Carbon support treatment effect on Ru/C catalyst performance for benzene partial hydrogenation

Article abstract of DOI:10.1016/j.apcata.2011.10.001

Ru/C catalysts were prepared from commercial activated carbon submitted to different treatments. The catalysts were prepared by incipient wetness impregnation, through an aqueous solution of the precursor RuCl ₃·xH₂O. After impregnation, some catalysts were submitted to direct reduction treatment under H₂ flow at the temperature of 150 °C, in order to evaluate the effects of activation. The supports were characterized by N₂ adsorption, Boehm and potentiometric titration. The X-ray photoelectron spectroscopy was used to study the supports and catalysts surfaces, while scanning electron microscopy allowed us to determine the chemical composition and observe the catalysts morphology. Ru/C catalysts performance was evaluated within the hydrogenation reaction of benzene in liquid phase, using a Parr reactor. The reaction was conducted under total pressure of 5.0 MPa of H₂, at a temperature of 100 °C with water in the reaction medium. The obtained results indicate that the Ru/C system catalytic performance is influenced for determined functional groups present on the activated carbon surface. The carbonyl groups decrease the activity and selectivity of the reactions, while an increase of the carboxylic groups leads to more active catalysts and the highest yield of cyclohexene.

Full text of DOI:10.1016/j.apcata.2011.10.001

Applied Catalysis A: General 409–410 (2011) 174–180
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Carbon support treatment effect on Ru/C catalyst performance for benzene
partial hydrogenation
Cristiane Zanuteloa,∗, Richard Landersb, Wagner Alves Carvalhoc, Antonio José Gomez Coboa
a Laboratory for Development of Catalytic Processes, Department of Chemical Systems Engineering, School of Chemical Engineering, University of Campinas, Cidade Universitária
Zeferino Vaz, CEP 13083-852, Campinas, SP, Brazil
b Laboratory of Surface Physics, Department of Applied Physics, ‘Gleb Wataghin’ Institute of Physics, University of Campinas, CEP 13083-970, Campinas, SP, Brazil
c Centro de Ciências Naturais e Humanas, Universidade Federal do ABC, Rua Santa Adélia, 166, CEP 09210-170, Santo André, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Ru/C catalysts were prepared from commercial activated carbon submitted to different treatments. The
catalysts were prepared by incipient wetness impregnation, through an aqueous solution of the precursor
RuCl3·xH2O. After impregnation, some catalysts were submitted to direct reduction treatment under H2
flow at the temperature of 150 ◦C, in order to evaluate the effects of activation. The supports were char-
acterized by N2 adsorption, Boehm and potentiometric titration. The X-ray photoelectron spectroscopy
was used to study the supports and catalysts surfaces, while scanning electron microscopy allowed us to
determine the chemical composition and observe the catalysts morphology. Ru/C catalysts performance
was evaluated within the hydrogenation reaction of benzene in liquid phase, using a Parr reactor. The
reaction was conducted under total pressure of 5.0 MPa of H2, at a temperature of 100 ◦C with water in
the reaction medium. The obtained results indicate that the Ru/C system catalytic performance is influ-
enced for determined functional groups present on the activated carbon surface. The carbonyl groups
decrease the activity and selectivity of the reactions, while an increase of the carboxylic groups leads to
more active catalysts and the highest yield of cyclohexene.
Received 21 May 2011
Received in revised form
24 September 2011
Accepted 1 October 2011
Available online 7 October 2011
Keywords:
Hydrogenation
Benzene
Ruthenium
Activated carbon
Cyclohexene
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
other catalytic supports. Nevertheless, the formation of active sites,
constituted by heteroatoms (O, N and H) is responsible for the
The partial hydrogenation of benzene is a chemical reaction
employed in the productive sector and environmental protection.
The cyclohexene obtained can be industrially used for several syn-
theses of compounds, as for instance nylon. Moreover, due to severe
restrictions regarding the presence of aromatic toxic compounds in
oil-derivative fuels, this reaction represents a promising alternative
treatment, without significant loss of octane [1]. Hydrogenation
reaction of benzene characteristically occurs in successive and
exothermic (benzene → cyclohexene → cyclohexane) stages and it
is highly favorable for cyclohexane production.
Therefore, through kinetic strategies higher selectivity can be
obtainedfor theintermediaryproduct. In suchreaction, ruthenium-
based catalysts have led to higher yields of cyclohexene [2,3],
especially in the presence of water in the reaction medium [4].
Although some studies highlight a better performance of ruthe-
nium mass catalysts [5], researches are still being carried out to
develop supported catalysts, aiming at reducing their costs and
exploring possible potentialities of metal–support interaction [6].
