Enhancement of biphenyl hydrogenation over gold catalysts supported on Fe-, Ce- and Ti-modified mesoporous silica (HMS)

Article abstract of DOI:10.1016/j.jcat.2009.07.011

Mesoporous metallosilicates (HMS-M; M = Ce, Fe, Ti) were used as supports for the preparation of Au catalysts, and were tested in the liquid-phase hydrogenation of biphenyl at 5 MPa and 488 K. Irrespective of the support, uniformly dispersed Au nanoparticles in range 3.2-6.5 nm were obtained. The highest turn over frequency (TOF), expressed per surface Au atom, was achieved on the Au/HMS-Fe, furthermore this catalyst gave the highest selectivity to the most saturated compound (bicyclohexyl with the highest cetane number) by means of enhancing the second aromatic-ring hydrogenation. From the catalyst activity-structure correlation, the highest activity of the Au/HMS-Fe catalyst is linked with: (i) the higher ratio of positively charged metallic gold Au^δ+/Si (XPS), and (ii) the higher stability of Au nanoparticles (HRTEM). A linear correlation between the activity (per gram of metal) of the catalysts and their ratio Au^δ+/Si is observed; however, Au/HMS-Ce catalyst displays a different behaviour in terms of activity per gram of metal exposed caused by the fact that ceria is not incorporated in the framework.

Full text of DOI:10.1016/j.jcat.2009.07.011

Journal of Catalysis 267 (2009) 30–39
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Enhancement of biphenyl hydrogenation over gold catalysts supported
on Fe-, Ce- and Ti-modified mesoporous silica (HMS)
Pedro Castaño a, T.A. Zepeda b, B. Pawelec c, Michiel Makkee a, J.L.G. Fierro c,
*
a Delft University of Technology, Catalysis Engineering, Julianalaan 136, NL2628 BL Delft, Netherlands
b Centro de Nanociencias y Nanotecnología – UNAM, Km. 107 Carretera Tijuana-Ensenada, CP. 22800, Ensenada, B.C., Mexico
c Instituto de Catálisis y Petroleoquímica, CSIC, c/Marie Curie 2, Cantoblanco, 28049 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Mesoporous metallosilicates (HMS–M; M = Ce, Fe, Ti) were used as supports for the preparation of Au cat-
alysts, and were tested in the liquid-phase hydrogenation of biphenyl at 5 MPa and 488 K. Irrespective of
the support, uniformly dispersed Au nanoparticles in range 3.2–6.5 nm were obtained. The highest turn
over frequency (TOF), expressed per surface Au atom, was achieved on the Au/HMS–Fe, furthermore this
catalyst gave the highest selectivity to the most saturated compound (bicyclohexyl with the highest
cetane number) by means of enhancing the second aromatic-ring hydrogenation. From the catalyst activ-
ity-structure correlation, the highest activity of the Au/HMS–Fe catalyst is linked with: (i) the higher ratio
of positively charged metallic gold Aud+/Si (XPS), and (ii) the higher stability of Au nanoparticles (HRTEM).
A linear correlation between the activity (per gram of metal) of the catalysts and their ratio Aud+/Si is
observed; however, Au/HMS–Ce catalyst displays a different behaviour in terms of activity per gram of
metal exposed caused by the fact that ceria is not incorporated in the framework.
Received 29 May 2009
Revised 16 July 2009
Accepted 17 July 2009
Available online 19 August 2009
Keywords:
Poly-aromatic
Liquid hydrogenation
Mesoporous silica
Au catalysts
Au–Ce catalyst
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
phene hydrodesulphurization [12,13]. The nature of the active
phase of Au catalyst is still controversial: the metallic Au0 [14] or
In the present decade, the environmental challenges for produc-
ing sufficient and better quality diesel fuel are accentuated to meet
the demand and the legislations. In particular, the polycyclic aro-
matic hydrocarbons (PAHs) have to be reduced because they are
hazardous for the humans and the environment [1,2]. The reduc-
tion of aromatics in diesel via hydrotreatment has been the basic
route for reaching the actual standards in terms of cetane number
(up to values of 51) and density(<845 kg mꢀ3). Traditional
the cationic Au3+ [15,16]. This could be due to the fact that it is dif-
ficult to detect Au3+ and positively charged metallic gold (Aud+
)
with some extended techniques as XPS [17]. A novel approach is
to embed Au3+ species in metal-organic frameworks [18].
Concerning the support, our previous study showed that Au
supported on SiO2 catalyst exhibited larger activity than the
c-
Al2O3-supported counterpart in the naphthalene hydrogenation
[9]. Those results prompt us to prepare nano-sized Au catalysts
supported on hexagonal mesoporous silica (HMS). In comparison
with SiO2, the HMS shows much weaker metal-support interac-
tion, but it offers a larger surface area and wormhole meso-
structure for diffusing reactants and products. Co(Ni)Mo/HMS
catalysts exhibited an excellent catalytic ability for hydrotreating
reactions [19–22]. The one-pot synthesis of HMS offers further
opportunities for the incorporation of transition metals into the
silica framework. The stabilization occurs by substitution of
Si4+ by Ti4+ and Al3+ [21,23]. On one hand, it has been demon-
strated that HMS–Fe support (i) is more hydrothermally stable
than pure HMS [24], (ii) is highly efficient for the hydroxylation
of phenol [25], and (iii) exhibits high activity for the alkylation
of benzene with benzyl chloride [26]. On the other hand, Au/Fex-
Oy catalysts were found to be effective in the liquid-phase
NiMo(P)/c-Al2O3 catalysts could not fulfil the required specifica-
tions so the challenge was to design stable hydroprocessing cata-
lysts based on noble metals (with higher activity) [3].
It is anticipated that the hydrogenation (HYD) rates on Au-sup-
ported catalysts are slower in comparison to that on Pt- or Pd-sup-
ported ones. This is because Au has a limited capacity to dissociate
H2 molecules due to the completely filled electronic d-band [4].
Several authors [5,6], however, demonstrated that nano-sized Au
prepared by deposition–precipitation (DP) exhibits improved
activity in the oxidation of CO at low temperatures. Recently, Au
catalysts have shown high activity and selectivity for a number
of reactions involving H2 [7,8], e.g. aromatic hydrogenation in the
presence of dibenzothiophene [9–11] and thiophene–dibenzothio-
hydrogenation of
a, b-unsaturated ketones [27]. Similar to Fe,
one might expect that the incorporation of Ce4+ ions into the
framework of HMS network might impact simultaneously on
* Corresponding author. Fax: +31 0 1527 85006.
E-mail address: jlgfierro@icp.csic.es (J.L.G. Fierro).
