Chemical Engineering Journal 250 (2014) 48–54

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

Ionic liquid extraction and catalytic oxidative desulfurization of fuels

using dialkylpiperidinium tetrachloroferrates catalysts
Wei Jiang a, Wenshuai Zhu a,⇑, Yonghui Chang a,b, Yanhong Chao a, Sheng Yin a, Hui Liu a, Fengxia Zhu a,
Huaming Li a,⇑
a
School of Chemistry and Chemical Engineering, Jiangsu University, Zhenjiang 212013, PR China
b
School of Chemistry and Chemical Engineering, Hainan Normal University, Haikou 571158, PR China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 A family of dialkylpiperidinium
tetrachloroferrate catalysts were
synthesized and characterized
systemically.
 Compared with the previous work,
the utilization of H2O2 was promoted.
 The interaction between DBT and
ionic liquid has been identified by
UV–Vis spectra.
 The oxidative product was
determined by GC–MS analysis.

a r t i c l e i n f o a b s t r a c t

Article history: A new family of dialkylpiperidinium tetrachloroferrate catalysts, such as [C2OHmpip]FeCl4,

Received 17 December 2013 [C4mpip]FeCl4, [C8mpip]FeCl4, and [C12mpip]FeCl4, were synthesized and characterized by FT-IR, Raman,
Received in revised form 15 March 2014 and ESI-MS. Their catalytic activities for oxidation of dibenzothiophene (DBT) in extraction and catalytic
Accepted 20 March 2014
oxidative desulfurization system (ECODS) were evaluated using different ionic liquids. Results indicated
Available online 28 March 2014
that the 97.1% removal of DBT was obtained with [C4mpip]FeCl4 as catalyst in ionic liquid [C8mim]BF4 at
30 °C in 60 min. The optimal H2O2/sulfur molar ratio was only 3.5:1, which suggested that the catalyst
Keywords:
was one of the most efficient catalysts reported so far. UV–Vis spectra provided an evidence that the
Dialkylpiperidinium
Tetrachloroferrates
lower desulfurization reactivity in PF 6 -containing ionic liquids was attributed to the strong interaction

Oxidative desulfurization between ionic liquid and DBT. Through the gas chromatography–mass spectrometer (GC–MS) analysis,
H2O2 dibenzothiophene sulfone was proved to be the only product of oxidation of DBT. Furthermore, the
Ionic liquid process of ECODS system was confirmed by GC–MS.
Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction temperature and pressure and consumes large amounts of hydrogen

Air pollution, caused by the emission of SOx, is one of the most
serious environmental problems in the world, and automobile
exhaust is the major source of SOx emanated from the combustion
of sulfur compounds in gasoline and diesel oil. Current
Hydrodesulfurization (HDS) technique must be operated at high
⇑ Corresponding authors. Tel.: +86 511 88791800; fax: +86 511 88791708.
E-mail addresses: zhuws@ujs.edu.cn (W. Zhu), lihm@ujs.edu.cn (H. Li).
[1–4]. Also some refractory sulfur compounds, such as dibenzothio-
phene (DBT) and its derivatives, are more difficult to be removed.
Thus, complementary desulfurization processes to HDS have been
investigated to achieve the lowest level of sulfur (<10 ppm) [5–8].
Oxidative desulfurization (ODS) has stolen the limelight owing
to mild operating conditions and no consumption of H2 [9–17].
Generally, ODS processes usually are achieved through two
steps. Firstly, sulfur compounds are selectively oxidized to
sulfoxides and sulfones, which would then be removed by

http://dx.doi.org/10.1016/j.cej.2014.03.074
1385-8947/Ó 2014 Elsevier B.V. All rights reserved.
 W. Jiang et al. / Chemical Engineering Journal 250 (2014) 48–54 49

