Synthesis and catalytic properties of molybdenum(VI) complexes with tris(3,5-dimethyl-1-pyrazolyl)methane

Article abstract of DOI:10.1021/ic102456z

The complex [MoO₂Cl{HC(3,5-Me₂pz)₃}] BF₄ (1) (HC(3,5-Me₂pz)₃ = tris(3,5-dimethyl-1- pyrazolyl)methane) has been prepared and examined as a catalyst for epoxidation of olefins at 55 °C using tert-butyl hydroperoxide (TBHP) as the oxidant. For reaction of cis-cyclooctene, epoxycyclooctane is obtained quantitatively within 5 h when water is rigorously excluded from the reaction mixture. Increasing amounts of water in the reaction mixture lead to lower activities (without affecting product selectivity) and transformation of 1 into the trioxidomolybdenum(VI) complex [{HC(3,5-Me₂pz)₃}MoO ₃] (4). Complex 4 was isolated as a microcrystalline solid by refluxing a suspension of 1 in water. The powder X-ray diffraction pattern of 4 can be indexed in the orthorhombic Pnma system, with a = 16.7349(5) A, b = 13.6380(4) A, and c = 7.8513(3) A. Treatment of 1 in dichloromethane with excess TBHP led to isolation of the symmetrical [Mo ₂O₄(μ₂-O){HC(3,5-Me₂pz) ₃}₂](BF₄)₂ (2) and unsymmetrical [Mo₂O₃(O₂)₂(μ2- O)(H ₂O){HC(3,5-Me₂pz)₃}] (3) oxido-bridged dimers, which were characterized by single-crystal X-ray diffraction. Complex 2 displays the well-known (Mo₂O₅)²⁺ bridging structure where each dioxidomolybdenum(VI) center is coordinated to three N atoms of the organic ligand and one μ2-bridging O atom. The unusual complex 3 comprises dioxido and oxidodiperoxo molybdenum(VI) centers linked by a μ2-bridgingO atom, with the former center being coordinated to the tridentate N-ligand. The dinuclear complexes exhibit a similar catalytic performance to that found for mononuclear 1. For complexes 1 and 2 use of the ionic liquids (ILs) 1-butyl-3-methylimidazolium tetrafluoroborate and N-butyl-3-methylpyridinium tetrafluoroborate as solvents allowed the complexes to be completely dissolved, and in each case the catalyst and IL could be recycled and reused without loss of activity.

Full text of DOI:10.1021/ic102456z

ARTICLE
pubs.acs.org/IC
Synthesis and Catalytic Properties of Molybdenum(VI) Complexes
with Tris(3,5-dimethyl-1-pyrazolyl)methane
Patrícia Neves,† Sandra Gago,†,‡ Salete S. Balula,†,§ Andrꢀe D. Lopes,|| Anabela A. Valente,*,† Luís Cunha-Silva,†,§
Filipe A. Almeida Paz,† Martyn Pillinger,† Jo~ao Rocha,† Carlos M. Silva,† and Isabel S. Gonc-alves*,†
†Department of Chemistry, CICECO, University of Aveiro, 3810-193 Aveiro, Portugal
‡Department of Chemistry, REQUIMTE, Faculty of Science and Technology, New University of Lisbon, 2829-516 Monte da Caparica,
Portugal
§REQUIMTE/Department of Chemistry and Biochemistry, Faculty of Science, University of Porto, 4169-007 Porto, Portugal
Faculty of Science and Technology, Department of Chemistry, Biochemistry and Pharmacy, University of the Algarve,
Campus de Gambelas, 8005-136 Faro, Portugal
S Supporting Information
b
ABSTRACT: The complex [MoO2Cl{HC(3,5-Me2pz)3}]BF4 (1) (HC(3,5-Me2pz)3 =
tris(3,5-dimethyl-1-pyrazolyl)methane) has been prepared and examined as a catalyst
for epoxidation of olefins at 55 °C using tert-butyl hydroperoxide (TBHP) as the
oxidant. For reaction of cis-cyclooctene, epoxycyclooctane is obtained quantitatively
within 5 h when water is rigorously excluded from the reaction mixture. Increasing
amounts of water in the reaction mixture lead to lower activities (without affecting
product selectivity) and transformation of 1 into the trioxidomolybdenum(VI)
complex [{HC(3,5-Me2pz)3}MoO3] (4). Complex 4 was isolated as a microcrystalline
solid by refluxing a suspension of 1 in water. The powder X-ray diffraction pattern of 4
can be indexed in the orthorhombic Pnma system, with a = 16.7349(5) Å, b =
13.6380(4) Å, and c = 7.8513(3) Å. Treatment of 1 in dichloromethane with excess
TBHP led to isolation of the symmetrical [Mo2O4(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2) and unsymmetrical [Mo2O3(O2)2(μ2-
O)(H2O){HC(3,5-Me2pz)3}] (3) oxido-bridged dimers, which were characterized by single-crystal X-ray diffraction. Complex 2
displays the well-known (Mo2O5)2þ bridging structure where each dioxidomolybdenum(VI) center is coordinated to three N
atoms of the organic ligand and one μ2-bridging O atom. The unusual complex 3 comprises dioxido and oxidodiperoxo
molybdenum(VI) centers linked by a μ2-bridging O atom, with the former center being coordinated to the tridentate N-ligand. The
dinuclear complexes exhibit a similar catalytic performance to that found for mononuclear 1. For complexes 1 and 2 use of the ionic
liquids (ILs) 1-butyl-3-methylimidazolium tetrafluoroborate and N-butyl-3-methylpyridinium tetrafluoroborate as solvents allowed
the complexes to be completely dissolved, and in each case the catalyst and IL could be recycled and reused without loss of activity.
’ INTRODUCTION
oxidative decarbonylation with an excess of SOCl2, I2, Br2, or
HNO3 leads to MoIII (X = Cl, Br, I) or MoVI (X = O) complexes
of the type [MoX3L].5,6 Very recently, Pombeiro and co-workers
reported that reaction of Li[Mo(CO)3(Tpms)] (Tpms = tris(1-
pyrazolyl)methanesulfonate) with I2 or HBF4 affords the
TpmsꢀMoII complexes [Mo(CO)3X(Tpms)] (X = I, H).7
The latter are oxidized with air to give the dinuclear MoV com-
plexes [{MoO(μ2-O)(Tpms)}2] and (in the presence of
CHCl3) [{MoOCl(Tpms)}2](μ2-O). Reaction of Li[Mo(CO)3-
(Tpms)] with AgBF4 in air gave the dinuclear MoVI complex
[{MoO2(Tpms)}2](μ2-O). Enemark and co-workers also noted
a tendency of TpmꢀMo mononuclear complexes to form di-
nuclear derivatives.6 Reaction of the cationic monooxidomoly-
bdenum(V) complex [MoOCl2{HC(3,5-Me2pz)3}]Cl with ca-
techol in the presence of triethylamine initially gave a solution
Scorpionate-type tris(pyrazolyl)methane (Tpm) ligands are
the neutral analogues of the anionic tris(pyrazolyl)hydroborates
(Tp).1 They are formally derived from Tp by replacing the apical
{BH}ꢀ anionic moiety with the isoelectronic CR group. Although
metal complexes based on Tpm ligands were first reported in
1970,2 they were much less studied than Tp systems mainly due
to poor synthetic routes with low yields. Development of
improved syntheses3 has led to a renewed interest in this
coordination chemistry, with applications in catalysis, photo-
chemistry, and biomedicine.1,4
In his landmark paper on the coordination chemistry of
poly(1-pyrazolyl)alkanes,2 Trofimenko reported the reaction of
Mo(CO)6 with the ligands (L) tris(1-pyrazolyl)methane [HC(pz)3]
and tris(3,5-dimethyl-1-pyrazolyl)methane [HC(3,5-Me2pz)3]
to give the tricarbonyl complexes [Mo(CO)3L]. These com-
plexes are useful starting materials for the preparation of other
complexes in higher oxidation states. For example, complete
Received: December 8, 2010
Published: March 25, 2011
r
2011 American Chemical Society
3490
dx.doi.org/10.1021/ic102456z Inorg. Chem. 2011, 50, 3490–3500
|

Inorganic Chemistry
ARTICLE
Chart 1
’ RESULTS AND DISCUSSION
Synthesis. The complex [MoO2Cl{HC(3,5-Me2pz)3}]BF4
(1) was synthesized by addition of 1 equiv of the ligand
HC(3,5-Me2pz)3 to a solution of [MoO2Cl(THF)3](BF4) in
THF (Chart 1). Slow diffusion of diethyl ether into a concen-
trated solution of 1 in acetonitrile gave a small crop of single
crystals suitable for X-ray diffraction. However, as described
below, analysis of the data revealed a structure containing a
symmetrical dioxido(μ-oxido)molybdenum(VI) complex, [Mo2O4-
(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2), rather than the mono-
nuclear dioxidomolybdenum(VI) complex; the presence of
residual amounts of water in the system may have accounted
for the conversion of a fraction of 1 into less soluble 2. In an
attempt to obtain a better understanding of the Mo species
formed during catalytic olefin epoxidation with complex 1 as the
catalyst precursor, reaction of 1 in CH2Cl2 with a large excess (ca.
150 equiv.) of the oxidant TBHP was performed separately. This
gave a yellow solution from which the unsymmetrical dinuclear
complex [Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] (3)
precipitated as a yellow solid upon stirring at room temperature.
Crystals of 3 suitable for X-ray diffraction were obtained by
recrystallization from 1,2-dichloroethane/diethyl ether/pentane.
The CH2Cl2 filtrate obtained in the above procedure was treated
with MnO2 to destroy the excess peroxide. Addition of diethyl
ether to the concentrated solution gave complex 2 (as confirmed
by infrared and Raman spectroscopies and elemental analysis) as
a colorless solid in 30% yield. Complex 1 may be transformed
into 2 via hydrolysis of the MoꢀCl bonds and dimerization of the
resultant metal species (the TBHP solution was used as received
and may contain up to 4% water). Several dimers of the type
[Mo2O4(μ2-O)Cl2(L)n] (L = monodentate or bidentate ligand)
have been obtained in an analogous fashion starting from
monomers of the type [MoO2Cl2(L)n/2] in water or in water-
containing systems and in the absence or in the presence of
TBHP.10 In an experiment to assess the stability of 1 toward
moisture, a suspension of the complex in water was refluxed for
2 h. A white precipitate was obtained, which was identified
(by infrared and Raman spectroscopies and elemental analysis)
as being the trioxidomolybdenum(VI) complex [{HC(3,5-
Me2pz)3}MoO3] (4), isolated in 51% yield. As reported pre-
viously,6 complex 4 is insoluble in all common organic solvents.