0021-9517/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.07.011

P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
31
Nomenclature
AwAu
C
atomic weight of gold in g/mol
Abbreviations and subscripts
BPH, CHB, BCH biphenyl, cyclohexylbenzyl, bicyclohexyl
concentration in g/L or mol/L
initial concentration in g/L or mol/L
effective diffusivity
particle diameter in nm or m
pseudo-first kinetic constant in gBPH/gcat ꢁ h
molecular weight in g/mol
Avogadro number
C0
DP
deposition–precipitation method
Deff
Dp
k
Mw
NA
ꢀrmax
t
S
V
Wc
X
Z
HYD
PAH
QTOF
TOF
hydrogenation
polycyclic aromatic hydrocarbons
turn over frequency in molBPH/molAu-total ꢁ s, Eq. (5)
turn over frequency in molBPH/molAu-exposed ꢁ s, Eq. (4)
weigh hourly space velocity in g/gcat ꢁ h
WHSV
maximum hydrogenation rate
time in h or s
Greek symbols
surface density of Au in atAu/cm2
volume of reactor in L
m
q
weight proportion of Au in the catalyst in gAu/gcat
density in g/cm3
weight of catalyst in gcat
conversion of biphenyl, Eq. (1)
conversion of aromatics, Eq. (2)
UWP
s
Weisz–Prater module, Eq. (6)
batch space time in gcat ꢁ h/gBPH, Eq. (3)
the redox and acidic properties of the support [28], enhancing
the hydrogenation reaction [29].
overnight at 323 K. Finally, calcination was performed at 573 K
for 3 h.
Within this background information, the present study investi-
gated the properties and hydrogenation performance of Au sup-
ported on HMS modified with Ce, Fe and Ti. Biphenyl HYD was
selected as the test reaction because it is a by-product of dibenzo-
thiophene hydrodesulphurization, and jointly with naphthalenes
forms the majority of aromatics in diesel. The structural and sur-
face characterization of the samples was performed by means of
different techniques (N2 adsorption–desorption isotherms, TPD-
NH3, HRTEM, X-ray photoelectron spectroscopy (XPS) and TPO/
TGA techniques).
2.2. Characterization techniques
2.2.1. Chemical analysis
Atomic Emission Spectroscopy (ICP-AES) was used for the
determination of the Au loading in the calcined catalysts by means
of a Perkin Elmer Optima 3300DV spectrometer. The solid samples
were digested (in a mixture of HF, HCl and HNO3) in a microwave
oven for 2 h. Then, aliquots of solution were diluted to 50 mL using
deionized water (18.2 m
X quality).
2. Experimental
2.2.2. N2 adsorption–desorption isotherms
N2 adsorption–desorption isotherms (at 77 K) were used for the
determination of surface area (BET method), pore volume and pore
size distribution (BJH method) of the pure supports and calcined
catalysts, degassed at 543 K in vacuum for 5 h, using a Micromer-
itics TriStar 3000 apparatus.
2.1. Support/catalyst preparation
Hexagonal mesoporous silica (HMS) was synthesized by neutral
S0I0 templating route, which is based on hydrogen bonding and
self-assembly between neutral primary amine surfactants (S0)
and a neutral inorganic precursor (I0) [30]. Tetraethylorthosilicate
(TEOS, 98% Aldrich) was used as a neutral silica precursor, whereas
dodecylamine (DDA, Aldrich 99%) and mesitylene (C9H12, Aldrich
98%) were used as a neutral structure director and a swelling or-
ganic agent, respectively, as firstly proposed by Kresge et al. [31].
Furthermore, the excess of HCl was employed and the synthesis
was performed at 335 K. The reaction products were filtered,
washed with distilled water and dried at room temperature for
24 h, followed by drying at 378 K for 2 h. Subsequently, the sample
was calcined in air at 823 K for 3.5 h with a heating rate of 2.5 K/
min. The metallosilicates (HMS–M; M = Ce, Fe, Ti) were prepared
by dissolving the appropriate amounts of cerium nitrate hexahy-
drate (Aldrich 99%), iron(III) nitrate nonahydrate (Aldrich 98%) or
tetrabutyl orthotitanate (Aldrich 98%) in TEOS in order to obtain
samples with a Si/M (M = Ce, Fe, Ti) atomic ratio of 40. The drying,
washing and calcination steps were the same as described above
for the HMS material.
2.2.3. Temperature-programmed desorption technique
Temperature-programmed desorption of ammonia (TPD-NH3)
was employed for the determination of the number of acid sites
of the supports, using a Micromeritics 2900 equipment provided
with TCD detector. After loading, the sample of 50 mg was pre-
treated in a He (Air Liquide, 99.996%) stream at 573 K for 1 h. Fol-
lowing this, the sample was cooled to 473 K and ammonia-
saturated in a stream of 5% NH3/He (Air Liquide) flow (50 mL/
min) for 1 h. Then, after catalyst equilibration in a He flow at
400 K for 0.5 h the ammonia was desorbed using a heating rate
of 10 K/min to 1350 K. The amount of desorbed ammonia was fol-
lowed by thermal conductivity detector (TCD).
2.2.4. X-ray diffraction (XRD)
The catalysts were characterized by powder X-ray diffractome-
try according to the step-scanning procedure (step size 0.02°; 0.5 s)
with a computerized Seifert 3000 diffractometer, using Ni-filtered
The supported Au/HMS, Au/HMS–Ce, Au/HMS–Fe and Au/HMS–
Ti catalysts were prepared by deposition–precipitation method
using the HAuCl4 as Au precursor. A solution of chloroauric acid
in water was slowly added to a stirred suspension of 2 g of support
in water. The pH of the solution was kept at 9.0 with aqueous
ammonia (Merck, 25%), and the solution was aged for 30 min at
room temperature. The chlorine ions were removed by washing
the sample with warm distilled water followed by drying in air
CuKa (k = 0.154 nm) radiation and a PW 2200 Bragg–Brentano h/2h
goniometer equipped with a bent graphite monochromator and an
automatic slit. The assignment of the various crystalline phases
was based on the JPDS powder diffraction file cards. The metal par-
ticle sizes were calculated from the line broadening of the most in-
tense peak using the Debye–Scherrer equation.

32
P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
2.2.5. UV–vis spectroscopy
ture of 488 K and stirring rate of 1200 rpm. Under these tailored
conditions, the concentration of hydrogen is 56.5 g/L [32,33]. The
analysis of un-reacted compounds and products was carried out
in a ChromPack CP9001 gas chromatograph equipped with a
50 m HP-1 column and a Flame Ionization Detector (FID).
Biphenyl (BPH) is hydrogenated in series to form cyclohexylb-
enzyl (CHB) and subsequently bicyclohexyl (BCH). In order to ac-
count for these sequenced hydrogenations in the kinetics two
different conversions have been defined:
The electronic spectra of the finely ground pure supports and
calcined Au catalysts were recorded in the range of 200–900 nm
at room temperature using a Varian Cary 5000 UV–vis spectropho-
tometer equipped with an integration sphere.
