Effect of the nature of sulfur compounds on their reactivity in the oxidative desulfurization of hydrocarbon fuels with oxygen over a modified CuZnAlO catalyst

Article abstract of DOI:10.1134/S0023158417010128

The reactivity of thiophene, dibenzothiophene (DBT), and 4,6-dimethyldibenzothiophene (4,6-DMDBT), which are the representatives of the main classes of sulfur compounds that are the constituents of diesel fractions, was studied in the course of their oxidative desulfurization with oxygen on a CuO/ZnO/Al₂O₃ catalyst modified with boron and molybdenum additives. At T ≥ 375°C, the reactivity increased in the order thiophene < DBT < 4,6-DMDBT. The degree of sulfur removal in the form of SO₂ from hydrocarbon fuel, which was simulated by a solution of 4,6-DMDBT in toluene, was 80%. Under the assumption of a first order reaction with respect to sulfur compound and oxygen, the apparent activation energies of the test processes were calculated. An attempt was made to reveal the role of the adsorption of sulfur compounds in the overall process of oxidative desulfurization with the use of X-ray diffraction analysis, X-ray photoelectron spectroscopy, and differential thermal and thermogravimetric analysis with the massspectrometric monitoring of gas phase composition.

Full text of DOI:10.1134/S0023158417010128

ISSN 0023-1584, Kinetics and Catalysis, 2017, Vol. 58, No. 1, pp. 58–72. © Pleiades Publishing, Ltd., 2017.
Original Russian Text © S.A. Yashnik, A.V. Salnikov, M.A. Kerzhentsev, A.A. Saraev, V.V. Kaichev, L.M. Khitsova, Z.R. Ismagilov, J. Yamin, O.R. Koseoglu, 2017, published
in Kinetika i Kataliz, 2017, Vol. 58, No. 1, pp. 62–77.
Effect of the Nature of Sulfur Compounds on Their Reactivity
in the Oxidative Desulfurization of Hydrocarbon Fuels
with Oxygen over a Modified CuZnAlO Catalyst
a,
a
a
a
a
b
S. A. Yashnik *, A. V. Salnikov , M. A. Kerzhentsev , A. A. Saraev , V. V. Kaichev , L. M. Khitsova ,
a, b
c
c
Z. R. Ismagilov , J. Yamin , and O. R. Koseoglu
a
Boreskov Institute of Catalysis, Siberian Branch, Russian Academy of Sciences, Novosibirsk, 630090 Russia
b
Institute of Coal Chemistry and Materials Science, Siberian Branch, Russian Academy of Sciences,
Kemerovo, 650000 Russia
c
Saudi Aramco Research and Development Center, 31311, Dhahran, Saudi Arabia
*
e-mail: yashnik@catalysis.ru
Received April 21, 2016
Abstract—The reactivity of thiophene, dibenzothiophene (DBT), and 4,6-dimethyldibenzothiophene
(
4,6-DMDBT), which are the representatives of the main classes of sulfur compounds that are the constitu-
ents of diesel fractions, was studied in the course of their oxidative desulfurization with oxygen on a
CuO/ZnO/Al O catalyst modified with boron and molybdenum additives. At T ≥ 375°C, the reactivity
increased in the order thiophene < DBT < 4,6-DMDBT. The degree of sulfur removal in the form of SO
2
3
2
from hydrocarbon fuel, which was simulated by a solution of 4,6-DMDBT in toluene, was 80%. Under the
assumption of a first order reaction with respect to sulfur compound and oxygen, the apparent activation
energies of the test processes were calculated. An attempt was made to reveal the role of the adsorption of sul-
fur compounds in the overall process of oxidative desulfurization with the use of X-ray diffraction analysis,
X-ray photoelectron spectroscopy, and differential thermal and thermogravimetric analysis with the mass-
spectrometric monitoring of gas phase composition.
Keywords: thiophene, dibenzothiophene, oxidative desulfurization in a vapor–gas phase, CuZnAl catalyst,
modifiers
DOI: 10.1134/S0023158417010128
INTRODUCTION
thermore, the alkyl-substituted derivatives of DBT
and naphthothiophene are characterized by high boil-
ing points; this leads to their concentration in high-
boiling fractions upon the direct distillation of oil [7].
The stringent requirements imposed on the quality
of motor fuels stimulate search for new methods for
the removal of sulfur-containing compounds from
them and an improvement of currently available pro-
Among the alternative processes of fuel desulfur-
cesses. Hydrodesulfurization in the presence of sup- ization, technologies based on the use of adsorption,
ported NiMo and CoMo sulfide catalysts is widely extraction, and selective oxidation are of the greatest
used for the purification of hydrocarbon fuels. Sul- interest [7]. An analysis of scanty patent [8] and scien-
fides, disulfides, mercaptans, thiophene, benzothio- tific [9, 10] literature showed that the process of desul-
phene (BT), and their alkyl-substituted derivatives can furization by the selective oxidation of sulfur-contain-
be easily removed by C–S bond hydrogenolysis [1–3]. ing compounds with atmospheric oxygen in the pres-
Sulfur-containing compounds, such as dibenzothio- ence of selective catalysts is sufficiently promising, in
phene (DBT) and its alkyl-substituted derivatives, particular, for the removal of stable heterocyclic com-
which are resistant to hydrogenolysis, are more diffi- pounds of sulfur from heavy oils. Its advantages consist
cult to remove. In the case of dialkyl-substituted in the possibility of using atmospheric oxygen as a
DBTs, this is related to steric hindrances, which cheap oxidizer and implementing the process at atmo-
appear due to the close arrangement of alkyl substitu- spheric pressure [4, 7, 11–13]. However, the oxidative
ents on the coordination of a sulfur-containing mole- process has an essential disadvantage, namely, the par-
cule through a lone electron pair of the sulfur atom to ticipation of aromatic hydrocarbons and branched
an active center of the catalyst [1–5]. However, these alkanes in partial and deep oxidation processes, which
compounds should be removed in order to obtain fuel lead to a loss of the hydrocarbon fraction and a
with a sulfur content of lower than 10 ppm [4, 6]. Fur- decrease in the cetane number.
58

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
59
The high oxygen affinity of sulfur-containing com- performed on them were studied by X-ray diffraction
pounds [14] and the smaller values of the irreversible analysis and X-ray photoelectron spectroscopy. The
oxidation potentials of sulfur compounds, as com- experimental data were compared with the results of
pared with the oxidation potentials of hydrocarbons, differential thermal and thermogravimetric analysis
form a theoretical basis of the oxidative desulfurization with the mass-spectrometric monitoring of gas phase
method [14]. Because of this, oxygen is selectively composition (DTA-TG-MS).
added to the organic derivatives of thiophene without
C–C and C–S bond cleavage to oxidize them to sul-
EXPERIMENTAL
Catalyst Synthesis
fones and sulfoxides [14]. Selective oxidation to sul-
fones and sulfoxides changes the polarity of sulfur
compounds [13] and makes it possible to readily
remove them by extraction or adsorption.
The CuO–ZnO–Al O catalyst was synthesized by
2
3
a precipitation method with the addition of a mixed
The high reactivity of heterocyclic sulfur-contain- solution of sodium hydroxide and sodium carbonate
ing compounds in their oxidation into sulfones and to a mixture of the solutions of corresponding nitrates
sulfoxides was confirmed based on an example of liq- (molar ratio, 12 : 4.2 : 14). The procedure of the syn-
uid-phase processes, where hydrogen peroxide [11–14] thesis was described in more detail previously [22].
or organic hydroperoxides [11, 12] served as oxidizing After calcination, the catalyst contained 50 wt % CuO,
agents. For example, in a homogeneous system based 15 wt % ZnO, 30 wt % Al O , and the balance water
2
3
on H O and formic acid [15–17], the reactivity of the
2−
2
2
and СО . Then, the resulting sample was modified
with the additives of MoO (5 wt %) and boron (1 wt %)
for increasing the sorption capacity and regulating the
acidity of the catalyst, respectively. The modified cat-
alyst was dried at 110 and calcined at 500°C.
