Simultaneous naphthalene and thiophene hydrogenation over Ni(X)-Pt/HMOR catalysts

Article abstract of DOI:10.1016/j.cattod.2014.08.032

A series of Ni(X)-Pt/HMOR catalysts (X = 0, 0.5, 1.0, 1.5 and 2.0 wt% of Ni and a fixed quantity of Pt = 1 wt%) were prepared and tested in the hydrogenation (HYD) of naphthalene with and without thiophene and also in the hydroconversion of thiophene. The catalysts were characterized using N₂ physisorption, X-ray diffraction (XRD), temperature-programmed reduction (TPR), transmission electron microscopy (TEM) and TEM-EDS elemental analysis. The TPR and TEM results indicate that Pt is mainly located inside the pore system of the mordenite, whereas Ni is located on both the external surface and inside the pore system of the mordenite zeolite. Naphthalene hydrogenation exhibited a high conversion for all the catalysts. In the presence of a sulfur compound and for naphthalene hydrogenation, all the tested catalysts present a fast deactivation. Ni/HMOR catalyst was less active and presents complete deactivation. However, the Ni(0.5)-Pt/HMOR and Pt/HMOR catalysts are able to convert thiophene and produce mainly tetrahydrothiophene. This hydrogenation activity was stable, at least for 12 h, showing that the catalytic system can preserve some of the hydrogenating capability. The values of thiophene hydroconversion and the selectivity to tetrahydrothiophene are much higher than the reported for similar systems. The hydrogenation activity is most likely due to the hydrogen spillover.

Full text of DOI:10.1016/j.cattod.2014.08.032

Catalysis Today 250 (2015) 12–20
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Simultaneous naphthalene and thiophene hydrogenation over
Ni(X)–Pt/HMOR catalysts
J.G. Téllez-Romeroa, R. Cuevas-Garcíaa,∗, J. Ramíreza, P. Castillo-Villalóna,
R. Contreras-Bárbarab, M.C. Salcedo-Lunac, R.I. Puente-Leec
a UNICAT, Depto. Ingeniería Química, Facultad de Química, Universidad Nacional Autónoma de México (UNAM), Ciudad Universitaria,
04510 Coyoacán, México, DF, Mexico
b Unidad de Energía Renovable, Centro de Investigación Científica de Yucatán, Mérida, CP 97200 Yucatán, Mexico
c USAI, Facultad de Química, UNAM, Ciudad Universitaria, 04510 Coyoacán, México, DF, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of Ni(X)–Pt/HMOR catalysts (X = 0, 0.5, 1.0, 1.5 and 2.0 wt% of Ni and a fixed quantity of Pt = 1 wt%)
were prepared and tested in the hydrogenation (HYD) of naphthalene with and without thiophene and
also in the hydroconversion of thiophene. The catalysts were characterized using N2 physisorption, X-ray
diffraction (XRD), temperature-programmed reduction (TPR), transmission electron microscopy (TEM)
and TEM-EDS elemental analysis. The TPR and TEM results indicate that Pt is mainly located inside the
pore system of the mordenite, whereas Ni is located on both the external surface and inside the pore
system of the mordenite zeolite. Naphthalene hydrogenation exhibited a high conversion for all the cat-
alysts. In the presence of a sulfur compound and for naphthalene hydrogenation, all the tested catalysts
present a fast deactivation. Ni/HMOR catalyst was less active and presents complete deactivation. How-
ever, the Ni(0.5)–Pt/HMOR and Pt/HMOR catalysts are able to convert thiophene and produce mainly
tetrahydrothiophene. This hydrogenation activity was stable, at least for 12 h, showing that the catalytic
system can preserve some of the hydrogenating capability. The values of thiophene hydroconversion
and the selectivity to tetrahydrothiophene are much higher than the reported for similar systems. The
hydrogenation activity is most likely due to the hydrogen spillover.
Received 15 November 2013
Received in revised form 8 August 2014
Accepted 28 August 2014
Available online 26 September 2014
Keywords:
Thio-tolerance
Thio-resistance
Naphthalene hydrogenation
Ni–Pt/HMOR catalyst
Pt/HMOR catalyst
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
It has been proved that the addition of a second metal to Pt in
the catalysts formulation enhances thio-resistance on noble metal
In recent years, several studies have been carried out to develop
hydrogenation catalysts with resistance to deactivation by sulfur
compounds (thio-resistance) or at least a catalyst able to operate
in the presence of low sulfur contents (thio-tolerance). Improve-
ment of the thio-resistance of noble metal-based catalysts can be
achieved by the formation of different alloys, by changes in the
acidity of the support and/or also by the selection of a support with
a pore system adequate to the reaction. Another related issue is
the production of self-containing sulfur heterocycles (e.g. tetrahy-
drothiophene) compounds that are a key part of the initial step for
the synthesis of physiologically active compounds, dyes, odorants
and oil additives [1].
hydrogenation catalysts as it has been reported for the Pt–Pd sys-
tem compared with a Pt-only catalyst [2]. In general, the second
metal should be a transition metal with similar characteristics to
the base metal; nonetheless, it is desirable for this second metal to
be less expensive, for example, nickel. The system Ni–Pt has been
studied for the reforming reaction (see [3] and references therein).
Nevertheless, only few works have been reported for the hydro-
genation reaction. As an example, Rynkowski et al. [4] reported that
for toluene hydrogenation the bimetallic catalyst Ni–Pt/SiO2–TiO2
was more active than the monometallic one with a similar content
of Pt.
It seems that the used support in the preparation of catalyst
presents at least two different effects: electronic effects associated
to the acidity [5], and effects associated to the porous structure.