2.2.6. HRTEM measurements
High resolution transmission electron microscopy (HRTEM)
studies of the spent catalysts were carried out using a JEM 2100F
microscope operating with a 200 kV accelerating voltage and fitted
with an INCA X-sight (Oxford Instruments) energy dispersive X-ray
microanalysis (EDX) system to verify semi-quantitative composi-
tion of supported phases. Au nanoparticles were differentiated in
all cases except for the Au/HMS–Ce catalyst where CeO2 clusters
show similar contrast as Au ones. In this case, the round morphol-
ogy of Au nanoparticles and the difference in the distance between
crystal planes (0.326 nm for CeO2 [111] and 0.235 nm for Au0
[111]) guided us to distinguish between the two metals.
C0BPH ꢀ CCHB ꢀ CBCH
X ¼
Z ¼
ð1Þ
ð2Þ
C0BPH
C0BPH ꢀ 1=2CCHB ꢀ CBCH
C0BPH
where C0BPH is the initial concentration of biphenyl, whereas CCHB
and CBCH are the concentrations of cyclohexylbenzene and bicy-
clohexyl, respectively. X characterizes the hydrogenation of biphe-
nyl, whereas Z characterizes the total aromatic hydrogenation in
this reaction in series, giving values of the conversion over the en-
tire range of operation.
2.2.7. X-ray photoelectron spectroscopy (XPS)
X-ray photoelectron spectroscopy (XPS) of the fresh reduced
and spent catalysts was carried out on a VG Escalab 200 R electron
spectrometer equipped with a hemispherical electron analyzer,
The batch space time (s) had been defined to compare the ki-
netic runs:
using
source.
a
Mg
Ka
(hm
= 1253.6 eV, 1 eV = 1.603 ꢂ 10ꢀ19 J) X-ray
!
Wc
V ꢁ C0BPH
s
¼
t
ð3Þ
2.2.8. Thermogravimetric analysis (TGA)
The quantity of coke deposited on the spent catalysts was deter-
mined by temperature-programmed oxidation (TPO) by thermo-
gravimetry (TGA/SDTA851 Mettler Toledo), measuring the weight
lost in the spent catalysts during oxidation. Each sample (ca.
30 mg) was previously heated in order to homogenize coke and re-
move the volatile compounds up to 773 K for 1 h in N2 (10 K/min;
200 mL/min). Once the sample was cooled to 303 K (5 K/min)
burning of coke was carried out by raising sample temperature
to a final temperature of 1173 K at a rate of 10 K/min in a 20%
O2/N2 gas mixture (50 mL/min). The data used correspond to tem-
peratures up to 823 K since the catalysts had been previously cal-
cined at this temperature and the HMS support could decompose.
The amount of coke (+ reactant/products mixture in the pores of
catalyst) present in the spent catalyst was calculated by the differ-
ence between initial weight and final weight of the catalyst sam-
ple, with the initial weight taken as the mass of the sample after
pre-treatment at 773 K under N2 at 1 h.
where V is the volume of reactive solution in the reactor (typically
0.15 L) and Wc is the weight of the catalyst.
The activity of the catalyst was determined on the basis of pseu-
do-first order kinetic linearization (ꢀLn(1 ꢀ X) = k ꢁ
s), which is a
much extended approach for this liquid-phase hydrogenation
[34]. The Turn Over Frequency (TOF) is the optimal method to com-
pare activity since it accounts for the number of accessible metallic
sites for hydrogenation. TOF had been calculated using the follow-
ing expression based on the assumption of cube-shaped particles
whose five of six faces were accessible:
kNADpAuqAu
5tAuMwBPHS
TOF ¼
ð4Þ
where NA was the number of Avogadro, DpAu is the diameter of Au
particles calculated from HRTEM results, qAu is the density
(19.3 gAu/cm3),
tAu is the Au weight proportion in the catalyst mea-
sured by ICP, MwBPH is the molecular weight of BPH and S is the sur-
face density (1.17 ꢂ 1015 atAu/cm2). Note that k has units inversed
2.3. Catalytic activity measurements
to
s
(gBPH/gcat ꢁ h). When the amount of accessible metallic sites
cannot be precisely calculated, then it is optimal to measure activity
The catalyst performance was measured in the liquid-phase
hydrogenation of biphenyl in a slurry-type reactor: 0.3 L batch stir-
red reactor, Medimex Inc. The reaction procedure consisted in
loading the catalyst (ca. 0.3 g) in the reactor and treating it in-situ
under a flow of hydrogen of 80 mL/min, atmospheric pressure and
523 K for 2 h. Then the liquid mixture was loaded. The liquid mix-
ture consisted in a solution (0.15 L) of n-tetradecane (Acros Organ-
ics, 99%), 10 g/L of n-hexadecane (Merck, 99%) as a reference
compound and 10–12 g/L of biphenyl (Acros Organics, 98%) as a
reactant. Traces of sulphur compounds were detected in the liquid
mixture and, therefore, it was treated prior in an adsorbing bed
(extrudates of Ni–NiO/Al2O3, atmospheric pressure, WHSV = 1 g/
gcat ꢁ h and 423 K) for decoupling the influence of poisoning on
the kinetics. By the adsorption treatment the concentration of sul-
phur in the feedstock was reduced from 0.2–0.6 ppm to 0.01–
0.02 ppm according to the results obtained with an internal stan-
dard solution of thiophene in a HP-4890 Series II gas-chromato-
graph, equipped with a Sulphur Chemiluminescence Detector
(SCD). The reaction conditions were: pressure of 5 MPa, tempera-
in terms of total metallic sites (in molBPH/molAu ꢁ h).:
kAwAu
AuMwBPH
QTOF ¼
ð5Þ
t
where AwAu is the atomic weight of gold. The absence of external
mass transfer limitations was assessed by selecting the optimal stir-
ring rate and catalyst particle size combination in order to attain
intrinsic kinetics with the more active commercial catalyst (per gcat
based on Pt–Pt [35]. The absence of internal mass transfer limita-
)
tion was verified by calculating the Weisz–Prater module (UWP
for a first-order kinetic reaction:
)
ðꢀrmaxÞDp2
UWP
¼
ð6Þ
Deff CBPH
where (ꢀrmax) is the maximum hydrogenation rate and Deff is the
effective diffusivity, derived from Wilke–Chang equation multiplied
by the porosity over the tortuosity. Moreover, the hydrogenation
runs were performed at conversions not higher than 25%, where

P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
33
the influence of the internal mass transfer limitation can be
neglected.
largest incorporation of its guest ions within the framework struc-
ture of the HMS material, whereas ceria might form some separate
clusters supported on the HMS but not embedded in its
framework.
