Deep oxidative desulfurization of dibenzothiophene with molybdovanadophosphoric heteropolyacid-based catalysts

Article abstract of DOI:10.1007/s11243-013-9792-7

Three hydrophobic Keggin-type heteropolyacid catalysts, [C ₃H₃N₂(CH₃)(C₂H ₄)]₅PMo₁₀V₂O₄₀ ([C ₂mim]PMoV), [C₃H₃N₂(CH ₃)(C₄H₈)]₅PMo₁₀V ₂O₄₀ ([C₄mim]PMoV) and [C₃H ₃N₂(CH₃)(C₆H₁₂)] ₅PMo₁₀V₂O₄₀ ([C₆mim]PMoV) , were synthesized by reacting molybdovanadophosphoric acid with imidazolium bromides, and characterized by spectroscopic methods. Their use as catalysts in the extractive catalytic oxidative desulfurization process using hydrogen peroxide as the oxidant and acetonitrile as phase transfer agent was studied. The catalytic properties decreased in the order: [C₆mim]PMoV > [C₄mim]PMoV > [C₂mim]PMoV. The main factors influencing the rate of removal of dibenzothiophene (DBT) were investigated, including reaction temperature, the amounts of catalyst, H₂O₂ and acetonitrile. Nearly 100 % sulfur removal rate was achieved under optimal conditions. The catalyst could be recycled six times with only a slight decrease in activity. A reaction mechanism for DBT oxidation is proposed, in which the Keggin anions first obtain active oxygen from H₂O₂, then the DBT is oxidized to dibenzothiophene sulfones.

Full text of DOI:10.1007/s11243-013-9792-7

Transition Met Chem (2014) 39:213–220
DOI 10.1007/s11243-013-9792-7
Deep oxidative desulfurization of dibenzothiophene
with molybdovanadophosphoric heteropolyacid-based catalysts
Jiangzhou Zhuang
Xiaoying Jin
•
Bing Hu
•
Junjun Tan
•
Received: 16 September 2013 / Accepted: 3 December 2013 / Published online: 20 December 2013
Ó Springer Science+Business Media Dordrecht 2013
Abstract Three hydrophobic Keggin-type heteropolyacid
catalysts, [C H N (CH )(C H )] PMo V O ([C mim]P-
At present, hydrodesulfurization (HDS) is widely used
in the petroleum refining industry. HDS can remove or-
ganosulfur compounds including thiols, thioethers and
disulfides efficiently but require high temperature, high
pressure and a suitable catalyst [2]. Unfortunately, some
aromatic sulfur compounds such as benzothiophene (BT),
dibenzothiophene (DBT) and 4,6-dimethyldibenzothio-
phene (4,6-DMDBT) are less reactive to HDS because of
steric hindrance [3, 4].
3
3
2
3
2
4
5
10
2
40
2
MoV), [C H N (CH )(C H )] PMo V O ([C mim]PMoV)
5
3
3
2
3
4
8
10
2
40
4
and [C H N (CH )(C H )] PMo V O ([C mim]PMoV),
12
3
3
2
3
6
5
10
2
40
6
were synthesized by reacting molybdovanadophosphoric
acid with imidazolium bromides, and characterized by
spectroscopic methods. Their use as catalysts in the extrac-
tive catalytic oxidative desulfurization process using hydro-
gen peroxide as the oxidant and acetonitrile as phase transfer
agent was studied. The catalytic properties decreased in the
order: [C mim]PMoV [ [C mim]PMoV [ [C mim]PMoV.
In order to achieve ultra low sulfur fuels, several alter-
native deep desulfurization approaches have been investi-
gated, such as adsorption [5], photocatalysis [6],
biodesulfurization [7] and oxidation [8, 9]. Among these
methods, extractive catalytic oxidative desulfurization
(ECODS) is regarded as one of the most promising tech-
niques due to its high sulfur removal rate and selectivity
under mild operating conditions (low reaction temperature
and pressure). To date, several oxidants have been studied,
including O [10], O [11], HNO /NO [12], t-BuOOH
6
4
2
The main factors influencing the rate of removal of dibenzo-
thiophene (DBT) were investigated, including reaction tem-
perature, the amounts of catalyst, H O and acetonitrile.
2
2
Nearly100 % sulfurremoval ratewas achieved under optimal
conditions. The catalyst could be recycled six times with
only a slight decrease in activity. A reaction mechanism for
DBT oxidation is proposed, in which the Keggin anions first
obtain active oxygen from H O , then the DBT is oxidized to
2
2
2
3
3
2
dibenzothiophene sulfones.
[13], KO [14] and H O . Among these, H O [15, 16] has
2 2 2 2 2
been widely used not only for its high oxidation activity,
but also because of its production of water as by-product.