3
organic compounds of sulfur correlated with an
increase in electron density on the sulfur atom in the
order BT < DBT < sulfides < thiols [16, 17]. In the
case of peroxide oxidation in the presence of heteroge-
3
neous catalysts, such as phosphotungstic acid on SiO
2
[
18], MoO /Al O [19], or Ti–zeolite BEA [20], the
3 2 3
Physicochemical Catalyst Characterization
reactivity increased in the following sequence: thio-
phene BT 4,6-dimethyldibenzothiophene
<
<
The chemical composition of the freshly prepared
(
4,6-DMDBT) < DBT. In some heterogeneous cata- catalyst was studied after its dissolution in a solution of
lytic systems, the number and length of alkyl substitu- sulfuric acid (45 wt %) by inductively coupled plasma
ents influence the ability of sulfur-containing com- atomic absorption spectrometry (ICP AES) on an
pounds to undergo oxidation with hydrogen peroxide. OPTIMA 4300DV spectrometer (Perkin Elmer, the
In particular, in the presence of a vanadium catalyst, the United States).
oxidizing activity of sulfur-containing compounds dis-
The textural characteristics of the sample (its spe-
solved in hexadecane [21] decreased with the number of
cific surface area S
, the pore volume V , and the
spBET
Σ
substituents in the order 4-MDBT > 2-methylthio-
pore diameter d) were studied by the low-temperature
phene > 2,5-dimethylthiophene > 4,6-DMDBT.
adsorption of nitrogen. The N adsorption–desorption
2
Information on the reactivity of sulfur-containing
compounds on their oxidation in a vapor–gas phase is
absent. An analysis of publications [9, 10] showed that
the concentration of oxygen in a reaction mixture
should be increased and the mass flow rate of fuel sup-
ply should be decreased in order to reach close effi-
ciencies in the removal of thiophene and DBT in the
presence of Cu/CeO , Pt/CeO , and CuZnAl cata-
isotherms were obtained on an ASAP-2400 automatic
adsorption instrument (Micromeritics, United States).
The carbon, hydrogen, and sulfur contents (CHS
analysis) in the spent catalysts were determined with
the aid of a Vario MicroCube analyzer (Elementar
Analysensysteme GmbH, Germany) equipped with a
thermal conductivity detector; the sample was burned
in an oxygen atmosphere, and the combustion prod-
ucts were separated by temperature-programmed
desorption.
The differential thermal and thermogravimetric
analysis with the mass-spectrometric monitoring of
gas phase composition (DTA-TG-MS) of the spent
catalysts was carried out on a NETZSCH STA 409 PG
instrument (Tetzsch-Geratebau GmbH, Germany). A
2
2
lysts. For example, at 300°C, the CuO/ZnO/Al O
2
3
catalyst afforded the conversion of thiophene and
DBT dissolved in isooctane at a level of 93%, but this
was achieved at an O /S ratio of 140 or 50 and a mass
2
–1
flow rate of 6 or 18 h , respectively [9]. This fact is
indicative of different reactivity of sulfur-containing
compounds in the oxidative desulfurization process.
The aim of this work was to study the reactivity of sample of 25 mg (particle size of <0.25 mm) was
thiophene, DBT, and 4,6-DMDBT dissolved in tolu- placed in a crucible, and a flow of oxygen
3
3
ene in an oxidative desulfurization reaction on the (15 cm /min) and helium (50 cm /min) was supplied
CuO/ZnO/Al O catalyst modified with boron and to it; the crucible was heated from 25 to 1000°C at a
2
3
molybdenum additives. To determine the contribution rate of 5 K/min. Gaseous products were analyzed on a
of the adsorption of these compounds to desulfuriza- NETZSCH QMS 403C mass spectrometer (Tetzsch-
tion, the catalysts after the adsorption and reaction Geratebau GmbH, Germany). The NIST Chemistry
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

6
0
YASHNIK et al.
Webbook database was used for identification pur- catalyst bed for preventing the gas-phase combustion
poses.
of toluene out of this bed. The liquid reaction products
were collected in a sample receptacle by their cooling
in a reflux condenser at 5°C.
The phase composition of the catalysts was deter-
mined by X-ray diffraction X-ray powder diffration
patterns were recorded on an HZG-4S diffractometer
The total sulfur content of toluene and the liquid
(
Freiberger Prazisionmechanik, Germany) with reaction products was determined with the aid of an
monochromatic CoK radiation in the 2θ range from ASE-2 X-ray energy-dispersive analyzer (Burevestnik,
α
1
0° to 70°. Phases were identified with the use of the Russia), which made it possible to measure sulfur con-
X-ray Powder Diffraction File JCPDS-ICDD.
tents from 0 to 1000 ppm with an accuracy of 10%. The
composition of individual hydrocarbons and sulfur-
containing compounds in the liquid reaction products
was calculated based on mass-spectrometric data
obtained on an Agilent 6890N mass spectrometer with
an Agilent 5973 mass-selective detector (Agilent Tech-
nologies, United States) and an HP-5ms capillary col-
umn (30 m × 0.25 mm × 0.25 μm). The NIST library
of spectra was used for the identification of compo-
nents based on their mass spectra. The gas phase was
analyzed on a Kristall 2000M chromatograph (Khro-
matek, Russia). A thermal conductivity detector was
used in the analysis for H , O , CO , and CH (SKT
The chemical composition of the catalysts was
studied by X-ray photoelectron spectroscopy (XPS)
on a SPECS’s instrument (Surface Nano Analysis
GmbH, Germany) equipped with a PHOIBOS-150
hemispherical analyzer, a nine-channel detector of
electrons, and an XR-50 source of characteristic
X-radiation with an Al–Mg dual anode using non-
monochromated MgK radiation (hν = 1253.6 eV).
α
The sample charging effect was taken into account
using an Al2p peak with the binding energy BE =
4
.05 eV. The relative concentrations of the elements in
2
2
2
4
the analysis zone were calculated from the integrated
XPS intensities with consideration for the photoion-
ization cross sections of the corresponding terms [23].
For a detailed analysis, the spectra were decomposed
into individual components. After Shirley background
carbon), and a flame photometric detector was used
for the determination of H S, COS, and SO (Hayesep
2
2
Q + 0.9% PTMSP).
In order to estimate the reactivity of a sulfur-con-
subtraction, the experimental curve was decomposed taining molecule toward its oxidation on the CuO–
into a number of lines, which corresponded to the ZnO–Al O catalyst, the efficiency of sulfur removal
2
3
photoemission of electrons from atoms in different
chemical environments. Data processing performed
with the aid of the CasaXPS program package [24].
The peak shapes were approximated by a symmetrical
function obtained by the multiplication of Gaussian
and Lorentzian functions.
from sulfur-containing molecules in a liquid phase,
their conversion into SO , the selectivity of oxygen
2
consumption for oxidation, and the toluene conver-
sion into CO were compared. The efficiency of sulfur
2
removal (X , %) was calculated as a change in its con-
S
0
S
centration in liquid fuel before (C , vol %) and after
C , vol %) reaction on a C . The conversion of the
For the preparation of the fresh and spent catalysts
0
S
(
for the analysis of their chemical and phase composi-
S
tion, a catalyst sample (2 g) was thoroughly ground in sulfur-containing molecule into SO (
,
%) was
X
2
S–SO2
order to average its composition. The textural charac-
teristics were determined by studying a fraction of
determined as a ratio of the amount of SO formed in
2
the reaction (CSO2, vol %) to the amount of sulfur
0.5–1.0 mm.
0
S
introduced into the reactor (C , vol %).
Catalyst Testing
The catalyst was tested in a reaction of the oxida-
0
C − C
S
S
X =
×100,
×100.
(1)
(2)
S
0
CS
tion of the sulfur-containing molecules of thiophene,
DBT, and 4,6-DMDBT dissolved in toluene (0.1 wt %
in terms of sulfur) with oxygen. The experiments were
carried out in a flow reactor placed in a furnace with a
fluidized bed of quartz sand for uniform temperature
distribution; the catalyst sample weight was 2 g (part-
C
SO2
0
XS−SO2
=
C
S
The total oxygen conversion is
ticle size of 0.5–1.0 mm). The molar ratio was O /S =
2
–1
1
20, the gas hourly space velocity (GHSV) was 3000 h ;
0
C
− C
O
2
O2
the weight hourly space velocity (WHSV) of fuel was
XO2
=
×100,
(3)
–1
0
6
h ; and the temperature range was 245–430°C. The
C
O2
temperature in the catalyst bed was measured with a
chromel–alumel thermocouple with an accuracy of
0
O
0
.5%. Model fuel was supplied to the reactor from bot- where C
and CO2 are the oxygen contents (vol %) of
2
tom to top. A mixture of oxygen (20.8 vol %) with the mixture of reactants before and after the reaction,
helium and toluene were separately supplied to the respectively.
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
61
(а)
(b)
Degree of desulfurization, %
Toluene conversion into CO , %
2
5
4
3
2
1
0
8
0
0
0
0
0
3
2
6
4
2
1
1
2
3
4
250 275 300 325 350 375 400 425
250 275 300 325 350 375 400 425
Temperature, °С
Temperature, °С
Fig. 1. The temperature dependences of (a) the degree of desulfurization of the liquid products and (b) the conversion of toluene
into CO in the processes of the oxidative desulfurization of (1) thiophene, (2) DBT, and (3) 4,6-DMDBT dissolved in toluene
2
and (4) the deep oxidation of toluene to CO in the absence sulfur-containing compounds on the modified CuZnAlO catalyst.
