(vs), 692 (s), 725 (m), 790 (m), 854 (vs), 873 (vs), 894 (vs), 931 (vs),
956 (s), 1062 (s), 1128 (w), 1153 (w), 1197 (w), 1276 (m), 1309 (s),
1398 (m), 1430 (m), 1519 (m), 1528 (sh), 1566 (m), 1631 (w), 2426
Complex 1 was also examined in acid catalysis and promis-
ing results were obtained for the alcoholysis of epoxides and
the acetalization of benzaldehyde under mild conditions.
Based on the ethanolysis reaction of styrene oxide, complex 1
was significantly more active than other recently examined oxi-
domolybdenum(V/VI) complexes bearing the ligand 4,4’-di-tert-
butyl-2,2’-bipyridine. The most important advantage of 1 is
that in most cases the product selectivity was 100% at high
conversions. Tests confirmed that the catalytic reactions oc-
curred in the homogeneous phase. Although the actual mech-
&
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
8
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
(m), 2512 (m), 2582 (m), 2630 (m), 2672 (m), 2717 (m), 2792 (m),
2879 (m), 2969 (m), 3127 (sh), 3135 cmÀ1 (vs).
ropeu de Desenvolvimento Regional) under the PT2020 Partner-
ship Agreement. The FCT and the European Union are acknowl-
edged for post-doctoral grants to M.M.A. (SFRH/BPD/89068/2012)
and T.R.A. (SFRH/BPD/97660/2013), co-funded by MCTES and the
European Social Fund through the program POPH of QREN.
Catalytic experiments
The catalytic tests were carried out at 358C or 708C, under air in
closed borosilicate batch reactors (5 mL capacity) equipped with a
magnetic stirrer (1000 rpm) and a valve for sampling. Typically, for
the epoxidation reactions, the reactor was loaded with reagents in
the molar proportion 1:100:152 (Mo/Cy8/oxidant); specifically,
0.018 mmol equivalents of Mo, 1.8 mmol cis-cyclooctene (Cy8),
2.74 mmol oxidant, and 1 mL co-solvent. The oxidant was 30% aq.
H2O2, 70% aq. tert-butylhydroperoxide (TBHPA) or 5.5m tert-butyl-
hydroperoxide in decane (TBHPD). The reactor containing 1, Cy8,
and co-solvent was pre-heated at 708C for 10 min, and afterwards
the oxidant solution (not pre-heated) was added to the reactor.
The timing of the reaction started from this instant.
Catalyst stability studies involved reusing the recovered catalyst
under similar reaction conditions to those used in the first run.
Prior to reuse, the solid was separated from the reaction mixture
by centrifugation at 3500 rpm, washed with the same solvent as
that used for the catalytic reaction, air-dried overnight under at-
mospheric conditions, and then under reduced pressure for 1 h at
608C.
Contact tests (CTs) were carried out as follows: compound 1 was
treated with H2O2/CH3CN for 24 h at 708C, or with TBHPD/TFT for
6 h at 708C. The conditions employed were similar to those used
for a normal catalytic test, albeit without the olefin. Subsequently,
the mixture was cooled to ambient temperature and centrifuged
(3500 rpm). The liquid phase was passed through a 13 mm syringe
fitted with a 0.20 mm PTFE filter. Cy8 was added to the liquid
phase in an amount to give an initial molar concentration of Cy8
equal to that for a normal catalytic test, and the solution was
stirred for 24 h at 708C. Attempts to isolate metal species from the
liquid phases of the tests (by addition of n-pentane, n-hexane, or
acetone, and leaving the solutions overnight at 48C) were unsuc-
cessful. The (undissolved) solid obtained from the CTs was washed,
dried overnight under atmospheric conditions, and then under re-
duced pressure for 1 h at 608C.
The alcoholysis and acetalization reactions were carried out at
358C by using the same type of reactor as used for the epoxida-
tion reactions. Each reactor was loaded with a 0.4m solution of
substrate (StyOx or PhCHO) in methanol or ethanol (1 mL), and
compound 1 (37.5 mmol Mo) or compound 2 (113 mmol Mo).
