Molybdenum(II) Complexes with α-Diimines: Catalytic Activity in Organic and Ionic Liquid Solvents

Article abstract of DOI:10.1002/ejic.201800700

The new [MoX(η³-C₃H₅)(CO)₂(α-diimine)] complexes with: (i) X = Br or triflate and α-diimine = 1,10-phenanthroline (phen) and dipyridophenazine (dppz); and (ii) X = Br and α-diimine = phen and dppz, with several substituents, are synthesized and characterized. The structures of [MoBr(η³-C₃H₅)(CO)₂(Cl-phen)] and [Mo(CF₃SO₃)(η³-C₃H₅)(CO)₂(dppz)] are determined by using single-crystal X-ray diffraction. These and three complexes of 2,2′-bipyridyl (bpy), and its two derivatives with Me and tBu substituents, are tested in the homogeneous catalytic epoxidation of several olefins in dichloromethane, exhibiting, in general, a good selectivity towards the respective epoxide and relatively low TOFs. For the first time, the oxidation of cis-cyclooctene with some of these catalysts is also conducted in a variety of room-temperature ionic liquids (RTILs). In the presence of [MoBr(η³-C₃H₅)(CO)₂(phen)], the conversions, in general, increase, compared with the reactions in organic solvents. Interestingly, different chemoselectivity is found when [C₆mim][Ntf₂] and [C₂mim][FAP] are used with diol (24–26 %). On the other hand, [MoBr(η³-C₃H₅)(CO)₂(L)] (L = Me-phen or dppz) exhibits much lower conversions in the RTILs tested than in common organic solvents.

Full text of DOI:10.1002/ejic.201800700

A Journal of
Accepted Article
Title: Molybdenum(II) complexes with alpha-diimines: catalytic activity
in organic and ionic liquid solvents
Authors: Marta Saraiva, Carla D Nunes, Vítor Félix, Ana P. C. Ribeiro,
Carlos Nieto de Castro, and Maria Jose Calhorda
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201800700
Link to VoR: http://dx.doi.org/10.1002/ejic.201800700

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Molybdenum(II) complexes with α-diimines: catalytic activity in organic
and ionic liquid solvents
Marta S. Saraiva,[a,b] Carla D. Nunes,[a,c] Vítor Félix, Ana P. C. Ribeiro, Carlos Nieto
[d]
[c]
[
c]
[a,b]*
de Castro, Maria José Calhorda
[
a]
Dr. Marta S. Saraiva, Dr. Carla D. Nunes, Prof. Dr. Maria José Calhorda
Centro de Química e Bioquímica, DQB, Faculdade de Ciências, Universidade de
Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
E-mail: mjc@ciencias.ulisboa.pt
http://intheochem.fc.ul.pt/
Dr. Marta S. Saraiva, Prof. Dr. Maria José Calhorda
[
b]
BioISI -Biosystems & Integrative Sciences Institute, Departamento de Química e
Bioquímica, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1749-
0
16 Lisboa, Portugal
[
[
c]
Dr. Carla D. Nunes, Dr. Ana P. C. Ribeiro, Prof. Dr. Carlos Nieto de Castro
Centro de Química Estrutural, Faculdade de Ciências e Instituto Superior Técnico,
Universidade de Lisboa, 1049-001 Lisboa, Portugal
d]
Prof. Dr. Vítor Félix
Department of Chemistry, CICECO - Aveiro Institute of Materials, University of
Aveiro, 3810-193 Aveiro, Portugal
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Abstract
3
The new [MoX(η -C
3
H
5
)(CO) (-diimine)] complexes with i) X = Br or triflate and
2
-diimine = 1,10-phenanthroline (phen) and dipyridophenazine (dppz), and ii) X = Br and
-diimine = phen and dppz with several substituents were synthesized and characterized.
3
The single crystal X-ray diffraction of [MoBr( -C
3
H
5
)(CO) (Cl-phen)] and
2
3
[
Mo(CF
3
SO
3
)( -C
3
H
5
)(CO) (dppz)] was determined. These and three complexes of 2,2’-
2
bipyridyl (bpy) and its two derivatives with Me and tBu substituents were tested in the
homogeneous catalytic epoxidation of several olefins in dichloromethane, exhibiting in
general a good selectivity toward the respective epoxide and relative low TOFs. For the
first time, the oxidation of cis-cyclooctene with some of these catalysts was also conducted
in a variety of room temperature ionic liquids (RTILs). In the presence of
3
[
MoBr(η -C
3
H
5
)(CO) (phen)], the conversions in general increased compared to the
2
reactions in organic solvents. Interesting was the different chemoselectivity found when
[
[
C
6
mim][Ntf
2
] and [C
2
mim][FAP] were used, with 24-26 % of diol. On the other hand,
(L)] (L = Me-phen or dppz) exhibited much lower conversions in the
3
MoBr(η -C
3
H
5
)(CO)
2
RTILs tested than in common organic solvents.
Keywords
Catalysis; molybdenum; -diimine; epoxidation; ionic liquids
Introduction
3
Molybdenum(II) complexes [MoX(η -C
3
H
5
)(CO)
2
(L) ] (X = Cl, Br; L = monodentate (n =
n
2
) or bidentate (n = 1) ligand) have attracted much attention owing to their structural
features and reactivity, as shown by two recent reviews.[1,2] They have proven to be very
reactive for a number of catalytic reactions such as imine aziridination, [3] oxidation of
triphenylphosphine with molecular oxygen, [4] allylic alkylations, [5] polymerization of
certain dienes, [6] oligomerisation of Ph(C≡C) Ph (n = 1,2), [7] and oxidation of olefins,
n
among others. [8] This family of complexes has shown moderate to high conversions and
selectivity for the epoxide for a variety of olefins using tert-butylhydroperoxide (TBHP, in
decane) as oxidant. Many studies have been carried out in the last few years in order to
understand the mechanism of these reactions, and the identification of the active species
formed from the Mo(II) catalyst precursors. [9] It has been shown that in contact with an
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
oxidant, such as H
2
O
2
or TBHP, they suffer an oxidative decarbonylation and form several
species depending on the experimental conditions. [10] Peroxide Mo(VI) complexes are
obtained with H , but TBHP may lead to dioxo Mo(VI) complexes or oligomers. The
2
O
2
catalytic activity of the Mo(II) complexes is therefore influenced not only by the ligands in
the metal coordination sphere, but also by the reaction conditions, such as oxidant,
temperature and solvent. [10]
These homogeneous catalysts were immobilized in several silicon based materials and in
[
8]
many situations their heterogeneous catalytic activity remained the same or increased. In
recent years, another alternative to improve the properties of catalysts consisted of using
room temperature ionic liquids (RTILs) as reaction media. Their physical properties, such
as low volatility and thermal stability, combined with the possibility of adding a second
phase to the reaction medium, has led to their application in a plethora of catalytic
reactions.[11,12] Biphasic catalysis is thus an alternative approach to heterogeneous catalysis
that maintains the benefits of homogeneous catalysis, high conversion and selectivity, and
still makes the separation of products from the reaction mixture easier. The recovery and
recycling of the catalysts become also more accessible and environmental friendly. RTILs
have been tested in epoxidation reactions with different catalysts such as Fe(III)porphyrin
complexes, methyltrioxorhenium(VII) (MTO), inorganic dioxomolybdenum(VI) systems,
5
}- -CH)(CO)
]. [13,14,15,16,17,18,19,20,21,22,23] The reaction in RTILs
1
or [Mo(η -C
often resulted in higher conversions than in standard organic solvent, specifically in
-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl) imide ([C mim][Ntf ]) and 1-
butyl-3-methylimidazolium hexafluorophosphate ([C mim][PF
]).[20,24, 25] In some cases
5
H
4
{CH(CH
2
)
n
3
1
4
2
4
6
the selectivity could also be enhanced by the presence of hydrophobic ionic liquids in
contrast with hydrophilic/hygroscopic ionic liquids that promote the formation of the
diol.[13,26,27]
In this work, we synthesized and characterized a family of complexes with 2,2’-bipyridine
(bpy), 1,10-phenanthroline (phen) and dipyridophenazine (dppz) derivatives and studied
their activity as homogeneous catalysts in epoxidation reactions, in dichloromethane, in
order to understand the influence of substituents of the -diimine ligands. In the second
part, we performed the same reactions in the presence of RTILs containing cations, such as
-
-
-), hexafluorophosphate (PF -
halogens (Cl , Br ), tetrafluoroborate (BF
4
6
),
bis(trifluoromethanesulfonyl)imide (Ntf -
), trifluoroacetate (TFA), and
2
tris(pentafluoroethyl)trifluoro phosphate (FAP). The discussion of the influence of the
RTILs will shed some light in the role and influence of these potential alternative solvents,
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
which can modify the outcome of the catalytic process. For the first time, an extensive
3
catalytic study of several RTILs with [MoX(η -C
3
H
5
)(CO) (-diimine)] complexes is
2
described.