2
The difference between the weights of toluene sup- where m and m are the weights of adsorbed sulfur
S
Cat
0
and the catalyst, respectively (g); G is the fuel feed rate
plied to the reactor Мt and released from the reactor
(
)
to the reactor (g/min); X is the sulfur conversion at
S,i
0
(
M ) divided by М : was considered as the toluene the ith stage of the experiment (%); and t is the sam-
t
t
conversion
pling time (min).
0
M − M
t
t
X =
.
(4)
RESULTS AND DISCUSSION
t
0
Mt
Catalytic Activity of the CuZnAlO Catalyst
in Oxidation Reactions
The toluene conversion was also monitored based
on the amount of released CO .
of Sulfur-Containing Compounds
2
Figure 1 shows the temperature dependences of the
efficiency of desulfurization of the liquid products and
the conversion of toluene. These data make it possible
to judge the activity of the modified CuZnAlO catalyst
in the oxidative desulfurization of sulfur-containing
molecules dissolved in toluene. Figure 2 illustrates the
changes in the efficiency of sulfur removal and the
The sorption of thiophene and DBT sulfur was
determined in a flow reactor by passing a vapor–gas
mixture through a catalyst bed with a weight of 2 g
until reaching a steady sulfur content of the liquid
reaction products. According to the results of measur-
ing the catalyst activity in the oxidative desulfurization
process, a sorption temperature of 375°C was chosen;
at this temperature, the high conversion of sulfur-con-
taining molecules into both liquid and gaseous prod-
release of gaseous SO in the process.
2
From the experimental data, it follows that thio-
ucts were reached. The other conditions of sorption phene is most difficult to remove (Fig. 1a, curve 1). Its
were the same as in the oxidative desulfurization reac- removal begins at T > 325°C. The process efficiency
–1
tion: a flow of helium, GHSV = 3000 h , and WHSV = slowly increased with temperature, and it was no
–1
6
h . The 1 mL samples of the liquid products were higher than 35–40% at 375–430°C. A further increase
taken for analysis at regular intervals of 10 min; they in the temperature is unreasonable because the deep
were diluted with toluene to 5 mL, and the concentra- oxidation of a substantial portion of toluene occurred
tions of sulfur were determined on the ASE-2 analyzer. in this range, and the toluene loss was as high as 3.7%
The sorption capacity of the catalyst for a sulfur-con- (Fig. 1b, curve 1). The degree of oxygen utilization in
taining compound in terms of sulfur was calculated the test temperature range was close to 100%. The for-
from the formula
mation of SO from thiophene increased with tem-
2
perature; however, the amount of SO released into
2
i=n
the gas phase (Fig. 2a, curve 1) was smaller than the
amount of converted thiophene (Fig. 2a, curve 2).
Furthermore, the oxidation of thiophene with the
release of sulfur dioxide did not reach steady-state val-
X t
∑
S,i i
mS
= GC ×10 × i=1
0
S
−4
a =
,
(5)
mCat
mCat
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

6
2
YASHNIK et al.
ues and exhibited a cyclic behavior, periodically
(а)
increasing (for example, to 25% at 430°C) and then
decreasing sharply (Fig. 2a). Based on these data, it is
possible to assume that, at temperatures above 300°C,
thiophene is adsorbed on the catalyst surface, and the
sulfur compound is oxidized and removed form the sur-
Sulfur conversion, %
4
25°
3
2
4
00°
3
75°
350°
4
3
2
0
0
0
325°
300°
250°
face as SO after the appearance of a monolayer coating.
2
The DBT conversion more rapidly increased with
temperature than the thiophene conversion (Fig. 1a,
curve 2). A maximum efficiency of DBT removal at
T > 380°C was 65–70%, which is close to the values
observed on the unmodified CuZnAlO catalyst [25].
However, in this case, the maximum conversion of
toluene reached 4.3% (Fig. 1b, curve 2). As in to the
oxidation of thiophene (Fig. 2a), the periodic accu-
mulation of sulfur on the catalyst surface with the sub-
sequent release into a gas phase in the form of sulfur
dioxide was observed in the case of the oxidation of
DBT (Fig. 2b). The DBT conversion with the release
1
0
Sorption
1
0
0
200 400 600 800 1000 1200 1400
(
b)
of SO at 380–430°C changed from 10–15 to 50–55%.
2
4
30°
3
2
4
05°
The maximum efficiency of sulfur removal from
1
00
3
80°
4
,6-DMDBT in a temperature range of 375–400°C
350°
was ~80% (Fig. 1a, curve 3). In contrast to reactions
8
6
4
2
0 3
00°
with thiophene and DBT, the formation of SO from
2
4
,6-DMDBT came into play at 325°C to reach 99.9% at
0
times and to decrease to 2–5% at T > 375°C (Fig. 2c).
In this case, the amount of sulfur removed from the
liquid products was also greater than the amount fixed
0
0
Sorption
in the form of SO , especially at low temperatures. For
Sorption
2
example, at 350°C, the degree of sulfur removal was
1
7
0%, and the degree of SO formation was 48%. It is
0
0
2
likely that, at low temperatures, a portion of sulfur
remained on the catalyst surface in an adsorbed state;
200
400
600
800
1000
then, it was oxidized and removed in the form of SO
2
at higher temperatures.
(
c)
Note that the degree of the deep oxidation of tolu-
ene in the absence of sulfur-containing compounds
was as high as 4.7–5.9% in a range of 375–430°C. In
the presence of thiophene in the reaction mixture, this
value was 3.7 or 4.3% in the presence of DBT or
4
30°
4
05°
3
3
80°
1
00
350°
325°
300°
8
0
0
0
0
0
4,6-DMDBT (Fig. 1b). In a range of 300–375°C, only
2
DBT exerted an inhibiting effect on the oxidation of
toluene, whereas this effect was completely absent in
the presence of thiophene and 4,6-DMDBT. It is
likely that the weak inhibition of the oxidation of tolu-
ene by sulfur-containing compounds is related to the
presence of a π-conjugated aromatic system in both
the sulfur-containing molecules and the toluene mol-
ecule; this is responsible for their high reactivity
toward oxidation by oxygen.
6
4
2
Sorption
Sorption
1
0
200 400 600 800 1000 1200 1400 1600 1800
Time, min
The oxygen conversion at temperatures of >300°C
changed insignificantly; however, as the temperature
was further increased to 325°C, it sharply increased to
reach almost 100% in the oxidation of mixtures con-
taining thiophene or 4,6-DMDBT. In the case of
DBT oxidation, this effect was observed at 375°C.
Note that, in this temperature range, the toluene con-
version also sharply increased to subsequently reach
Fig. 2. Oxidative desulfurization of (a) thiophene, (b) DBT,
and (c) 4,6-DMDBT dissolved in toluene on the modified
CuZnAlO catalyst: (1) desorption with oxidation to gaseous
SO , (2) substrate sorption from a liquid phase, and (3) tem-
2
perature changes in the in the course of the process.
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
63
constant values. Thus, a decrease in the 4,6-DMDBT Table 1. Activation energies of the oxidative desulfurization
reactions of thiophene, DBT, and 4,6-DMDBT in the pres-
ence of the modified CuZnAlO catalyst and the accumula-
tion of sulfur in the course of substrate sorption at 375°C in
the absence oxygen
conversion at T > 375°C may be caused by the occur-
rence of a competing reaction of toluene oxidation.
Below 375°C, 4,6-DMDBT was oxidized somewhat
more easily than toluene. A comparison of the oxida-
tion dynamics of thiophene and toluene and the con-
sumption of oxygen showed that, on the contrary,
thiophene was oxidized more difficultly than toluene.