The progress of the catalytic reactions was monitored by using a
Varian 3800 GC equipped with a Bruker BR-5 CB capillary column
(30 mꢂ0.25 mmꢂ0.25 mm) and a flame ionization detector, and H2
was used as the carrier gas; quantifications were based on calibra-
tions using internal standards (undecane, 1-octene, and nonane for
the substrates Cy8, StyOx, and PhCHO, respectively). The reaction
products were identified by GC-MS [Trace GC 2000 Series (Thermo
Quest CE Instruments)-DSQ II (Thermo Scientific)], equipped with a
DB-1 capillary column (30 mꢂ0.25 mmꢂ0.25 mm), and He as the
carrier gas.
Conflict of interest
The authors declare no conflict of interest.
Keywords: alcoholysis
· epoxidation · molybdenum · N
ligands · peroxido ligands
[1] S.-Q. Bai, D. J. Young, T. S. A. Hor, Chem. Asian J. 2011, 6, 292–304.
[2] R. Contreras, A. Flores-Parra, E. Mijangos, F. Tꢃllez, H. Lꢄpez-Sandoval, N.
Barba-Behrens, Coord. Chem. Rev. 2009, 253, 1979–1999.
˘
˘
[3] C. Pettinari, A. Tabacaru, S. Galli, Coord. Chem. Rev. 2016, 307, 1–31.
[4] J. G. Haasnoot, Coord. Chem. Rev. 2000, 200–202, 131–185.
[5] G. Aromꢅ, L. A. Barrios, O. Roubeau, P. Gamez, Coord. Chem. Rev. 2011,
255, 485–546.
[6] C. R. Groom, I. J. Bruno, M. P. Lightfoot, S. C. Ward, Acta Crystallogr. Sect.
B 2016, 72, 171–179.
[7] a) P. J. Hagrman, R. L. LaDuca, H. J. Koo, R. Rarig, R. C. Haushalter, M. H.
Whangbo, J. Zubieta, Inorg. Chem. 2000, 39, 4311–4317; b) T. R. Amar-
ante, P. Neves, A. A. Valente, F. A. A. Paz, M. Pillinger, I. S. GonÅalves, J.
Catal. 2016, 340, 354–367.
[8] a) A. B. Lysenko, G. A. Senchyk, J. Lincke, D. Lꢆssig, A. A. Fokin, E. D.
Butova, P. R. Schreiner, H. Krautscheid, K. V. Domasevitch, Dalton Trans.
2010, 39, 4223–4231; b) G. A. Senchyk, A. B. Lysenko, H. Krautscheid,
K. V. Domasevitch, Inorg. Chem. Commun. 2011, 14, 1365–1368; c) O. V.
Sharga, A. B. Lysenko, M. Handke, H. Krautscheid, E. B. Rusanov, A. N.
Chernega, K. W. Kramer, S. X. Liu, S. Decurtins, A. Bridgeman, K. V. Do-
masevitch, Inorg. Chem. 2013, 52, 8784–8794; d) B. Liu, Y.-H. Yu, W.-H.
Jiang, G.-F. Hou, J.-S. Gao, Inorg. Chem. Commun. 2015, 54, 12–15;
e) A. B. Lysenko, G. A. Senchyk, K. V. Domasevitch, J. Hauser, D. Fuhr-
mann, M. Kobalz, H. Krautscheid, P. Neves, A. A. Valente, I. S. GonÅalves,
Inorg. Chem. 2015, 54, 8327–8338; f) A. B. Lysenko, G. A. Senchyk, L. V.
Lukashuk, K. V. Domasevitch, M. Handke, J. Lincke, H. Krautscheid, E. B.
Rusanov, K. W. Krꢆmer, S. Decurtins, S.-X. Liu, Inorg. Chem. 2016, 55,
239–250; g) A. B. Lysenko, G. A. Senchyk, K. V. Domasevitch, M. Kobalz,
H. Krautscheid, J. Cichos, M. Karbowiak, P. Neves, A. A. Valente, I. S. Gon-
Åalves, Inorg. Chem. 2017, 56, 4380–4394.
[9] J.-Y. Piquemal, E. Briot, J.-M. Brꢃgeault, Dalton Trans. 2013, 42, 29–45.
