Al-pillared montmorillonite as a support for catalysts based on ruthenium sulfide in HDS reactions

Article abstract of DOI:10.1016/j.cattod.2011.12.020

Sulfided ruthenium supported catalysts based on an Al pillared montmorillonite, with different Al contents, have been prepared and characterized by XRD, S_BET, TEM, XPS, and tested in the hydrodesulfurization (HDS) of dibenzothiophene (DBT). The role of the Al content in the catalytic activity of RuS₂ based catalysts was studied. The activity and selectivity of the catalysts studied in the HDS of DBT suggest that Al pillared montmorillonite could be of interest as a low cost support for this reaction.

Full text of DOI:10.1016/j.cattod.2011.12.020

Catalysis Today 187 (2012) 88–96
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Al-pillared montmorillonite as a support for catalysts based on ruthenium
sulfide in HDS reactions
A. Romero-Péreza, A. Infantes-Molinab, A. Jiménez-Lópeza, E. Roca Jalilc,
K. Sapagc, E. Rodríguez-Castellóna,∗
a Dpto. de Química Inorgánica, Cristalografía y Mineralogía (Unidad Asociada al ICP-CSIC), Facultad de Ciencias, Universidad de Málaga, Campus de Teatinos, 29071 Málaga, Spain
b Instituto de Catálisis y Petroleoquímica, CSIC, c/Marie Curie, 2, Cantoblanco, 28049 Madrid, Spain
c Laboratorio de Sólidos Porosos, Instituto de Física Aplicada (INFAP), CONICET – Universidad Nacional de San Luís, Argentina
a r t i c l e i n f o
a b s t r a c t
Article history:
Sulfided ruthenium supported catalysts based on an Al pillared montmorillonite, with different Al
contents, have been prepared and characterized by XRD, SBET, TEM, XPS, and tested in the hydrodesulfur-
ization (HDS) of dibenzothiophene (DBT). The role of the Al content in the catalytic activity of RuS2 based
catalysts was studied. The activity and selectivity of the catalysts studied in the HDS of DBT suggest that
Al pillared montmorillonite could be of interest as a low cost support for this reaction.
© 2011 Elsevier B.V. All rights reserved.
Received 14 June 2011
Received in revised form
22 November 2011
Accepted 12 December 2011
Available online 14 January 2012
Keywords:
Al pillared montmorillonite
Hydrodesulfurization, Ruthenium sulfide
Dibenzothiophene
1. Introduction
incorporation of small quantities (0.3–0.8 wt.%) of noble metals (Rh,
Ru, Pd and Pt) to a Mo/Al2O3 system produced an enhancement [6]
The presence of sulfur in fossil fuels is one of the most impor-
tant causes of environmental pollution. The burning of diesel in
combustion engines releases sulfur into the environment as SOx,
and when it combines with atmospheric water, it becomes what is
known as ‘acid rain’ [1]. As a consequence, irreversible ecological
harm is done to the biodiversity of the planet, contributing to the
degradation of our environment. The main process in the petro-
chemical industry for the removal of sulfur from crude oil is the
hydrotreating process, which is carried out with Co(Ni)MoS sul-
fided catalysts [2,3] as active phases and ␥-alumina as a support.
To date, these catalysts have been used due to the excellent activity
shown. Notwithstanding, the increasing demand of diesel oil, the
lower quality of the remaining petroleum reserves and the tighter
environmental restrictions with regards to sulfur content in diesel
fuel has led to the search for highly efficient catalytic phases which
can almost completely remove any sulfur containing molecules. In
the last few decades, a plethora of articles has been devoted to the
hydrotreating activity of Co(Ni)MoS based catalysts using different
supports, promoters or synthetic approaches in order to improve
the activity and stability of the hydrodesulfurization (HDS) cata-
lysts, although the active phase has hardly been modified [4,5]. The
of the catalytic activity due to synergic effects. Unsupported tran-
sition metal sulfides were plotted into a curve called the “volcano
plot” where the HDS activity per mole of metal versus the M
S
bond strength was plotted, the most active being the RuS2 phase
[7].
Several unsupported transition metal sulfides (TMS) have been
synthesized and tested in the HDS reaction of DBT [8]. In general,
TMS catalysts are stable under the severe conditions employed in
the hydrotreating process, this fact being of utmost importance [9].
Among all the TMS studied, RuS2 was the most active catalyst. Raje
et al. [10] studied the removal of individual sulfur compounds of
a coal-derived naphtha with several bulk TMS. They found that
ruthenium and rhodium were the most active with total sulfur
removal levels being greater than 90%. Moreover, in a simultaneous
removal of sulfur, nitrogen and oxygen compounds from a coal-
derived naphtha, RuS2 displayed the highest activity for all three
of the hydrotreating processes [11]. When supported, its catalytic
activity and stability depend strongly on the preparation method.
Owing to this there are various points to consider during its prepa-
ration since according to the literature [12,13] RuS2 must have a
pyrite-type structure, called laurite, which should be highly dis-
persed in the support. To this end, experimental variables such as
the precursor salt employed, the sulfiding gas mixture, the sulfiding
temperature and the catalyst support employed influence the cat-
alytic activity of the prepared materials [12,13]. In this sense, it is
∗
Corresponding author. Tel.: +34 952131873; fax: +34 952132000.
E-mail address: castellon@uma.es (E. Rodríguez-Castellón).
