Influence of iridium content on the behavior of Pt-Ir/Al2O 3 and Pt-Ir/TiO2 catalysts for selective ring opening of naphthenes

Article abstract of DOI:10.1016/j.apcata.2012.12.015

The influence of Ir content on the properties of Pt-Ir/Al₂O ₃ and Pt-Ir/TiO₂ catalysts for selective ring opening of naphthenes was studied. It was found that these catalysts display a strong Pt-Ir interaction but only a weak metal-support interaction. Catalyst acidities depend on the metal loading, but opposite effects were observed on alumina (decrease) or titania (increase) as the metal loading increased. The results obtained from test reactions (cyclohexane dehydrogenation and cyclopentane hydrogenolysis) showed that titania supported catalysts were less active than their alumina supported counterparts. This behavior could be due to the partial blockage of metallic sites by migrated TiO_x species and the sinterization of metallic phase during the reduction step. The methylcyclopentane ring opening reaction was found to occur through a partially selective mechanism, and an increase in activity as the Ir loading increased was observed. The selective mechanism was favored by higher total metal loadings, possibly due to an increase in the size of metallic aggregates. Alumina supported catalysts present higher ring opening selectivities. The activity for decalin ring opening increased both with metal loading and reaction temperature level.

Full text of DOI:10.1016/j.apcata.2012.12.015

Applied Catalysis A: General 453 (2013) 167–174
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Influence of iridium content on the behavior of Pt-Ir/Al2O3 and
Pt-Ir/TiO2 catalysts for selective ring opening of naphthenes
María A. Vicericha, Viviana M. Beniteza, Catherine Especelb, Florence Epronb,
Carlos L. Piecka,∗
a Instituto de Investigaciones en Catálisis y Petroquímica (INCAPE) (FIQ-UNL, CONICET), Santiago del Estero 2654, S3000AOJ, Santa Fe, Argentina
b Institut de Chimie des Milieux et Matériaux de Poitiers, IC2MP, Université de Poitiers, 4 rue Michel Brunet, 86022 Poitiers Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The influence of Ir content on the properties of Pt-Ir/Al2O3 and Pt-Ir/TiO2 catalysts for selective ring
opening of naphthenes was studied. It was found that these catalysts display a strong Pt–Ir interaction
but only a weak metal–support interaction. Catalyst acidities depend on the metal loading, but opposite
effects were observed on alumina (decrease) or titania (increase) as the metal loading increased. The
results obtained from test reactions (cyclohexane dehydrogenation and cyclopentane hydrogenolysis)
showed that titania supported catalysts were less active than their alumina supported counterparts.
This behavior could be due to the partial blockage of metallic sites by migrated TiOx species and the
sinterization of metallic phase during the reduction step. The methylcyclopentane ring opening reaction
was found to occur through a partially selective mechanism, and an increase in activity as the Ir loading
increased was observed. The selective mechanism was favored by higher total metal loadings, possibly
due to an increase in the size of metallic aggregates. Alumina supported catalysts present higher ring
opening selectivities. The activity for decalin ring opening increased both with metal loading and reaction
temperature level.
Received 19 September 2012
Received in revised form 7 December 2012
Accepted 11 December 2012
Available online 26 December 2012
Keywords:
Selective ring opening
Pt–Ir
Decalin
Methylcyclopentane
© 2012 Elsevier B.V. All rights reserved.
1. introduction
Onyestyák et al. [3] and McVicker et al. [4] studied the ring open-
ing reaction for alkyl substituted C6 C10 naphthenes and found
Diesel fuel production technology is very complex as it implies
the careful selection and mixing of different petroleum fractions
to fulfill strict specifications [1]. The most important parameter
to measure diesel quality is the cetane index (CI). Current diesel
engines are designed to run on fuels with 40–55 CI values. A
common way to increase CI values of petroleum fractions is the
hydrogenation of its aromatic components, but this improvement
is moderate as the CI of naphthenes are relatively low [2]. This lim-
itation is particularly important in the upgrading of aromatic rich
petroleum fractions such as LCO (light cycle oil) from fluid catalytic
cracking (FCC). The additional opening of at least one naphthenic
ring is necessary in this case to obtain reasonable CI values.
There are two main types of hydrocracking reactions for naph-
thenes. External C C bonds of hydrocarbon chains attached to
naphthenic rings can be broken resulting in lower molecular weight
products. On the other hand, cyclic C C bonds can also be broken
(ring opening reaction, RO), then the molecular mass of products is
nearly the same as for reactants in this case.
a remarkable selectivity of Ir catalysts towards the opening of
five-carbon naphthenic rings. The ring opening rates significantly
decreased as the alkyl substitution degree increased, and were
roughly proportional to the number of secondary C C bonds. Pt
catalysts were more active for the scission of substituted C
C
bonds. RO rates were nevertheless sensitive to the cis/trans ratio
of methyl substituted cyclopentanes. On the other hand, the acid-
catalyzed naphthenic ring opening is believed to proceed through
an entirely different mechanism on Brønsted surface sites, starting
with protolytic cracking followed by chain reactions of the resulting
carbenium ions [5,6]. Kubicka et al. [7,8] found that catalyst acid-
ity has a strong influence on the selective ring opening of bicyclic
naphthenes.
The simple methylcyclopentane (MCP) ring opening reaction
has been widely used as a test reaction mainly for Pt based catalysts
[9–15].
