Influence of calcination on metallic dispersion and support interactions for NiRu/TiO2 catalyst in the hydrodeoxygenation of phenol

Article abstract of DOI:10.1016/j.cattod.2018.11.007

Bio-oil upgrading by hydrotreatment has been considered as a renewable route for fuels production and potential substitutes for fossil oil distillates. Traditional hydrodesulfurization catalysts (sulfided CoMo and NiMo) used in petroleum refining have been evaluated in hydrodeoxygenation reactions with high selectivity to aromatics. However, significant deactivation has been commonly observed, due to the low sulfur content in bio-oil. Thus, the development of different active and stable catalytic materials is needed. In this regard, we have previously reported a synergetic effect between Ni and Ru bimetallic materials supported on Al₂O₃, TiO₂ and ZrO₂ in phenol HDO with promising activity and selectivity results. In particular, materials supported TiO₂ on displayed the highest activities. Therefore, the aim of this work was to evaluate the effect of calcination on dispersion and metal-support interactions and their impact on activity and selectivity for NiRu/TiO₂ catalysts in HDO of phenol as a model compound for bio oil. Catalysts were characterized by hydrogen chemisorption, Temperature- programmed reduction, X-ray photoelectron spectroscopy, High resolution electron microscopy. Additionally, hydrogenation of benzene was used as a structure unsensitive reaction in order to obtain more information about metallic dispersion. A highly active NiRu catalyst was obtained when calcination was avoided prior to reduction. The calcination procedure induced the formation of metallic oxides and it provoked less dispersed NiRu species as compared with catalysts without this previous treatment. On the contrary, the formation of metallic species from direct reduction of precursor salts contributed to produce highly dispersed species.

Full text of DOI:10.1016/j.cattod.2018.11.007

Accepted Manuscript
Title: Influence of calcination on metallic dispersion and
support interactions for NiRu/TiO catalyst in the
2
hydrodeoxygenation of phenol
Authors: O.U. Vald e´ s-Mart ´ı nez, C.E. Santolalla-Vargas, V.
Santes, J.A. de los Reyes, B. Pawelec, J.L.G. Fierro
PII:
S0920-5861(18)31231-8
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2018.11.007
CATTOD 11741
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
9 August 2018
29 October 2018
2 November 2018
Please cite this article as: Vald e´ s-Mart ´ı nez OU, Santolalla-Vargas CE, Santes V, de
los Reyes JA, Pawelec B, Fierro JLG, Influence of calcination on metallic dispersion
and support interactions for NiRu/TiO catalyst in the hydrodeoxygenation of phenol,
2
Catalysis Today (2018), https://doi.org/10.1016/j.cattod.2018.11.007
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Influence of calcination on metallic dispersion and support interactions
for NiRu/TiO catalyst in the hydrodeoxygenation of phenol
2
a
b
b
a
O.U. Valdés-Martínez , C. E. Santolalla-Vargas , V. Santes , J. A. de los Reyes *, B.
c
c
Pawelec and J.L.G. Fierro
a
Área de Ingeniería Química. Universidad A. Metropolitana-Iztapalapa, Av. FFCC R.
Atlixco 189 , 09340 Ciudad de México, México
b
Departamento de Biociencias e Ingeniería, Centro Interdisciplinario de Investigaciones y
Estudios sobre Medio Ambiente y Desarrollo (CIIEMAD), Instituto Politécnico Nacional,
C.P. 07340, Ciudad de México, México
c
Instituto de Catálisis y Petroleoquímica, CSIC, C/ Marie Curie 2, Cantoblanco, E-28049,
Madrid, Spain
*
Corresponding author at:
Universidad A. Metropolitana, Área de Ingeniería Química, Av. FFCC R. Atlixco 186,
Vicentina, 09340 Distrito Federal, México.
Tel: +52 5558044600 ext. 1247, Fax: +52 5556122479.
Email address: jarh@xanum.uam.mx

Graphical Abstract
Highlights


The calcination procedure through the formation of metallic oxides induced less
dispersed NiRu species.
Chemisorption, benzene hydrogenation and HRTEM characterization pointed
H
2
out that dispersion improved when catalyst was only dried and not calcined prior to
reduction.

Calcination induced sintering of metal particles leading to a decrease of HDO
activity
Abstract
Bio-oil upgrading by hydrotreatment has been considered as a renewable route for fuels
production and potential substitutes for fossil oil distillates. Traditional
hydrodesulfurization catalysts (sulfided CoMo and NiMo) used in petroleum refining have
been evaluated in hydrodeoxygenation reactions with high selectivity to aromatics.
However, significant deactivation has been commonly observed, due to the low sulfur
content in bio-oil. Thus, the development of different active and stable catalytic materials is
needed. In this regard, we have previously reported a synergetic effect between Ni and Ru
bimetallic materials supported on Al
2
O
3
, TiO
2
and ZrO
2
in phenol HDO with promising
on displayed the
activity and selectivity results. In particular, materials supported TiO
2