Results and discussion
Chemical studies
3
Molybdenum(II) complexes [MoX(η -C
3
H
5
)(CO)
2
(L-L)] with L-L = α-diimine were
3
] [28] with several α-diimines
synthesized from reaction of [MoBr(η -C
3
H
5
)(CO) (CH CN)
2
3
2
in 1:1 ratio in dichloromethane at room temperature: 2,2’-bipyridyl (bpy) (1), 4,4’-
dimethyl-2,2’-dipyridyl (Me-bpy) (2), 4,4’-di-tert-butyl-2,2’-dipyridyl (tBu-bpy) (3), 1,10-
phenanthroline (phen) (4), 2,9-dimethyl-phenanthroline (Me-phen) (5), 2,9-di-tert-butyl-
1
,10-phenanthroline (tBu-phen) (6), 2,9-dichloro-1,10-phenanthroline (Cl-phen) (7), 2,9-
di-trichloromethyl-1,10’-phenanthroline (tCl-phen) (8), dipyridophenazine (dppz) (9), and
2
,9-dichloro-dipyridophenazine (Cl-dppz) (10). Reaction of complex 9 with Tl[CF
3
SO ]
3
3
afforded the complex [Mo(CF
3
SO
3
)(η -C
3
H
5
)(CO) (dppz)] (11) after substitution of
2
bromide by the triflate anion, as reported for similar complexes. [29] All the complexes are
shown in Scheme 1. Complexes 1,[30] 2,[31] 3,[10] and 4 [30] were prepared earlier.
Scheme 1. Complexes 1-11.
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
1
The new complexes prepared (5, 6, 7, 8, 9, 10, and 11) were characterized by FTIR, H and
13C NMR, and elemental analysis. Complexes 7 and 11 were also structurally characterized
by single-crystal X-ray diffraction. Some of the complexes, namely those with bulkier
substituents or bearing Cl substituents (6, 8, and 10), decomposed easily.
The CO stretching modes are the most characteristic feature of the FTIR spectra of all the
complexes (Table S1). In general, the highest energy vibration (symmetric stretching) is
shifted to lower values when the two nitrile ligands of A are replaced by the -diimine
ligands. The same is observed for the second CO stretching mode in most of the
complexes. The general trend reflects increased back donation from the metal to the
carbonyls, indicating that the -diimines are weaker -acceptors and stronger -donors
than carbonyls. The main exceptions to this trend are found in complexes 7, 10, and 11. In
7
and 10, it is assigned to electronegative substituents in the -diimine ligands. The
changes in CO frequencies are more pronounced than in the previously studied family
3
[
Cp'Mo( -C
3
H
5
)(CO) ], where Cp' represents a substituted cyclopentadienyl or indenyl
2
ligand.[32] Characteristic bands of the α-diimines ligands, such as C-H (aromatic C-H
groups), C=N and C-N, slightly shifted relative to those of the free ligands are also
observed.
1
13
The H and C NMR spectra showed the characteristic signals of the -diimines as well as
the allyl groups (see Experimental and Computational studies), shifted from their values in
3
[
MoBr(η -C
3
H
5
)(CO)
2
(CH
3
CN) ]. For most complexes there are several isomers in
2
solution exhibiting a well-known fluxional behaviour.[30,31]
Crystallography
3
The molecular structures of complexes [MoBr( -C
3
H
5
)(CO) (Cl-phen)] (7),
2
3
[
Mo(CF
3
SO
3
)( -C
3
H
5
)(CO)
2
(dppz)].CH
2
Cl (11), determined by single-crystal X-ray
2
diffraction are shown in Figure 1 with the relevant atomic notation scheme adopted.
Selected bond distances and angles around the molybdenum(II) center are given in Table
S2.
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
3
3
Figure 1. Molecular structure of [MoBr( -C
)(CO) (Cl-phen)] (7), and [Mo(CF SO )( -C
thermal ellipsoids drawn at the 50% probability level.
3
H
5
)(CO)
2
(Cl-phen)] (7, top) [MoBr( -
3
C
3
H
5
2
3
3
3
H
5
)(CO) (dppz)] (11, bottom) with
2
In both complexes the coordination geometry is pseudo-octahedral with the
3
MoBr( -C
3
H
5
)(CO) moiety adopting a fac disposition, and the allyl ligand exo to two
2
cis-carbonyl ligands. In complex 7, the remaining two equatorial positions are occupied by
the Cl-phen chelating ligand with Mo-N bond lengths of 2.314(2) and 2.318(2) Å, and the
3
bromine is trans to  -C
3
H . In contrast, in complex 11, the dppz ligand fulfils an
5
equatorial and axial coordination positions with Mo-N bond lengths of 2.2674(15) and
.2136(14) Å, respectively. The CF monodentate ligand occupies the remaining
SO -
2
3
3
equatorial site with Mo-O bond length of 2.2674(13) Å. These two alternative octahedral
coordination scenarios correspond to the equatorial and the axial preferred isomers shown
in Scheme 3. [1]
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Scheme 3. The two most stable equatorial and axial isomers of complexes
3
[
MoX(η -C
3
H
5
)(CO) (L-L)].
2
There is not a clear preference for each of them, though stronger -acceptor -diimines
tend to favor the axial isomer, but the equatorial is found in many more structures than the
axial. Interestingly, no axial tungsten analogues have been reported. These aspects were
recently reviewed. [1]
In the crystal structure of 7 there are discrete dimers, in which two molecules of
3
[
MoBr(η -C
3
H
5
)(CO) (Cl-phen)] are self-assembled in a centrosymmetric arrangement via
2
C-Hphen···Br bonding interactions with C···Br distances of 3.837(3) Å and a C-H···Br
angle of 146º. In addition, as can be seen in Figure 2, the Cl-phen ligands adopt an
antiparallel disposition with the phenanthroline rings at a centroid-centroid distance of
3
.753 Å, which also suggests the existence of - stacking interactions in the crystal
packing. Similar dimers have been observed for Mo(II) derivatives of other -diimine
ligands.[1,9]
a
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
b
3
Figure 2. (a) Self-assembly of [MoBr(η -C
3
H
5
)(CO) (Cl-phen)] (7) complexes in dimers
2
by C-H···Br bonding contacts (green dashed lines) assisted by -  stacking interactions
between the phenanthroline rings (top). (b) Relevant crystal packing features of
3
[
Mo(CF
3
SO
3
)( η -C
3
H
5
)(CO)
2
(dppz)].CH
2
Cl (11) showing the - stacking between the
2
dppz extended aromatic rings and the C-H···O bonding contacts (green dashed lines,
bottom).
The crystal structure of 11 reveals a more complicated pattern in the - stacking of the
dppz ligands (Figure 2b). Indeed, as observed for 7, two molecules of [Mo(CF
3
SO
)(η3-
3
C
3
H
5
)(CO) (dppz)] form a dimer, concomitantly stabilized by -  interactions and C-
2
H···O hydrogen bonding contacts. The former ones occur between rings A and B (or A’
and B’) sited at a centroid-centroid relative distance of 3.752 Å, while the second ones
comprise two C-H···O centrosymmetric bonding contacts between C-H groups from allyl
and phen ligands and CF
3
SO
3
monodentate ligands, with C···O distances of 3.250 and
3
.334 Å respectively, and corresponding C-H···O angles of 150 and 133º. Subsequently,
two dimers are assembled by - stacking interactions with centroid-centroid longer
distances between the six-membered rings B and B’ of 3.633 Å.
Computational studies
DFT calculations[33] were performed using the ADF program[34] to find the structural
preferences for the complexes 1-11 and compare with the experimentally observed when
available. In the absence of steric constraints, the energy difference is very small, the
equatorial isomer being preferred (Table S4, Figure S1, Supporting Information). This
happens for the bpy, Me-bpy, tBu-bpy, phen, and dppz complexes 1, 2, 3, 4, and 9, with a
-1
0
.1-0.5 kcal mol difference, and has been confirmed by the single crystal X-ray structures
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
of 1 and 4.[30] The introduction of Cl or Me substituents adjacent to the nitrogen donor
atom reinforces the preference for the equatorial isomer, namely 5.39 (2), 5.99 (7), and
-1
6
.22 kcal mol (10). On the other hand, for the bulkiest tBu (6) and CCl (8) substituents,
3
only the equatorial isomer could be modeled, since it is impossible to avoid an excessive
repulsion between one of the substituents and allyl hydrogens in the axial isomer. Finally,
-1
complex 11 displays a small preference for the axial isomer (1.69 kcal mol ), reproducing
what has also been observed for the bpy and phen analogues,
3
3
)(CO)
(phen)], [35] and other
[
Mo(CF
3
SO
3
)(η -C
3
H
5
)(CO)
2
(bpy)] and [Mo(CF
3
SO
3
)(η -C
3
H
5
2
related complexes.[29] In all the examples, the substitution of Br by triflate stabilizes the
axial isomer. These preferences are reflected in the binding energies of the diamine
ligands, E
b
, defined as E
b
= Ecomplex – (E
L
+ EMo’), where Ecomplex, E and EMo’ are
L
3
the energies of the complex, the ligand and the MoBr(η -C
3
H
5
)(CO) fragment with their
2
-1
optimized geometries. These reach -40 kcal mol for 1, 2, 3, 4, 9, and 11 and drop to ~ -35
-1
(
5), -25 (7), -24 (10) kcal mol in the complexes with ortho substituents Me or Cl at the N-
N ligands when the substituents are the bulky tBu or CCl , the binding energies are
3
positive, since the ligand has to change its geometry to bind to the metal, even in the
equatorial isomer (Figure S1).