0920-5861/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2011.12.020

A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
89
AlCl3·6H2O 0.2 M solution with NaOH 0.5 M (OH−/Al3+ molar ratio
of 2). NaOH was added dropwise to the aluminum chloride solu-
tion at 60 ◦C under vigorous stirring. The mixture was then stirred
overnight at room temperature to obtain the pillaring agent. The
solution used for pillaring contains Al13 Keggin polycations (the
precursor of the pillars) that can be ion exchanged with the charge
compensating cations of the montmorillonite clay. A montmoril-
lonite suspension of 3% (w/v) was prepared and stirred for an hour.
The pillaring agent was then added dropwise while stirring at room
temperature. The mixture obtained was stirred for approximately
1 h and left to settle for 12 h to allow the solid to be deposited at
the bottom. The solid was recovered and flushed in order to remove
residual salts. Washing was carried out with distilled-water dial-
ysis membranes until the conductivity was that of the distilled
water. Finally the solid was dried at 60 ◦C overnight and calcined
at 500 ◦C for 1 h at a heating rate of 10 ◦C min−1, thus obtaining
the aluminum-pillared clay. The mequiv. Al3+ g−1 dried clay ratios
employed to synthesized the aluminum-pillared clays were of 5,
10, 15 and 20 and the obtained solids denoted as PILC Al 5, PILC Al
10, PILC Al 15, and PILC Al 20, respectively. The Si/Al ratio of the
parent montmorillonite is 2.67, while that for pillared-clays ranges
between 2.29 and 1.97.
The incorporation of ruthenium into the aluminum-pillared clay
was carried out by the incipient wetness impregnation method.
First an aqueous solution of ruthenium(III) chloride (RuCl3.nH2O),
with the corresponding amount of ruthenium, was added to the
pelletized support (0.85–1 mm). All the catalysts prepared here had
a constant ruthenium loading of 7 wt.% (0.000745 mol gsupport−1).
Once the supports were impregnated, they were air-dried, giv-
ing the catalyst precursors denoted as prec Al x, where x is the
mequiv. Al3+ g−1 clay (x = 5, 10, 15 and 20). Finally, they were sul-
fided in situ at atmospheric pressure with a H2S/N2 (10/90, v/v)
flow of 60 mL min−1 and heated from room temperature (rt.) to the
sulfidation temperature (Ts) (2 h) at a heating rate of 10 ◦C min−1
to obtain the sulfided catalysts. The catalysts will be referred as Al
5, Al 10, Al 15 and Al 20, accordingly.
suggested that calcination be avoided after impregnation, and that
RuCl3 be used as a precursor salt and a sulfiding mixture be used in
the absence of hydrogen in order to avoid the formation of metallic
ruthenium. The latter is very difficult to sulfide [12,14] and requires
a higher sulfidation temperature in order to form highly stable RuS2
particles with a pyrite-type structure presenting a better exposure
of the crystallographic planes (111) and (210), on which the reac-
tion takes place [15,16]. In reference to this, De los Reyes et al.
[13] pointed out that the faces formed at a low sulfiding tempera-
ture are more active for biphenyl (BP) hydrogenation (HYD) while
those formed at higher temperatures are more active for thiophene
HDS [13]. Considering the influence of the support employed, the
most widely used is alumina [12,13,17], although SBA-15 [18] and
MCM-41-type mesoporous materials [19] and MgF2 [14] have been
used with improved performances, mainly due to the mesoporous
texture of SBA-15 and MCM-41.
In recent years, pillared clays (PILC) [20] have been studied as
adsorbents, supports and acid catalysts not only for their perfor-
mance but also due to their low cost [21,22]. One of the most
important applications for these materials has been as a cata-
lyst support for the selective catalytic reduction (SCR) of NO by
NH3 [23–25]. Pillared clays have also been used as a support for
NiMo catalysts for HDS and HDN reactions [26], where both the
impregnation method and the order of incorporation of the met-
als (Ni and Mo) influenced the dispersion of the active phases. To
date, information about ruthenium sulfide catalysts supported on
alumina-pillared clays has been scarce in the literature. However,
an aluminum-pillared bentonite was used as a ruthenium catalyst
support for the 1-butene hydrogenation reaction [27] and Pérez-
Zurita et al. [28] used a commercial montmorillonite clay as a
support for RuS2 catalysts. The formation of the aluminum pillars
and the introduction of the metal into the interlayer space were
done simultaneously. The results obtained indicated that after the
pillaring process, an increase in the surface area of the pillared clays,
as well as a good dispersion of ruthenium, were observed. They
tested the RuS2 catalysts in a thiophene HDS reaction and their
results showed conversions of between 3 and 21% at 280 ◦C.
In this work, supported RuS2 catalysts on aluminum-pillared
clays, with different aluminum contents were synthesized in order
to study the role of the clay in the formation and activity of the RuS2
phase in the DBT HDS reaction and the influence of the aluminum
content on the catalytic activity.
2.3. Characterization techniques
X-ray diffraction patterns (XRD) of the precursor, sulfided and
spent catalysts were obtained with an X’Pert PRO MPD Philips
diffractometer (PANanalytical), using monochromatic CuK␣ radia-
˚
tion (ꢀ = 1.5406 A). The K␣1 radiation was selected with a Ge (1 1 1)
primary monochromator. The X-ray tube was set at 45 kV and
40 mA.