For the first effect, it was reported that the use of acidic supports,
such as zeolites [5], ASAs [6] or fluorinated alumina [7], produces
an enhanced catalytic activity in hydrogenation and hydrogenolysis
reactions and provides some thio-resistance to Pt-based catalysts.
∗
Corresponding author. Tel.: +52 56225368; fax: +52 56225366.
E-mail addresses: cuevas@unam.mx, rcuevasg@prodigy.net.mx
(R. Cuevas-García).
http://dx.doi.org/10.1016/j.cattod.2014.08.032
0920-5861/© 2014 Elsevier B.V. All rights reserved.

J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
13
The explanation given for the thio-resistance was the creation of
positively charged metal clusters in contact with support acidic
sites; Dalla Betta [8] and Sachtler [5] called this phenomenon “metal
electron deficiency.” For the second effect, it has been proposed
[9] that a support with a special pore system arrangement, com-
posed of small and large interconnected pores, could improve the
thio-resistance. The idea was that some metallic clusters were
deposited in the small pores, so that sulfur organic compounds
cannot access to these clusters which remain in the metallic state.
Then the clusters generate hydrogen spillover which helps to keep
all the metallic sites active. It has been established that the system
Pd supported on mordenite (HMOR) (Pd/HMOR) presents higher
thio-resistance than a Pd/Y zeolite [10]. Other study conducted by
Contreras et al. [11] concluded that the Pt/HMOR catalyst was more
active than Pt/HFAU or Pt/HMFI for the hydrogenation of thiophene.
It seems that the MOR is a suitable support to confer to Pt clusters
the thio-resistance property.
In this work, we study the effects of the addition of an inexpen-
sive metal, nickel, to a Pt/HMOR catalyst with the aim of analyzing
the thio-resistance and the catalytic performance of the system
Ni(X)–Pt/HMOR catalysts (X = 0, 0.5, 1.0, 1.5 and 2.0 wt% of Ni and
a fixed quantity of Pt = 1 wt%). To do so, the catalysts were tested
in naphthalene hydrogenation in the presence and absence of a
sulfur compound (thiophene). In addition, the hydroconversion of
thiophene was done and the products distribution was carefully
analyzed to evaluate the extent of the hydrogenation reaction route
to tetrahydrothiophene.
Micromeritics TriStar 3000 apparatus. Before the measurements,
the samples were pretreated overnight at 543 K under vacuum
(350 Pa). Specific area was obtained following the BET procedure.
The BJH method was used to determine the distribution of meso-
pores from both the adsorption and desorption branches and the
t-plot method was used to estimate the microporous volume.
2.2.2. X-ray diffraction (XRD)
X-ray powder diffraction patterns of HMOR, Ni/HMOR, Pt/HMOR
and Ni(X)–Pt/HMOR catalysts on the oxide state were collected on
a SIEMENS-D5000 diffractometer equipped with Cu K␣ radiation
˚
(ꢀ = 1.54 A). The X-ray diffraction data were obtained in the range
of 2ꢁ between 5◦ and 80◦ with a 0.020◦ step size.
2.2.3. Temperature-programmed reduction (TPR)
The HMOR, Ni/HMOR, Pt/HMOR and Ni(X)–Pt/HMOR samples
in the oxide state were studied by TPR. The TPR apparatus was
a characterization ISRI RIG 100 system equipped with a thermal
conductivity detector (TCD). For the analysis, a sample was heated
from room temperature to 1073 K with a heating rate of 10 K/min
in H2 flow. Before the TPR measurements, the sample was placed
in a quartz tubular reactor and then it was heated in N2 flow
(30 mL/min) from room temperature to 673 K for 2 h (heating rate
of 10 K/min). The sample was cooled to room temperature, while
the detector was stabilized under 30 mL/min of an argon/hydrogen
gas mixture (Praxair, 30/70 Ar/H2 volumetric ratio, >99% purity)
flow.
2. Experimental
2.1. Catalyst preparation
2.2.4. Transmission electron microscopy (TEM)
The TEM studies were carried out using a JEOL model JEM-2010
apparatus with an Oxford X-ray energy dispersion spectrometry
(EDS) device. Reduced samples and spent samples were employed
for this analysis.
Commercial NH4–mordenite, CVB 21A, was supplied by Zeolyst
International with 500 m2/g of BET surface area, a SiO2/Al2O3 ratio
of 20 and 0.08 wt% of Na2O content. The zeolite was changed to
its acid form (HMOR) by calcination at 823 K for 6 h in a static
atmosphere. The preparation of the Pt catalysts was made using
the ion exchange (IE) technique with the following procedure: the
HMOR zeolite was contacted with 100 mL of an aqueous solution
of [Pt(NH3)4](NO3)2 (R.A. Sigma-Aldrich) containing the required
amount of Pt to obtain a nominal content of 1.0 wt% Pt. The pH of
the suspension was held at 8.0 by adding a 1 N NH4OH solution. The
suspension was stirred at room temperature for 192 h. Solids were
separated from the liquid by centrifugation. The obtained material
was dried at room temperature for 48 h and then calcined in oxygen
flow at 723 K for 2 h. One sample, without any further modification,
was separated and called Pt/HMOR. The remaining of the Pt/HMOR
preparation was used to synthesize Ni(X)–Pt/HMOR catalysts. Ni
was incorporated by incipient wetness impregnation technique
(IWI). A known volume of an aqueous solution with the required
amount of Ni(NO3)2·6H2O (A.R. J.T. Baker) was incorporated over
different Pt/HMOR samples in order to provide the required Ni
nominal content. After the impregnation, the samples were dried at
room temperature for 48 h and calcined at 723 K for 2 h in a static
atmosphere. The prepared catalysts was labeled Ni(X)–Pt/HMOR
where X represents the weight % of Ni load (0.5, 1.0, 1.5 and 2.0),
the quantity of Pt was fixed an equal to wt. 2% nominal content.