Hydrogen peroxide is also cheaper and safer than organic
peroxides and peracids. Ionic liquids have been attracted
much attention for use as phase transfer agents (PTA),
based on two characteristics; namely good extraction per-
formance and very low vapor pressure [8, 17]. However,
Swatloski et al. [18] have pointed out that ionic liquids are
not always green. Some ionic liquids can decompose under
the conditions used for their preparation, which giving the
fluoride hydrates. Moreover, the preparation of ionic liq-
uids can be complicated. Acetonitrile is a good solvent
which has been used in oil refining and other industries and
is not expensive. Therefore, acetonitrile was selected as a
PTA in this work.
Introduction
The increasingly severe environmental issues caused by the
widespread and increasing use of fuel oil have led to an
urgent need for sulfur emission reduction agents. Stringent
regulations intended to reduce the sulfur content in fuel oil
as low as 10 ppm have been introduced [1].
J. Zhuang ꢀ B. Hu (&) ꢀ J. Tan ꢀ X. Jin
School of Chemistry and Chemical Engineering, Hubei
University of Technology, Wuhan 430068, People’s Republic of
China
e-mail: hbwh04@163.com
123

2
14
Transition Met Chem (2014) 39:213–220
With respect to catalysts for ECODS, heteropoly-based
ionic liquids have attracted considerable attention. Li et al.
19–21] have described a series of amphiphilic catalysts
such as [(C H ) N(CH ) ] [PW O ], [C H N
molybdate and sodium metavanadate were purchased from
Sinopharm Chemical Reagent Co., China. The reagents were
AR grade and used without further purification. N-methylim-
idazole (purity [99 %) was purchased from Kaile chemical
factory, China, and used after distillation under reduced
pressure.
[
1
8
37 2
3 2 3
12 40
18 37
(
CH ) ] [P W O ], [C H N(CH ) ] [P W O ], [C
3 3 6 2 18 62 18 37 3 3 10 2 17 61 18
H N(CH ) ] [a-H P W O ] and [C H N(CH ) ] Na
3
7
3 3 12
2
2
12 48
18 37
3 3 5
2
[
PW O ], all of which show high catalytic activity and
11 39
selectivity. Wenshuai Zhu et al. [22] reported several
Keggin-type polyoxometalate-based ionic liquid hybrid
materials such as 1-(3-sulfonic acid)propyl-3-methyl imi-
dazolium phosphotungstate ([MIMPS] PW O ꢀ2H O),
Synthesis of molybdovanadophosphoric acid and ionic
liquids
The molybdovanadophosphoric acid and ionic liquids were
synthesized according to the literature procedures [26, 27].
3
12 40
2
1
-butyl-3-methyl
imidazolium
phosphomolybdate
(
[Bmim] PMo O ) and 1-butyl-3-methyl imidazolium
3 12 40
In a typical synthesis of [C mim]Br, ethyl bromide
2
silicotungstate ([Bmim] SiW O ). Of these, [MIM-
12 40
(0.1 mol) was added dropwise to N-methylimidazole
(0.1 mol) under a nitrogen atmosphere at room temperature
in 15 min, followed by stirring for 24 h. After the reaction,
the filter product was washed with ethyl acetate, dried at
60 °C under vacuum for 8 h, then cooled to room tem-
perature. The product was isolated as a white solid. The
preparations of [C mim]Br and [C mim]Br were similar to
3
PS] PW O ꢀ2H O presented the best catalytic activity
3
12 40
2
with 100 % sulfur removal in the ionic liquid 1-octyl-3-
methyl imidazolium hexafluorophosphate ([Omim]PF6).
Huang et al. [23] synthesized a catalyst by reacting octa-
decyltrimethylammonium bromide with phosphotungstic
acid. This catalyst showed high selectivity toward oxida-
tion of DBT. These studies also showed that the Keggin
structure hybrid acid anion could show a better catalytic
activity while replacing hydrogen cations [24].
4
6
[C mim]Br, except that ethyl bromide was replaced by
2
butyl and hexyl bromides, respectively, and the reaction
temperatures were 70 and 80 °C, respectively. The pro-
ducts of the reactions with [C mim]Br and [C mim]Br
Molybdovanadophosphoric heteropoly acids present high
activity for oxidation reactions. In our previous work [25],
we synthesized [C H N (CH )] PMo V O which was
4
6
were yellowish and brown liquids, respectively.