2. Experimental
Transmission electron micrographs of the precursor and
sulfided catalysts were obtained using a Philips CM 200 Supertwin-
DX4 microscope. Samples were dispersed in ethanol and a drop of
the suspension was put on a Cu grid (300 mesh).
The textural properties (SBET, Vp, dp) of the sulfided catalysts
were obtained from the N2 adsorption-desorption isotherms at
−196 ◦C measured with a Micromeritics ASAP 2020 apparatus.
Prior to the measurements, samples were outgassed overnight
at 200 ◦C and 10−2 Pa. Surface areas were determined using the
2.1. Materials
The support used in this study was an aluminum-pillared clay,
prepared by pillaring a natural montmorillonite (from the Alto Valle
region, Argentina), used as received, with Al3+ ions. The formula of
the parent montmorillonite is Si4Al1.36Mg0.27Fe0.25O10(OH)2Na0.53
,
where the cation exchange capacity is 0.89 mequiv./g of clay. Ruthe-
nium(III) chloride, RuCl3·nH2O (∼41 wt.% Ru, from Fluka) was used
as the ruthenium precursor salt. The chemical products used in the
reactivity study were dibenzothiophene (Aldrich 98%) in cis-, trans-
decahydronaphthalene (Sigma–Aldrich 98%). The gases employed
were H2S/N2 10/90 (v/v) (Air Liquide 99.99%), He (Air Liquide
99.99%), H2 (Air Liquide 99.999%) and N2 (Air Liquide 99.9999%).
Brunauer–Emmett–Teller equation and a nitrogen molecule cross
2
˚
section of 16.2 A . The pore size distribution was calculated by
applying the NLDFT method. The total pore volume was calculated
from the adsorption isotherm at P/P0 = 0.95, according to the Gur-
vich rule. The microporous volume was calculated using the ␣-plot
method.
2.2. Preparation of catalysts
X-ray photoelectron spectra of the sulfided and spent cat-
alysts were collected using
a Physical Electronics PHI 5700
Four aluminum-pillared clays with different mequiv. Al3+ g−1
dried clay ratios were used as supports for the RuS2 catalysts.
The synthesis of the aluminum-pillared clay was carried out from
a pillaring agent, which was prepared from the hydrolysis of an
spectrometer with non-monochromatic Al K␣ radiation (300 W,
15 kV, and 1486.6 eV) with a multi-channel detector. Spectra of
pelletized samples were recorded in the constant pass energy
mode at 29.35 eV, using a 720 ␮m diameter analysis area. Charge

90
A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
referencing was measured against adventitious carbon (C 1s at
284.8 eV). A PHI ACCESS ESCA-V6.0 F software package was used
for acquisition and data analysis. A Shirley-type background was
subtracted from the signals. Recorded spectra were always fitted
using Gaussian–Lorentzian curves in order to determine the bind-
ing energies of the different element core levels more accurately.
2.4. Catalytic activity measurements
For the catalytic test, the HDS of DBT was chosen, which was
performed in a high-pressure fixed-bed continuous-flow stainless
steel catalytic reactor (9.1 mm in diameter, and 230 mm in length),
operated in the down-flow mode. The reaction temperature was
measured with an interior placed thermocouple in direct contact
with the catalyst bed. The organic feed was prepared by dissolv-
ing the proper amount of DBT (10,000 ppm) in cis-, trans-decalin.
Each solution was supplied by means of a Gilson 307SC piston
pump (model 10SC). For the activity tests, 0.5 g of catalyst was used
(particle size 0.85–1.00 mm) and was diluted with silicon carbide
(0.85 mm) to 3 cm3. Prior to the activity test, the catalyst precur-
sors were sulfided at atmospheric pressure with a H2S/N2 (10/90,
v/v) flow of 60 mL min−1 by heating from rt. to 500 ◦C (2 h) with
a heating rate of 10 ◦C min−1. The sulfiding mixture in absence
of hydrogen avoids the formation of metallic ruthenium which is
more difficult to sulfide. Catalytic activities were measured at dif-
ferent temperatures (260–360 ◦C), under 3.0 MPa of H2, with a flow
rate of 100 mL min−1 and with hourly space velocities (WHSV) of
32 h−1. The evolution of the reaction was monitored by collecting
liquid samples after 60 min at the desired reaction temperature.
These liquid samples were kept in sealed vials and subsequently
analyzed by gas chromatography (Shimadzu GC-14B, equipped
with a flame ionization detector and a capillary column, TBR-14,
coupled to an automatic Shimadzu AOC-20i injector).
Fig. 1. HDS conversion of RuS2 based catalysts supported on aluminum-pillared
clays as a function of the reaction temperature.
There is a compensatory effect, the isokinetic temperature was cal-
culated as described by Exner [29] by plotting log k versus T−1, and
the found value was 292 ◦C. At temperatures higher than 292 ◦C
clear different catalytic behaviors were observed. The Al 15 catalyst
reaches the highest DBT conversion value at 360 ◦C (90%), followed
by Al 10 (77%) and Al 5 (74%). These data suggest that the higher
the aluminum content, the better the results, with the exception of
Al 20, that exhibits a lower conversion value (57%).
The increasing activity with temperature can be attributed to
the creation of either coordinatively unsaturated sites (CUS) or
vacancies around the metal; such sites are formed by metal sulfide
reduction in a large excess of hydrogen. A CUS site has the function
of electron-withdrawing, interacting with electrodonating organic
molecules such as the DBT molecule [30–32].
