Application of VIVO(acac)2 type complexes in the desulfurization of fuels with ionic liquids

Article abstract of DOI:10.1016/j.cattod.2012.03.037

The acetylacetonate complexes of oxovanadium(IV): VO(X-acac)₂ (X = Cl, CH₃ and CH₃CH₂) were structurally characterized in several alkylimidazolium ionic liquids by visible absorption and EPR spectroscopies. VO(Me-acac)₂ and VO(Et-acac)₂ showed solvatochromism in the selected ionic liquids (a λ₁ shift of ca. 120 nm was observed) and most complexes showed two species in the anisotropic EPR spectra in several ionic liquids. It was shown that the ionic liquids anions have coordinating ability and the basicity order found was: NTf₂^- < OTf^- < BF₄ ^-, in agreement with previous findings. An extraction and catalytic oxidation desulfurization (ECODS) system composed of VO(X-acac)₂, H₂O₂ and ionic liquids was tested in the removal of benzothiophene (BT), dibenzothiophene (DBT) and 4,6-dimethyldibenzothiophene (4,6-DMDBT) from a model oil (n-octane) at ambient temperature. The influence of the catalyst, ionic liquid, substrate and the molar ratio of the oxidant (H₂O₂) and the S-containing molecule were examined. The reaction rates of the oxidative desulfurization reaction were found to increase with the molar ratio of H₂O₂ and the S-containing molecule, as expected, however a ratio of 4 was enough to achieve >95% conversion to sulfone. The ionic liquid that showed the best performance with all substrates was bmimNTf₂. In bmimOTf and bmimBF₄ the activity of all catalysts in the ECODS system was low to moderate, which is probably due to the higher coordinating ability of these anions.

Full text of DOI:10.1016/j.cattod.2012.03.037

Catalysis Today 196 (2012) 119–125
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
IV
Application of V O(acac) type complexes in the desulfurization of fuels with
2
ionic liquids
a
b
c
a,∗
Andreia Mota , Nataliya Butenko , Jason P. Hallett , Isabel Correia
a
Centro Química Estrutural, Instituto Superior Técnico, TU Lisbon, Av. Rovísco Pais, 1049-001 Lisbon, Portugal
Departamento de Química, Bioquímica e Farmácia, Universidade do Algarve, Campus de Gambelas, 8005-139 Faro, Portugal
Chemistry Department, Imperial College London, London SW7 2AZ, UK
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
The acetylacetonate complexes of oxovanadium(IV): VO(X-acac)2 (X = Cl, CH3 and CH CH ) were
3
2
Received 31 October 2011
Received in revised form 14 March 2012
Accepted 19 March 2012
Available online 23 April 2012
structurally characterized in several alkylimidazolium ionic liquids by visible absorption and EPR spec-
troscopies. VO(Me-acac)2 and VO(Et-acac)2 showed solvatochromism in the selected ionic liquids (a ꢀ1
shift of ca. 120 nm was observed) and most complexes showed two species in the anisotropic EPR spec-
tra in several ionic liquids. It was shown that the ionic liquids anions have coordinating ability and the
−
−
−
basicity order found was: NTf2 < OTf < BF4 , in agreement with previous findings.
Keywords:
An extraction and catalytic oxidation desulfurization (ECODS) system composed of VO(X-acac)2, H2O2
and ionic liquids was tested in the removal of benzothiophene (BT), dibenzothiophene (DBT) and 4,6-
dimethyldibenzothiophene (4,6-DMDBT) from a model oil (n-octane) at ambient temperature. The
influence of the catalyst, ionic liquid, substrate and the molar ratio of the oxidant (H2O2) and the S-
containing molecule were examined. The reaction rates of the oxidative desulfurization reaction were
found to increase with the molar ratio of H2O2 and the S-containing molecule, as expected, however
a ratio of 4 was enough to achieve >95% conversion to sulfone. The ionic liquid that showed the best
performance with all substrates was bmimNTf2. In bmimOTf and bmimBF4 the activity of all catalysts in
the ECODS system was low to moderate, which is probably due to the higher coordinating ability of these
anions.
Ionic liquids
Vanadium complexes
ECODS
Desulfurization of fuels
©
2012 Elsevier B.V. All rights reserved.
1
. Introduction
dibenzothiophenes (DBTs), and particularly substituted DBTs. This
limitation led to the search for alternative processes, which will
operate under moderate conditions and without H2 and expensive
catalysts [2]. Due to short reaction time at ambient conditions, high
efficiency and selectivity, oxidation followed by extraction with
ionic liquids is regarded to be a promising alternative.