3
3
2
3
5
10 2 40
used in oxidative desulfurization, giving a sulfur removal
rate of 99.1 %. According to the literature [9], the number of
carbons in the cationic alkyl groups can affect the catalytic
properties. With this in mind, we carried out the present study
as follows. Three molybdovanadophosphoric heteropoly-
Synthesis of molybdovanadophosphoric
heteropolyacid-based catalysts
Molybdovanadophosphoric heteropolyacid-based catalyst
[C mim]PMoV was synthesized according to the literature
2
acid-based catalysts were synthesized, namely [C mim]P-
2
[28]. A solution of molybdovanadophosphoric acid (1 mmol)
in deionized water (50 ml) was added dropwise to a solution
MoV, [C mim]PMoV and [C mim]PMoV, by reacting
4
6
H PMo V O with the respective ionic liquids; 1-ethyl-3-
10
of [C mim]Br (5 mmol) in deionized water (50 ml),
2
5
2
40
methyl imidazolium bromide ([C mim]Br), 1-butyl-3-
2
whereupon a mustard yellow precipitate appeared immedi-
ately. This was filtered off, washed with deionized water and
dried at 60 °C under vacuum. 1H NMR (600 MHz, DMSO-
d6): d = 1.44 (3H, t), 3.90 (3H, s), 4.25 (2H, t), 7.71 (1H, s),
7.80 (1H, s), 9.09 (1H, s).
methyl imidazolium bromide ([C mim]Br) and 1-hexyl-3-
4
methyl imidazolium bromide ([C mim]Br). The perfor-
6
mance of these catalysts for the oxidative desulfurization of
DBT was investigated via an ECODS process using H O as
2
2
oxidant and acetonitrile as PTA. The ECODS conditions
were also optimized, including reaction temperature, cata-
lyst amount, as well as the volume ratios of model oil to H O
[C mim]PMoV was prepared as a yellow solid by the
4
same method as for [C mim]PMoV. 1H NMR (600 MHz,
2
DMSO-d6): d = 0.90 (3H, t), 1.27 (2H, m), 1.77 (2H, t), 3.89
(3H, s), 4.21 (2H, t), 7.71 (1H, s), 7.80 (1H, s), 9.09 (1H, s).
2
2
[
V(S):V(O)] and model oil to acetonitrile [V(S):V(PTA)].
[C mim]PMoV was obtained as a yellow solid accord-
6
ing to the method used for [C mim]PMoV. 1H NMR
2
Experimental
(600 MHz, DMSO-d6): d = 0.86 (3H, t), 1.24 (6H, m),
1.79 (2H, m), 3.89 (3H, s), 4.20 (2H, t), 7.71 (1H, s), 7.78
Materials
(1H, s), 9.08 (1H, s).
Dibenzothiophene (DBT, 99 %) was purchased from Across
Organics, USA. Hydrogen peroxide (30 % wt.), acetonitrile, n-
octane, ethyl acetate, ethyl bromide, butyl bromide, hexyl
bromide, sodium phosphate monobasic dihydrate, sodium
Characterization of the samples
FTIR spectra were acquired from KBr pellets with a
Nicolet 8700 FTIR spectrophotometer in the range of
1
23

Transition Met Chem (2014) 39:213–220
215
-
00–4,000 cm . Powder XRD patterns were measured
1
-1
4
showed four peaks at 1,057, 958, 872 and 787 cm , cor-
responding to the P–O, M = Ot, M-Ob and M-Oc bonds
(M = Mo or V, Ot = terminal oxygen, Ob = bridging
oxygen between corner-sharing octahedra, Oc = bridging
oxygen between edge-sharing octahedra) [29]. The four
with a Philips Analytical X-ray diffractometer using a Cu
˚
Ka radiation (1.5406 A) source at 40 kV and 30 mA from
5
to 808 with a scan rate of 0.5 °/min. UV–vis spectra were
obtained on a TU-901 spectrophotometer in the range of
00–600 nm using acetonitrile as the solvent. 1H NMR
5
-
2
prominent FTIR features due to PMo V O
0 2 40
were
1
-
1
spectra were recorded on a Bruker AV600 spectrometer
using DMSO-d6 as the solvent.
clearly marked in the region of 700–1,100 cm which
indicated the primary Keggin structure was intact in these
catalysts [25]. The remaining FTIR spectral features in the
regions of 1,130–1,650, 2,800–3,200 and 3,400–
Catalytic ECODS of model oil
-
1
3
,500 cm in spectra (e–f) are similar to those observed in
Model oils with an S-content of 500 ppm were prepared by
dissolving DBT in n-octane. The oxidative desulfurization
catalytic tests were performed in a flask equipped with a
stirrer and condenser. The oxidative reaction was carried
out with 10 ml of model oil, a certain amount of H O ,
spectra (a–c) in Fig. 1 for pure imidazolium bromide ionic
liquids [28]. Hence, the three molybdovanadophosphoric
heteropolyacid-based catalysts retain their Keggin struc-
ture. The small difference of absorption peaks among the
catalysts may relate to the cationic structures [1].