The stability of the catalysts was studied by following the reac-
tion for 12 h on stream at 360 ◦C. The evolution of the conversion
with the reaction time is plotted in Fig. 2. From this figure sev-
eral conclusions can be drawn: the Al 15 and Al 10 catalysts that
gave the highest initial activities, suffer a slight deactivation with
time on stream, the conversion decreasing from 85% to 60% for
the Al 15 catalyst and from 73% to 56% for the Al 10 one; Al 20
gives the lowest conversion values but without any deactivation
detected during the test; and the catalyst with the lowest aluminum
For these catalysts, the main products of the HDS reaction were
biphenyl (BP), cyclohexylbenzene (CHB), bicyclohexyl (BCH), ben-
zene (B) and cyclohexane (CH). The total conversion of the HDS
reaction was calculated from the ratio of converted DBT/initial DBT.
The selectivity to the different reaction products was calculated
considering BP, CHB, B and CH as the only products obtained from
HDS. Assuming that the reaction is pseudo-first order, then the
reaction rate constants of HDS (kHDS) were calculated according
to Eq. (1):
ꢀ
ꢁ
F
kHDS = −
ln(1 − x)
(1)
W
in which, F is the feed rate of dibenzothiophene (mol min−1), W is
the catalyst mass (g) and x is the fractional conversion.
3. Results and discussion
3.1. Catalytic activity in the HDS of DBT
RuS2 based catalysts supported on aluminum-pillared clays
were tested in the dibenzothiophene (DBT) hydrodesulfuriza-
tion (HDS) reaction at different temperatures (260–360 ◦C), under
3.0 MPa of H2, with a flow rate of 100 mL min−1 and with hourly
space velocities (WHSV) of 32 h−1. Fig. 1 depicts the evolution of
the conversion for all the catalysts.
As can be clearly seen from this figure, the catalytic activ-
ity strongly depends on the support used. The support acidity
determined by NH3-TPD measurements, through out acidity val-
ues ranging from 330 to 360 ␮mol NH3 g−1 for all the pillared clays,
ruling out the support acidity as an important factor to under-
stand the activity of the ruthenium sulfide phase. At low reaction
temperatures, all the catalysts present similar conversion values,
but differences arise when the reaction temperature is increased.
Fig. 2. Evolution of the conversion at 360 ◦C, as a function of the reaction time.

A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
91
Table 1
Rate constants for the sulfided catalysts in the DBT HDS reaction.
Catalyst
Reaction temperature (◦C)
Stability test at
360 ◦C (12 h)
260
280
300
320
340
360
k × 105 (mol gcat−1 min−1
)
Al 5
0.15
0.25
0.15
0.32
0.46
0.58
0.45
0.69
1.68
1.05
1.12
0.73
2.16
2.39
2.51
1.61
2.69
3.22
4.10
1.78
4.02
4.35
6.36
2.43
3.6
2.4
2.6
1.9
Al 10
Al 15
Al 20
content, Al 5, proved to be the most active and stable catalyst, with
constant conversion of 70%.
Comparing the diffractograms, it can be seen that the Al 15 cat-
alyst has the least well defined and least intense RuS2 diffraction
peaks, indicating a lower and less crystalline active phase forma-
tion, i.e., a better dispersion. Thus, considering the main diffraction
line of RuS2 phase located at 2ꢁ = 31.8◦, and by applying the Scherrer
equation, the particle size was calculated. The particle size follows
the order: Al 20 (10.9 nm) > Al 5 (7.5 nm) > Al 10 (6.9 nm) > Al 15
(5.9 nm).
In order to know the effect of the reduction conditions employed
in the HDS reaction on the RuS2 phase stability, the X-ray diffrac-
tion patterns of spent catalysts were also recorded (Fig. 4b). The
diffractograms of spent catalysts show a considerable decrease in
the intensity of the diffraction lines from the RuS2 phase, indicat-
ing that a loss of the diffraction domains occurs during the catalytic
test, probably due to the partial reduction of ruthenium sulfide into
metallic ruthenium as previously reported [17,18]. Catalyst Al 10
suffers the greatest decrease in the RuS2 diffraction signal intensity
and catalyst Al 15 has a broad shoulder at ca. 2ꢁ = 43◦ corresponding
to the [1 0 1] plane of the metallic phase, indicating the formation
of this phase during the catalytic test. Catalyst Al 10 does not have
such a shoulder leading us to conclude that a partial reduction of
the RuS2 phase occurs but to a lesser extent than that observed in
the case of Al 15. The formation of metallic ruthenium cannot be
ruled out in the other catalysts, since particles with a dimension
smaller than the detection limit of XRD could have been formed.
From conversion values and by applying Eq. (1), the rate con-
stants at all reaction temperatures and after the stability tests were
calculated, and the corresponding values are shown in Table 1. The
kHDS increases with the temperature, presenting Al 15 catalyst the
highest catalytic activity at 360 ◦C. However, when this catalyst is
maintained at 360 ◦C for 12 h its activity decays, decreasing the kHDS
a 60%; followed by Al 10 with a loss of activity of 45%; while Al 5
and Al 20 catalysts hardly change their initial activity at 360 ◦C, i.e.,
they both are the most stable ones.
