Surface investigation by X-ray photoelectron spectroscopy of Ru-Zn catalysts for the partial hydrogenation of benzene

Article abstract of DOI:10.1016/j.mcat.2019.110710

Partial hydrogenation of benzene with ruthenium-supported catalysts is an interesting alternative route to produce cyclohexene, an intermediate in the production of nylon 6.6 and other fine chemicals. In this work, ruthenium chloride solution was added to xZnO/Al₂O₃ by wetness impregnation. The supports were derived from hydrotalcite compounds and tested on the partial hydrogenation of benzene. The catalyst with 10 wt.% of ZnO (Ru10ZnAl) presented the best catalytic performance reaching total conversion in 40 min of the reaction, and a maximum yield of 35% using pure benzene (without water, salts or other additives). The samples were characterized by X-ray photoelectron spectroscopy (XPS), X-ray fluorescence (XRF), X-ray diffraction (XRD), temperature programmed reduction (TPR) and H₂ Chemisorption. Metallic and partially reduced ruthenium surface species were identified on the surface of the Ru10ZnAl catalyst and the electronic vicinity of Ru is proposed to be responsible for promoting the cyclohexene desorption, revealing a synergistic effect. In this sense, the electronic surroundings modification of surface zinc species with ruthenium species, followed by reduction, was evidenced by using the Modified Auger Parameter (MAP). These species played an important role in the reaction.

Full text of DOI:10.1016/j.mcat.2019.110710

MolecularCatalysisxxx(xxxx)xxxx
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Surface investigation by X-ray photoelectron spectroscopy of Ru-Zn catalysts
for the partial hydrogenation of benzene
Arthur H.A. Gonçalvesa, João Carlos S. Soaresa,b, Lúcia R. Raddi de Araújob, Fátima M.Z. Zotinb,
Fabiana M.T. Mendesa, Alexandre B. Gaspara,*
a Instituto Nacional de Tecnologia, Av. Venezuela, 82, Sl. 518, Rio de Janeiro, Brazil
b Instituto de Química, UERJ, Rua S. Francisco Xavier, 524, Rio de Janeiro, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords:
Partial hydrogenation of benzene with ruthenium-supported catalysts is an interesting alternative route to
produce cyclohexene, an intermediate in the production of nylon 6.6 and other fine chemicals. In this work,
ruthenium chloride solution was added to xZnO/Al2O3 by wetness impregnation. The supports were derived
from hydrotalcite compounds and tested on the partial hydrogenation of benzene. The catalyst with 10 wt.% of
ZnO (Ru10ZnAl) presented the best catalytic performance reaching total conversion in 40 min of the reaction,
and a maximum yield of 35% using pure benzene (without water, salts or other additives). The samples were
characterized by X-ray photoelectron spectroscopy (XPS), X-ray fluorescence (XRF), X-ray diffraction (XRD),
temperature programmed reduction (TPR) and H2 Chemisorption. Metallic and partially reduced ruthenium
surface species were identified on the surface of the Ru10ZnAl catalyst and the electronic vicinity of Ru is
proposed to be responsible for promoting the cyclohexene desorption, revealing a synergistic effect. In this sense,
the electronic surroundings modification of surface zinc species with ruthenium species, followed by reduction,
was evidenced by using the Modified Auger Parameter (MAP). These species played an important role in the
reaction.
X-ray photoelectron spectroscopy
Ruthenium-zinc interaction
Hydrotalcite route
Cyclohexene
Benzene hydrogenation
1. Introduction
cyclohexane, i.e. −23 kJ mol−1 and −98 kJ mol−1, respectively [2].
Reduced ruthenium catalysts are the most active for this reaction
Benzene is a key organic compound for the chemical industry and is
obtained mostly by fossil resources. Due to its high toxicity, new global
policies have been applied in order to prevent the addition of this
substance to fuels such as gasoline [1]. These constraints lead to an
excess of benzene, which tends to reduce its price. On the other hand,
benzene could be highly reactive in an appropriate catalytic system for
different reactions, generating new substances that are industrially
important.
Hydrogenation reactions of petroleum-derived chemicals have been
extensively studied, since they are crucial to obtain many important
intermediate compounds. One example is the hydrogenation of ben-
zene, an attractive route to produce cyclohexene or cyclohexane, which
are used in the synthesis of adipic acid and other fine chemicals [2–4].
The main challenge in the partial hydrogenation of benzene is to
avoid the adsorption of cyclohexene on the ruthenium surface, in-
hibiting its further hydrogenation to cyclohexane. That is due to the
unfavorable thermodynamics to obtain this compound – benzene hy-
drogenation to cyclohexene has a ΔG0 4.3 times higher than to
[5,6]. Additives such as zinc sulfate and water are used in order to
avoid cyclohexene adsorption on the catalyst surface [7,8]. Zinc and
water additions promote the formation of a stagnant film over the
catalyst surface and, during the reaction, both hydrogen and benzene
must overcome this barrier to reach the active center [9,10]. Thus, the
hydrophilicity of the catalyst may reduce the adsorption extent of the
cyclohexene formed, preventing its subsequent hydrogenation to cy-
clohexane, increasing the selectivity to cyclohexene [11]. However, the
addition of water has disadvantages; it hinders the separation of pro-
ducts, making the cost of the process more expensive. Furthermore, a
possible corrosion of the reactor is a problem with this system due to
the pH of the aqueous phase to be acidic, and a gradual deactivation of
the catalyst may occur due to a progressive corrosion products ad-
sorption on the catalyst surface, from the reactor wall, mainly of iron
[12,13].
The patented Asahi process uses four phases in order to promote
cyclohexene yields: gaseous (hydrogen), organic (benzene), aqueous
(ZnSO4 in aqueous solution) and solid (bulk ruthenium catalyst). [14].
⁎ Corresponding author:
E-mail address: alexandre.gaspar@int.gov.br (A.B. Gaspar).
https://doi.org/10.1016/j.mcat.2019.110710
Received 21 July 2019; Received in revised form 6 October 2019; Accepted 5 November 2019
2468-8231/©2019ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:ArthurH.A.Gonçalves,etal.,MolecularCatalysis,https://doi.org/10.1016/j.mcat.2019.110710

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
In this sense, many authors also have been dedicated to achieving a
supported catalyst of ruthenium that is active and selective as the
commercial one. They have observed improvement in activity and se-
lectivity to cyclohexene using zinc salts added to the liquid medium as
well as to the catalyst. Hu and Chen [15] prepared Ru-Zn/SiO2 catalysts
by the simultaneous addition of ruthenium chloride and zinc nitrate to
a SiO2 support using dry impregnation. The ruthenium content was
5 wt.% and zinc content varied from 0.3 to 3.0 wt.%. According to the
authors, small amounts of Zn retarded the reduction of ruthenium oxide
and increased the consumption of H2 up to 1 wt.% Zn, suggesting a
partial reduction of ZnO to Zn metal. The benzene hydrogenation was
carried out at 150 °C, 33.6 bar, 2.5 g of catalyst, 0.62 M NaOH, 75 mL of
benzene and 100 mL of water. However, while the selectivity was in-
creased after zinc addition, the catalyst activity was reduced, probably
due to a Ru sites coating through a zinc layer during the reduction. In
fact, the highest cyclohexene yield was observed at 5%Ru-1%Zn/SiO2
(31% at 55% conversion), which suggests a faster inhibition of hydro-
genation and/or cyclohexene than cyclohexane desorption.