2
2
catalyst and acetonitrile. The ECODS process was studied
at different temperatures. After reaction, the upper oil layer
was decanted at room temperature and analyzed by GC-
FID with an internal standard (Agilent 7890A equipped
The Keggin structures of H PMo V O and the cata-
5 10 2 40
lysts were further confirmed by the XRD patterns. The
XRD patterns of H PMo V O and the catalysts are
5
10 2 40
shown in Fig. 2. The main positions (2h) of the XRD peaks
are in the region of 7–10°, 16–22° and 25–34°, especially
the strongest peak in the first band, i.e., 7–10°. According
to the literature [30, 31], the reflection at 2h = 9° is a
typical feature of the Keggin structure. Meanwhile, the
XRD patterns of the catalysts after the catalytic reaction
showed similar features as the parent H PMo V O ,
with
a
capillary column (HP-5, 30 m 9 0.32 mm
i.d. 9 0.25 lm)). The catalyst which dispersed in the lower
phase was collected by distilling under vacuum at 60 °C
for 4 h, dried and used in the recycling experiments.
5
10 2 40
Results and discussion
confirming that the catalysts retain their structure (Fig. 3).
The differences between H PMo V O and the molyb-
5
10 2 40
Characterization of the catalysts
dovanadophosphoric heteropolyacid-based catalysts may
be attributed to the imidazolium cations.
FTIR spectra are quite useful to identify structural and
bonding changes. The FTIR spectra of the imidazolium
bromide ionic liquids H PMo V O and molybdovana-
The UV–vis spectra of H PMo V O and the catalysts
5 10 2 40
are illustrated in Fig. 4. The pure H PMo V O exhibits
10 2 40
5
two absorption bands at 214 and 312 nm, attributed to the
Keggin structure and in agreement with the results reported
previously [25]. The band at 214 nm is assigned to a
O ? Mo transition, while the band at 312 nm is most
likely attributed to the incorporation of vanadium into the
5
10 2 40
dophosphoric heteropolyacid-based catalysts are presented
5
-
in Fig. 1. The PMo V O
0 2 40
Keggin primary structure
1
g
f
d
c
e
d
c
b
a
b
a
4
000
3000
2000
1000
-1
Wavenumber (cm )
10
20
30
40
2
50
60
70
80
θ (°)
Fig. 1 FTIR spectra of samples: a [C
c [C mim]Br, d H PMo10 40, e [C
and g [C mim]PMoV
2
4
mim]Br, b [C mim]Br,
6
5
2
V O
2
mim]PMoV, f [C
4
mim]PMoV
Fig. 2 XRD
b [C
patterns
of
samples:
mim]PMoV and d [C
5 2 40
a H PMo10V O ,
mim]PMoV
6
2
mim]PMoV, c [C
4
6
123

2
16
Transition Met Chem (2014) 39:213–220
1
00
c
8
6
0
0
b
[
[
[
C mim]PMoV
2
C mim]PMoV
4
C mim]PMoV
6
a
40
1
0
20
30
40
θ (°)
50
60
70
80
2
2
0
2
0
40
60
80
100
120
140
160
180
2
Fig. 3 XRD patterns of samples after reaction: a [C mim]PMoV,
Time (min)
b [C
mim]PMoV and c [C mim]PMoV
4 6
Fig. 5 Influence of catalyst on BDT conversion. Conditions: 10 ml
model oil, T = 50 °C, m(catalyst) = 0.04 g, V(S):V(O):V(PTA) =
0
0
0
0
0
0
0
0
.7
.6
.5
.4
.3
.2
.1
.0
2
0:1:4
[
[
[
C mim]PMoV
2
C mim]PMoV
100
4
C mim]PMoV
6
H PMo V O
5
10
2
40
8
6
0
0
0
0
0
0
0
.02 g
.03 g
.04 g
.05 g
.06 g
40
20
2
00
300
400
500
600
Wavelength (nm)
0
Fig. 4 UV–vis spectra of samples: H
mim]PMoV and [C mim]PMoV
5
PMo10V
2
O40, [C
2
mim]PMoV,
0
20
40
60
80
100 120 140 160 180
[
C
4
6
Time (min)
Fig. 6 Influence of amount of catalyst on DBT conversion. Condi-
tions: 10 ml model oil, T = 50 °C, V(S):V(O):V(PTA) = 20:1:4
Keggin ion [32]. These bands are shifted to 215 and
10 nm, respectively, in the spectra of the catalysts. These
3
shifts might be related to intermolecular electronic inter-
actions between the imidazolium cations and molybdova-
nadophosphoric anions [28]. According to the results of the
FTIR, XRD and UV–vis studies, it is clear that the catalysts
retain the Keggin structure of the molybdovanadophos-
phoric anion.
literature [9]. The longer alkyl chain should render the cata-
lyst more hydrophobic and therefore more soluble in both the
acetonitrile and the oil phases. Moreover, the optimum molar
quantities of the catalyst follow the order [C
mim]PMoV [
mim]P-
2
[C mim]PMoV [[C mim]PMoV. Therefore, [C
4
6
6
MoV was selected for further optimization of the catalytic
conditions.