However, the solid obtained by the above procedure was micro-
crystalline, and its powder X-ray diffraction pattern could be
indexed in the orthorhombic Pnma system, with a = 16.7349(5) Å,
b = 13.6380(4) Å, and c = 7.8513(3) Å (see the Supporting
Information for full details).
containing the mononuclear MoV complex [MoOX2{HC(3,5-
Me2pz)3}]ꢀ (X2 = dianion of catechol), but upon standing of
the reaction mixture the unsymmetrical dimer [Mo2O4(OC6H4-
O){HC(3,5-Me2pz)3}] was isolated.
Reaction of [MoO2X2(THF)2] (X = Cl, Br) with HC(pz)3 or
HC(3,5-Me2pz)3 leads to replacement of both coordinated sol-
vent molecules and one of the halo ligands to give [MoO2-
X(L)]X.8 These complexes are moderately active catalysts for the
epoxidation of cis-cyclooctene with tert-butyl hydroperoxide
(TBHP) as the oxidizing agent, giving 1,2-epoxycyclooctane as
the only product. Santos et al. concluded that the organic ligand
remained at least partially attached to the metal in the catalytically
active center and determined its activity. We recently showed that
the analogous complexes [MoO2Cl{HC(bim)3}]X (X = Cl, BF4,
PF6) containing the tridentate ligand tris(benzimidazolyl)me-
thane exhibit a similar catalytic performance for olefin epox-
idation.9 The epoxide selectivity was very good with certain
substrates, and the right combination of complex (X = Cl), ionic
liquid (as solvent), and oxidant allowed efficient recovery of the
catalyst/solvent mixture for use in subsequent reaction runs.
When the complex [MoO2Cl{HC(bim)3}]PF6 was used for the
epoxidation of cis-cyclooctene with TBHP and no cosolvent, an
insoluble solid recovered at the end of a catalytic run was ten-
tatively identified as the dioxido(μ-oxido)molybdenum(VI) di-
mer [Mo2O4(μ2-O){HC(bim)3}2](Y)2 (Y = Cl, PF6), which
is analogous to the dinuclear TpmsꢀMoVI complex described
in ref 7.
Infrared and Raman Spectroscopy. The infrared and Raman
spectra of 1 in the solid state exhibit strong bands at 951 ( 1 and
917 ( 1 cmꢀ1, which are assigned to the symmetric and
asymmetric ModO stretching vibrations, respectively, of the
cis-[MoO2]2þ unit. Several intense bands between 1200 and
1600 cmꢀ1 are assigned to internal ligand modes from the
tridentate tris(3,5-dimethylpyrazolyl)methane ligand (e.g., the
characteristic pyrazolyl ring stretching vibration at 1563 cmꢀ1),
while a very strong and broad absorption centered at 1060 cmꢀ1
is attributed to ν(BꢀF) of tetrafluoroborate anions. Complex 2
exhibits a very similar IR spectrum in the range of 900ꢀ
1600 cmꢀ1, with the ModO stretching vibrations appearing at
916 and 945 cmꢀ1 and ν(BꢀF) at about 1070 cmꢀ1. A medium-
intensity band at 361 cmꢀ1 in both the IR and the Raman spectra
of 1 is assigned to a MoꢀCl stretching mode. As expected, this
The studies outlined above suggested to us that more work
needed to be done to understand the catalytic performance of
complexes of the type [MoO2X(L)]X bearing Tpm ligands. In
the present work the complex [MoO2Cl{HC(3,5-Me2pz)3}]BF4
(1) has been prepared and tested in catalytic olefin epoxidation.
With the aim of obtaining a better understanding of the cata-
lytically active species formed under the reactions conditions
used, complex 1 was treated with excess TBHP, which resulted
in the isolation of the symmetrical [Mo2O4(μ2-O){HC(3,5-
Me2pz)3}2](BF4)2 (2) and unsymmetrical [Mo2O3(O2)2(μ2-
O)(H2O){HC(3,5-Me2pz)3}] (3) dimers. The single-crystal
X-ray structures of 2 and 3 are described. The catalytic
behaviors of all three complexes are compared, and the
influence of water on formation of the different species is
discussed.
3491
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
Table 1. Bond Lengths (in Angstroms) for the Six
Molybdenum Coordination Environments Present in
[Mo2O4(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2)
Table 2. Bond Angles (in degrees) for the Six Molybdenum
Coordination Environments Present in [Mo2O4(μ2-O){HC-
(3,5-Me2pz)3}2](BF4)2 (2)
Mo(1)ꢀO(1)
Mo(1)ꢀO(2)
Mo(1)ꢀO(3)
Mo(1)ꢀN(1)
Mo(1)ꢀN(3)
Mo(1)ꢀN(5)
Mo(2)ꢀO(3)
Mo(2)ꢀO(4)
Mo(2)ꢀO(5)
Mo(2)ꢀN(7)
Mo(2)ꢀN(9)
Mo(2)ꢀN(11)
Mo(3)ꢀO(6)
Mo(3)ꢀO(7)
Mo(3)ꢀO(8)
Mo(3)ꢀN(13)
Mo(3)ꢀN(15)
Mo(3)ꢀN(17)
1.683(5)
1.664(6)
1.892(5)
2.296(7)
2.231(6)
2.304(6)
1.886(5)
1.690(6)
1.693(5)
2.314(6)
2.204(6)
2.325(7)
1.896(5)
1.683(6)
1.688(5)
2.219(6)
2.289(7)
2.289(6)
Mo(4)ꢀO(6)
Mo(4)ꢀO(9)
Mo(4)ꢀO(10)
Mo(4)ꢀN(19)
Mo(4)ꢀN(21)
Mo(4)ꢀN(35)
Mo(5)ꢀO(11)
Mo(5)ꢀO(12)
Mo(5)ꢀO(13)
Mo(5)ꢀN(25)
Mo(5)ꢀN(27)
Mo(5)ꢀN(29)
Mo(6)ꢀO(11)
Mo(6)ꢀO(14)
Mo(6)ꢀO(15)
Mo(6)ꢀN(31)
Mo(6)ꢀN(33)
Mo(6)ꢀN(35)
1.888(5)
1.696(5)
1.689(6)
2.291(6)
2.234(6)
2.290(6)
1.890(5)
1.709(5)
1.691(5)
2.213(6)
2.319(7)
2.308(7)
1.887(5)
1.698(5)
1.680(5)
2.314(7)
2.310(6)
2.230(6)
O(1)ꢀMo(1)ꢀO(3)
O(1)ꢀMo(1)ꢀN(1)
O(1)ꢀMo(1)ꢀN(3)
O(1)ꢀMo(1)ꢀN(5)
O(2)ꢀMo(1)ꢀO(1)
O(2)ꢀMo(1)ꢀO(3)
O(2)ꢀMo(1)ꢀN(1)
O(2)ꢀMo(1)ꢀN(3)
O(2)ꢀMo(1)ꢀN(5)
O(3)ꢀMo(1)ꢀN(1)
O(3)ꢀMo(1)ꢀN(3)
O(3)ꢀMo(1)ꢀN(5)
N(1)ꢀMo(1)ꢀN(5)
N(3)ꢀMo(1)ꢀN(1)
N(3)ꢀMo(1)ꢀN(5)
O(3)ꢀMo(2)ꢀN(7)
O(3)ꢀMo(2)ꢀN(9)
O(3)ꢀMo(2)ꢀN(11)
O(4)ꢀMo(2)ꢀO(3)
O(4)ꢀMo(2)ꢀO(5)
O(4)ꢀMo(2)ꢀN(7)
O(4)ꢀMo(2)ꢀN(9)
O(4)ꢀMo(2)ꢀN(11)
O(5)ꢀMo(2)ꢀO(3)
O(5)ꢀMo(2)ꢀN(7)
O(5)ꢀMo(2)ꢀN(9)
O(5)ꢀMo(2)ꢀN(11)
N(7)ꢀMo(2)ꢀN(11)
N(9)ꢀMo(2)ꢀN(7)
N(9)ꢀMo(2)ꢀN(11)
103.3(2) O(6)ꢀMo(3)ꢀN(13)
160.8(2) O(6)ꢀMo(3)ꢀN(15)
89.9(2) O(6)ꢀMo(3)ꢀN(17)
85.7(2) O(7)ꢀMo(3)ꢀO(6)
104.3(3) O(7)ꢀMo(3)ꢀO(8)
103.7(3) O(7)ꢀMo(3)ꢀN(13)
89.5(3) O(7)ꢀMo(3)ꢀN(15)
91.1(3) O(7)ꢀMo(3)ꢀN(17)
163.2(3) O(8)ꢀMo(3)ꢀO(6)
85.8(2) O(8)ꢀMo(3)ꢀN(13)
156.8(2) O(8)ꢀMo(3)ꢀN(15)
86.7(2) O(8)ꢀMo(3)ꢀN(17)
78.0(2) N(13)ꢀMo(3)ꢀN(15)
76.4(2) N(13)ꢀMo(3)ꢀN(17)
75.2(2) N(15)ꢀMo(3)ꢀN(17)
89.1(2) O(6)ꢀMo(4)ꢀN(19)
157.2(2) O(6)ꢀMo(4)ꢀN(21)
85.1(2) O(6)ꢀMo(4)ꢀN(23)
103.3(2) O(9)ꢀMo(4)ꢀO(6)
104.1(3) O(9)ꢀMo(4)ꢀN(19)
86.8(3) O(9)ꢀMo(4)ꢀN(21)
91.5(2) O(9)ꢀMo(4)ꢀN(23)
162.3(2) O(10)ꢀMo(4)ꢀO(6)
102.8(3) O(10)ꢀMo(4)ꢀO(9)
161.3(2) O(10)ꢀMo(4)ꢀN(19)
89.9(2) O(10)ꢀMo(4)ꢀN(21)
160.4(2)
89.8(2)
87.9(2)
103.5(2)
103.0(3)
88.8(2)
159.9(2)
88.8(3)
102.7(2)
88.9(2)
88.2(3)
161.6(2)
74.6(2)
77.1(2)
76.6(2)
88.2(2)
160.6(2)
90.2(2)
102.4(3)
161.2(2)
89.8(3)
87.3(3)
102.4(2)
104.6(3)
87.9(3)
88.7(3)
band is absent for 2. Complex 1 exhibits a medium-intensity band
at 762 cmꢀ1 in the IR spectrum, which is assigned to a ligand
mode. The corresponding band in the spectrum of 2 appears with
much higher intensity, which is attributed to overlap of the ligand
mode with a ν(MoꢀOꢀMo) vibration. This latter assignment is
fully consistent with literature data for other examples of dioxido-
(μ-oxido)molybdenum(VI) dimers, which typically exhibit the
ν(MoꢀOꢀMo) mode as a strong band in the 750ꢀ800 cmꢀ1
region of the IR spectrum.7,10 For complex 3 two well-separated
and strong bands are observed at 760 and 794 cmꢀ1 in the IR
spectrum; the former band is assigned to a ligand mode, while the
latter is assigned to ν(MoꢀOꢀMo). The cis-[MoO2]2þ frag-
ment in 3 gives rise to two strong ModO stretching vibrations at
915 and 954 cmꢀ1 in the IR spectrum (920 and 948 cmꢀ1 in the
Raman), in agreement with that observed for complex 1. In
addition, the bands at 985 [strong in the Raman (not observed
for 1), weaker in the IR], 862 (medium in the IR), and 567
(medium in the IR) are assigned to ν(ModO), ν(OꢀO), and
ν[MoꢀO(O2)], respectively, from the oxidodiperoxo fragment.