appropriate extractants or adsorbents [18–22]. Organic solvents (DBT), and 4,6-dimethyldibenzothiophene (4,6-DMDBT) were
are usually utilized as extractants, which would cause further envi- purchased from Sigma–Aldrich. The ILs 1-butyl-3-methylimidazo-
ronmental and security problems as these solvents are generally lium tetrafluoroborate ([C4mim]BF4), 1-n-octyl-3-methylimidazo-
flammable and volatile organic compounds. Instead of the organic lium tetrafluoroborate ([C8mim]BF4), 1-butyl-3-methylimidazolium
solvents, ionic liquids (ILs) viewed as ‘‘green solvents’’ can address hexafluorophosphate ([C4mim]PF6) and 1-n-octyl-3-methylimida-
this issue in ODS processes [4,23–28]. zolium hexafluorophosphate ([C8mim]PF6) were purchased from
The interest in ILs has been growing rapidly due to the unique Shanghai Chengjie Chemical Co., Ltd. Other materials were
physical and chemical properties such as non-volatility, good ther- commercial reagent grade and were obtained from Shanghai
mal stability and recyclability [29–33]. Furthermore, the two steps Sinopharm Chemical Co., Ltd.
of ODS can be reduced to one by using ILs extraction and catalytic
oxidative desulfurization (ECODS) [34–36]. Recently, our group has
reported several tetrachloroferrate (FeCl 4 ) based ILs, such as
2.2. Preparation of the dialkylpiperidinium tetrachloroferrate catalysts
[C4mim]FeCl4 [37], [Et3NH]FeCl4 [38], [BPy][FeCl4] [39] and
([Rmpip]FeCl4)
[(C8H17)3CH3N]FeCl4 [40], in ECODS system. In these processes, ILs
acted as reaction media and extractants and 30 wt% H2O2 was
The preparation of dialkylpiperidinium tetrachloroferrate cata-
oxidant. They all showed good extraction and catalytic activity, how-
lysts are illustrated in Scheme 2. Firstly, an acetonitrile solution
ever, there were still some drawbacks that should be overcome. For
of 1-Methylpiperidine (mpip, 0.1 mol) and chloralkane (RCl,
example, the amount of ILs was quite large when the former two ILs
0.1 mol) was stirred at 60 °C for 48 h. The solvent was removed
were used as extractants (m(model oil)/m([C4mim]FeCl4) = 3:1,
under reduced pressure and remaining solid was washed with ace-
V(model oil)/V([Et3NH]FeCl4) = 5:2). For the latter two ILs, much
tone and diethyl ether to give a white powder dialkylpiperidinium
more H2O2 was required (n(H2O2)/n(S) > 8). We also have been stud-
chloride ([Rmpip]Cl) ([C8mpip]Cl, a yellow powder). Secondly,
ied an ECODS system with [(CH3)4N]FeCl4 as catalyst in ionic liquid
[Rmpip]FeCl4 was synthesized by stirring [Rmpip]Cl with and equi-
1-butyl-3-methylimidazolium tetrafluoroborate ([C4mim]BF4)
molar amount of ferric chloride in ethyl alcohol at room tempera-
[41], but large amount of H2O2 was still mandatory to achieve deep
ture for 4 h. The solvents were removed under reduced pressure
desulfurization. This implied the insufficient catalytic activity of
and the residue dried in vacuo at 80 °C for 24 h.
the previous catalysts. Therefore, more efficient catalysts should
The FT-IR spectra of all the catalysts (KBr pellets) were recorded
be developed from both industrial and green chemistry perspec-
with a Nicolet Nexus 470 FT-IR instrument. Ultraviolet–visible
tives. Furthermore, it will be interesting to study the effect of differ-
(UV–Vis) spectra were performed on UV-2450 spectrophotometer
ent catalytic systems or various types of catalysts on oxidative
(Shimadzu Corporation, Japan). Electrospray ionization mass
desulfurization.
(ESI–MS) spectra were collected on a LXQ Linear Ion Trap Mass
In this work, a new family of dialkylpiperidinium tetrachloro-
Spectrometer. Raman tests were carried out by using Thermo Sci-
ferrate catalysts, such as [C2OHmpip]FeCl4, [C4mpip]FeCl4,
entific DXR Smart Raman spectrometer equipped with a 532 nm
[C8mpip]FeCl4, and [C12mpip]FeCl4 (Scheme 1), were synthesized
excitation. NMR spectra were recorded on Bruker AV-400 spec-
and utilized as catalysts in the ECODS process with H2O2 as oxi-
trometer (Germany) using dimethyl sulfoxide (DMSO) and (CH3)4Si
dant. It was found that [C4mpip]FeCl4 exhibited highest activity
as the solvent and the reference, respectively.
in the oxidation of DBT and the sulfur removal could reach 97.1%
Data for [C2OHmpip]Cl: 1H NMR (400 MHz, DMSO), d: 5.78
in 60 min at 30 °C, in which ionic liquid 1-n-octyl-3-methylimida-
(t, 3JHH = 5.32 Hz, 1H), 3.81 (m, 2H), 3.48 (m, 6H), 3.14 (s, 3H),
zolium tetrafluoroborate ([C8mim]BF4) was used as extactant. The
1.78 (m, 4H), 1.53 (m, 2H). 13C NMR (100 MHz, DMSO), d: 19.86,
amount of H2O2 was only 0.273 mmol (H2O2/sulfur molar = 3.5),
21.03, 48.81, 54.88, 61.17, 63.92. ESI–MS: [C2OHmpip]+ 144.2.
which was much lower than that used in the previous work [40–
Data for [C4mpip]Cl: 1H NMR (400 MHz, DMSO), d: 3.37 (m, 6H),
42]. The difference of the desulfurization activities in various ILs,
3.04 (s, 3H), 1.76 (m, 4H), 1.63 (m, 2H), 1.52 (m, 2H), 1.30 (m, 2H),
such as [C4mim]BF4, [C8mim]BF4, 1-butyl-3-methylimidazolium
0.93 (t, 3JHH = 7.40 Hz), 3H). 13C NMR (100 MHz, DMSO), d: 14.01,
hexafluorophosphate ([C4mim]PF6) and 1-n-octyl-3-methylimida-
19.75, 19.79, 21.15, 24.46, 47.37, 60.21, 62.45. ESI–MS: [C4mpip]+
zolium hexafluorophosphate ([C8mim]PF6), were testified through
156.2.
UV–Vis spectra. Besides, the recycling of catalytic system contain-
Data for [C8mpip]Cl: 1H NMR (400 MHz, DMSO), d: 3.42 (m, 6H),
ing [C8mim]BF4 and [C4mpip]FeCl4 was also investigated. ECODS
3.07 (s, 3H), 1.74 (m, 4H), 1.63 (m, 2H), 1.52 (m, 2H), 1.24 (m, 2H),
process was researched and validated by GC–MS.
0.83 (t, 3JHH = 7.04 Hz), 3H). 13C NMR (100 MHz, DMSO), d: 14.33,
19.83, 21.07, 21.51, 22.50, 26.40, 29.00, 31.64, 47.21, 59.99,
2. Experimental 62.53. ESI–MS: [C8mpip]+ 212.4.
Data for [C12mpip]Cl: 1H NMR (400 MHz, DMSO), d: 3.2 (m, 6H),
2.1. Materials 3.01 (s, 3H), 1.77 (m, 4H), 1.64 (m, 2H), 1.53 (m, 2H), 1.25 (m, 18H),
0.86 (t, 3JHH = 7.0 Hz), 3H). 13C NMR (100 MHz, DMSO), d: 14.40,
1-Methylpiperidine was commercially available from Aladdin 19.77, 21.18, 21.43, 22.55, 26.37, 29.01, 29.17, 29.30, 29.40,
Chemical Co., Ltd. Benzothiophene (BT), dibenzothiophene 31.75, 47.40, 60.30, 62.81. ESI–MS: [C12mpip]+ 268.4.