Activated carbon is considered an inert material if compared to
acidity or basicity of the solid, as well as for their redox prop-
erties [7]. Particularly, the presence of oxygen on the carbon
surface may influence the catalyst hydrophilicity and, as a conse-
quence, influence the yield of the desired product, affecting the
precursor–support interaction during the catalyst preparation. In
liquid reaction medium, water and hydrophilic catalysts have been
widely utilized to inhibit the hydrogenation of the formed cyclo-
hexene [8]. The main role of water is to withdraw cyclohexene
from the surface of the hydrophilic catalyst, reducing its undesired
hydrogenation. By studying benzene hydrogenation in liquid state,
applying Pt catalysts prepared from activated carbon modified with
nitric acid and N2, Aksoylu et al. [9] noticed that such treatments
led to more active and selective catalysts. In this context, the pur-
pose of this article is to study the influence of treatment applied
to support in the performance of Ru/C catalysts to partial benzene
hydrogenation in liquid state.
2. Experimental
2.1. Preparation of supports and catalysts
∗
This study used a commercial carbon (Clarimex 061) pro-
duced by Brascarbo Agroindustrial Ltd. This solid, denominated
Corresponding author. Tel.: +55 19 35210398; fax: +55 19 3521 3894.
E-mail address: crisvox@feq.unicamp.br (C. Zanutelo).
0926-860X/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.10.001

C. Zanutelo et al. / Applied Catalysis A: General 409–410 (2011) 174–180
175
Table 1
Table 2
Treatments applied to original carbon C(OR).
Activated carbon textural characteristics.
Support
Treatment
Support
Sg (m2/g)
Pv (cm3/g)
Micro
C(N2/300)
Heating at 10 ◦C/min under a flow of 250 mL of N2/min up
to the temperature of 300 ◦C, maintained for 7 h
Meso
Total
C(OR)
900
0.13
0.10
0.13
0.16
0.09
0.08
0.14
0.008
0.63
0.59
0.60
0.58
0.27
0.28
0.56
0.082
0.76
0.69
0.73
0.74
0.36
0.36
0.70
0.09
C(N2/400)
C(HN25)
Heating at 10 ◦C/min, under a flow of 250 mL of N2/min, up
to the temperature of 400 ◦C, maintained for 7 h
C(N2/300)
C(N2/400)
C(HN25)
C(HN90)
C(HN90-N2/300)
C(H2O2)
790
870
910
480
460
870
70
Oxidation in aqueous solution of 2 mol of HNO3/L, under
stirring at 25 ◦C for 3 h. Washing in distilled water until
reaching neutral pH and drying at 100 ◦
C
C(HN90)
Oxidation in aqueous solution of 8.6 mol of HNO3/L, under
reflux at 90 ◦C for 20 h. Washing in distilled water until
C(KOH)
reaching neutral pH and drying at 100 ◦
C
C(HN90-N2/300) Oxidation in aqueous solution of 8.6 mol of HNO3/L, under
reflux at 90 ◦C for 20 h. Washing in distilled water until
reaching neutral pH and drying at 100 ◦C. Heating at
10 ◦C/min, under a flow of 250 mL of N2/min, up to the
temperature of 300 ◦C, maintained for 7 h
Total = micropores + mesopores.
2.3. Catalytic tests
C(H2O2)
Oxidation in aqueous solution of 2 mol of H2O2/L, under
Benzene hydrogenation reactions were performed in a slurry
type Parr reactor with total capacity of 300 mL. A 200 mg catalyst
mass was introduced into the reactor, along with 30 mL of water
and 5 mL of n-heptane as internal standard. The catalyst activation
was performed in situ, at 100 ◦C, under flow of H2 for 45 min. Next,
the reactor was charged at 5 MPa of working pressure, and 25 mL
of benzene were introduced into the reaction medium. During the
reaction, a pressure accumulator tank and regulator valve permit-
ted the maintenance of a constant hydrogen pressure inside the
reactor. The velocity of 1.500 rpm was adopted for being consid-
ered sufficient to avoid mass transport limitation and the reaction
was conducted at the constant temperature of 100 ◦C. During the
reaction, organic phase aliquots were drawn (around 6 samples per
test) through gas phase chromatography, in order to determine its
composition. Chromatographic analyses used CG 151-40 capillary
column of 25 m length in phase FI-53 of polydimethylsiloxane, con-
nected to a chromatograph with flame ionization detector, model
HP 5890.
stirring at 25 ◦C for 3 h Washing in distilled water until
reaching neutral pH and drying at 100 ◦
C
C(KOH)
Oxidation in aqueous solution of 16 mol of KOH/L, under
stirring maintained for 10 min, at room temperature.