3. Results
The wide-angle XRD patterns of the calcined catalysts are
shown in Fig. 1. As seen in this figure, all calcined catalysts exhibit
peaks positioned at 2h values of 38.3° and 44.6°, which are attrib-
uted to [111] and [200] planes, respectively, of Au° crystals. Addi-
tionally, the Au/HMS–Ce shows the peaks positioned at 2h values
of 26.8° and 55.6°, which are indicative of the planes of CeOx par-
ticles, respectively. On the other hand, from the peaks for Au/HMS–
Fe, the evidence for the presence of Fe2O3 was deduced from a
broad peak line at 2h = 23.0°. Au/HMS–Ti catalyst does not show
any evidence of TiO2 crystallites.
3.1. Characterization of the fresh catalysts
3.1.1. Chemical, textural and acidic properties
Some physico-chemical properties of the calcined catalysts are
summarized in Table 1. Chemical analysis revealed that Au content
of Au/HMS–M (M = Ce, Fe, Ti) catalysts is a little larger than that of
the Au/HMS sample (3.02–3.44 vs. 2.24 wt.%), considering the fact
that the nominal content in all catalyst is 3.5 wt.%, we have to indi-
cate that the efficiency of the Au-loading method increases after
HMS modification with Ce, Fe or Ti. Furthermore, the results pre-
sented in Table 1 show that the efficiency of the Au-loading meth-
od increases with the Ce, Fe or Ti concentration, particularly for the
Au/HMS–Ce catalyst. The textural properties of the calcined cata-
lysts were evaluated from the nitrogen adsorption–desorption iso-
therms at 77 K (not shown here). According to the IUPAC
classification, all supports show the N2 isotherms of type IV and
H1 hysteresis loops, which indicate the presence of textural
meso-pores and uniform cylindrical pores. Au/HMS and Au/HMS–
Ti catalysts show two capillary condensation steps: the first one
at P/P0 ꢃ 0.40 indicates the presence of framework meso-porosity,
whereas the second one at P/P0 ꢃ 0.85 is due to textural inter-par-
ticle meso-porosity or macro-porosity. Moreover, both Au/HMS–Ce
and Au/HMS–Fe samples show smaller sharpness of their adsorp-
tion isotherms indicating poorer pore uniformity.
3.1.3. UV–vis spectroscopy
To gain insight into the effect of HMS doping with Ce, Fe and Ti,
the UV–vis diffuse reflectance spectra of the pure supports were
recorded at room temperature in the range of 200–900 nm
(Fig. 2). The electronic spectrum of Au/HMS–Ce shows a wide
and intense absorption band in the region of 270–350 nm. Its inter-
pretation is not easy due to the large band width and the specific
reflectance, which is frequently observed in the Ce spectra. How-
ever, one could obtain information on surface coordination and
oxidation states of Ce ions by measuring their d–d and f–d transi-
tions and the oxygen–metal ion charge transfer bands [37]. The
absorption band observed in the range of 270–350 nm could be
associated with the electronic transition with charge transfer
(O2ꢀ ? Ce4+) associated to Ce ions in SiO2 [38], pointing again that
ceria has not been incorporated in the framework but forms clus-
ters on the surface of the HMS. On the other hand, the UV–vis spec-
trum of HMS–Fe shows a broad band of absorption at ca. 248 nm,
For all catalysts, the specific area (SBET), cumulative pore volume
(Vtotal) and pore diameter are compiled in Table 1. From this table,
the SBET and Vtotal values follow similar order (Au/HMS–Ti > Au/
HMS > Au/HMS–Fe > Au/HMS–Ce). Both Au/HMS and Au/HMS–Ti
samples show much broader pore size distribution (in the range
2–10 nm) than their Au/HMS–Fe and Au/HMS–Ce counterparts
(in range 2–6 nm).
Au0
The total concentration of acid sites (expressed as molNH3/m2) is
shown in Table 1. From this table, the acidity of the catalysts fol-
lows the trend: Au/HMS > Au/HMS–Ti > Au/HMS–Fe > Au/HMS–
Ce. Thus, the HMS material modification with Ti and Ce ions led
to both a lower and a larger decrease in overall support acidity.
CeOX
(44.6°)
(26.8°)
Au0
(38.2°)
CeOX
(55.6°)
Au/HMS-Ce
Au/HMS-Ti
3.1.2. X-ray diffraction
Fe O3
The hexagonal arrangement of all synthesized supports was
confirmed by the low-angle XRD patterns (not shown here). The
unique peak at ca. 1.8° was due to the diffraction plane [100].
The absence of higher order reflections, viz. [110], [200] and
[210], could be due to finite size effects of very fine particle mor-
phology or due to the more disordered hexagonal framework
structure of the samples [36]. For all supports, the d100-spacing cal-
culated from the X-ray patterns follows the trend: HMS–Ce
(232.0°)
Au/HMS-Fe
Au/HMS
10
20
30
40
50
60
70
2 Theta (°)
(6.94 nm) > HMS
(6.89 nm) > HMS–Ti
(6.72 nm) > HMS–Fe
Fig. 1. XRD of the fresh Au/HMS–M (M = [–], Ce, Fe, Ti) catalysts.
(5.96 nm). Such a trend indicates that the HMS–Fe showed the
Table 1
Some characteristics of the calcined catalysts and supports.
Au (wt.%)a
Ce2O3/ Fe2O3/TiO2 (wt.%)a
SBET (m2/g)
Vtotal (cm3/g)
db (nm)
Acid sitesc (mmolNH3/m2)
b
b
Samples
Au/HMS
2.24
3.44
2.84
3.02
–
521
387
460
551
1.1
0.6
0.8
1.2
8.2
6.7
7.0
8.9
0.70
0.34
0.46
0.62
Au/HMS–Ce
Au/HMS–Fe
Au/HMS–Ti
11.2
5.8
3.1
a
As determined by chemical analysis (ICP technique) for the oxide precursors.
Textural properties of the calcined catalysts as determined by N2 adsorption–desorption isotherms: SBET: specific surface area as calculated by BET equation; Vtotal: total
b
pore volume; d: average pore diameter.
c
Total amount of the acid sites of the pure supports.

34
P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
Au/HMS
Au/HMS-Ce
274
248
support
support
fresh callcined
spent
fresh calcined
spent
520
223
528
200
400
600
800 200
400
600
800
Wavelength (nm)
Wavelength (nm)
211
Au/HMS-Ti
Au/HMS-Fe
248
support
support
fresh calcined
spent
fresh calcined
spent
337
540
528
200
400
600
800 200
400
600
800
Wavelength (nm)
Wavelength (nm)
Fig. 2. UV-vis spectra of the support, fresh and spent Au/HMS–M (M = [–], Ce, Fe, Ti) catalysts.
confirming the incorporation into the HMS framework [39]. The
observed shoulder at around 350 nm is characteristic of small FeOx
nanocrystallites [40]. The comparison of band intensities clearly
indicates the major incorporation of Fe3+ ions into framework of
HMS material. Finally, the HMS–Ti support shows one intense band
at 211 nm which is usually assigned to isolated Ti4+ ions in a nearly
tetrahedral coordination [41].