highest activities. Therefore, the aim of this work was to evaluate the effect of calcination
on dispersion and metal-support interactions and their impact on activity and selectivity for
NiRu/TiO catalysts in HDO of phenol as a model compound for bio oil. Catalysts were
2
characterized by hydrogen chemisorption, Temperature- programmed reduction, X-ray
photoelectron spectroscopy, High resolution electron microscopy. Additionally,
hydrogenation of benzene was used as a structure unsensitive reaction in order to obtain
more information about metallic dispersion. A highly active NiRu catalyst was obtained
when calcination was avoided prior to reduction. The calcination procedure induced the
formation of metallic oxides and it provoked less dispersed NiRu species as compared with
catalysts without this previous treatment. On the contrary, the formation of metallic species
from direct reduction of precursor salts contributed to produce highly dispersed species.
Keywords
Nickel-Ruthenium; Hydrodeoxygenation; Phenol; Calcination; Dispersion
Introduction
A continuous effort to mitigate the strong dependence of fossil fuels has motivated the research
on new and alternative fuels derived from renewable resources. Lignocellulosic biomass can
be transformed by fast pyrolysis into a product called bio-oil. This product has a high
oxygen content that leads to chemical instability, immiscibility with fossil fuels and
corrosion, preventing the direct application of this liquid in the transportation sector [1]. In
this sense, upgrading bio-oils to liquid hydrocarbons is the clue in the development of a
feasible route to obtain sustainable fuels from lignocellulose biomass [2].
Hydrodeoxygenation (HDO) is one of the main approaches to achieve this goal [3][4]. One
relevant challenge in this process is to design an optimal catalyst, which should be active,
stable and selective, allowing C-O bond scission with minimal H consumption and carbon
2
loss. The design of an adequate catalyst for HDO of bio-oil or model compounds requires
the understanding of reaction routes. For example, two main parallel pathways have been
proposed for phenol HDO[5]: (i) the direct deoxygenation (DDO), involves the CAR–O
cleavage by hydrogenolysis to produce benzene; and (ii) the prehydrogenation of the
aromatic ring (HYD) that leads to formation of cyclohexanol as an intermediate [6],

followed by dehydration of cyclohexanol to yield cyclohexene that is then hydrogenated to
cyclohexane. Cyclohexene and methylcyclopentene could be then hydrogenated to the
corresponding saturated compounds.
Due to the similarity of HDO with hydrodesulfurization (HDS), typical HDS catalyst
(NiMoS, CoMoS supported on alumina) have been reported as active [6,7][8]. However,
the lack of sulfur-containing compounds causes deactivation by the oxygen -sulfur
exchange between sulfide active sites and oxygen in the mixture[9][6]. In this regard, the
development of different active and stable catalytic materials is needed. Thus, metals like
Fe, Cu and Ni have been widely studied in HDO. For instance, reduced nickel has shown to
be highly active in aromatics hydrodeoxygenation [10]. Ni over oxidic supports exhibited
the best performance in HDO in comparison with other metals, even as noble metals like Pt
[
11]. In a previous work we have reported a synergetic effect between Ni and Ru bimetallic
materials supported on Al , TiO and ZrO in phenol HDO with promising activity and
2
O
3
2
2
selectivity results. [12]. For these materials, the presence of ruthenium enhanced nickel
0
reducibility which was related to hydrogen dissociation of Ru and a spillover effect with
Ni [13], that could improve the hydrogen dissociation. This is needed for hydrogenation
reactions that take place on metallic sites [9][14]. In addition to that, materials supported on
TiO
and Al
TiO that induced the interaction of oxygen with the support, weakening the C-O bond and
promoting the HDO reactions.[15][16].
2
have exhibited an outstanding performance as compared with other supports like SiO
2
2
O
3
[12][15]. Some authors have attributed this feature to the oxophilic nature of
2
Moreover, Ni-Ru bimetallic materials have been widely studied for other reactions, like CO
methanation [17][13][18][19], CO reforming [20][21] and hydrogen production
22][23][24]. Most of these Ni-Ru systems were calcined at temperature values above 773
2
[
K, leading to the formation of the respective metals oxides. On one hand, it is well known
that calcination of Ni catalyst may induce strong interactions of the support with metal
entities. These Ni species became more difficult to reduce as calcination temperature
increased[25][26][27]. Furthermore, Louis et al. [28] indicated that calcination before
reduction for Ni catalysts had an important influence on particle size distribution. These
authors found that a reduction treatment without calcination induced the formation of

smaller particles than the reduction of calcined materials when Ni nitrate was used as
precursor. On other hand, for Ru, other authors have reported that the reduction of
impregnated ruthenium occurred at lower temperature when the precursor was not calcined
as compared with the calcined solids. Also, reduction temperature and particle size
increased with calcination temperature [29][30].
Nevertheless, the impact of direct reduction without calcination on HDO reactions has not
been reported, to our knowledge, for Ni Ru bimetallic materials. It can be proposed that
differences between both materials may be expected in terms of dispersion for the active
phases (Ni and Ru) and metal support interactions [26][27][25][29][30]. Therefore, the aim
of this work was to evaluate the effect of calcination on these parameters and their impact
on activity and selectivity for NiRu/TiO catalysts in HDO of phenol as a model compound
2
for bio oil. Catalysts were characterized by hydrogen chemisorption, Temperature-
programed reduction, X-ray photoelectron spectroscopy, High resolution electron
microscopy. Additionally, hydrogenation of benzene was used as a structure unsensitive
reaction in order to obtain more information about metallic dispersion.
1
. Experimental
1
.1.Synthesis
TiO
2
was synthesized by sol-gel method as described elsewhere [31]. Specific area was
2
-1
S
BET =93 m g . XRD characterization revealed the presence of the anatase phase
exclusively.
Bimetallic Ni - Ru catalysts were prepared by wet co-impregnation using an aqueous
solution of nickel(II) nitrate hexahydrate (99.999 % Ni(NO
3
)
2
·6H
2
O; Aldrich), and
ruthenium(III) nitrosyl nitrate (1.5 wt.-%Ru/vol., Ru(NO)(NO
)
3 3
; Aldrich) . Nominal metal
loadings were 6 wt.-% Ni and 0.6 wt.-% Ru similar to the published by Tada et al [17].
Solids were dried at 393 K for 12 h then, one material was calcined at 773 K in static air
while the other was not calcined to assess the influence of this process on the catalytic
activity. Materials were reduced in-situ before reaction in a 60 ml/min H
2
flow at 773 K for
calcined catalysts and 623 K for the uncalcined catalyst. Reduction temperatures were