A catalyst with 2.0 wt% of nickel, Ni/HMOR, was prepared by IWI
using the same preparation procedure.
2.3. Catalyst activity measurements
2.3.1. General procedure
The reaction was carried out in a high-pressure fixed-bed down-
flow reactor. Typically, 0.15 g of catalyst were packed into the
reactor, and then reduced at 673 K under hydrogen (Praxair 99.99%)
atmosphere for 4 h with a heating rate of 10 K/min. Ni/HMOR,
Pt/HMOR and Ni(0.5)–Pt/HMOR catalysts were subjected to the fol-
lowing reaction schedule: (1) naphthalene hydrogenation reaction
(HYD) took place at 493 K and 5.52 MPa of hydrogen. The feed was
a solution of 5.2 wt% of naphthalene (A.R. J.T. Baker) dissolved in
n-decane (A.R. Sigma-Aldrich). (2) After a time of stream (TOS) of
12 h, the feed was changed to another solution containing 500 ppm
of sulfur as thiophene (Fluka) + 5.2 wt% of naphthalene dissolved in
n-decane. (3) Finally, after TOS of 24 h, the catalysts were reduced
again and the HYD of naphthalene (without thiophene) was per-
formed once more.
2.3.2. Analysis of reaction products
The reaction samples were collected every hour and analyzed by
gas chromatography. A VARIAN CP-3800 chromatograph equipped
with two detectors: flame ionization detector (FID), for the quan-
tification of organic compounds, and pulsed flame photometric
detector (PFPD), for identifying the sulfur compounds. The identi-
fication of the reaction products was made by GC–MS (HP 61800B
GCD System) equipped with an electron capture detector (ECD) and
using a PONA column of 50 m.
2.2. Catalyst characterization
2.2.1. N2 adsorption–desorption isotherms
The pore-size distribution and specific area measurements were
made by nitrogen adsorption–desorption data acquired with a

14
J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
3. Results and discussion
3.1. Characterization techniques
3.1.1. Textural properties
The N2 adsorption–desorption isotherms (see supplemental
information) of all catalysts display similar hysteresis loops indi-
cating the presence of the same porous structure. The isotherms
can be described as a combination of types I and IV, as defined by
IUPAC [12]. Table 1 shows the textural properties of the catalysts,
a slight decrease in the specific area was observed, however, these
changes are within the experimental error. The BJH method was
applied to the adsorption and desorption hysteresis loop branches
(see supplemental information). For both cases, pore size distribu-
tion remained practically the same with the addition of the metals.
The constancy of the surface area and pore size distributions indi-
cates that the addition of Ni and Pt to the zeolite not give rise to
blocking of the porous system.
3.1.2. X-ray diffraction (XRD)
Fig. 1. TPR profiles for HMOR, Pt/HMOR, Ni/HMOR and Ni(X)–Pt/HMOR catalysts.
The XRD patterns (not shown, see additional information) of all
catalysts in the oxide state display sharp peaks in the 2ꢁ values of
6◦ ≤ 2ꢁ ≤ 30◦ associated to the HMOR structure [13,14]. None of the
peaks associated to PtO, Pt0 and Ni2O3 (2003 JCPDS cards 85-0714,
89-7382, 14-0481, respectively) were observed. In the case of PtO,
the position of some of associated HMOR peaks present the same
positions of the PtO peaks, this problem obstruct the compound
identification. Another possible cause of this behavior is the low
metal content in the samples. Rynkowski et al. [4] reported that
they did not detect Ni or Pt species on SiO2–TiO2 with a content
of 2.5 wt%. In addition, it is reported that Pt presents a detection
limit at 3.0 wt% [15]. In the case of NiO (see supplemental informa-
tion), according to the 2003 JCPDS cards 89-7131 (NiO monoclinic)
and 89-7390 (NiO rhombohedral), the three more intense peaks are
located at 2ꢁ = 43.4◦, 37.25◦ and 62.9◦ listed from the high inten-
sity to low intensity order. In this case, it can be observed that
two (37.25◦ and 62.9◦) peaks appear in areas associated with the
peaks of HMOR. The third peak is located at 2ꢁ = 43.4 (most intense
peak). At this position, the appearance of a peak is observed, slightly
above the noise level. It is possible to say that there is an incipient
formation of NiO.
Pt oxides in the zeolite channels are reduced at slightly higher
temperatures (temperature of reduction, Tr > 350 K). Therefore, the
observed shoulder at around 400 K indicates the presence of some
Pt species inside the pore system. This suggests the reduction of
some Pt species in high interaction with the support; most likely,
PtO in the zeolite. Lerner [17] signaled that the reductions of Pt
oxide species are complete at 573 K.
The TPR profile of monometallic Ni/HMOR shows a non-
symmetrical peak with a maximum at 713 K. This peak could
be assigned to the reduction of NiO to Ni0 [20]. The asymmetry
at higher temperatures could be ascribed to some Ni species in
stronger interaction with the support.