Various ionic liquids have been tested for the extraction of sul-
fur compounds from n-dodecane (model of diesel oil) [3,4], model
fuel and commercial gasoline [5,6]. Wasserscheid [3,4] promoted
imidazolium alkylsulfates, whereas Nie [7,8] favored imidazolium
dialkylphosphates, as candidates for extractive desulfurisation
technologies. Holbrey [9] showed that the partition ratio for
dibenzothiophene between dodecane and ionic liquids is strongly
influenced, and can be controlled, by the selection of the cation
(methylpyridinium < pyridinium ∼ imidazolium ∼ pyrrolidinium)
and to a much lesser extent the anion of the ionic liquid. Ionic liquids
with ethanoate and thiocyanate anions gave the best extraction
performance with each cation. Hansmeier et al. showed recently
[10] that the pyridinium-based ionic liquids [3-mebupy]N(CN)2,
The use of fossil fuels generates sulfur oxides, which are respon-
sible for smog, acid rain and respiratory problems in humans. Sulfur
also hinders the performance of catalytic converters of car exhausts.
To limit these effects, it is necessary to reduce the sulfur emissions
and thus, the sulfur in the crude oil must be removed at the refin-
ing stage. The oil industry is facing increasing pressure to remove
organic sulfur compounds from transportation fuels, since gov-
ernments have been implementing increasingly strict legislation
[
1].
Currently the removal of sulfur-containing compounds from
fuels is achieved by hydrodesulfurization (HDS). This technology
can desulfurize aliphatic and acyclic sulfur-containing compounds
on the industrial scale. However, to obtain fuels with the low
levels of sulfur-containing compounds required nowadays, very
harsh conditions have to be applied in HDS, namely high tem-
peratures and pressures, since it is not the best process for the
removal of aromatic thiophenes such as benzothiophenes (BTs),
[
4-mebupy]N(CN)2 and also [bmim]C(CN)3 are superior to sul-
folane (which is a benchmark extraction solvent for sulfur removal
from petrochemical streams) and also outperformed the ionic
liquids reported previously in the literature.
∗ Corresponding author. Tel.: +351 218419264.
E-mail address: icorreia@ist.utl.pt (I. Correia).
0
920-5861/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2012.03.037

1
20
A. Mota et al. / Catalysis Today 196 (2012) 119–125
However, and as a general rule, in the absence of an oxidant,
UV–vis Lambda 35 spectrophotometer and the temperature was
controlled with a Peltier controller from Perkin-Elmer. The EPR
spectra were recorded at 77 K (on glasses made by freezing
solutions in liquid nitrogen) and some at room temperature
with a Bruker ESP 300E X-band spectrometer. The GC sam-
ples were analysed on a GC–MS from Perkin Elmer model
Clarus 600 C.
ionic liquids have failed to provide high levels of sulfur removal.
The introduction of an oxidation step prior to the ionic liquid
extraction is a much more appealing approach since the oxi-
dation increases the polarity of the aromatic sulfur compounds,
which are then extracted with much higher partition ratios to
the ionic liquid phase. This process has been described using p.e.
H O /ethanoic acid [11]. However, in the absence of a catalyst very
2
2
high oxidant/sulfur rations have to be applied, in order to achieve
high sulfur removal. Several catalysts have been tested, which can
remove high amounts of sulfur with lower oxidant/sulfur ratios:
p.e. tungsten and molybdenum complexes [12,13], vanadium oxide
2
.4. Vis and EPR spectra
The visible electronic and EPR samples were prepared under
nitrogen with schlenk techniques to avoid vanadium oxidation, and
exposure of the ionic liquid to water.