Drying at 100 ◦C for 14 h, followed by heating at 110 ◦C for
14 h, under flow of 150 mL of N2/min up to 300 ◦C,
maintained for 1 h. Washing in aqueous solution of 0.5 mol
HCl/L and later in hot water. Washing in distilled water at
room temperature, until reaching neutral pH, and drying at
100 ◦C [10]
original carbon C(OR), was submitted to the treatments described in
Table 1.
Ru/C catalysts were prepared by applying the incipient impreg-
nation, using an aqueous solution of RuCl3·xH2O from Aldrich
Chem. Co., so as to obtain a metal mass fraction of 5 wt%. After the
impregnation, the solids were left at room temperature for about
12 h, and dried in a heater at 80 ◦C for 24 h. Next, some of the cat-
alysts were submitted to direct reduction treatment at 150 ◦C for
3 h, under a flow of a N2/H2 (98%/2%) mixture.
The initial reaction rate was determined from benzene con-
sumption. Benzene conversion (X) was calculated according to its
classical definition, while yield of cyclohexene (R) was calculated
considering the following definition:
2.2. Characterization of supports and catalysts
No of mols of formed cyclohexene
R =
N2 adsorption, Boehm titration and point of zero charge (PZC)
were utilized to characterize the supports, while X-ray photoelec-
tron spectroscopy (XPS) was applied to characterize supports and
catalysts.
No of initial mols of benzene
Thus, cyclohexene selectivity (S) can be obtained through the
relation S = R/X.
N2 adsorption (Brunauer, Emmet, Teller method) was per-
formed at its normal boiling temperature (−196 ◦C), in
a
3. Results and discussion
Micromeritics Tristar dynamic equipment model ASAP 2010. The
adjustment of experimental points was performed by employing
the classic BET model.
3.1. Carbon textural characteristics
Based on neutralization (acid–base) [11], Boehm titration was
used to quantify some of the functional groups existing on carbon
surface.
PZC, defined as the pH at which carbon surface has neutral
charge, was determined by the potentiometric titration, utilizing
a digital Analyser pHmeter model 300M.
Table 2 shows the values for specific surface area (Sg) and pore
volume (Pv) obtained for the studied carbons.
With regard to the reference original carbon C(OR), the achieved
results indicate that treatments performed with nitric acid at 90 ◦C
and with potassium hydroxide lead to a significant reduction of the
carbon specific area, especially in the base. Such effect comes along
with an accentuated reduction of the total pore volume. According
to Castilla et al. [12] and Castelló et al. [13], a severe oxidation may
destroy the carbon porous structure, considerably increasing the
number of acid groups and making the basic groups disappear.
According to Castilla et al. [12], treatment with nitric acid may to
lead to carbons with pore of walls thinner and therefore more eas-
ily destroyed. According to these authors, two mechanisms can act
in the process of breaking of the walls of the pores, after treatment
with a strong oxidizing agent. The first way is through the oxida-
tion of surface terminal groups, where a decrease in the area of the
coal may occur due the formation of acid mellitus [C6(COOH)6]. The
second mechanism occurs through mechanical destruction of the
XPS spectra were obtained with a HA100 VSW hemispherical
analyzer, operating in stable transmission mode (44 eV transition
energy). Al K␣ radiation (1486.6 eV) was applied and the chamber
pressure was lower than 2 × 10−8 mbar.
The scanning electron microscopy with microprobe for spec-
trometry of Energy Dispersive X-ray analysis (SEM + EDX) was
used aiming at quantifying the main elements found in the cata-
lysts as well as to observing their morphology. The analyses were
performed with a Leica device model LEO 440i. The initial stage
consisted of metalizing the catalyst to form a thin layer of gold
˚
atoms on it with 92 A film thickness. The analyses were performed
by employing 3 MA current for 180 s.

176
C. Zanutelo et al. / Applied Catalysis A: General 409–410 (2011) 174–180
Table 3
Table 4
Distribution of surface functional groups.
pH at point of zero charge (PZC).