CH2CH2OH C4H9 C8H17 C12H25

N N N N

FeCl4- FeCl4- FeCl4- FeCl4-

[C2OHmpip]FeCl4 [C4mpip]FeCl4 [C8mpip]FeCl4 [C12mpip]FeCl4

Scheme 1. Structure of the dialkylpiperidinium tetrachloroferrate catalysts.

50 W. Jiang et al. / Chemical Engineering Journal 250 (2014) 48–54

R R
A a C-O
N N N
60oC FeCl3
+ RCl Cl FeCl4 C-H
CH3CN O-H

Transmittance /%
C2H5OH b
mpip [Rmpip]Cl [Rmpip]FeCl 4
R = C2H4OH or C4H 9 or C 8H17 or C12H 25 c

Scheme 2. Synthetic pathway to [C2OHmpip]FeCl4, [C4mpip]FeCl4, [C8mpip]FeCl4

d
and [C12mpip]FeCl4.

CH2
ESI–MS of catalysts [Rmpip]FeCl4 are shown as follows:
4000 3500 3000 2500 2000 1500 1000 500
+
[C2OHmpip]FeCl4: [C2OHmpip] 144.2, [FeCl4] 197.7.  Wavenumber /cm -1
[C4mpip]FeCl4: [C4mpip]+ 156.2, [FeCl4] 197.7.
[C8mpip]FeCl4: [C8mpip]+ 212.4, [FeCl4] 197.7. B 332
[C12mpip]FeCl4: [C12mpip]+ 268.4, [FeCl4] 197.7.

2.3. Oxidative desulfurization procedure

a
Model oils were prepared by dissolving DBT, BT and 4,6-DMDBT

Intensity
in n-octane to give solutions with a sulfur content of 500, 250 and
250 ppm, respectively. In addition, tetradecane was added into the b
model oil as an internal standard. The extraction and catalytic oxi-
dative desulfurization experiments were performed in a stirred c
40 mL two-necked flask, to which catalyst, IL, model oil were
added in turn. Until the extractive equilibrium was reached after d
10 min, a certain amount of 30 wt% H2O2 was injected to the flask.
Then the mixture was stirred vigorously at 30 °C for 60 min. The 450 400 350 300 250 200
-1
model oil was separated by decantation after reaction for further Wavenumber /cm
analysis via gas chromatography–flame ionization detection (GC–
FID) (Agilent 7890A, HP-5 column, 30 m long  0.32 mm inner Fig. 1. (A) FT-IR spectra of catalysts. (B) Raman spectra of catalysts. (a)
[C2OHmpip]FeCl4; (b) [C4mpip]FeCl4; (c) [C8mpip]FeCl4 and (d) [C12mpip]FeCl4 .
diameter (id)  0.25 lm film thickness). The sulfur removal was
calculated as in Eq. (1), where C0 (ppm) is the initial sulfur concen-
tration in the model oil, Ct (ppm) is the transient sulfur concentra-
tion at any time t (min). The oxidized S-compound was 200–450 cm1, indicating their similar molecular structure. The
characterized Gas Chromatography–Mass Spectrometer (GC–MS) identical bands at 332 cm1 is assigned to the totally symmetric
(Agilent 7890/5975C-Gas Chromatography (GC)/Mass Selective Fe–Cl stretching vibration of FeCl
4 in the literature [43,44]. Thus,

Detector (MSD); HP-5 MS column, 30 m  250 lm i.d.  0.25 lm). it can be confirmed that the tested samples contain the same
[FeCl4] anion.
Sulfur removal ð%Þ ¼ ð1  C t =C 0 Þ  100 ð1Þ

3.2. Effect of different desulfurization systems on DBT removal

3. Results and discussion
3.1. Characterizations of dialkylpiperidinium tetrachloroferrates
catalysts ([Rmpip]FeCl4)
The FT-IR spectra of catalysts are shown in Fig. 1A. The similar
absorption peaks of catalysts were observed in the region of 4000–
400 cm1, which indicated that the four catalysts had similar cat-
ions. The bands at 2953 and 2873 cm1 come from the stretching
vibration absorption of C–H. The bands at 1462 and 1385 cm1
are due to the bending vibration of C–H. The characteristic peaks
of [C2OHmpip]FeCl4 at 3548 and 1090 cm1 are attributed to
the stretching vibration of O–H and C–O, respectively. For
[C8mpip]FeCl4 and [C12mpip]FeCl4, the bands found at about
723 cm1 arise from the bending vibration of –(CH2)n– (n P 4).
The FT-IR of the intermediate [Rmpip]Cl was also measured in
Fig. S1, the main characteristic bands of [Rmpip]FeCl4 are consis-
tent with that of [Rmpip]Cl, indicating that the cation is unchanged
during the preparation of catalysts.
Raman spectra of [C2OHmpip]FeCl4, [C4mpip]FeCl4, [C8mpip]FeCl4
and [C12mpip]FeCl4 were carried out in Fig. 1B. The Raman spectra
of the four catalysts are very well in keeping with each other in
Three different desulfurization systems including extraction,
extraction coupled with catalytic oxidation and catalytic oxidation
are shown in Table 1. DBT was selected as a representative sulfur
compound in fuel because it was one of the main refractory sulfur
compounds in the HDS treatment. The conversion of DBT in model
oil was used to calculate the removal of sulfur. The desulfurization
efficiency of extraction or catalytic oxidation was very poor. Obvi-
ously, the removal of DBT increased sharply with the extraction
coupled with catalytic oxidation. The desulfurization efficiency
could reach 92.1% with [C4mim]BF4 as extractant and 97.1% for
[C8mim]BF4. However, when [C4mim]PF6 and [C8mim]PF6 were
used, the sulfur removals were only 61.6% and 72.2%, respectively.
But the desulfurization efficiency was still higher than extraction
desulfurization. These results showed that the extraction and cat-
alytic oxidative desulfurization (ECODS) system was vastly supe-
rior to other two desulfurization systems, and the IL with BF 4
anion was more beneficial to the oxidation of DBT than the IL with
PF6 anion when [C4mpip]FeCl4 was used as catalysts.
The different sulfur removals among the ILs may attribute to the
interaction between ILs and DBT. UV–Vis spectra of DBT and DBT
extracted in ILs had been used to determine the interaction.
 W. Jiang et al. / Chemical Engineering Journal 250 (2014) 48–54 51