[
14] and iron complexes [15,16]. These studies have shown the
relevance of the topic and the need to develop studies on the appli-
cation of the oxidation/extraction procedure for the desulfurization
of fuels, which is still underdeveloped. Among the different chem-
2.5. Desulfurization experiments
ical oxidants H O2 is the best option since it has a high content
2
As a general procedure the desulfurization experiments were
of active oxygen, gives water as the only by-product and is com-
mercially available. When used in conjunction with a transition
metal catalyst it forms peroxo species, which oxidize the sulfur
compounds. Our approach comes from our previous studies on the
behavior of VO(acac)2 in ionic liquids [17]. This complex and its
derivatives are easily obtained, soluble in a huge number of sol-
vents and ionic liquids, and since they are Lewis acids they are able
to activate peroxides. So, as a development of previous studies on
the structure of vanadium acetylacetonate complexes in ionic liq-
carried out in glass vials in which reagents were put together by
the following order: the selected catalyst, 0,5 ml of the ionic liquid,
1
ml of the octane solution containing the sulfur compound (the
substrate) and tetradecane (the internal standard) in a 1:1 molar
ratio. The reactions were initiated with the addition of the oxi-
dant (aqueous H O ). The mixtures were stirred and GC samples
2
2
were taken from the organic phase (after decantation) at differ-
ent time intervals: 15 or 30 min and 2 h, to eppendorfs containing
a small amount of triphenylphosphine (to deactivate the remain-
uids [17], we characterized the behavior of VO(X-acac) complexes
2
ing H O2 and stop the reaction). Stirring was kept constant in all
2
in alkylimidazolium ILs through spectroscopic techniques and then
experiments.
tested the complexes in the ECODS system with H O . Our goal was
2
2
to use room temperature, short reaction times and low H O /sulfur
ratios.
2
2
2.6. Analysis
Calibration curves were made with standards containing dif-
2
. Experimental
ferent molar ratios of the sulfur compound and tetradecane
(the internal standard), prepared in octane. The sulfur con-
tent of the standards and samples was analyzed using GC
2.1. Materials and reagents
All chemicals used were of analytical reagent grade. 1-
◦
with selected programs for temperature ranges of 80–250 C
Methylimidazole was purchased from Acros Organics and distilled
from potassium hydroxide; 1-chlorobutane was purchased from
Acros Organics and distilled from phosphorus pentoxide. Lithium
with
a
column from “SGE Analytical Science”, type BPX5
(30 m × 0.22 mm).
bis(trifluoromethylsulfonyl)imide [Li(NTf )] and lithium triflu-
2
3
. Results and discussion
oromethanesulfonate [Li(OTf)] were purchased from Apollo
Scientific and used as received. Bis(acetylacetonate)vanadium(IV)
was also obtained from Acros Organics. Dibenzotiophene and
3.1. Characterization the VO(X-acac)2 complexes in ionic liquids
4
,6-dimethylbenzothiophene were purchased from Aldrich and 1-
The complexes studied are shown in Scheme 1. They were
benzotiophene from Fluka. Octane and tetradecane were provided
from Aldrich and H O2 (30%) was from Panreac.
All syntheses and sample preparations were performed under
anaerobic conditions using standard Schlenk techniques. The
preparation and spectral data of the ionic liquids have been
described elsewhere and are included in SI [18,19].
synthesised (except 1 that is commercially available) according
to published procedures [20–22] and the solid state and solution
characterization proved its formulation and purity. The ionic
liquid ions chosen for this study are depicted in Scheme 2. The
procedure used in its preparation and purification (washing with
water, addition of charcoal, and filtration through acidic alumina)
afforded colorless liquids, suitable for spectroscopic studies. The
2
2
.2. Synthesis of the VIVO complexes
VO(Cl-acac) , VO(Me-acac)2 and VO(Et-acac)2 were prepared
2
according to literature procedures [20–22].
Analysis of VO(Cl-acac) : found: C 36.0 and H 3.5. Calc. for
2
C10H12O5Cl V: C 35.96 and H 3.62.
2
Analysis of VO(Me-acac) : found: C 49.1 and H 6.4. Calc. for
2
C12H18O5V: C 49.16 and H 6.19.
Analysis of VO(Et-acac) : found: C 52.2 and H 7.1. Calc. for
2
C14H22O5V: C 52.34 and H 6.90.
2.3. Instruments
1H NMR spectra were recorded on a Bruker 300 or 400 MHz
spectrometer. UV–vis spectra were recorded on a Perkin-Elmer
Scheme 1. Formula of the studied complexes. 1: R = H, VO(acac)2; 2: R = Cl, VO(Cl-
acac)2; 3: R = CH3, VO(Me-acac)2; 4: R = CH2CH3, VO(Et-acac)2.

A. Mota et al. / Catalysis Today 196 (2012) 119–125
121
Scheme 2. Molecular formula of the ionic liquid ions used in this study.
analytic characterization showed the absence of impurities, such as
residual chloride, which might change enormously their solvation
properties.
the ionic liquids used. Complex 2 shows poor solubility in most
ionic liquids and therefore low solvatochromism.