Yuan et al. [16] observed the same effect when studying Ru-Zn/
ZrO2 catalysts using theoretical and experimental approaches. Ru-Zn/
ZrO2 samples were prepared by co-precipitation of ZrOCl2 and RuCl3
with KOH followed by ZnSO4 impregnation and reduction (Zn content:
2.4–3.0 wt.%). The catalysts were evaluated in the benzene hydro-
genation under the conditions: 2.0 g of catalyst, 80 mL of benzene,
150 mL of water and ZnSO4 in solution (Zn2+ = 0.35–0.55 M). The
authors observed that the activity decreased, but the selectivity for
cyclohexene increased, as a result of the decrease of the benzene and
cyclohexene adsorption on the catalysts, influenced by the Zn and water
addition in the reaction medium. The best result was obtained with a Zn
content of 2.72 wt.% with a cyclohexene yield of up to 44%.
Attempts to use other materials as support have also been made.
Zhang et al. [2] prepared ruthenium-zinc catalysts in hydroxyapatite.
Ru-Zn/HAP bimetallic catalysts (2.5 wt.% Ru) were prepared by the ion
exchange method using RuCl3 and ZnSO4. The benzene hydrogenation
was carried out in a 6 mL of a teflon coated stainless steel autoclave
reactor fitted with a magnetic stirrer under the following conditions:
0.5 mL of benzene, 20 mg of catalyst, and 1.5 mL of an aqueous solution
of sodium hydroxide. A yield of cyclohexene around 33.0% was ob-
tained by Ru-Zn/HAP (1: 1 Ru/Zn molar ratio) at 150 0C and 50 bar.
The authors reported that both metallic zinc and Zn2+ play an im-
portant role in increasing selectivity of cyclohexene. Compared to other
supports (MgO, CeO2 and ZrO2), hydroxyapatite gave better activity
and selectivity probably due to their adsorption capacity, high hydro-
philicity and dispersion of Ru-Zn nanoparticles, and the synergistic ef-
fect of metallic Zn and Zn2+ cations.
studied the total hydrogenation of benzene using hydrotalcites (Ru-HT)
obtained by partial substitution of Mg2+ or Al3+ cations by ruthenium
metal (1.0 wt.% Ru) in octahedral layers. The reactions were carried out
at 120 0C and 60 bar and the catalyst was not previously reduced. The
authors observed complete conversion with 100% selectivity to cyclo-
hexane after 2 h of reaction. No cyclohexene was observed, even with
addition of water to the reaction medium.
Fukuhara et al. [18] prepared Ru-HT catalyst with 15.3 wt.% Ru and
performed the partial hydrogenation of benzene, observing the influ-
ence of Zn and NaOH on both activity and selectivity. The highest yield
of cyclohexene (26%) was obtained with Ru - HT with zinc and NaOH
solution. Similar behavior was observed when the catalyst was sub-
jected to a 5000C heat treatment with nitrogen before the experiment.
The adsorption and desorption phenomena are important steps of
any heterogeneous catalytic process, so it is crucial to identify the
surface chemical species that are involved in the benzene to cyclo-
hexene conversion. X-ray Photoelectron Spectroscopy (XPS) is a pow-
erful technique capable of detecting the elements present on a target
surface and the different species of all elements, except hydrogen [25].
However, in XPS carbon is found as an impurity for almost every
analysis. In addition, C1s photoelectron peak coexists with Ru3d region,
turning the analysis more complex. Therefore, whenever the Ru3d
photoelectron peak is analyzed, C1s components must be taken in ac-
count in the peak model [26,27].
In this context, the aim of the present work was to prepare Ru/Zn/
Al catalysts by the hydrotalcite route and study their performance in the
partial hydrogenation of benzene without additives. A deeply XPS study
of the surface of these catalysts was carried out in order to investigate
the catalytic performance towards of the reaction, evidencing the role
of zinc species. Aware of the complexity of Ru3d region, metallic ru-
thenium and ruthenium oxide were used as standard materials.
2. Experimental
2.1. Catalysts preparation
The xZnO-Al2O3 supports (where x was 10 or 50 wt.% of zinc oxide)
were prepared by coprecipitation, with temperature and pH control,
following the methodology used for the synthesis of hydrotalcites [20].
Thus, two aqueous solutions were prepared for each support using the
precursor salts (Merck) Zn(NO3).6H2O (0.2 and 0.9 M) and Al
(NO3).9H2O (2.6 and 1.5 M), in order to obtain the concentrations of 10
or 50 wt.% of ZnO, respectively. These solutions were mixed, trans-
ferred to a burette and added slowly to 200 mL of distilled water,
maintained at temperature of 70 °C and stirring at 400 rpm. The pH was
stabilized at 7.0 throughout addition of a solution (1:1) of KOH (1.6 M)
and K2CO3 (1.6 M). After the addition, stirring was kept for 4 h. The
final solution was allowed to stand for about 12 h. After this period, the
solid was purified by washing at pH 7.0, followed by centrifugation and
drying in an oven at 120 °C for 18 h. Using this methodology, the ma-
terials obtained were denominated 10ZnAl and 50ZnAl, which contain
10 and 50 wt.% of ZnO, corresponding to Zn/Al molar ratios of 0.07
and 0.63, respectively. Zinc oxide was obtained by precipitation of
ZnCl2 (36.7 g in 365 mL of deionized H2O) with urea (8.3 g in 61 mL of
deionized H2O), adapted from the method described by Chuah et al.
[28]. The precipitate was aged for 96 h with pH control at approxi-
mately 10.0, in reflux apparatus with magnetic stirring and ultrasonic
bath. Subsequently, the resulting solution was washed by washing at pH
7 and drying in an oven at 120 °C for 18 h. γ-Al2O3, also used as support,
was a commercial Harshaw Al-3916 P from Engelhard.
Among the different supports used, there are no reports in the open
literature about the use of mixed oxides prepared by hydrotalcite route
for the partial benzene hydrogenation without additives. Hydrotalcite,
also called layered double hydroxides, are mixed metal hydroxides
3+
which general formula is [M2+
M
(OH)2][Am−
]x/m.nH2O, where x is
(1-x)
x
the ratio M3+/(M2++M3+). M2+ and M3+ are, respectively, divalent
and trivalent cations positioned at the octahedral sites in the hydroxyl
layers and Am- is the interlayer anion.
Hydrotalcite type compounds can be calcined above 450 0C gen-
erating mixed oxides with high surface area, giving rise to highly dis-
persed metal crystals [17]. The redox and acid-base properties of these
mixed oxides will depend on the composition of M2+ and M3+ cations.
These materials have been studied for many applications in catalysis,
such as hydrogenation, hydrogenolysis and condensation reactions
[18–21]. They are promising to associate zinc with ruthenium aiming to
achieve higher selectivity to cyclohexene since Zn2+ is one option for
these substituent cations. The catalyst separation from the reaction
products provides the possibility of its reuse, increasing the pro-
ductivity of the process.
All supports were calcined at 400 °C for 5 h at a rate of 10 °C min−1
and were maintained in the granulometric range between 200 and 270
mesh. After the preparation of all supports, an aqueous solution of
RuCl3.xH2O was added by wetness impregnation and, thereafter, cal-
cined at 400 °C for 5 h at 10 °C min-1 in a muffle under atmospheric air
originating Ru/xZnO-Al2O3 (x = 10 or 50 wt.% ZnO, Ru10ZnAl or
Hydrotalcite based on Mg and Al can be changed by the partial or
total substitution of these cations [17,22–24]. Sharma et al. [17]
2

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
Ru50ZnAl), Ru/Al2O3 (RuAl) and Ru/ZnO (RuZn) catalysts (Ru
=1.0 wt.%). Before X-ray diffraction (XRD) and XPS measurements, an
enough amount of each catalyst was reduced at 250 °C, for 1 h, under
pure H2 flow of 50 mL min−1, cooled with He up to -80 °C and then
passivated with a mixture of O2/He (50 mL min−1) during 30 min.