selected acording to TPR analyses and in order to avoid formation of metal titanate species
in the uncalcined sample. It is highly possible that the higher reduction temperature used in
calcined catalyst could also induce some sintering.
1
.2.Hydrogen Chemisorption
Materials were reduced before analysis (623 K for uncalcined and 773 K for calcined).
Once reduced and cooled, the reactor was evacuated with Ar. Chemisorption was carried
out by pulses which were injected continuously until TCD signal do not vary indicating that
hydrogen molecules covered metallic surface.
1
.3.Hydrogenation of benzene
The hydrogenation of benzene was chosen as a structure unsensitive reaction in order to
obtain information about differences in metallic dispersion. Materials were reduced before
experiments. Reaction was carried in a continuous down flow reactor at atmospheric
pressure. Catalysts were reduced in situ before reaction at the same conditions used for
phenol HDO (60 ml/min H
2
flow at 773 K for calcined catalysts and 623 K for the
uncalcined catalyst), then reactor was cooled down to 323 K . Benzene was introduced to
-3
-1
the reactor using H
2
stream as carrier through a saturator (WHSV = 2.4X10 h ). The
reactor effluents were analyzed by a GC HP 5890 with a FID detector.
1
.4.High-resolution transmission electron microscopy
High-resolution transmission electron microscopy (HRTEM) analyses for reduced catalysts
were performed with a JEOL-2100F TEM (200 kV) instrument equipped with a Link ISIS
micro-analysis system. Its resolution was 0.19 nm. The high-resolution images were
captured digitally with a Gatan CCD digital camera. Samples were reduced under
activation conditions as described previously and isolated under Ar atmosphere to avoid air
contact. Before analyses, the reduced samples were finely ground, ultrasonically dispersed
in ethanol and collected on a carbon-coated copper grid.

1
.5.Temperature-programmed reduction
The temperature-programmed reduction (TPR) experiments for solids were carried out in
an AMI-90 apparatus (Altamira) equipped with a thermal conductivity detector (TCD).
About 50 mg of a sample were placed in a quartz sample cell (U-shaped) for each analysis.
The samples were pretreated in situ at 373 K for 1h under Ar flow to remove physisorbed
impurities. The reduction step was performed under a stream of (10 vol.-%) H /Ar (50
2
3
-1
-1
cm ·min ), with a heating rate of 10 K min up to 1073 K. A thermal conductivity detector
was used to determine variations in the hydrogen composition of the output current. A
moisture trap was used to prevent measurements interferences.
1
.6.X-ray photoelectron spectroscopy
The X-ray photoelectron spectroscopy (XPS) studies for freshly reduced catalysts were
performed in a VG scalab 200R spectrometer equipped with a hemispherical analyzer and
Mg k (h= 1253.6 eV) X-ray source. The calcined un uncalcined samples were in situ
–
5
reduced in H at the mentioned conditions and then degassed (< 10 Pa). All the data were
2
acquired using monochromatic Mg K radiation (1253.6 eV). The analyzed surface on the
sample was 300 x 700 nm, and it allowed a representative characterization of the whole
sample. The BE values for Ni 2p and Ru 3d spectra (±0.2 eV) were determined by
computer fitting using Casa XPS software of the measured spectra applying a Shirley
background subtraction and Gaussian-Lorentzian decomposition parameters with 30/70
Gaussian/Lorentzian proportion. The BE were corrected taking the C 1s (248.4 eV) signal
as a reference.
1
.7.Temperature-programmed desorption of amonia
The acidity for the catalysts was determined using the isothermal adsorption of NH
3
as
obtained on a Micrometrics 2900 equipment. Before the analysis, the materials were in situ
reduced in hydrogen under conditions described before. Then the solids were cooled and
saturated with NH
between 273-923 K (10 K∙min ) in He flow. A Thermal Conductivity Detector was used to
measure the NH desorption. The area under desorption curve of the desorption profile was
3
at 373 K with a (5%) NH
3
/Ar flow. Desorption process was carried out
–1
3

used to quantify the acid sites. The acid site strength classification was based on the peak
temperature range as follows: weak, 273K-523 K; medium, 523-673 K; and strong: 673-
7
23 K; according to previously published results [12].
1
.8.Hydrodeoxygenation of phenol
The hydrodeoxygenation of phenol was carried out in a high pressure laboratory scale set
up equipped with a down flow fixed bed catalytic reactor. For each experiment ca. 0.15 g of
catalyst (particle diameter 0.25–0.30 mm) were mixed with SiC (3:1) diameter 0.25 mm).
The catalyst was reduced in situ at 623 K (for uncalcined) and 773 K (for calcined) 2 h in
H
2
under atmospheric pressure. The temperature was selected after TPR experiment, as the
temperature were metal was expected to be completely reduced. After reduction, H
2
pressure was increased to the desired value (3.0 MP), and the catalytic bed was heated up to
the temperature of the HDO reaction (593 K). The liquid feed (2 vol% of phenol dissolved
in dodecane) was injected by a high-pressure HPLC Knauer pump (WHSV = 2.57 h−1) into
a hydrogen stream (40 mL min−1). The reactor effluents were condensed, and liquid
samples were analyzed by a GC Agilent 6890A with a FID detector.
1
.9.TGA
The amount of deposited coke on the catalysts tested in HDO of phenol reaction was
determined with a thermogravimetric TGA/SDTA851 equipment (Mettler Toledo),
measuring the weight change in the coked catalysts during oxidation. Each sample (ca. 30
mg) was previously heated in inert atmosphere remove the volatile compounds up to 773 K
-1
-1
for 1 h in N
2
(10 K min ; 200 mL min flow-rate). Once the sample was cooled to 300 K
-1
(
5 K min ) burning of coke was carried out by raising sample temperature to a final
-1
/N
gas mixture (50 mL min-1).
temperature of 1073 K at a rate of 10 K min in a 20% O
2
2
2
. Results and discussion
2
.1.Hydrogen Chemisorption
Hydrogen chemisorption was performed to compare exposed metallic areas for calcined
and uncalcined bimetallic NiRu catalysts supported on TiO . Table 1 gives the amount of
2