Fig. 2 also shows the UV–vis spectra of the calcined and spent
catalysts (with Au). The absorption band at 248 nm observed in
the spent Au/HMS and Au/HMS–Fe samples could be associated
to the presence of charged gold species that are probably localized
in the perimeter of the interface of Au particles with the support
[42]. Contrary to the calcined catalysts, the UV–vis spectra of the
spent Au catalysts show a band at ca. 530 nm which is usually as-
signed to plasmon or Au nanoparticles [43]. However, as this band
is absent in the fresh catalysts, it can be reasonably assigned to
coke residues [44]. Considering the higher adsorption of the deac-
tivation catalyst (Fig. 2), it seems improbable that the leaching of
Ce, Ti and Fe from the HMS is occurring during the reaction.
Fig. 4 shows a micrograph of the Au/HMS–Ce catalyst where it
can be observed that the ceria is not implanted in the framework
of HMS but forming clusters on its surface in agreement with the
UV–vis and XRD results. In the light of these results, two types of
catalyst can be differentiated: on one hand, the Au supported on
HMS, HMS–Ti and HMS–Fe, on the other hand the bimetallic Au–
Ce supported on HMS since both metals form interacting domains
on the surface of the support.
Fig. 4 also shows the intimate contact between CeO2 and Au
nanoparticles. However, around half of the population of Au parti-
cles remains isolated from CeO2 particles. The CeO2 particle size
distribution has been determined for HMS–Ce support (prior to
Au loading), indicating that most of the CeO2 particles are around
4 nm but they have a broad distribution up to sizes larger than
10 nm. On top and although the procedure has been explained in
the experimental section, differentiating Au from CeO2 is not al-
ways straightforward; EDX analysis could prove the existence of
Au and Ce in a single point but the contrast is not sufficient to
determine the boundaries of the nanoparticles. As a result, we
can conclude that the values of Au particle size for Au/HMS–Ce cat-
alyst could be less accurate.
As compared with fresh reduced catalysts (Fig. 3), their spent
counterparts showed a more heterogeneous distribution of Au0
particles, with the Au/HMS–Fe catalyst being the one which shows
the lowest increase of the Au0 particle size during the reaction.
3.1.4. HRTEM measurements
The morphology of the surface of the reduced and the spent cat-
alysts was studied by HRTEM technique. The size histograms of
both forms of catalysts are compared in Fig. 3, the micrographs
demonstrate that the Au nanoparticles have a rounded shape.
Au/HMS–Ti catalyst shows ꢄ70% of population of particles in the
1–2 nm range, ꢄ60% of the particles of Au in the reduced Au/
HMS and Au/HMS–Fe catalysts are in the 2–3 nm size range, while
Au/HMS–Ce reduced catalyst shows the highest mean value and
distribution of Au particle sizes (5.5 1.8 nm).
3.1.5. XPS measurements
The electronic properties of the support modifiers (Ti, Fe and
Ce) and the Au were studied by XPS technique. The XP spectra of
Au 4f core electron level of the fresh reduced and spent catalysts
are shown in Figs. 5A and B. The binding energies values of the

P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
35
100
100
80
Au/HMS
Au/HMS-Ce
reduced:d = 2.5 0.9 nm
spent: d = 3.2 1.6 nm
reduced:d = 5.5 1.8 nm
spent: d =5.5 1.7 nm
80
60
40
20
0
60
40
20
0
0
1
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
Particle size (nm)
Particle size (nm)
100
80
100
80
Au/HMS-Fe
reduced:d = 3.2 1.6 nm
spent: d = 3.2 1.5 nm
Au/HMS-Ti
reduced:d = 1.5 0.5 nm
spent: d = 3.2 1.7 nm
60
40
20
0
60
40
20
0
0
1
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
Particle size (nm)
Particle size (nm)
Fig. 3. Histograms of particle size distribution in the reduced and spent catalysts as obtained by HRTEM.
After the reaction, the catalysts displayed an additional Au 4f dou-
blet whose most intense Au 4f7/2 peak appeared at a BE of 85.3 eV,
which can be assigned to very small positively charged metallic
gold (assigned to Aud+) interacting with support [45], see Table 2.
Au/HMS–Ti and Au/HMS–Ce catalysts show much less presence
of Aud+ species by XPS, however, a direct evidence of AuOꢀ, AuO2
and AuOHꢀ ion clusters on supported Au catalysts has been re-
cently reported by a time-of-flight secondary-ion mass-spectros-
copy (TOF-SIMS), whereas the XPS of the same samples did not
detect those species [17].
For the fresh reduced and spent Au/HMS–Ce catalysts (Fig. 6),
the XP spectra of Ce 3d core electron level exhibited eight peaks
corresponding to the spin-orbit doublet and satellite lines charac-
teristic of oxidized state. The spin-orbit splitting is 18.4 eV. The Ce
3 d5/2 and 3d3/2 showed the main features at 882.2, 884.6, 892.7
and 898.7 eV corresponding to v, v0,v00 and v000, respectively, and
the features at 902.2, 904.7, 907.4, and 916.9 correspond to u, u0,u00
and u000 components, respectively [46]. Analyzing the features char-
acteristic of each Ce4+ and Ce3+ species, it is evident that the pro-
portion of the Ce4+ is three times higher than that of the Ce3+
.
The spectra of the fresh reduced and spent Au/HMS–Fe samples
are compared in Fig. 6. For both fresh and used catalysts, the Fe
2p3/2 binding energy at 710.9 eV (or 710.6 eV for spent catalyst)
and the signal shape indicated the presence of Fe3+ ions. The slight
increase of the BE value (711.4 eV) of Fe 2p3/2 peak in the spent
sample suggests the formation of some hydroxyl groups on the
iron surface [47]. For both fresh reduced and spent forms of the
Au/HMS–Ti catalyst (spectra not shown here), the component with
BE of Ti 2p3/2 at 460.0 eV indicates unambiguously that Ti4+ ions
are in a tetrahedral arrangement (Si–O–Ti), whereas the minor Ti
2p3/2 component at a BE of 458.5 eV is characteristic of titanium
ions in an octahedral (Ti–O–Ti) coordination [48]. The proportion
of these octahedral Ti–O–Ti species decreases after reaction sug-
Fig. 4. Micrograph of a metallic cluster formed by Au (B) and CeO2 (A) for the Au/
HMS–Ce reduced catalyst.
most intense Au 4f7/2 peak are listed out in Table 2 along with the
binding energies of the Ce 3d5/2, Fe 2p3/2 and Ti 2p3/2 core levels
corresponding to Au/HMS–M (M = Ti, Fe, Ce), respectively. For all
catalysts, the Au 4f7/2 binding energy (BE) was found at 83.7 eV
(83.8 eV for Au/HMS–Fe, and this BE is typical of metallic Au [6]).