Possible products from these RO reactions are n-hexane (nC6),
2-methylpentane (2MP), 3-methylpentane (3MP) and light C1 C5
hydrocarbons. Galperin et al. [11] explained these results on the
basis of the coexistence of two simultaneous RO mechanisms: a
selective one leading to 2MP and 3MP on high size metal parti-
cles, and a non-selective path with additional formation of nC6 on
∗
Corresponding author. Tel.: +54 342 4533858; fax: +54 342 4531068.
E-mail address: pieck@fiq.unl.edu.ar (C.L. Pieck).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.12.015

168
M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
smaller metal particles. 2MP and 3MP would be produced only on
metallic sites, but the formation of cyclohexane, benzene and nC6
would require the presence of both metallic and acid sites on the
catalyst. Current information in the open literature indicates that
supported Ir catalysts are more active and selective due to their
low ability to break the exocyclic chains attached to naphthenic
rings. The distribution of products obtained from C5 naphthenes
is better than that obtained from C6 naphthenes. There are well
known catalysts with high activity, selectivity and stability for the
isomerization of cyclohexane to methylcyclopentane [12] which
could be coupled to RO catalysts to improve their performance, but
an unavoidable isomerization of existing linear chains would cause
undesirable cetane number losses.
Some patents regarding selective ring opening claim the
advantages of Ir-based catalysts using mesoporous supports with
medium acidity which is regulated by the controlled addition of
alkaline ions [16]. A modification of the selective ring opening (SRO)
process coupled with catalytic assisted sulfur traps (CAST) has been
recently reported, which presents some advantages over the tra-
ditional hydrocracking process to upgrade refinery cuts used for
diesel fuels formulation [17]. A drawback of this approach is that a
virtually complete removal of S present in feeds is required as cat-
alytic hydrogenolysis on metals is extremely sensitive to sulphur
poisoning.
In a previous work [18], we reported the performance of Pt-
Ir/Al2O3 catalysts with and without Mg addition prepared by
catalytic reduction method for SRO of decalin. These catalysts
present a high interaction between Pt and Ir due to the prepa-
ration method used and are very active for SRO. In order to gain
insight about the role of the support and the preparation method
on the catalyst performance, we now make a comparative study of
Pt-Ir catalysts supported on alumina or titania prepared by coim-
pregnation. The main goal of the experimental work is to find
suitable catalysts for the SRO of mono and polycyclic alkanes with
good activity, selectivity and stability. Model molecules selected
to test catalyst performance were cyclopentane, methylcyclopen-
tane, cyclohexane and decalin. The influence of support acidity is
also studied.
2.2. Temperature-programmed desorption of pyridine (TPD)
The amount and strength distribution of acid sites were assessed
by means of the temperature-programmed desorption of pyridine
(Py). 200 mg of catalyst were first immersed for 4 h in a closed vial
containing pure pyridine (Merck, 99.9%). Then the vial was opened
and the excess base was allowed to evaporate at room conditions
until apparent dryness. The sample was then loaded into a quartz
tube of 0.4 cm diameter over a quartz wool plug. A constant flow of
nitrogen (40 cm3 min−1) was kept through the sample. A first step
of desorption of weakly physisorbed base and sample stabilization
was performed by heating the sample at 110 ◦C for 2 h. Then the
temperature of the oven was raised to a final value of 550 ◦C at a
heating rate of 10 ◦C min−1. The reactor outlet was directly con-
nected to a gas chromatograph equipped with a flame ionization
detector.
2.3. Temperature-programmed reduction (TPR)
This technique allows gathering information about the inter-
action of the metal components by means of the measurement of
the hydrogen consumption during the reduction of surface oxide
species at constant heating rates. The temperature at which reduc-
tion occurs and the number of reduction peaks depend on the
oxidation state of the metals, the interaction of the oxides among
them and with the support, and the possible catalytic action of Pt
or other elements present or generated during reduction. These
tests were performed in Ohkura TP2002 equipment provided with
a thermal conductivity detector. Prior to each TPR test the catalyst
samples were pretreated in situ by heating under flowing air at
300 ◦C during 1 h. Then they were heated from room temperature
to 700 ◦C at 10 ◦C min−1 under a controlled flow gas stream of 5.0%
hydrogen in argon.
2.4. Hydrogen chemisorption
The metallic accessibility was determined from H2 chemisorp-
tion measurement, carried out in a pulse chromatographic system
equipped with a thermal conductivity detector. The catalysts were
first reduced under H2 at the 300 or 500 ◦C for 1 h, then evacuated
at the same temperature under Ar for 2 h and cooled down to room
temperature. Pulses of H2 were injected at room temperature, every
minute up to saturation and the amount of hydrogen required was
calculated (HC1). After 30 min of purging under pure Ar, in order to
eliminate the reversible part of the chemisorbed hydrogen, a new
series of pulses was injected over the sample, and the hydrogen
consumption (HC2) was calculated again. The irreversible part was
taken as HC = HC1 − HC2 and allows estimating the metallic accessi-
bility taking the stochiometric ratio between hydrogen and surface
atoms (H/Ms) equal to one.