Amount
Activation energy,
Compound
of adsorbed sulfur,
mg /g *
kJ/mol
S
Cat
Using experimental data on the conversion of sul-
fur-containing compounds in a liquid phase (Fig. 1),
we calculated the apparent activation energies of the
Thiophene
DBT
31 ± 6
42 ± 5
50 ± 5
1.3
0.63
–
oxidation of the sulfur-containing compounds 4,6-DMDBT
(
Table 1). Because the oxidation reactions of sulfur-
*
Calculated from Eq. (5).
containing compounds are nonzero-order with
A dash denotes that sorption was not determined.
respect to oxygen and the concentration of O changed
2
considerably even at low substrate conversions, we
based on the assumption of the first orders of reaction
with respect to the sulfur-containing compound and
oxygen and believed that the reactor operated in the
plug-flow mode:
hydrogen peroxide, an increase in the number of aro-
matic rings led to an increase in their reactivity, and the
presence of alkyl substituents in the 4- and 4,6-posi-
tions increased the reactivity of DBT derivatives. It was
hypothesized that the reactivity of a sulfur compound
toward oxidation is caused by two factors that exert
opposite effects on the rate of reaction. The positive
(6) inductive effect of alkyl substituents increases the rate of
oxidation, and the close arrangement of these substitu-
ents to the sulfur atom creates steric hindrances for an
attack of electrophilic particles on a lone electron pair of
1
k =
0
O
0
τ C − CS
(
)
)
⎤
2
0
O2
0
⎡
XS
−
C
CS
0
S
0
(
C
−1 −1
×
⎢ln
+ ln
⎥, L mol s ,
(
X −1
)
C ⎥
O
2
⎦
⎢
⎣
S
where τ is the reaction mixture–catalyst contact time. sulfur and thus slows down the reaction. Both of these
effects indicate that hydrogen peroxide directly oxidizes
As can be seen in Table 1, the activation energy
sulfur in the composition of a sulfur compound.
somewhat increases in the order thiophene
(
(
31 kJ/mol) < DBT (42 kJ/mol) < 4,6-DMDBT
In the case of high-temperature oxidation with
50 kJ/mol). It is believed that the rate of adsorption of oxygen on a heterogeneous catalyst, difference in
sulfur-containing compounds on the catalyst surface reactivity can be caused by not only the electronic
is limited by pore-diffusion resistance, whereas the structure peculiarities of a sulfur compound but also
oxidation of adsorbed compounds occurs in the the structure and reactivity of an intermediate formed
kinetic region. Under the conditions of the liquid- upon substrate adsorption on the catalyst surface. An
phase oxidation of DBT, 4,6-DMDBT, and C -DBT analysis of experimental data [9] showed that, as the
3
by tetrabutyl peroxide in the presence of a 16 wt % experiment time was increased from 1 to 5 h, the sulfur
MoO /Al O catalyst, an activation energy of 32 ± contents of both the catalyst and liquid fuel increased.
3
2
3
A low degree of oxygen consumption (4% in a reaction
with thiophene or 19% in a reaction with DBT [9]) at
a high efficiency of sulfur removal (93%) was also
indicative of the sorption of sulfur compounds. Our
experimental data given in Fig. 2 show that, indeed,
sulfur-containing compounds were sorbed on the
modified CuZnAlO catalyst.
2
kJ/mol was observed previously [19], and this energy
was about 60 kJ/mol in the oxidation of them by
hydrogen peroxide on a Ti/zeolite catalyst [26]. These
values are close to the activation energies of the
adsorption of thiophene, DBT, and 4,6-DMDBT
obtained in this work.
Thus, the reactivity of the sulfur-containing com-
pounds in their high-temperature oxidation by oxygen
We attempted to compare the sorption of thiophene
with the formation of sulfur dioxide on the modified and DBT on the catalyst surface in the absence and in
CuZnAlO catalyst increases in the order thiophene < the presence of oxygen in a reaction atmosphere. Under
DBT < 4,6-DMDBT. Analogous regularities were similar experimental conditions in the absence of oxy-
described for the low-temperature oxidation of sulfur gen, the duration of thiophene and DBT sorption at
compounds dissolved in liquid hydrocarbons or diesel 375°C was 220 and 110–120 min, respectively. The
fractions to sulfones and sulfoxides by hydrogen perox- sorption capacity of the catalyst for thiophene and DBT
ide in the presence of formic acid solutions [14, 16, 17, determined from Eq. (5) with consideration for the
2
7] and homogeneous [16] or heterogeneous [14] cata- amount of sulfur remained in toluene and the total
lysts and also for oxidation with an ozone–oxygen mix- amount of sulfur passed over the catalyst was 1.3 and
ture [28]. In the majority of cases, on the oxidative 0.63 mg of sulfur per gram of catalyst, respectively
desulfurization of benzothiophene homologues with (Table 1). In this case, according to the CHS analysis
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

6
4
YASHNIK et al.
Table 2. Main characteristics of the initial and spent catalysts after performing the oxidation and sorption of sulfur-con-
taining molecules dissolved in toluene (0.1 wt %) on them
Element concentration, wt %
S-containing
compound
2
3
Phase composition
SBET, m /g Vpore, cm /g
C
H
–
S
Initial catalyst
Absent
(Cu,Zn)Al O (а = 7.896 Å)
2
4
CuO
50
0.16
0.10 ± 0.03
0.05 ± 0.02
MoO3
Catalyst after oxidation
Thiophene
DBT
(Cu,Zn)Al O (а = 8.089 Å)
2 4
Cu (Н50.5 = 600)
Cu O (Н = 25)
30
0.20
0.14
0.18
7.22 ± 0.03 0.30 ± 0.01 0.85 ± 0.04
3.90 ± 0.03 0.42 ± 0.01 1.96 ± 0.02
5.61 ± 0.07 0.30 ± 0.01 1.55 ± 0.03
2
49.5
(Cu,Zn)Al O (а = 8.095 Å)
2
4
40
Cu O (Н = 160)
2
49.5
4,6-DMDBT
(Cu,Zn)Al O , (а = 8.070 Å)
2 4
Cu (Н50.5 = 400)
Cu O (Н = 70)
58
2
49.5
Catalyst after sorption at 375°C
Thiophene
DBT
CuAl O (33–448),
2 4
(
а = 8.083 Å), traces
55
45
0.16
0.17
3.02 ± 0.03 0.42 ± 0.01
0.17 ± 0.01
Cu (4–836)
CuAl O (33–448),
2
4
(
а = 8.083 Å), traces
Cu (4–836)
A dash denotes that the value is below the detection limit.
3.60 ± 0.04 0.44 ± 0.01 0.13 ± 0.01
data, the sulfur content of the catalyst was close to its also undergo oxidation by catalyst lattice oxygen and
sorption capacity for thiophene (0.17 and 0.13 wt %, desorption into toluene solution in the form of sul-
respectively, Table 2), whereas these values were dif- fones, sulfoxides, or SO . In the experiments on the
2
ferent by a factor of 2 in the case of DBT (0.12 and absorption of DBT, the presence of the insignificant
0
.06 wt %, respectively, cf. Tables 1 and 2). A compar- amounts of SO in the gas phase was detected; how-
2
ison of the results of CHS analysis also showed that, in ever, this was not observed in the experiences with
the absence of oxygen, the sorption capacity of the cat- thiophene. Third, the temperatures at which maxi-
alyst was lower than the amount of sulfur accumulated mum catalytic activity and sorption were reached were
on the catalyst in the course of oxidative desulfurization 430 and 375°C, respectively, and an increase in the
by a factor of 5–15 (Table 2). Furthermore, from the temperature increased the sorption capacity of the cat-
C/S atomic ratio in the catalyst after the sorption of a alyst because its deeper layers underwent sulfidizing.
sulfur compound (Table 2), it follows that the toluene
solvent was adsorbed on the catalyst surface in addition
to the sulfur compound.
Analysis of Spent Catalysts
Table 2 summarizes the chemical and phase com-
position and the textural characteristics of the catalysts
before and after their testing in the processes of the
oxidation and sorption of sulfur-containing com-
pounds. Earlier, we found [29] that the initial CuO–
Discrepancies in the sulfur contents of the catalyst
can be explained by several reasons. First, in the
course of sorption in the absence of oxygen, sulfur is
accumulated on the surface in the form of sulfur com-
pounds, and more stable compounds, for example,
metal sulfates, which are decomposed at temperatures
above 750°C, are formed as a result of their oxidations
by oxygen. Second, adsorbed thiophene is oxidized to
ZnO–Al O catalyst contained the highly dispersed
2
3
spinel (Cu,Zn)Al O and oxide CuO phases. We
2
4
related the absence of a crystallized ZnO phase to the
2+
sulfur dioxide only by gaseous oxygen, whereas the stabilization of Zn cations in the lattice of aluminum
adsorbed molecules of DBT and 4,6-DMDBT can oxide and the formation of spinel ZnAl
O
and fine
2
4
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
65
Table 3. Atomic ratios between elements in the near-surface layers of the fresh and spent catalysts according to XPS data
Ratio
Catalyst
Cu/Al
Zn/Al
Cu/Zn
Mo/Al
B/Al
O/Al
S/Al
Initial
0.31
0.09
0.22
0.13
0.16
0.08
0.15
0.08
1.87
1.12
1.52
1.49
0.09
0.10
0.15
0.12
0.21
0.19
0.11
0.16
2.9
2.4
3.3
2.6
0
After reaction with thiophene
After reaction with DBT
After reaction with 4,6-DMDBT
0.02
0.29
0.13
ZnO particles with sizes smaller than 5 nm. The intro- Cu2p , Zn2p , S2p, Mo3d , and B1s binding ener-
3/2
3/2
5/2
duction of molybdenum and boron additives into the gies, respectively.
catalyst led to the formation of highly dispersed MoO
3
Table 3 indicates that a considerable amount of sul-
fur was accumulated in the catalyst after a reaction
with DBT; the S/Al atomic ratio was ~0.29. This value
is greater than that found based on the CHS analysis
and ICP AAS data by a factor of almost 3 (S/Al =
particles and the modification of the Al O surface
2
3
with borate anions.