In the case of the Ni(X)–Pt/HMOR samples, in general, the
addition of the general characteristics of Pt/HMOR and Ni/HMOR
catalyst thermograms are present; but the temperature of reduc-
tion maximum changes to lower values. Besides, the peak
associated to nickel oxide is much wider. The shift of the nickel
oxide temperature of reduction (Ni/HMOR, 713 K) toward lower
temperatures in Pt presence is not unexpected. In the literature, it
is known that the presence of a second noble metal on a Ni cata-
lyst enhances the reducibility of the Ni species. This behavior was
reported in the case of Ni–Pt catalysts on different supports [20,21],
for the system Ni–Pd [22], and in the case of Ni/Al2O3 catalysts
with different metal such as Pd, Pt, Au, Ir, Rh and Ru [23]. Sev-
eral authors attributed this reduction effect to hydrogen spillover
from the previously reduced Pt to the nickel oxide via the support
[21,23]. The further shift from 687 K (Ni(0.5)–Pt/HMOR) to 625 K
(Ni(2.0)–Pt/HMOR) is a result of the increment of the oxide parti-
cle size caused by the increase in the nickel content. It has been
reported that the increase in particle size can be associated with a
decrease in the interaction between the metallic clusters and the
support [24]. In the bimetallic catalysts, the asymmetry of the NiO
reduction peak subsists. Li et al. [3] reported similar results for the
Ni(0.9)–Pt(0.2)/␥-Al2O3 catalyst. However, the asymmetry of the
reduction peak increases with the content of nickel. The asymme-
try suggests that several other species of Ni with high interaction
with the support are present; moreover, the higher reduction tem-
perature of the emerging peak (ca. 700 K) is probably related to NiO
particles inside the zeolite cavities.
3.1.3. Temperature-programmed reduction (TPR)
For the Pt/HMOR system, Yuvaraj et al. [16] reported the reduc-
tion of the Pt oxide species at low temperatures, from 77 to 273 K.
Our TPR apparatus cannot work at temperatures below 273 K and,
therefore, we were not able to detect the low-temperature reduc-
tion peaks of Pt species. In addition, one striking feature in the
thermograms is the presence of a large peak with a maximum
located around 330 K. This peak is present in all the samples even
in the support (HMOR), as can be seen at the bottom of Fig. 1. The
presence of this peak has already been reported and was associated
with the diffusion of hydrogen into the zeolite and the elution of
the inert gas from the pore system into the H2/Ar reduction mix-
ture [17,18]. In our case, because the HMOR has a dual pore size
system, it is possible that a second peak, associated to the gas elu-
tion from the side pockets of the mordenite, could appear. Indeed,
a second small and broad peak was observed from 400 to 500 K on
the pure zeolite thermogram. No further attention was paid to this
two “apparent” reduction peaks.
The thermogram of the Pt/HMOR catalysts (Fig. 1) shows the
“apparent” reduction peak at 333 K, but with a shoulder at 390 K. On
similar Pt/HMOR catalysts, the group of Sachtler [17,19] reported
a reduction peak at 330 K due to the PtO2 reduction. In addition,
according to Yuvaraj [16], Pt oxide located in the external surface of
zeolites is reduced at temperatures, ca. 350 K, while well-dispersed
3.1.4. Transmission electron microscopy (TEM)
TEM micrographs of the Pt/HMOR, Ni(X)–Pt/HMOR and
Ni/HMOR catalysts in the reduced state are shown in Fig. 2. The
particle size distribution for each catalyst is shown in Fig. 3. The

J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
15
Table 1
Textural properties of catalysts obtained from N2 adsorption–desorption isotherms.
Catalyst
Micropore surface
area (m2/g)
External surface
area (m2/g)
Total areaa (m2/g)
Micropore volumeb
(cm3/g)
HMOR
Pt/HMOR
380
360
371
365
366
359
362
61
63
61
59
59
58
61
441
423
432
424
425
417
423
0.177
0.167
0.172
0.169
0.170
0.167
0.168
Ni(0.5)–Pt/HMOR
Ni(1.0)–Pt/HMOR
Ni(1.5)–Pt/HMOR
Ni(2.0)–Pt/HMOR
Ni/HMOR
a
BET method.
t-plot method.
b
Fig. 2. TEM micrographs of Pt/HMOR, Ni(X)–Pt/HMOR and Ni/HMOR catalysts.

16
J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
Fig. 5. Naphthalene distribution products for Ni/HMOR catalyst. Operating condi-
tions: 493 K and hydrogen pressure of 5.52 MPa.
Fig. 3. Particle size distribution for the prepared catalysts.
TEM-EDS analysis detected platinum, thus confirming the exist-
ence of a highly dispersed very small (<4 A) Pt clusters. For the
˚
˚
bimetallic Ni(X)–Pt/HMOR catalysts, the TPR analysis suggested
that the metal–support interaction diminishes with the Ni content.
Eswaramoorthi et al. [26], who worked with the Ni–Pt system sup-
ported on beta zeolite, made similar observations. Therefore, in our
case, the decrease in the metal–support interaction can be related
to an increase in the Ni cluster size caused, in turn, by the increase
in the nickel content.
majority of Pt clusters present a size of about 10 A. Some of the Pt
clusters have even smaller sizes; but they appear in the micrograph
with ill-defined borders; so to avoid a statistical error in the mea-
sured size distribution, all the smaller clusters were reported with
˚
sizes of less than 10 A.
From the TEM observations of the Ni/HMOR sample (see
Figs. 3 and 4), the observed size of nickel clusters are larger than
40 A. In other words, from the XRD results, we concluded that there
˚
is an incipient formation of NiO. These results indicate that the
reduction step causes sintering of Ni. Richarson et al. [25] shown
that the reduction at 573 K causes transformation of crystals of
NiO which size is about 3 nm into metallic Ni crystals with sizes
larger than 20 nm. In addition, from the size of the nickel clusters
compared with the size of HMOR zeolite main cavities (6.5 A × 7 A),
it can be inferred that the major fraction of Ni is located on the
external surface of the zeolite.
3.2. Catalytic test
3.2.1. Naphthalene hydrogenation reaction
The catalysts employed for this reaction were Ni(0.5)–Pt/HMOR,
Pt/HMOR and Ni/HMOR. Test results are presented in Fig. 4 for the
naphthalene HYD in the presence and absence of thiophene. In this
figure, three sections can be observed: (1) naphthalene hydrogena-
tion; (2) naphthalene hydrogenation in thiophene presence and (3)
naphthalene hydrogenation after catalyst regeneration.