We also calculated the chemical shifts for the two isomers of complex 11. For the
equatorial isomer there are only three allylic signals (H
relative to TMS), while for the equatorial there are five (H
.42, 1.62 ppm). The experimental spectrum displays the H
m
3.14, H
4.40, H
signal at 4.25 ppm, in good
s
2.87, and H
a
1.41 ppm
m
s
5.60, 3.26, and H
a
1
m
agreement with the 4.40 calculated for the axial isomer. The other signals observed could
be assigned to the syn and two anti protons. The pattern corresponding to the equatorial
isomer does not fit the experimental spectrum, and there is no evidence of fluxional
behavior.
Catalytic studies. Epoxidation in conventional organic solvents (dichloromethane).
Two sets of substrates were studied. The first one comprises simple olefins, such as cis-
cyclooctene (cy8), styrene (sty) and 1-octene (1-oct), and the second one multifunctional
olefins, namely, cis-hex-3-en-1-ol (cis), trans-hex-2-en-1-ol (trans) and R-(+)-limonene
(R-lim).
3
The complexes 1-11, as well as [Mo(CF
3
SO
3
)(η -C
3
H
5
)(CO) (bpy)] (12) and
2
3
)(CO)
(phen)] (13),[35] were tested as catalysts in the epoxidation of
[
Mo(CF
3
SO
3
)(η -C
3
H
5
2
cis-cyclooctene (cy8) and styrene (sty), and complexes 4-9, 11-13 in the epoxidation of
9
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
cis-hex-3-en-1-ol (cis), trans-hex-2-en-1-ol (trans), and R(+)-limonene (R-lim). The
catalytic activity of complexes of bpy and its substituted analogues (1-3) was reported
before[10] but they were tested in the same conditions as 1-13 for comparison. All
experiments were performed at 328 K with dichloromethane as solvent, dibutyl ether
(DBE) as the internal standard and tert-butylhydroperoxide (TBHP) as the oxidant. The
conversion and selectivity are presented in Table 2 for cis-cyclooctene (cy8), styrene (sty),
cis-hex-3-en-1-ol (cis), and trans-hex-2-en-1-ol (trans), and in Table 3 for R-(+)-limonene
(
R-lim). Blank runs (without catalyst and in the presence of oxidizing agent) gave virtually
no conversion of starting material. The oxidation products detected in all the catalytic runs
see also next section) are shown in Scheme 4.
(
Scheme 4. Substrates studied and oxidation products detected.
1
0
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Table 2. Olefin epoxidation of cis-cyclooctene (cy8), and styrene (sty) catalyzed by
complexes 1-13, and of cis-hex-3-en-1-ol (cis) and trans-hex-2-en-1-ol (trans) by
complexes 4-9.[a]
cy8
sty
Entry
Conv. %
b]
Select. %
[c]
Conv
%[b]
94
90
23
98
97
99
92
14
99
80
6
Select. %
Complex
TOF[d]
TOF[d]
[
[c]
68
44
27
52
61
92
40
78
61
2
1
2
3
4
5
6
7
8
9
1
2
35
34
100
100
100
100
100
100
100
100
100
100
100
100
100
4
4
127
3
3
90
121
2
6
4
18
139
319
306
42
4
5
67
47
16
278
427
25
176
52
51
75
6
89
7
99
8
99
9
96
397
90
1
1
1
1
1
0
1
2
3
10
11
12
13
100[e]
43
1
48
1
5
1
61
8
5
1
cis
trans
Select. %
Entry
Conv. %
Select. %
[c]
Conv
%[b]
90
Complex
TOF[d]
TOF[d]
[
b]
[c]
100
93
1
1
1
1
1
1
4
5
6
7
8
9
4
5
6
7
8
9
53
63
71
88
89
5
100
284
3
0
13
10
140
65
2
100
83
100
3
94
92
100
86
48
0
99
100
100
83
100
99
100
47
[
(
a] All reactions were carried in dichloromethane in the presence of 2 eq. of oxidant
TBHP) and 1 mol % of catalyst at 328 K; [b] Calculated after 24 h of reaction; [c]
d
Calculated as ‘’Yield of epoxide’’/ ‘’Conversion’’x 100 % after 24 h 95 % at 30 min; [d]
Calculated at 0.17 h as mol molMo h ; [e] at 30 min.
-1
-1
1
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
The only formed product for the epoxidation of cy8 and cis, with these complexes as
catalysts, was the respective epoxide. For sty, the epoxide and benzaldehyde were
observed, and for trans the products detected were the epoxide and the ketone.
All studied complexes showed catalytic activity in the epoxidation of cy8 and were 100 %
selective toward the epoxide. The highest conversions (99-100 %) were obtained with the
catalysts containing Cl substituents, namely complexes 7 (Cl-phen), 8 (tCl-phen), and 10
(
Cl-dppz). In particular complex 10 presented a conversion of 100 % after only 30 min of
reaction. Complexes 9 (dppz), 3 (tBu-bpy), and 8 (tCl-phen) reached conversions of 89 –
6 % and the others were modest. In a related work, complexes 1 and 3 catalyzed the
9
epoxidation of cy8 in the presence of TBHP with 100 % selectivity and conversions of 19
and 59 %, respectively, after 24 hours at 343 K in acetonitrile.[10] These values are lower
than those in Table 2 (entries 1 and 3), probably due to the different conditions (solvent
dichloromethane and temperature 328 K in this work).
The bulkier substituents led to high conversions (entries 3, 6, 8), which is interesting since
they occupy the 4,4’ positions of bpy and the 2,9 positions of phen. The bpy complex has
higher activity that the phen one (entries 1 and 4), but the trend is reversed when
substituents are introduced (entries 2, 3 vs. 5, 6). The dppz complexes are particularly
active (entries 9, 10), the unsubstituted complex 9 (dppz) being much more active than the
analogous 1 (bpy) and 4 (phen). The substitution of the halogen Br by the triflate ligand
(CF
3
SO
3
) resulted in an increase of the catalytic activity of the bpy (35 % for 1, 48 % for
1
1) and phen complexes (18 % for 4, 61 % for 13), and a significant decrease for the dppz
one (96 % for 9, 43 % for 11).
The best catalyst for the epoxidation of styrene is complex 6 (tBu-phen) with a conversion
of 99 % and a selectivity of 92 % for the epoxide. The catalysts 4, 5, 6, 7, and 9 convert 92
-
99 % after 24 h of reaction, but the selectivity remains below 62 % for all of them except
6
. The complexes with the bulky tBu and tCl substituents in the ligands, tBu-phen (6), and
tCl-phen (8), display the highest selectivity for the epoxide. The triflate complexes 11-13
showed very little activity for styrene epoxidation and no activity at all for epoxidation of
cis and trans, indicating that the substitution of Br by CF
3
SO reduces the catalytic activity
3
of the complexes.
Complexes 4-9 are very selective in the epoxidation of cis, but the conversions are
relatively low, reaching 88 – 89 % for the Cl containing catalysts 7 and 8, and being almost
null for 9 (5 %). The epoxidation of trans is achieved with slightly lower selectivities
1
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
(down to 93, 92, and 83 % in entries 15, 16, 19 and 100 % for the others) but better
conversions. These are higher than 90 % for all complexes except 5 and 9 (Table 2).
The turnover frequencies (TOF) after 10 min reaction time show that the initial catalytic
activity covers a wide range of values. Apart from some exceptions, the reactions are
initially slow, but after 24 h quite good conversions are achieved, as discussed above.
Despite this, for cyclooctene (cy8) epoxidation, for instance, TOFs ranging between 2 and
27 mol molMo h-1 were reported for all the complexes after 10 minutes reaction. In the
-1
4
case of the epoxidation of cyclooctene the higher TOF values were observed for complexes
, 8 and 10. The TOF values were, respectively, 427 mol molMo h-1 for complex 8, 278
-
1
7
mol molMo h-1 and 176 mol molMo h-1 for complexes 7 and 10 (Table 2, entries 8, 7, and
0). These values were similar to those found in the literature for related Mo(II)
complexes with different ligands. [36] There are chlorine substituents in complexes 7, 8 and
0 . Their  donation capability may help to stabilize the active Mo(VI) species and make
its formation faster. Similar trends were obtained for the epoxidation of cis-hex-3-en-1-ol
cis), and trans-hex-2-en-1-ol (trans), with complexes 7 and 8 presenting the best turnover
-1
-1
1
1
(
-1
-1
frequencies. For the epoxidation of cis they reached 86 and 48 mol molMo h ,
respectively, for 7 and 8, while in the epoxidation of trans the observed TOF values were
40 mol molMo h-1 and 65 mol molMo h-1. Since the Mo(II) complexes are initially
-
1
-1
1
[
8]
oxidized to active Mo(VI) catalytic species, losing the carbonyl ligands, the weakening
of the Mo-C bond, inferred from IR data, may speed up their formation and give rise to
high TOFs.