Support
Concentration of surface functional groups (meq/g)
Support
pH at PZC
C(OR)
2.75
2.40
2.44
2.13
1.87
2.35
2.25
8.50
CX
L
Ph
CN
C(N2/300)
C(N2/400)
C(HN25)
C(HN90)
C(HN90-N2/300)
C(H2O2)
C(OR)
0.58
0.57
0.49
0.77
1.60
0.92
0.99
0.38
0.03
0.13
0.03
0.05
0.30
0.35
0.11
0.92
0.49
0.39
0.71
0.58
0.52
0.98
0.50
0.74
1.4
1.2
1.3
1.2
0.46
0.36
0.98
0.45
C(N2/300)
C(N2/400)
C(HN25)
C(HN90)
C(HN90-N2/300)
C(H2O2)
C(KOH)
C(KOH)
Cx = carboxylic groups; L = lactones; Ph = phenolic groups; CN = carbonyls.
Determining pH at PZC through potentiometric titration of the
different supports allows the identification of changes on the orig-
inal carbon surface. In amphoteric carbon, such as the original
carbon C(OR), the surface is positively charged when pH < PZC and
negatively charged when pH > PZC.
pores due to surface tension of the oxidant solution. In the case
of carbon treated with KOH, pores with thinner walls may have
been originated from the breaking of bonds in the matrix carbonic
with rearrangement of the aggregates, leading to the collapse and
destruction of pores. In addition, the use of a high amount of KOH
in the preparation of activated carbon may inhibit the formation
of CO2 and of potassium compounds, preventing the gasification
of carbon with the consequent formation of porosity [13], usually
observed for low concentrations of base.
The achieved results for the different carbon supports are shown
in Table 4.
Prepared carbons show lower PZC values than the original one
C(OR), except for C(KOH) solid. The carbons obtained through nitric
acid, C(HN25) and C(HN90) treatment show the lowest values for
pH of PZC; the value diminishes as the acid treatment temperature
increases. In the case of carbon prepared with potassium hydroxide,
C(KOH), there is an elevated increase of pH value of PZC, which
indicates a strong basic characteristic of this support.
On the other hand, the treatments applied barely affected the
textural characteristics of the original carbon C(OR).
Results obtained for pH of PZC lead to the following
3.2. Characteristics of carbon surface
acidity
order:
C(HN90) > C(HN25) > C(H2O2) > C(HN90-
N2/300) > C(N2/300) ≈ C(N2/400) > C(OR) > C(KOH).
Table 3 shows the distribution of surface functional groups
determined by Boehm titration, as verified in the carbons studied
here. The results indicate that adsorptive carbons have acid sur-
face functions, in the form of non-carbonylic (carboxylic groups,
lactones and phenolic groups) and carbonylic (carbonyls) groups.
The level of acid strength of acid functional groups is ordered as
follows: carboxylics > lactones > phenolic groups > carbonyls.
With reference to exclusively thermal processes (physical), the
comparison between the applied treatments show that wet acti-
vation processes (chemical) create oxygenated groups in a more
uniform way on the activated carbon surface, which means that
the quantity of non-carbonylic groups (carboxylic groups, lactones
and phenolic groups) is very similar to the quantity of carbonylic
groups. Among wet activation process, the nitric acid treatment
at 90 ◦C leads to the acquisition of the highest quantity of non-
carbonylic acid sites on the carbon surface (2.42 meq/g).
The acidity of the different functional groups of carbon is not
only a function of the surface group nature (pKa fixed value), but
also of their carbon chain position. Usually, such groups facilitate
the adsorption process, contributing to the increase in the adsor-
bent material hydrophilicity [14]. In addition, they can promote
ruthenium dispersion on the activated carbon, leading to more
active catalysts Fig. 1.
(a)
1
1,4x10
1,2x10
1,0x10
8,0x10
6,0x10
4,0x10
C(OR) C1s
1
1
0
0
0
0
2,0x10
0,0
0
-2,0x10
300
298
296
294
292
290
288
286
284
282
280
binding energy (eV)
1
1,6x10
1,4x10
1,2x10
1,0x10
8,0x10
6,0x10
4,0x10
2,0x10
(b)
1
1
1
0
0
0
0
C(HN90) C1s
0,0
0
-2,0x10
300
298
296
294
292
290
288
286
284
282
280
Binding energy (eV)
Fig. 1. Thermodynamics aspects of the benzene hydrogenation reaction.
Fig. 2. XPS spectra for C 1s for activated carbon C(OR) and C(HN90).

C. Zanutelo et al. / Applied Catalysis A: General 409–410 (2011) 174–180
177
(a)
1,8x101
1,6x101
1,4x101
1,2x101
1,0x101
8,0x100
6,0x100
4,0x100
Ru/C(OR)NR
C1s+Ru3d
2,0x100
0,0
-2,0x100
300
295
290
285
280
275
Binding energy (eV)
(b)
1,2x101
1,0x101
8,0x100
6,0x100
4,0x100
Ru/C(OR)R150
C1s+Ru3d
2,0x100
0,0
300
295
290
285
280
275
Binding energy (eV)
Fig. 3. The spectra of XPS for C 1s and Ru 3d for Ru/COR)NR and Ru/C(OR)R150 catalysts.