With regards to the selectivity values, the literature fully
describes the dibenzothiophene HDS reaction taking place along
two different pathways [33]. The first one, called the direct desul-
furization route (DDS), gives biphenyl (BP) as the main product;
and the second one, called the hydrogenation (HYD) pathway, gives
cyclohexylbenzene (CHB) and bicyclohexyl (BCH) as the main prod-
ucts. Here, in all cases, the products detected were BP, CHB, BCH
with cyclohexane (CH) and benzene (B) in minor quantities, com-
ing from the cracking of the former. The BP/CHB ratio is determined
by the pathways followed. In our case this ratio was in all cases close
to 7 at 360 ◦C, indicating the DDS route as dominant for these cat-
alytic systems with high temperatures. However, at lower reaction
temperatures (<320 ◦C) the reaction products BCH (bicyclohexil), B
(benzene), and CH (cyclohexane) were also observed.
The evolution of the selectivity for the ruthenium sulfide based
catalysts as a function of reaction temperature is depicted in Fig. 3.
In all cases, and as previously pointed, the product coming from
the DDS route, BP, is the main reaction product whose formation
increases with temperature. The CHB production goes down as
expected, because the hydrogenation is not favored at high tem-
peratures. However, the selectivity presented by Al 20 is slightly
different (Fig. 3D), since it is the only catalyst that produces BCH
at low temperatures (10–20%) up to 300 ◦C. From 260 to 300 ◦C the
selectivity to BCH decreases and that of CH increases, suggesting
the cracking of BCH molecule as responsible for CH formation over
Al 20 catalyst.
3.2.2. N2 adsorption-desorption at −196 ◦C
The textural properties of the supports and catalysts were
obtained by N2 adsorption-desorption isotherms at −196 ◦C. The
shape of the adsorption-desorption isotherms (not shown here)
were classified, according to IUPAC, as a combination of I type
isotherm at low relative pressures and IIb type isotherm at higher
relative pressures, with an hysteresis loop H4 type [34]. Such
isotherms are characteristic of slit-shaped pores or assemblies of
platy particles. The calculated specific surface area, micropore vol-
ume and total pore volume of the supports and sulfided catalysts
are compiled in Table 2. After the pillaring process, the clays suffer
3.2. Characterization results
Table 2
Textural properties of the supports and sulfided catalysts.
3.2.1. XRD
a
b
SBET (m2/g)
VT (cm3/g)
Vꢂp (cm3/g)
The X-ray diffraction technique allows us to identify the differ-
ent crystalline phases present. In order to know if the RuS2 phase
has been formed, the X-ray diffraction profiles of both fresh and
spent catalysts after temperature testing were recorded and are
shown in Fig. 4.
Supports
Natural montomorillonite
PILC AL 5
PILC AL 10
PILC AL 15
PILC AL 20
64
184
279
283
285
0.11
0.15
0.18
0.19
0.17
0.01
0.06
0.10
0.11
0.10
The alumina pillared supports present similar d0 0 1 values
(18.0–18.5 A), typical of this type of materials. The diffractograms
˚
Sulfided catalysts
of freshly sulfided catalysts (Fig. 4a) show several diffraction peaks.
In all cases, the diffraction lines located at 2ꢁ = 4.8, 19.8, 35.1 and
61.9◦ are due to the aluminum-pillared clay pure support; while
those found at 2ꢁ values of 27.5◦, 31.8, 45.7 and 54.1◦, belong to the
[1 1 1], [2 1 0], [2 1 1], and [3 1 1] crystallographic planes of the RuS2
phase with a pyrite-type structure (PDF Card No. 00-012-0737).
Al 5
118
148
179
149
0.09
0.16
0.12
0.10
0.03
0.04
0.06
0.05
Al 10
Al 15
Al 20
a
VT Pore total volumen.
Vꢂp Microporous volumen.
b

92
A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
Fig. 3. Selectivity of (A) Al 5; (B) Al 10; (C) Al 15; (D) Al 20 catalysts, where B, BCH, BP, CHB and CH stand for benzene, bicyclohexyl, biphenyl, cyclohexylbenzene and
cyclohexane, respectively.
an important increase in the surface area with regards to the par-
ent montmorillonite. The pillared clays present very similar values
with the exception of PILC AL 5 with the lowest surface area. Con-
sidering the Si/Al ratio obtained from the chemical composition of
the pillared clays, the sample PILC Al 5 has the highest value and
the change in the ratio for the other samples is not very important
and suggests that the addition of higher amounts of Al3+ ions has
no new effects and therefore the textural properties of the clays
hardly change.
that the micropore volume undergoes a considerable decrease after
metal incorporation, confirming the previous statement. Catalyst
Al 5 suffers a smaller decrease in surface area with regards to its
pristine support, although the textural properties of this support
are the lowest of the catalysts studied. The pore size distribu-
tion, calculated by the NLDFT method [35] ranged in all cases
˚
between 11 and 12 A and those of the pure supports ranged from
˚
14 to 16 A.
After the incorporation of the precursor salt and subsequent sul-
fidation, the specific surface area of the sulfided catalysts is lower
than that of the pristine supports. This is attributed to the loca-
tion of the active phase at the entrance of the porous structure,
causing a partial blockage of the interlayer space of the aluminum-
pillared clay. From the data compiled in Table 2, it can be seen
3.2.3. TEM
Transmission electron microscopy was performed in order to
elucidate the size, morphology and distribution of the active phase.
Fig. 5 shows the micrographs of both sulfided and spent catalysts.