The IR spectrum of 3 contains no band for ν(BꢀF), which is
consistent with the complex being neutral. Complex 4 exhibits
three ν(ModO) absorptions at 899, 861, and 840 cmꢀ1, in
agreement with that reported by Enemark and co-workers for the
same complex, which was prepared in 40% yield by exhaustive
oxidation of [{HC(3,5-Me2pz)3}Mo(CO)3] with nitric acid.6
Crystal Structure Descriptions. The single crystals obtained
as described above were analyzed by X-ray diffraction, allowing a
full crystallographic description. The two crystal structures are
formulated as [Mo2O4(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2)
88.8(3) O(10)ꢀMo(4)ꢀN(23) 160.1(2)
77.7(2) N(21)ꢀMo(4)ꢀN(19)
74.4(2) N(21)ꢀMo(4)ꢀN(23)
76.3(2) N(23)ꢀMo(4)ꢀN(19)
76.3(2)
75.3(2)
77.0(2)
87.7(2)
85.5(2)
O(11)ꢀMo(5)ꢀN(25) 157.5(2) O(11)ꢀMo(6)ꢀN(31)
O(11)ꢀMo(5)ꢀN(27)
O(11)ꢀMo(5)ꢀN(29)
87.2(2) O(11)ꢀMo(6)ꢀN(33)
85.7(2) O(11)ꢀMo(6)ꢀN(35) 157.2(2)
O(12)ꢀMo(5)ꢀO(11) 103.1(2) O(14)ꢀMo(6)ꢀO(11) 103.2(2)
O(12)ꢀMo(5)ꢀN(25)
O(12)ꢀMo(5)ꢀN(27)
90.2(2) O(14)ꢀMo(6)ꢀN(31) 162.3(3)
85.7(3) O(14)ꢀMo(6)ꢀN(33)
89.4(3)
89.9(3)
O(12)ꢀMo(5)ꢀN(29) 160.7(2) O(14)ꢀMo(6)ꢀN(35)
O(13)ꢀMo(5)ꢀO(11) 103.1(2) O(15)ꢀMo(6)ꢀO(11) 103.6(2)
O(13)ꢀMo(5)ꢀO(12) 104.5(3) O(15)ꢀMo(6)ꢀO(14) 103.5(3)
O(13)ꢀMo(5)ꢀN(25)
90.7(3) O(15)ꢀMo(6)ꢀN(31)
87.2(3)
O(13)ꢀMo(5)ꢀN(27) 163.1(2) O(15)ꢀMo(6)ꢀN(33) 161.8(3)
O(13)ꢀMo(5)ꢀN(29)
N(25)ꢀMo(5)ꢀN(27)
N(25)ꢀMo(5)ꢀN(29)
N(29)ꢀMo(5)ꢀN(27)
89.8(3) O(15)ꢀMo(6)ꢀN(35)
75.6(2) N(33)ꢀMo(6)ꢀN(31)
76.5(2) N(35)ꢀMo(6)ꢀN(31)
77.5(2) N(35)ꢀMo(6)ꢀN(33)
91.1(2)
77.4(2)
75.5(2)
75.9(2)
and [Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] 3H2O (3).
the Supporting Information). Due to the similar features of the
three crystallographically independent complex units in 2, the
structural discussion will be mainly focused on the cationic
complex which is formed by Mo(1) and Mo(2). Unless other-
wise stated, these assumptions are valid for the remaining
crystallographically independent complexes. Selected bond
lengths and angles for the six molybdenum coordination envir-
onments present in 2 are listed in Tables 1 and 2.
3
Complex 2 crystallizes in the triclinic space group P1 having
three crystallographically independent cationic [Mo2O4(μ2-O)-
{HC(3,5-Me2pz)3}2]2þ units and six BF4 anions composing
ꢀ
the asymmetric unit (asu), whereas complex 3 crystallizes in the
monoclinic space group P21/c with a neutral complex [Mo2-
O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] and three water
molecules of crystallization comprising the asu (Figure S2 in
3492
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
Table 3. Selected Bond Lengths (in Angstroms) and
Angles (in degrees) for the Molybdenum Coordination
Environments Present in [Mo2O(O2)2(μ2-O)(H2O){HC-
(3,5-Me2pz)3}] 3H2O (3)
3
{Mo(1)N3O3}
{Mo(2)O7}
Mo(1)ꢀO(1)
1.6986(15) Mo(2)ꢀO(3)
1.7213(15) Mo(2)ꢀO(4)
1.8098(14) Mo(2)ꢀO(5)
2.2865(17) Mo(2)ꢀO(6)
2.3289(17) Mo(2)ꢀO(7)
2.3363(17) Mo(2)ꢀO(8)
2.0259(14)
1.6785(18)
1.9589(17)
1.9408(17)
1.9538(16)
1.9345(16)
2.3152(17)
1.477(2)
1.476(2)
94.54(7)
87.13(6)
131.31(7)
89.45(6)
Mo(1)ꢀO(2)
Mo(1)ꢀO(3)
Mo(1)ꢀN(1)
Mo(1)ꢀN(3)
Mo(1)ꢀN(5)
Mo(2)ꢀO(1W)
O(5)ꢀO(6)
O(7)ꢀO(8)
O(1)ꢀMo(1)ꢀO(2) 103.77(7)
O(1)ꢀMo(1)ꢀO(3) 105.10(7)
O(1)ꢀMo(1)ꢀN(1) 88.58(7)
O(1)ꢀMo(1)ꢀN(3) 158.91(7)
O(1)ꢀMo(1)ꢀN(5) 87.82(7)
O(2)ꢀMo(1)ꢀO(3) 104.69(7)
O(2)ꢀMo(1)ꢀN(1) 88.58(7)
O(2)ꢀMo(1)ꢀN(3) 88.93(6)
O(2)ꢀMo(1)ꢀN(5) 158.59(6)
O(3)ꢀMo(1)ꢀN(1) 158.36(7)
O(3)ꢀMo(1)ꢀN(3) 87.42(6)
O(3)ꢀMo(1)ꢀN(5) 89.17(6)
N(1)ꢀMo(1)ꢀN(3) 74.97(6)
N(1)ꢀMo(1)ꢀN(5) 74.42(6)
N(3)ꢀMo(1)ꢀN(5) 75.24(6)
O(3)ꢀMo(2)ꢀO(4)
O(3)ꢀMo(2)ꢀO(5)
O(3)ꢀMo(2)ꢀO(6)
O(3)ꢀMo(2)ꢀO(7)
O(3)ꢀMo(2)ꢀO(8)
133.34(7)
O(3)ꢀMo(2)ꢀO(1W) 82.18(6)
O(4)ꢀMo(2)ꢀO(5)
O(4)ꢀMo(2)ꢀO(6)
O(4)ꢀMo(2)ꢀO(7)
O(4)ꢀMo(2)ꢀO(8)
102.25(9)
100.40(9)
101.45(8)
101.44(8)
O(4)ꢀMo(2)ꢀO(1W) 176.71(7)
O(5)ꢀMo(2)ꢀO(6)
O(5)ꢀMo(2)ꢀO(7)
O(5)ꢀMo(2)ꢀO(8)
44.52(7)
156.24(8)
130.33(7)
Figure 1. Schematic representation of the (a) [Mo2O4(μ2-O){HC(3,5-
Me2pz)3}2]2þ and (b) [Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}]
dinuclear complexes present in compounds 2 and 3, respectively. The
labeling scheme for all atoms composing the molybdenum coordination
environments is provided. Thermal ellipsoids are drawn at the 50%
probability level, and for clarity, all H atoms bonded to carbon have been
omitted. For details on the represented bond lengths and angles, see
Tables 1ꢀ3.
O(5)ꢀMo(2)ꢀO(1W) 77.70(7)
O(6)ꢀMo(2)ꢀO(7)
O(6)ꢀMo(2)ꢀO(8)
131.28(7)
88.54(7)
O(6)ꢀMo(2)ꢀO(1W) 81.90(8)
O(7)ꢀMo(2)ꢀO(8)
44.62(7)
O(7)ꢀMo(2)ꢀO(1W) 78.54(7)
O(8)ꢀMo(2)ꢀO(1W) 80.90(7)
ModO. The cis and trans internal octahedral (N,O)ꢀMoꢀ(N,
O) angles are found in the 74.4(2)ꢀ104.6(3)° and 156.8(2)ꢀ
163.2(3)° ranges, respectively (Table 2), considerably spreading
around the expected values for ideal octahedral environments.