2. Experimental
2.1. Catalyst preparation methods
␥-Al2O3 (Cyanamid Ketjen CK-300, pore volume = 0.5 cm3 g−1
Sg = 180 m2 g−1
35–80 mesh) was calcined during 4 h at
500 ◦C under flowing dry air. TiO2 (pore volume = 0.8 cm3 g−1
,
,
,
Sg = 80 m2 g−1, 35–80 mesh) was calcined during 2 h a 500 ◦C
under flowing dry air. The X-ray diffraction of prepared titania
shows mainly an anatase structure with some traces of rutile
phase. Samples were let in contact with the amount of aqueous
solutions of metallic precursors (H2PtCl6 and H2IrCl6) required for
the desired metal loading during 1 h at ambient temperature in
order to get an uniform distribution of metallic salts onto the sup-
port. The resulting mixture was kept at 70 ◦C until a dry solid was
obtained. The drying process was completed in an oven (overnight
at 120 ◦C) and the samples were then stabilized by calcination
under flowing air during 4 h at 400 ◦C, and then cooled down to
ambient temperature under N2. Samples were finally reduced
under flowing H2 at 300 ◦C during 4 h. All heating and cooling steps
were done at 10 ◦C min−1. The concentration of aqueous solutions
of metallic precursors salts was adjusted to obtain the following
metal weight loadings on the finished catalysts: 1.0%Pt-0.3%Ir,
1.0%Pt-1.0%Ir, 1.0%Pt-2.0%Ir, 1.0%Pt and 0.3%Ir. Hereafter, the
catalysts are labeled Pt(x)-Ir(y)/Al2O3 or Pt(x)-Ir(y)/TiO2 where x
and y denote the content (wt%) of Pt and Ir, respectively.
2.5. Transmission electron microscopy (TEM) measurements
TEM measurements were performed on a JEOL 2100 electron
microscope operating at 200 kV with a LaB6 source and equipped
with a Gatan Ultra scan camera. All the samples were embedded in a
polymeric resin (spur) and cut into a section as small as 40 nm with
an ultramicrotome equipped with a diamond knife. Cuts were then
deposited on an Al grid previously covered with a thin layer of car-
bon. Average particle sizes were determined by measuring at least
140 particles for each sample analyzed, from at least five differ-
ꢀ
ꢀ
ent micrographs, using the following formula: d = nidi3
(di = diameter of the particle ni). Microanalysis of Pt and Ir was
carried out by energy dispersive X-ray spectroscopy (EDX) in the
nanoprobe mode.
/
nidi2

M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
169
2.6. Cyclopentane (CP) hydrogenolysis
The reaction was performed in a glass reactor (length 10 cm,
diameter 1 cm) under the following conditions: catalyst mass
150 mg, temperature 300 ◦C, pressure 0.1 MPa, H2 flow rate
40 cm3 min−1, cyclopentane flow rate 0.483 cm3 h−1. Cyclopentane
was fed to the reactor using a syringe pump. Before the reac-
tion was started, the catalysts were conditioned under flowing H2
(60 cm3 min−1, 300 ◦C, 1 h). The reaction products were analyzed by
online gas chromatography using a ZB-1 capillary column. Values
of initial conversion at 10 min time-on-stream are reported.
2.7. Cyclohexane (CH) dehydrogenation
The reaction was performed in a glass reactor (length 10 cm,
diameter 1 cm) under the following conditions: catalyst mass
50 mg, temperature 300 ◦C, pressure 0.1 MPa, H2 flow rate
36 cm3 min−1, cyclohexane flow rate 0.727 cm3 h−1. The CH was
introduced into the reactor using a syringe pump. Before the reac-
Fig. 1. TPR profiles for Pt-Ir/Al2O3 catalysts.
tion was started, the catalysts were reduced with H2 (60 cm3 min−1
,
300 ◦C, 1 h). The reaction products were analyzed by online gas
chromatography using a ZB-1 capillary column. All points reported
are mean values obtained by averaging 12 consecutive measure-
ments equally spaced along the run time. No significant catalyst
deactivation was observed in any run.
2.8. Methylcyclopentane (MCP) ring opening (RO)
The reaction was performed in a glass reactor (length 10 cm,
diameter 1 cm) at atmospheric pressure. Before the reaction, sam-
ples were reduced with H2 at 300 ◦C for 1 h. The conditions
used were: catalyst mass = 135 mg, H2 flow rate = 36 cm3 min−1
,
MCP flow = 0.362 cm3 min−1, reaction temperature = 250 ◦C, reac-
tion time = 2 h. The reaction products were analyzed on a Varian CX
3400 gas chromatograph equipped with a capillary column Phen-
omenex ZB-1 connected online.
Fig. 2. TPR profiles for Pt-Ir/TiO2 catalysts.
content increased. This fact indicates that its metallic phase con-
tains Pt–Ir ensembles or at least that Pt and Ir strongly interacts.
The monometallic Pt and Ir catalysts supported on TiO2 (Fig. 2)
show reduction peaks centered at 215 and 255 ◦C, respectively.
The bimetallic Pt(1.0)-Ir(0.3)/TiO2 and Pt(1.0)-Ir(1.0)/TiO2 catalysts
have large reduction peaks at about 170–190 ◦C and very broad
reduction bands at 300–600 ◦C. Large peaks at lower temperatures
could be due to the presence of Pt–Ir ensembles while the wider
reduction zones can be certainly attributed to reduction of the sup-
port [22]. The Pt(1.0)-Ir(2.0)/TiO2 catalyst has two reduction peaks:
the first peak at 205 ◦C could be due to Pt-Ir ensembles rich in Pt
and the second peak at 291 ◦C would correspond to reduction of
Pt–Ir ensembles rich in Ir or segregated Ir species [22].