In all cases the lamellar structure of aluminum-pillared clays are
clearly observed.
Fig. 4. X-ray diffraction profiles of fresh and spent catalysts after temperature test.

A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
93
Fig. 5. TEM micrographs of fresh (A) Al 5; (C) Al 10; (E) Al 15; (G) Al 20 and spent catalysts (B) Al 5; (D) Al 10; (F) Al 15; (H) Al 20.

94
A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
Table 3
Spectral parameters of fresh and spent catalysts.
Sample
Binding Energy (eV)
Ru 3p3/2
Atomic ratio
Ru/(Al + Si)
S 2p3/2
S/Ru
2−
S2
RuS2
Run+
Fresh catalyst
Al 5
Al 10
Al 15
Al 20
461.5 (91.8)
461.8 (85.9)
461.6 (85.8)
461.6 (89.8)
463.1 (8.2)
465.5 (14.1)
463.9 (14.2)
464.4 (10.2)
162.6
162.8
162.8
162.4
0.12
0.04
0.14
0.06
2.9
2.8
2.5
2.3
Spent catalyst
Al 5
Al 10
Al 15
Al 20
461.8 (90.8)
461.8 (84.1)
462.1 (83.7)
461.6 (89.8)
464.8 (9.2)
465.1 (15.9)
464.6 (16.3)
464.1 (10.2)
162.6
162.7
162.9
162.5
0.15
0.07
0.12
0.20
1.8
1.3
1.1
2.1
In brackets the percentage of each species on the support surface.
The general trend observed is the presence of highly dispersed
RuS2 particles and some agglomerates which are bigger than the
interlayer space. Particles lower than 10 nm are observed in all
cases that are located outside the pores as long as the pore size is
for Al 10 and Al 15 catalysts, along with a decrease in the S 2p sig-
nal intensity for the Al 15 catalyst. This indicates a loss of sulfur
during the test, as previously found from the XRD results. In fact,
by considering the S/Ru atomic ratio values, which are higher than
the theoretical one (2.0) for the fresh catalysts, and the decrease in
these values after the catalytic test, we can confirm the loss of sulfur
during the catalytic test. This is a much more important considera-
tion for catalysts Al 10 and Al 15 (1.3 and 1.1), which could explain
the loss of activity found for these systems during the stability test.
On the other hand, the most stable catalysts (Al 5 and Al 20) have a
S/Ru atomic ratio close to 2 after the catalytic test, indicating that
the pyrite structure on these two catalysts is more stable.
˚
11–12 A, in all cases. Moreover, the micrographs show a probable
restructuring of the active phase for catalysts Al 10 and Al 15, where
a change of the active phase morphology is noticeable after the cat-
alytic reaction (Fig. 4D and F). This is also apparent for catalyst Al
20 but this sample is less affected than the others. Meanwhile, the
Al 5 sample was the least affected morphologically and hence its
catalytic activity has probably hardly been affected.
3.2.4. XPS
The chemical state of the elements present on the catalyst sur-
face was studied by XPS. To this end O 1s, Si 2p, Al 2p, Ru 3p, S 2p
and Cl 2p signals were analyzed for sulfided and spent catalysts.
The O 1s, Al 2p, and Si 2p BE values were 538.2 eV, 75.4 eV and
103.4 eV, respectively. These values remained practically constant
in all cases and also after the catalytic test, demonstrating the high
stability of the Al pillared clay during the reaction. The Cl 2p signal
was not detected in any case, which indicates that chloride species
are removed during the sulfidation process. The Ru 3p3/2, S 2p3/2
binding energy values, as well as the Ru/(Si + Al) and S/Ru atomic
ratios of sulfided and spent catalysts, are shown in Table 3.
The Ru 3p core level spectra of sulfided catalysts (Fig. 6A) can
be decomposed into two contributions. The first ranges between
461.5 and 462.1 eV, and is ascribed to a RuS2 compound [36]; the
second ranges between 463.1 and 465.5 eV, being attributed to Run+
species [37–39]. The S 2p signal was studied to establish whether
S22− ions were formed after the sulfiding procedure. The S 2p spec-
trum (Fig. 6B) gives an asymmetrical signal fitted into one doublet, S
2p3/2 and S 2p1/2 coming from the spin–orbit splitting. The BE value
of the S 2p3/2 component ranges between 162.4 and 162.9 eV, and
3.3. Discussion
The hydrodesulfurization of DBT over ruthenium sulfide
supported on Al pillared clays mainly involves the direct desulfu-
rization route (DDS) with biphenyl being the main product found
in all cases. Hence the hydrogenolysis reaction is dominant over
the hydrogenation reaction. All RuS2 catalysts are active in the DBT
HDS reaction. Nonetheless, Al 10 and Al 15 suffer a considerable
decrease in their catalytic activity after 6 h on stream, while Al 5 and
Al 20 catalysts were found to be the most stable, with Al 5 giving a
constant conversion of ca. 70% after 12 h on stream. These catalytic
results are of the up most importance when considering the search
for a support which avoids a difficult synthetic procedure.