In the asymmetrical dinuclear complex [Mo2O3(O2)2(μ2-
O)(H2O){HC(3,5-Me2pz)3}], Mo(1) has a similar environ-
ment to that described for 2, being coordinated to three N atoms
of the organic ligand, two terminal O atoms, and a μ2-bridging O
atom (Figure 1b), exhibiting a highly distorted octahedral coor-
dination environment with the Mo(1)ꢀO,N bond lengths and
(N,O)ꢀMoꢀ(N,O) angles being comparable to those of 2
(Table 3). The Mo(2) center in 3 is absolutely distinct, co-
ordinating to one oxido group, two peroxo moieties, one water
molecule, and a μ2-bridging oxido group, in an overall hepta-
coordination environment, {MoO7}, which geometrically re-
sembles a distorted pentagonal bipyramid, for which the equa-
torial plane is composed of the four O atoms from the two peroxo
groups [O(5), O(6), O(7), and O(8)], with Mo(2)ꢀO distances
in the 1.9345(16)ꢀ1.9589(17) Å range, and by the μ2-bridging
O atom, with an Mo(2)ꢀO(3) distance of 2.0259(14) Å
(Table 3 and Figure 1b). This center is displaced by ca. 0.33 Å
from the equatorial plane [defined by O(3), O(5), O(6), O(7),
and O(8)] in the direction of the apical O(4) oxido group. The
axial positions of the pentagonal bipyramid are occupied by the
latter group and by the O atom of the coordinated water
molecule, with the bond distance being 2.3152(17) Å (Table 3
and Figure 1b). The cis and trans OꢀMo(2)ꢀO angles are
found in the 44.52(7)ꢀ102.25(9)° and 131.28(7)ꢀ176.71(7)°
The molecular structures of the cationic and neutral dinuclear
complexes are depicted in Figure 1. In both complexes the MoVI
centers are bridged by an oxygen atom (μ2-bridging oxido
group), imposing Mo Mo distances of 3.7666(1), 3.7702(1),
3 3 3
and 3.7707(1) Å [average of 3.691(1) Å] and MoꢀOꢀMo
angles of 170.118(3)°, 171.347(4)°, and 173.064(3)° [average of
171.510(4)°] for the three crystallographically independent
complex residues in 2 and a Mo(1) Mo(2) distance of
3 3 3
3.6513(1) Å and Mo(1)ꢀO(3)ꢀMo(2) angle of 144.264(3)°
in 3. The shorter intermetallic distance in 3 is attributed to the
features of the coordination environment of the Mo(2) center
which is less sterically hindered. The two MoVI centers in
[Mo2O4(μ2-O){HC(3,5-Me2pz)3}2]2þ have identical distorted
octahedral coordination environments, {MoN3O3}, being con-
nected to three N atoms of one HC(3,5-Me2pz)3 ligand, two
terminal oxido groups, and a μ2-bridging oxido group (Figure 1a).
The high degree of distortion observed in the octahedral
geometries is confirmed by the bond distances and internal
octahedral angles: as expected, for all crystallographically inde-
pendent moieties, the ModO distances range between 1.664(6)
and 1.709(5) Å and are markedly shorter than the MoꢀOꢀMo
distances of the μ2-O bridges which are instead found in the
1.886(5)ꢀ1.896(5) Å range. Additionally, the lengths of the
trans MoꢀN bonds with respect to the terminal ModO bonds
[from 2.289(6) to 2.325(7) Å] are greater than those which are
trans to the μ2-O bridges [2.213(6)ꢀ2.234(6) Å], which is
probably a direct consequence of the well-known trans effect of
3493
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
Table 4. Hydrogen-Bonding Geometries
(distances in Angstroms and angles in degrees) for
[Mo2O(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] 3H2O (3)a
3
DꢀH
A
d(D A)
<(DHA)
3 3 3
3 3 3
O(1W)ꢀH(1W) O(2W)
2.812(2)
2.712(3)
2.862(2)
2.996(2)
2.778(2)
2.881(3)
2.854(3)
2.856(3)
3.435(3)
2.976(3)
168(2)
172(2)
172(2)
145(2)
152(3)
173(3)
171(3)
171(3)
142(2)
149(2)
3 3 3
O(1W)ꢀH(2W) O(3W)
3 3 3
O(2W)ꢀH(3W) O(7)i
3 3 3
O(2W)ꢀH(3W) O(8)i
3 3 3
O(2W)ꢀH(4W) O(2)
3 3 3
O(3W)ꢀH(5W) O(2W)ii
3 3 3
O(3W)ꢀH(6W) O(4W)iii
3 3 3
O(4W)ꢀH(7W) O(5)
3 3 3
O(4W)ꢀH(7W) O(6)
3 3 3
O(4W)ꢀH(8W) O(6)iv
3 3 3
a Symmetry transformations used to generate equivalent atoms: (i) x, ꢀy
þ 1/2, z ꢀ 1/2; (ii) ꢀx, ꢀy þ 1, ꢀz þ 1; (iii) x, ꢀy þ 1/2, z þ 1/2; (iv)
ꢀx, ꢀy, ꢀz þ 1.
Figure 2. Mixed ball-and-stick (cationic complexes) and space-filling
(BF4 anions) representation of the crystal packing of [Mo2O4(μ2-
O){HC(3,5-Me2pz)3}2](BF4)2 (2), viewed in perspective along the
[100] crystallographic direction. All H atoms have been omitted for
clarity.
ꢀ
weak hydrogen bonds (not shown; C O distances found
3 3 3
between ca. 3.07 and 3.19 Å), leads to voids which are occupied
ꢀ
by the BF4 counterions (Figure 2 and Figure S3 in the
Supporting Information). The crystal packing further induces
the presence of a number of weak CꢀH F contacts connect-
3 3 3
ꢀ
ing the cationic complexes to BF4 (not shown); all C
F
3 3 3
distances are in the 2.925ꢀ3.169 Å range.
Water molecules play a decisive role in the crystal packing of
compound 3. Indeed, as depicted in Figure 3a, the O atoms of the
peroxo groups [O(5), O(6), O(7), and O(8)] and the coordi-
nated water molecule [O1(W)] composing the coordination
sphere of Mo(2) are engaged in a series of strong and highly
directional OꢀH O hydrogen-bonding interactions with the
3 3 3
uncoordinated water molecules [O2(W), O3(W), and O4(W)]
(Figure 3a and Table 4). These are engaged in cooperative water-
to-water interactions [O(3W)ꢀH(5W) O(2W)ii and O-
3 3 3
(3W)ꢀH(6W) O(4W)iii; Figure 3b and Table 4] and form
3 3 3
angular clusters of the type (H2O)3. Two neighboring water
clusters interact with the coordinated water molecules to origi-
nate a cyclic cluster comprising eight water moieties, (H2O)8,
which can be systematically represented by the graph set motif
R66 (12). Recently, various reports have described distinct types
of discrete water clusters such as tetramers,13 hexamers,14
octamers,15 decamers,16 dodecamers,17 and other larger agglom-
erates.18 A number of polymeric water clusters, mainly chains
(1D clusters),19 some layers (2D clusters),20 and a handful of
frameworks (3D clusters),21 are also described in the literature.22
A remarkable feature of compound 3 concerns the fact that
Figure 3. Schematic representation of the hydrogen bonds (light-blue
dashed lines) found in [Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}]
3
3H2O (3): (a) hydrogen-bonding interactions involving all water
molecules and interconnecting neighboring complexes (atoms of the
asymmetric unit are represented in ball-and-stick mode); (b) (H2O)8
cluster, where the O atoms shown as red and orange represent the
uncoordinated and coordinated water molecules, respectively
[symmetry transformations used to generate equivalent atoms: (i) ꢀx,
ꢀy þ 1, ꢀz þ 1; (ii) ꢀx, y þ 1/2, ꢀz þ 1/2; (iii) x, ꢀy þ 1/2, z þ 1/2].
For clarity, only the H atoms associated with the water molecules are
represented. Table 4 collects all relevant details on the hydrogen-
bonding geometries.
the OꢀH O hydrogen-bonding interactions connecting each
3 3 3
individual complex to the water molecules of crystallization led to
formation of a 2D hydrogen-bonded network as shown in Figure 4a.
Layers are built up by parallel rows of individual [Mo2O3(O2)2(μ2-
O)(H2O){HC(3,5-Me2pz)3}] complexes (pointing in opposite
directions) and interconnected by the water molecules of crystal-
lization. Individual layers close pack along the [100] direction of
the unit cell (Figure 4b) with the structural cohesion of this
packing being reinforced by the presence of a number of coop-
ranges. The wider ranges for these angles arise from the presence
of the peroxo groups which contribute, at each coordinating
position, two oxygen donor atoms. The geometric features of the
Mo(2) center, including bond distances and angles, are compar-
able to those reported for related dinuclear Mo complexes.11
The crystal packing of dinuclear [Mo2O4(μ2-O){HC(3,5-
erative weak CꢀH O hydrogen bonds (not shown) involving
3 3 3
the CH groups of one layer and the O atoms (from the oxido and
peroxo groups and water molecules) of the adjacent one.
Catalytic Tests. The catalytic performance of [MoO2Cl{HC(3,5-
Me2pz)3}]BF4 (1) was investigated in the reaction of cis-cyclooctene
Me2pz)3}2]2þ cationic complexes in 2, mediated by CꢀH
O
3 3 3
3494
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
Figure 4. Ball-and-stick representation of the 2D hydrogen-bonded networks (infinite layers) involving the [Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-
Me2pz)3}] complexes and the water molecules of crystallization: (a) single layer with the complexes of adjacent rows pointing above and below the plane
of the layer; (b) perspective view along the [001] direction of the unit cell showing the packing of layers, with neighboring nets represented in different
colors. Hydrogen bonds are represented as dashed lines, and for clarity, only the H atoms of water molecules are shown.