Catalytic ECODS properties
Next, the effect of amount of [C mim]PMoV in the
6
The sulfur removal performance of the three molybdova-
nadophosphoric heteropolyacid-based catalysts in the EC-
catalytic process was investigated. Increasing the amount
of catalyst improved the sulfur removal performance. As
can be seen from Fig. 6, the curve for 0.02 g of catalyst
shows a different final percentage of sulfur removal com-
pared to the other quantities. When the amount of catalyst
is over 0.02 g, the rate of reaction increases notably, and a
higher extent of sulfur removal is achieved in a shorter
reaction time. However, the plots in Fig. 6 indicate that
ODS process is displayed in Fig. 5. [C mim]PMoV shows
6
the best catalysis, followed by [C mim]PMoV and
4
[
C mim]PMoV with sulfur removal efficiencies of 98.5,
2
9
6.6 and 94.1 %, respectively. Hence, the efficiency of
sulfur removal increases with increasing alkyl chain length;
a similar situation has been described previously in the
1
23

Transition Met Chem (2014) 39:213–220
217
100
1
00
8
0
8
6
4
2
0
0
0
0
0
60
40
0
.2 ml
0.3 ml
0
0
o
3
0 C
.4 ml
.5 ml
o
40 C
o
0.6 ml
50 C
20
o
6
7
0 C
o
0 C
0
0
20
40
60
80
100 120 140 160 180
0
20
40
60
80
100 120 140 160 180
Time (min)
Time (min)
2 2
Fig. 7 Influence of volume of H O on the removal of DBT.
Conditions: 10 ml model oil, T = 50 °C, m(catalyst) = 0.04 g,
V(S):V(PTA) = 5:1
Fig. 8 Influence of reaction temperature on the removal of DBT.
Conditions: 10 ml model oil, m(catalyst) = 0.04 g,
V(S):V(O):V(PTA) = 20:1:4
there is a threshold value for the amount of catalyst, after
which the influence of catalyst amount becomes less sig-
nificant. Therefore, an amount of 0.04 g was selected for
subsequent investigations.
temperature is raised to 60 and 70 °C. At these tempera-
tures, H O decomposes rapidly, hence there is insufficient
2
2
oxidant, and the extent of sulfur removal is less than that
obtained at 40 °C in 180 min. The curve obtained at 50 °C
exhibits the highest overall extent of sulfur removal, and
the reaction is close to equilibrium within 120 min. Thus,
Next, the effects of varying the amount of H O were
2
2
investigated. As can be seen from Fig. 7, the rate of reac-
tion in the first 20 min increases upon increasing the vol-
ume of H O from 0.2 to 0.3 ml and then decreases for
5
0 °C is taken as an optimum temperature for sulfur
2
2
removal in these experiments.
volumes of H O over 0.3 ml. However, the ultimate
2
2
extent of sulfur removal increases from 81.0 to 99.6 % on
increasing the volume of H O from 0.2 to 0.5 ml. Zhang’s
Influence of the volume of phase transfer agent
2
2
and Li’s groups found a similar pathway for the oxidation
of DBT to dibenzothiophene sulfones (DBTOs) that could
be used to explain the reaction in this ECODS process [1,
In a typical ECODS process, the PTA plays the role of
partitioning the S-containing components and provides an
environment for the ODS reaction [23, 33]. Therefore, the
amount of acetonitrile was studied in order to find out its
influence on the sulfur removal performance. As shown in
2
5]. H O as the oxidant can continuously supply active
2 2
oxygen to the heteropoly anions. Then, DBT is oxidized to
DBTOs by reaction with the heteropoly anions in the phase
interface. Increasing the amount of H O could accelerate
2
2
1
00
80
60
the oxidation of DBT, but also increase the amount of H O,
2
which could hinder the reaction of DBT [33]. It may be that
the rate of reaction is governed by the diffusion of H O . If
2
2
so, then when the diffusion limit is reached, adding more
H O will not increase the rate of reaction further, but may
2
2
increase the amount of H O. Overall, 0.5 ml of H O
2 2
2
1
10:3
1
1
10:6
0:2
seems to be the optimum amount under our experimental
conditions.