Characterization results from XRD and XPS reveal that the
formation of RuS2 with a pyrite-type structure occurs after the sul-
fidation process. From XRD patterns it can be inferred that the least
stable catalysts are those that suffer a greater sulfur loss during the
catalytic test, diminishing the intensity of RuS2 diffraction lines or
even showing diffraction signals arising from the presence of metal-
lic ruthenium, as in the case of Al 15. This is also confirmed by XPS,
where the partial reduction of the RuS2 phase during the catalytic
test is clearly observed as shown by the S/Ru atomic ratios found
for the spent catalysts, explaining the loss of activity exhibited in
catalysts Al 10 and Al 15. Moreover, micrographs of the spent cat-
alysts (Fig. 4D and F) show a possible change in the morphology of
the active phase which could be related to the decrease in catalytic
activity of catalysts Al 10 and Al 15. This result is a consequence of
using high temperatures and strong reducing conditions, i.e. severe
reaction conditions which could lead to a restructuring of the active
phase [2].
2−
is assigned to S2 forming a pyrite-type structure [12,13,40] such
as RuS2.
The influence of the clay used as a support on the Ru 3p and
S 2p core level spectra reveals that catalyst Al 5 gives the most
intense Ru 3p and S 2p signals, indicating a surface enrichment
of such compounds on the catalyst surface. Contrary to this, cat-
alyst Al 10 gives the weakest intensities of these signals due to a
greater dispersion of the active phase on this catalyst after sulfida-
tion. Meanwhile, catalysts Al 15 and Al 20 show a similar ruthenium
and sulfur distribution. The analysis of the Ru 3p and S 2p core level
spectra for spent catalysts are shown in Fig. 6B and D. It can be seen
that BE (Table 3) slightly changed after the catalytic test. The main
differences found are the increase in the intensity of the Ru 3p3/2
band assigned to Run+ species (∼463.5 eV) and the correspondent
increase of the percentages of such species on the catalyst surface
Catalysts Al 5 and Al 20 showed an interesting but contradictory
catalytic behavior. On one hand, catalyst Al 5 has the least Al3+ g−1
clay ratio (Si/Al = 2.29) and catalyst Al 20 the highest Al3+ g−1 clay
ratio (Si/Al = 1.94). Al 5 catalyst was the least affected by the incor-
poration of the precursor salt (RuCl3·3H2O), since the decrease in

A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
95
Fig. 6. Ru 3p3/2 core level spectra of: (A) sulfided catalysts and (B) spent catalysts after stability test. S 2p core level spectra of: (C) sulfided catalysts and (D) spent catalysts
after stability test.
the surface area with regards to its pristine support was the lowest.
According to XRD, TEM micrographs, this catalyst also maintains its
morphology during the test. While XPS reveals that not great sul-
fur lost occurs during the test, pointing to the stability of the active
phase. On the other hand, Al 20 catalyst, possessing the highest
amount of aluminum incorporated, exhibits the lowest conversion
although a more stable behavior than Al 10 and Al 15. The decrease
in the surface area is hardly the same than Al 10 and Al 15 ones,
TEM micrographs indicate that the RuS2 phase is less dispersed but
less affected by the catalytic test, while XRD and XPS point to a
higher stability of the RuS2 pyrite structure due to a lower sulfur
loss during the run with regards to Al 10 and Al 15. These results
suggest that the amount of aluminum incorporated is clue to stabi-
lize the RuS2 phase, being the catalysts presenting the lowest and
the highest aluminum content the most stable catalysts.
In order to highlight the importance of the results presented
here, where a natural material such as a pillared montmorillonite
worked out to be a good support in hydrotreating reactions, the
activity of these catalysts can be compared with that presented
RuS2 supported on a commercial ␥-Al2O3. Thus, in a previous work
[19], it has been prepared a ruthenium sulfided catalyst supported
on a commercial support such as ␥-Al2O3 and tested under the same
conditions exposed here. The catalyst presented a kHDS at 340 ◦C
of 1.48 mol gcat−1 min−1, where the catalyst was more active, and
much lower than the values attained here with all the catalysts at
this temperature (Table 1).
observed to decrease. This is probably, due to the reduction of the
active phase and a change in its morphology as a consequence of the
severe reaction conditions used in this catalytic process. Catalyst Al
20 presented the lowest catalytic activity, possibly attributed to a
fairly dispersed active phase. However, the Al 5 sample showed a
good catalytic performance due to a well dispersed active phase
that inferred the highest catalytic stability throughout the test.
Acknowledgements
We gratefully acknowledge the support from the Ministry of Sci-
ence and Innovation, Spain (MICINN, Espan˜a) through the project
MAT2009-10481, the regional government (JA) through the Excel-
lence projects (P06-FQM-0166 and P07-FQM-5070) and FEDER
funds. A.R.P. thanks the CONACyT (México) for its financial support
(Scholarship No. 189933). A.I.M. also thanks the MICINN, Spain, for
a Juan de la Cierva contract.
References
[1] C.S. Hsu, P.R. Robinson, Environmental Pollution Control, Practical Advances in
Petroleum Processing, Springer, New York, 2006, pp. 395–447.
[2] R.R. Chianelli, G. Berhault, B. Torres, Catal. Today 147 (2009) 275–286.
[3] M. Breysse, P. Afanasiev, C. Geantet, M. Vrinat, Catal. Today 86 (2001)
5–16.
[4] M. Egorova, R. Prins, J. Catal. 224 (2004) 278–287.
[5] C. Song, X. Ma, Appl. Catal. B 41 (2003) 207–238.
[6] Z. Vít, J. Cinibulk, D. Gulková, Appl. Catal. A 272 (2004) 99–107.