(Cy) with tert-butyl hydroperoxide (5ꢀ6 M in decane, denoted
TBHPorg) at 55 °C using a Mo:olefin:oxidant molar ratio of
0.01:1.0:1.7. Complex 1 is not completely soluble under the reaction
conditions used, originating a yellow opaque mixture. Epoxycyclooc-
tane (CyO) was obtained as the only product in 97% yield at 24 h
reaction (Table 5 and Figure 5). Blank tests indicated that the
reaction did not take place to a significant extent in the absence of
complex (4% Cy conversion at 24 h), in the absence of oxidant (0%),
or when the ligand HC(3,5-Me2pz)3 was used instead of the MoVI
complex (3%). Hence, the simultaneous presence of the molybde-
num species and the oxidant was required for the catalytic epoxidation
to take place. On the basis of the CyO yield at 24 h these results are at
least as good if not better than those reported in the literature for
other ionic complexes that differ from 1 in the counterion or in the
tridentate organic Lewis base ligand used as catalysts in the same
reaction under similar conditions: [MoO2Cl{HC(3,5-Me2pz)3}]Cl
gave ca. 75%,8 [MoO2Cl(Bn3Me3-tame)]BF4 gave 45% (Bn3Me3-
tame = N,N0,N00-tribenzyl-1,1,1-tris(methylaminomethyl)ethane),23
and [MoO2Cl{HC(bim)3}]BF4 gave ca. 93% (HC(bim)3 = tris-
(benzimidazolyl)methane).9
As indicated above, the reaction mixture after 24 h consisted of
a solid suspended in a yellow liquid. The solid and liquid phases
were separated as described in the Experimental Section, and the
recovered beige solid (hereafter referred to as 1-TBHPorg) was
used in a second reaction run of 24 h. The kinetic curves were
essentially identical for the two runs, as were the CyO yields at 6
and 24 h (Figure 5 and Table 5). GC-MS analysis of the liquid
phase from the first run revealed the presence of free HC(3,5-
Me2pz)3. The ATR FT-IR spectrum of 1-TBHPorg exhibited
some differences in the ranges of 800ꢀ1150 and 300ꢀ400 cmꢀ1
when compared with complex 1, suggesting that the complex was
not completely stable under the reaction conditions used (Figure
S5 in the Supporting Information). As will be discussed below, these
changes can be attributed to the water sensitivity of 1 (the TBHP was
used as received and may contain up to 4% water, which means that
the H2O:Mo molar ratio could be as high as 50:1 under the reaction
conditions used) and its transformation into other species, namely,
complex 4 and possibly also complexes 2 and 3. Accordingly, when
the catalytic run was repeated but with water removed from the Cy/
Table 5. Catalytic Performance of Compounds 1ꢀ4 in the
Epoxidation of cis-Cyclooctene at 55 °C
epoxide yield at 6/24 h (%)
catalyst
oxidant
solventa
nas
run 1
run 2
1
TBHPorg
87/97
53/78
48/72
100/100
49/72
26/41
22/39
88/100
79/95
54/81
48/72
88/99
46/76
38/71
86/98
64/96
60/88
[BMIM]BF4
[BMpy]BF4
dryTBHPorg nas
TBHPaq
nas
[BMIM]BF4
[BMpy]BF4
nas
2
TBHPorg
88/99
76/99
54/83
[BMIM]BF4
[BMpy]BF4
nas
TBHPaq
TBHPorg
TBHPorg
3
4
nas
nas
[BMIM]BF4
a nas = no additional solvent.
TBHPorg system (by using molecular sieves) prior to addition of 1
and heating to 55 °C (experiment denoted as 1/Cy/dryTBHPorg)
not only was an increase in the reaction rate observed (61%/100%
CyO yield at 10 min/5 h compared with 23%/86% for 1/Cy/
TBHPorg) but also the ATR FT-IR spectrum of the recovered solid
was identical to that for 1 (not shown).
In order to explore the versatility of complex 1, similar catalytic
tests (using olefin/dryTBHPorg reaction mixtures) were carried
out for cyclic and linear olefins, namely, R-(þ)-limonene, 1-octene,
and trans-2-octene. Whereas R-(þ)-limonene and trans-2-octene
were completely converted at 6 h reaction, conversion of 1-octene
was 21%. With the exception of limonene, the epoxide was always
the only reaction product. In the case of limonene as sub-
strate, selectivity to 1,2-epoxy-p-menthane and 1,2-8,9-diepoxy-
p-menthane was 70% and 22%, respectively, and p-mentha-1,
3495
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
strongly reduced in the spectrum of 1-TBHPaq, leaving only
the fairly narrow ligand mode at 1041 cmꢀ1. Toward higher
frequencies (1200ꢀ1600 cmꢀ1) no changes occurred with
respect to the bands assigned to internal ligand modes. All of
these differences were also evident (but to a lesser degree) in the
ATR FT-IR spectrum of 1-TBHPorg (Figure S5 in the Support-
ing Information). The absorption bands observed in the range of
300ꢀ1200 cmꢀ1 in the ATR FT-IR spectrum of 4 (Figure S5 in
the Supporting Information), which was prepared by the treat-
ment of complex 1 with water, match the new bands described
above for 1-TBHPorg and 1-TBHPaq, indicating that the
changes observed in the spectra of the recovered solids (com-
pared with 1) are mainly due to the progressive formation of 4,
with the relative amount of this species increasing with the water
content of the reaction mixture.
The solid 4 was tested in the Cy reaction with TBHPorg at
55 °C (Table 5 and Figure 5). An initial induction period of at
least 30 min was observed, possibly due to the very poor
solubility of the solid in the reaction medium, and the epoxide
yields at 6/24 h (46%/76%) were much lower than those
observed for 1 (87%/97%); CyO was always the only reaction
product. The ATR FT-IR spectrum of the solid recovered after
the 24 h run (Figure S6 in the Supporting Information) was
identical to that for 4, showing that the compound is stable under
the reaction conditions used.
Figure 5. Epoxidation of cis-cyclooctene at 55 °C in the absence of
cosolvent for 1 with TBHPorg [run 1 (—) and run 2 (0)], 1 with
dryTBHPorg (2), 1 with TBHPaq ((), 2 with TBHPorg [run 1 (þ) and
run 2 ( ꢁ )], 3 with TBHPorg (O), and 4 with TBHPorg (9).
8-dien-7-ol was formed with 8% selectivity at 100% conversion.
These results further indicate regioselectivity toward the epox-
idation of the endocyclic double bond relative to the exocyclic
one. The higher reactivity of more nucleophilic substrates cor-
roborates with the mechanism proposed for dioxomolybdenum-
(VI) complexes and alkylhydroperoxides as oxidants in that the
higher electronic density of an olefinic double bond should favor
the nucleophilic attack on an electrophilic oxidizing species
formed from the reaction between the molybdenum complex
and the oxidant.24
A further catalytic test for Cy epoxidation was carried out at
55 °C with complex 1 and 1,2-dichloroethane as cosolvent.
Under these conditions (which, apart from the reaction tem-
perature, are similar to those used to prepare 2 and 3) the
reaction mixture consisted of a yellow solid in suspension. The
ATR FT-IR spectrum of the yellow solid recovered after the 24 h
run was almost identical to that for the dinuclear complex 3.
Hence, formation of complex 3 in the catalytic system 1/Cy/
TBHPorg seems likely and would explain the presence (noted
above) of free ligand in the liquid phase.
The catalytic performance of complex 1 for Cy epoxidation
was subsequently examined using 70% aqueous TBHP (TBHPaq)
as the oxidant instead of TBHPorg. The CyO yields at 6 and 24 h
were much lower than those obtained with TBHPorg (CyO was
always the only reaction product) (Table 5 and Figure 5). There
are several possible reasons for these activity differences: (i)
water may act as a coordinating solvent and inhibit the olefin
reaction; (ii) the complex may be water sensitive; (iii) TBHP
may be present in both phases of the liquidꢀliquid biphasic
system, and mass transfer limitations may exist. The ATR FT-IR
spectrum of the insoluble solid recovered after catalysis (1-
TBHPaq) exhibited several notable differences (in the ranges
of 750ꢀ1150 and 300ꢀ400 cmꢀ1) when compared with com-
plex 1, suggesting that the complex was not stable under the
reaction conditions used (Figure S5 in the Supporting Informa-
tion). Thus, the initially strong ν(ModO) band at 916 cmꢀ1 was
reduced to weak intensity, and new bands appeared at 898 (m)
and 847 (s) cmꢀ1. In the region 1000ꢀ1100 cmꢀ1 the broad
ν(BꢀF) absorption centered around 1060 cmꢀ1 for 1 was
To completely dissolve complex 1 and TBHPorg, the ionic
liquids (ILs) 1-butyl-3-methylimidazolium tetrafluoroborate
([BMIM]BF4) and N-butyl-3-methylpyridinium tetrafluorobo-
rate ([BMPy]BF4) were used as solvents. The two ILs gave
comparable catalytic results, suggesting that the cation of the IL
has only a minor influence on the Cy reaction (Figure 6 and
Table 5). CyO was always the only product, obtained in 48ꢀ53/
72ꢀ78% yield at 6/24 h reaction. These results are on a par with
those reported in the literature for [MoO2Cl(L)]Y/IL catalytic
systems: for L = HC(bim)3,9 40ꢀ80% CyO yield at 24 h; for L =
Bn3Me3-tame,23 60ꢀ79% CyO yield at 30 h, and the correspond-
ing diol was formed as a byproduct in 11ꢀ14% yield. The metal
complex/IL systems were recycled after centrifugation and
extraction of CyO with n-hexane (which is immiscible with the
IL and does not dissolve complex 1); the complete extraction of
Cy and CyO from the metal complex/IL system was confirmed
by GC, and ICP-AES analyses indicated that no measurable
amounts of molybdenum were extracted. The recovered cata-
lyst/IL mixtures were recharged with olefin and TBHPorg, and
the catalytic reactions were monitored for a further 24 h at 55 °C
(Figure 6 and Table 5). For both recovered catalyst/IL mixtures,
higher CyO yields were reached at 6/24 h in the second runs
(88ꢀ96% at 24 h), possibly due to the accumulation of TBHP in
the recycled complex/IL mixture, accounting for higher initial
oxidant:olefin and oxidant:molybdenum molar ratios. Never-
theless, the efficiency in recovering the metal complex/IL mix-
tures without loss of metal is a major issue for the excellent
catalytic performance observed.
As described in the synthesis section above, reaction of 1 in
CH2Cl2 with a large excess (150 equiv) of the oxidant TBHP
(5ꢀ6 M in decane) resulted in isolation of the dinuclear
complexes 2 and 3. The fact that species 4 was not isolated in
these experiments may be because the reactions were carried out
at room temperature with a solvent, while the catalytic reactions
that gave solids containing 4 were carried out at 55 °C without a
cosolvent, and reaction of 1 with water was carried out under
reflux. The catalytic performance of 2 and 3 for reaction of Cy
3496
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
of 1 gave single crystals of the symmetrical dioxido(μ-
oxido)molybdenum(VI) complex, [Mo2O4(μ2-O){HC(3,5-
Me2pz)3}2](BF4)2 (2), rather than the mononuclear complex
1, which is in agreement with several studies in the literature that
indicate a tendency of TpmꢀMo mononuclear complexes to
form dinuclear derivatives. When complex 1 is dissolved in
dichloromethane and treated with excess TBHP at room tem-
perature, the unusual unsymmetrical [Mo2O3(O2)2(μ2-O)-
(H2O){HC(3,5-Me2pz)3}] (3) precipitates from the reaction
mixture and complex 2 can be isolated from the liquid phase.
When used as catalysts for the liquid-phase epoxidation of cis-
cyclooctene with TBHP as the oxidant, complexes 1ꢀ3 exhibit
comparable performance, giving epoxycyclooctane as the only
product. Taking into account the conditions under which 2 and 3
were obtained (starting from 1), these dinuclear complexes may
be present in the 1/Cy/TBHP system and contribute to the
catalytic behavior. However, under the catalytic reaction condi-
tions used, transformation of 1 into the less active compound 4 is
significant, especially in the presence of water. Formation of 4 can
be avoided and the catalytic activity maximized by rigorously
drying the substrate/oxidant mixture prior to addition of the
catalyst.