Table 1
Effect of different desulfurization systems on DBT removal.a A 255 DBT
DBT+ [C4 mim]BF4
Entry ILs S-removal (%)
235 DBT+ [C8 mim]BF4
Extractionb Extraction + Catalytic
oxidation oxidation
1 [C4mim]BF4 20.4 92.1

Abs.
2 [C4mim]PF6 17.3 61.6
3 [C8mim]BF4 36.9 97.1
4 [C8mim]PF6 37.1 72.2
5 5.8
a
Experimental conditions: n([C4mpip]FeCl4) = 0.078 mmol, model oil = 5 mL,
IL = 1 mL, n(H2O2) = 0.273 mmol, T = 30 °C, t = 60 min.
b
t = 10 min.
200 250 300 350 400
Wavelength /nm
As shown in Fig. 2A, an absorption band at 235 nm could be
observed for DBT in n-octane [45]. After DBT was extracted into
[C4mim]BF4 and [C8mim]BF4, this band shifted to 255 nm, suggest- B 235 DBT
ing that the structure of DBT was distorted in ILs. This might be [C 4 mim]PF6+DBT
attributed to the polarization of DBT by fluorine ion [46]. The [C 8 mim]PF6+DBT
absorption band at 235 nm disappeared in Fig. 2B when DBT was
extracted in
[C4mim]PF6 and [C8mim]PF6, indicating stronger interaction

Abs.
between PF 6 containing ILs and DBT. Thus, we conjectured that
the strong interaction led to hard oxidation of DBT.

3.3. Effect of piperidinium cations on DBT removal

Four kinds of catalysts with different piperidinium cations were

synthesized to investigate the effect of the cations of catalyst on 200 250 300 350 400
the removal of DBT. As shown in Fig. 3, the final desulfurization Wavelength /nm
efficiency at 60 min with the four catalysts was all greater than
95%. However, they had appreciable differences in the initial stage. Fig. 2. (A) UV–Vis spectra of DBT, DBT extracted in [C4mim]BF4 and in [C8mim]BF4.
(B) UV–Vis spectra of DBT, DBT extracted in [C4mim]PF6 and in [C8mim]PF6.
The catalyst with [C4mpip]+ cation gave the fastest reaction rate,
following by the catalyst with [C8mpip]+ and [C12mpip]+ cation.
The difference may be attributed to the influence of steric hin-
100
drance of different piperidinium cations [18]. The catalytic activity
of [C2OHmpip]FeCl4 was also lower than that of [C4mpip]FeCl4,
90
which was probably due to the hydroxyl that may trap the active
radical [47]. Compared with the previous work using FeCl3 and
Sulfur removal /%

80
[(CH3)4N]FeCl4 as catalysts [41,42], the dialkylpiperidinium tetra-
chloroferrates catalysts showed much higher catalytic activity [C4mpip] +
because less amount of H2O2 was used. H2O2/sulfur (O/S) molar 70
[C8 mpip] +
ratio was reduced from 6:1 for FeCl3 and 11.6:1 [(CH3)4N]FeCl4 to
3.5:1. Therefore, we could conclude that the ECODS system with 60 [C12 mpip] +
dialkylpiperidinium tetrachloroferrates catalysts was, to the best [C2OHmpip] +
of our knowledge, one of the most efficient systems using ionic liq- 50
uids reported so far.
40

3.4. Effect of reaction temperature and time on DBT removal 10 20 30 40 50 60

The effect of temperature and time on the ECODS system was
given Fig. 4. The sulfur removal reached 97.1%, 93.9% and 88.8%
at 60 min at 30 °C, 40 °C and 50 °C, respectively. The result sug-
gested that a higher reaction temperature was unfit for oxidation
of DBT due to the fact that the high temperature resulted in more
decomposition of H2O2 [40]. Therefore, there existed two parallel
reactions: the oxidation of DBT and thermal decomposition of
H2O2. However, the sulfur removal decreased in the order
50 °C > 40 °C > 30 °C at 10 min and 40 °C > 30 °C > 50 °C at 20 min.
This indicated that the higher reaction temperature could acceler-
ate the oxidation of DBT but it also made H2O2 decomposed more
severely. After 70 min, the sulfur removal increased a little to
97.1%, 94.0% and 88.8% at 30 °C, 40 °C and 50 °C, respectively.
Therefore, it was favorable to conduct the desulfurization reaction
at 30 °C in 60 min.
Reaction time /min
Fig. 3. Effect of piperidinium cations on DBT removal. Experimental conditions:
n(catalyst) = 0.078 mmol, model oil = 5 mL, [C8mim]BF4 = 1 mL, n(H2O2) = 0.273 -
mmol, T = 30 °C.
3.5. Effect of H2O2/DBT (O/S) molar ratio on DBT removal
The oxidation of DBT under various H2O2/DBT (O/S) molar ratios
were conducted at 30 °C to study the effect of the dosage of oxidant
on the oxidative properties. As shown in Table 1 and Table 2, O/S
molar ratio had a strong effect on the removal of DBT. Without
H2O2, the extraction desulfurization efficiency was only 36.9%. As
the O/S molar ratio increased from 1:1 to 4:1, the removal of
DBT from model oil increased from 73.4% to 97.2%. So it was clear
that the sulfur removal reached a plateau of 97.1% when the O/S
52 W. Jiang et al. / Chemical Engineering Journal 250 (2014) 48–54