Fig. 1 shows the Vis spectra obtained for complex 4 dissolved
in the ionic liquids and Table 1 the position of the lowest energy
band and the type of anisotropy observed in the EPR spectra. The
Vis spectra measured for complexes 2 and 3 are included in the SI.
The d–d absorption spectra of vanadyl complexes are inter-
preted considering C4V symmetry for the square pyramid, and
three transitions are expected, considering the following energy
levels [23]:
The complexes were characterized by visible electronic and
EPR spectroscopies. The Vis spectra and particularly the position
of the lowest energy band give information on the coordinating
ability of the ionic liquid anion, which may change the reac-
tivity of the complex. The EPR spectrum is a fingerprint for
equatorial coordination since the presence of different complexes
in solution is detected due to the different coupling contri-
butions of each equatorial donor group. Thus, the analysis of
anisotropic EPR spectra can give detailed information on the num-
ber of species present in solution; the symmetry and coordination
geometry of the vanadium complex; and the identity of the equa-
torial ligands through the hyperfine coupling constant Az (or
A||).
b2(dxy) < e(dxz, dyz) < b (d
) < a (dz2)
1
1
x2−y2
Since it is considered that b2 is an almost pure vanadium 3dxy
orbital and e(dxz, dyz) is a combination of vanadium and oxygen
orbitals, the lowest energy transition (ꢀ ) – from b (dxy) to e(dxz
,
1
2
The characterization by Vis spectroscopy showed that com-
plexes 3 and 4 are solvatochromic, showing a large shift of the
lowest energy absorption band: ca. 120 nm for both complexes, in
dyz) – should be sensitive to axial perturbations from the solvent
[24]. Thus, upon dissolution, the five coordinated square pyrami-
dal VO(X-acac)2 complex may become six coordinated, with the

1
22
A. Mota et al. / Catalysis Today 196 (2012) 119–125
Table 1
Electronic absorption characterization of the complexes. Low energy absorption band (ꢀ1), difference in the 1st and 2nd absorption band (ꢀ1a − ꢀ2), EPR anisotropy and
difference in hyperfine coupling constants Ax − Ay.
Complex/ionic liquid
bbimNTf2
bmimNTf2
C3OmimNTf2
bm2imNTf2
bmimOTf
bmimBF4
VO(Cl-acac)2 2
ꢀ
1a (nm)
–
–
–
–
–
–
778.5
Axial
763.5
Axial
785
Axial
Anisotropy
VO(Me-acac)2 3
ꢀ
ꢀ
1a (nm)
1a − ꢀ2 (nm)
675
80
Rhombic
6.3
680.5
81
Rhombic
9.4
689.5
89
Rhombic
8.7
733
–
Axial
–
780
–
Axial
–
801
–
Axial
–
Anisotropy
4
−1
−1
|Ax − Ay| (× 10 cm
)
)
VO(Et-acac)2 4
ꢀ
ꢀ
1a (nm)
1a − ꢀ2 (nm)
678
80
Rhombic
8.8
681.5
80
Rhombic
9.4
686.5
92
Rhombic
8.8
757.5
–
Axial
–
774.5
–
Axial
–
798.5
–
Axial
–
Anisotropy
4
|Ax − Ay| (× 10 cm
O
V
O
V
O
V
O
O
O
O
O
An
O
O
O
O
O
O
An
O
trans
cis
5-coordinated
Scheme 3. Structure of the proposed species formed in solution. An = ionic liquid
anion. The 5-coordinated complex can be distorted into a trigonal bipyramid.
anisotropy. This suggests a distortion from the square pyramid to a
trigonal bipyramid and a 5-coordinated structure (see Scheme 3). In
−
the ILs of the anions with the highest coordinating power (OTf and
−
BF4 ) the EPR spectra are axial, probably due to an octahedral
geometry.
Fig. 1. Electronic absorption spectra of VO(Et-acac)2 (complex 4) in the visible
In terms of the number of species present in solution, VO(Cl-
acac)2 shows two species in all the measured ionic liquids, in
different ratios. In bmimOTf they are ca. 1:1. Moreover, the spin
Hamiltonian parameters are roughly the same as the ones obtained
for VO(acac)2 [17]. Complexes 3 and 4 show two species in most
of the ionic liquids. Fig. 2 shows the measured spectra at 77 K for
complex 3 (the others are included in SI). The parameters for the
two species shift slightly in the various ionic liquids, but overall we
can say that they correspond to the same type of complexes. Con-
region measured in different ionic liquids.
addition of a solvent molecule in the axial position (or an ionic
liquid anion) [24–27], and this will affect the position of the low
energy band. Hydrogen bonding to the oxo group will also affect
this band.