These catalysts received the letters RP referencing the reduction fol-
lowed by passivation treatments. Powdered RuO2 and metallic Ru
(Aldrich) were used in the XPS analysis as reference materials.
Table 1
Chemical analysis by XRF, TPR-H2 consumption, H2 chemisorption and RuO2
particle diameter for all calcined samples.
Sample
Cl/Rua Al/(Zn + Al)b TPR-H2
H2 Chemisorption DRu
(μmols
(μmols gRu−1
)
(nm)
mgRu−1
)
c
RuAl
Ru10ZnAl 3.3
Ru50ZnAl 3.4
3.1
1
15.2
47.3
14.3
17.8
113
828
60
29
8
44
24
0.94
0.63
0
RuZn
0
41
2.2. Catalyst characterization
a
Molar%.
Atomic ratio.
The structural characteristics of the samples were obtained by X-ray
diffraction in a Rigaku Miniflex diffractometer equipped with CuKα
radiation of 1.541 Å. The parameters were 2θ from 10° to 70°, 0.02° per
step, and scan time of 1 s per step. RuO2 particle diameters for all
calcined catalysts were calculated by using the Scherrer equation.
Chemical analysis of the calcined catalysts was performed by the X-ray
fluorescence technique with a Bruker S8 Tiger spectrometer. It was
assembled with a rhodium tube and operated at 4 kW. The textural
analysis of the catalysts was determined by the N2 adsorption in a
Micromeritics ASAP 2020 equipment. The samples were pretreated at
300 °C and isotherms were obtained at −196 °C. Temperature pro-
b
c
Stoichiometric reduction for RuO2: 19.8 and RuCl3: 14.8 μmol s H2 mg
Ru−1
.
3. Results and discussion
Chemical analysis of the calcined samples, obtained by using XRF, is
shown in Table 1. After impregnation and calcination at 400 °C, all
aluminum-containing samples presented an atomic Cl/Ru ratio between
3.1 and 3.4, close to the stoichiometric 3.0. RuCl3.xH2O was used as the
precursor salt and, according to the literature, alumina has the char-
acteristic of retaining chloride ions [31–34]. Concerning textural ana-
lyses, RuAl, Ru10ZnAl and Ru50ZnAl presented specific areas of 220,
297 and 211 m2 g−1, respectively. Accordingly, the pore volumes were
0.47, 0.56 and 0.33 cm3 g−1. All catalysts presented type IV isotherms
regarding mesoporous materials with hysteresis loop type H2. All
samples presented a monomodal distribution of pore diameters with a
mean of approximately 50 Å. The RuZn catalyst presented a very low
grammed reduction (TPR) analysis was carried out using
a
Micromeritics AutoChem II equipment. The materials were dried at
150 °C, for 1 h, with nitrogen. After this step, the reduction was pro-
grammed from room temperature to 500 °C (10 °C min−1), under 10%
H2/N2 flow (50 mL min−1). H2 chemisorption was performed at the
same equipment. After TPR analysis, the samples were purged with N2
flow of 50 mL min−1 and the H2 pulses were carried out at 100 °C.
Surface analysis of all catalysts was carried out by X-Ray
Photoelectron Spectroscopy (XPS). It was used a hemispherical spec-
trometer PHOIBOS 150 from SPECS, appareled with X-Ray Gun (model
XR-50) with a non-monochromatic Al Kα/Mg Kα source. The power
rating of the anode was 10 W and for survey spectra, pass energy was
50 eV, while it was 20 eV for the regions of interest. Due to the coex-
istence of C1s peak in the region of Ru3d, the aluminum containing
samples spectra were calibrated using Al 2p peak (74.4 eV) [29] while
the standards spectra were calibrated using the C1s peak at 284.6 eV.
The background used to investigate the species of each spectrum was
Shirley baseline and all mathematical treatment was performed using
the Tags quantification available in the CasaXPS software version
2.3.16 [30]. The pressure in the analysis chamber was below 10−9
mbar during data acquisition.
specific area (< 10 m²
g
−1) as well as the smallest pore volume
(0.01 cm3 g−1).
Fig. 1 shows the TPR profiles of the calcined catalysts. RuAl and
Ru10ZnAl exhibited a reduction peak around 140 °C, which can be at-
tributed to the RuCl3 reduction [35]. On the other hand, Ru50ZnAl and
RuZn TPR profiles did not show the peak related to RuCl3 reduction.
The reduction peak observed in the temperature range between 190 °C
and 220 °C was attributed to the RuO2 reduction [35,36]. RuAl catalyst
presented a shoulder around 160 °C that can be related to well-dis-
persed Ru species [35]. Last but not least, Ru10ZnAl presented a broad
reduction peak between 275 °C and 375 °C, which can be related to the
reduction of zinc species. Furthermore, the H2 consumption by TPR for
this catalyst was higher than the stoichiometric, as shown in Table 1.
The H2 chemisorption data, which are commonly used to determine
the metallic dispersion, are also given in Table 1. The order for the H2
chemisorption was Ru10ZnAl > > > RuAl > Ru50ZnAl > RuZn.
2.3. Partial hydrogenation of benzene
The catalytic tests were performed in a PARR 4566 batch reactor
with a 300 mL vessel. The catalysts were activated in situ with 50 mL
min−1 flow of pure H2 at 250 °C for 1 h. After reduction, in order to
maintain an inert atmosphere, the reactor was purged with pure N2 and
cooled down to reach the reaction temperature, which was 150 °C.
Then, 100 mL of pure benzene was added to 0.5 g of catalyst. The H2
pressure was kept at 50 bar with an agitation of 600 rpm during the 3 h
of each reaction. Aliquots were collected in an interval of 10 min in the
first hour and thereafter, every 30 min. The products were identified
and quantified in a HP6890 gas chromatograph assembled with a NST-
100 column and FID detector. The catalysts performances were eval-
uated by observing the conversion of benzene, selectivity and yield to
cyclohexene. These were calculated by the following expressions:
Conversion (%) = (mols of reacted benzene/mols of initial benzene) ×
100
Selectivity (%) = (mols of obtained cyclohexene/mols of reacted ben-
zene) × 100
Fig. 1. TPR profiles of all calcined catalysts.
Yield (%) = (Conversion (%) × Selectivity (%))/100
3

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
Fig. 2. TPR profiles of calcined Ru10ZnAl and 10ZnAl.
Fig. 4. XRD patterns of the calcined aluminum-containing catalysts. *:RuO2; #:
ZnAl2O4; dashed line: γ-Al2O3.
These results indicate that the ruthenium particles are more dispersed
on the 10ZnAl support surface, while the other catalyst presented a
lower metallic dispersion.
presented in Fig. 4. The reflections at 2θ = 37.2°, 46.0° and 66.9°
(JCPDS 1-088-2112), observed for RuAl, are characteristic of the γ-
Al2O3 support. It is noted that RuAl, Ru10ZnAl and Ru50ZnAl presented
the three peaks relative to crystalline RuO2, which are 2θ = 28°, 35°
and 54° (JCPDS-40-1290). The spinel formation (ZnAl2O4) was ob-
served for both Ru10ZnAl and Ru50ZnAl catalysts with the main peaks
observed at 2θ = 31.2°, 36.7°, 44.7°, 49.1°, 59.3° and 65.3° (JCPDS-05-
0669) [40]. In addition, Ru50ZnAl catalyst also presented a displace-
ment of the main line (2θ = 36.7°) to 2θ = 36.2°. This suggests that for
this sample segregation of ZnO occured. XRD result of RuZn is pre-
sented in Fig. 5 and only two peaks corresponding to RuO2 – i.e.