36
P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
B
A
Au 4f
Au 4f
δ
Au/HMS
reduced
Au/HMS
spent
Au/HMS-Ti
reduced
Au/HMS-Ce
spent
Au/HMS-Ti
spent
Au/HMs-Fe
reduced
Au/HMS-Fe
spent
Au/HMs-Ce
reduced
80
84
BE (eV)
88
92
80
84
88
BE (eV)
92
Fig. 5. XPS spectra of Au 4f core level for reduced (A) and spent (B) Au/HMS–M catalysts.
Table 2
Binding energies (eV) of core electrons and surface atomic ratios of the reduced (573 K, 1 h) and spent catalysts.
Catalyst
Si 2p
Au 4f
Ce 3d (Fe or Ti 2p)
Au/Si at (Aud+/Si)
M/Si at (M = Ce, Fe, Ti)
Fresh reduced
Au/HMS
Au/HMS–Ce
Au/HMS–Fe
Au/HMS–Ti
103.4
103.4
103.4
103.4
83.7
83.7
83.8
83.7
–
0.0052
0.0045
0.0044
0.0047
–
882.5
710.9
458.5 (76) 460.0 (24)
0.022
0.014
0.017
Spent
Au/HMS
103.4
103.4
103.3
103.4
84.0 (89) 85.3 (11)
83.9 (90) 85.2 (10)
83.8 (82) 85.3 (18)
83.9 (90) 85.2 (10)
–
0.0061 (0.00067)
0.0022 (0.00022)
0.0041 (0.00078)
0.0025 (0.00025)
–
Au/HMS–Ce
Au/HMS–Fe
Au/HMS–Ti
882.0
709.6 (43) 711.4 (57)
458.6 (41) 459.9 (59)
0.010
0.010
0.022
gesting that some hydroxyl groups around the coordination sphere
of Ti4+ are formed. Considering the information from UV–vis
(Fig. 2) and HRTEM (Fig. 3) results, the smallest size of the Au0 par-
ticles formed on Au/HMS–Ti catalyst could be linked with the large
amount of TiO2 species present on the support surface after reac-
tion as detected by XPS.
From the Au/Si surface atomic ratios evaluated from XP spectra,
it is clear that the cation Mn+ incorporated into HMS–M has a
strong influence on the distribution of the supported Au species
(Table 2). Considering the Au/Si atomic ratio of the fresh reduced
samples, the observed trend is: Au/HMS > Au/HMS–Ti ꢃ Au/HMS–
Ce > Au/HMS–Fe. Thus, Au exposure on all the cation-modified
HMS–M substrates decreases with respect to the HMS one. This
observation is in agreement with HRTEM analysis of the particle
size and density.
iour of all catalysts was well modelled by the pseudo-first-order
kinetics with respect to the batch space time ( ) from Eq. (3). Even
s
though pseudo-first-order kinetic models have some limitations
with respect to integral models [35], they are still useful for com-
paring the intrinsic activities close to differential conditions (con-
version <20%). The kinetic performances of Au/HMS–M catalysts
are displayed in Fig. 7 by means of the linear pseudo-first-order ki-
netic plot. Derived kinetic constants (in inverse units as s) are sum-
marized in the sub-graph of Fig. 7. In terms of the pseudo-first
kinetic constant (k), Au/HMS–Fe catalyst displays the highest activ-
ity towards biphenyl hydrogenation. The value of the Weisz–Prater
module of this experimental run (calculated using the rate in
Fig. 7) was 0.045 assuring the minimal contribution of internal
mass transfer limitations during the reaction.
It is expected that the acid sites could be involved in the hydro-
genation mechanisms [49]. However, the direct contribution of the
support to the hydrogenation rates is negligible based on (i) the
low conversion (2.5%) obtained with the bare support (HMS) at
the same conditions and (ii) the fact that the observed activity of
the catalysts is not a function of their acidity (Table 1).
3.2. Catalyst testing in biphenyl liquid-phase hydrogenation
The biphenyl liquid-phase hydrogenation was used to deter-
mine the activity of the supported Au catalysts. The kinetic behav-

P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
37
Ce 3d
ν
Ce 3d
ν
ν'
ν''
ν'
υ
υ'
υ'
ν'''
υ
υ'' Au/HMS-Ce
υ'' Au/HMS-Ce
reduced
spent
υ'''
870
885
900
915
870
885
900
915
BE (eV)
BE (eV)
Fe3+
Fe2+
Fe3+
Fe 2p
Fe 2p
Au/HMS-Fe
spent
Au/HMS-Fe
reduced
704
712
720
BE (eV)
728
710
720
BE (eV)
730
Fig. 6. XPS spectra of the Fe 2p and Ce 3d core electron levels for reduced Au/HMS–Fe and Au/HMS–Ce catalysts.
0.30
0.25
0.20
0.15
0.10
0.05
0.00
1.5
1.0
0.5
0.0
coke precursors
soft coke
Au/HMS
Au/HMS-Ce
Au/HMS-Fe
Au/HMS-Ti
Au/HMS-Ti
Catalyst
Au/HMS-Ce
Au/HMS-Fe
Fig. 8. Comparison of the type of coke formed on the different catalysts (TPO).
Au/HMS
0.0 0.1 0.2 0.3 0.4 0.5 0.6
-1
-1
cat
k (gBPH
g
h )
of the fresh-spent catalysts (Fig. 3) allows us to conclude that there
is certain sintering of the Au/HMS-Ce and Au/HMS–Ti catalysts,
while the Au particle size distribution of Au/HMS and Au/HMS–
Fe catalysts remains almost unaffected, in particular the Fe-con-
taining catalyst.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
-1
BPH
Batch space time (gcat h g
)
Fig. 7. Pseudo-first kinetic performance of the studied Au catalysts and their kinetic
constants calculated from the slopes. Reaction conditions: 5 MPa, 488 K.
The spent catalysts were characterized also by temperature-
programmed oxidation (TPO) experiments. Fig. 8 shows the weight
loss percentage of the samples corresponding to the temperature
of combustion: 300–500 K, coke precursors and 500–800 K, ‘‘soft
coke”. The total amount of carbonaceous deposits in all the cata-
lysts is similar; ꢄ1.4% and the ratio of coke precursors to soft coke
is also very similar indicating that neither the incorporation of
metals into the HMS framework nor the Au properties affect the
coke deposition.
3.3. Characterization of the spent catalyst
The properties of the spent catalysts have been previously com-
pared to those of the fresh catalysts by UV–vis spectroscopy
(Fig. 2), HRTEM (Fig. 3) and XPS (Fig. 5) techniques. The compari-
son of the size distribution of Au particles (by processing HRTEM)

38
P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
genation rates (k1 and k2) and k2 in particular, leading to higher
4. Discussion
activity and selectivity of the most interesting product (BCH).