XPS analyses performed for supports C(OR), C(N2/300) and
C(HN90) reveal the main occurrence of graphitic carbon (284.7 eV),
correspondent to non-functional carbon, Csp2 and Csp3, being Csp2
the carbon from the aromatic chain and Csp3 the aliphatic carbon in
hydrocarbonates. The existence of phenolic groups (282.6 eV), car-
bonyls (287.6 eV), carboxylic groups (289.1 eV), and nitrogenated
groups bonded to carbon and electron ␲ bonds in aromatic car-
bons (291.0 eV) is also noticed [15]. Quantifying the distribution
of elements C, O and N was based on the intensity of characteris-
tic peaks of the elements present and feature sensitivity factors (C
1s = 1.00; O 1s = 2.850; N 1s = 1.770).
accordance with the increase in quantity of lactones observed
through Boehm titration.
However, such increase may not be detected through XPS anal-
ysis because the characteristic binding energies of lactone and
phenolic groups are very similar. Nevertheless, the performed
characterization demonstrates the conclusion that the treatments
applied to the original carbon C(OR) change the solid surface, pro-
moting an increase in the concentration of the surface functional
groups.
3.3. Chemical composition of the catalysts
The spectra of XPS for C1 s for the activated carbons C(OR) and
C(HN90) are shown in Fig. 2(a and b).
XPS analyses were also performed with the following catalysts:
Ru/C(OR)NR, Ru/C(OR)R150, Ru/C(HN25)NR and Ru/C(HN90)NR.
The results obtained for ruthenium in non-reduced (282 eV)
Ru/C(OR) catalyst suggest that this metal is in the form of RuCl3
(281.8 eV). The spectra of XPS for C 1s and Ru 3d for the different
catalysts are shown in Figs. 3 and 4(a and b).
In the case of C(OR) and C(N2/300) supports, there is no occur-
rence of nitrogenated compounds found in support C(HN90). In this
solid, nitrogen is formed as NO2 groups and nitrogenated functions.
Such result is in accordance with what Salame and Bandosz [14–16]
observed – the study verified the presence of nitronium and nitrate
ions on the surface of the carbon prepared with nitric acid.
Furthermore, results reveal that support C(N2/300) does not
show significant changes on its surface in comparison to the
original carbon C(OR). At first impression, this would not be in
However, in catalyst Ru/C(OR) reduced to 150 ◦C, ruthenium
shows a binding energy of 281.6 eV, very close to this element
in stages RuO2 (281.5 eV) and RuCl3. In this case, the absence of
Ru0 (280.0 eV) is due to a limited oxidation of this element in the

178
C. Zanutelo et al. / Applied Catalysis A: General 409–410 (2011) 174–180
(a)
2,5x101
2,0x101
1,5x101
1,0x101
Ru/C(HN25)NR
C1s+Ru3d
5,0x100
0,0
300
295
290
285
280
275
Binding energy(eV)
(b)
1,0x101
8,0x100
6,0x100
4,0x100
Ru/C(HN90)NR
C1s+Ru3d
2,0x100
0,0
300
295
290
285
280
275
Binding energy (eV)
Fig. 4. The spectra of XPS for C 1s and Ru 3d for Ru/C(HN25)NR and Ru/C(HN90)NR catalysts.
surface of the metallic particle, induced by the catalyst exposure to
air [17].
supports treated with nitric acid (HNO3) at 25 ◦C and 90 ◦C, the
quantity of chlorine diminishes as the treatment temperature
rises.
Drastic reduction of chlorine level in Ru/C(HN90) catalyst may
be due to the presence of a great quantity of carboxylic groups
and different nitrogenated groups. In the absence of chlorine, the
functional surface groups are favorable to the maintenance of the
coordinated ruthenium.
For Ru/C catalysts whose supports were treated with nitric acid,
C(HN25) and C(HN90), the XPS results suggest that metal ruthe-
nium is in the forms of RuO2 and/or RuO3 (282.5 eV). As a matter
of fact, both catalysts show different characteristics from the other
analyzed solids, especially for the nitrogen present in two distinct
types of nitrogenated functional groups (around 404.5 eV).
The presence of these two nitrogenated functional groups may
contribute to an increase in the support acidity and promote the
formation of Ru–NO2 bond, modifying ruthenium behavior on the
catalyst surface.