100 100

95 90
Sulfur removal /%

Sulfur removal /%
90 80

70
85
0.156 mmol
o
30 C 60 0.078 mmol
80 0.039 mmol
40 o C
0.02 mmol
50 o C 50
75
40
70
30
10 20 30 40 50 60 70 10 20 30 40 50 60
Reaction time /min Reaction time /min

Fig. 4. Effect of temperature and reaction time on sulfur removal. Experimental Fig. 5. Effect of the amount of the catalyst on removal of DBT. Experimental
conditions: n([C4mpip]FeCl4) = 0.078 mmol, model oil = 5 mL, [C8mim]BF4 = 1 mL, conditions: model oil = 5 mL, [C8mim]BF4 = 1 mL, n(H2O2) = 0.273 mmol, T = 30 °C.
n(H2O2) = 0.273 mmol.

100 DBT
Table 2 BT
Effect of H2O2/DBT (O/S) molar ratio on removal of DBT.a 4,6-DMDBT
90

Sulfur removal /%
n(H2O2)/n(S) 1:1 2:1 3:1 3.5:1 4:1
S-removal (%) 73.4 91.7 95.8 97.1 97.2 80
a
Experimental conditions: n([C4mpip]FeCl4) = 0.078 mmol, T = 30 °C, t = 60 min,
70
model oil = 5 mL, [C8mim]BF4 = 1 mL.

60

50
molar ratio reached 3.5:1. Stoichiometrically, 2 mol of H2O2 are
consumed for oxidation of the sulfur-containing compounds to 40
their corresponding sulfones. However, a little excess H2O2 was
required for promoting the equilibrium reaction due to the H2O2 10 20 30 40 50 60
decomposition. Therefore, the optimal O/S molar ratio was 3.5:1, Reaction time /min
taking into account the economic point of view.
Fig. 6. Oxidation of different sulfur-containing compounds. Experimental condi-
tions: n([C4mpip]FeCl4) = 0.078 mmol, n(H2O2) = 0.273 mmol, model oil = 5 mL,
3.6. Effect of the amount of the catalyst on DBT removal [C8mim]BF4 = 1 mL, T = 30 °C.

The effect of the amount of the catalyst [C4mpip]FeCl4 on desul-

furization is shown in Fig. 5. The results indicated that the amount
of the catalyst had a vital effect on the removal of DBT. When
0.02 mmol catalyst was used, the sulfur removal increased slowly
with increasing reaction time. The sulfur removal reached 81.1%
with 0.039 mmol catalyst was used. Further increasing the amount
of catalyst from 0.078 mmol mg to 0.156 mmol promoted the
desulfurization reaction and the sulfur removal increased to
97.1% and 97.3%. Obviously, the excessive catalyst was not neces-
sary and economical in the process. Thus, 0.078 mmol was chosen
as the most suitable quantity in this study.
3.7. Effect of the feature of the substrates
The reactivity of different sulfur-containing compounds, includ-
ing DBT, BT and 4,6-DMDBT, was estimated in the ECODS system
using [C4mpip]FeCl4 as catalyst. The results in Fig. 6 displayed that
sulfur removal of DBT, BT and 4,6-DMDBT were 97.1%, 76.6% and
56.3%, respectively. From the data, it can be seen that the intrinsic
properties of the sulfur-containing compounds were a critical fac-
tor for desulfurization reaction. These differences mainly came
from two factors, that is, the electron density and steric hindrance
on sulfur atoms. As calculated by Otsuki et al. [48], the electron
density data on the sulfur atom of DBT, BT and 4,6-DMDBT is
5.758, 5.739 and 5.760, respectively. Therefore, the lowest reactiv-
ity of 4,6-DMDBT was mainly affected by the steric hindrance of
the methyl groups, which was a main obstacle for the approach
of the sulfur atom to the catalytic active species. For DBT and BT,
lower electron density on the sulfur atom of BT led to a lower reac-
tivity. Therefore, the catalytic oxidation reactivity of the different
substrates was in the following order: DBT > BT > 4,6-DMDBT.
3.8. Regeneration/recycling of catalytic system
The recycling performance of the catalytic system containing
[C4mpip]FeCl4 and [C8mim]BF4 was investigated on the removal
of DBT in model oil. The oil, water and the IL phase with catalyst
was in upper layer, middle layer and lower layer respectively after
each reaction. The model oil could be separated by decantation
from the triphasic system. Water, residual H2O2 and oil were
removed under reduced pressure at 50 °C for 8 h. Then, fresh
model diesel and H2O2 were introduced for the next cycle under
the same conditions. From the data in Table 3, it was found that
the desulfurization system could be recycled for four times with-
out an obvious decrease in the catalytic activity. However, the sul-
fur removal decreased to 87.0% after recycling for five times.
According to our observation in Fig. 7, as the recycling times
increased, the IL phase became more and more turbid, which
was ascribed to the oxidative product of DBT in the reaction pro-
cess. Therefore, the continuous decreasing desulfurization effi-
ciency may arise from the increasing oxidative product.
 W. Jiang et al. / Chemical Engineering Journal 250 (2014) 48–54 53

Table 3 that DBT could be extracted by [C8mim]BF4 from the oil phase into
Recycling performance of the catalytic system containing [C4mpip]FeCl4 and the IL phase, where they were further oxidized. In order to further
[C8mim]BF4.a
research the reaction process of ECODS system, sulfur compounds
Cycle 1 2 3 4 5 in model oil and in IL phase were measured by GC–MS analysis.
S-removal (%) 97.1 96.6 92.9 90.2 87.0 After reaction, the upper layer was separated and IL phase was
a
Experimental conditions: model oil = 5 mL, n([C4mpip]FeCl4) = 0.078 mmol, extracted by diethyl ether, and then the model oil and the resulting
[C8mim]BF4 = 1 mL, n(H2O2) = 0.273 mmol, T = 30 °C, t = 60 min. diethyl ether phase were analyzed by GC–MS. The results were
shown in Fig. 8. In oil phase, only DBT was detected while both
DBT and DBTO2 were found in IL phase. Therefore, we could con-
clude that the sulfur compounds such as DBT were extracted into
IL firstly, and then was oxidized to DBTO2.