We may also consider the nature of the active species. Gamelas et [10] studied the nature of
the oxidized species derived from the chloride analogues of bpy complexes 1-3, and
showed that it depends on the oxidant and the substituents. With TBHP in decane, a dimer
[
{MoO
L)][MoO
catalytic reactions, complex 1 should have been converted to a dimer [{MoO
2
Cl(L-L)}
2
(-O)] forms, while TBHP in water leads to {[MoO
(R = H, 1) or [Mo ] (R = tBu, 3). In the conditions of our
Br(L-L)}
]. In another work, we showed that the 1,4-(2,6-
dimethyl)phenyl-2,3-dimethyldiazabutadiene complex was oxidized to [{MoO
O) ]. Reactions of complexes 4, 5, and 9 with TBHP (3 days) were studied and a
comparison of FTIR spectra suggested that 4 and 9 could have been converted in
MoO(O (L-L)], while 5 appeared to convert into [Mo 24(Me-phen) ]. The formation of
different active species may account for the different initial reaction rates (and TOFs). On
3
(L-
3
(H O)]}
2
n
8
O24(L-L)
4
2
2
(-
[
9]
O)] or [{MoO
2
(L-L)}
2
(-O)
2
2
(L-L)} (-
2
2
[
2
)
2
8
O
4
1
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
the other hand, the complexes with ortho substituents do not form very stable allylic
complexes, as we showed above (Computational studies), some being viable only as the
equatorial isomer. Therefore they should react faster with TBHP to form the less crowded
dimer [{MoO
2
(L-L)}
2
(-O) ].
2
In the case of styrene epoxidation, the observed values of turnover frequencies were
completely different, suggesting that the catalytic reaction is ruled by different parameters.
The highest TOFs of 397 mol molMo h-1, 319 mol molMo h-1 and 306 mol molMo h-1
-
1
-1
-1
were obtained with complexes 9, 5 and 6, respectively, being higher than those reported for
3
CN)
] by some of us previously. [36]
[
MoBr(η -C
3
H
5
)(CO)
2
(CH
3
2
The epoxidation of R-(+)-limonene (R-lim) was catalyzed with high conversions (99 – 100
) by catalysts 7, 8, and 9 (Table 3, entries 4-6), the phen complex 4 being almost
%
inactive. Two main products, Z-Limox and E-Limox, were formed in the presence of
complexes 4-9. While complex 4 led to a racemic mixture, 5 (Me-phen) was very selective
toward Z-Limox (91 %), and 8 (tCl-phen) toward E-Limox (90 %). Complexes 11-13 led
to conversions below 50 %, showing a high selectivity (89 % or higher) toward
dihydrocarvone.
3
Table 3. Epoxidation of R(+)-limonene (R-lim) catalyzed by [MoBr(η -C
3
H
5
)(CO) (L)]
2
3
complexes 4-9 and [Mo(CF
3
SO
3
)(η -C
3
H
5
)(CO)
2
(L)] (11-13). [a]
Select. %
Select %
Select %
Entry Complex Conv.%[b]
TOF[d]
Z-limox[c]
E-limox[c]
dihydrocarvone[c]
1
2
3
4
5
6
7
8
9
4
5
9
71
45
91
4
50
1
0
0
1
3
6
56
79
85
90
9
0
68
391
194
19
9
7
99[e]
99
15
10
59
0
0
8
0
9
100
46
0
11
12
13
0
89
93
100
21
0
0
15
0
31
0
0
[
a] All reactions were carried in dichloromethane in the presence of 2 eq. of
oxidant (TBHP) and 1 mol % of catalyst at 328 K; [b] Calculated after 24 h of
reaction; [c] Calculated as ‘’Yield of product’’/ ‘’Conversion’’x100 % after 24
-1
-1
.
h; [d] Calculated at 0.17 h as mol molMo h ; [e] 99 % at 1 h
1
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Catalytic studies. Epoxidation in ionic liquids.
Different ionic liquids were used as solvents in the epoxidation of cis-cyclooctene, in order
to assess their influence on the catalytic reaction, as well as to take advantage of the
biphasic features of the system. The ionic liquids tested consist of different families, based
on imidazolium, pyridinium, pyrrolidinium, piperidinium and phosphonium cations,
-
-
-),
associated with several anions, such as halogens (Cl , Br ), tetrafluoroborate (BF
hexafluorophosphate (PF ), trifluoroacetate
-), bis(trifluoromethanesulfonyl)imide (Ntf -
TFA), and tris(pentafluoroethyl)trifluoro phosphate (FAP), as shown in Figure 3. The
4
6
2
(
length of the aliphatic chain in the cation was varied.
Figure 3. Schematic structure of the ionic liquids used.
1
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Complex 4 was the catalyst chosen for the preliminary catalytic studies (Scheme 1),
despite its low catalytic activity, owing to being cheap, easy to synthesize, and more stable
than most of the other complexes tested. The complex was dissolved in the ionic liquid,
heated to 55 ºC, followed by the addition of TBHP, the internal standard, and the substrate
(see details in Experimental Section). With most of the ionic liquids tested, two phases
were observed; the lower phase consisted of the ionic liquid, the catalyst and the oxidant,
while the upper phase contained the substrate, the internal standard and the product. The
upper layer remained colourless throughout the reaction, while the lower phase gained a
yellow coloration due to the formation of the Mo(VI) active species. This indicates that the
reactions occurred as heterogeneous catalysis because the catalyst remained in a different
phase. The results of the catalysis are presented in Table 4.
Most of the ionic liquids tested maintain or reduce the conversion observed after 24 h,
when compared with the conversion obtained in dichloromethane (see also Figure S3 in
Supporting Information). The ionic liquid, which leads to the highest conversions, is 1-
butyl-1-methylpiperidinium bis(trifluoromethylsulfonyl)imide ([C
4
mpip][Ntf ]), Table 3,
2
entry 19), with conversions of 45 % and 34 % for the first and second runs, respectively,
much higher than the conversions in organic solvents (11- 20 %, entries 1-3), and being
still 100 % selective toward the epoxide. The same cannot be said for [C
2
mim][Ntf ],
2
[
C
6
mim][Ntf ], [C mim][BF ] and [C mim][FAP] (entries 4, 6, 10, 13), since in their
2
4
4
2
presence secondary products, such as the respective ketone, diol, or both (Scheme 4), were
observed. The formation of cyclooctane-1,2-diol may be promoted by the presence of
water in the catalytic reaction, which can be indicative that [C
6
mim][Ntf ], and
2
[
C
2
mim][FAP] have higher water concentration than the other RTILs studied.
1
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Table 4. Olefin epoxidation of cis-cyclooctene (cy8) catalyzed by
3
[
MoBr(η -C
3
H
5
)(CO) (phen)] (4) in a number of different organic and ionic liquids as
2
solvents. [a]
Conv. %
b]
Epoxide
%[c]
Ketone
%[c]
0
Diol
Entry
Solvent
TOF[d]
[
%[c]
0
1
2
3
4
5
6
7
8
9
CH
Acetone
CHCl
2
Cl
2
18
11
100
2
2
100
0
0
3
20
100
0
0
8
[C
[C
[C
[C
2
4
6
8
mim][Ntf
mim][Ntf
mim][Ntf
mim][Ntf
2
2
2
2
]
]
]
]
30/32[e]
9/4
92/96
8/4
0/0
7/4
0/0
0/0
0/0
17/12
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
24/18
0/0
0/0
0/0
0/0
0/0
0/0
26/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
0/0
2/5
2/2
2/4
2/7
2/6
3/3
2/0
3/15
6/2
2/7
2/1
1/0
3/2
5/5
4/4
4/8
0
100/100
69/75
23/47
16/12
14/17
18/4
100/100
100/100
100/100
83/88
[C10mim][Ntf
2
]
]
[C12mim][Ntf
2
1
1
1
1
1
1
1
1
1
1
2
2
0
1
2
3
4
5
6
7
8
9
0
1
[C
4
mim][BF
4
]
18/17
9/16
[C
[C
4
mim][PF
mim][PF
6
]
]
100/100
100/100
74/100
100/100
100/100
100/100
100/100
100/100
100/100
100/100
100/100
6
6
12/7
[C
2
mim][FAP]
mim][FAP]
38/24
12/18
11/9
[C
4
[C
4
mim][TFA]
dmim][Br]
py][Ntf
pyr][Ntf
mpip][Ntf
29)][Cl]
29)][Ntf
[C
4
15/17
23/35
34/30
45/34
14/12
17/15
[C
[C
[C
4
2
]
4
2
]
4
2
]
[P(C
6
H
13
13
)
3
(C14
H
[P(C
6
H
)
3
(C14
H
2
]
0
[
(
a]All reactions were carried in 0.5 ml of ionic liquid in the presence of 2 eq. of oxidant
TBHP) and 1 mol % of catalyst at 328 K; [b] Calculated after 24 h of reaction; [c]
Calculated as ‘’Yield of product’’/ ‘’Conversion’’ x 100 % after 24 h; [d] Calculated at
-1
-1
0
.17 h as mol molMo h ; [e] Two cycles were performed of all the reactions in ionic
liquids, by separation of the ionic liquid/catalyst phase through addition of diethyl ether.