[10] M. Herbert, F. Montilla, E. ꢇlvarez, A. Galindo, Dalton Trans. 2012, 41,
6942–6956.
[11] P. Neves, A. C. Gomes, F. A. A. Paz, A. A. Valente, I. S. GonÅalves, M. Pillin-
ger, Mol. Catal. 2017, 432, 104–114.
[12] C. Djordjevic, J. L. Gundersen, B. A. Jacobs, E. Sinn, Polyhedron 1989, 8,
541–543.
[13] K. Most, N. C. Mçsch-Zanetti, D. Vidovic, J. Magull, Organometallics
´
2003, 22, 5485–5490.
[14] a) T. Arumuganathan, M. Volpe, B. Harum, D. Wurm, F. Belaj, N. C.
Mçsch-Zanetti, Inorg. Chem. 2012, 51, 150–156; b) O. M. El-Kadri, M. J.
Heeg, C. H. Winter, J. Organomet. Chem. 2009, 694, 3902–3911; c) G.
Lyashenko, R. Herbst-Irmer, V. Jancik, A. Pal, N. C. Mçsch-Zanetti, Inorg.
Chem. 2008, 47, 113–120; d) O. M. El-Kadri, M. J. Heeg, C. H. Winter,
´
Dalton Trans. 2006, 1943–1953; e) K. Most, J. Hoßbach, D. Vidovic, J.
Magull, N. C. Mçsch-Zanetti, Adv. Synth. Catal. 2005, 347, 463–472; f) G.
Zhu, J. M. Tanski, G. Parkin, Polyhedron 2003, 22, 199–203; g) K. Most, S.
Kopke, F. Dall’Antonia, N. C. Mçsch-Zanetti, Chem. Commun. 2002,
1676–1677.
Acknowledgments
This work was developed in the scope of the project (Associate
Laboratory) CICECO-Aveiro Institute of Materials-POCI-01–0145-
FEDER-007679 [FCT (Fundażo para a CiÞncia e a Tecnologia) ref.
UID/CTM/50011/2013], financed by national funds through the
FCT/MEC and when applicable co-financed by FEDER (Fundo Eu-
[15] a) D. Gajan, N. Rendꢄn, K. M. Wampler, J.-M. Basset, C. Copꢃret, A.
Lesage, L. Emsley, R. R. Schrock, Dalton Trans. 2010, 39, 8547–8551;
b) C. L. Dezelah IV, O. M. El-Kadri, M. J. Heeg, C. H. Winter, J. Mater.
Chem. 2004, 14, 3167–3176.
[16] V. Arjunan, S. Thirunarayanan, M. K. Marchewka, S. Mohan, J. Mol. Struct.
2017, 1145, 211–221.
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
9
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[17] a) T. R. Amarante, P. Neves, A. C. Coelho, S. Gago, A. A. Valente, F. A. A.
Paz, M. Pillinger, I. S. GonÅalves, Organometallics 2010, 29, 883–892;
b) M. Grzywa, W. Nitek, W. Łasocha, J. Mol. Struct. 2008, 888, 318–326;
c) V. Conte, B. Floris, Dalton Trans. 2011, 40, 1419–1436; d) P. Neves, S.
Gago, S. S. Balula, A. D. Lopes, A. A. Valente, L. Cunha-Silva, F. A. A. Paz,
M. Pillinger, J. Rocha, C. M. Silva, I. S. GonÅalves, Inorg. Chem. 2011, 50,
3490–3500; e) “Oxidodiperoxidomolybdenum Complexes: Properties
and Their Use as Catalysts in Green Oxidations”: F. Montilla, A. Galindo
in Elsevier Reference Module in Chemistry, Molecular Sciences and Chemi-
cal Engineering (Ed.: J. Reedijk), Elsevier, Waltham, MA, 2017, pp. 1–27.