4. Conclusion

Deep oxidative desulfurization has been achieved by the extrac-

tion and oxidative desulfurization system [C4mpip]FeCl4/[C8mim]BF4
in the presence of H2O2. The reaction conditions, including reaction
time, temperature, H2O2/DBT molar ratio, amount of catalyst, and
Fig. 7. The photos of the catalytic system containing [C4mpip]FeCl4 and
alkyl chain of catalyst cation, have been investigated in detail.
[C8mim]BF4. (a) before reaction; (b) after first cycle and (c) after fifth cycle. Compared with the previous work, [C4mpip]FeCl4 showed better
catalytic activity because of the low dosage of H2O2. It is notewor-
thy that the strong interaction between PF 6 containing ionic liq-
uids and DBT led to a low desulfurization efficiency, which was
A proved by UV–Vis spectra. Through the GC–MS analysis, the pro-
184.0
cess of ECODS system could be confirmed that DBT was firstly
extracted into IL phase and oxidized to DBTO2.
DBT
Acknowledgements
Abundance

s
4 6 8 10 12 14 16
Retention time /min The authors are grateful for financial supported by the National
Nature Science Foundation of China (Nos. 21106055, 21276117,
21266007), the Natural Science Foundation of Jiangsu Province
(Nos. BK2011506, BK2012697), Postdoctoral Foundation of China
139.0 and Jiangsu Province (No. 1301001A).
92.1 152.0

Appendix A. Supplementary material

60 80 100 120 140 160 180 200
m/z Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.cej.2014.03.074.
B DBT 216.0
DBTO2 References

[1] H.Y. Lu, W.Z. Ren, W.P. Liao, W. Chen, Y. Li, Z.H. Suo, Aerobic oxidative
desulfurization of model diesel using a B-type Anderson catalyst
s [(C18H37)2N(CH3)2]3Co(OH)6Mo6O183H2O, Appl. Catal., B 138 (2013) 79–83.
Abundance