In the first section, the initial conversion of naphthalene hydro-
genation is around 94% for all the tested catalysts. For Pt/HMOR,
the conversion of naphthalene remains almost constant during
12 h of time on stream (TOS). In contrast, the conversion of the
Ni(0.5)–Pt/HMOR catalyst starts to decrease after 7 h of TOS, and
for the Ni/HMOR catalyst the deactivation begins earlier.
When the thiophene is added to the feed, naphthalene conver-
sion decreases drastically <2% for all the catalysts. This shows the
deactivation caused by the sulfur compound.
˚
˚
In the case of Ni(X)–Pt/HMOR catalyst (see Fig. 3), the observed
˚
metallic clusters present sizes between 5 and 100 A. A few metal
˚
clusters whose sizes are higher than 100 A can be found outside
the zeolite. By analogy with the Ni/HMOR catalysts, these outside
clusters appear to be associated to Ni. The TEM results show that
the main cluster size increases with the Ni content. In addition,
there are some small particles formed of Pt clusters. To explore
the possibility of the presence of very small Pt clusters, a TEM-
EDS analysis was done for the Ni(0.5)–Pt/HMOR catalyst in a zone
˚
˚
(150 A × 150 A) where no metallic particles were observed. The
After the regeneration of the catalysts, naphthalene conver-
sion is almost recovered to the original values for Pt/HMOR and
Ni(0.5)–Pt/HMOR samples. The Pt/HMOR catalyst presents a stable
operation period up to seven additional hours of TOS. In contrast,
the conversion of Ni(0.5)–Pt/HMOR catalyst decreases to values
around 21.1% with a TOS of 31 h. Unsurprisingly, the Ni/HMOR cat-
alyst was not fully regenerated, getting only a conversion of 57%
followed by a fast deactivation.
To explain the deactivation process, it is important to analyze
the products distribution (see Figs. 5–7). Yield was defined as moles
of a particular product generated per mole of initial key reactant
[27].
The products for naphthalene HYD reaction are tetralin, trans-
decalin, cis-decalin and products coming from acid site reactions
such as hydrogenolysis/ring opening and hydrocracking. These acid
reaction products are reported together in the distribution product
figures as “other compounds.”
The Ni/HMOR catalyst (Fig. 5) presents an initial naphthalene
conversion of 95%; hydrogenation reaction reaches 90%, tetralin
Fig. 4. Hydrogenation of naphthalene with and without thiophene. Operating con-
ditions: 493 K and hydrogen pressure of 5.52 MPa.

J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
17
Fig. 6. Naphthalene distribution products for Pt/HMOR catalyst. Operating condi-
Fig. 8. Thiophene global conversion in the presence of naphthalene with respect to
tions: 493 K and hydrogen pressure of 5.52 MPa.
TOS. Operating conditions: 493 K and hydrogen pressure of 5.52 MPa.
being the main product. The yield for hydrocracking (HYC) reaction
was about 6%.
at 523 K after 0.5 h of TOS. However, less attention was paid to the
reaction behavior of the sulfur compound.
The product distribution of Pt/HMOR is presented in Fig. 6. This
catalyst presents a total (98.8%) of naphthalene conversion. Respect
to the HYD reaction the product yields was 0.8% of tetralin, 30.4%
of cis-decalin, 48.1% of trans-decalin; the HYD reaction was almost
complete to the most hydrogenated products. The acid reaction:
hydrogenolysis/ring opening and HYC reactions produces a yield
of 19.5%. This behavior is observed up to 12 h of TOS.
The addition of nickel to Pt/HMOR catalysts (see Fig. 7)
causes a different distribution of products. The products from the
acid-catalyzed hydrocracking reaction decay with TOS and the
deactivation process become evident after 7 h of TOS. The yield to
tetralin increased rapidly along with a decrement of the decalin
production, suggesting that there is a decrement in the quality
or quantity of the hydrogenation sites. This deactivation can be
explained by coke deposition on the metallic sites. The deactiva-
tion process is more evident in the Ni/HMOR catalyst where the
main product was always tetralin.
After the pure naphthalene HYD reaction period, thiophene was
added to the reaction feed. As expected, a deactivation caused by
sulfur occurs and consequently the values of naphthalene conver-
sion fall to values lower than 2% (Fig. 8). Similar behavior was
reported but for much smaller amounts of sulfur compound. For
example, Simon et al. [28] used 50 ppm of thiophene added to the
feed and reported benzene conversion values less than 1% after
10 min of TOS for Pt/MOR catalysts. Similarly, Seoane et al. [29]
observed catalytic activity decay for the ethylbenzene HYD for a
Ni(8 wt%)/MOR catalyst in the presence of 100 ppm of thiophene
In Fig. 8, the thiophene conversion to tetrahydrothiophene(THT)
is presented. In this reaction, the catalysts Ni(0.5)–Pt/HMOR and
Pt/HMOR catalysts present high (up to 50%) conversions and did not
exhibit deactivation. The fact that the conversion of thiophene was
up to 50% with a yield of 50% to THT for the Ni(0.5)–Pt/HMOR cata-
lyst led us to do a more detailed study of the reaction of thiophene
hydroconversion.
Regarding the thiophene hydrogenation to assess the possi-
ble effect of the coke deposition on the deactivation, a reaction
with naphthalene and thiophene was performed with fresh and
used catalysts. The results for the fresh Ni(0.5)–Pt/HMOR and
Ni/HMOR samples were very similar to those obtained with the
spent catalysts (Fig. 8), although there was a slight increase of
5% in the thiophene conversion when the Pt/HMOR was tested
using the fresh catalyst. This indicated that the deactivation caused
by coke during the naphthalene HYD period is not significant
in this scale of TOS. For the fresh catalysts and for naphthalene
hydrogenation, the conversion decreases in the following order:
Pt/HMOR > Ni(0.5)–Pt/HMOR > Ni/HMOR.