[7] R.P. Toulhoat, S. Kasztelan, G. Kresse, J. Hafner, Catal. Today 50 (1999)
629–636.
4. Conclusions
[8] T.A. Pecoraro, R.R. Chianelli, J. Catal. 67 (1981) 430–445.
[9] Y. Aray, A.B. Vidal, J. Rodriguez, M.E. Grillo, D. Vega, D.S. Coll, J. Phys. Chem. C
113 (2009) 19545–19557.
[10] P. Raje, S.-J. Liaw, B.H. Davis, Appl. Catal. A 150 (1997) 319–342.
[11] P. Raje, S.-J. Liaw, B.H. Davis, Appl. Catal. A 150 (1997) 297–318.
Ruthenium sulfide based catalysts supported on aluminum-
pillared clays were active in the HDS reaction of DBT. During
catalytic testing, the catalytic activity for Al 10 and Al 15 was

96
A. Romero-Pérez et al. / Catalysis Today 187 (2012) 88–96
[12] J.A. De Los Reyes, S. Göbölös, M. Vrinat, M. Breysse, Catal. Lett. 5 (1990)
17–24.
[13] J.A. De Los Reyes, M. Vrinat, C. Geantet, M. Breysse, Catal. Today 10 (1991)
645–664.
[29] O. Exner, Nature 227 (1970) 366–367.
[30] R.A. Pis-Diez, H. Jubert, J. Mol. Catal. 83 (1993) 219–235.
[31] G. Berhault, M. Lacroix, M. Breysse, F. Maugé, J.C. Lavalley, H. Nie, L. Qu, J. Catal.
178 (1998) 555–565.
[14] M. Wojciechowska, M. Pietrowski, B. Czajka, Catal. Today 65 (2001) 349–353.
[15] F. Fréchard, P. Sautet, Surf. Sci. 336 (1995) 149–165.
[32] M. Breysse, G. Berhault, S. Kasztelan, M. Lacroix, F. Maugé, G. Perot, Catal. Today
66 (2001) 15–22.
[16] T. Cai, Z. Song, J.A. Rodriguez, J. Hrbek, J. Am. Chem. Soc. 126 (2004) 8886–8887.
[17] J. Quartararo, S. Mignard, S. Kasztelan, J. Catal. 192 (2000) 307–315.
[18] A. Romero-Pérez, A. Infantes-Molina, E. Rodríguez-Castellón, A. Jiménez-López,
Appl. Catal. B 97 (2010) 257–268.
[33] K.G. Knudsen, B.H. Cooper, H. Topsøe, Appl. Catal. A 189 (1999) 205–215.
[34] J. Rouquerol, F. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids
Principles, Methodology and Applications, Academic Press, London, 1999, pp.
355–399 (Chapter 11).
[19] D. Eliche-Quesada, E. Rodríguez-Castellón, A. Jiménez-López, Microporous
Mesoporous Mater. 99 (2007) 268–278.
[35] G. Leofanti, M. Podovan, G. Tozzola, B. Venturelli, Catal. Today 41 (1998)
207–219.
[20] K. Sapag, S. Mendioroz, Colloids Surf. A 187–188 (2001) 141–149.
[21] A. Gil, L.M. Gandía, Catal. Rev. Sci. Eng. 42 (2000) 145–212.
[22] J.M. Adams, R.W. Mccabe, in: B.K.G. Bergaya, G. Theng, Lagaly (Eds.), Handbook
of Clays Science, vol. 1, Elsevier, Amsterdam, 2006, pp. 541–581.
[23] R.Q. Long, R.T. Yang, Appl. Catal. B 24 (2000) 13–21.
[36] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in: J. Chastain (Ed.), Hand-
book of X-ray Photoelectron Spectroscopy, Perkin-Elmer, Minnesota, 1992, pp.
114–115.
[37] V. Mazzieri, F. Coloma-Pascual, A. Arcoya, P.C. L’Argentière, N.S. Fígoli, Appl.
Surf. Sci. 210 (2003) 222–230.
[24] H.J. Chae, I.S. Nam, S.B. Hong, Stud. Surf. Sci. Catal. 146 (2003) 697–700.
[25] J. Arfaoui, L.K. Boudali, A. Ghorbel, G. Delahay, Catal. Today 142 (2009) 234–238.
[26] P. Salerno, S. Mendioroz, A. López-Agudo, Appl. Catal. A 259 (2004) 17–28.
[27] M. Lenarda, L. Storaro, R. Ganzerla, R. Bertoncello, J. Mol. Catal. Mol. A: Chem.
144 (1999) 151–158.
[38] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochman, Advanced Inorganic Chem-
istry, John Wiley and Sons Inc., USA, 1999, pp. 1012–1013.
[39] S.E. Livingstone, in: J.C. Bailar, M.J. Emeléus, R. Nyholm, A.F. Trotman-Dickenson
(Eds.), Comprehensive Inorganic Chemistry, vol. 3, Pergamon Press, Oxford,
1973, pp. 1191–1193.
[28] M.J. Pérez-Zurita, G. Pérez Quintana, G.J. Biomorgi, C.E. Scott, Prepr. Symp. Am.
Chem. Soc. Div. Fuel Chem. 48 (1) (2003) 90–91.
[40] P.C.H. Mitchell, C.E. Scott, J.P. Bonnelle, J.G. Grimblot, J. Catal. 107 (1987)
482–487.