Table 5
Global atomic ratios (EDX) and surface atomic ratios (XPS) to Ru/C catalysts.
For the existing elements in the analyzed catalysts, the atomic
ratios of the surface were calculated from the XPS spectra, exper-
imentally obtained to the energy state of Ru 3d5/2, Cl 2p and C 1s.
The results are shown in Table 5, along with global values obtained
by EDX, for comparison purpose.
The obtained atomic ratios confirm a fine concordance between
EDX and XPS analysis results. The results indicate that reduc-
tion treatment lessens the chlorine level in Ru/C(OR) catalyst
[18]. In the case of non-reduced Ru/C systems, prepared from
Catalyst
Atomic ratio
Cl/Ru
Ru/C
EDX
EDX
XPS
XPS
Ru/C(OR)NR
3.1
2.9
n.d.
n.d.
3.3
2.2
3.8
0.15
0.0061
0.074
n.d.
0.0063
0.0085
0.0039
0.051
Ru/C(OR)R150
Ru/C(HN25)NR
Ru/C(HN90)NR
n.d.
n.d. = not determined; NR = non-reduced; R150 = reduced to 150 ◦C.

C. Zanutelo et al. / Applied Catalysis A: General 409–410 (2011) 174–180
179
Table 6
Values of rate constants for the hydrogenation reaction of benzene with Ru/C cata-
lysts non-reduced.
Catalyst
k1 (m s−1
)
k2 (m s−1
)
k3 (m s−1
)
Ru/C(OR)NR
24
28
29
30
35
12
18
33
36
37
32
75
88
14
16
50
220
100
120
70
74
31
Ru/C(N2/300)NR
Ru/C(N2/400)NR
Ru/C(HN25)NR
Ru/C(HN90)NR
Ru/C(HN90-N2/300)NR
Ru/C(H2O2)NR
Ru/C(KOH)NR
54
110
of carboxylic groups defined for this support, which is consistent
with that seen in Figs. 6 and 7.
Studies conducted with activated carbons [20] showed that sev-
eral acidic carbon–oxygen groups provide a hydrophilic character
of carbon and reduce the adsorption of benzene, which is non-
polar compound. When acidic surface groups are removed, and
carbonyl functional groups and quinones predominate, the adsorp-
tion of benzene increases. A greater adsorption of benzene is due
to the interaction between the ␲ electron cloud of the aromatic
ring of benzene and the partial positive charge of the carbonyl car-
bon atom. Complexes formed between benzene and carbonyl and
quinone sites occur within the micropores of the carbon. Thus, the
benzene present in the reaction medium can be easily adsorbed on
Fig. 5. Yield of cyclohexene throughout the reaction with non-reduced Ru/C cata-
lysts.
3.4. Catalytic performances in benzene hydrogenation
Ru/C catalysts were evaluated with the purpose of study-
ing the influence of treatment applied to support in the
catalytic performance in hydrogenation reaction of liquid
benzene.
When Ru/C(OR) catalyst was reduced to 150 ◦C, a decrease on
cyclohexene performance was noticed throughout the reaction,
in relation to the non-reduced catalyst. Such effect is due to the
elimination of chloride by reduction treatment, since the pres-
ence of chloride may enhance the hydrophilic characteristic of the
solid [19]. Therefore, the catalytic tests were conducted with non-
reduced catalysts.
Results demonstrated in Fig. 5 indicate that all Ru/C catalysts,
whose supports were treated, present higher cyclohexene yields
than the catalyst prepared with the original carbon Ru/C(OR).
The increased cyclohexene yields can be explained based on an
increase of the catalysts hydrophilicity, due to the higher quantity
of surface oxygenated groups, formed by the treatments applied to
the support, according to the results of carbon characterization.
In the case of supports treated with nitric acid, the obtained cat-
alysts, Ru/C(HN25) and Ru/C(HN90), show a similar behavior with
a high yield of cyclohexene for low conversions. Such differenti-
ated behavior is related to the presence of nitrogenated functional
groups formed by the treatment with nitric acid.
Fig. 6. Relationship between cyclohexene maximum yield and total quantity of
carbonylic functional groups on the carbon surface.
It is also important to observe the high yields obtained with
Ru/C(KOH) catalyst, which may be related to the low specific area
of the support [19].
As can be seen in Fig. 1, the constant k1 is related to the rate of
partial hydrogenation of benzene and, consequently, the formation
of cyclohexene. In turn, the parameter k2 is associated with the rate
of hydrogenation of cyclohexene, for the formation of cyclohexane.
Since k3 is a constant rate on the direct hydrogenation of benzene
to form cyclohexane.