3.2.2. Thiophene hydroconversion
For this reaction, fresh samples of all the synthetized catalysts
were used with a feed consisting of thiophene, 500 ppm of sul-
fur content, dissolved in n-decane. The total conversion obtained
for the reaction is shown in Fig. 9. The Pt/HMOR catalyst shows
a high initial conversion, but presents a fast deactivation process
Fig. 7. Naphthalene distribution products for Ni(0.5)–Pt/HMOR catalyst. Operating
Fig. 9. Thiophene global conversion for the prepared catalysts. Operating condi-
conditions: 493 K and hydrogen pressure of 5.52 MPa.
tions: 493 K and hydrogen pressure of 5.52 MPa.

18
J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
within 3 h of TOS, thereafter the global conversion becomes stable
at 27% for TOS of 12 h. In contrast, the Ni/HMOR catalyst shown the
smallest initial conversion, approximately 26%, as it was expected,
it undergone a great deactivation. The bimetallic Ni(X)–Pt/HMOR
catalysts shown initial conversions between 40 and 60% and did
not present deactivation, in contrast to the monometallic Ni/HMOR
and Pt/HMOR catalysts, after 3 h of TOS from this time period, the
total conversion for the Ni(X)–Pt/HMOR catalysts is higher than the
monometallic Pt/HMOR. In principle, higher metal load must be
reflected in higher initial activity; however, the Ni(2.0)–Pt/HMOR
and Ni(1.5)–Pt/HMOR catalysts did not exhibited better activ-
ity. The monometallic Ni/HMOR catalyst is not thio-resistant and
thio-tolerant, whereas the bimetallic Ni(X)–Pt/HMOR catalysts
are thio-tolerant for the thiophene hydrogenation reaction up to
500 ppm of sulfur, values much higher than the reported previously
for similar reactions using different catalysts.
Fig. 10. Yield of all the catalysts for thiophene hydrogenation at 12 h of TOS. Oper-
The products of the hydroconversion of thiophene were tetrahy-
drothiophene (THT), cis- and trans-butene, isobutene, butane and
H2S. Based on the obtained products and in the known mechanisms
of the HDS reaction, the reaction pathway is proposed in Scheme 1.
For the quantification of the products of the HYD route, it must be
taken into account that THT, being more reactive than thiophene,
is easily hydrodesulfurizated. Actually, the main HYD products are
THT and n-butane. However, n-butane also comes from the direct
HDS reaction and because of this it is difficult to measure the how
much of it comes from the HYD or HDS routes. Therefore, we report
only the THT formation as HYD reaction route yield. In this way, the
reported HYD yields are lower than the real ones.
Since we are using mordenite as support, a concern related to
the effect of the acid sites of the zeolite on the thiophene hydrocon-
version can arise. To evaluate this, a test of n-decane with the pure
mordenite support in the same reaction conditions of naphthalene
HYD was made. The HYC reaction conversion of n-decane was less
than 0.8%. Taking into account the aromatic character of naphtha-
lene and thiophene along with the small value of the n-decane
conversion, the effect of the support acidity per se on the HYC reac-
tions can be neglected. This result is explained also explained due
to the low reaction temperature (493 K) used in this work.
The product distribution on the hydroconversion of thiophene
for all the catalysts at 12 h of TOS is shown in Fig. 10. For all the
evaluated catalysts, the main product was THT. The hydrocracking
and isomerization reactions are mainly associated with the dual
metallic-acidic sites of the catalyst [30]. The direct HDS and that
occurring after hydrogenation of thiophene (HDS)’ must be asso-
ciated to metallic platinum and platinum sulfides; for instance,
Chianelli [31] reported activity in HDS for PtS and an almost neg-
ligible HDS activity for NiS [30,32]. Moreover, Mashkina et al. [33]
published several works on thiophene hydrogenation to THT on
transition metal sulfides; this author pointed out that Ni/Al2O3 cat-
alyst presents low activity for the reaction. Moreover, the same
ating conditions: 493 K and hydrogen pressure of 5.52 MPa.
author indicated that in the case of NiS/Al2O3 for the thiophene
hydroconversion, the selectivity to HYD (THT product) was only
4% [34]. Another factor related with the hydrogenation reaction by
transition metal sulfide (TMS) is the operating temperature. The
TMS work at temperatures around 573 K and our operating tem-
perature was 493 K. It is possible that HDS takes place on PtS but
not on NiS. An important issue is to identify on which phase the
thiophene HYD to THT reaction takes place. It seems, likely, that
the thiophene hydrogenation took place on some kind of metallic
particles.
After the analysis of the clusters size distribution from the TEM
micrographs and the TPR experiments, we conclude that some of
Pt clusters are located in the zeolite channels and some outside
the pore system. Regarding the Ni particles, the results show that
the major portion of them are located in the external surface of
the mordenite and only a small portion inside the HMOR zeolite
cavities. No one of the characterizations that we performed show
the presence of Ni–Pt alloys.
Metallic Ni is easily poisoned by H2S; Bartholomew et al. [35,36]
reported that the deactivation of metallic nickel catalysts takes
place at H2S contents as low as 1–10 ppb. In our case, taking into
account the thiophene hydroconversion during the activity test
there at least 125 ppm H2S which are produced in the reaction.
With this sulfur content, our catalysts should have been deactivated
drastically; however, it was not the case.