It has also been proposed [28] that a distortion from square
pyramid towards a trigonal bipyramide geometry breaks the
degeneracy of the e(dxz, dyz) orbitals resulting in the appearance
est
sidering the A|| for the coordination of the 4 equatorial carbonyl
of four bands in the Vis absorption spectra (bands Ia, I , II and
−4
−1
b
groups (168.4 × 10 cm ) we propose the outer species, whose
III) and rhombicity in the EPR spectra of vanadyl compounds. The
distortion of the geometry may be evaluated by analysis of the
wavelength difference between the central transitions in the Vis
spectrum, |ꢀ − ꢀ | and also by the difference between the EPR
−4
−1
parameters very between 168.1 and 164.7 × 10 cm , to be this
1
b
2
hyperfine coupling parameters |Ax − Ay|. The results suggest that
(
see Table 1) in bbimNTf , bmimNTf and C OmimNTf there is
2 2 3 2
a slight distortion of the geometry, towards a trigonal bipyramid
in the five-coordinated complexes and in the other group of ILs the
complexes are six-coordinated and approach an octahedral geome-
try. In terms of coordinating ability the following order is proposed:
−
−
−
NTf2 < OTf < BF4 , which was the same found for VO(acac)2 [17].
It might seem unexpected, since the anions that are used to
make ionic liquids are usually described as non-coordinating, that
any difference between the ILs is seen. However, since the anions
are in the neighbourhood and there is no other source of potential
electron pairs as the solvent is in huge excess, they will interact with
the metal centre to some extent. This is essentially a concentration
effect.
The EPR spectra of complexes 2–4 were measured at 77 K (in
ionic liquid glasses) and Table 2 shows the spin Hamiltonian param-
eters obtained by simulation [29] for all complexes. In the less
−
coordinating ionic liquids (containing NTf2 anions) the EPR spectra
show asymmetry in the perpendicular region, presenting rhombic
Fig. 2. EPR spectra of VO(Me-acac)2 measured in different ionic liquids at 77 K.

A. Mota et al. / Catalysis Today 196 (2012) 119–125
123
Table 3
ECODS experiments with complex 1 (5 mol%) in bmpyNTf2 with different H2O2:DBT
ratios.
H2O2/DBT
Conversion (15 min)
Conversion (2 h)
2
4
6
8
52.6
74.3
88.1
53.3
82.3
92.0
94.5
92.2
one. The inner one must therefore correspond to coordination of
the ionic liquid anion at the equatorial position since no other type
of complex is envisioned.
Overall, the spectroscopic experiments show that all ionic liquid
anions have coordinating ability and that three types of species can
be formed depending on the basicity of the anion: the trans complex
containing the anion in the axial position, the cis complex with the
anion in equatorial position and the 5-coordinated complex (see
Scheme 3).
DFT studies with VO(acac)2 in mmimOTf, presented in a previ-
−
ous report [17], confirmed that the anion OTf is able to coordinate
to VO(acac)2 with the trans isomer being more stable than the cis
by 4.8 kcal/mol. It also showed the importance of the hydrogen
bond donation ability from the cation, showing that the hyper-
fine coupling constant Az for the cis isomer changes more than
−
4
−1
upon consideration of the cation association to
1
0 × 10 cm
the VO(acac) OTf complex. This explains the observed shift in the
2
Hamiltonian hyperfine parameters in the different ionic liquids.
3.2. Desulfurization of model oil
Since VO(X-acac)2 complexes show high solubility in a wide
range of ionic liquids and are good Lewis acids we tested them
in the desulfurization of model fuel with ionic liquids. Among the
different chemical oxidants used in oxidative desulfurization (ODS)
the best option is hydrogen peroxide. However, in order to make an
ODS process competitive with deep HDS we still need to improve
the catalytic activity at low H O /S ratios and the mass transfer
2
2
in the biphasic system containing the oil fraction and polar phase.