According to XRD data, copper metal and/or
Cu O were formed on the catalyst after performing the
2
oxidation reactions of thiophene, DBT, or 4,6-DMDBT 0.11). The S/Al ratio in the near-surface layers of cat-
on it. After the sorption of thiophene or DBT in the alysts on which 4,6-DMDBT was oxidized was also
absence of oxygen, the reduction of Cu(II) cations to substantially (by a factor of about 2) higher than this
a metallic state was also detected; it was apparently ratio in the bulk. Consequently, sulfur was accumu-
caused by the action of hydrocarbons at elevated tem- lated predominantly in the near-surface layers. At the
peratures. Regardless of the nature of a sulfur com- same time, the sulfur content of the near-surface layer
pound, the crystallized phases of metal sulfides, sul- after the oxidation of thiophene was very low (S/Al =
fites, and sulfates were not detected in the composi- 0.02). The zinc content of this layer remained
tion of spent catalysts by XRD analysis. However, the unchanged in the course of a reaction with DBT,
application of CHS analysis, XPS, and DTA-TG-MS whereas the layer was depleted of zinc cations upon
allowed us to detect the accumulation of sulfur and the oxidation of thiophene and 4,6-DMDBT. Fur-
carbon as surface sulfur- and carbon-containing com- thermore, a tendency toward surface enrichment in
pounds, whose concentrations depended on the struc- molybdenum and depletion of copper and boron was
ture of a substrate. According to the CHS analysis observed. For example, in the case of DBT oxidation,
data, the S/C ratio decreased in the order DBT (0.19) > the Mo/Al ratio increased from 0.1 to 0.15; the Cu/Al
4
,6-DMDBT (0.10) > thiophene (0.04). A compari- ratio decreased by a factor of 1.5, and B/Al was halved
son of S/C ratios in the spent catalysts and sulfur-con- (from 0.2 to ~0.1). The change of copper concentra-
taining molecules showed that, upon the oxidation of tion in the near-surface layers is reasonable because of
thiophene, the catalyst surface was enriched in carbon- a growing tendency toward the formation of a CuAl
O
2
4
containing compounds because these ratios in the cata-
lyst and in the thiophene molecule are strongly different
spinel phase in the course of reaction. The formation
of CuAl O was detected by XRD as an increase in the
2
4
(0.04 and 0.25, respectively).
spinel lattice parameters. It is likely that the diffusion
The specific surface area of the modified of boron into the catalyst volume was related to its
CuZnAlO catalysts insignificantly changed in the ability to accomplish the anionic modification of alu-
course of a reaction. Usually, the formation of amor- minum oxide and to form highly dispersed aluminob-
phous carbon increases the catalyst surface area, orate phases under thermal action. There are pub-
whereas the condensation of deposited hydrocarbon lished data on the formation of Al B O aluminobo-
4
2
9
products, on the contrary, decreases it. The carbon rate phases on the heat treatment of aluminum oxide
content and the degree of condensation of hydrocar- modified with boron [30, 31] but at a higher concen-
bon deposits (C/H) calculated based on the results of tration of boron (>5 wt % [30]). The formation of alu-
CHS analysis depend on the nature of a sulfur-con- minoborate phases retards the phase transitions of
taining compound and increase in the order DBT low-temperature aluminum oxides [32], and their sur-
(
3.9%, 0.75) < 4,6-DMDBT (5.6%, 1.5) < thiophene face modification with borate anions changes the
(
7.2%, 2). The specific surface area of spent catalysts acid–base properties of Al O [30, 33, 34]. It is well
2
3
also increased in this order.
known that, on the addition of boron in an amount to
The chemical composition of the near-surface lay- 1.3 wt % to aluminum oxide, it was initially adsorbed
ers of modified CuZnAlO catalysts was determined by on strong acid sites with a characteristic absorption
XPS analysis. Tables 3 and 4 summarize the observed band frequency of 3750 cm , which corresponds to
atomic ratios between the elements in the near-surface terminal Al–OH groups [33]. At a boron content
layers of fresh and spent catalysts and the Al2p, higher than 1 wt %, these sites are almost absent from
–1
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

6
6
YASHNIK et al.
Table 4. The Al2p, Cu2p3/2, Zn2p3/2, S2p, Mo3d5/2, and B1s binding energies and the Auger parameters α(Cu) in the fresh
and spent catalysts
Binding energy and α(Cu), eV
Catalyst
Al2p
Cu2p3/2
α(Cu)
Zn2p3/2
Mo3d5/2
B1s
S2p
Initial
74.05
74.05
933.4
932.3
1851.1
1849.5
1021.94
1021.85
232.3
232.2
191.8
191.7
–
After reaction with thiophene
168.8
934.7
After reaction with DBT
72.36
932.5
935.0
1849.5
1849.5
1022.11
1021.82
232.4
232.3
191.6
191.8
164.0
168.7
74.05
After reaction with 4,6-DMDBT
74.05
932.5
163.8
169.0
935.2
A dash denotes that the compound was not detected.
the B O /Al O system [33], and medium-power acid
After the oxidative desulfurization reactions of
2
3
2
3
sites are formed instead of them [34, 35]. The assump- thiophene and 4,6-DMDBT on the catalyst, the Zn2p
−
2
binding energies of 1021.9 and 1021.8 eV, respectively,
were detected. These values are close to BE for the
fresh catalyst. After the interaction of the catalyst with
DBT, the binding energy somewhat increased to
tion of catalyst modification with ВО anions is con-
sistent with the shape of the IR spectra of a fresh cata-
lyst, which exhibit an absorption band at 3690–
–
1
3
695 cm , which corresponds to B–OH stretching
1
022.1 eV. The experimental binding energies do not
–1
vibrations [36], and bands at 1240 and 1340 cm ,
which relate to vibrations in the BO tetrahedrons [35].
exclude the presence of zinc cations in the spent cata-
4
lysts as the constituents of spinel, the sulfate ZnSO ,
4
and the sulfide ZnS. The binding energy values of
Figure 3 shows the Al2p, Zn2p, Cu2p , B1s, and
3/2
1
4
022.0–1023.1 eV [44, 45] and 1021.3–1021.7 eV [46,
Mo3d XPS spectra of the fresh catalyst and the sam-
ples that participated in the oxidative desulfurization
of thiophene, DBT, and 4,6-DMDBT. According to
the XPS and XRD analysis data, aluminum in the
7] were reported in the literature for zinc as a constit-
uent of the sulfate and ZnS, respectively.
The Cu2p spectra of catalysts after interactions
3/2
fresh and spent catalysts mainly occurred as a constit- with thiophene, DBT, and 4,6-DMDBT exhibited
two peaks: an intense peak at 932.2–932.5 eV and a
small peak at 934.6–935.0 eV with a corresponding
shake-up satellite in the region of higher binding ener-
gies. The Auger parameter for the spent catalysts was
uent of spinels, in which the Al2p binding energies
(
BE) are 73.3–74.1 eV [37, 38]. Zinc and copper
occurred in oxidized states in the fresh catalyst. The
observed Zn2p3 and Cu2p binding energies are
1
/2
3/2
1
849.5 eV, which is characteristic of CuSO . The low
022.0 and 933.2–933.4 eV, respectively. The former is
4
2+
Cu2p binding energy in the spent catalysts (932.2–
3/2
close to the binding energy of Zn cations in complex
oxides, where BE = 1021.4–1021.9 eV [38, 39]. The
shape of the spectra (the presence of a shake-up satel-
9
32.5 eV) attributed to copper sulfate is determined by
the differential charging effect. Thus, in the catalyst
after a reaction performed on it, copper occurred in
lite peak at 941–944 eV), the high Cu2p3 binding
/2
the forms of the copper sulfate CuSO and the copper
4
energy (933.2–933.4 eV), and the value of the Auger
parameter α (1851.4–1851.7 eV) unambiguously indi-
cate that, in the initial catalyst, copper occurred is in
oxide CuO.
The states of boron and molybdenum remained
state Cu(II), probably, in the form of CuO (933.6– unchanged in the course of reaction. The B1s spectrum
9
34.6 eV [40, 41]), or it entered into the composition of the spent catalysts was identical to the spectrum of
the initial catalyst, and they corresponded to the anions
of the spinel (Cu,Zn)Al O . The B1s spectra of the
2
4
−
ВО ; only a decrease in the B1s line intensity was
fresh catalyst have a symmetrical peak at 191.4–
2
6+
1
91.8 eV, which corresponds to the presence of boron detected. The state of molybdenum was close to Mo .