According to Marécot et al. [37], the Ni deactivation mecha-
nism is the following: first, H2S will be adsorbed on Ni0-forming
superficial sulfides, quickly, it forms polysulfides and after, it slowly
attacks deeper layers to form bulk sulfides, NiS. The deactivation is
a function that depends strongly on the nature of the sulfur com-
pounds and the nature of the catalysts [38]. For the Ni/HMOR, it
Scheme 1. Reaction pathways for the thiophene hydroconversion.

J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
19
is expected an almost complete deactivation, assuming that Ni is
4. Conclusions
located mainly outside the zeolite pore system and, thus, easily
accessible to sulfur compounds. In the bimetallic Ni(X)–Pt/HMOR
catalysts, for every metallic Ni cluster located in the external zeo-
lite surface, it could be assumed that initially, the same deactivation
mechanism takes place. To test this possibility, a TEM-EDS elemen-
tal analysis was made on the Ni(2.0)–Pt/HMOR catalyst after the
thiophene hydroconversion reaction, in order to detect if S is asso-
ciated to Ni or Pt. Several big particles (>100 nm) located at the
external surface of the zeolite are observed by TEM. In this parti-
cles, the TEM-EDS analysis detected Ni and S. As well, the TEM-EDS
analysis of small particles located inside the zeolite porous network
was analyzed; Pt and sulfur were observed and in the detection
limit Ni.
The bimetallic Ni(X)–Pt/HMOR catalysts show a global conver-
sion and THT production bigger than Pt/HMOR. At short times,
this behavior can be explained because there are more metallic
phases, where Pt and Ni are involved. However, at TOS greater
than 3 h, the activity preserves a higher value than the Pt-only
catalyst. Since the Pt content is the same (1 wt%) therefore some
nickel compound must provide the additional hydrogenation activ-
ity. Moreover, the Ni/HMOR catalyst presents a very low thiophene
conversion. This behavior proved that all the metallic phases
associated to Ni located on the external surface of the support
must be, almost, completely deactivated and corroborates the
reported low activity of NiS [31,32,34]. The conclusion is that Pt
protects from deactivation some portion of Ni in the bimetallic
catalysts, Ni(X)–Pt/HMOR, maintaining a small amount of the Ni
in a somewhat metallic state or, most likely, at least generating
some thiophene hydrogenation active sites in a highly reduced
sulfide.
In this work was tested the possibility of the substitution of some
of the Pt content with Ni in a hydrogenation catalyst and tested its
activity in the presence of a sulfur compound, thiophene.
In the Ni(X)–Pt/HMOR catalysts, apparently, Pt mainly is located
inside the pore system of the mordenite, while the main portion of
Ni appears to be located on the external surface and only a small
part in the pore system. For the naphthalene hydrogenation reac-
tion (without thiophene), the most active catalyst was Pt/HMOR
but it deactivates in the presence of thiophene. For the thiophene
hydrogenation, the Ni(0.5)–Pt/HMOR catalyst is the most active and
less prone to deactivation. In the presence of a sulfur compound,
all the tested catalysts present a fast deactivation for naphtha-
lene hydrogenation. However, the Ni(0.5)–Pt/HMOR and Pt/HMOR
catalysts are able to convert thiophene to produce mainly tetrahy-
drothiophene. This hydrogenation activity was stable, at least for
12 h, showing that the catalytic system can preserve some of its
hydrogenating capability. The hydrogenation activity is most likely
due to the hydrogen spillover originated in the Pt particles con-
tained in the small zeolite pores, which on the other side is not
efficient enough to regenerate completely the hydrogenation activ-
ity of the metallic Ni phase.
Acknowledgement
J.G. Téllez-Romero is very grateful for the financial support
provided by Consejo Nacional de Ciencia y Tecnología (CONACyT,
México) through a fellowship with register number 202853.
Appendix A. Supplementary data
Published literature presents three hypotheses that can explain
the detected activity toward the hydrogenation of a sulfur com-
pound with the bimetallic catalysts: (1) It is known that an increase
in the support acidity leads to an increase in the thio-resistance.
Simon et al. [30] proposed a model where the thio-tolerance of Pt
can be increased by adjusting the Brønsted acid site concentration
of the support to an optimal value. However, in our case all the
samples were prepared on the same support HMOR and although
it is known that ion metal exchange can alter the zeolite acidity, the
content of Pt and the procedure of preparation of our samples is the
same. Taking into account that Ni is located mainly outside the zeo-
lite, not a great change of acidity is expected. (2) The formation of an
alloy was used also as explanation for thio-resistance, as we discuss
earlier we cannot exclude the possibility of the Ni–Pt alloy pres-
ence. However, there are not reports about a favorable influence of
Ni–Pt alloy in a hydrogenation reaction. (3) The spillover from well-
dispersed Pt. Song et al. [9,10] found that Pt or Pd/HMOR catalysts
were more active and exhibits higher sulfur tolerance than the cor-
respondent catalysts on Y zeolite. According to these authors, this
is due to the hydrogen spillover, which allows for the regenera-
tion of the Pt clusters in mordenite, via the interconnected pore
system of the support. In our case, if this happens, the spillover
would come from well-dispersed Pt clusters to coordinately unsat-
urated sites (CUS) on the deactivated nickel or platinum sulfide
phase.
Supplementary material related to this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.cattod.
2014.08.032.
References
[1] A.V. Mashkina, Chem. Heterocycl. Compd. 38 (5) (2002) 503–523.
[2] H. Yasuda, Y. Yoshimura, Catal. Lett. 46 (1–2) (1997) 43–48.
[3] D. Li, Y. Nakagawa, K. Tomishige, Appl. Catal. A: Gen. 408 (1–2) (2011) 1–24.
[4] J. Rynkowski, D. Rajski, I. Szyszka, J.R. Grzechowiak, Catal. Today 90 (2004)
159–166.