2.9. Decalin ring opening (RO)
All SRO experiments were performed in a stainless steel,
autoclave-type stirred reactor. The reaction conditions were:
temperature = 300–325–350 ◦C, hydrogen pressure: 3 MPa, stirring
rate = 1360 rpm, volume of decalin = 25 cm3, catalyst loading = 1 g
and reaction time = 6 h. A sample was taken at the end of the
experiments and it was analyzed using a Varian 3400 CX gas chro-
matograph equipped with a capillary column (Phenomenex ZB-5)
and FID.
The pyridine thermodesorption technique provides valuable
information about the support acidity and acid strength distribu-
tion. The area under the desorption pattern is proportional to the
total acidity. Relative total acidity values, taking as reference the
monometallic Pt/Al2O3 catalyst, are shown in Table 1. The acidity
3. Results and discussion
The TPR profiles of Pt(1.0)-Ir(y) catalysts supported on Al2O3 or
TiO2 are shown in Figs. 1 and 2, respectively. Fig. 1 (Al2O3 supported
catalysts) shows that the monometallic Pt/Al2O3 catalyst has a
well-defined reduction peak centered at about 200 ◦C. We have
previously reported that Pt oxide species on alumina are reduced
at 220–250 ◦C [19,20]. The somewhat lower reduction temperature
reported here could be due to the absence of HCl in the impregna-
tion step. The monometallic Ir/Al2O3 catalyst has a wide reduction
zone centered at 240 ◦C. Such broad reduction pattern could be due
to the presence of several Ir species with different particle sizes.
Carnevillier et al. [21] reported that reduction at low temperature
is due to large particles of Ir oxides species, while reduction at high
temperature is due to Ir oxides species well dispersed onto the sup-
port [21]. Bimetallic Pt(1.0)-Ir(y)/Al2O3 catalysts have also broad
reduction peaks which are shifted to lower temperatures as the Ir
Table 1
Normalized pyridine TPD areas of Pt(1.0)-Ir(y)/Al2O3 and Pt(1.0)-Ir(y)/TiO2 catalysts.
Catalyst
Relative areas of pyridine TPD
Pt(1.0)/Al2O3
1.00
1.03
1.10
1.12
0.90
0.57
0.49
0.46
0.47
0.51
Pt(1.0)-Ir(0.3)/Al2O3
Pt(1.0)-Ir(1.0)/Al2O3
Pt(1.0)-Ir(2.0)/Al2O3
Ir(0.3)/Al2O3
Pt(1.0)/TiO2
Pt(1.0)-Ir(0.3)/TiO2
Pt(1.0)-Ir(1.0)/TiO2
Pt(1.0)-Ir(2.0)/TiO2
Ir(0.3)/TiO2

170
M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
Table 2
Mean conversion values of cyclohexane, initial (10 min time-on-stream) conversion
values of cyclopentane, hydrogen chemisorption and average particle size deduced
from TEM analysis of Pt(1.0)-Ir(1.0)/TiO2 catalyst reduced at 300, 400 and 500 ◦C.
Experiment
Reduction temperature (◦C)
300
400
500
Cyclohexane dehydrogenation (conversion %)
Cyclopentane hydrogenolysis (conversion %)
Hydrogen chemisorption (H/M %)
40.4
79.0
40
32.4
52.4
40
18.4
11.7
16
Average particle diameter (nm, from H/M)
Average particle diameter (nm, from TEM)
2.0
1.6
2.0
2.3
5.3
2.3
The cyclohexane dehydrogenation it is a non-demanding (struc-
ture insensitive) reaction so the morphological details of metallic
phase have no major influence. As shown in Fig. 3, the dehydro-
genating activity of Pt monometallic catalysts is higher than that
found for Ir monometallic catalysts. The better performance of Pt
monometallic catalysts was expected, since it is known that this
metal is more active for dehydrogenation than Ir [23–26] and the
metal loading of Ir monometallic catalysts is lower. In agreement
with the results reported by Ali et al. [24] who studied Pt–Ir sup-
ported on Al2O3 and Capula Colindres et al. [25] who studied Pt–Ir
supported on TiO2, bimetallic Pt–Ir catalysts are clearly more active
than monometallic ones for both series and the activity increases as
the Ir loading increases. This could simply be due to the increase of
total metal loading as Ir loading increases. Here, it can be observed
that Al2O3 supported catalysts are more active than TiO2 supported
ones. This behavior can also be due to partial blockage of Pt and Ir
species by TiOx (x < 2) migrated during reduction step. Such par-
tial coverage of the metal by the oxide support diminishes both
hydrogen and CO chemisorption, and is believed to occur during
the catalyst reduction step [27–30].
Hydrogenolysis is known to be a structure sensitive or “demand-
ing” reaction so highly dispersed metal crystallites have low activity
for such reaction [31]. It is known that Ir has a higher hydrogenolytic
activity than Pt [32], and also that Pt–Ir bimetallic ensembles are
much more active than any of these metals alone [23]. Fig. 4
shows that both for alumina or titania bimetallic catalysts series,
the hydrogenolytic activity increases as the Ir content increases.
The activity of monometallic Ir catalysts is higher than that of
monometallic Pt ones [32], as expected, even when the Pt loading is
higher. The activity is higher for alumina supported catalysts than
for titania supported ones. This result could be due to the occur-
rence of some migration of TiOx (x < 2) species towards the active
Pt–Ir phase and/or an increase of metallic particle size despite the
relatively low reduction temperature level used in the activation
steps.