+
The cation 1-butyl-1methylimidazolium ([C
4
mim] ) was associated with five different
-, BF
-, PF
,TFA , and FAP . A comparison of the results of the reactions
-
-
-
anions Ntf
2
4
6
1
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
catalyzed by complex 4 in the presence of these five RTILs (Figure 4) with those in
organic solvents /Table 3) reveals a decrease of the conversions, except for [C mim][BF
but the selectivity for the epoxide remained 100 %. For the latter RTIL, the conversion is
maintained (18 % and 17 % in runs 1 and 2, respectively, compared to 18 % in CH Cl ),
however the selectivity decreases with the formation of the respective ketone, which can be
explained by the hydrophilic / hygroscopic nature of the [C mim][BF
] ionic liquid.[13]
4
4
],
2
2
4
4
Figure 4. Conversion of cis-cyclooctene into its epoxide with complex 4 in [C4mim][anion].
Ionic liquids with cations such as 1-butylpyridinium, 1-butylpyrrolidinium, 1-butyl-1-
methylpiperidinium, associated with the Ntf -
anion were also tested (Figure 5), as well as
2
-
-
tris(n-hexyl)-n-tetradecylphosphonium with Cl and Ntf anions (Table 3, entries 17-21).
2
Figure 5. Conversion of cis-cyclooctene into its epoxide with complex 4 in [C
4
cation]
[
Ntf ].
2
1
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
The selectivity toward the epoxide remained 100 % and, for [C
mpip][Ntf ], the conversions are higher than in organic solvents (Table 3). No specific
trend can be associated with the change in length of the aliphatic chain (C , C , C , C
and C12) for the imidazolium family of ionic liquids with the Ntf -
anion (Figure 6).
4
pyr][Ntf
2
] and
[
C
4
2
2
4
6
8
, C10
2
Figure 6. Conversion of cis-cyclooctene into its epoxide with complex 4 in [cation][Ntf ].
2
The most interesting result in the epoxidation of cy8 is the formation of diol (24 % and 18
in the first and second runs) when using [C mim][Ntf ] (Table 3, entry 6), which suggest
%
6
2
that a suitable choice of ionic liquid might help to tune the reactivity of the catalyst. On the
other hand, the selectivity is low, as three products are formed. The TOF values in
cyclooctene epoxidation catalyzed by complex 4 are not sensitive to the solvent nature
(normal or ionic), as shown in Table 4.
The increase in the alkyl chain within the same anion thus promotes different results
Figure 6). However, the effect of changing the anion from Ntf -
to FAP is not the same
-
(
2
+
+
when the cation is [C
2
mim] or [C mim] (Figure 7). For both anions, the increase in the
4
alkyl chain from 2 to 4 carbon atoms lowers the conversion, in runs 1 and 2. Interestingly,
a direct comparison of [C mim]Ntf ] and [C mim][FAP] shows that the conversion
2
2
4
increased from the first to the second cycle. Further studies are required, but viscosity may
play an important role in forming the active species from the pre-catalyst 4.
1
9
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Figure 7. Comparison of the conversion of cis-cyclooctene into its epoxide with complex 4
in [C cation] [anionNtf
], with N = 2, 4 and anion = Ntf - and FAP-.
n
2
2
In order to obtain more information about the role of the catalyst, complexes
3
3
[
MoBr(η -C
were tested as catalysts in the same reaction conditions in the presence of selected RTILs
Table 5).
3
H
5
)(CO)
2
(Me-phen)] (5) and [MoBr(η -C
3
H
5
)(CO)
2
(dppz)] (9) (Scheme 1)
(
Table 5. Olefin epoxidation of cis-cyclooctene (cy8) catalyzed by
3
[
MoBr(η -C
3
H
5
)(CO) (L)] (L = Me-phen (5), dppz (9)) in dichloromethane and ionic
2
liquids as solvents.[a]
Complex
Solvent
CH Cl
mim][Ntf
py][Ntf
mpip][Ntf
CH Cl
mim][Ntf
Conv. %[b]
Epoxide %[c]
100
TOF[d]
2
2
67
35
47
5
[
C
2
2
]
100
5
[
C
4
2
]
22/37[e]
20/36
96
100/100
100/100
100
0
[
C
4
2
]
0
2
2
25
0
9
[
C
2
2
]
6
100
[
(
a] All reactions were carried in 0.5 ml of ionic liquid in the presence of 2 eq. of oxidant
TBHP) and 1 mol % of catalyst at 328 K; [b] Calculated after 24 h of reaction; [c]c
Calculated as ‘’Yield of product’’/ ‘’Conversion’’x100 % after 24 h; [d] Calculated at
-1
-1
0
.17h as mol molMo h ; [e] Two cycles were performed of all the reactions in ionic
liquids, by separation of the ionic liquid/catalyst phase through addition of diethyl ether.
The performance of these catalysts in the epoxidation of cis-cyclooctene is much better in
CH
2
Cl
2
, with higher conversions than in the tested ionic liquids, [C
2
mim][Ntf
2
],
[
C
4
py][Ntf
2
] and [C mpip][Ntf ], contrary to what was observed with complex 4 (see also
4
2
2
0
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Figure S3 in Supporting Information). Instead of enhancing the catalytic activity of the
complex, as happened with 4, their presence seems to stop the catalytic reaction and the
conversions decrease significantly, even though the selectivity is again 100 % for the
epoxide. These results show that the role of an ionic liquid depends on many factors and
further studies must be undertaken to shed light on this problem. Also, TOF values became
lower when the solvent changed from CH
2
Cl to an ionic liquid (Table 5).
2
The mechanism to obtain the Mo(VI) active species from the Mo(II) precursor and how it
catalyzes olefin epoxidation in the presence of TBHP has been proposed before.[8,9] The
nature of the oxidized species has been shown to vary significantly with the solvent, with
TBHP in decane forming dimeric species but in water affording oligo- or polymeric
species, as mentioned above.[10] Although many further studies will have to be performed,
the strong electrostatic interactions introduced by RTILs will certainly interfere, as has
been shown in other catalytic reactions.[37]
Conclusions
3
A family of [MoX(η -C
H
3 5
)(CO)
2
(-diimine)] complexes was obtained for X = Br and
- or 4,4’-tBu -bpy, 2,9-Me -, 2,9-tBu -, 2,9-
-dppz, and X = triflate and bpy, phen or dppz, providing a
three -diimines (phen, dppz, bpy), 4,4’-Me
2
2
2
2
Cl -, or 2,9-(CCl -, and 2,9-Cl
2
3
)
2
2
variety of steric and electronic effects. Their effect was tested in the homogeneous catalytic
epoxidation of cis-cyclooctene and styrene (sty) in dichloromethane. The highest
conversions of cis-cyclooctene (100 % epoxide formed) and TOFs were associated with
bulky substituents in the catalyst, either in 4,4’ positions of bpy or in 2,9 positions of phen,
and presence of chloride (Cl or CCl
3
). The oxidation of styrene is less selective
(benzaldehyde formed besides the epoxide), conversions are lower and the best TOF is
achieved with the dppz complex. For these two substrates, and also cis-hex-3-en-1-ol,
trans-hex-2-en-1-ol and R-(+)-limonene, the substitution of bromide by triflate quenches
the catalysis (very low conversions, almost zero TOFs). The Mo(II) complexes are
catalysis precursors, which are oxidized by TBHP affording Mo(VI) oxide active species,
which may differ depending on conditions and dinitrogen ligand. Probably the steric
hindrance of the 2,9 substituents of phen and dppz contributes to accelerate this reaction, as
it is more relevant in the Mo(II) than in the Mo(VI) species. DFT calculations were
performed to have a better idea about the most stable isomer for each complex. In the
2
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
search for more efficient and environment friendly catalysts, the epoxidation reaction of
cis-cyclooctene was chosen to analyse the effect of using ionic liquids instead of organic
solvents with three of the previous catalysts. The reaction with
3
[
MoBr(η -C
3
H
5
)(CO)
2
(phen)] afforded better conversions in some of the RTILs. The loss
mim][Ntf ] and [C mim][FAP] is interesting in the
of chemoselectivity observed in [C
6
2
2
sense that they open a route for the diol. However, although the TOF values are very low
in all cases, these not yet conclusive results seem promising.