–
as a constituent of (BO ) metaborates (191.7–192.5 eV
The S2p spectrum after the oxidative desulfuriza-
2
[
41]). Boron in the oxide B O is usually characterized tion of DBT and 4,6-DMDBT exhibited two peaks at
2
3
1
1
64.0 and 168.7 eV and a peak with a binding energy of
by BE = 193.6 eV [42]. The Mo3d spectra are ade-
quately described by a Mo3d –Mo3d doublet at a
68.9 eV after the interaction of the catalyst with thio-
5/2
3/2
phene. The peak at 164.0 eV belongs to elemental and
Mo3d5/2 binding energy of 231.8–232.4 eV, which cor-
2
n
−
5+
6+
responds to Mo (230.8–231.6 eV) and Mo (232.7– polysulfide sulfur S S (2 < n < 8), and the binding
233.2 eV) [43].
energy of sulfur in this state is 163.6–164.2 eV [48].
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
67
Intensity, arb. units
Al2p
Zn2p
Cu2p3/2
74.05
933.4
1021.8
Cu3p
4
4
4
3
2
3
2
3
2
1
1
1
8
4 82 80 78 76 74 72 70 68
1050 1040 1030 1020
045 1035 1025 1615
955 945 940 935 930 925
Binding energy, eV
1
Intensity, arb. units
В1s
Mo3d
S2p
232.3
168.8
191.8
163.9
4
3
4
3
4
3
2
2
1
2
1
1
196 194 192 190 188
240 238 236 234 232 230 228
175
170
165
160
Binding energy, eV
Fig. 3. The Al2p, Zn2p, Cu2p3/2, B1s, Mo3d, and S2p XPS spectra of (1) the fresh catalyst and the catalysts after the oxidative
desulfurization of (2) DBT, (3) thiophene, and (4) 4,6-DMDBT performed on them. The spectra were normalized to the inte-
grated intensity of the corresponding Al2p spectra.
The second peak at 168.7 eV characterizes sulfur as a catalyst surface in the form of elemental sulfur and
2
−
polysulfides. The formation of bulk crystallized sul-
fides in the spent modified CuZnAlO catalysts was not
detected by XRD and XPS analysis. At the same time,
it is well known that bulk CuO and CuAl O are com-
constituent of the sulfate groups SO : the S2p bind-
(
4
)
ing energy from 168.8 to 169.5 eV corresponds to sulfur
6+
in the state S [48, 49]. However, the observed S2p
binding energy for sulfur in the composition of sulfides
is usually 161.1–161.7 eV [49, 50]. Such values of BE
were not detected in the spent catalysts.
2
4
pletely sulfidized with hydrogen sulfide under reduc-
ing conditions (for example, with the use of a mixture
of 15% H S/H [51]) at 350°C. Sulfidizing occurs
2
2
Thus, the cations of Al, B, and Mo did not change
their state in the course of the oxidative desulfuriza-
tion of both thiophene and dibenzothiophenes, while
0
through the step of Cu(II) reduction to Cu [51]; in
this case, the supported Cu/Al O catalysts are
2
3
Cu and Zn were converted into surface sulfates in the reduced and sulfidized more rapidly than the bulk
course the reactions. Sulfur also accumulated on the samples of Cu [51].
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

6
8
YASHNIK et al.
(а)
(b)
TG, %
04
DTA, μV
DTG
0.25
0.20
0.15
TG, %
DTA, μV
DTG
0.8
1
0
0
.0
.5
1.5
1
105
DTA
1.0
0.5
0
0
0
0
0
–0.2
–0.4
–0.6
–0.8
.6
.4
.2
1
03
02
01
00
100 TG
0
0
0
.10
.05
–0.5
–1.0
–1.5
1
95 DTA
90
–
–
0.5
1.0
1
DTG
–0.05
0.10
–0.15
1
–2.0
8
5 DTG
–
–1.5
–2.0
–2.5
–
–
2.5
3.0
9
9 TG
80
H2O
H2O
O2
O2
CO2
CO
CO2
CO
SO3 × 300
SO3
SO2
SO3
200
400
600
800
1000
200
400
600
800
1000
Temperature, °С
Temperature, °С
Fig. 4. DTA-TG-MS curves of the modified CuZnAlO catalyst (a) after the sorption of DBT on it and (b) after the oxidative
desulfurization of DBT.
The results of the analysis of the initial and spent remained in the catalyst in the form of metal sulfates.
catalysts by a DTA-TG-MS technique, which were The accumulation of hydrocarbon fragments on the
described in detail previously [29], showed that sur- surface of the spent catalyst was caused by the adsorp-
face metal sulfides/polysulfides and sulfates were tion of both sulfur-containing molecules and the
formed in the course of thiophene, DBT, and products of the aromatic polycondensation of toluene.
Amorphous carbon and the polycondensation prod-
ucts were removed from the surface at 300–340 and
4
,6-DMDBT oxidation. They were removed in the
form of SO in a flow of an oxygen-containing mixture
2
470–520°C, respectively. CO and H O were released
at temperatures of 100–150 and 650–800°C, respec-
tively (Figs. 4b and 5b). In the course of desorption,
the sample weight decreased and endothermic effects
were observed. An important result of DTA is a con-
clusion on the predominance of sulfides/polysulfides,
which include carbon and hydrogen in addition to sul-
fur and decompose at low temperatures with the
release of SO , CO , and H O, in the catalyst compo-
2 2
into a gas phase and the liberation of heat occurred
Figs. 4b and 5b). Note that, in the case of DBT oxi-
(
dation, a greater portion of accumulated carbon was
released as CO at 300–340°C (Fig. 4b), whereas car-
2
bon was removed at both 300–340 and 470–520°C in
the oxidation of thiophene (Fig. 5b). Furthermore, at
2
70–295°C, the oxidation of Cu O to CuO was
2
2
2
2
observed; this was reflected in an increase in the sam-
ple weight (TG curves) and an exothermic effect
sition after the oxidation of thiophene (Fig. 5b). It is
believed that the catalyst contained thiophene
and/or its partial oxidation products; therefore, its
extraction from a toluene solution occurs through
adsorptive desulfurization and subsequent oxidative
transformations. According to the DTA data, sulfur
was released from the samples after the oxidation of
(DTA curves).
In contrast to the oxidative desulfurization process,
the adsorption of thiophene and DBT on the modified
catalyst in the absence of oxygen was accompanied by
the accumulation of a considerably smaller amount of
sulfur (Table 2). On the adsorption of DBT, 0.13 wt %
DBT and 4,6-DBT predominantly in the form of SO
2
in a temperature range of 650–800°C (Fig. 4b). sulfur was accumulated, whereas 1.96 wt % was accu-
Therefore, DBT and 4,6-DBT were decomposed in mulated on oxidation; at the same time, carbon was
the course of oxidative desulfurization, and sulfur deposited in approximately equal amounts (3.6 and
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
69
(а)
(b)
TG, %
05
04 DTA
03
02
01
00 DTG
DTA, μV
DTG
TG, %
106
DTA, μV DTG
2.5
1
0
0
.0
.5
1
0.25
0.20
1
0.2
2.0
0.15
0.10
0.05
0
1
104
1.5
0
–
0.5
–1.0
1.5
–2.0
DTG
1
.0
1
0.5
0
–0.5
–1.0
–1.5
–
–
–
0.2
–
102
1
DTA
TG
0.4
0.6
1
–0.05
–
–
2.5
3.0
9
9 TG
100
–0.10
H2O
O2
H2O
O2
CO2
CO
CO2
CO
SO2
SO3
SO2
SO3
200
400
600
800
1000
200
400
600
800 1000
Temperature, °С
Temperature, °С
Fig. 5. DTA-TG-MS curves of the modified CuZnAlO catalyst (a) after the sorption of thiophene on it and (b) after the oxidative
desulfurization of thiophene.