[5] W.M.H. Sachtler, A.Y. Stakheev, Catal. Today 12 (2–3) (1992) 283–295.
[6] H.R. Reinhoudt, Fuel Process. Technol. 61 (1) (1999) 133–147.
[7] S. Jongpatiwut, Catal. Lett. 122 (3) (2008) 214–222.
[8] R.A. Dalla Betta, M. Boudart, Proc. 5th Int. Congr. Catalysis, Palm Beach, Miami,
FL, 1972, pp. 1329.
[9] C. Song, Chemtech 29 (1999) 26–30.
[10] C. Song, A.D. Schmitz, Energy Fuels 11 (1997) 656–661.
[11] R. Contreras, R. Cuevas-García, J. Ramírez, L. Ruiz-Azuara, A. Gutiérrez-
Alejandre, I. Puente-Lee, P. Castillo-Villalón, C. Salcedo-Luna, Catal. Today 130
(2–4) (2008) 320–326.
[12] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquérol, T.
Siemieniewska, Pure Appl. Chem. 57 (4) (1985) 603–619.
[13] M.M.J. Treacy, J.B. Higgins, MOR - Mordenite, in: Collection of Simulated XRD
Powder Patterns for Zeolites, 5th ed., Elsevier, Amsterdam, 2007, pp. 284–285.
[14] IZA, Collection of Simulated XRD Powder Patterns for Zeolites, 2001, Available
from: http://www.iza-structure.org/databases/.
[15] I.E. Wachs, L.E. Fitzpatrick, Characterization of Catalytic Materials, University
of Michigan, Butterworth-Heinemann, 1992.
The hydrogen spillover is, apparently, not able to regenerate the
sulfides to a completely metallic phase. The best catalyst was Ni
(0.5)–Pt/HMOR. In principle, the hydrogenation activity must be
related with the metallic content, but our results show that for Ni
contents bigger than 0.5 wt% an activity decrement is observed. The
TEM results show that the crystal size increases with the Ni content;
therefore, the activity decrease on the Ni(X)–Pt/HMOR catalysts
where X > 0.5 wt% can be explained as a decline in the Ni disper-
sion; in addition, if a nickel cluster reaches certain size, it would be
converted to NiS.
[16] S. Yuvaraj, T.H. Chang, C.T. Yeh, J. Catal. 221 (2004) 466–473.
[17] B.A. Lerner, B.T. Carvill, W.M.H. Sachtler, J. Mol. Catal. 77 (1992) 99–108.
[18] S. Yuvaraj, T.H. Chang, C.T. Yeh, J. Phys. Chem. B 107 (2003) 4971–4974.
[19] D.J. Ostgard, L. Kustov, K.R. Poeppelmeier, W.M.H. Sachtler, J. Catal. 133 (1992)
342–357.
[20] C. Raab, J.A. Lercher, J.G. Goodwin Jr., J.Z. Shyu, J. Catal. 122 (1990) 406–414.
[21] R.M. Jao, T.B. Lin, J.R. Chang, J. Catal. 161 (1996) 222–229.
[22] Y. Mukainakano, K. Yoshida, K. Okumura, K. Kunimori, K. Tomishige, Catal.
Today 132 (2008) 101–108.
[23] K. Yoshida, N. Begum, S. Ito, K. Tomishige, Appl. Catal. A: Gen. 358 (2009)
186–192.
[24] J.T. Richarson, R. Scates, M.V. Twigg, Appl. Catal. A: Gen. 246 (2003) 137–150.

20
J.G. Téllez-Romero et al. / Catalysis Today 250 (2015) 12–20
[25] P. Can˜izares, A. de Lucas, F. Dorado, A. Durán, I. Asencio, Appl. Catal. A: Gen. 169
[32] W.J.J. Welters, G. Vorbeck, H.W. Zandbergen, J.W. Dehaan, V.H.J. Debeer, R.A.
Vansanten, J. Catal. 150 (1994) 155–169.
(1998) 137–150.
[26] I. Eswaramoorthi, A. Geetha Bhavani, N. Lingappan, Appl. Catal. A: Gen. 253
[33] A.V. Mashkina, T.S. Sukhareva, React. Kinet. Catal. Lett. 67 (1) (1999)
103–110.
(2003) 469–486.
[27] J.J. Carberry, Chemical and Catalytic Reaction Engineering, McGraw-Hill, United
States of America, 1976.
[28] L.J. Simon, J.G. van Ommen, A. Jentys, J.A. Lercher, Catal. Today 73 (2002)
[34] A.V. Mashkina, L.G. Sakhaltueva, Kinet. Catal. 43 (1) (2002) 107–114.
[35] C.H. Bartholomew, Mechanisms of nickel catalyst poisoning, in: B. Delmon, G.F.
Froment (Eds.), Studies in Surface Science and Catalysis, Elsevier, Amsterdam,
1987, pp. 81–104.
105–112.
[29] X.L. Seoane, A. Arcoya, J.A. González, N. Travieso, Ind. Eng. Chem. Res. 28 (1989)
[36] C.H. Bartholomew, Appl. Catal. A: Gen. 212 (2001) 17–60.
[37] P. Marécot, E. Paraiso, J.M. Dumas, J. Barbier, Appl. Catal. A: Gen. 80 (1992)
79–88.
[38] P. Marécot, E. Paraiso, J.M. Dumas, J. Barbier, Appl. Catal. A: Gen. 80 (1992)
89–97.
260–264.
[30] L.J. Simon, M. Rep, J.G. van Ommen, J.A. Lercher, Appl. Catal. A: Gen. 218 (2001)
161–170.
[31] R.R. Chianelli, M. Daage, M.J. Ledoux, Adv. Catal. 40 (1993) 177–231.