Experimental
General Considerations. All reagents, including 2,2’-dipyridyl (bpy), 4,4’-dimethyl-2,2’-
dipyridyl (Me-bpy), 4,4’-di-tert-butyl-2,2’-dipyridyl (tBu-bpy), 1,10-phenanthroline
(phen), and 2,9-dimethyl-1,10-phenanthroline (Me-phen) were obtained from Sigma-
Aldrich and used as received. The ligands 2,9-di-tert-butyl-1,10-phenanthroline (tBu-phen)
[38]
,9-dichloro-1,10-phenanthroline (Cl-phen),[39] 2,9-di-trichloromethyl-1,10’-
2
phenanthroline (tCl-phen),[40] dipyridophenazine (dppz),[41] and 2,9-dichloro-
dipyridophenazine (Cl-dppz)[39] were prepared according to literature procedures. The
ionic liquids used were synthesized following reported methods,[42,43] except for the FAP
RTILs that were kindly given by Merck. The solvents used in the complexation reactions
were dried and deoxygenated using normalized procedures (CH
CaH , diethyl ether and hexane over Na/benzophenone), distilled under nitrogen, and kept
over 4 Å molecular sieves (3 Å for acetonitrile).
The molybdenum complexes were prepared using standard Schlenk techniques under
2
Cl
2
and CH
3
CN over
2
3
],[28] [MoBr(η -C
3
nitrogen. Complexes [MoBr(η -C
3
H
5
)(CO)
2
(CH
3
CN)
2
3
H
5
)(CO)
2
(L)] (L =
bpy,[30] Me-bpy,[31]tBu-bpy,[10] phen)[30] and [Mo(CF
SO
)(η -C
3
H
5
)(CO)
(L-L)] (L-L = -
3
3
3
2
diimines) [29,35] were prepared according to the literature. They were characterized by
1
13
FTIR, H and C NMR and elemental analyses.
Infrared spectra were obtained on a Nicolet 6700 FTIR spectrophotometer as KBr pellets.
1
13
H and C NMR spectra were obtained with a Bruker Avance-400 spectrometer. Chemical
shifts are quoted in ppm from TMS. Microanalyses for C, H, and N were performed at
CACTI, University of Vigo.
2
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Syntheses
3
[
MoBr(η -C
3
H
5
)(CO) (L)] (5, 6, 7, 8, 9, 10). A solution of the ligand (1.2 mmol) in
2
3
dichloromethane was added to a stirred solution of [MoBr(η -C
3
H
5
)(CO)
2
(CH
3
CN) ] (1.0
2
mmol, 0.355 g) in dichloromethane. The mixture was left stirring in inert atmosphere at
room temperature for 3 hours. The precipitate formed was filtered, washed with diethyl
ether and dried under vacuum.
3
[
MoBr(η -C
3
H
5
)(CO) (2,9-dimethyl-1,10-phenanthroline)] (5)
2
Yield: 410 mg (85 %); elemental analysis calcd (%) for MoBrN
2
O
2
19
C H17 (481.21): C
4
7.42, H 3.56, N 5.82; found: C 47.01, H 3.67, N 5.90.
-1
FTIR (KBr, /cm ): 3042 (m), 2972 (w), 2968 (w), 2923 (w), 1936 (s), 1865 (s), 1626 (w),
587 (m), 1502 (s), 1444 (s), 1376 (m), 1158 (s), 1023 (m), 861 (s), 796 (s), 731 (s), 637
m), 554 (s).
1H NMR (400.10 MHz, CDCl
.31 (s, 6H, H /H
d, 2H, H /H
13C NMR (100.25 MHz, CDCl
1
(
3
3
3
, t.a., δ ppm): 1.29 (d, 2H,  -C
3
H
5
), 3.22 (m, 2H,  -C
3
5
H ),
3
3
1
8
), 3.39 (d, 1H, ƞ -C
3
H
5
), 7.70 (d, 2H, H
3
/H
6
), 7.83 (s, 2H, H
4
/H ), 8.28
5
(
2
7
)
3
3
, t.a., δ ppm): 29.4 (C
1
/C
8
), 51.6 ( -C
3
5
H ), 126.3
(C
3
/C
4
/C
5
/C
6
), 137.7 (C
2
/C ).
7
3
[
MoBr(η -C
3
H
5
)(CO) (2,9-di-tert-butyl-1,10-phenanthroline)] (6)
2
Yield: 410 mg (72 %).
-1
FTIR (KBr, /cm ): 3231 (w), 3067 (w), 2966 (w), 2931 (w), 2887 (w), 1929 (s), 1851 (s),
1
636 (s), 1605 (s), 1393 (m).
1H NMR (400.10 MHz, CDCl
, t.a., δ ppm): 1.27 (d, 2H, ƞ -C
3
H
), 3.31 (s, 18H, H1/H8),
), 8.33 (s, 4H,
3
3
5
3
3
3
3
.42 (d, 2H,  -C
3
H
5
), 3.75 (m, 1H,  -C
/H ).
3
H
5
), 3.92 (m, 1H,  -C
3
H
5
H
3
/H
4
/H
5
/H
6
), 9.06 (s, 2H, H
2
7
3
[
MoBr(η -C
3
H
5
)(CO) (2,9-dichloro-1,10-phenanthroline)] (7)
2
Yield: 400 mg (76 %); elemental analysis calcd (%) for MoBrCl
9.11, H 2.12, N 5.37; found: C 38.52, H 2.84, N 5.40.
2
N
2
O
2
17
C H11 (522.05): C
3
-1
FTIR (KBr, /cm ): 3223 (w), 2924 (w), 2851 (w), 1944 (s), 1873 (s), 1639 (s), 1619 (s),
341 (w).
1
2
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
1
3
3
H NMR (400.10 MHz, DMF, t.a., δ ppm): 1.3 (d, 2H, ƞ -C
3
H
5
), 3.22 (d, 2H, ƞ -C
3
5
H ),
3
3
.39 (m, 1H, ƞ -C
3
H
5
), 8.3 (d, 2H, H
3
/H
6
), 8.4 (s, 2H, H
4
/H
5
), 8.9 (d, 2H, H
2
/H ).
7
1
3
3
3
C NMR (100.25 MHz, DMF, t.a., δ ppm): 56.0 ( -C
3
H
5
), 75.0 ( -C
3
5
H ), 128.0
(C
3
/C
4
/C
5
/C
6
), 141.0 (C
2
/C ).
7
3
[
MoBr(η -C
3
H
5
)(CO) (2,9-di-trichloromethyl-1,10-phenanthroline)] (8)
2
Yield: 470 mg, 68%; elemental analysis calcd (%) for MoBrCl
3
N
2
2
O C
19
H
11 + 2 CH
2
Cl
2
(751.37): C 29.41, H 1.76, N 3.27; found: C 29.42, H 2.0, N 4.08.
-1
FTIR (KBr, /cm ): 3215 (s), 3035 (s), 2962 (m), 2922 (m), 2860 (m), 1927 (s), 1846 (s),
1
635 (m), 1605 (m), 1392 (m), 1348 (m).
1
6
3
3
H NMR (400.10 MHz, DMSO-d , t.a., δ ppm): 1.24 (m, 2H,  -C
3
H
5
), 4.14 (d, 1H,  -
3
3
C
3
H
5
), 4,22 (d, 2H,  -C
H
3 5
), 4,47 (m, 2H,  -C
3
H
5
), 7.66 (d, 1H, H
4
), 7.95 (s, 1H, H ),
5
8
.28 (d, 1H, H ), 8.5 (d, 1H, H
6
3
), 8.77 (d, 1H, H
7
), 8.85 (d, 1H, H ).
2
1
3
6
C NMR (100.25 MHz, DMSO-d , t.a., δ ppm): 72.3 (C4), 119.2 (C3), 121.1 (C6), 139.6
(C2/C7).
HR-MS (ESI): Calcd for Mo(C14
H
6
N
2
Cl
6
)(C
3
H
5
)(CO)
2
(CH
3
OH)
2
(H
2
O)
2
(HCOOH)
.
+
[
M+H] m/z = 754.875575, found m/z = 754.81042.
3
[
MoBr(η -C
Yield: 460 mg (83 %); elemental analysis calcd (%) for MoBrN
597.70): C 47.22, H 2.70, N 9.37; found: C 47.01, H 2.59, N 9.45.
3
H
5
)(CO)
2
(dipyridophenazine)] (9)
4
O
2
C
23
H
15 + 0.5 CH
2
Cl
2
(
-1
FTIR (KBr, /cm ): 3049 (w), 3028 (w), 2925 (w), 1933 (s), 1864 (s), 1600 (w), 1575 (w),
1
464 (w), 1418 (m), 1361(m), 1139 (w), 1096 (w), 845 (w), 818 (m), 779 (m), 731 (m),
82 (w), 510 (w), 477 (w).