2θ = 28° and 54° – can be well distinguished for RuZn. The peak at
2θ = 35° is probably superimposed by the very intense ZnO char-
acteristic peaks. The RuO2 diameter calculated by the Scherrer equa-
tion, for the calcined catalysts is presented in Table 1. The particle size
for Ru50ZnAl is approximately 6 times larger than the found for
Ru10ZnAl. This observation agrees with the H2 chemisorption analysis,
also presented in Table 1, showing that ruthenium tends to aggregate
on the 50ZnAl support surface, while for the 10ZnAl support more
dispersed ruthenium particles are observed.
In order to investigate if the aforementioned reduction region in the
Ru10ZnAl catalyst profile – 275 °C–375 °C – is in fact associated to the
consumption of H2 by zinc species, a TPR analysis of the 10ZnAl-cal-
cined support was performed. Fig. 2 shows that there is a reduction
around 400 °C, which suggests a partial reduction of zinc oxide in the
support. After ruthenium impregnation, the reduction temperature of
ZnO is shifted to lower temperatures. The difference between the
Ru10ZnAl catalyst and the support, regarding the temperature of re-
duction of ZnO, probably is due to the interaction of Ru and support.
The presence of ruthenium is responsible for promoting the strong
metal-support interaction (SMSI) and/or spillover effects, phenomena
that may cause the partial reduction of zinc, already reported in lit-
erature [2,15,37–39]. That may explain the high H2 consumption result
for the Ru10ZnAl catalyst, Table 1.
X-ray diffractograms of the 10ZnAl and 50ZnAl supports before
calcination are shown in Fig. 3. Considering the Al/(Al + Zn) atomic
ratio presented in Table 1, Zn-Al hydrotalcites are not expected, since
according to Cavani et al. [22], they are only formed in the range of
0.20 and 0.33 of Al/(Zn + Al) atomic ratio. The 10ZnAl support showed
the typical boehmite structure (JCPDS 1-088-2112). No hydrotalcite
phase was observed on this support. Besides, the absence of the ZnO
phase (JCPDS 36-1451) suggests that this species is amorphous or well
dispersed in the boehmite structure. However, the 50ZnAl support
presented typical structure of hydrotalcite [23,24].
The ruthenium reduction was also investigated by X-ray diffraction
and it was used to reinforce the hypothesis of the partial reduction of
the support for the Ru10ZnAl catalyst due to the interaction between
ruthenium and zinc. Diffractograms of Ru10ZnAl and Ru10ZnAl-RP are
shown in Fig. 6. For Ru10ZnAl-RP, the peaks related to the ruthenium
oxide disappeared while two new showed up. At 2θ = 44° (JCPDS-06-
XRD patterns of all aluminum containing calcined catalysts are
Fig. 3. XRD patterns of the Zn-Al supports before calcination.
Fig. 5. XRD pattern of calcined RuZn catalyst. *:RuO2.
4

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
Fig. 6. XRD patterns of Ru10ZnAl and Ru10ZnAl-RP samples.*:RuO2 ; #: Zn0;
º:Ru0.
0663), the peak of metallic ruthenium appeared, and reduced zinc was
observed at 2θ = 42° (JCPDS-04-0831). These observations corroborate
with the hypothesis of partial reduction of zinc in Ru10ZnAl catalysts,
also evidenced by TPR analysis. Additionally, these results indicated
that the reduction, followed by the passivation procedure were effective
to stabilize the metallic particles of ruthenium.
XPS characterization of the Ru 3d region is admittedly complex,
mainly due to the overlapping of C1s and Ru 3d regions. Alternatively,
the Ru3p region can be used in order to support the Ru3d analysis.
Therefore, Ru3p regions of commercial metallic ruthenium and ruthe-
nium (IV) oxide were recorded. Ru3p was detected by observing the
doublet separation of approximately 22.2 eV [41] for all samples.
Table 2 shows the binding energies for the Ru3p3/2 and Ru3d5/2 species
compilation found for each reference and catalyst sample in this work.
As can be seen, the Ru3p3/2 binding energies for metallic ruthenium
and RuO2 samples were 461.4 eV and 462.5 eV, respectively, agreeing
with those observed by Morgan [41]. All calcined catalysts presented
Ru3p3/2 binding energies values lower than the RuO2 reference sample,
indicating that the supported ruthenium particles are surrounded by
less electronegative atoms such as Al and Zn. After RP procedure, the
Ru3p3/2 peak of all catalysts showed significant shifts to lower binding
energies values, confirming, at least, a partial reduction of the ruthe-
nium particles.
According to the literature, metallic ruthenium and RuO2, an
asymmetric line shape gives the best fit for the 3d3/2 and 3d5/2 lines
processing. A study considering the line shape asymmetry for different
ruthenium compounds including metallic ruthenium and RuO2, ob-
taining very accurate peak models was performed by Morgan [41].
Thus, in the present work, all Ru3d peak fittings were carried out by
applying the line shapes parameters reported by this author. As metallic
ruthenium and RuO2 have very similar Ru3d doublet separation values,
Fig. 7. Ru3d XPS spectrum for metallic ruthenium (a) and RuO2 (b) standards.
this parameter was set to 4.17 eV for all Ru3d doublets in each peak
model and the area ratio was set to 0.67. The residual plot for each peak
model, supplied by the CasaXPS software [30], showed that all em-
ployed parameters resulted in very accurate peak models.
Fig. 7a shows the Ru3d XPS spectrum for the commercial metallic
ruthenium powder. Ru0 was observed at 280.0 eV in a very good
agreement with the literature [26,41,42]. In addition, the adventitious
carbon was observed at 284.6 eV. The Ru3d spectrum for the other
reference material, ruthenium (IV) oxide powder (commercial), is ex-
hibited in Fig. 7b. Two Ru3d5/2 components were observed at 280.4 eV
and 282.5 eV B.E. The lower B.E. peak was attributed to Ru4+ and also
is in good agreement with the literature [26,41]. Many authors that
have studied this oxide have assigned the photoelectron peak at higher
binding energies to Ru6+ [42–44]. However, others explained that the
aforementioned peak would not be corresponding to the RuO3, but
satellite peaks, plasmon peaks and/or different surface terminations
[41,45–47]. Therefore, in this work, whenever a Ru4+ species related to
the ruthenium (IV) oxide was observed for the supported ruthenium
catalysts, its satellite was considered in the peak model. In addition,
two components related to adventitious carbon were found, as ex-
pected.
Table 2
Compilation of all Ru3d5/2 and Ru3p3/2 peaks for the ruthenium samples by
XPS.