Figure 6. Epoxidation of cis-cyclooctene with TBHPorg or TBHPaq
using an IL as cosolvent at 55 °C. For TBHPorg: 1/[BMIM]BF4 [run 1
(9) and run 2 (0)] and 1/[BMPy]BF4 [run 1 (2) and run 2 (Δ)]. For
TBHPaq: 1/[BMIM]BF4 ()) and 1/[BMPy]BF4 (().
with TBHPorg was similar to that for 1, with CyO being the only
product formed in 88/100% yield at 6/24 h (Table 5). Only the
initial reaction rates (based on conversions up to 1 h) were
significantly different for the three complexes: conversion at
10 min decreased in the order 1 (23%) > 2 (13%) > 3 (8%). The
reaction mixtures at 24 h consisted of a beige (for 2) or yellow
(for 3) solid suspended in a yellow liquid. The solids were
recovered (giving 2-TBHPorg and 3-TBHPorg), and 2-TBHPorg
was used in a second run, giving a kinetic curve that was
essentially coincident with that for the first run. Unlike 1 and
1-TBHPorg, the ATR FT-IR spectra of 2-TBHPorg and 3-
TBHPorg were quite similar to those for 2 and 3, respectively
(Figure S6 in the Supporting Information), indicating that the
dinuclear complexes are more stable than 1 under the reaction
conditions used. Complexes 2 and 3 were treated with water as
described above for 1, giving the solids referred to as 2-H2O and
3-H2O. The IR spectrum of 2-H2O was identical to that for 4,
while the spectrum of 3-H2O was similar to that for 3 (Figure S5
in the Supporting Information).
For complex 2 reaction of Cy with TBHPorg was also carried
out using the ILs [BMIM]BF4 or [BMPy]BF4 as solvent, which
gave CyO yields at 24 h of 95% and 81%, respectively (Table 5).
Both of the metal complex/IL systems could be recycled (as
described above for 1) and reused with no loss of catalytic
performance (Table 5). A similar catalytic test performed for
complex 4/[BMIM]BF4 gave 38/71% conversion at 6/24 h
(selectivity to CyO was always 100%); complex 4 was insoluble
intheIL(Table5). TheATRFT-IRspectrumofthesolidrecovered
from this reaction (not shown) was similar to that for 4, suggesting
that the complex is stable in the IL mixture. The slower epoxidation
reactions for the systems with ILs in comparison to those without
additional cosolvent may be partly due to mass transfer limitations.
’ EXPERIMENTAL SECTION
Materials and Methods. Microanalyses for CHN were performed
at the University of Aveiro. Quantification of molybdenum in the
samples was performed at the University of Aveiro (Lina Carvalho,
Eugꢀenio Soares) by inductively coupled plasma optical emission spec-
troscopy (ICP-OES) after acid digestion of the sample (10% experi-
mental error). Quantification of Cl was carried out using a Dionex
2000i/SP ion chromatography workstation with a conductivity detector
(C. Patinha, University of Aveiro). An isocratic elution with a NaHCO3ꢀ
Na2CO3 solution was employed together with a Dionex AS4 anion
separator column. Transmission IR spectra were recorded on a Mattson
7000 FT-IR spectrometer. Attenuated total reflectance (ATR) FT-IR
spectra were measured on the same instrument equipped with a Specac
Golden Gate Mk II ATR accessory having diamond top-plate and KRS-5
focusing lenses. FT-Raman spectra were recorded on a Bruker RFS 100
spectrometer with a Nd:YAG coherent laser (λ = 1064 nm). 1H NMR
spectra were acquired at room temperature using a Bruker CXP 300
spectrometer and referenced to TMS.
All air-sensitive preparations and manipulations were carried out
using standard Schlenk techniques under nitrogen with solvents dried by
standard procedures. The reagents 3,5-dimethylpyrazole (99%, Aldrich),
K2CO3 (Riedel de H€aen), tetrabutylammonium hydrogen sulfate (97%,
Sigma-Aldrich), and MoO2Cl2 (Aldrich) were purchased from com-
mercial sources and used as received. Thallium tetrafluoroborate was
prepared from Tl2CO3 and HBF4 in water and vacuum dried at 80 °C
prior to use. The TBHP solution (5.0ꢀ6.0 M in decane, containing up to
4% water) was used as received.
Tris(3,5-dimethyl-1-pyrazolyl)methane [HC(3,5-Me2pz)3].
The ligand HC(3,5-Me2pz)3 was prepared as described in the literature
with minor changes.3a A mixture of dimethylpyrazole (11.54 g, 0.12
mol), anhydrous K2CO3 (82.93 g, 0.60 mol), and (Bu4N)HSO4 (2.04 g,
6 mmol) was stirred and refluxed in CHCl3 (100 mL) for 19 h. The
resultant dark brown mixture was filtered, and the solid was washed with
hot CH2Cl2. The organic solution was evaporated to dryness, and the
crude product was purified by column chromatography (silica) using
diethyl ether/hexane (1:1) as eluent. After concentrating the yellow
solution, it was stored in a fridge overnight and the resultant precipitate
was filtered and washed with pentane. Yield: 4.0 g, 34%. Anal. Calcd for
C16H22N6 (298.39): C, 64.40; H, 7.43; N, 28.16. Found: C, 64.22; H,
’ CONCLUDING REMARKS
The mononuclear complex [MoO2Cl{HC(3,5-Me2pz)3}]BF4
(1) can be obtained in good yield by replacement of the
coordinated solvent molecules in [MoO2Cl(THF)3](BF4) by
the tris(3,5-dimethyl-1-pyrazolyl)methane) ligand. Recrystallization
3497
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
Table 6. Crystal and Structure Refinement Data for Com-
pounds [Mo2O4(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2) and
H, 4.04; N, 15.32; Cl, 6.46. Found: C, 35.51; H, 4.68; N, 15.00; Cl, 6.98.
Selected IR (KBr, cmꢀ1): 3139 (m), 3015 (m), 2975 (m), 2932 (m),
2871 (m), 1563 (vs), 1459 (s), 1415 (s), 1377 (s), 1303 (s), 1259 (s),
1124 (m), 1082 (sh), 1060 (vs), 1047 (s), 987 (sh), 951 (s), 916 (vs),
857 (m), 837 (m), 806 (m), 762 (s), 708 (s), 519 (s), 490 (s), 399 (m),
378 (m), 361 (s), 314 (m). Selected Raman (cmꢀ1): 3140, 3015, 2933,
1562, 1461, 1382, 1299, 1160, 1123, 1063, 1048, 1018, 952, 944, 918,
763, 597, 492, 483, 411, 401, 361, 282, 248, 208. 1H NMR (300 MHz,
DMSO-d6): δ (ppm) = 8.30 (s, 1H, CH), 5.94 (s, 3H, H-4), 2.06 (s, 9H,
CH3 in position 3), 1.94 (s, 9H, CH3 in position 5).
[Mo2O4(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2). A large excess
(ca. 150 equiv.) of TBHP (5.0ꢀ6.0 M in decane) was added to a
solution of 1 (140 mg, 0.26 mmol) in CH2Cl2 (15 mL). The resultant
yellow solution was stirred for 24 h at room temperature, during which
time a yellow solid precipitated. The mixture was filtered, and MnO2 was
added to the filtrate to destroy the excess peroxide. After filtering off the
solution it was concentrated and mixed with diethyl ether. The colorless
solid that precipitated was filtered, washed with diethyl ether and
n-pentane, and vacuum dried. Yield: 0.04 g, 30%. Anal. Calcd for
C32H44B2F8Mo2N12O5 (1042.26): C, 36.88; H, 4.26; N, 16.13. Found: C,
36.65; H, 4.60; N, 16.07. IR (KBr, cmꢀ1): 3129 (w), 2980 (sh), 2896 (w),
2778 (w), 2666 (w), 1562 (vs), 1459 (s), 1415 (m), 1379 (vs), 1303 (s),
1263 (s), 1071 (sh), 1046 (vs), 993 (w), 945 (s), 916 (vs), 859 (s), 766 (vs),
704 (m), 630 (m), 533 (m), 521 (m), 490 (m), 484 (sh), 394 (s), 323 (s).
1H NMR (300 MHz, DMSO-d6): δ (ppm) = 8.31 (s, 2H, CH), 5.95 (s, 6H,
H-4), 2.07 (s, 18H, CH3 in position 3), 1.95 (s, 18H, CH3 in position 5).
[Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] 3H2O (3)
3
2 (SQUEEZE)
3
formula
C
32H44B2F8
Mo2N12O5
C16H30N6
O12Mo2
690.32
Mr [g molꢀ1
]
1042.26
crystal description
crystal size [mm]
T [K]
colorless blocks
0.14 ꢁ 0.12 ꢁ 0.12
150(2)
yellow plates
0.20 ꢁ 0.16 ꢁ 0.06
180(2)
cryst syst
space group
a [Å]
triclinic
monoclinic
P21/c
P1
17.9727(5)
18.8894(5)
23.4086(7)
73.354(2)
75.827(2)
68.6830(10)
7004.4(3)
6
12.7169(6)
15.6739(8)
13.9182(7)
b [Å]
c [Å]
R [deg]
β [deg]
γ [deg]
V [Å3]
115.405(2)
2505.9(2)
4
Z
F
calcd [g cmꢀ3
]
1.483
1.830
F(000)
3156
1392
μ [mmꢀ1
]
0.618
1.071
[Mo2O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] 3H2O (3). The
3
θ range [deg]
3.55ꢀ25.35
ꢀ21 e h e 21
ꢀ22 e k e 22
ꢀ28 e l e 25
190 551
3.55ꢀ29.13
ꢀ17 e h e 17
ꢀ21 e k e 21
ꢀ19 e l e 18
66 423
yellow precipitate obtained in the above procedure was washed with
diethyl ether, vacuum dried, and recrystallized from 1,2-dichloroethane/
diethyl ether/pentane. Yield: 0.024 g, 27%. Anal. Calcd for C16H30Mo2-
N6O12 (690.32): C, 27.84; H, 4.38; N, 12.17. Found: C, 28.08; H, 3.55;
N, 12.12. IR (KBr, cmꢀ1): 3139 (w), 3068 (sh), 2976 (w), 2921 (w),
2848 (sh), 1564 (s), 1460 (s), 1414 (m), 1380 (w), 1302 (s), 1266 (s),
1156 (sh), 1121 (w), 1083 (w), 1046 (s), 985 (m), 954 (vs), 915 (s), 862
(m), 794 (m), 760 (m), 702 (w), 629 (w), 567 (w), 489 (w), 434 (w),
394 (w), 354 (w), 320 (m). Raman (cmꢀ1): 3142, 2929, 1463, 1382,
index ranges
no. of reflns
collected
no. of independent
reflns
25 019 (Rint = 0.0769) 6731 (Rint = 0.0357)
1
final R indices
[I > 2σ(I)]a
R1 = 0.0836
R1 = 0.0247
1304, 986, 948, 920, 597, 286, 205. H NMR (300 MHz, DMSO-d6):
δ (ppm) = 8.31 (s, 1H, CH), 5.95 (s, 3H, H-4), 2.08 (s, 9H, CH3 in
position 3), 1.95 (s, 9H, CH3 in position 5).