5
1
3
3
H NMR (400.10 MHz, DMF, t.a., δ ppm): 1.27 (m, 2H,  -C
3
H
5
), 3.00 (m, 1H,  -C
3
5
H ),
3
3
9
.19 (d, 2H,  -C
3
H
5
), 7.97 (m, 4H, H
2
/H
5
/H
6
/H
9
), 8.45 (m, 1H, H
7
), 8.47 (m, 1H, H ),
4
.29 (d, 2H, H
3
/H
8
), 9.80 (d, 1H, H10), 9.85 (d, 1H, H ).
1
1
3
3
3
3
C NMR (100.25 MHz, DMF, t.a., δ ppm): 54.2 ( -C
3
H
5
), 55.8 ( -C
3
H ), 73.7 ( -
5
C
3
H
5
), 129.9 (C
4
/C
7
), 132.1 (C
2
/C
5
/C
6
/C
9
), 135.2 (C /C10).
1
2
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
3
[
MoBr(η -C
3
H
5
)(CO)
2
(2,9-dichloro-dipyridophenazine)] (10)
Yield: 490 mg (79 %).
1H NMR (400.10 MHz, CDCl
, t.a., δ ppm): 1.43 (d, 2H, ƞ -C
3
H
5
), 2.92 (m, 1H,  -C
3
3
3
3
5
H ),
3
3
.09 (d, 2H, ƞ -C
3
H
5
), 7.83 (d, 2H, H
/H ), 9.76 (d, 1H, H
13C NMR (100.25 MHz, CDCl
, t.a., δ ppm): 125.8 (C
3
/H
8
), 7.99 (t, 1H, H
6
), 8.08 (t, 1H, H ), 8.39 (d, 1H,
5
H
7
), 9.63 (d, 2H, H
2
9
4
).
3
3
/C
8
), 129.5 (C
7
), 131.2 (C
5
/C ),
6
1
36.6 (C
2
/C
9
), 137.2 (C
4
).
3
[
Mo(CF
3
SO
3
)( η -C
3
H
5
)(CO)
) (0.353 g, 1 mmol) was added to a solution of
(dipyridophenazine)] (9) (0.555 g, 1 mmol) in dichloromethane (20
2
(dipyridophenazine)] (11)
Thallium triflate (TlCF
3
SO
3
3
[
MoBr(η -C
3
H
5
)(CO)
2
mL), and the mixture was refluxed overnight. A white solid of TlBr was formed and
filtered with celite. The solid was washed three times with dichloromethane. The filtrate
was concentrated, and n-hexane was added. After a few days, the formation of red crystals
was observed.
Yield: 460 mg (73 %); elemental analysis calcd (%) for MoN
4
O
5
SF
3
C
24
2
H15 + 1.5 CH Cl
2
(751.80): C 40.74, H 2.41, N 7.45; found: C 40.49,H 2,55, N 8.16.
-1
FTIR (KBr, /cm ): 3071 (w), 2924 (w), 2854 (w), 1948 (s), 1859 (s), 1497 (m), 1419 (m),
1
5
362 (m), 1309 (s), 1234 (s), 1218 (s), 1167 (s), 1027 (s), 811 (s), 775 (s), 730 (s), 631 (s),
80 (s).
1H NMR (400.10 MHz, CDCl
, t.a., δ ppm): 1.73 (d, 2H,  -C
3
H
5
), 3.89 (d, 2H, ƞ -C
3
3
3
3
5
H ),
3
4
.24 (m, 1H,  -C
3
H
5
), 8.01 (t, 2H, H
5
/H
6
), 8.13 (t, 2H, H
2
/H
9
), 8.40 (d, 2H, H /H ), 9.59
4
7
(
s, 2H, H
3
/H
8
), 9.88 (d, 2H, H /H10).
1
13C NMR (100.25 MHz, CDCl
, t.a., δ ppm): 60.9 ( -C
3
H
5
), 73.8 ( -C
3
3
3
3
5
H ), 126.5
(C2/C9), 130.0 (C4/C7), 132.4 (C5/C6), 136.6 (C1/C10), 143.3 (C3/C8).
Reactions of complexes 4, 5, and 9 with (tert-butylhydroperoxide (TBHP)
0 mmol of TBHP was added dropwise to a solution of 1mmol of the complex in
dichloromethane and was stirred for three hours. MnO was added to the mixture and
1
2
stirred for three days. Another sample using complex 4 was also stirred for three hours.
The solution was filtered, concentrated, and hexane was added to the solution. The yellow
powder formed was filtered and dried under vacuum. The FTIR spectra taken are shown in
Figure S2 of Supporting Information.
2
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Complex 4 (3 hours)
-1
FTIR (KBr, /cm ): 935 (s) (Mo=O), 901 (s) (Mo=O), 848 (s) (Mo-O-Mo), 725 (m), 648 (m).
1H NMR (400.10 MHz, CD
Cl , t.a., δ ppm): 8.01 (t, 2H, H2/H2’), 8.07 (s, 2H, H4/H4’),
.64 (d, 2H, H3/H3’), 9.75 (s, 2H, H1/H1’).
Complex 4
FTIR (KBr, /cm ): 921 (s) (Mo=O); 891 (s) (O-O); 870 (O-O); 845 (s) (O-O); 845 (s) (O-
); 725 (s); ((Mo(O2)2).
2
2
8
-1
O
Complex 5
FTIR (KBr, /cm ): 945 (s) (Mo=O); 907 (s) (Mo=O); 862 (m) (Mo-O-Mo); 836 (m), 795 (s)
Mo-O-Mo), 720 (s, br) ((Mo-O-Mo); 668 (m); 636 (m).
-1
(
Complex 9
FTIR (KBr, /cm ): 931 (s) (Mo=O); 891 (O-O); 843 (w); 818 (O-O); 645 (Mo(O2)2); 618
-1
(Mo(O2)2); 563 (s) ((Mo(O2)2).
Crystallography
Suitable single crystals of 7 and 11 for X-structure determination were obtained by
recrystallization from a solution of dichloromethane/n-hexane. A selected crystal of these
compounds was mounted on a goniometer and the X-ray diffraction data were collected
with monochromated Mo-K radiation using a Bruker Kappa APEX II diffractometer
equipped with CCD area detector and a low temperature-device. The crystals of both
complexes were kept at 180(2) K during the data collection. Data were corrected for
Lorentz and polarization effects. Absorption corrections were applied using the
SADABS.[12] The structures were solved by directed methods with SHEXLS[44] and refined
by least-squares methods using the ShelXL-2016/4[45] within the Olex2 software suite.[46]
All non-hydrogen atoms were refined with anisotropic thermal displacements. Hydrogen
atomic positions of the allyl ligand in both complexes were obtained from the difference
Fourier maps and refined with individual isotropic temperature factors. The remaining non-
hydrogen atom were calculated geometrically and refined using the riding model with
thermal parameters Uiso values set at 1.2 times the equivalent isotropic Ueq of the C atoms
to which they are attached. Crystal data are summarized in Table S3(Supporting
Information). The molecular graphics were made with Mercury[47] and PyMOL,[48]
respectively.
2
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
CCDC-1833524 and 1833525 contain(s) the supplementary crystallographic data for this
paper (complexes 7 and 11). These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/data request/cif.
Computational studies.
The Amsterdam Density Functional program package (ADF)[34] was used to perform the
Density Functional Theory calculations.[33 ] The geometries were optimized with gradient
correction, without symmetry constraints, using the Local Density Approximation of the
correlation energy (Vosko-Wilk-Nusair), [49] and the Generalized Gradient Approximation
(
Becke’s [50] exchange and Perdew’s [51] correlation functionals). Relativistic effects were
treated with the ZORA approximation. [52]The core orbitals were frozen for O, C, F, and N
1s) and S (1s, 2s, 2p). Triple  Slater-type orbitals (STO) were used to describe the
(
valence shells O, C, F, and N (2s and 2p) and S (3s, 3p). A set of two polarization
functions was added to O, C, F, N, and S (single , 3d, 4f). Triple  Slater-type orbitals
(STO) were used to describe the valence shells of H (1s) augmented with two polarization
functions (single  2s, 2p). The structures were modeled after those of compounds 7 and 11
described above. Three-dimensional representations of the structures were obtained with
Chemcraft. [53]
Catalytic studies. Epoxidation in organic solvents.