Sample
Ru3p3/2
Ru3d5/2 Binding Energy (eV)
Binding Energy (eV)
Ru0
Ru4+
Ru4+SAT
Ruδ+
The prepared RuAl and Ru10ZnAl catalysts were analyzed by XPS
and the Ru3d photoelectron region is exhibited on Figs. 8 and 9, in
which the calcined samples are presented in Fig. 8 and the ones, which
were reduced and passivated (RP), are presented in Fig. 9. The corre-
sponding binding energies are shown in Table 2. The goal was to in-
vestigate surface modification prior and after RP process. After fitting
the data for the RuAl catalyst, two peaks - Ru4+ at 280.5 eV and its
satellite at 282.5 eV - were observed (Fig. 8a). The Ru10ZnAl catalyst
RuO2
Metallic Ru
RuAl
462.5
461.4
461.9
461.4
461.6
461.2
462.1
461.0
–
280.4
–
280.5
–
280.3
–
280.0
280.8
282.5
–
282.5
–
282.4
–
282.8
283.0
–
–
–
280.0
–
279.9
–
279.8
–
279.2
RuAl-RP
281.5
281.2
281.0
281.9
281.9
Ru10ZnAl
Ru10ZnAl-RP
Ru50ZnAl
Ru50ZnAl-RP
5

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
Fig. 8. Ru3d XPS spectrum of RuAl (a) and Ru10ZnAl (b).
presented Ru4+ at 280.3 eV and its satellite at 282.4 eV (Fig. 8b). It is
important to note that, differently than the RuAl catalyst, a third peak
appeared at 281.2 eV (Ruδ+), which suggests an interaction between
ruthenium and the support in the Ru10ZnAl catalyst. This effect was
previously discussed during the analysis of XRD and TPR results and has
already been reported in literature [2,15,37–39]. Also, the coexistence
of Ru-Cl species within the Ru 3d5/2 (Ru4+) satellite peak cannot be
discarded because residual chloride was found at 199.1 eV, which is
characteristic of metal-chloride interaction. RuCl3.xH2O was employed
as the precursor salt. Therefore, even after the calcination step, residual
chlorine was detected, which was also confirmed by TPR and XRF
techniques. On the other hand, after the reduction and passivation
procedures (Fig. 9a and b), only Ru0 (279.9 and 279.8 eV) and Ruδ+
(281.5 and 281.0 eV) were detected for both RuAl-RP and Ru10ZnAl-RP
catalysts, respectively. The difference of 0.5 eV between the so-called
Ruδ+ species among the samples RuAl-RP and Ru10ZnAl-RP suggests
that the interaction of ruthenium with the support is different for each
catalyst. In addition, although RuAl-RP and Ru10ZnAl-RP have similar
ruthenium surface species, it is apparent that the peak related to Ruδ+
is relatively more pronounced for the Ru10ZnAl-RP sample than that
observed for the RuAl sample.
Fig. 9. Ru3d XPS spectrum of RuAl-RP (a) and Ru10ZnAl-RP (b).
significantly lower than the found for the reference samples in this
work. Larichev et al. [48] studied Ru/MgO catalysts and observed a
similar phenomenon related to a differential charging effect. According
to Grünert et al. [49], differential charging may be expected with large
particles. This may explain the difference in the values found for these
species for the Ru50ZnAl samples. In addition, it is notable that the
binding energy values for the Ruδ+ species of Ru50ZnAl and Ru50ZnAl-
RP samples are shifted toward higher binding energies values compared
to the Ru10ZnAl and Ru10ZnAl-RP samples. This shift could be related
to the presence of Ru-Cl species, once the calcined Ru50ZnAl sample
presented a surface Cl/Ru ratio, by XPS, approximately 7.4 times higher
than the calcined Ru10ZnAl. Last but not least, a peak at 280.8 eV was
detected and considered as non-reduced RuO2, because of the satellite
feature at 283.0 eV. This result was very similar to that observed by
Morgan [41], which noted a Ru 3d5/2 satellite feature 1.9 eV distant
than the Ru 3d5/2 peak, for a RuO2 sample. The presence of RuO2 on
surface of the Ru50ZnAl-RP sample indicates that, even after the RP
procedure, it was not possible to reduce all the RuO2, stabilizing the Ru0
particles, probably due to the RuO2 particle aggregation (observed by
the low H2 chemisorption and, consequently, the larger particle size). A
careful quantitative analysis by XPS, was performed by using the Ru3d
peak models of the RP samples. The quantification of Ru3d region,
expressed as the relative percentage of ruthenium surface species on the
catalysts is presented in Table 3. Ru4+ surface species did not appear
for the RuAl-RP and Ru10ZnAl-RP samples after RP process. Ru10ZnAl-
RP sample presented the higher Ruδ+ relative percentage among the
other RP samples. It can be observed that the Ru50ZnAl-RP sample
The electronic surroundings on the surface of Ru50ZnAl and
Ru50ZnAl-RP catalysts were also investigated. The Ru3d XPS spectrum
for the calcined Ru50ZnAl sample (Fig. 10a) presented Ru4+ at
280.0 eV and its respective satellite at 282.8 eV. Similar to the calcined
Ru10ZnAl sample, a peak at 281.9 eV (Ruδ+) was also detected. After
the RP procedure (Fig. 10b), metallic ruthenium was observed at
279.2 eV. It is important to note that the binding energy values corre-
sponding to Ru4+ (Ru50ZnAl) and Ru0 (Ru50ZnAl-RP) were
6

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
Fig. 11. Zn2p3/2 shift (a) and ZnLMM Auger (b) for the hydrotalcite-like sam-
ples.
Fig. 10. Ru3d XPS spectrum of Ru50ZnAl (a) and Ru50ZnAl-RP (b).
Table 4
Table 3
Zn2p3/2, ZnLMM peaks and Modified Auger parameter between the hydro-
talcite-like supports and their respective reduced ruthenium catalysts.
Surface chemical analysis by XPS for all RP samples.
Sample
Relative percentage (%)
Ruδ+/(Ru0+ Ru4+) atomic ratio
Sample
Zn2p3/2
ZnLMMb
MAPc
a
Ru0
Ru4+
Ruδ+
10ZnAl
Ru10ZnAl-RP
50ZnAl
1022.0
1022.4
1022.2
1021.9
987.2
987.5
987.5
987.8
2009.2
2009.9
2009.7
2009.7
RuAl-RP
Ru10ZnAl-RP
Ru50ZnAl-RP
82.2
53.8
29.0
–
–
39.1
17.8
46.2
31.9
0.22
0.86
0.47
Ru50ZnAl-RP
a
Binding energy (BE - eV).
Kinetic energy (KE - eV).
MAP = Zn2p3/2 (BE) + ZnLMM (KE).
b
c
presented a Ru4+ relative percentage of 39.1%, which was the major
oxidation state found on the surface of the samples studied. Table 3 also
shows the Ruδ+/(Ru0++Ru4+) atomic ratio calculated by the TAG
methodology, based on the CasaXPS software manual [30]. The cal-
culated Ruδ+/(Ru0++Ru4+) atomic ratio for the Ru10ZnAl-RP sample
was 0.86, approximately the double that found for the Ru50ZnAl-RP
sample.
In order to investigate if the ruthenium impregnation and the RP
procedure changed the chemical environment of the zinc on the sup-
ports surfaces, high resolution spectra were performed in the regions of
Zn2p photoelectron peak (Fig. 11a) and ZnLMM Auger peak (Fig. 11b)
for the 10ZnAl and 50ZnAl supports and Ru10ZnAl-RP and Ru50ZnAl-
RP catalysts, respectively. Table 4 shows the values of the Zn2p3/2 peak
for all samples. It can be observed that there is a shift of 0.4 eV to higher
binding energies of the Zn2p3/2 peak after the RP procedure for the
Ru10ZnAl catalyst. Also, the Modified Auger Parameter (in this work
abbreviated as MAP), which is the sum of the Auger peak (ZnLMM)
kinetic energy and the main photoelectron peak (Zn2p3/2) binding en-
ergy [50], was used to confirm the Ru-Zn interaction in the Ru10ZnAl
catalyst after the ruthenium impregnation and RP procedure. As can be
seen in Table 4, the MAP value for the 10ZnAl support was 2009.2 eV,
which is characteristic of ZnAl2O4, and it is in good agreement with the
literature [50,51]. However, the MAP value found for the Ru10ZnAl-RP
sample was 2009.9 eV (0.7 eV higher than the observed for the 10ZnAl
support). On the contrary, the MAP value for the Ru50ZnAl-RP sample
remained the same as for the support 50ZnAl (2009.7 eV). The differ-
ence of 0.5 eV between the MAP values for the supports 10ZnAl and
50ZnAl is coherent. Strohmeier et al. [52] varied the zinc contents in
Zn/Al2O4 catalysts and observed an increasing in the MAP values for
7

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
hydrogenation reaction. These reaction rates cannot be directly com-
pared with the literature because data were obtained at distinct con-
ditions, such as pressures and temperatures. However, Cobo et al. [56]
worked with Ru/Al2O3 and Ru/CeO2 catalysts in different particle sizes
under pressure of 50 bar H2 and 100 °C and achieved maximum activity
of 1.7 mols min−1 gRu−1. Milone et al. [57] found that Ru/Al2O3
catalysts synthesized with the RuCl3 precursor led to better results in
cyclohexene selectivity and activity in the benzene hydrogenation re-
action than catalysts prepared with the ruthenium acetylacetonate (Ru
(acac)3) and nitrosyl nitrate precursors. (Ru(NO)(NO3)), reaching an
activity of 1.7 mol min−1 gRu−1
.