o o [2] Y.W. Hu, Q.H. He, Z. Zhang, N.D. Ding, B.X. Hu, Oxidative desulfurization of
4 6 8 10 12 14 16
dibenzothiophene with hydrogen peroxide catalyzed by selenium(IV)-
Retention time /min 187.0 containing peroxotungstate, Chem. Commun. 47 (2011) 12194–12196.
[3] H.Y. Song, J.J. Gao, X.Y. Chen, J. He, C.X. Li, Catalytic oxidation-extractive
139.1160.0 desulfurization for model oil using inorganic oxysalts as oxidant and Lewis
acid–organic acid mixture as catalyst and extractant, Appl. Catal., A 456 (2013)
115.1 67–74.
79.1 [4] J.H. Xu, S. Zhao, W. Chen, M. Wang, Y.F. Song, Highly efficient extraction and
oxidative desulfurization system using Na7H2LaW10O3632H2O in bmim BF4 at
room temperature, Chem. Eur. J. 18 (2012) 4775–4781.
[5] F. Lin, D.G. Wang, Z.X. Jiang, Y. Ma, J. Li, R.G. Li, C. Li, Photocatalytic oxidation of
20 40 60 80 100 120 140 160 180 200 220 thiophene on BiVO4 with dual co-catalysts Pt and RuO2 under visible light
m/z irradiation using molecular oxygen as oxidant, Energy Environ. Sci. 5 (2012)
6400–6406.
Fig. 8. (A) GC–MS analysis of sulfur compounds in model oil after reaction. (B) GC– [6] C. Sentorun-Shalaby, S.K. Saha, X.L. Ma, C.S. Song, Mesoporous-molecular-
MS analysis of sulfur compounds in IL after reaction. sieve-supported nickel sorbents for adsorptive desulfurization of commercial
ultra-low-sulfur diesel fuel, Appl. Catal., B 101 (2011) 718–726.
[7] Z. Li, C.P. Li, Y.S. Chi, A.L. Wang, Z.D. Zhang, H.X. Li, Q.S. Liu, U. Welz-Biermann,
3.9. The proposed reaction process of ECODS system Extraction process of dibenzothiophene with new distillable amine-based
protic ionic liquids, Energy Fuels 26 (2012) 3723–3727.
[8] Y. Chen, S. Zhao, Y.-F. Song, An efficient heterogeneous catalyst based on highly
Comparative experiments using [C4mpip]FeCl4/H2O2 as desul-
dispersed Na7H2LaW10O3632H2O nanoparticles on mesoporous silica for deep
furization systems showed almost no sulfur removal in the absence desulfurization, Appl. Catal., A 466 (2013) 307–314.
of IL (Table 1, entry 5), which suggested that oxidation does not [9] X.M. Yan, P. Mei, L. Xiong, L. Gao, Q.F. Yang, L.J. Gong, Mesoporous titania–
take place without IL. With addition of [C8mim]BF4, sulfur removal silica–polyoxometalate nanocomposite materials for catalytic oxidation
desulfurization of fuel oil, Catal. Sci. Technol. 3 (2013) 1985–1992.
increased significantly, and this suggested that DBT had been [10] F.T. Li, Y. Liu, Z.M. Sun, Y. Zhao, R.H. Liu, L.J. Chen, D.S. Zhao, Photocatalytic
oxidized in the presence of [C8mim]BF4. This experiment showed oxidative desulfurization of dibenzothiophene under simulated sunlight
54 W. Jiang et al. / Chemical Engineering Journal 250 (2014) 48–54
irradiation with mixed-phase Fe2O3 prepared by solution combustion, Catal.
Sci. Technol. 2 (2012) 1455–1462.
[11] N.F. Tang, Y.N. Zhang, F. Lin, H.Y. Lu, Z.X. Jiang, C. Li, Oxidation of
dibenzothiophene catalyzed by [C8H17N(CH3)3]3H3V10O28 using molecular
oxygen as oxidant, Chem. Commun. 48 (2012) 11647–11649.
[12] M. Zhang, W.S. Zhu, S.H. Xun, H.M. Li, Q.Q. Gu, Z. Zhao, Q. Wang, Deep
oxidative desulfurization of dibenzothiophene with POM-based hybrid
materials in ionic liquids, Chem. Eng. J. 220 (2013) 328–336.
[13] W.H. Wang, G. Li, W.G. Li, L.P. Liu, Synthesis of hierarchical TS-1 by caramel
templating, Chem. Commun. 47 (2011) 3529–3531.
[14] M. Riad, S. Mikhail, Oxidative desulfurization of light gas oil using zinc
catalysts prepared via different techniques, Catal. Sci. Technol. 2 (2012) 1437–
1446.
[15] J.J. Gao, W.Y. Ma, L.P. Yuan, Y.F. Dai, C.X. Li, Catalytic oxidative desulfurization
mechanism in Lewis–Brønsted complex acid, Appl. Catal., A 467 (2013) 187–
195.
[16] H.W. Zheng, Z. Sun, X.L. Chen, Q.D. Zhao, X.H. Wang, Z.J. Jiang, A micro
reaction-controlled phase-transfer catalyst for oxidative desulfurization based
on polyoxometalate modified silica, Appl. Catal., A 467 (2013) 26–32.
[17] J. Zhang, A.J. Wang, Y.J. Wang, H.Y. Wang, J.Z. Gui, Heterogeneous oxidative
desulfurization of diesel oil by hydrogen peroxide: catalysis of an amphipathic
hybrid material supported on SiO2, Chem. Eng. J. 245 (2014) 65–70.
[18] H.Y. Lu, W.Z. Ren, H.Y. Wang, Y. Wang, W. Chen, Z.H. Suo, Deep desulfurization
of diesel by ionic liquid extraction coupled with catalytic oxidation using an
Anderson-type catalyst [(C4H9)]4N4NiMo6O24H6, Appl. Catal., A 453 (2013)
376–382.
[19] W.S. Zhu, G.P. Zhu, H.M. Li, Y.H. Chao, Y.H. Chang, G.Y. Chen, C.R. Han,
Oxidative desulfurization of fuel catalyzed by metal-based surfactant-type
ionic liquids, J. Mol. Catal. A: Chem. 347 (2011) 8–14.
[20] C.S. Song, X.L. Ma, New design approaches to ultra-clean diesel fuels by deep
desulfurization and deep dearomatization, Appl. Catal., B 41 (2003) 207–238.
[21] G.H. Gao, S.F. Cheng, Y. An, X.J. Si, X.L. Fu, Y.M. Liu, H.J. Zhang, P. Wu, M.Y. He,
Oxidative desulfurization of aromatic sulfur compounds over titanosilicates,
ChemCatChem 2 (2010) 459–466.
[22] Z.E.A. Abdalla, B.S. Li, Preparation of MCM-41 supported (Bu4N)4H3(PW11O39)
catalyst and its performance in oxidative desulfurization, Chem. Eng. J. 200
(2012) 113–121.
[23] G.R. Yu, J.J. Zhao, D.D. Song, C. Asumana, X.Y. Zhang, X.C. Chen, Deep oxidative