3
.9 wt %, respectively). Carbon was mainly contained
Analysis of the Composition of the Oxidation Products
of Sulfur-Containing Compounds Dissolved in Toluene
in condensation products (probably, this was related to
the adsorption and polycondensation of toluene), and
it was removed from the surface in a wider temperature
The main gaseous products of the oxidative desul-
furization of thiophene, DBT, and 4,6-DMDBT dis-
range of 200–550°C (Figs. 4a and 5a, curves for CO ). solved in toluene were CO and SO . Unreacted sul-
As in the cases of thiophene and DBT sorption, sulfur- fur-containing molecules and their cracking products
containing compounds occurred in an adsorbed state
because their removal in an oxygen-containing mix-
ture occurred predominantly at 100–130°C with the
2
2
2
predominated in the composition of the liquid reac-
tion products, whereas we failed to detect H S, sul-
2
fones, and sulfoxides by gas chromatography and gas
chromatography–mass spectrometry (GC–MS)
analysis. As measured by coulometric titration, the
water content of the liquid products obtained at 430°C
did not exceed 270 ppm, which is somewhat higher
release of SO into the gas phase (Figs. 4a and 5a,
curves for SO ). However, it is likely that an amount of
DBT was strongly adsorbed on the surface and oxi-
2
2
dized to metal sulfates, which were removed at T > than the water content of the initial hydrocarbon fuel
7
00°C. This conclusion was made based on TG data (240 ppm).
on an increase and loss of sample weight in the speci-
Table 5 summarizes the results of the GC–MS
fied temperature range (Fig. 4a). Thus, the results of analysis of the hydrocarbon composition of the liquid
the DTA-TG-MS of catalysts after the sorption of sul- products of the oxidative desulfurization of toluene at
4
30°C on the modified CuZnAlO catalyst. In addition
fur compounds on them in the absence of oxygen con-
firmed the assumption of the adsorptive removal of
sulfur from thiophene and the oxidative desulfuriza-
tion of DBT and its derivatives. The oxidative transfor-
mations of adsorbed thiophene occur under the action
of molecular oxygen, whereas a fraction (no smaller
to toluene, benzene and its alkyl- and aryl-substituted
derivatives, such as 1,3-dimethylbenzene, ethylben-
zene, diphenylmethane, and the isomers of benzyltol-
uene (1-methyl-2-(phenylmethyl)-benzene, 1-me-
thyl-3-(phenylmethyl)-benzene, and 1-methyl-4-
(
phenylmethyl)-benzene) were present. These com-
than 10%) of DBT and its derivatives underwent oxi- pounds are the alkylation and aromatic condensation
dation by the oxygen of the catalyst.
products of toluene:
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

7
0
YASHNIK et al.
Table 5. Composition of the initial hydrocarbon fuels and the products of their oxidative desulfurization at 430°C (accord-
ing to the GC–MS analysis data)
Composition of hydrocarbon fuel, rel. %
Retention
time, min
thiophene–
toluene
4,6-DMDBT–
toluene
Component
Formula
DBT–toluene
initial product initial product initial product
Sulfur dioxide
Benzene
SO2
2.375
2.47
–
0.20
2.61
96.41
0.12
–
0.15
3.91
1.23
82.87
0.11
0.10
8.90
0.39
–
–
–
0.25
0.33
–
–
0.25
–
0.28
2.27
–
C H
6
6
Thiophene
Toluene
C H S
2.54
–
4
4
C H
3.66
90.72
0.13
0.54
–
73.42
0.27
1.02
11.07
0.58
0.69
0.37
0.44
–
86.02
0.04
0.04
–
84.55
0.04
0.04
6.15
0.24
–
7
8
Ethylbenzene
C H
5.28
8
10
10
1,3-Dimethylbenzene
C H
5.43
8
Benzaldehyde
Benzofuran
Benzoic acid
BT
C H O
7.26
–
7
6
C H O
7.99
–
–
–
8
6
C H O
2
11.47
11.87
15.13
16.00
17.35
17.48
17.54
17.66
19.01
19.55
20.12
20.79
22.12
23.39
–
–
–
7
6
C H S
–
–
–
–
0.04
0.04
0.03
0.08
0.14
0.39
0.25
0.04
–
8
6
Biphenyl
C H
0.04
–
0.09
0.09
–
0.13
–
0.02
–
1
2
10
Diphenylmethane
Dibenzofuran
C H
13
12
C H O
–
–
–
–
1
2
8
1-Methyl-3-(phenylmethyl)-benzene
C H
0.47
0.08
–
0.17
0.37
0.25
0.08
0.73
0.12
–
0.54
0.10
–
1.29
4.00
2.12
0.56
2.39
–
1
4
14
14
14
1-Methyl -2-(phenylmethyl)-benzene C H
–
14
1-Methyl -4-(phenylmethyl)- benzene C H
–
14
Benzophenone
C H O
–
–
–
1
3
10
2-Methyl-benzophenone
C H O
–
–
–
1
4
12
12
trans-Stilbene
C H
–
–
0.04
0.51
0.34
2.13
196
1
4
DBT
C H S
–
7.84
–
1.78
–
–
1
2
8
4
4
-МDBT
C H S
–
–
–
1
3
10
,6-DMDBT
C H S
–
970
–
680
–
985
–
397
13.66
1052
1
4
12
Total sulfur, ppm*
*
According to the X-ray energy-dispersive analysis data.
Dashes denote that the compounds were not detected or their concentrations were lower than the limits of detection.
Sulfur-containing compounds and their concentrations in the hydrocarbon fuel (rel. %) are boldfaced.
PhCOPh
°
°
C2, Cat., ~400°C
PhCHO
−H2O
O, H'
PhCOPh
PhCH Ph
2
−H2
+
H2O, C2, Cat, 300°C
Cat. <300°C
PhCH3
PhOH
H C
3
PHCH2
−H2
3/2O2, Cat., <300°C
PhCHOH
PhH
−H2O
−CO2
Scheme 1.
In the products of the oxidative desulfurization of a in trace amounts; it can be formed as a result of the
mixture of toluene–DBT, biphenyl was also detected oxidative conversion of dibenzothiophene sulfone:
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

EFFECT OF THE NATURE OF SULFUR COMPOUNDS ON THEIR REACTIVITY
SO + 12CO + 4H O
71
2
2
2
+O2
+
2H
/
H
O
S
−
SO
O
O
+
H2O
H2S
+
2H2
+
−
C4H6
S
S
O
+H2O, Al2O3, 400−500°C
+H2S
O
Scheme 2.
In the liquid products of DBT oxidation, some reaction (Scheme 2), which occurs on Al O in the
2
3
amounts of benzothiophene were detected in addition presence of water vapor at high temperatures [54].
to unreacted sulfur compounds, and BT, DBT, and
4
-MDBT were detected in the products of 4,6-DMDBT
CONCLUSIONS
In this work, we found that the nature of a sulfur-
oxidation. BT is formed as a result of the cracking of
DBT (Scheme 2), and DBT and 4-MDBT are formed
through the oxidation of the methyl groups of containing molecule has an essential effect on its reac-
4
,6-DMDBT into carboxyl groups (–COOH) with tivity and, correspondingly, on the efficiency of sulfur
their subsequent decarboxylation:
,6-DMDBT + 3О → DBT(СООН) + 2Н О,
removal from it by selective oxidation with oxygen in
the presence of the modified CuZnAlO catalyst. At
temperatures higher than 375°C, the reactivity of the
test substrates in the liquid-phase process of oxidative
desulfurization increases in the order thiophene <
DBT < 4,6-DMDBT. Sulfur is most readily removed
from the molecule of 4,6-DMDBT; in this case, the
efficiency of removal is about 80%.
4
2
2
2
DBT(СООН) → DBT + 2СО .
2
2
DBT can further undergo oxidative desulfuriza-
tion. Note that, in the presence of the modified
CuZnAlO catalyst, the formation of sulfones and sulf-
oxides was not observed, whereas they were detected,
We assumed that thiophene loses sulfur predomi-
although in trace amounts, upon the reaction per- nantly in the process of adsorption; the subsequent
formed on the unmodified catalyst sample [52].
oxidation of the adsorbed molecules by molecular
oxygen is accompanied by the formation of copper and
Benzaldehyde, which results from the oxidation of
the methyl group of toluene (Scheme 1), is the main
oxygen-containing hydrocarbon in the composition of
oxidation products. Other oxygen-containing com-
pounds can be present in trace amounts (no more than
zinc sulfates and the release of SO and hydrocarbons.
2
Adsorbed DBT and its derivatives can be oxidized by
both catalyst lattice oxygen and molecular oxygen with
the release of SO and the formation of hydrocarbons.
2
5–7 ppm). In this case, benzoic acid is an oxidation
product of toluene, and benzophenone and 2-methyl-
benzophenone are the condensation products of benz-
aldehyde and phenol (2-methylphenol), which are
also formed as a result of the occurrence of the consec-
utive oxidation, decarboxylation, and hydration reac-
tions of toluene in the presence of oxygen. The trans-
formations of toluene, which are shown in Scheme 1,
occur with sufficiently high selectivity in the presence
ACKNOWLEDGMENTS
This work was supported by ARAMCO (contract
no. 6600022642). We are grateful to O.S. Efimova for
performing CHS analysis and to P. Lapina for per-
forming GC–MS analysis at the Analytical Study
Center FRC CCC SB RAS (Kemerovo).
Al O and MoO in a temperature range of 300–450°C
2
3
2
REFERENCES
. Ma, X., Sakanishi, K., and Mochida, I., Ind. Eng.
Chem. Res., 1994, vol. 33, p. 218.
. Ma, X., Sakanishi, K., and Isoda, T., Ind. Eng. Chem.
Res., 1995, vol. 34, p. 748.
[
53]; moreover, at T ≥ 400°C, the reactions were
1
shifted toward the formation of toluene, benzalde-
hyde, and polyaromatic compounds [53].
2
Furan and benzofuran, which were also detected in
insignificant amounts (smaller than 10 ppm) by GC–
MS analysis, are the products of the conversion of cor-
responding sulfur-containing aromatic compounds
3. Kabe, T. and Ishiharam, A., Htajima, Ind. Eng. Chem.
Res., 1992, vol. 31, p. 1577.