Fig. 3. Average cyclohexane conversion for Pt-Ir/Al2O3 and Pt-Ir/TiO2 catalysts.
of bimetallic Pt-Ir/Al2O3 catalysts slightly increases as the Ir content
increases, and an opposite trend was found for bimetallic Pt-Ir/TiO2
catalysts. Taking into account the acidic character of Ir oxides, the
acidity behavior of alumina supported bimetallic catalysts is easily
understood. The apparently anomalous behavior of total acidity as
the Ir content increases found for titania supported bimetallic cata-
lysts may be due to a partial blockage of acid sites as a consequence
of migration of TiOx (x < 2) species during the catalyst reduction
procedure [22]. On the other hand, it can be seen that catalysts
supported on Al2O3 are more acidic than those supported on TiO2.
It must be recalled here that some suppression of acidity may be
desirable in order to favor the endocyclic hydrogenolysis over the
exocyclic bonds breaking. Exocyclic C C bonds rupture may be due
to cracking reactions on the strongest support acid sites. An exces-
sive acidity level could also lead to linear paraffins isomerization
with the subsequent cetane number losses.
Figs. 3 and 4 show values of reactant conversion for cyclohexane
(CH) dehydrogenation and cyclopentane (CP) hydrogenolysis test
reactions for both catalyst series. Cyclohexane dehydrogenation
runs proceeded with 100% selectivity to benzene and no catalyst
deactivation was detected. Therefore, reported conversion values
(Fig. 3) are mean values along 1 h time on stream (12 points). On
the other hand, the cyclopentane hydrogenolysis activity rapidly
declined during the runs due to unavoidable coke deposition so
that the conversion data shown in Fig. 4 were those found at 10 min
time-on-stream.
Additional experiments to determine the influence of reduc-
tion temperature on cyclopentane and cyclohexane conversion, i.e.
determination of metallic accessibility (by hydrogen chemisorp-
tion) and particle size (measured by TEM), were done in order
to verify the TiOx (x < 2) migration or sinterization hypothesis.
Table 2 shows cyclopentane and cyclohexane conversion for the
Pt(1.0)-Ir(1.0)/TiO2 catalyst activated at three reduction tempera-
ture levels. Also, the metallic accessibility and the average particle
diameter of the same catalysts are reported. According to TPR
results (Fig. 2), it is expected that all Pt and Ir are reduced to
their metallic states at these temperature levels. It can be observed
that both hydrogenolytic and dehydrogenation activity decrease as
reduction temperature increases for this TiO2 supported catalyst.
On the other hand, the reduction temperature level has no influ-
ence on the hydrogenolysis or dehydrogenation activity for Al2O3
supported bimetallic catalysts (results not shown). The decrease
in metallic accessibility (i.e. lower hydrogen chemisorption capaci-
ties) leads to lower hydrogenolytic and dehydrogenation activities,
Fig. 4. Cyclopentane conversion for Pt-Ir/Al2O3 and Pt-Ir/TiO2 catalysts at 10 min
time-on-stream.

M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
171
Fig. 5. Representative TEM images and particle size distributions of Pt(1.0)-Ir(1.0)/TiO2 catalyst.
being the former reaction more affected due to its demanding
nature. This behavior observed for TiO2 supported catalysts could
be explained by the previously mentioned migration of TiOx (x < 2)
species during the reduction step [27–30,33–35]. Such phenomena
could create new active sites on Pt–TiO2 interphase and could
also partially destroy the Pt–Ir ensembles known to be responsible
for hydrogenolytic activity [35]. On the other hand, Tauster et al.
[33] reported that the electronic properties of noble metals can be
strongly affected by their interaction with the TiO2 surface, this
interaction being favored by high-temperature treatments. How-
ever, it is also possible to explain the lower metallic activity as a
consequence of simple thermal sinterization of Pt and Ir during
high temperature treatments.
Fig.
5 shows representative TEM images of the Pt(1.0)-
Ir(1.0)/TiO2 catalyst, as well as the corresponding particle size
distributions obtained for the three reduction temperature lev-
els. The observations of the three samples by TEM lead to similar
results. Two population types of metallic particles are observed:
(i) Pt particles well dispersed onto TiO2, with sizes in the order
of 1 nm, (ii) Ir particles on certain TiO2 crystallites. These aggre-
gates of Ir particles may be more numerous and/or the particles
are larger for the sample treated at 500 ◦C. The histograms and the

172
M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
to benzene are comparable for both monometallic catalysts. There
is no production of significant amounts of deep hydrogenolysis
products (C1 C5) for these catalysts. An increase in Ir loading of
bimetallic catalysts causes an increase in light hydrocarbon, 2MP
and 3MP formation, selectivity towards 2MP reaching the highest
value for the catalysts with the same loading of Pt and Ir. Bimetallic
catalysts also have a significant higher selectivity towards C6
isomers as compared to monometallic ones, in the case of the TiO2
support. Both mono and bimetallic Al2O3 supported catalysts have
comparable selectivities to benzene. This selectivity is much lower
in bimetallic catalysts supported on TiO2 than in the corresponding
monometallic ones. The lower rate of aromatics formation on the
TiO2 supported catalysts could be attributed to their lower acidity.
Maire et al. [35] concluded that product distribution in RO reac-
tions depends on the operating reaction mechanisms. According to
a non-selective reaction path that would occur on small metallic
particles, with the same breaking probability for any of the cyclic
bonds, a product ratio nC6:2MP:3MP of 2:2:1 is expected. On the
other hand, if a selective mechanism where the breaking of substi-
tuted endocyclic bonds is forbidden operates, such products ratio
would be 0:2:1. Such mechanism would proceed on metallic aggre-
gates of higher size. An alternative mechanism, partially selective,
with an intermediate product distribution ratio, could also be pos-
sible.