chemisorbed hydrogen per gram of catalyst. It was observed that the amount of gas retained
by the calcined solid was 4.2 H2 mol/g of catalyst , while for the uncalcined sample was
4
5.8 H2 mol/g of catalyst (ca. 9 times). This fact indicated that calcination affected
considerably the metallic dispersion.
2
.2.Hydrogenation of benzene
Benzene hydrogenation was carried out as a structure unsensitive reaction in order to obtain
more information about differences in metallic dispersion. Since hydrogenation reactions
only take place on metallic sites[32], differences in hydrogenation rates are only related to
differences in metalic dispersion for catalyst with identical metal loadings. Table 1 gives
reaction rates for benzene hydrogenation for both NiRu catalysts. Uncalcined catalyst
exhibited a 8 fold higher activity than that for the calcined one, in good agreement with the
H
2
chemisorption results. These results clearly indicated that metallic particles in
uncalcined NiRu/TiO showed higher dispersed than in the calcined NiRu/TiO
2
2
.
2
.3.High-resolution transmission electron microscopy
To explore the effect of calcination on the morphology and surface distribution for metallic
crystallites, high-resolution images for uncalcined and calcined NiRu/TiO catalysts were
2
analyzed (Figure 1 ). In general, arbitrary agglomerated crystalline particles were observed.
From Fast Fourier Transform (FFT) performed on the images, these particles presented
mainly d-spacing of 3.52 and 1.89 Ǻ, that could be indexed to the (101) and (200)
crystalline planes on the anatase-TiO lattice. The presence of Ni and Ru was confirmed by
2
EDX and indexation of crystalline metal particles in HRTEM images. In this case, small
particles with d-spacing 2.04 Ǻ were indexed to the (111) crystalline plane of face-centered
cubic Ni lattice (cubic-Ni; JCPD-ICDD-PDF: 004-0850). Meanwhile, the d-spacing of 2.09
and 2.3 nm could be indexed as (101) and (100) crystalline planes of hexagonal close-
packed Ru lattice (hexagonal-Ru; JCPD-ICDD-PDF: 006-0663). Figure 1 shows the
particle size distributions for the uncalcined and calcined solids respectively,.It was
confirmed that calcination influenced strongly metallic dispersion. Therefore, these metallic
particles for uncalcined catalyst were almost 4.6 times smaller than those observed on the
calcined one. Taking into account that the statistic calculation of the particle size is based

on the information obtained from a large number of TEM images having low resolution.
Thus, we assume that the particle size obtained correspond mainly to metallic Ni phase.
It was not possible to identify the formation of bimetallic clusters or zones of close contact
between Ni and Ru by using this technique, as previously reported by others[17]. As a
consequence, it is impossible to speculate on the possible influence of Ru on the catalyst
activity.
2
.4.Temperature-programmed reduction
Figure 2 shows the reduction profiles for uncalcined and calcined samples. The TPR
profile for calcined NiRu/TiO
2
catalysts (Figure 2 a)) exhibited a sharp peak centered at
0
4
19 K. It can be assigned to the RuO
2
→ Ru reduction in good agreement with the
literature[19][33]. The peak located at 680 K was asigned to NiO reduction, whereas the
X
reduction process between 470-570 K could be related to the reduction of bimetallic
species[19]. Above 780 K, all reducible species were transformed into their metallic state,
since reduction before reaction was carried out at 723 K. Thus, one may assume that Ni0
0
and Ru species were present during reaction. For the uncalcined-NiRu sample (Figure 2
b)), a sharp reduction peak at 483 K can be assigned to the reduction of Ru and Ni in the
3
+
0
impregnated ruthenium nitrosyl complex (Ru →Ru ) [34] and nickel hexahydrate complex
2
+
0
(
Ni →Ni ) [35][36] occurring simultaneously. The sharp shape of the reduction peak
suggested a uniform distribution of metal entities. Furthermore, the next low intense peak
was associated with a reduction of Ni complex in low interaction with the support[34][35].
0
Pre-reduction for reaction was performed at 623 K, thus the active phases were also Ni and
0
Ru . Based on the form and position of peaks, we assume that calcination induced stronger
metal support interaction than the drying, which promoted particle growth, as reported by
other authors[26][27][25].
2
.5.X-ray photoelectron spectroscopy
Figure 3 gives Ni 2p1/2 and Ni 2p3/2 XPS levels spectra for a) calcined and b) uncalcined
samples. Both spectra are composed of two adjacent bands with a maximum at 851 eV and
8
55 eV and their satellites. Appropriate decomposition was carried out for the spectra,

0
2+
considering bands associated with Ni and Ni (≈852.2 eV and ≈855eV respectively).
0
2+
Notable differences related with the amounts of Ni and Ni can be observed between both
catalysts (Table 2). For the calcined catalyst, 77% of Ni in the solid was reduced while 23
%
was at the oxidic state. Thus, the reduction conditions were not optimal to reduce
completely NiO as produced from calcination. Besides, only 60% of Ni was found in
metallic state and 40% as not reduced Ni species for the uncalcined sample. The reduction
conditions neither were enough for the total reduction of Ni. Table 2 gives XPS results for
the Ru 3d level. For both materials, the XPS results for the Ru 3d showed that this metal
was completely reduced after treatments.
2
.6.Temperature-programmed desorption
Table 3 gives the weak medium and strong acid sites densities as determined from NH
3
TPD experiments. This table shows that the total acid sites concentration was similar for
both materials. However the uncalcined solid exhibited 26% more weak and medium
strength acid sites according to the strength classification based on temperature desorption
range (273K-523 K; medium, 523-673 K; and strong: 673-723 K). A decrease of acidity
induced by calcination at high temperatures has been widely reported [37][27]. The higher
concentration of acid sites for the uncalcined sample can be explained in terms of
remaining OH groups on the carrier after impregnation, and the presence of NiOx species
originated from the precursors reduction process. Calcination of the catalyst, as generally
accepted, promotes the dehydroxylation of the support. In addition to the improved
dispersion, the higher weak and medium acid site density may also beneficiate the HDO
activity. These sites are involved in dehydration reactions needed to the cleavage of oxygen
as water. In this regard, a linear correlation between acid sites density and HDO reaction
rate was reported before [12][38].
In this work the uncalcined catalyst reduced at 773 K exhibited a larger amount of medium
strength acid sites than its calcined counterpart reduced at 623 K. For the strong strength
acid sites, the contrary trend occurs. This is expected because the calcination of NiRu/TiO2
sample led probably to formation of the metal titanate species, which were not formed by
direct reduction of the uncalcined catalyst.