4
2
0
0
0
0:4
0:5
Influence of reaction temperature
In principle, increasing the reaction temperature could both
enhance DBT oxidation and accelerate H O2 thermal
2
0
20
40
60
80
100
120
140
160
180
decomposition [34]. As shown in Fig. 8, the curves for 30
and 40 °C have similar trends; the extent of reaction
increases slowly with time. However, the initial sulfur
removal rate appears markedly different when the
Time (min)
Fig. 9 Influence of volume of phase transfer agent on the removal of
DBT. Conditions: 10 ml model oil, T = 50 °C, m(catalyst) = 0.04 g,
V(S):V(O) = 20:1
123

2
18
Transition Met Chem (2014) 39:213–220
4
3
2
1
0
1
00
90
1
1
1
1
1
0:2
0:3
0:4
0:5
0:6
8
7
6
5
0
0
0
0
2
R = 0.95
0
.4
0.5
0.6
0.7
0.8
0.9
0
10
20
30
40
50
60
C - C
A
Time (min)
I
Fig. 10 Pseudo-first-order kinetics of oxidation using different
volumes of acetonitrile. Conditions: 10 ml model oil, T = 50 °C,
m(catalyst) = 0.04 g, V(S):V(O) = 20:1
Fig. 11 Relationship between sulfur removal rate and the concen-
tration difference of water
slight decline in activity. However, its performance declined
significantly if the lower layer was used for the next run
without any further treatment (Table 2). It seems possible
that the remaining water in the acetonitrile phase might
hinder DBT oxidation. Based on the two-film theory, the
mass transfer rate is directly proportional to the concentra-
tion difference. A hypothesis is proposed, such that the
concentration difference can be obtained by the subtracting
water concentration in the acetonitrile phase from that in the
interface, and DBT at the interface is rapidly oxidized. This
leads to the following equation;
Table 1 Rate constant k and correlation coefficient R for different
volumes of acetonitrile
V(S):V(PTA)
10:2
10:3
10:4
10:5
10:6
-
1
)
k (min
0.0211
0.9894
0.0425
0.9872
0.0601
0.9825
0.0654
0.9999
0.0655
0.9998
R
Fig. 9, the sulfur removal rates are fairly invariant obvi-
ously with the increasing V(S)/V(PTA) ratio beyond 10:3.
The rate of reaction under different volumes of phase
transfer agent was calculated according to the pseudo-first-
order equation;
S / CI ꢁ CA
where S is the extent of sulfur removal, and C and C are
ð3Þ
I
A
the water concentrations in the interface and acetonitrile
ꢁ
dC=dt ¼ kC
ð1Þ
ð2Þ
phase, respectively. The value of C is taken as 1, while CA
I
is taken as the mean of the water concentrations at the start
time and at 90 min. As shown in Fig. 11, there is an
approximate linear relationship between S and C - C ,
lnðC =CÞ ¼ kt
0
where C and C are the sulfur concentrations at the start
0
I
A
time and at time t (min), respectively, and k is the first-
supporting the hypothesis.
-
order-rate constant (min ). Plots of ln(C /C) versus time
1
0
were linear, as shown in Fig. 10. The rate constant k and
correlation coefficient R are listed in Table 1; the value of
k increases greatly as the volume ratio is changed from
Proposed mechanism for DBT oxidation
It is well known that the polyoxometallic anions can play
catalytic roles [9, 16, 20, 35]. In this ECODS process, a
probable reaction mechanism similar to the one proposed
1
0:2 to 10:5, but the increase is much smaller when the
volume ratio is above 10:5. On the basis of these results, a
volume ratio of 10:5 was chosen as the optimal condition.
Table 2 DBT conversion (%) over fresh and reused [C
6
mim]PMoV
Recycling capability
Drying
Run
1
Further experiments were conducted to investigate the pos-
sibility of recycling the catalyst. After an initial experiment,
the upper layer was taken off by decantation as completely as
possible, the lower layer was distilled under vacuum at 60 °C
for 4 h, and the dried residue was used in the next run. It was
found that the catalyst could be recycled six times with only a
2
3
4
5 6
Yes
No
100
100
99.1
92.5
98.8
81.5
98.8
72.4
96.3
63.5
96.1
55.7
Reaction conditions: 50 °C, 90 min, 10 ml model oil, m(cata-
lyst) = 0.04 g, V(S):V(O):V(PTA) = 20:1:6
1
23

Transition Met Chem (2014) 39:213–220
219
Scheme 1 Proposed reaction
mechanism involving V-peroxo
species
by Arichi et al. [29] is proposed, as shown in Scheme 1. In
this mechanism, the V-containing heteropoly anions are
oxidized by H O in step 1 and peroxide heteropoly anions
2. Eßer J, Wasserscheid P, Jess A (2004) Deep desulfurization of oil
refinery streams by extraction with ionic liquids. Green Chem
6
:316–322
2
2
3
4
5
. Otsuki S, Nonaka T, Takashima N, Qian W, Ishihara A, Imai T,
Kabe T (2000) Oxidative desulfurization of light gas oil and
vacuum gas oil by oxidation and solvent extraction. Energy Fuels
are generated in step 2. DBT is oxidized by the peroxide
heteropoly anions in step 3, after which the original state of
catalyst is restored in step 4.