Markedly different from other mentioned studies, the present work
did not use water – which is used to increase the cyclohexene yields
[12,58]. According to Foppa and Dupont [58], the addition of water
improves the cyclohexene selectivity because of a difference in solu-
bility, where benzene is six-fold more soluble than cyclohexene,
125 mol m−3 and 21 mol m−3, respectively. Thus, when added, a
stagnant water layer is formed surrounding the catalyst surface, pro-
moting the feeding of benzene and desorption of the formed cyclo-
hexene. However, mass transport phenomena play an important role
when the reactional medium is composed with four phases in order to
promote cyclohexene yields: gaseous (hydrogen), organic (benzene),
aqueous (ZnSO4 in aqueous solution) and solid (bulk ruthenium cata-
lyst). Struijk et al. [12,59] studied the phenomena of external mass
transport and pore diffusion limitations of partial benzene hydrogena-
tion using Carberry numbers and the Wheeler-Weisz group calculations
and experimental data. The authors noted that the liquid/solid trans-
port of hydrogen is the reaction rate controlling step in the first hour of
the reaction and the mass transfer limitation of benzene gradually in-
creases, becoming the limiting step in the reaction sequence. These
constrains reduce the overall benzene hydrogen rate but increase cy-
clohexene yields as reported by many authors. Besides, the addition of
water may be considered a disadvantage to the process because an
additional step of organic / aqueous phase separation must necessarily
be included.
Fig. 12. Conversion of benzene (a) and selectivity to cyclohexene (b) in func-
tion of time for all catalysts.
In the present paper, in the absence of water, the most selective
catalyst to cyclohexene – Ru10ZnAl – presented the highest Ruδ+
/
the catalysts with zinc loadings above 20 wt.%, probably due to the ZnO
formation on surface, also observed by XRD in this work for the
Ru50ZnAl sample. These results suggest that even with a low content of
ruthenium on the catalyst surface (nominal content of 1 wt.%), the
presence of this metal changes the electronic environment of the zinc
for the Ru10ZnAl catalyst. Therefore, the ruthenium impregnation as-
sociated with the RP procedure were responsible to provoke an electron
transfer between zinc and the reduced ruthenium species, as already
discussed by Peng et al. [53].
The conversions of benzene and selectivity to cyclohexene in func-
tion of the time for all catalysts are illustrated in Fig. 12a and b. After
10 min, the Ru10ZnAl catalyst reached a maximum yield to cyclo-
hexene (35%). This catalyst reached 100% conversion in approximately
40 min of reaction. RuAl was also active for this reaction but converted
only 20% of the benzene during 180 min of reaction and presented a
maximum yield of 0.2% at 10 min. Ru50ZnAl and RuZn were not active
for this reaction. Suryawanshi et al. [54] obtained a maximum yield to
cyclohexene of 3.0% using a 4%Ru/Al2O3 catalyst in 30 min of reac-
tion. Zhang et al. [55] varied the zinc content in Ru-Zn catalysts sup-
ported in a MCM-41 zeolite and reached a maximum yield to cyclo-
hexene of 3.5% in 35 min.
(Ru0+Ru4+) atomic ratio by XPS after reduction, among the other
reduced catalysts in this work. At first, one can suggest that the higher
the metallic ruthenium dispersion on the surface, better would be the
performance of the catalyst in the partial hydrogenation of benzene,
once hydrogenation reactions require metallic sites in heterogeneous
catalysis. However, Mazzieri et al. [60,61] studied supported ruthe-
nium catalysts by XPS, and observed that other species detected in
higher binding energies than Ru0, ascribed as Ruδ+, contributed to
greater cyclohexene selectivity values. Similar results were reported by
Fan et al. [62], which tested Ru-Co-B supported in alumina catalysts in
partial hydrogenation of benzene. The authors obtained a yield of cy-
clohexene of 28.8% (62.7% conversion and 45.7% selectivity) at 150 °C
and 50 bar H2 after 30 min of reaction but using water (3 mL benzene
and 4 mL water). Fan et al. [62] identified by XPS, that the presence of
cobalt have induced electron transfers between Ru and Co, generating
electron-deficient ruthenium particles that are favorable to achieve
better cyclohexene yields. Both, Mazzieri et al. [61] and Fan et al. [62]
observed electron-deficient Ruδ+ in RuCoB/γ-Al2O3 and Ru/γ-Al2O3,
respectively, and ascribed the presence of this species to an improve-
ment of the selectivity to cyclohexene because cyclohexene is weakly
adsorbed and easily desorbed on electron-deficient ruthenium species.
Thus, the higher activity and cyclohexene selectivity of Ru10ZnAl could
be related to its higher Ruδ+/(Ru0+Ru4+) atomic ratio after reduction.
Zhang et al. [2] reported that an interaction between Ru and Zn
atoms would be due to the ruthenium electronic characteristics. The
difference in electronegativity values between Ru and Zn is consider-
able. Ruthenium is 0.6 eV more electronegative than zinc. Thus, ru-
thenium tends to attract zinc electrons, creating an interaction between
the two metals, which promotes the reduction of Zn2+ particles.
The reaction rates were calculated as described by Cobo et al. [56].
Thus, RuAl catalyst presented a reaction rate of 1.5 mol min−1 gRu−1
and Ru10ZnAl catalyst presented a much higher activity of 1075.7 mols
min−1 gRu−1. The remarkable results obtained with Ru10ZnAl catalyst
were
(1075.7
repeated
twice
with
very
good
reproducibility
111.9 mol min−1 gRu−1). In addition, the catalyst was re-
used and approximately the same activity was obtained (1156.8 mol
min−1 gRu−1) and no Ru leaching was observed by XRF. The Ru50ZnAl
and RuZn catalysts showed no catalytic activity on benzene partial
8

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
Recently, a few theoretical studies using molecular modeling have been
performed in order to outline the effects of the Ru-Zn alloy on the
partial hydrogenation of benzene. Yuan et al. [16,63] determined by
Density Functional Theory (DFT), that zinc acts in two ways: reducing
the ruthenium strength of chemisorbing cyclohexene and causing the
repulsion of hydrogen atoms available to hydrogenate cyclohexene.
These phenomena prevent the hydrogenation of cyclohexene to cyclo-
hexane, increasing the selectivity to cyclohexene. Therefore, the pre-
sence of zinc and its surface species modification is beneficial to the
partial hydrogenation of benzene.
15 (2013) 152–159, https://doi.org/10.1039/c2gc36596k.
[3] J.C.J. Bart, S. Cavallaro, Ind. Eng. Chem. Res. 54 (2015) 1–46, https://doi.org/10.
1021/ie5020734.