[18] Characterization data for 3: elemental analysis calcd (%) for
C2N3H3Mo2O10 (420.9): C 5.71, N 9.98; found: C 5.67, N 9.84; selected
FTIR (ATR): n˜ =525 (s, br), 571 (m), 614 (s), 745 (w, br), 861 (m), 909 (s),
946 (m), 1002 (w), 1055 (m), 1175 (w), 1320 (m), 1435 (w), 1537 cmÀ1
(m); selected FT-Raman: n˜ =317 (m), 367 (w), 389 (w), 571 (m), 607 (w),
679 (w), 703 (w), 916 (m), 955 (vs), 1054 (vw), 1315 (m), 1432 (vw),
1536 cmÀ1 (vw); 1H NMR (300 MHz, D2O, 258C, TMS): d=8.35 ppm (s,
2H, CH). Tentative structural interpretation based on these data: The el-
emental analysis results are consistent with an empirical formula ap-
proximating to Mo2O10(trz). The observation of a singlet in the 1H NMR
spectrum suggests that both trz protons are equivalent and therefore
the trz ligand most likely adopts a bridging N1/N2 coordination mode
to two equivalent Mo centers. The simplest species compatible with
this coordination mode would be a dinuclear complex. Peroxomolybde-
num complexes containing the dinuclear moiety {Mo2O2(m-O2)2(O2)2}
(Mo2O10) are well known and usually contain an additional ligand such
as sulfate or phosphate that acts as a bipodal ligand to the two Mo
atoms. Considering that this is also the preferred coordination mode of
triazole, the complex [Mo2O2(m-O2)2(O2)2(trz)], containing an N1,N2-
bridging trz, can be proposed as a plausible structure for 3.
[19] A. C. Gomes, P. Neves, L. Cunha-Silva, A. A. Valente, I. S. GonÅalves, M.
Pillinger, Catal. Sci. Technol. 2016, 6, 5207–5218.
[20] a) P. Silva, F. Vieira, A. C. Gomes, D. Ananias, J. A. Fernandes, S. M. Bruno,
R. Soares, A. A. Valente, J. Rocha, F. A. A. Paz, J. Am. Chem. Soc. 2011,
133, 15120–15138; b) A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Chem.
Eur. J. 2010, 16, 8530–8536.
[21] a) S. S. Kahandal, S. R. Kale, S. T. Disale, R. V. Jayaram, Catal. Sci. Technol.
2012, 2, 1493–1499; b) G. B. Payne, C. W. Smith, J. Org. Chem. 1957, 22,
1682–1685; c) C. J. Carrasco, F. Montilla, E. ꢇlvarez, M. Herbert, A. Galin-
do, Polyhedron 2013, 54, 123–130.
[22] K. Jeyakumar, D. K. Chand, Synthesis 2008, 5, 807–819.
[23] For the alcoholysis or acetalization reactions using 1 or 2, PXRD and
ATR FTIR spectral data for the solid phases recovered by centrifugation
after 24 h catalytic runs were in agreement with those for the as-pre-
pared catalysts (not shown here).
[24] a) A. Herbst, A. Khutia, C. Janiak, Inorg. Chem. 2014, 53, 7319–7333;
b) A. Dhakshinamoorthy, M. Alvaro, H. Garcia, Adv. Synth. Catal. 2010,
352, 3022–3030.
[25] A. S. Poyraz, C.-H. Kuo, E. Kim, Y. Meng, M. S. Seraji, S. L. Suib, Chem.
Mater. 2014, 26, 2803–2813.
Manuscript received: February 9, 2018
Revised manuscript received: March 21, 2018
Accepted manuscript online: March 22, 2018
Version of record online: && &&, 0000
&
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
10
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
A structurally unparalleled molybde-
num(VI) complex has been isolated
from the reaction of MoO3 with H2O2 in
the presence of 1,2,4-triazole. The com-
plex leads to selective conversions in
catalytic olefin epoxidation, mild acid-
catalyzed ring-opening of epoxides, and
acetalization of benzaldehyde by using
alcohols as solvents and nucleophiles.
M. M. Antunes, T. R. Amarante,
A. A. Valente,* F. A. Almeida Paz,
I. S. GonÅalves, M. Pillinger*
&& – &&
A Linear Trinuclear Oxidodiperoxido-
molybdenum(VI) Complex with Single
Triazole Bridges: Catalytic Activity in
Epoxidation, Alcoholysis, and
Acetalization Reactions
ChemCatChem 2018, 10, 1 – 11
www.chemcatchem.org
11
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