2
.7.Hydrodeoxygenation of phenol
Table 4 gives the phenol hydrodeoxygenation activity and selectivity for calcined and
uncalcined catalyst. The HDO reaction rate for the calcined catalyst was 17.8 mol of
-1
-1
-1
-1
phenol* g of catalyst * s while for the uncalcined was 149 mol of phenol* catalyst * s
8.3 higher than the calcined one). Regarding HDO selectivities, the main products
(
identified by GC were cyclohexane and methylcyclopentane, with a small production of
benzene. Table 4 gives the HYD to DDO ratio. It can be seen that the HYD route was
preferred over DDO for both catalysts. In these materials phenol activation may occur via a
heterolytic dissociation of the O–H bond when phenol is absorbed on a surface oxygen site
of suport, near to a metallic site. This site stabilizes the phenoxide ion. Consequently, the
phenoxide ion will be in close interaction with the metal where the hydrogen is free to
react[11]. After hydrogenation take place, an OH group of ciclohexanol may be easily
dehydrated to yield cyclohexene molecule. It is commonly accepted that dehydration of
alcohols occurs on acid cites. Therefore, a high density of acid sites may induce a fast
dehydration of cyclohexanol[11][12]. In this work there was not correlation between the
acidities of the pre-reduced catalysts samples and their HYD/DDO selectivities ratio
calculated at the same phenol conversion. This is probably because the larger amount of
medium strength acid sites of the uncalcined sample was compensated by the effect of the
larger amount of strong acid sites of its calcined counterpart (Table 3). The absence of the
correlation between the catalysts acidities and HDO/DDO selectivities ratio was reported
previously[12].
Since catalysts were reduced before reaction, metallic Ni and Ru were the active phases for
0
0
both materials. Ni and Ru were obtained by the reduction of NiO and RuO for calcined
catalyst, as a results of calcination respectively. Besides, for uncalcined catalyst metallic Ni
and Ru were produced by the reduction of the residual impregnated species from Ni(NO
3 2
)
and Ru(NO)(NO . In this regard, the main differences between both material may be
3 3
)
dispersion of the same actives phases (Ni and Ru) and metal support interaction.
Additionally, particle growth by high temperature treatment and strong metal support
interaction as a consequence of calcination may explain the lower

activity[26][27][25][29][30]. Also, uncalcined Ru materials were reported as more active
previously[30].
Table 4 present also the estimated TOF values for both catalysts, the difference between
them may be indicative than metal dispersion is not enough to explain the main impact of
calcination process. However experimental errors should be taken into account when
comparing both values. It should be considered that Ru content in the calcined sample
could be slightly lower than in the uncalcined sample, since Ru precursors could be
oxidized at temperatures up to 673 K to a volatile form such as RuO [39].Additionally, the
4
larger density of acid sites in the uncalcined catalyst could contribute to a higher catalytic
activity for this sample as compared with the calcined one. . Current research is been
carried out to rule out this possibility.
2
.8.TGA
After catalyst drying, a burning of coke was carried out by raising sample temperature to a
-1
final value of 1073 K at a rate of 10 K min in a 20% O
2
/N gas mixture. This technique
2
confirmed that the main catalyst deactivation in this reaction was due to deposition of
carbonaceous residues/coke on the catalyst surface. Therefore, one might expect that the
availability of hydrogen on metal sites should be lower. The weight change during
oxidation of the spent catalysts in a 20% O
2
/N
2
mixture (TGA profiles) is shown in Figure
4
. At low temperature region, for all catalysts the weight gain is observed due to oxidation
2
+
2+
of metallic Ni( Ru) to Ni (Ru ). At high temperature region, the weight loss is due to the
gasification of carbon. The percentages of total mass gain and loss corresponding to
oxidation of metal species and coke, respectively, are shown in Table 5.
As seen in Figure 4, there is some change in the intensity and position of the peaks for all
samples induced by the different size of the metal particles and type of support. In good
agreement with study by Luo et al.[40], for calcined NiRu/TiO
carbon filamentous occurs at temperature above 943 K. Noticeably, the uncalcined
NiRu/TiO exhibits totally different TG profile: removal of carbon occurs at much lower
2
catalyst, the removal of
2
temperature (about 573 K) and occurs in two steps indicating the formation of two types of
coke and/or its different location on the catalyst surface[41] The catalyst regeneration by

oxidation is easy because of a low amount of coke formed and its more favourable
composition. Considering the study by Ibáñez et al. [41] we assume that coke formed has
aliphatic nature (non aromatic nature).
Summarizing, we conclude that the morphology of the carbon is essentially affected by the
catalyst pre-treatment (drying vs. calcination) and the size of the Ni/Ru particles. However,
the same amount of coke was formed on the calcined and non-calcined samples. Pre-
reduction of the uncalcined catalysts was more favorable to the catalyst stability because
formed coke, which could be more easily removed by the catalyst oxidation at much lower
temperature than in case of the calcined sample.
3
. Conclusions
Calcination of NiRu catalyst supported on TiO induced sintering of metal particles and
2
dispersion was significantly lower when compared with solids not calcined prior to
reduction. Calcination also induced a decrease of weak and medium acid sites. Both effects
influenced strongly higher HDO activities (8.3) for uncalcined material as compared with
the calcined solid. Finally, TGA showed that prereduction of uncalcined catalyst was more
favorable for the catalyst stability because formed coke could be easily removed by
catalyst oxidation at much lower temperature than in case of the calcined catalyst.
4
. Acknowledgements
The authors acknowledge financial support from CONACYT-México (Grant 237857). We
also thank Instituto Politécnico Nacional (Proyecto SIP 20181105).