The trend in activity cannot be correlated with the acidic prop-
erties (Table 1), the textural properties (Table 1) or more impor-
tantly with the Au particle morphology (Table 2, Figs. 3 and 4).
The most active catalyst in terms of hydrogenation rate per
gcat(k), per gAu-exposed(TOF) and gAu-total (QTOF) is Au/HMS–Fe.
Moreover Au/HMS–Fe catalyst enhances the second aromatic
hydrogenation (k2) over the first one (k1). Interestingly, the ob-
served QTOF correlates linearly with the exposure of Aud+ species.
For Au/HMS–Ce catalyst, the Ce is not incorporated in the frame-
work creating nano-sized clusters of Au and CeO2. The resulting
interaction makes the differentiation of Au from CeO2 cumbersome
in HRTEM (yielding not reliable DpAuvalues), furthermore this
interaction is able to influence the electronic properties of Au with-
out forming Aud+ species (Fig. 5) enhancing the aromatic hydroge-
nation. The lower activity of Au/HMS–Ti catalyst can be explained
taking into account the blockage of Au nanoparticles caused by
titania coverage and/or Au sintering, as derived from the lower
Au/Si atomic ratio of the spent catalyst with respect to the fresh
counterpart (XPS results, Table 2).
All the Au-supported catalysts (Au/HMS–M) proved to be active
in the liquid-phase hydrogenation of biphenyl, the activity per
gram of catalyst or metal is, however, substantially lower than that
of Pt–Pd/Al2O3 catalyst upon similar reaction conditions [35]. This
means that the implementation of Au-supported catalyst is not
justified in terms of its activity, but could be justified in terms of
stability.
The reaction rates expressed as TOF (in molecule of biphenyl
converted per second and per exposed atom of Au, calculated using
Eq. (4)) and QTOF (in molecule of biphenyl converted per second
and per total atom of Au, calculated using Eq. (5)) are plotted in
Fig. 9 vs. the exposure of Aud+ species (measured by XPS, Table
2). As seen, Au/HMS–Fe catalyst shows the higher aromatic hydro-
genation rate, the modification of the catalyst with Ti decreased
the activity with respect to the Au/HMS catalyst. Besides, a linear
correlation is observed for both TOF and QTOF with respect to
the Aud+ exposure, except for the Au/HMS–Ce catalyst which shows
an higher TOF value than expected (in red, Fig. 9). This outlier
comes from the particular state of the metallic phases in the Au/
HMS–Ce catalyst, which form clusters of Au–CeO2 onto the HMS
support (Fig. 4), thus making the measurement of DpAu less accu-
rate, which is required for calculating TOF in Eq. (4).
5. Conclusions
Bicyclohexyl (BCH) is the most interesting product due to its ce-
tane number. Fig. 10 represents the selectivity of BCH versus BPH
conversion (X) for all the catalysts, and the bare HMS support. Con-
sidering the fact that this is a reaction in series with two reaction
rates (k1 and k2), Fig. 10 indicates that at the same BPH conversion
level, k2 is faster relative to k1 for the Au/HMS–Fe catalyst (in the
order: Au/HMS–Fe > Au/HMS–Ce > Au/HMS–Ti > Au/HMS). As a re-
sult, the doping of Fe in the Au/HMS catalyst enhances both hydro-
Au nanoparticles supported on modified HMS with Ce, Fe and Ti
were synthesized by deposition–precipitation method, character-
ized (fresh and spent) and tested in the liquid-phase biphenyl
hydrogenation. The obtained results can be summarized as
follows:
ꢅ The modification of HMS by Ce and Fe leads to Au-supported
catalysts that are more stable in terms of sintering. The enhance-
ment of biphenyl hydrogenation with respect to Au/HMS cata-
lyst only occurs when Fe is incorporated into the framework of
HMS.
2.0
Au/HMS-Fe
Au/HMS-Ce
1.5
Au/HMS
1.0
100
Y = 20%
0.5
Au/HMS-Ti
80
0.0
0
2
4
6
8
10
Y = 15%
Au3+/Si (10-4 at.)
60
Au/HMS-Fe
6
5
4
3
Y = 10%
40
Au/HMS
Au/HMS-Fe
Au/HMS
Y = 5%
Au/HMS-Ce
Au/HMS-Ti
HMS
Au/HMS-Ti
20
0
Y = 1%
Au/HMS-Ce
0
2
4
6
8
10
0
5
10
15
20
25
Au3+/Si (10-4 at.)
BPH conversion, X (%)
Fig. 9. (a) Linear correlation between TOF and the Aud+/Si atomic ratio as measured
by XPS (b) Linear correlation between QTOF and the Aud+/Si atomic ratio as
measured by XPS.
Fig. 10. Selectivity of bicyclohexyl (BCH) versus biphenyl (BPH) conversion of the
catalyst used and the bare support. Reaction conditions at 488 K and 5 MPa. The
grey lines represent the yields (Y) of BCH.

P. Castaño et al. / Journal of Catalysis 267 (2009) 30–39
39
[12] A.M. Venezia, V. La Parola, G. Deganello, B. Pawelec, J.L.G. Fierro, J. Catal. 215
(2003) 317.
ꢅ The critical parameter when designing Au supported on HMS
catalyst is the exposure of Aud+ species. Indeed, a linear correla-
tion between this parameter and the activity, in terms of QTOF,
is presented.
ꢅ In the Au/HMS–Ce catalyst, Ce is not incorporated into the
framework of HMS leading to clusters of Au and CeO2. This
causes the difficulty of calculating accurate values of DpAu from
HRTEM.
ꢅ The modification of HMS by Ti does not lead to any enhance-
ment of the hydrogenation performance of the Au/HMS catalyst
due to the sintering of Ti (confirmed by XPS) which blocks part
of the Au surface.
[13] A.M. Venezia, V. La Parola, V. Nicoli, G. Deganello, J. Catal. 212 (2002) 56.
[14] A. Hugon, L. Delannoy, C. Louis, Gold Bull. 41 (2008) 127.
[15] X. Zhang, H. Shi, B. Xu, Angew. Chem. Int. Ed. 44 (2005) 7132.
[16] X. Zhang, H. Shi, B.-Q. Xu, Catal. Today 122 (2007) 330.
[17] L. Fu, N.Q. Wu, J.H. Yang, F. Qu, D.L. Johnson, M.C. Kung, H.H. Kung, V.P. Dravid,
J. Phys.Chem. B 109 (2005) 3704.
[18] X. Zhang, F.X. Llabrés i Xamena, A. Corma, J. Catal. 265 (2009) 155.
[19] R. Nava, J. Morales, G. Alonso, C. Ornelas, B. Pawelec, J.L.G. Fierro, Appl. Catal. A:
Gen. 321 (2007) 58.
[20] B. Pawelec, S. Damyanova, R. Mariscal, J.L.G. Fierro, I. Sobrados, J. Sanz, L.