[4] Y. Ma, Y. Huang, Y. Cheng, L. Wang, X. Li, Appl. Catal. A: Gen. 484 (2014)
154–160, https://doi.org/10.1016/j.apcata.2014.07.015.
[5] M.C. Shoenmaker-Stolk, J.W. Verwijs, J.A. Don, J.J.F. Scholten, Appl. Catal. 29
(1987) 73–90, https://doi.org/10.1016/S0166-9834(00)82608-1.
[6] H. Kubicka, J. Catal. 12 (1968) 223–237, https://doi.org/10.1016/0021-9517(68)
90102-4.
[7] W. Wang, H. Liu, T. Wu, P. Zhang, G. Ding, S. Liang, T. Jiang, B. Han, J. Mol. Catal.
A Chem. 355 (2012) 174–179, https://doi.org/10.1016/j.molcata.2011.12.013.
[8] E.V. Spinacé, J.M. Vaz, Catal. J. Commun. 4 (2003) 91–96, https://doi.org/10.
1016/S1566-7367(02)00270-4.
[9] H. Nagahara, M. Ono, M. Konishi, Y. Fukuoka, Appl. Surf. Sci. 121/122 (1997)
448–451, https://doi.org/10.1016/S0169-4332(97)00325-5.
[10] J. Struijk, M. d’Angremond, W.J.M. Lucas-de Regt, J.J.F. Scholten, Appl. Catal. A
Gen. 83 (1992) 263–295, https://doi.org/10.1016/0926-860X(92)85039-E.
[11] P.T. Suryawanshi, V.V. Mahajani, J. Chem. Technol. Biotechnol. 69 (1997)
154–160, https://doi.org/10.1002/(SICI)1097-4660(199706)69:2<154::AID-
JCTB670>3.0.CO;2-1.
[12] J. Struijk, R. Moene, T. van der Kamp, J.J.F. Scholten, Appl. Catal. A: Gen. 89
(1992) 77–102, https://doi.org/10.1016/0926-860X(92)80079-R.
[13] F. Mizukami, S. Niwa, S. Ohkawa, A. Katayama, Stud. Surf. Sci. Catal. 78 (1993)
337–344, https://doi.org/10.1016/S0167-2991(08)63338-8.
[14] Asahi Chemical Industry Co. Ltd., US Patent, 4,734,536 (1988).
[15] S.C. Hu, Y.W. Chen, Ind. Eng. Chem. Res. 40 (2001) 6099–6104, https://doi.org/
10.1021/ie0007926.
[16] P. Yuan, B. Wang, Y. Ma, H. He, Z. Cheng, W. Yuan, J. Mol. Catal. A-Chem. 309
(2009) 124–130, https://doi.org/10.1016/j.molcata.2009.05.006.
[17] F. Cavani, F. Trifirò, A. Vaccari, Catal. Today 11 (1991) 173–301, https://doi.org/
10.1016/0920-5861(91)80068-K.
[18] S.K. Sharma, K.B. Sidhpuria, R.V. Jasra, J. Mol. Catal. A: Chem. 335 (2011) 65–70,
https://doi.org/10.1016/j.molcata.2010.11.015.
[19] M. Tu, J. Shen, J. Therm. Anal. Calorim. 58 (1999) 441–446, https://doi.org/10.
1023/A:1010171725592.
[20] A.H. Padmasri, A. Venugopal, V.D. Kumari, K.S. Rama Rao, P.K. Rao, J. Mol. Catal.
A-Chem. 188 (2002) 255–265, https://doi.org/10.1016/S1381-1169(02)00356-4.
[21] H. Fukuhara, F. Matsunaga, Y. Nakashima, Eur. Patent 323 192 (1989), to Mitaui
Petrochem. Ind.
[22] P. Gao, F. Li, H. Zhan, N. Zhao, F. Xiao, W. Wei, L. Zhong, H. Wang, Y. Sun, J. Catal.
298 (2013) 51–60, https://doi.org/10.1016/j.jcat.2012.10.030.
[23] D.L. Carvalho, R.R. Avillez, M.T. Rodriges, L.E.P. Borges, L.G. Appel, Appl. Catal. A:
Gen. 415–416 (2012) 96–100, https://doi.org/10.1016/j.apcata.2011.12.009.
[24] L.C. Meher, R. Gopinath, S.N. Naik, A.K. Dalai, Ind. Eng. Chem. Res. 48 (2009)
1840–1846, https://doi.org/10.1021/ie8011424.
4. Conclusions
Ruthenium supported in mixed oxide synthesized from hydrotalcite
route with different Zn contents were tested in the partial hydrogena-
tion of benzene. Even though the typical hydrotalcite phase was not
observed for the catalyst support precursor with 10 wt.% of ZnO con-
tent, it showed the best performance reaching a maximum yield of 35%
at 10 min and 100% of conversion at approximately 40 min of reaction.
XPS measurements detected that even after reduction, electron defi-
cient ruthenium species (Ruδ+) remained on the surface of the cata-
lysts, which was attributed to the ruthenium-support interaction.
A careful quantification was performed to estimate the surface ru-
thenium species distribution prior to the benzene partial hydrogena-
tion. Ru10ZnAl-RP presented the highest Ruδ+ content proving that,
not only Ru0, but also Ruδ+ are required to increase cyclohexene yields.
In addition, this catalyst showed a Zn2p3/2 main peak displacement to
higher binding energies when reduced, indicating a stronger Ru-Zn
interaction. This phenomenon promoted a partial reduction of the
support (reducing zinc) – detected by XRD and TPR analysis – which
makes the surface of the catalyst more hydrophilic, promoting the cy-
clohexene yields.
The Modified Auger Parameter (MAP) concept by XPS was crucial to
elucidate the surface chemical modification of zinc particles after ru-
thenium impregnation and RP procedure for the Ru10ZnAl catalyst,
confirming the Ru-Zn interaction. However, there was no difference in
MAP values between the 50ZnAl support and Ru50ZnAl-RP. These
observations confirmed that the presence of zinc was a good promoter
for ruthenium catalysts to enhance cyclohexene selectivity in the ben-
zene partial hydrogenation, but only in the presence of Ru-Zn interac-
tion.
[25] F.M.T. Mendes, Introdução à Técnica de Espectroscopia Fotoeletrônica por Raios X,
Synergia, Rio de Janeiro, (2011).
[26] C.D. Wagner, W.M. Rigs, L.E. Davis, J.F. Molder, Handbook of X-Ray
Photoelectronic Spectroscopy, Perkin-Elmer Corporation, Minnesota, 1979.
[27] F.M.T. Mendes, Introdução à Espectroscopia Fotoeletrônica por Raios X-
Tratamento dos Dados Gerados - Tutorial do Software Casaxps, Synergia, Rio de
Janeiro, (2015).
[28] G.K. Chuah, S. Jaenicke, K.S. Chan, Appl. Catal. A: Gen. 145 (1996) 267–284,
https://doi.org/10.1016/0926-860X(96)00152-4.
In summary, XPS analysis showed its powerful capability to provide
understanding of the surface chemical modifications of Ru/Zn/Al cat-
alyst. The XPS results also collaborated to explain how the catalyst with
10%wt ZnO content has shown interesting catalytic performance in the
partial hydrogenation of benzene.
[29] C. Elmasides, D.I. Kondarides, W. Grünert, X.E. Verykios, J. Phys. Chem. B 103
(1999) 5227–5239, https://doi.org/10.1021/jp9842291.
[30] CasaXPS: Processing Software for XPS, AES, SIMS and More, (2018) (Accessed July
19 2019), http://www.casaxps.com.