Table 6 gives the values of the constants k1, k2 and k3, obtained
from the catalytic tests performed, considering a first-order reac-
tions.
The results showed that the catalyst Ru/C(OR)NR, prepared with
the original untreated carbon, has the higher value for the con-
stant k3 (220 m s−1), which is much larger than the constants k1
(24 m s−1) and k2 (36 m s−1). This result indicates a high rate of
direct hydrogenation of benzene to cyclohexane, to the detriment
of the consecutive hydrogenation. Such behavior may be due both
to the large amount of carbonyl groups, and the low concentration
Fig. 7. Relationship between cyclohexene maximum yield and total number of car-
boxylic functional groups on the carbon surface.

180
C. Zanutelo et al. / Applied Catalysis A: General 409–410 (2011) 174–180
the catalyst surface, leading to the direct hydrogenation of benzene
to cyclohexane.
in the macropore volume, favoring the diffusion of the cyclohex-
ene formed in the pores and, consequently, avoiding its undesired
hydrogenation at cyclohexane.
In turn, in the case of the catalysts Ru/C(N2/300)NR,
Ru/C(N2/400)NR and Ru/C(KOH)NR the constants k1 and k2
have values close to those obtained for the catalyst reference
Ru/C(OR)NR. However, a lower value for the constant k3 for these
catalysts shows a decrease in the rate of direct hydrogenation of
benzene, leading to a higher yield of cyclohexene. However, it
should be noted that in the case of the catalyst Ru/C(KOH)NR, the
highest selectivity of cyclohexene can also be due to lower spe-
cific surface area of solid. According to Rodrigues and Cobo [19], Ru
catalysts with low surface area are more selective for the partial
hydrogenation of benzene in the liquid phase.
For the catalysts treated with HNO3, Ru/C(HN25)NR and
Ru/C(HN90)NR, k1 values remain around 30 m s−1. However, in
relation to previous cases, the values of k2 (about 80 m s−1) increase,
while k3 (around 70 m s−1) decreases. This result indicates an
increased rate of partial hydrogenation of benzene at the expense
of its direct hydrogenation to cyclohexane.
Since the catalysts Ru/C(HN90-N2/300)NR and Ru/C(H2O2)NR
had the lowest values for the constants determined. This result
indicates a catalytic activity lower than that of other catalysts. How-
ever, the average value obtained for the relative selectivity between
the formation of cyclohexene and cyclohexane, as reflected by the
ratio k1/(k2 + k3), is around 0.26. This value is much higher than in
the case of the solid reference Ru/C(OR)NR (0.09) and greater than
the calculated for the other catalysts (around 0.20). The increase in
selectivity at the expense of catalytic activity, suggests an increased
hydrophilic character of these catalysts, induced by the treatments
applied to the supports. In the system under study, the water added
to the reaction medium has the role of the cyclohexene formed
away from the catalytic surface, preventing the undesired hydro-
genation to cyclohexane. In this case, a catalyst more hydrophilic
is more easily engaged by a layer of water, which decreases the
rate of mass transfer of benzene to the surface of the catalyst and
therefore decreases the reaction rate observed.
With regard to catalytic activity, it was observed that the
increase in the quantity of carboxylic groups tends to increase the
initial reaction rate. Such effect is associated with an increase of
ruthenium dispersion in the catalyst. As noted by Zhu et al. [21,22],
ruthenium dispersion increases according to the quantity of car-
boxylic surface groups. Moreover, highly acid supports favor the
catalytic activity of this reaction [23,24]. Adsorbed benzene in acid
active sites of the support may also react with the adsorbed hydro-
gen on the metallic surface and consequently raise the catalytic
activity. For this reason, supports with great quantity of carboxylic
groups (strong acids) on their surface may contribute to a higher
catalytic activity in benzene hydrogenation reaction.
4. Conclusions
The achieved results in the present study reveal that treatments
applied to activated carbon may influence the performance of Ru/C
catalysts in liquid benzene partial hydrogenation reaction.
The catalytic performance of Ru/C system is influenced by cer-
tain functional groups existing on the surface of the activated
carbon. Carbonyl groups decrease the reaction activity and selectiv-
ity, whereas an increase of carboxylic groups leads to more active
catalysts and higher yields of cyclohexene.
Acknowledgments
To the State of São Paulo Research Foundation (FAPESP)
(2006/04142-0) and to the Coordination for Improvement of Higher
Level Personnel (CAPES) for the financial support and for the schol-
arship granted to Cristiane Zanutelo.
References
[1] E. Diestzsch, P. Claus, D. Honicke, Top. Catal. 10 (1–2) (2000) 99–106.