wR2 = 0.2531
R1 = 0.1122
wR2 = 0.0508
R1 = 0.0408
[{HC(3,5-Me2pz)3}MoO3] (4). A suspension of 1 (0.85 g, 1.55
mmol) in water (40 mL) was refluxed for 2 h. A white precipitate was
obtained, which was washed with water, acetone, and pentane and
vacuum dried. Yield: 0.35 g, 51%. Anal. Calcd for C16H22MoN6O3
(442.32): C, 43.45; H, 5.01; N, 19.00; Mo, 21.69. Found: C, 43.58; H,
5.05; N, 18.84; Mo, 22.20. IR (KBr, cmꢀ1): 3133 (m), 3103 (m), 2999
(w), 2987 (w), 2956 (m), 2921 (m), 1565 (vs), 1461 (s), 1409 (m),
1385 (s), 1304 (vs), 1260 (vs), 1115 (s), 1042 (vs), 992 (s), 899 (vs),
861 (vs), 840 (sh), 781 (s), 703 (vs), 667 (m), 631 (m), 602 (m),
490 (m), 384 (s), 366 (m), 350 (m), 328 (w), 275 (m). Raman
(cmꢀ1): 3135, 2987, 2919, 1566, 1465, 1391, 1057, 898, 862, 849,
603, 351, 242.
final R indices
(all data)a
wR2 = 0.2712
wR2 = 0.0572
largest diff peak and
3.774 and ꢀ2.835
0.461 and ꢀ0.823
hole [eÅꢀ3
]
a R1 = ∑ Fo| ꢀ |Fc /∑|Fo|. wR2 = (∑[w(Fo ꢀ Fc2)2]/∑[w(Fo )])1/2
.
2
2
7.50; N, 28.68. Selected IR (KBr, cmꢀ1): 3131 (m), 3039 (m), 2984 (s),
2956 (s), 2924 (s), 2869 (m), 1562 (vs), 1471 (s), 1449 (s), 1414 (s),
1376 (s), 1347 (s), 1320 (s), 1274 (s), 1252 (s), 1139 (s), 1031 (s), 986
(m), 971 (m), 881 (m), 862 (s), 827 (s), 778 (s), 762 (m), 709 (m), 699
(s), 624 (s). Selected Raman (cmꢀ1): 3128, 3096, 2985, 2957, 2925,
2875, 2747, 1561, 1450, 1388, 1026, 591, 217, 194. 1H NMR (300 MHz,
DMSO-d6): δ (ppm) = 8.30 (s, 1H, CH), 5.94 (s, 3H, H-4), 2.06 (s, 9H,
CH3 in position 3), 1.94 (s, 9H, CH3 in position 5).
Single-Crystal X-ray Diffraction. Crystallographic data collec-
tion and structure refinement details are summarized in Table 6. Single
crystals of [Mo2O4(μ2-O){HC(3,5-Me2pz)3}2](BF4)2 (2) and [Mo2-
O3(O2)2(μ2-O)(H2O){HC(3,5-Me2pz)3}] 3H2O (3) were harvested
3
from the crystallization vials and immediately immersed in highly
viscous FOMBLIN Y perfluoropolyether vacuum oil (LVAC 140/13,
Sigma-Aldrich) to avoid degradation caused by evaporation of the
solvent. The crystals were mounted on a Hampton Research CryoLoop
with the help of a Stemi 2000 stereomicroscope equipped with Carl
Zeiss lenses.25 Data were collected on a Bruker X8 Kappa APEX II CCD
area-detector diffractometer (Mo KR graphite-monochromated radia-
tion, λ = 0.71073 Å) controlled by the APEX2 software package26 and
equipped with an Oxford Cryosystems Series 700 cryostream monitored
[MoO2Cl{HC(3,5-Me2pz)3}]BF4 (1). One equivalent of TlBF4
was added to a solution of MoO2Cl2 (0.50 g, 2.51 mmol) in THF
(15 mL), resulting in the immediate precipitation of colorless TlCl. After
stirring the mixture for 15 min the pale yellow solution was filtered off
and treated with HC(3,5-Me2pz)3 (0.75 g, 2.51 mmol). An off-white
precipitate formed, and the mixture was stirred for 1 h at room tem-
perature. After evaporating the solvent under reduced pressure, the solid
was washed three times with diethyl ether and vacuum dried. Yield:
0.90 g, 65%. Anal. Calcd for C16H22BClF4MoN6O2 (548.58): C, 35.03;
3498
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
remotely using the software interface Cryopad.27 Images were processed
using the software package SAINTþ,28 and data were corrected for
absorption by the multiscan semiempirical method implemented in
SADABS.29 While the structure of 2 was solved using the Patterson
synthesis algorithm implemented in SHELXS-97,30 allowing the im-
mediate location of the Mo centers, that of 3 was instead solved by direct
methods using SHELXS-97,30 permitting the direct location of the
majority of the heaviest atoms. For the two structures all remaining non-
hydrogen atoms were located from difference Fourier maps calculated
from successive full-matrix least-squares refinement cycles on F2 using
SHELXL-97.30a,31 All non-hydrogen atoms were successfully refined
using anisotropic displacement parameters with the exception of the six
crystallographically independent BF4ꢀ anions in 2.
For the two structures H atoms bound to carbon were located at their
idealized positions using appropriate HFIX instructions in SHELXL: 43
for the aromatic, 13 for the tertiary carbons, and 137 for the terminal
ꢀCH3 methyl groups belonging to the HC(3,5-Me2pz)3 ligands. All
these atoms were included in subsequent refinement cycles in riding-
motion approximation with isotropic thermal displacements parameters
(Uiso) fixed at 1.2 (aromatic and tertiary) or 1.5 (ꢀCH3 groups) times
Ueq of the attached carbon atom. In 3 the hydrogen atoms associated
with water molecules (coordinated and uncoordinated) were markedly
visible in the difference Fourier maps and included in the final structure
CCDC-771365 (3) contain the supplementary crystallographic data for
this paper (excluding structure factors). These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
Catalysis. The liquid-phase epoxidation of cis-cycloctene (Cy) was
carried out at atmospheric pressure in borosilicate microreactors
equipped with a magnetic stirrer and a valve for sampling. Typically,
the reactors were loaded with the metal complex (in an amount
equivalent to 16 μmol of molybdenum), 1.6 mmol of Cy, and 2.75
mmol of oxidant, and immersed in a thermostatted oil bath at 55 °C (this
instant is taken as time zero). The oxidants used were 5.0ꢀ6.0 M tert-
butylhydroperoxide in decane, which contains up to 4% water (denoted
TBHPorg), and 70% aqueous TBHP (TBHPaq). The reactions were
carried out without adding a cosolvent or using 1,2-dichloroethane
(2 mL) or an ionic liquid, 1-butyl-3-methylimidazolium tetrafluorobo-
rate ([BMIM]BF4) or N-butyl-3-methylpyridinium tetrafluoroborate
([BMPy]BF4) (100 μL), as cosolvent. For some catalytic tests water was
removed from the Cy/TBHP solution using 4 Å molecular sieves that
had been previously activated by heating at 180 °C under vacuum for 16
h; after the molecular sieves were separated from the Cy/TBHP
solution, the metal complex was added and the reaction mixture was
immersed in the thermostatted oil bath. After a 24 h-run the solid
and liquid phases were separated via centrifugation and filtration
(0.2 μm Cronus PTFE membrane). The solid was thoroughly
washed with n-pentane and dried at room temperature overnight
prior to characterization/reuse. Catalytic reactions were monitored
by gas chromatography (V3900 GC-FID) using a capillary column
(DB-5, 30 m ꢁ 0.25 mm).
model during subsequent refinement stages with the OꢀH and H
H
3 3 3
distances restrained to 0.95(1) and 1.55(1) Å, respectively. This
procedure ensures a chemically reasonable geometry for these mol-
ecules. Atoms were treated using a riding-motion approximation with
U
iso fixed at 1.5Ueq of the parent oxygen atom.
In 2 a total of six crystallographically independent BF4ꢀ anions were
directly located in difference Fourier maps and included in the structural
model with full site occupancy and rigid geometric constraints
’ ASSOCIATED CONTENT
(BꢀF and F F tetrahedral distances constrained to common refine-
S
Supporting Information. Crystallographic data for com-
b
3 3 3
able values) in order to ensure chemically reasonable environments for
these moieties. This structure contains inner voids [the largest centered
at (1/2 0 1/2), which most likely contain highly disordered solvent
molecules (diethyl ether and n-pentane). Attempts to locate and model
these solvent molecules were unfruitful due to the considerable smeared-
out electron density located inside the cavities. In addition, the lack of
well-defined electron density peaks further promoted unstable structural
refinements even when rigid geometric constraints were employed for
the included solvent moieties. The structural reliability factors con-
verged, at this stage, to R1 [I > 2σ(I)] = 0.1471 and wR2 (all data) =
0.2886. PLATON32 estimated that the residual solvent-accessible vo-
lume encloses about 1093 Å3 (ca. 15.6% of the total volume of the
unit cell); see Figure S3 in the Supporting Information. The original data
set was mathematically treated using SQUEEZE33 in order to remove
the contribution of these highly disordered molecules. It was ultimately
estimated that the aforementioned cavities would contain ca. 527
electrons. The calculated solvent-free reflection list was used for further
structural refinements, which converged to the solvent-free structure
reported in this manuscript having the reliability factors summarized in
Table 6.
pounds 2 and 3 in CIF format, representations of the asymmetric
units of 2 and 3, crystal packing of 2, unaccounted solvent-
accessible area in the crystal structure model of 2, details of the Le
Bail whole-powder-diffraction-pattern profile fitting for 4, and
ATR FT-IR spectra. This material is available free of charge via
the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
* E-mail: atav@ua.pt (A.A.V.); igoncalves@ua.pt (I.S.G.).