The complexes prepared were tested as catalysts in the epoxidation of cis-cyclooctene
(cy8), styrene (sty), cis-hex-3-en-1-ol (cis), trans-hex-2-en-1-ol (trans), R-(+)-limonene (R-
lim). The catalytic oxidation tests were carried out at 328 K in air in a reaction vessel
equipped with a magnetic stirrer and a condenser. In a typical experiment the vessel was
loaded with 100 % of the olefin, internal standard (dibutylether (DBE)), 1 % of catalyst,
2
00 % of oxidant (tert-butylhydroperoxide (TBHP)), and 3 ml of dichloromethane as
solvent. Addition of the oxidant determines the initial time of the reaction. The course of
the reaction was monitored by collecting samples at 10 min, 30 min, 1 h, 1h30 min, 2, 4, 6,
8
and 24 h of reaction. These samples were diluted in dichloromethane and treated with
MnO , which will destroy the oxidant and thus stop the reaction, and then analyzed in a
2
Shimadzu QP2010-Plus GC-MS with a capillary column (TRB-5MS (for cy8 and sty
substrates), TRB-1MS (for cis, trans and R-lim substrates)) for the determination of the
conversions and selectivity of the products.
2
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
Epoxidation of cis-cyclooctene, styrene, cis-hex-3-en-1-ol, trans-hex-2-en-1-ol, R-(+)-
limonene
Substrate (800 mg: 7.3 mmol of cis-cyclooctene, 7.7 mmol of styrene, 7.99 mmol of cis-3-
hex-3-en-1-ol and trans-hex-2-en-1-ol, 5.87 mmol of R-(+)-limonene, 800 mg dibutyl ether
(internal standard), 1 mol % of catalyst, 2.65 ml of tert-butylhydroperoxide (5.0-6.0 M in
decane) and 3 ml of CH Cl
2
2
.
Catalytic studies. Epoxidation in ionic liquids.
Complexes 4, 5 and 9 were also tested as catalysts in the epoxidation of cis-cyclooctene
(cy8) in the presence of different ionic liquids. The catalytic tests were carried out in the
same way as mentioned above. The only difference was the substitution of the 3 mL of
dichloromethane by 0.5 mL of the room temperature ionic liquid.
Recycling studies
n-Hexane was added to the reaction mixture, and the ionic liquid / catalyst phase was
collected, washed with n-hexane and dried in vacuum. A second reaction vessel equipped
with a magnetic stirrer was then loaded with the RTIL / catalyst phase, heated to 328 K and
TBHP was added. The remaining procedure was similar to the first run.
Acknowledgements
We thank Fundação para a Ciência e a Tecnologia for financial support
(UID/MULTI/00612/2013 and UID/QUI/00100/2013). MSS thanks FCT for fellowship
SFRH/BPD/88082/2012, APCR for fellowship SFRH/BPD/90883/2012, and MJC for
fellowship SFRH/BSAB/135473/2017. The crystallographic studies were supported by
CICECO – Aveiro Institute of Materials (UID/CTM/50011/2013) financed by National
Funds through the FCT/MEC and co-financed by QREN-FEDER through COMPETE
under the PT2020 Partnership.
References
[
[
[
1] M. J. Calhorda, P. J. Costa, Coord. Chem. Rev. 2017, 344, 83–100.
2] D. E. Ryan, D. J. Cardin, F. Hartl, Coord. Chem. Rev. 2017, 335, 103–149.
3] D. Morales, J. Pérez, L. Riera, V. Riera, R. Corzo-Suárez, S. García-Granda, D. Miguel,
Organometallics, 2002, 21, 1540–1545.
[
4] V.S. Joshi, M. Nandi, H. Zhang, B.S. Haggerty, A. Sarkar, Inorg. Chem. 1993, 32, 1301–
1303.
2
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800700
[
5] (a) B. M. Trost, M. Lautens, J. Am. Chem. Soc. 1982, 104, 5543–5545; (b) B. M. Trost, M.
Lautens, J. Am. Chem. Soc. 1987, 109, 1469–1478; (c) B. M. Trost, C. A. Merlic, J. Am.
Chem. Soc. 1990, 112, 9590–9600.
[
[
6] F. Dawans, J. Dewailly, J. Meunier-Piret, P. Piret, J. Organomet.Chem. 1974, 76, 53-63.
7] J. W. Goodyear, C. W. Hemingway, R. W. Harrington, M. R. Wiseman, B. J. Brisdon, J.
Organomet. Chem. 2002, 664, 176–181.
[
[
8] C. D. Nunes, M. J. Calhorda, Inorg. Chim. Acta 2015, 43, 122–131.
9] J. C. Alonso, P. Neves, M. J. Pires da Silva, S. Quintal, P.D. Vaz, C. Silva, A. A. Valente, P.
Ferreira, M. J. Calhorda, V. Félix, M. G. B. Drew, Organometallics 2007, 26, 5548–5556.
10] C. A. Gamelas, A. C. Gomes, S. M. Bruno, F. A. Almeida Paz, A. A. Valente, M. Pillinger,
C. C. Romão, I. S. Gonçalves, Dalton Trans. 2012, 41, 3474–3484.
[
[
[
[
[
[
[
[
[
[
11] T. Welton, Coord. Chem. Rev. 2004, 248, 2459–2477.
12] Q. Zhang, S. Zhang, Y. Deng, Green Chem. 2011, 13, 2619–2637.
13] D. Betz, A. Raith, M. Cokoja, F. E. Kühn, ChemSusChem 2010, 3, 559 – 562.
14] K. Srinivas, A. Kunar, S. Chauhan, Chem. Commun. 2002, 2456 –2457.
15] Z. Li, C. Xia, Tetrahedron Lett. 2003, 44, 2069 –2071.
16] Z. Li, C. Xia, M. Ji, Appl. Catal., A 2003, 252, 17– 21.
17] J. Dupont, R. Souza, P. Suarez, Chem. Rev. 2002, 102, 3667 –3692.
18] R. Saladino, R. Bernini, V. Neri, C. Crestini, Appl. Catal., A 2009, 360, 171 –176.
19] A. Valente, Z. Petrovski, L. C. Branco, C. A. M. Afonso, M. Pillinger, A. D. Lopes, C. C.
Romão, C. D. Nunes, I. S. Gonçalves, J. Mol. Catal. A: Chem. 2004, 218, 5 –11.
20] F. E. Kühn, J. Zhao, M. Abrantes, W. Sun, C. Afonso, L. Branco, I. S. Gonçalves, M.
Pillinger, C. C. Romão, Tetrahedron Lett. 2005, 46, 47–52.
[
[
[
[
[
[
[
21] M. Herbert, A. Galindo, F. Montilla, Catal. Commun. 2007, 8, 987 –990.
22] C. Bibal, J. C. Daran, S. Deroover, R. Poli, Polyhedron 2010, 29, 639– 647.
23] K. R. Jain, F. E. Kühn, Dalton Trans. 2008, 2221 –2227.
24] J. P. Hallett, T. Welton, Chem. Rev. 2011, 111, 3508–3576.
25] R. Sheldon, Chem. Commun. 2001, 2399–2407.
26] M. Herbert, F. Montilla, R. Moyano, A. Pastor, E. Álvarez, A. Galindo, Polyhedron, 2009,
2
8, 3929–3934.
27] M. Herbert, F. Montilla, A. Galindo, R. Moyano, A. Pastor, E. Álvarez, Dalton Trans. 2011,
0, 5210–5219.
[
4
[
[
28] H. tom Dieck, H. Friedel, J. Organomet. Chem. 1968, 14, 375–385.
29] M. S. Saraiva, S. Quintal, F. C. M. Portugal, T. A. Lopes, V. Félix, J. M. F. Nogueira, M.
Meireles, M. G. B. Drew, M. J. Calhorda, J. Organomet. Chem. 2008, 693, 3411–3418.
[
[
[
30] J. R. Ascenso, C. G. de Azevedo, M. J. Calhorda, M. A. A. F. de C. T. Carrondo, P. Costa, A.
R. Dias, M. G. B. Drew, A. M. Galvão, C. C. Romão, V. Félix, J. Organomet. Chem. 2001,
632, 197–208.
31] J. R. Ascenso, C. G. de Azevedo, M. J. Calhorda, M. A. A. F. de C. T. Carrondo, P. Costa, A.
R. Dias, M. G. B. Drew, A. M. Galvão, C. C. Romão, V. Félix, J. Organomet. Chem. 2001,
632, 197–208.
32] I. S. Gonçalves, L. F. Veiros, C. A. Gamelas, C. Cabrita, M. J. Calhorda, C. F.G.C. Geraldes,
J. Green, E. Packham, M. G.B. Drew, V. Félix, A. G. Santos, C. C. Romão, J. Organomet.
Chem. 2015, 792, 154–166.
[
[
33] R. G. Parr, W. Yang, Density Functional Theory of Atoms and Molecules; Oxford University
Press: New York, 1989.
34] a) G. te Velde, F. M. Bickelhaupt, S. J. A. van Gisbergen, C. Fonseca Guerra, E. J. Baerends,
J. G. Snijders, T. Ziegler, J. Comp. Chem., 2001, 22, 931 – 967; b) C. Fonseca Guerra, J. G.
2
9
This article is protected by copyright. All rights reserved.