Ionic liquids have the ability to act simultaneously as the oxidation
solvent and the extraction media (ECODS). As substrates we tested
several aromatic thiophenes: benzothiophene (BT), dibenzothio-
phene (DBT) and 4,6-dimethyldibenzothiophene (4,6-DMDBT) and
as model oil we used n-octane.
Dibenzothiophene was chosen as a sulfur compound represen-
tative of the main refractory sulfur-containing compounds in fuel
by HDS treatment. Thus, most of the catalytic experiments were
performed with this substrate. The experimental conditions and
results for this substrate are included in Tables 3 and 4. Depending
on the chosen ionic liquid and its miscibility with water, either bi-
phasic (model oil and IL) or tri-phasic (model oil, IL and aqueous
H O ) systems were obtained.
2
2
By analyzing the data from Table 4 we can see that 30 min
are enough, in most cases, to achieve very high conversion val-
−
−
ues (>90%) in ionic liquids with NTf2 anions (or PF6 ) and with
most complexes. These ILs are water immiscible and thus, tri-phasic
systems are obtained in the ECODS reaction system.
The screening of the H O :DBT ratio in bmpyNTf with 5 mol% of
2
2
2
VO(acac) (see Table 3) shows that as little as 2:1 ratio was enough
2
to obtain >80% conversion. In the following experiments we used a
ratio of 4 to assure that enough peroxide was present to obtain high
conversion values. In bmimNTf2 1 mol% of complex 1 was enough
to get a conversion of 90% (see entries 2 and 3 in Table 4).
In most experiments complexes 3 and 4 showed very similar
−
−
catalytic activity: high conversions in NTf2 and PF6 ionic liquids
and moderate in bmimBF4 and bmimOTf. Complex 2 also showed

1
24
A. Mota et al. / Catalysis Today 196 (2012) 119–125
Table 4
3.4. Oxidation of benzothiophene (BT)
ECODS experiments with all catalysts with DBT.
Entry Catalyst IL
Catalyst H2O2/ Conversion
Conversion
(2 h)
For this substrate complex 1 shows 100% conversion when
mol% catalyst were used in bmimNTf2 (see Table 6). In all other
(
mol%)
DBT
(30 min)
3
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
2
2
3
3
3
3
3
3
4
4
4
4
4
4
bmimNTf2 10
9.9
4
4
95.9
92.3
77.2
–
95.9
95.7
90.6
95.9
92.0
47.8
50.3
50.0
52.1
95.8
96.0
94.7
95.3
56.3
58.0
95.7
96.0
94.8
80.0
experiments, in which we varied the catalyst, the ionic liquid or the
amount of catalyst, the conversions were between 55 and 75%.
bmimNTf2
bmimNTf2
2.5
1
bm2imNTf2 10
10
bmpyNTf2
bmimBF4
bmimOTf
bmimBF4
bmimOTf
bmimNTf2
bmimPF6
bbimNTf2
bmpyNTf2
bmimBF4
bmimOTf
bmimNTf2
bmimPF6
bbimNTf2
bmpyNTf2
5
5
5
6
6
5
5
6
5
5
6
5
6
6
6
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
74.3
48.9
48.5
45.9
39.5
95.7
94.8
95.0
90.0
54.5
56.8
95.4
96.0
94.5
64.5
4. Conclusions
The spectroscopic studies presented in this paper show that
most of the VO(X-acac) complexes have high solubility in ionic liq-
2
1
1
1
1
1
1
1
1
1
1
uids and are good solvatochromic probes. The basicity order found
−
−
−
was: NTf2 < OTf < BF4 , in agreement with our previous findings
with VO(acac)2 [17] and with determinations by others authors
with different metal probes [30–32]. Coordination of the ionic liq-
uid anion was observed by both Vis and EPR spectroscopies. Two
isomers were found by EPR, which correspond to two types of iso-
mers: trans and cis. In a previous study with VO(acac)2 [17] DFT
studies showed the higher stability of the trans isomers (when com-
pared to the cis) and confirmed the importance of the hydrogen
bond donation ability from the cation, showing that the hyper-
fine coupling constant Az for the cis isomer changes more than
Table 5
ECODS experiments with several catalysts with 4,6-DMDBT and H2O2: 4,6-
DMDBT = 4.