5
. References
[
[
[
[
[
1]
2]
3]
4]
5]
G.W. Huber, S. Iborra, A. Corma, Synthesis of Transportation Fuels from Biomass:ꢀ
Chemistry, Catalysts, and Engineering, Chem. Rev. 106 (2006) 4044–4098.
doi:10.1021/cr068360d.
M. Stöcker, Biofuels and biomass-to-liquid fuels in the biorefinery: Catalytic conversion of
lignocellulosic biomass using porous materials, Angew. Chemie - Int. Ed. 47 (2008) 9200–
9211. doi:10.1002/anie.200801476.
C. Zhao, Y. Kou, A.A. Lemonidou, X. Li, J.A. Lercher, Highly selective catalytic
conversion of phenolic bio-oil to alkanes, Angew. Chemie - Int. Ed. 48 (2009) 3987–3990.
doi:10.1002/anie.200900404.
A. Gutierrez, R.K. Kaila, M.L. Honkela, R. Slioor, A.O.I. Krause, Hydrodeoxygenation of
guaiacol on noble metal catalysts, Catal. Today. 147 (2009) 239–246.
doi:10.1016/J.CATTOD.2008.10.037.
F.E. Massoth, P. Politzer, M.C. Concha, J.S. Murray, J. Jakowski, J. Simons, Catalytic
Hydrodeoxygenation of Methyl-Substituted Phenols:ꢀ Correlations of Kinetic Parameters
with Molecular Properties, J. Phys. Chem. B. 110 (2006) 14283–14291.
doi:10.1021/jp057332g.
[
[
6]
7]
V.N. Bui, D. Laurenti, P. Afanasiev, C. Geantet, Hydrodeoxygenation of guaiacol with
CoMo catalysts. Part I: Promoting effect of cobalt on HDO selectivity and activity, Appl.
Catal. B Environ. 101 (2011) 239–245. doi:10.1016/J.APCATB.2010.10.025.
A. Popov, E. Kondratieva, L. Mariey, J.M. Goupil, J. El Fallah, J.-P. Gilson, A. Travert, F.
Maugé, Bio-oil hydrodeoxygenation: Adsorption of phenolic compounds on sulfided
(
Co)Mo catalysts, J. Catal. 297 (2013) 176–186. doi:10.1016/J.JCAT.2012.10.005.
D.C. Elliott, Historical Developments in Hydroprocessing Bio-oils, Energy & Fuels. 21
2007) 1792–1815. doi:10.1021/ef070044u.
[
[
[
8]
9]
(
E. Furimsky, Catalytic hydrodeoxygenation, Appl. Catal. A Gen. 199 (2000) 147–190.
doi:10.1016/S0926-860X(99)00555-4.
10] P.M. Mortensen, J.D. Grunwaldt, P.A. Jensen, A.D. Jensen, Influence on nickel particle size
on the hydrodeoxygenation of phenol over Ni/SiO2, Catal. Today. 259 (2016) 277–284.
doi:10.1016/j.cattod.2015.08.022.
[
[
11] P.M. Mortensen, J.D. Grunwaldt, P.A. Jensen, A.D. Jensen, Screening of catalysts for
hydrodeoxygenation of phenol as a model compound for bio-oil, ACS Catal. 3 (2013) 1774–
1785. doi:10.1021/cs400266e.
12] O.U. Valdés-Martínez, V.A. Suárez-Toriello, J.A. d. los Reyes, B. Pawelec, J.L.G. Fierro,
Support effect and metals interactions for NiRu/Al2O3, TiO2and ZrO2catalysts in the
hydrodeoxygenation of phenol, Catal. Today. 296 (2017) 219–227.
doi:10.1016/j.cattod.2017.04.007.
[
13] S. Tada, D. Minori, F. Otsuka, R. Kikuchi, K. Osada, K. Akiyama, S. Satokawa, Effect of
Ru and Ni ratio on selective CO methanation over Ru-Ni/TiO 2, Fuel. 129 (2014) 219–224.
doi:10.1016/j.fuel.2014.03.069.
[
[
14] X. Zhu, L. Nie, L.L. Lobban, R.G. Mallinson, D.E. Resasco, E ffi cient Conversion of m ‑
Cresol to Aromatics on a Bifunctional Pt / HBeta Catalyst, (2014).
15] P.M. de Souza, R.C. Rabelo-Neto, L.E.P. Borges, G. Jacobs, B.H. Davis, D.E. Resasco, F.B.
Noronha, Hydrodeoxygenation of Phenol over Pd Catalysts. Effect of Support on Reaction
Mechanism and Catalyst Deactivation, ACS Catal. 7 (2017) 2058–2073.
doi:10.1021/acscatal.6b02022.
[
16] P.M. De Souza, L. Nie, L.E.P. Borges, F.B. Noronha, D.E. Resasco, Role of oxophilic
supports in the selective hydrodeoxygenation of m-cresol on Pd catalysts, Catal. Letters. 144