desulfurization of diesel fuels by acidic ionic liquids, Ind. Eng. Chem. Res. 50
(2011) 11690–11697.
[24] H.S. Gao, C. Guo, J.M. Xing, J.M. Zhao, H.Z. Liu, Extraction and oxidative
desulfurization of diesel fuel catalyzed by a Bronsted acidic ionic liquid at
room temperature, Green Chem. 12 (2010) 1220–1224.
[25] J.Z. Gui, D. Liu, Z.L. Sun, D.S. Liu, D. Min, B. Song, X.L. Peng, Deep oxidative
desulfurization with task-specific ionic liquids: an experimental and
computational study, J. Mol. Catal. A: Chem. 331 (2010) 64–70.
[26] Y.S. Chi, C.P. Li, Q.Z. Jiao, Q.S. Liu, P.F. Yan, X.M. Liu, U. Welz-Biermann,
Desulfurization by oxidation combined with extraction using acidic room-
temperature ionic liquids, Green Chem. 13 (2011) 1224–1229.
[27] F.T. Li, C.G. Kou, Z.M. Sun, Y.J. Hao, R.H. Liu, D.S. Zhao, Deep extractive and
oxidative desulfurization of dibenzothiophene with C5H9NOSnCl2 coordinated
ionic liquid, J. Hazard. Mater. 205 (2012) 164–170.
[28] M. Zhang, W.S. Zhu, H.M. Li, S.H. Xun, W.J. Ding, J.J. Liu, Z. Zhao, Q. Wang,
One-pot synthesis, characterization and desulfurization of functional
mesoporous W-MCM-41 from POM-based ionic liquids, Chem. Eng. J. 243
(2014) 386–393.
[29] E. Lissner, W.F. de Souza, B. Ferrera, J. Dupont, Oxidative desulfurization of
fuels with task-specific ionic liquids, ChemSusChem 2 (2009) 962–964.
[30] Y. Nie, Y.X. Dong, L. Bai, H.F. Dong, X.P. Zhang, Fast oxidative desulfurization of
fuel oil using dialkylpyridinium tetrachloroferrates ionic liquids, Fuel 103
(2013) 997–1002.
[31] W.D. Liang, S. Zhang, H.F. Li, G.D. Zhang, Oxidative desulfurization of simulated
gasoline catalyzed by acetic acid-based ionic liquids at room temperature, Fuel
Process. Technol. 109 (2013) 27–31.
[32] Y.L. Gu, Multicomponent reactions in unconventional solvents: state of the art,
Green Chem. 14 (2012) 2091–2128.
[33] S.H. Ren, Y.C. Hou, W.Z. Wu, X.T. Chen, J.L. Fan, J.W. Zhang, Effect of H2O on the
desulfurization of simulated flue gas by an ionic liquid, Ind. Eng. Chem. Res. 48
(2009) 4928–4932.
[34] W.S. Zhu, W.L. Huang, H.M. Li, M. Zhang, W. Jiang, G.Y. Chen, C.R. Han,
Polyoxometalate-based ionic liquids as catalysts for deep desulfurization of
fuels, Fuel Process. Technol. 92 (2011) 1842–1848.
[35] W.S. Zhu, H.M. Li, Q.Q. Gu, P.W. Wu, G.P. Zhu, Y.S. Yan, G.Y. Chen, Kinetics and
mechanism for oxidative desulfurization of fuels catalyzed by peroxo-
molybdenum amino acid complexes in water-immiscible ionic liquids, J.
Mol. Catal. A: Chem. 336 (2011) 16–22.
[36] X.C. Chen, D.D. Song, C. Asumana, G.R. Yu, Deep oxidative desulfurization of
diesel fuels by Lewis acidic ionic liquids based on 1-n-butyl-3-
methylimidazolium metal chloride, J. Mol. Catal. A: Chem. 359 (2012) 8–13.
[37] H.M. Li, W.S. Zhu, Y. Wang, J.T. Zhang, J.D. Lu, Y.S. Yan, Deep oxidative
desulfurization of fuels in redox ionic liquids based on iron chloride, Green
Chem. 11 (2009) 810–815.
[38] W.S. Zhu, J.T. Zhang, H.M. Li, Y.H. Chao, W. Jiang, S. Yin, H. Liu, Fenton-like ionic
liquids/H2O2 system: one-pot extraction combined with oxidation
desulfurization of fuel, RSC Adv. 2 (2012) 658–664.
[39] W.S. Zhu, P.W. Wu, L. Yang, Y.H. Chang, Y.H. Chao, H.M. Li, Y.Q. Jiang, W. Jiang,
S.H. Xun, Pyridinium-based temperature-responsive magnetic ionic liquid for
oxidative desulfurization of fuels, Chem. Eng. J. 229 (2013) 250–256.
[40] Y.Q. Jiang, W.S. Zhu, H.M. Li, S. Yin, H. Liu, Q.J. Xie, Oxidative desulfurization of
fuels catalyzed by Fenton-like ionic liquids at room temperature,
ChemSusChem 4 (2011) 399–403.
[41] W. Jiang, W.S. Zhu, H.M. Li, J. Xiong, S.H. Xun, Z. Zhao, Q. Wang, Deep oxidative
desulfurization of fuels catalyzed by magnetic Fenton-like hybrid catalysts in
ionic liquids, RSC Adv. 3 (2013) 2355–2361.
[42] J.T. Zhang, W.S. Zhu, H.M. Li, W. Jiang, Y.Q. Jiang, W.L. Huang, Y.S. Yan, Deep
oxidative desulfurization of fuels by Fenton-like reagent in ionic liquids, Green
Chem. 11 (2009) 1801–1807.
[43] S. Hayashi, H.O. Hamaguchi, Discovery of a magnetic ionic liquid bmim FeCl4,
Chem. Lett. 33 (2004) 1590–1591.
[44] X.W. Pei, Y.H. Yan, L.Y. Yan, P. Yang, J.L. Wang, R. Xu, M.B. Chan-Park,
A magnetically responsive material of single-walled carbon nanotubes
functionalized with magnetic ionic liquid, Carbon 48 (2010) 2501–2505.
[45] H.Y. Lu, Y.N. Zhang, Z.X. Jiang, C. Li, Aerobic oxidative desulfurization of
benzothiophene, dibenzothiophene and 4,6-dimethyldibenzothiophene using
an Anderson-type catalyst [(C18H37)2N(CH3)2]5IMo6O24, Green Chem. 12
(2010) 1954–1958.
[46] B.Y. Zhang, Z.X. Jiang, J. Li, Y.N. Zhang, F. Lin, Y. Liu, C. Li, Catalytic oxidation of
thiophene and its derivatives via dual activation for ultra-deep desulfurization
of fuels, J. Catal. 287 (2012) 5–12.
[47] W. Jiang, W.S. Zhu, H.M. Li, Y.H. Chao, S.H. Xun, Y.H. Chang, H. Liu, Z. Zhao,
Mechanism and optimization for oxidative desulfurization of fuels catalyzed
by Fenton-like catalysts in hydrophobic ionic liquid, J. Mol. Catal. A: Chem. 382
(2014) 8–14.
[48] S. Otsuki, T. Nonaka, N. Takashima, W.H. Qian, A. Ishihara, T. Imai, T. Kabe,
Oxidative desulfurization of light gas oil and vacuum gas oil by oxidation and
solvent extraction, Energy Fuels 14 (2000) 1232–1239.