−
4
−1
1
0 × 10 cm upon consideration of the cation association to the
VO(acac) OTf complex. This corroborates our spectroscopic stud-
2
Entry
Catalyst
IL
Catalyst
mol%)
Conversion
(30 min)
Conversion
(2 h)
ies, which show the presence of both isomers in most of the ionic
liquids and a shift in the hyperfine coupling constants, which is
probably due to hydrogen bond donation from the ionic liquid
cation.
(
1
2
3
4
5
6
7
8
9
1
1
1
1
1
3
3
3
4
4
4
bmimNTf2
bmimNTf2
bmimNTf2
bmimPF6
bbimNTf2
bmimNTf2
bmimBF4
bmimOTf
bmimNTf2
bmimBF4
bmimOTf
1
3
6
6
6
5
5
5
5
5
5
27.4
–
35.1
35.1
80.8
59.1
47.4
92.4
16.3
36.4
60.2
14.0
33.6
23.8
46.1
33.6
92.0
16.4
14.1
41.8
–
The complexes were used as catalysts in the oxidation of thio-
phenes from model oil at ambient temperature. Overall, the ECODS
studies showed that the VO(X-acac)2 complexes show very high
catalytic activity for the oxidation of the S-containing molecules in
−
the ionic liquids containing the least coordinating anions: NTf2
−
1
1
0
1
and PF6
.
24.6
The best ionic liquid for the process is bmimNTf , in which
2
conversions >95% (with BT and DBT) were achieved, when 5 mol%
of complexes 1 and 3 were used. Moreover, the use of catalysts
allowed the use of low H O :substrate ratios.
Table 6
ECODS experiments with several catalysts with BT and H2O2:BT = 4.
2
2
We believe that the use of these catalysts, which are readily
available, in other ionic liquids (preferably cheaper) may improve
the desulfurization of fuels at ambient temperature with low oxi-
dant:substrate ratios and are an attractive alternative method to
current HDS.
Entry
Catalyst
IL
Catalyst
mol%)
Conversion
(30 min)
Conversion
(2 h)
(
1
2
3
4
5
6
7
8
9
1
1
1
3
3
3
4
4
4
bmimNTf2
bmimNTf2
bmimNTf2
bmimNTf2
bmimBF4
bmimOTf
bmimNTf2
bmimBF4
bmimOTf
1
2
3
5
5
5
5
5
5
71.6
72.9
73.2
60.4
61.9
59.5
64.9
57.3
57.4
71.4
73.4
100.0
61.5
61.8
66.4
63.6
57.5
56.4
Acknowledgment
The work was financially supported by Funda c¸ ão para a Ciência
e Tecnologia (project PTDC/QUI-QUI/098516/2008).
moderate conversions in these two ILs. In general all complexes
show very good activity in ionic liquids containing NTF2 and PF6
Appendix A. Supplementary data
−
−
anions and low reaction times were observed.
Supplementary
can be
http://dx.doi.org/10.1016/j.cattod.2012.03.037.
data
found,
associated
in the
with
online
this
version,
arti-
at
cle
3.3. Oxidation of 4,6-dimethyldibenzothiophene (4,6-DMDBT)
This substrate was also tested since it presents high steric hin-
References
derance and is the most difficult refractory compound in HDS.
Table 5 shows the results obtained in experiments in which a
H O :4,6-DMDBT ratio of 4 was used. In bmimOTf and bmimBF
[
[
[
1] B. Pawelec, R.M. Navarro, J.M. Campos-Martin, J.L.G. Fierro, Catalysis Science
and Technology 1 (2011) 23–42.
2] J.M. Campos-Martin, M.C. Capel-Sanchez, P. Perez-Presas, J.L.G. Fierro, Journal
of Chemical Technology and Biotechnology 85 (2010) 879–890.
3] J. Esser, P. Wasserscheid, A. Jess, Green Chemistry 6 (2004) 316–322.
2
2
4
the conversion values obtained are low for all the studied catalysts,
and in most other ILs they are only moderate. However complex
shows a conversion of 92% in bmimNTf2 with 5 mol% of cata-
lyst. So, we can conclude that this is the best IL to perform these
experiments and that even this difficult substrate can be efficiently
oxidized and removed in the ECODS system.
3
[4] A. Bosmann, L. Datsevich, A. Jess, A. Lauter, C. Schmitz, P. Wasserscheid, Chem-
ical Communications (2001) 2494–2495.
[
[
5] S.G. Zhang, Z.C. Zhang, Green Chemistry 4 (2002) 376–379.