4. Stanislaus, A., Marafi, A., and Mohan, S.R., Catal.
Today, 2010, vol. 153, p. 1.
(
thiophene, BT, and DTB) according to the Yur’ev
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

7
2
YASHNIK et al.
5
. Babich, I.V. and Moulijn, J.A., Fuel, 2003, vol. 82, no. 6, 32. Saih, Y. and Segawa, K., Appl. Catal., A, 2009, vol. 353,
p. 607.
p. 258.
6
7
. Song, C., Catal. Today, 2003, vol. 86, p. 211.
. Ismagilov, Z.R., Yashnik, S.A., Kerzhentsev, M.A.,
33. Flego, C. and Parker, W.O., Appl. Catal., A, 1999,
vol. 185, p. 137.
Parmon, V.N., Bourane, A., Al-Shahrani, F.M., 34. Ferdous, D., Dalai, A.K., and Adjaye, J., Appl. Catal., A,
Hajji, A.A., and Koseoglu, O.R., Catal. Rev. Sci. Eng.,
011, vol. 53, p. 199.
2004, vol. 260, p. 137.
2
35. Delmastro, A., Gozzelino, G., Mazza, D., Vallino, M.,
Busca, G., and Lorenzelli, V., J. Chem. Soc., Faraday
Trans., 1992, vol. 88, p. 2065.
8. Ford, J.F., Rayne, T.A., and Adlington, D.G., US Pat-
ent 3341448, 1967.
3
6. Sibeijin, M., van Veen, J.A.R., Bliek, A., and Moulijn, J.A.,
9. Gao, L., Tang, Y., Xue, Q., Liu, Ye., and Lu, Y., Energy
Fuels, 2009, vol. 23, p. 624.
J. Catal., 1994, vol. 145, p. 416.
3
7. Gong, L., Ye, Z., Lu, J., Zhu, L., Huang, J., Gu, X.,
1
0. Lu, Y., Wang, Y., Gao, L., Chen, J., Mao, J., Xue, Q.,
Liu, Y., Wu, H., Gao, G., and He, M., ChemSusChem,
008, vol. 1, p. 302.
1. Rang, H., Kann, J., and Oja, V., Oil Shale, 2006,
vol. 23, no. 2, p. 164.
2. Anisimov, A.V. and Tarakanova, A.V., Ross. Khim. Zh.,
008, vol. 52, no. 4, p. 32.
3. Sharipov, A.Kh. and Nigmatullin, V.R., Pet. Chem.,
005, vol. 45, no. 6, p. 371.
4. Ito, E. and Rob van Veen, J.A., Catal. Today, 2006,
vol. 116, p. 446.
and Zhao, B., Vacuum, 2010, vol. 84, p. 947.
2
38. Li, L., Fang, L., Zhou, X.J., Liu, Z.Y., Zhao, L., and
Jiang, S., J. Electron Spectrosc. Relat. Phenom., 2009,
vol. 173, p. 7.
1
3
9. Venugopal, A., Palgunadi, J., Deog, J.K., Joo, O.-S.,
and Shin, C.-H., J. Mol. Catal. A: Chem., 2009,
vol. 302, p. 20.
1
2
1
4
0. Otamiri, J.C., Andersson, S.L.T., and Andersson, A.,
2
Appl. Catal., 1990, vol. 65, p. 159.
1
4
1. Bukhtiyarov, V.I., Kaichev, V.V., and Prosvirin, I.P.,
Top. Catal., 2005, vol. 32, p. 3.
15. Luk’yanitsa, V.G. and Gal’pern, G.D., Izv. Akad. Nauk
4
2. Arzac, G.M., Rojas, T.C., and Fernández, A., Appl.
SSSR, Ser. Khim., 1956, vol. 1, p. 130.
Catal., B, 2012, vol. 128, p. 39.
16. De Filippis, P. and Scarsella, M., Energy Fuels, 2003,
4
3. Ong, C.W., Huang, H., Zheng, B., Kwok, R.W.M.,
Hui, Y.Y., and Lau, W.M., J. Appl. Phys., 2004, vol. 95,
p. 3527.
vol. 17, no. 6, p. 1452.
17. Otsuki, S., Nonaka, T., Takashima, N., Quian, W.,
Ishihara, A., Imai, T., and Kabe, T., Energy Fuels,
4
4. Olsson, C.-O.A., Mathieu, H.-J., and Landolt, D.,
2000, vol. 14, p. 1232.
Surf. Interface Anal., 2002, vol. 34, p. 130.
18. Yan, X.-M., Lei, J.-H., Liu, D., Wu, Y.-Ch., and
4
5. Zhiyong, Y., Bensimon, M., Sarria, V., Stolitchnov, I.,
Jardim, W., Laub, D., Mielczarski, E., Mielczarski, J.,
Kiwi-Minsker, L., and Kiwi, J., Appl. Catal., B, 2007,
vol. 76, p. 185.
Liu, W., Mater. Res. Bull., 2007, vol. 42, no. 11, p. 1905.
1
9. Ishihara, A., Wang, D., Dumeignil, F., Amano, H.,
Qian, E.W., and Kabe, T., Appl. Catal., A, 2005,
vol. 279, p. 279.
4
6. Lu, S.W. and Schmidt, H.K., Mater. Res. Bull., 2008,
20. Hulea, V., Fajula, F., and Bousquet, J., J. Catal., 2001,
vol. 43, p. 583.
vol. 198, p. 179.
47. Bär, M., Ennaoui, A., Klaer, J., Kropp, T., Sáez-
Araoz, R., Allsop, N., Schock, I.L.H.-W., and Lux-
Steiner, M.C., J. Appl. Phys., 2003, vol. 99, p. 123503.
21. Caero, L.C., Hernández, E., Pedraza, F., and Murri-
eta, F., Catal. Today, 2005, vols. 107–108, p. 564.
4
8. Susac, D., Zhu, L., Teo, M., Sode, A., Wong, K.C.,
Wong, P.C., Parsons, R.R., Bizzotto, D., Mitchell, K.A.R.,
and Campbell, S.A., J. Phys. Chem. C, 2007, vol. 111,
p. 18715.
22. US Patent 5753102.
23. Scofield, J.H., J. Electron Spectrosc. Relat. Phenom.,
1976, vol. 8, p. 129.
24. Fairley, N. http://www.casaxps.com.
4
9. Appay, M.-D., Manoli, J.-M., Potvin, C., Muhler, M.,
2
5. Yashnik, S.A., Kerzhentsev, M.A., Sal’nikov, A.V.,
Ismagilov, Z.R., Bourane, A., and Koseoglu, O.R.,
Kinet. Catal., 2015, vol. 56, no. 4, p. 466.
Wild, U., Pozdnyakova, O., and Paal, Z., J. Catal.,
2004, vol. 222, p. 419.
5
0. Ma, H., Han, J., Fu, Y., Song, Y., Yu, C., and Dong, X.,
2
2
2
2
6. Jiang, Z., Lu, H., Zhang, Y., and Li, C., Chin. J. Catal.,
Appl. Catal., B, 2011, vol. 102, p. 417.
2011, vol. 32, p. 707.
5
1. Strohmeier, B.R., Leyden, D.E., and Field, R.S.,
7. Krivtsov, E.B. and Golovko, A.K., Pet. Chem., 2014,
J. Catal., 1985, vol. 94, p. 514.
vol. 54, no. 1, p. 51.
5
2. Ismagilov, Z.R., Kerzhentsev, M.A., Yashnik, S.A.,
Khairulin, S.R., Kuznetsov, V.V., Salnikov, A.V., Par-
mon, V.N., Bourane, A., and Koseoglu, O.R., Eurasian
Chem.Technol. J., 2015, vol. 17, no. 2, p. 119.
8. Krivtsov, E.B. and Golovko, A.K., Izv. Tomsk. Politekh.
Univ., 2012, vol. 321, no. 3, p. 157.
9. Yashnik, S.A., Sal’nikov, A.V., Kerzhentsev, M.A.,
Khitsova, L.M., Malysheva, V.Yu., Teryaeva, T.N., and 53. Ganesan, K. and Pillai, C.N., J. Catal., 1989, vol. 119,
Ismagilov, Z.R., Vestn. KuzGTU, 2014, vol. 5, p. 96.
0. Ni, J., Chen, L., Lin, J., and Kawi, S., Nano Energy, 54. Elderfield, R.C., Heterocyclic Compounds, New York:
012, vol. 1, p. 674.
Wiley, 1951.
1. Colorio, G., Vedrine, J.C., Auroux, A., and Bon-
netot, B., Appl. Catal., A, 1996, vol. 137, p. 55.
p. 13.
3
2
3
Translated by V. Makhlyarchuk
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017