[2] V. Mazzieri, F.P. Coloma, M. Gonzalez, P. L‘Argentieri, N. Fígoli, React. Kinet.
Catal. Lett. 76 (1) (2002) 53–59.
[3] J. Bu, J.L. Liu, X.Y. Chen, J.H. Zhuang, S.R. Yan, M.H. Qiao, H.Y. He, K.N. Fan, Catal.
Commun. 9 (2008) 2612.
[4] J.W. Silva, A.J.G. Cobo, Appl. Catal. A: Gen. 252 (1) (2003) 9–16.
[5] G.Y. Fan, R.X. Li, J.X. Li, H. Chen, Catal. Commun. 9 (2007) 1394.
[6] M. Soede, E.J.A.X. Van de Sandt, M. Makkee, J.J.F. Scholten, Heterogen. Catal.
Fine Chem. III (1993) 345–352.
[7] A.G. Moura, Influence of lanthanide and titanium in the partial hydrogena-
tion of benzene with the catalyst based on ruthenium, PhD thesis, University
Campinas, 2004.
[8] J. Struijk, M. d‘Angremond, W.J.M.L. d‘Legt, J.J.F. Scholten, Appl. Catal. A: Gen.
89 (2) (1992) 263–295.
[9] A. Aksoylu, M. Freitas, J.L. Figueiredo, Appl. Catal. A: Gen. 192 (1) (2000) 29–42.
[10] A. Ahmadpour, D.D. Do, Carbon 34 (4) (1996) 471–479.
[11] H.P. Boehm, Carbon 40 (2) (2001) 145–149.
[12] C.M. Castilla, M.V.R. López, M. Carrasco, Carbon 38 (4) (2000) 1995–2001.
[13] J.M.D.L. Castelló, D.C. Amorós, A.L. Solano, Carbon 45 (2007) 2529.
[14] I.I. Salame, T.J. Bandosz, J. Colloid Interface Sci. 240 (1) (2001) 252–258.
[15] A.P. Terzik, G. Rychlicki, S. Biniak, J.P. Lukaszewicz, J. Colloid Interface Sci. 257
(2003) 13.
3.5. Influence of the functional groups in the Ru/C catalysts
performance
Aiming at evaluating the influence of the different functional
oxygenated groups, found on the surface of activated carbon, on the
performance of Ru/C; the establishment of a relation between the
quantity of these groups and the catalytic performance in benzene
hydrogenation reaction was realized.
Fig. 6 represents the achieved relation between cyclohexene
maximum yield and the total quantity of carbonylic groups on the
carbon surface area.
In this case, results show that the increase in the number of car-
bonylic functional groups tends to reduce cyclohexene maximum
yield. The same effect was observed in catalytic activity, indicating
that a great number of carbonylic groups may disfavor ruthenium
dispersion on the support, leading to less active catalysts.
Fig. 7 shows the achieved relationship between the maximum
yield of cyclohexene and the total quantity of carboxylic functional
groups on the carbon surface. The results indicate that an increase
in carboxylic functional groups tends to increase cyclohexene yield.
It is important to point out that the highest maximum yields of
cyclohexene observed in supports C(KOH), C(HN90) and C(HN90-
N2/300) are due to the decrease of these solids specific surface area
along with the reduction in these supports pore volume. The reduc-
tion in micropore and mesopore volume may result in an increase
[16] P. Jen, Y. Hsu, S. Lin, Catal. Today 123 (2007) 133.
[17] C. Milone, G. Néri, A. Donato, M.G. Musolino, L. Mercadante, J. Catal. 159 (2)
(1996) 253–258.
[18] P.C. Zonetti, R. Landers, A.J.G. Cobo, Appl. Surf. Sci. 254 (21) (2008) 6849–6853.
[19] M.F.F. Rodrigues, A.J.G. Cobo, Catal. Today 149 (2010) 321–325.
[20] R.C. Bansal, M. Goyal, Activated Carbon Adsorption, Ed. T&F Group, New York,
2005, 497 p.
[21] H. Zhu, W. Han, H. Liu, Catal. Lett. 115 (1–2) (2007) 13–18.
[22] S.D. Lin, M.A. Vannice, J. Catal. 143 (1993) 539–553.
[23] L. Li, Z. Zhu, Z. Yan, G. Lu, L. Rintoul, Appl. Catal. A: Gen. 320 (2007) 166.
[24] Y. Xue, G. Lu, Y. Guo, Y. Wang, Z. Zhang, Appl. Catal. B: Environ. 79 (2008) 262.