[31] B. Lin, R. Wang, J. Lin, S. Du, X. Yu, K. Wei, Catal. Commun. 11 (2007) 1838–1842,
https://doi.org/10.1016/j.catcom.2007.02.021.
[32] T. Narita, H. Miura, M. Ohira, H. Hondou, K. Sugiyama, T. Matsuda, R.D. Gonzalez,
Appl. Catal. 32 (1987) 185–190, https://doi.org/10.1016/S0166-9834(00)
80624-7.
Declaration of Competing Interest
[33] J. Okal, M. Zawadzki, Appl. Catal. B: Environ. 89 (2009) 22–32, https://doi.org/10.
1016/j.apcatb.2008.11.024.
[34] M.M.V.M. Souza, D.A.G. Aranda, M. Schmal, J. Catal. 204 (2001) 498–511, https://
doi.org/10.1006/jcat.2001.3398.
[35] P. Betancourt, A. Rives, R. Hubaut, C.E. Scott, J. Goldwasser, Appl. Catal. A: Gen.
170 (1998) 307–314, https://doi.org/10.1016/S0926-860X(98)00061-1.
[36] A. Arcoya, X.L. Seoane, L.M. Gómez-Sainero, Appl. Surf. Sci. 211 (2003) 341–351,
https://doi.org/10.1016/S0169-4332(03)00354-4.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
[37] D. Li, N. Ichikuni, S. Shimazu, T. Uematsu, Appl. Catal. A: Gen. 19 (1999) 227–235,
https://doi.org/10.1016/S0926-860X(98)00335-4.
[38] S.R. Morris, R.B. Moyes, P.B. Wells, R. Whyman, J. Catal. 96 (1985) 23–31, https://
doi.org/10.1016/0021-9517(85)90356-2.
[39] J. Wang, Y. Wang, S. Xiea, M. Qiao, H. Li, K. Fan, Appl. Catal. A: Gen. 272 (2004)
29–36, https://doi.org/10.1016/j.apcata.2004.04.038.
[40] E.L. Foletto, S. Battiston, J.M. Simões, M.M. Bassaco, L.S.F. Pereira, E.M.M. Flores,
E.I. Müller, Microporous Mesoporous Mater. 163 (2012) 29–33, https://doi.org/10.
1016/j.micromeso.2012.06.039.
The author acknowledges the Catalysis Division/INT staff for the
XRD, TPR and FRX analyses, the Center of Characterization in
Nanotechnology for Materials and Catalysis (CENANO)/INT for the XPS
analysis and to CNPq (PCI), CAPES and FAPERJ for the financial sup-
port.
[41] D.J. Morgan, Surf. Interface Anal. 47 (2015) 1072–1079, https://doi.org/10.1002/
References
sia.5852.
[42] K.S. Kim, N. Winograd, J. Catal. 35 (1974) 66–72, https://doi.org/10.1016/0021-
9517(74)90184-5.
[43] W.E. O’Grady, L.J. Atanasoska, F.H. Pollak, H.L. Park, J. Electroanal. Chem. 178
(1984) 61–68, https://doi.org/10.1016/S0022-0728(84)80023-6.
[1] E. Dietzsch, P. Claus, D. Hönike, Top. Catal. 10 (2000) 99–106, https://doi.org/10.
1023/A:1019164001135.
[2] P. Zhang, T. Wu, T. Jiang, W. Wang, H. Liu, H. Fan, Z. Zhang, B. Han, Green Chem.
9

A.H.A. Gonçalves, et al.
MolecularCatalysisxxx(xxxx)xxxx
[44] L.J. Atanasoska, W.E. O’Grady, R.T. Atanasoski, F.H. Pollak, Surf. Sci. 202 (1988)
142–166, https://doi.org/10.1016/0039-6028(88)90066-0.
[45] Y.J. Kim, Y. Gao, S.A. Chambers, Appl. Surf. Sci. 120 (1997) 250–260, https://doi.
org/10.1016/S0169-4332(97)00233-X.
[54] P.T. Suryawanshi, V.V. Mahajani, J. Chem. Technol. Biotechnol. 69 (2) (1997)
154–160, https://doi.org/10.1002/(SICI)1097-4660(199706)69:23.3.CO.
[55] T. Zhang, Z. Wang, Q. Zhao, F. Li, W. Xue, J. Nanomater. (2015) 670896, https://
doi.org/10.1155/2015/670896.
[46] H. Over, A.P. Seitsonen, E. Lundgren, M. Smedh, J.N. Andersen, Surf. Sci. 504
(2002) L196–L200, https://doi.org/10.1016/S0039-6028(01)01979-3.
[47] K. Reuter, M. Scheffler, Surf. Sci. 490 (2001) 20–28, https://doi.org/10.1016/
S0039-6028(01)01214-6.
[56] R.S. Supino, R. Landers, A.J.G. Cobo, Appl. Catal. A: Gen. 452 (2013) 9–16, https://
doi.org/10.1016/j.apcata.2012.11.034.
[57] C. Milone, G. Neri, A. Donato, M.G. Musolino, L. Mercadante, J. Catal. 159 (2)
(1996) 253–258, https://doi.org/10.1006/jcat.1996.0086.
[48] Y.V. Larichev, B.L. Moroz, I.P. Prosvirin, V.A. Likholobov, I. Bukhtiyarov, Chem.
[58] L. Foppa, J. Dupont, Chem. Soc. Rev. 44 (2015) 1886–1897, https://doi.org/10.
Sustain. Dev. 11 (2003) 155–160.
1039/c4cs00324a.
[49] W. Grünert, X ray photoelectron spectroscopy, in: B.N. Weckhuysen, P. Van der
Voort, G. Cartana (Eds.), Spectroscopy of Transition Metals Ions on Surface, 2000,
pp. 269–304 chap. 5.
[59] J. Struijk, M. d’Angremond, W.J.M. Lucas-de-Regt, J.J.F. Scholten, Appl. Catal., A:
Gen. 83 (1992) 263–295, https://doi.org/10.1016/0926-860X(92)85039-E.
[60] V. Mazzieri, F. Coloma-Pascual, A. Arcoya, P.C. L’Argentière, N.S. Fígoli, Appl. Surf.
Sci. 210 (2003) 222–230, https://doi.org/10.1016/S0169-4332(03)00146-6.
[61] V. Mazzieri, N.S. Fígoli, F. Coloma-Pascual, P.C. L’Argentière, Catal. Lett. 102
(2005) 79–82, https://doi.org/10.1007/s10562-005-5206-6.
[50] C.D. Wagner, L.H. Gale, R.H. Raymond, Anal. Chem. 51 (1979) 466–482, https://
doi.org/10.1021/ac50040a005.
[51] C.D. Wagner, A. Joshi, J. Electron Spectrosc. 47 (1988) 283–313, https://doi.org/
10.1016/0368-2048(88)85018-7.
[62] G. Fan, W. Jiang, J. Wang, R. Li, H. Chen, X. Li, Catal. Commun. 10 (2008) 98–102,
https://doi.org/10.1016/j.catcom.2008.08.002.
[52] B.R. Strohmeier, D.M. Hercules, J. Catal. 86 (2) (1984) 266–279, https://doi.org/
10.1016/0021-9517(84)90372-5.
[53] Z. Peng, X. Liu, H. Lin, Z. Wang, Z. Li, B. Li, Z. Liu, S. Liu, J. Mater. Chem. A 4
(2016) 17694–17703, https://doi.org/10.1039/C6tA08529F.
[63] P. Yuan, B. Wang, Y. Ma, H. He, Z. Cheng, W. Yuan, J. Mol. Catal. A-Chem. 301
(2009) 140–145, https://doi.org/10.1016/j.molcata.2008.11.028.
10