1
4:1232–1239
. Ram ´ı rez-Verduzco LF, Torres-Garc ´ı a E, G o´ mez-Quintana R,
Gonz a´ lez-Pe n˜ a V, Murrieta-Guevara F (2004) Desulfurization of
diesel by oxidation/extraction scheme: influence of the extraction
solvent. Catal Today 98:289–294
. Kim JH, Ma X, Zhou A, Song C (2006) Ultra-deep desulfuriza-
tion and denitrogenation of diesel fuel by selective adsorption
over three different adsorbents: a study on adsorptive selectivity
and mechanism. Catal Today 111:74–83
Conclusions
Three hydrophobic molybdovanadophosphoric hetero-
polyacid-based catalysts were synthesized by reacting
molybdovanadophosphoric acid with ionic liquids and
investigated as catalysts for the ECODS process. The cat-
alyst with the longest carbon chain showed the highest
catalytic activity. The optimum temperature for the cata-
lytic reaction was 50 °C; above this, the decomposition of
6
7
. Zhao D, Li F, Sun Z (2008) Study on photosensitized oxidative
desulfurization of thiophene by riboflavin. J Fuel Chem Technol
36:161–164
. Soleimani M, Bassi A, Margaritis A (2007) Biodesulfurization of
refractory organic sulfur compounds in fossil fuels. Biotechnol
Adv 25:570–596
H O was too rapid for efficient catalysis. The extent of
2
2
8. Lu L, Cheng S, Gao J, Gao G, He M (2007) Deep oxidative
desulfurization of fuels catalyzed by ionic liquid in the presence
sulfur removal was close to 100 % under the optimum
conditions. The catalyst could be used six times with only a
slight decline in activity, provided it was dried before reuse
each time. These results may help in the development of
hydrophobic catalysts for deep oxidative desulfurization.
2 2
of H O . Energy Fuels 21:383–384
9
. Jiang X, Li H, Zhu W, He L, Shu H, Lu J (2009) Deep desul-
furization of fuels catalyzed by surfactant-type decatungstates
2 2
using H O as oxidant. Fuel 88:431–436
10. L u¨ H, Gao J, Jiang Z, Yang Y, Song B, Li C (2007) Oxidative
desulfurization of dibenzothiophene with molecular oxygen using
emulsion catalysis. Chem Commun 150–152
1. Wang J, Zhao D, Li K (2010) Oxidative desulfurization of
dibenzothiophene using ozone and hydrogen peroxide in ionic
liquid. Energy Fuels 24:2527–2529
2. Tam PS, Kittrell JR, Eldridge JW (1990) Desulfurization of fuel
oil by oxidation and extraction. 1. Enhancement of extraction oil
yield. Ind Eng Chem Res 29:321–324
3. Ishihara A, Wang D, Dumeignil F, Amano H, Qian EW, Kabe T
(2005) Oxidative desulfurization and denitrogenation of a light
gas oil using an oxidation/adsorption continuous flow process.
Appl Catal A 279:279–287
Acknowledgments We gratefully acknowledge the financial sup-
port of Hubei Provincial Department of Education of China
1
1
1
(
D20121401) for this work.