Petrov, J. Catal. 223 (2004) 86.
[21] T.A. Zepeda, B. Pawelec, J.L.G. Fierro, T. Halachev, J. Catal. 242 (2006) 254.
[22] T.A. Zepeda, B. Pawelec, J.L.G. Fierro, T. Halachev, Appl. Catal. B: Environ. 71
(2007) 223.
ꢅ The Au/HMS–Fe is the most promising catalyst for the hydroge-
nation of aromatics due to its enhanced stability and second aro-
matic-ring hydrogenation rate.
[23] T. Chiranjeevi, P. Kumar, S.K. Maity, M.S. Rana, G.M. Dhar, T. Rao, Micropor.
Mesopor. Mater. 44 (2001) 547.
[24] N.Y. He, J.M. Cao, S.L. Bao, Q.H. Xu, Mater. Lett. 31 (1997) 133.
[25] H. Liu, G.Z. Lu, Y.L. Guo, Y. Guo, J.S. Wang, Micropor. Mesopor. Mater. 108
(2008) 56.
[26] K. Bachari, J.M.M. Millet, B. Benaichouba, O. Cherifi, F. Figueras, J. Catal. 221
(2004) 55.
[27] C. Milone, R. Ingoglia, L. Schipilliti, C. Crisafulli, G. Neri, S. Galvagno, J. Catal.
236 (2005) 80.
[28] S.C. Laha, P. Mukherjee, S.R. Sainkar, R. Kumar, J. Catal. 207 (2002) 213.
[29] B. Campo, M. Volpe, S. Ivanova, R. Touroude, J. Catal. 242 (2006) 162.
[30] P.T. Tanev, T.J. Pinnavaia, Science 267 (1995) 865.
[31] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992)
710.
[32] M.R. Riazi, Y.A. Roomi, Chem. Eng. Sci. 62 (2007) 6649.
[33] M.R. Riazi, J.H. Vera, Ind. Eng. Chem. Res. 44 (2005) 186.
[34] S. Toppinen, T. Salmi, T.K. Rantakyla, J. Aittamaa, Ind. Eng. Chem. Res. 36
(1997) 2101.
However, further investigation on the nature of the interaction
Au–CeO2 could yield more explanations with regard to the perfor-
mance of Au/HMS–Ce catalyst. The intrinsic activity per gmetal (ex-
posed, TOF; and total, QTOF) of all Au/HMS–M catalysts is
substantially lower than that of Pt–Pd-supported catalyst so that
new approaches are required for the application of Au-supported
catalyst in hydroprocessing of aromatics.
Acknowledgments
[35] P. Castaño, D. van Herk, M.T. Kreutzer, J.A. Moulijn, M. Makkee, Appl. Catal. B:
Environ. 88 (2009) 213.
[36] P.T. Tanev, M. Chibwe, T.J. Pinnavaia, Nature 368 (1994) 321.
[37] B.M. Weckhuysen, R.A. Schoonheydt, Catal. Today 49 (1999) 441.
[38] R.X. Li, S. Yabe, M. Yamashita, S. Momose, S. Yoshida, S. Yin, T. Sato, Mater.
Chem. Phys. 75 (2002) 39.
[39] Y. Wang, Q.H. Zhang, T. Shishido, K. Takehira, J. Catal. 209 (2002) 186.
[40] K. Bachari, J.M.M. Millet, P. Bonville, O. Cherifi, F. Figueras, J. Catal. 249 (2007)
52.
[41] S. Perathoner, P. Lanzafame, R. Passalacqua, G. Centi, R. Schlogl, D.S. Su,
Micropor. Mesopor. Mater. 90 (2006) 347.
[42] A.N. Pestryakov, V.V. Lunin, A.N. Kharlanov, N.E. Bogdanchikova, I.V.
Tuzovskaya, Europ. Phys. J. D 24 (2003) 307.
[43] P. Claus, A. Bruckner, C. Mohr, H. Hofmeister, J. Am. Chem. Soc. 122 (2000)
11430.
[44] F. Bauer, H.G. Karge, Characterization of coke on zeolites, in: H.G. Karge, J.
Weitkamp (Eds.), Molecular Sieves. Science and Technology, Characterization
II, Vol. 5, Springer-Verlag, Berlin-Heidelberg, 2007, p. 282.
[45] D.A. Bulushev, I. Yuranov, E.I. Suvorova, P.A. Buffat, L. Kiwi-Minsker, J. Catal.
224 (2004) 8.
[46] Z.Q. Zou, M. Meng, L.H. Guo, Y.Q. Zha, J. Hazar. Mater. 163 (2009) 835.
[47] P. Reyes, H. Rojas, J.L.G. Fierro, Appl. Catal. A: Gen. 248 (2003) 59.
[48] M.C. Capel-Sanchez, J.M. Campos-Martin, J.L.G. Fierro, M.P. de Frutos, A.P. Polo,
Chem. Commun. (2000) 855.
Financial support by CSIC, Comunidad de Madrid (project
CCG07-CSIC/ENE-1884) is gratefully acknowledged. T.A.Z. wishes
to express his gratitude to the CNyN-UNAM (Mexico) for a post-
doctoral bursary. The authors are grateful to G. Hosford, Dr. M.
Peña and Dr. R. Guil-López for their help.
References
[1] S.K. Samanta, O.V. Singh, R.K. Jain, Trends Biotechnol. 20 (2002) 243.
[2] P.J. Tsai, T.S. Shih, H.L. Chen, W.J. Lee, C.H. Lai, S.H. Liou, Atmos. Environ. 38
(2004) 333.
[3] A. Stanislaus, B.H. Cooper, Catal. Rev.Sci. Eng. 36 (1994) 75.
[4] G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng. 41 (1999) 319.
[5] M. Haruta, N. Yamada, T. Kobayashi, S. Iijima, J. Catal. 115 (1989) 301.
[6] M.P. Casaletto, A. Longo, A.M. Venezia, A. Martorana, A. Prestianni, Appl. Catal.
A: Gen. 302 (2006) 309.
[7] A.S.K. Hashmi, G.J. Hutchings, Angew. Chem. Int. Ed. 45 (2006) 7896.
[8] P. Claus, Appl. Catal. A: Gen. 291 (2005) 222.
[9] A.M. Venezia, V. La Parola, B. Pawelec, J.L.G. Fierro, Appl. Catal. A: Gen. 264
(2004) 43.
[10] B. Pawelec, A.M. Venezia, V. La Parola, S. Thomas, J.L.G. Fierro, Appl. Catal. A:
Gen. 283 (2005) 165.
[11] B. Pawelec, A.M. Venezia, V. La Parola, E. Cano-Serrano, J.M. Campos-Martin,
J.L.G. Fierro, Appl. Surf. Sci. 242 (2005) 380.
[49] J. Kanai, J.A. Martens, P.A. Jacobs, J. Catal. 133 (1992) 527.