The formation of aromatics on noble metal supported catalysts
was studied by Ponec and co-workers [37]. They were able to deter-
mine the contribution of metal and acid sites of sulfur modified
Pt/Al2O3, Pt-Co/Al2O3 and Pt-Re/Al2O3 catalysts and found that
2MP and 3MP are formed exclusively on catalyst metal sites while
cyclohexane, benzene and nC6 required the presence of both metal
and acid sites. Then, in the Ir supported catalyst the reaction could
be assumed to proceed through a similar bifunctional mechanism.
It was also found that for the monometallic Pt supported catalysts
benzene and nC6 formation are accompanied by the production
of some isomers such as 2MP and 3MP showing a more complex
reaction pattern where hydrogenolysis has a more important role.
From the selectivity values shown in Table 3, it can be con-
cluded that all bimetallic catalysts included in this study exhibit
an intermediate behavior regarding reaction mechanisms as the
nC6/3MP ratio lies between 0 and 2. However, the selective mecha-
nism becomes more important as the Ir loading increases as long as
the nC6/3MP ratio decreases as the Ir content increases. The higher
C1 C5 selectivities found for bimetallic catalysts can be attributed
to the higher hydrogenolytic activity of Pt-Ir ensembles. For TiO2
supported catalysts, Pt(1.0)-Ir(1.0)/TiO2 has the highest RO selec-
tivity (nC6 + 2MP + 3MP). On the other hand, the monometallic Pt
on alumina catalyst has the higher RO selectivity for that series.
Decalin conversion and selectivity towards RO products found
for all catalysts are presented in Table 4. The decalin reaction
products were classified considering the criterion used by Santiku-
naporn et al. [2] and Chandra Mouli et al. [38]. The products of
reaction are lumped according to:
Fig. 6. MCP conversion as function of time-on-stream for the Pt-Ir/Al2O3 and Pt-
Ir/TiO2 catalysts.
average particle size (Table 2) show that increasing the tempera-
ture treatment from 300 ◦C to 400 ◦C produces big changes in the
particles size. However, a higher increase in the reduction tempera-
ture does not modify the average particle size. The particle size was
also calculated from the H/Pt ratio using the equation proposed by
White [36] the values are reported in Table 2. The sample treated at
500 ◦C shows big differences between the particle sizes calculated
from TEM and H/Pt ratio due to the migration of TiOx (x < 2) parti-
cles. Considering the results reported on Table 2 and Fig. 5, it can
be concluded that the migration of TiOx (x < 2) particles occurs at
high temperature while at lower temperatures the metallic phase
is only sinterized.
The value of the H/M ratio as obtained by hydrogen chemisorp-
tion of Pt(1.0)-Ir(1.0)/Al2O3 catalyst (catalyst with equal metal
charge than that reported on Table 2) was 80%. This value yields
a particle diameter of 1.0 nm.
The results obtained for the MCP ring opening reaction are
shown in Fig. 6. It can be observed that reactant conversion
increases as Ir loading in the bimetallic systems increases and
the most active catalysts suffer a significant time on stream deac-
tivation by coking. Monometallic catalysts for both series have
comparable activity levels. TiO2 supported bimetallic catalysts are
more active than those supported on Al2O3. This somewhat unex-
pected behavior could be due to the lower reduction temperature
level used for TiO2 bimetallic catalysts which would lead to a lower
migration of TiOx (x < 2) species and a lower sinterization of the
metallic phase.
Values of selectivities towards reaction products taken at 15 min
of time on stream (i.e. with very low amount of coke deposited
on catalyst surface) are shown in Table 3. It can be observed that
monometallic Ir catalysts have a good selectivity towards nC6.
Monometallic Pt catalysts have a lower selectivity towards nC6 and
a higher selectivity to hexane isomers (2MP and 3MP). Selectivities
(i) Cracking products (C1 C9 products): 2-methyl butane,
hexane, 2,3-dimethyl pentane, 3-methylpentane, 5-methyl, 1-
hexene, methylcyclopentane, propylcyclopentane, 2-methyl-
propylcyclopentane, 1,1-dimethylcyclopentane, cyclohexane,
methylcyclohexa-ne, propylcyclohexane, cis 1-ethyl-2-
methylcyclohexane, trans 1-ethyl-4-methylcyclo-hexane,
1-ethyl-3-methylcyclohexane, 1,1,4-trimethylcyclohexane.
(ii) Ring opening (RO) C10 products: alkylcyclohexanes, alkyl-
cyclopentanes cyclohexenes or benzenes (for example:
1-methyl-2-propylcyclohexane,
diethylcyclo-hexane,
cis
and trans 1,1,3,5-tetramethylcyclohexane, 2-methylpropy-
lbenzene).

M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
173
Table 3
Selectivity to 2MP, 3MP, nC6, C1 C5 and benzene (Bz) of Pt-Ir(y)/Al2O3 and Pt-Ir(y)/TiO2 catalysts at 15 min time-on-stream for the SRO of MCP.