References
1
. Zhang H, GaoJ, MengH, LiC(2012)Removal ofthiophenicsulfurs
using an extractive oxidative desulfurization process with three new
phosphotungstate catalysts. Ind Eng Chem Res 51:6658–6665
123

2
1
20
Transition Met Chem (2014) 39:213–220
4. Chan NY, Lin TY, Yen TF (2008) Superoxides: alternative oxi-
dants for the oxidative desulfurization process. Energy Fuels
3 3 4 2
[C18H37N(CH ) ] [H NaPW10O36] catalyst assembled in emul-
sion droplets. J Catal 239:369–375
2
2:3326–3328
5. Ding Y, Zhu W, Li H, Jiang W, Zhang M, Duan Y, Chang Y
2011) Catalytic oxidative desulfurization with a hexatungstate/
aqueous H /ionic liquid emulsion system. Green Chem 13:
210–1216
25. Li J, Hu B, Hu C (2013) Deep desulfurization of fuels by het-
eropolyanion-based ionic liquid. Bull Korean Chem Soc 34:
225–230
26. Tsigdinos GA, Hallada CJ (1968) Molybdovanadophosphoric
acids and their salts. I. Investigation of methods of preparation
and characterization. Inorg Chem 7:437–441
27. Huddleston JG, Visser AE, Reichert WM, Willauer HD, Broker
GA, Rogers RD (2001) Characterization and comparison of
hydrophilic and hydrophobic room temperature ionic liquids
incorporating the imidazolium cation. Green Chem 3:156–164
28. Rajkumar T, Ranga Rao G (2008) Characterization of hybrid
molecular material prepared by 1-butyl 3-methyl imidazolium
bromide and phosphotungstic acid. Mater Lett 62:4134–4136
29. Arichi J, Eternot M, Louis B (2008) Synthesis of V-containing
Keggin polyoxometalates: versatile catalysts for the synthesis of
fine chemicals? Catal Today 138:117–122
1
(
2
O
2
1
1
6. Zhang B, Jiang Z, Li J, Zhang Y, Lin F, Liu Y, Li C (2012)
Catalytic oxidation of thiophene and its derivatives via dual
activation for ultra-deep desulfurization of fuels. J Catal 287:
5
–12
1
1
1
7. Zhao D, Wang Y, Duan E, Zhang J (2010) Oxidation desulfur-
ization of fuel using pyridinium-based ionic liquids as phase-
transfer catalysts. Fuel Process Technol 91:1803–1806
8. Swatloski RP, Holbrey JD, Rogers RD (2003) Ionic liquids are
not always green: hydrolysis of 1-butyl-3-methylimidazolium
hexafluorophosphate. Green Chem 5:361–363
9. Li C, Jiang Z, Gao J, Yang Y, Wang S, Tian F, Sun F, Sun X,
Ying P, Han C (2004) Ultra-deep desulfurization of diesel: oxi-
dation with a recoverable catalyst assembled in emulsion. Chem
A Eur J 10:2277–2280
30. Moffat JB, McMonagle JB, Taylor D (1988) Microporous het-
eropoly oxometalate heterogeneous catalysis. Solid State Ion
26:101–108
2
0. Zhang Y, L u¨ H, Wang L, Zhang Y, Liu P, Han H, Jiang Z, Li C
(2010) The oxidation of benzothiophene using the Keggin-type
31. Kozhevnikov IV (1998) Catalysis by heteropoly acids and mul-
ticomponent polyoxometalates in liquid-phase reactions. Chem
Rev 98:171–198
32. Rana S, Mallick S, Parida KM (2011) Selective oxidation of
benzaldehyde by molecular oxygen over molybdovanadophos-
phoric acid supported MCM-41. J Porous Mater 19:397–404
33. Zhang M, Zhu W, Xun S, Li H, Gu Q, Zhao Z, Wang Q (2013)
Deep oxidative desulfurization of dibenzothiophene with POM-
based hybrid materials in ionic liquids. Chem Eng J 220:328–336
34. Zhang J, Wang A, Li X, Ma X (2011) Oxidative desulfurization
lacunary polytungstophosphate as catalysts in emulsion. J Mol
Catal A Chem 332:59–64
1. Zhang Y, Wang L, Zhang Y, Jiang Z, Li C (2011) Ultra-deep
oxidative desulfurization of fuel oil catalyzed by Dawson-type
polyoxotungstate emulsion catalysts. Chin J Catal 32:235–239
2. Zhu W, Huang W, Li H, Zhang M, Jiang W, Chen G, Han C
2
2
2
(2011) Polyoxometalate-based ionic liquids as catalysts for deep
desulfurization of fuels. Fuel Process Technol 92:1842–1848
3. Huang D, Wang YJ, Yang LM, Luo GS (2006) Chemical oxi-
dation of dibenzothiophene with a directly combined amphiphilic
catalyst for deep desulfurization. Ind Eng Chem Res 45:
of dibenzothiophene and diesel over [Bmim]
279:269–275
3
PMo12O40. J Catal
35. Zhu J, Wang P, Lu M (2012) Catalytic wet air oxidation of
aromatic compounds: degradation in molybdovanadophosphoric
polyoxometalates micellar system under room temperature con-
ditions. J Iran Chem Soc 10:123–129
1
880–1885
2
4. Lu H, Gao J, Jiang Z, Jing F, Yang Y, Wang G, Li C (2006)
Ultra-deep desulfurization of diesel by selective oxidation with
1
23