(
2014) 2005–2011. doi:10.1007/s10562-014-1337-y.
[
[
[
17] S. Tada, R. Kikuchi, K. Wada, K. Osada, K. Akiyama, S. Satokawa, Y. Kawashima, Long-
term durability of Ni/TiO2 and Ru–Ni/TiO2 catalysts for selective CO methanation, J.
Power Sources. 264 (2014) 59–66. doi:10.1016/J.JPOWSOUR.2014.04.075.
18] S. Tada, R. Kikuchi, A. Takagaki, T. Sugawara, S. Ted Oyama, S. Satokawa, Effect of metal
addition to Ru/TiO2catalyst on selective CO methanation, Catal. Today. 232 (2014) 16–21.
doi:10.1016/j.cattod.2013.09.048.
19] S. Tada, R. Kikuchi, A. Takagaki, T. Sugawara, S.T. Oyama, K. Urasaki, S. Satokawa,
Study of RuNi/TiO2catalysts for selective CO methanation, Appl. Catal. B Environ. 140–
141 (2013) 258–264. doi:10.1016/j.apcatb.2013.04.024.
[
[
[
20] C. Crisafulli, S. Scire, R. Maggiore, S. Minico, S. Galvagno, CO reforming of methane over
2
Ni – Ru and Ni – Pd bimetallic catalysts, Catal. Letters. 59 (1999) 21–26.
21] C. Crisafulli, S. Scirè, S. Minicò, L. Solarino, Ni – Ru bimetallic catalysts for the CO 2
reforming of methane, Appl. Catal. 225 (2002) 1–9. doi:10.1016/S0926-860X(01)00585-3.
22] E. Salehi, F.S. Azad, T. Harding, J. Abedi, Production of hydrogen by steam reforming of
bio-oil over Ni/Al2O3catalysts: Effect of addition of promoter and preparation procedure,
Fuel Process. Technol. 92 (2011) 2203–2210. doi:10.1016/j.fuproc.2011.07.002.
23] V. Modafferi, G. Panzera, V. Baglio, F. Frusteri, P.L. Antonucci, Propane reforming on Ni-
Ru/GDC catalyst: H2production for IT-SOFCs under SR and ATR conditions, Appl. Catal.
A Gen. 334 (2008) 1–9. doi:10.1016/j.apcata.2007.10.006.
24] S.C. Baek, K.W. Jun, Y.J. Lee, J.D. Kim, D.Y. Park, K.Y. Lee, Ru/Ni/MgAl2O4catalysts for
steam reforming of methane: Effects of Ru content on self-activation property, Res. Chem.
Intermed. 38 (2012) 1225–1236. doi:10.1007/s11164-011-0462-0.
25] F.W. Chang, M.S. Kuo, M.T. Tsay, M.C. Hsieh, Effect of calcination temperature on
catalyst reducibility and hydrogenation reactivity in rice husk ash-alumina supported nickel
systems, J. Chem. Technol. Biotechnol. 79 (2004) 691–699. doi:10.1002/jctb.1029.
26] J.G. Seo, M.H. Youn, J.S. Chung, I.K. Song, Effect of calcination temperature of
mesoporous nickel-alumina catalysts on their catalytic performance in hydrogen production
by steam reforming of liquefied natural gas (LNG), J. Ind. Eng. Chem. 16 (2010) 795–799.
doi:10.1016/j.jiec.2010.05.010.
[
[
[
[
[
27] L. Smoláková, M. Kout, E. Koudelková, L. Čapek, Effect of Calcination Temperature on the
Structure and Catalytic Performance of the Ni/Al
2
O
3
and Ni–Ce/Al
2
O Catalysts in
3
Oxidative Dehydrogenation of Ethane, Ind. Eng. Chem. Res. 54 (2015) 12730–12740.
doi:10.1021/acs.iecr.5b03425.
[
[
[
[
[
[
28] C. Louis, Z.X. Cheng, M. Che, Characterization of nickel/silica catalysts during
impregnation and further thermal activation treatment leading to metal particles, J. Phys.
Chem. 97 (1993) 5703–5712. doi:10.1021/j100123a040.
29] T. LI, S. WANG, D. GAO, S. WANG, Effect of support calcination temperature on the
catalytic properties of Ru/Ce0.8Zr0.2O2 for methanation of carbon dioxide, J. Fuel Chem.
Technol. 42 (2014) 1440–1446. doi:10.1016/S1872-5813(15)60001-9.
30] L. Zhao, J. Zhou, H. Chen, M. Zhang, Z. Sui, X. Zhou, Carbon nanofibers supported Ru
catalyst for sorbitol hydrogenolysis to glycols: Effect of calcination, Korean J. Chem. Eng.
27 (2010) 1412–1418. doi:10.1007/s11814-010-0257-9.
31] S. Núñez, J. Escobar, A. Vázquez, J.A. de los Reyes, M. Hernández-Barrera, 4,6-Dimethyl-
dibenzothiophene conversion over Al2O3–TiO2-supported noble metal catalysts, Mater.
Chem. Phys. 126 (2011) 237–247. doi:10.1016/J.MATCHEMPHYS.2010.11.032.
32] F. Fajardie, J.-F. Tempére, G. Djèga-Mariadassou, G. Blanchard, Benzene Hydrogenation as
a Tool for the Determination of the Percentage of Metal Exposed on Low Loaded Ceria
Supported Rhodium Catalysts, J. Catal. 163 (1996) 77–86. doi:10.1006/JCAT.1996.0306.
33] I.M. Leo, M.L. Granados, J.L.G. Fierro, R. Mariscal, Sorbitol hydrogenolysis to glycols by
supported ruthenium catalysts, Chinese J. Catal. 35 (2014) 614–621. doi:10.1016/S1872-