Catalyst
Selectivity (%)
2MP
Molar ratio
n-C6/3MP
3MP
nC6
C1 C5
Bz
RO
2MP/3MP
Pt(1.0)/Al2O3
18
22
35
31
3
14
19
20
21
2
40
24
16
13
62
0
1
5
8
1
28
34
24
27
32
72
65
71
65
67
2.98
1.28
0.74
0.63
39.87
1.31
1.16
1.73
1.43
2.13
Pt(1.0)-Ir(0.3)/Al2O3
Pt(1.0)-Ir(1.0)/Al2O3
Pt(1.0)-Ir(2.0)/Al2O3
Ir(0.3)/Al2O3
Pt(1.0)/TiO2
16
48
65
54
0
7
25
25
30
0
41
13
5
6
75
0
3
3
8
0
36
12
2
2
25
64
86
95
90
75
6.34
0.53
0.18
0.20
–
2.51
1.95
2.60
1.81
–
Pt(1.0)-Ir(0.3)/TiO2
Pt(1.0)-Ir(1.0)/TiO2
Pt(1.0)-Ir(2.0)/TiO2
Ir(0.3)/TiO2
Table 4
Decalin conversion (X%) and RO selectivity after 6 h of reaction for Pt-Ir(y)/Al2O3 and Pt-Ir(y)/TiO2 catalysts.
Catalyst
300 ◦
C
325 ◦
C
350 ◦
C
X (%)
RO (%)
X (%)
RO (%)
X (%)
RO (%)
Pt(1.0)/Al2O3
3.3
3.9
7.6
7.1
2.7
7.2
22.5
38.2
47.4
21.8
4.3
4.4
9.5
10.1
7.2
12.1
60.0
65.9
40.8
23.5
11.3
13.4
26.9
28.5
15.1
10.3
21.0
22.8
28.4
8.0
Pt(1.0)-Ir(0.3)/Al2O3
Pt(1.0)-Ir(1.0)/Al2O3
Pt(1.0)-Ir(2.0)/Al2O3
Ir(0.3)/Al2O3
Pt(1.0)/TiO2
0.7
1.9
2.2
6.5
2.4
39.7
37.5
48.1
41.3
0.1
6.2
2.1
5.4
10.2
7.9
7.7
24.0
45.1
30.6
0.3
10.8
15.1
21.5
24.6
10.0
4.5
10.6
9.43
23.8
3.68
Pt(1.0)-Ir(0.3)/TiO2
Pt(1.0)-Ir(1.0)/TiO2
Pt(1.0)-Ir(2.0)/TiO2
Ir(0.3)/TiO2
(iii) Ring contraction (RC) products: 2,2,3-trimethyl bicy-
clo[2.2.1] heptane, 2,6,6-trimethyl bicyclo[3.1.1]heptane,
1,1-bicyclopentyl, spiro[4.5]decane, 3,7,7-trimethyl bicy-
clo[4.1.0]heptane.
(iv) Other products: 1-methylindan, cis and trans decahy-
dronaphthalene, 1,2-dihydronaphthalene, 1,2,3,4-tetrahydro-
naphthalene and include others heavy dehydrogenation prod-
ucts.
on metal loading, but opposite effects on alumina (decrease) and
titania (increase) were observed as the metal load increased.
Test reactions for cyclohexane dehydrogenation and cyclopen-
tane hydrogenolysis showed that alumina supported catalysts were
more active. This behavior can be due to the partial blockage of
metallic sites by migrated TiOx (x < 2) species during reduction steps
and by the sinterization of metallic phase.
The RO reaction of methylcyclopentane was found to occur
through an intermediate mechanism, partially selective with an
increase of activity as the Ir loading increased. The selective mech-
anism was favored by higher total metal loadings, possibly due to
an increase in the size of metallic aggregates. Titania supported cat-
alysts exhibited higher RO selectivities. The activity for decalin RO
increased both with metal loading and reaction temperature level.
For both bimetallic catalyst series decalin conversion increases
as the Ir loading and temperature level increase as found for the
MCP test reactions. Alumina supported catalysts are more active
than the TiO2 supported ones possibly due to the fore mentioned
partial metal blocking by TiOx (x < 2) species or sintering of the
metallic phase. The combined effect of higher metal activity and
higher acid amount in the case of the alumina supported catalysts
leads to a higher activity for decalin ring opening in comparison
to the titania supported catalysts. The best decalin conversion are
obtained with Pt(1.0)-Ir(2.0) catalysts while the best RO selectiv-
ity were found for Pt(1.0)-Ir(1.0) catalysts in both support series.
Regarding temperature level, the best RO selectivities are obtained
at different temperatures. At higher temperatures, dehydrogena-
tion reactions occur on this type of catalysts become important,
the presence of naphthalene being an indication of this undesir-
able situation [39]. Moreover, cracking reactions are increased as
the temperature is increased while ring contraction reactions are
decreased.
References
[1] Kirk-Othmer, Encyclopedia of Chemical Technology, vol. 17, 3rd ed., John Wiley
& Sons, Inc., New York, 1982, pp. 119–143.
[2] M. Santikunaporn, J.E. Herrera, S. Jongpatiwut, D.E. Resasco, W.E. Alvarez, E.L.
Sughrue, J. Catal. 228 (2004) 100–113.
[3] G. Onyestyák, G. Pál-Borbély, H.K. Beyer, Appl. Catal. A 229 (2002) 65–74.
[4] G.B. McVicker, M. Daage, C.W. Touvelle, M.S. Hudson, D.P. Klein, W.C. Baird Jr.,
B.R. Cook, J.G. Chen, S. Hantzer, D.E.W. Vaughan, E.S. Ellis, O.C. Feeley, J. Catal.
210 (2002) 137–148.
[5] A. Corma, F. Mocholi, A.V. Orchillés, G.S. Koermer, R.J. Madon, Appl. Catal. 67
(1991) 307–324.