6] C.P. Huang, B.H. Chen, J. Zhang, Z.C. Liu, Y.X. Li, Energy Fuel 18 (2004)
1862–1864.

A. Mota et al. / Catalysis Today 196 (2012) 119–125
125
[
[
[
7] Y. Nie, C.X. Li, A.J. Sun, H. Meng, Z.H. Wang, Energy Fuel 20 (2006) 2083–2087.
8] X.C. Jiang, Y. Nie, C.X. Li, Z. Wang, Fuel 87 (2008) 79–84.
9] J.D. Holbrey, I. Lopez-Martin, G. Rothenberg, K.R. Seddon, G. Silvero, X. Zheng,
Green Chemistry 10 (2008) 87–92.
[20] S.S. Amin, K. Cryer, B.Y. Zhang, S.K. Dutta, S.S. Eaton, O.P. Anderson, S.M. Miller,
B.A. Reul, S.M. Brichard, D.C. Crans, Inorganic Chemistry 39 (2000) 406–416.
[21] D.C. Crans, A.R. Khan, M. Mahroof-Tahir, S. Mondal, S.M. Miller, A. la Cour,
O.P. Anderson, T. Jakusch, T. Kiss, Journal of the Chemical Society – Dalton
Transactions (2001) 3337–3345.
[
10] A.R. Hansmeier, G.W. Meindersma, A.B. de Haan, Green Chemistry 13 (2011)
1
907–1913.
[22] M. Mahroof-Tahir, D. Brezina, N. Fatima, M.I. Choudhary, Atta-ur-Rahman, Jour-
nal of Inorganic Biochemistry 99 (2005) 589–599.
[
[
11] W.H. Lo, H.Y. Yang, G.T. Wei, Green Chemistry 5 (2003) 639–642.
12] W.S. Zhu, H.M. Li, X. Hang, Y.S. Yan, H.D. Lu, J.X. Xia, Energy Fuel 21 (2007)
[23] C.J. Ballhausen, H.B. Gray, Inorganic Chemistry 1 (1962) 111–122.
[24] I. Bernal, P.H. Rieger, Inorganic Chemistry 2 (1963) 256.
[25] A. Urbanczyk, M.K. Kalinowski, Monatshefte fur Chemie 114 (1983) 1311–1319.
[26] A. Urbanczyk, M.K. Kalinowski, Monatshefte fur Chemie 121 (1990) 879–881.
[27] J. Selbin, T.R. Ortolano, Journal of Inorganic and Nuclear Chemistry 26 (1964)
37–40.
[28] E. Garribba, G. Micera, D. Sanna, Inorganica Chimica Acta 359 (2006)
4470–4476.
[29] A. Rockenbauer, L. Korecz, Applied Magnetic Resonance 10 (1996) 29–43.
[30] M.J. Muldoon, C.M. Gordon, I.R. Dunkin, Journal of the Chemical Society – Perkin
Transactions 2 (2001) 433–435.
2
514–2516.
[
[
[
[
[
13] L.N. He, H.M. Li, W.S. Zhu, J.X. Guo, X. Jiang, J.D. Lu, Y.S. Yan, Industrial and
Engineering Chemistry Research 47 (2008) 6890–6895.
14] D. Xu, W.S. Zhu, H.M. Li, J.T. Zhang, F. Zou, H. Shi, Y.S. Yan, Energy Fuel 23 (2009)
5
929–5933.
15] H.M. Li, W.S.A. Zhu, Y. Wang, J.T. Zhang, J.D. Lu, Y.S. Yan, Green Chemistry 11
2009) 810–815.
(
16] J.T. Zhang, W.S. Zhu, H.M. Li, W. Jiang, Y.Q. Jiang, W.L. Huang, Y.S. Yan, Green
Chemistry 11 (2009) 1801–1807.
17] I. Correia, A. Mota, J.P. Hallett, M.L. Kuznetsov, Physical Chemistry Chemical
Physics 13 (2011) 15094–15102.
18] I. Correia, T. Welton, Dalton Transactions (2009) 4115–4121.
19] L. Cammarata, S.G. Kazarian, P.A. Salter, T. Welton, Physical Chemistry Chemical
Physics 3 (2001) 5192–5200.
[31] J. Bartosik, A.V. Mudring, Physical Chemistry Chemical Physics 12 (2010)
4005–4011.
[32] S. Pitula, A.V. Mudring, Physical Chemistry Chemical Physics 12 (2010)
7056–7063.
[
[