’ ACKNOWLEDGMENT
We are grateful to the Fundac-~ao para a Ci^encia e a Tecnologia
(FCT), the Programa Operacional “Ci^encia e Inovac-~ao 2010”
(POCI 2010), the Orc-amento do Estado (OE), and the Fundo
Europeu de Desenvolvimento Regional (FEDER) for funding
through the projects PTDC/QUI/71198/2006 and PTDC/
QUI/65805/2006. P.N., S.S.B., and S.G. thank the FCT for
grants. We are also grateful to the FCT for financial support
toward the purchase of the single-crystal diffractometer.
The oxidation states of the molybdenum centers in 2 were further
investigated using PLATON,32 following closely the theoretical
models of Brese and O’Keeffe34 and Brown and Altermatt.35 The
sums of the bond valences at each atomic position are as follows
(considering a þ6 oxidation state for each metal center): Mo(1)
þ6.26; Mo(2) þ6.09; Mo(3) þ6.18; Mo(4) þ6.10; Mo(5) þ6.00;
Mo(6) þ6.09.
’ REFERENCES
(1) (a) Bigmore, H. R.; Lawrence, S. C.; Mountford, P.; Tredget,
C. S. Dalton Trans. 2005, 635–651. (b) Pettinari, C.; Pettinari, R. Coord.
Chem. Rev. 2005, 249, 525–543.
(2) Trofimenko, S. J. Am. Chem. Soc. 1970, 92, 5118–5126.
(3) (a) Julia, S.; del Mazo, J. M.; Avila, L.; Elguero, J. Org. Prep. Proc.
Int. 1984, 16, 299–307. (b) Reger, D. L.; Grattan, T. C.; Brown, K. J.;
Little, C. A.; Lamba, J. J. S.; Rheingold, A. L.; Sommer, R. D.
The last difference Fourier map synthesis showed the following: for 2,
the highest peak (3.774 eÅꢀ3) and deepest hole (ꢀ2.835 eÅꢀ3) located
at 0.07 and 0.12 Å from F(1) and F(20), respectively; for 3, the highest
peak (0.461 eÅꢀ3) and deepest hole (ꢀ0.823 eÅꢀ3) located at 0.72 and
0.66 Å from C(6) and Mo(2), respectively. CCDC-764428 (2) and
3499
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500

Inorganic Chemistry
ARTICLE
J. Organomet. Chem. 2000, 607, 120–128. (c) Goodman, M. S.; Bateman,
M. A. Tetrahedron Lett. 2001, 42, 5–7.
(26) APEX2, Data Collection Software, version 2.1-RC13; Bruker
AXS: Delft, The Netherlands, 2006.
(4) (a) Reger, D. L. Comments Inorg. Chem. 1999, 21, 1–28. (b)
Garcia, R.; Paulo, A.; Santos, I. Inorg. Chim. Acta 2009, 362, 4315–4327.
(5) Dilsky, S. J. Organomet. Chem. 2007, 692, 2887–2896.
(6) Dhawan, I. K.; Bruck, M. A.; Schilling, B.; Grittini, C.; Enemark,
J. H. Inorg. Chem. 1995, 34, 3801–3808.
(7) Dinoi, C.; da Silva, M. F. C. G.; Alegria, E. C. B. A.; Smoleꢀnski, P.;
Martins, L. M. D. R. S.; Poli, R.; Pombeiro, A. J. L. Eur. J. Inorg. Chem.
2010, 2415–2424.
(8) Santos, A. M.; K€uhn, F. E.; Bruus-Jensen, K.; Lucas, I.; Rom~ao,
C. C.; Herdtweck, E. J. Chem. Soc., Dalton Trans. 2001, 1332–1337.
(9) Gago, S.; Balula, S. S.; Figueiredo, S.; Lopes, A. D.; Valente, A. A.;
Pillinger, M.; Gonc-alves, I. S. Appl. Catal. A: Gen. 2010, 372, 67–72.
(10) (a) Arzoumanian, H.; Bakhtchadjian, R.; Agrifoglio, G.; Atencio,
R.; Brice~no, A. Transition Met. Chem. 2006, 31, 681–689. (b) Pereira,
C. C. L.; Balula, S. S.; Paz, F. A. A.; Valente, A. A.; Pillinger, M.;
Klinowski, J.; Gonc-alves, I. S. Inorg. Chem. 2007, 46, 8508–8510. (c)
Gago, S.; Neves, P.; Monteiro, B.; Pess^ego, M.; Lopes, A. D.; Valente,
A. A.; Paz, F. A. A.; Pillinger, M.; Moreira, J.; Silva, C. M.; Gonc-alves, I. S.
Eur. J. Inorg. Chem. 2009, 4528–4537.
(11) (a) Djordjevic, C.; Gundersen, J. L.; Jacobs, B. A.; Sinn, E.
Polyhedron 1989, 8, 541–543. (b) Grzywa, M.; Nitek, W.; Lasocha, W.
J. Mol. Struct. 2008, 888, 318–326. (c) Le Carpentier, J.-M.; Mitschler,
A.; Weiss, R. Acta Crystallogr., Sect. B: Struct. Sci. 1972, 28, 1288–1298.
(12) Bernstein, J.; Davis, R. E.; Shimoni, L.; Chang, N. L. Angew.
Chem., Int. Ed. 1995, 34, 1555–1573.
(27) Cryopad, Remote monitoring and control, version 1.451; Oxford
Cryosystems: Oxford, U.K., 2006.
(28) SAINTþ, Data Integration Engine, version 7.23a; Bruker AXS:
Madison, WI, 2005.
(29) Sheldrick, G. M. SADABS, Bruker/Siemens Area Detector
Absorption Correction Program, version 2.01; Bruker AXS: Madison,
WI, 1998.
(30) (a) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, 64, 112–122. (b) Sheldrick, G. M. SHELXS-97, Program for Crystal
Structure Solution; University of G€ottingen: G€ottingen, Germany, 1997.
(31) Sheldrick, G. M. SHELXL-97, Program for Crystal Structure
Refinement; University of G€ottingen: G€ottingen, Germany, 1997.
(32) (a) Spek, A. L. Acta Crystallogr., Sect. A: Found. Crystallogr.
1990, 46, C34. (b) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7–13.
(33) Van der Sluis, P.; Spek, A. L. Acta Crystallogr., Sect. A: Found.
Crystallogr. 1990, 46, 194–201.
(34) Brese, N. E.; O’Keeffe, M. Acta Crystallogr., Sect. B: Struct. Sci.
1991, 47, 192–197.
(35) Brown, I. D.; Altermatt, D. Acta Crystallogr., Sect. B: Struct. Sci.
1985, 41, 244–247.
(13) Lakshminarayanan, P. S.; Kumar, D. K.; Ghosh, P. Inorg. Chem.
2005, 44, 7540–7546.
(14) Zhu, W. H.; Wang, Z. M.; Gao, S. Dalton Trans. 2006, 765–768.
(15) Prasad, T. K.; Rajasekharan, M. V. Cryst. Growth Des. 2006,
6, 488–491.
(16) Barbour, L. J.; Orr, G. W.; Atwood, J. L. Nature 1998, 393,
671–673.
(17) Neogi, S.; Savitha, G.; Bharadwaj, P. K. Inorg. Chem. 2004, 43,
3771–3773.
(18) (a) Ghosh, S. K.; Bharadwaj, P. K. Inorg. Chem. 2004,
43, 6887–6889. (b) Soares-Santos, P. C. R.; Cunha-Silva, L.; Paz,
F. A. A.; Ferreira, R. A. S.; Rocha, J.; Carlos, L. D.; Nogueira, H. I. S.
Inorg. Chem. 2010, 49, 3428–3440. (c) Cunha-Silva, L.; Mafra, L.;
Ananias, D.; Carlos, L. D.; Rocha, J.; Paz, F. A. A. Chem. Mater. 2007,
19, 3527–3538.
(19) Reger, D. L.; Semeniuc, R. F.; Pettinari, C.; Luna-Giles, F.;
Smith, M. D. Cryst. Growth Des. 2006, 6, 1068–1070.
(20) Shi, F. N.; Cunha-Silva, L.; Ferreira, R. A. S.; Mafra, L.;
Trindade, T.; Carlos, L. D.; Paz, F. A. A.; Rocha, J. J. Am. Chem. Soc.
2008, 130, 150–167.
(21) (a) Huang, Y. G.; Gong, Y. Q.; Jiang, F. L.; Yuan, D. Q.; Wu,
M. Y.; Gao, Q.; Wei, W.; Hong, M. C. Cryst. Growth Des. 2007,
7, 1385–1387. (b) Kristiansson, O. Eur. J. Inorg. Chem. 2002,
2355–2361. (c) Rocha, J.; Shi, F. N.; Paz, F. A. A.; Mafra, L.; Sardo, M.;
Cunha-Silva, L.; Chisholm, J.; Ribeiro-Claro, P.; Trindade, T. Chem.—Eur.
J. 2010, 16, 7741–7749.
(22) (a) Ludwig, R. Angew. Chem., Int. Ed. 2001, 40, 1808–1827. (b)
Mascal, M.; Infantes, L.; Chisholm, J. Angew. Chem., Int. Ed. 2006,
45, 32–36.
(23) Petrovski, Z.; Valente, A. A.; Pillinger, M.; Dias, A. S.; Rodrigues,
S. S.; Rom~ao, C. C.; Gonc-alves, I. S. J. Mol. Catal. A: Chem. 2006,
249, 166–171.
(24) (a) K€uhn, F. E.; Groarke, M.; Bencze, E.; Herdtweck, E.;
Prazeres, A.; Santos, A. M.; Calhorda, M. J.; Rom~ao, C. C.; Gonc-alves,
I. S.; Lopes, A. D.; Pillinger, M. Chem.—Eur. J. 2002, 8, 2370–2381. (b)
^
Veiros, L. F.; Prazeres, A.; Costa, P. J.; Rom~ao, C. C.; K€uhn, F. E.;
Calhorda, M. J. Dalton Trans 2006, 1383–1389. (c) Mitchell, J. M.;
Finney, N. S. J. Am. Chem. Soc. 2001, 123, 862–869. (d) Groarke, M.;
Gonc-alves, I. S.; Herrmann, W. A.; K€uhn, F. E. J. Organomet. Chem. 2002,
649, 108–112.
(25) Kottke, T.; Stalke, D. J. Appl. Crystallogr. 1993, 26, 615–619.
3500
dx.doi.org/10.1021/ic102456z |Inorg. Chem. 2011, 50, 3490–3500