2067(14)60086-3.
[
[
34] A.M. Serrano-Sánchez, P. Steltenpohl, M.P. González-Marcos, J.A. González-Marcos, J.R.
González-Velasco, Evaluation of the oxidizing potential of Ru and Ru-Pt/ZrO2 catalysts for
NO oxidation in the presence of C3H6 and O2, Chem. Pap. 58 (2004) 41–47.
35] C.E. Santolalla-Vargas, V.A. Suárez Toriello, J.A. de los Reyes, D.K. Cromwell, B.
Pawelec, J.L.G. Fierro, Effects of pH and chelating agent on the NiWS phase formation in
NiW/γ-Al2O3 HDS catalysts, Mater. Chem. Phys. 166 (2015) 105–115.
doi:10.1016/J.MATCHEMPHYS.2015.09.033.
[
[
36] V.A. Suárez-Toriello, C.E. Santolalla-Vargas, J.A. de los Reyes, A. Vázquez-Zavala, M.
Vrinat, C. Geantet, Influence of the solution pH in impregnation with citric acid and activity
of Ni/W/Al2O3 catalysts, J. Mol. Catal. A Chem. 404–405 (2015) 36–46.
doi:10.1016/J.MOLCATA.2015.04.005.
37] H. Xu, M. Sun, S. Liu, Y. Li, J. Wang, Y. Chen, Effect of the calcination temperature of
cerium-zirconium mixed oxides on the structure and catalytic performance of WO3/CeZrO2
monolithic catalyst for selective catalytic reduction of NOx with NH3, RSC Adv. 7 (2017)
24177–24187. doi:10.1039/C7RA03054A.
[
[
[
[
38] Y. Elkasabi, C.A. Mullen, A.L.M.T. Pighinelli, A.A. Boateng, Hydrodeoxygenation of fast-
pyrolysis bio-oils from various feedstocks using carbon-supported catalysts, Fuel Process.
Technol. 123 (2014) 11–18. doi:10.1016/J.FUPROC.2014.01.039.
39] P.W. Cains, K.C. Yewer, S. Waring, Volatilization of Ruthenium, Caesium and Technetium
from Nitrate Systems in Nuclear Fuel Processing and Waste Solidification, Radiochim.
Acta. 56 (1992) 99–104. doi:10.1524/ract.1992.56.2.99.
40] J.Z. Luo, Z.L. Yu, C.F. Ng, C.T. Au, CO2/CH4 Reforming over Ni–La2O3/5A: An
Investigation on Carbon Deposition and Reaction Steps, J. Catal. 194 (2000) 198–210.
doi:10.1006/JCAT.2000.2941.
41] M. Ibáñez, M. Artetxe, G. Lopez, G. Elordi, J. Bilbao, M. Olazar, P. Castaño, Identification
of the coke deposited on an HZSM-5 zeolite catalyst during the sequenced pyrolysis–
cracking of HDPE, Appl. Catal. B Environ. 148–149 (2014) 436–445.
doi:10.1016/J.APCATB.2013.11.023.

Figure 1. HRTEM images and particle size distribution for NiRu/TiO
2
catalysts a) calcined
and b) uncalcined
Figure 2. TPR profiles for NiRu/TiO catalysts a) calcined and b) uncalcined
2
Figure 3. XPS spectra comparison for reduced NiRu/TiO catalysts a) calcined and b)
2
uncalcined
Figure 5. TG profiles for the reduced NiRu/TiO a) calcined and b) uncalcined catalysts
2
after catalytic evaluation in the HDO of phenol.
Figure 1
Figure 2

Figure 3

Figure 4

Table 1 H
2
Chemisorption results, benzene hydrogenation rates and calculated average particle size
for reduced NiRu TiO catalysts
2
Catalyst
NiRu/TiO
(
2
NiRu/TiO
(uncalcined)
2
calcined)a
b
H
2
chemisorbed H
2
[mol H
2
/ g of catalyst]
4.2
45.8
8.3
c
1
.17
Benzene HYD Rate [mol Benzene/ g of catalyst s]
Average particle size (nm)d
9.9
2.14
a
Pre-reduced at 773 K
Pre-reduced at 623 K
Calculated at steady-state conditions
Estimated by HRTEM
b
c
d
Table 2 XPS binding energies and relative percentage of species for reduced NiRu/TiO
2
catalysts.
Catalyst
NiRu/TiO
(
2
NiRu/TiO
(uncalcined)
2
calcined) a
b
0
Ni 2p3/2 Ni (%)
852.2 eV (77) 852.6 eV (60)
854.9 eV (23) 855.5 eV (40)
280.1 eV(100) 280.1 eV(100)
2+
Ni 2p3/2 Ni (%)
Ru 3d5/2 Ru0
a
After reduction at 773 K
After reduction at 623 K
b
Table 3. Acid sites density for reduced NiRu/ TiO
2
catalysts
Amount of acid sites (mol NH ·gcat)
3
Catalyst
NiRu/TiO
calcined)a
2
NiRu/TiO
(uncalcined)
2
b
(
Weak (273 K-523K)
Medium (523 K-673 K)
Strong (> 673 K)
58
45
19
64
67
6
a
After reduction at 773 K
After reduction at 623 K
b

Table 4 Catalytic activity and selectivity of reduced NiRu /TiO Catalyst
2
NiRu/TiO
(
2
NiRu/TiO
2
HDO od Phenol
calcined)a (uncalcined)
b
c
Phenol HDO Rate [mol/g of catalyst s]
17.8
149
d
Selectivity (HYD/DDO)
4.59
0.83
4.61
0.68
-1
TOF [s ]
a
After reduction at 773 K
After reduction at 623 K
Obtained at steady-state conditions
Calculated at same phenol conversion(70%)
b
c
d
Table 5 Percentages of mass gain and loss corresponding to oxidation of metal species and coke
deposition (from TG/DTA) of spent NiRu/TiO2 catalyst tested in HDO of phenol.
NiRu/TiO
2
NiRu/TiO
(uncalcined)
2
TG/TGA data
(
calcined)
Weight gain (%) a
Weight loss (%) b
2.4
1.1
2.8
0.5 + 0.6
a
Oxidation of Ni0 and Ru0
b
Coke