[6] J. Abbot, J. Catal. 123 (1990) 383–395.
[7] D. Kubicka, N. Kumar, P. Maki-Arvela, M. Tiitta, V. Niemi, H. Karhu, T. Salm, D.Y.
Murzin, J. Catal. 222 (2004) 65–79.
[8] D. Kubicka, N. Kumar, P. Maki-Arvela, M. Tiitta, V. Niemi, H. Karhu, T. Salm, D.Y.
Murzin, J. Catal. 227 (2004) 313–327.
[9] F.G. Gault, Adv. Catal. 30 (1981) 1–95.
4. Conclusions
[10] A. Djeddi, I. Fechete, F. Garin, Appl. Catal. A 413–414 (2012) 340–349.
[11] L.B. Galperin, J.C. Bricker, J.R. Holmgren, Appl. Catal. A 239 (2003) 297–304.
[12] W.C. Baird Jr., J.G. Chen, G.B. McVicker, US Patent 6,623,625 (2003).
[13] A. Djeddi, I. Fechete, F. Garin, Top. Catal. 55 (2012) 700–709.
[14] P. Samoila, M. Boutzeloit, C. Especel, F. Epron, P. Marécot, Appl. Catal. A 369
(2009) 104–112.
It was found that Al2O3 or TiO2 supported Pt-Ir catalysts pre-
pared by the coimpregnation technique have a strong Pt-Ir interac-
tion and only a weak metal-support interaction, as evidenced by the
low reduction temperature levels observed for the bimetallic cat-
alysts despite the type of support used. Catalyst acidities depend
[15] P. Samoila, M. Boutzeloit, C. Especel, F. Epron, P. Marécot, J. Catal. 276 (2010)
237–248.

174
M.A. Vicerich et al. / Applied Catalysis A: General 453 (2013) 167–174
[16] K. Shimizu, T. Sunagawa, C.R. Vera, K. Ukegawa, Appl. Catal. A 206 (2001) 79–86.
[17] J.A. Kowalski, C.N. Elia, J.G. Santiesteban, M. Acharya, M.A. Daage, A.B. Dandekar,
D. Sinclair, M. Kalyanaraman, L. Zhang, European Patent 2242721 (2010).
[18] S.A. D’Ippolito, V.M. Benitez, P. Reyes, M.C. Rangel, C.L. Pieck, Catal. Today 172
(2011) 177–182.
[27] J.B.E. Anderson, R. Burch, J.A. Cairns, J. Catal. 107 (1987) 351–363.
[28] K. Hayek, R. Kramer, Z. Paál, Appl. Catal. A 162 (1997) 1–15.
[29] F. Pesty, H.P. Steinrück, T.E. Madey, Surf. Sci. 339 (1995) 83–95.
[30] Z.M. Liu, M.A. Vannice, Surf. Sci. 350 (1996) 45–59.
[31] J.M. Parera, N.S. Fígoli, in: G.J. Antos, A.M. Aitani, J.M. Parera (Eds.), Catalytic
Naphtha Reforming: Science and Technology, Marcel Dekker Inc., New York,
1995, pp. 45–78.
[19] V.A. Mazzieri, J.M. Grau, J.C. Yori, C.R. Vera, C.L. Pieck, Appl. Catal. A 354 (2009)
161–168.
[20] L.S. Carvalho, C.L. Pieck, M.C. Rangel, N.S. Fígoli, J.M. Grau, P. Reyes, J.M. Parera,
Appl. Catal. A 269 (2004) 91–103.
[21] C. Carnevillier, F. Epron, P. P.Marecot, Appl. Catal. A 275 (2004) 25–33.
[22] P. Panagiotopoulou, A. Christodoulakis, D.I. Kondarides, S. Boghosian, J. Catal.
240 (2006) 114–125.
[23] V.M. Benitez, M. Boutzeloit, A. Mazzieri, C. Especel, F. Epron, C.R. Vera, P. Maré-
cot, C.L. Pieck, Appl. Catal. A 319 (2007) 210–217.
[24] L.I. Ali, A.A. Ali, S.M. Aboul-Fotouh, A.K. Aboul-Gheit, Appl. Catal. A 177 (1999)
99–110.
[32] B.H. Davis, Catal. Today 53 (1999) 443–516.
[33] S.J. Tauster, S.C. Fung, R.L. Garten, J. Am. Chem. Soc. 100 (1978) 170–175.
[34] D. Poondi, M.A. Vannice, J. Mol. Catal. 124 (1997) 79–89.
[35] G. Maire, G. Plouidy, J.C. Prudhomme, F.G. Gault, J. Catal. 4 (1965) 556–569.
[36] M.G. White, Heterogeneous Catalysts, Prentice Hall, England Clifts, NJ, 1991, p.
88.
[37] M.J. Dees, M.H.B. Bol, V. Ponec, Appl. Catal. 64 (1990) 279–295.
[38] K. Chandra Mouli, V. Sundaramurthy, A.K. Dalai, J. Mol. Catal. A: Chem. 304
(2009) 77–84.
[25] S. Capula Colindres, J.R. Vargas García, J.A. Toledo Antonio, C. Angeles Chavez,
J. Alloys Compd. 483 (2009) 406–409.
[39] K. Chandra Mouli, V. Sundaramurthy, A.K. Dalai, Z. Ring, Appl. Catal. A 321
(2007) 17–26.
[26] G.B. McVicker, P.J. Collins, J.J. Ziemiak, J. Catal. 74 (1982) 156–172.