Catalytic behavior of size-controlled palladium nanoparticles in the hydrodechlorination of 4-chlorophenol in aqueous phase

Article abstract of DOI:10.1016/j.jcat.2012.06.009

Unsupported Pd nanoparticles of controlled size were tested as catalyst in liquid-phase hydrodechlorination (303-323 K, 1 atm) using 4-chlorophenol (4-CP) as target compound. The Pd nanoparticles were synthesized by chemical reduction, using ethanol and m

Full text of DOI:10.1016/j.jcat.2012.06.009

Journal of Catalysis 293 (2012) 85–93
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Catalytic behavior of size-controlled palladium nanoparticles in the
hydrodechlorination of 4-chlorophenol in aqueous phase
⇑
J.A. Baeza, L. Calvo , M.A. Gilarranz, A.F. Mohedano, J.A. Casas, J.J. Rodriguez
Sección Departamental de Ingeniería Química, C/ Francisco Tomás y Valiente 7, Universidad Autónoma de Madrid, 28049 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 December 2011
Revised 5 June 2012
Accepted 8 June 2012
Available online 18 July 2012
Unsupported Pd nanoparticles of controlled size were tested as catalyst in liquid-phase hydrodechlorina-
tion (303–323 K, 1 atm) using 4-chlorophenol (4-CP) as target compound. The Pd nanoparticles were syn-
thesized by chemical reduction, using ethanol and methanol as reducing agents and poly(N-vinyl-2-
pyrrolidone) (PVP) as capping agent. The size of the nanoparticles and the Pdn+/Pd0 ratio decreased with
increasing alcohol concentration and PVP/Pd ratio, both being lower for ethanol medium.
High 4-CP conversion values (80–100%) were achieved at low Pd concentration (2.45 ꢀ 10ꢁ3 g/ L).
Phenol was the only reaction by-product detected in contrast to the previous results with supported
Pd catalysts, where the active phase–support interaction in 4-CP HDC led to obtain also cyclohexanone
and cyclohexanol as by-products in equivalent experimental conditions. The smaller nanoparticles
showed higher activity due to the higher available surface (m2/gcat). Thus, the smaller nanoparticles syn-
thesized in ethanol medium ranged between 2.7 and 2.8 nm and yielded activity values between 16.7 and
39.1 mmol/gcat min, whereas the smaller particles obtained in methanol medium were in the 3.1–4.2 nm
range and exhibited activity values of 20.1–25.7 mmol/gcat min. However, the large nanoparticles exhib-
ited higher activity per unit of catalyst surface, for example, 0.34–0.43 mmol/min m2 in the case of those
synthesized in methanol medium. On the other hand, higher activity was observed for the nanoparticles
synthesized in methanol medium when equivalent nanoparticle size was compared. Activation energy
values around 100 kJ/mol were obtained for Pd nanoparticles of different characteristics, significantly
higher than the value reported for supported Pd catalysts.
Keywords:
Colloidal nanoparticles
Palladium
Particle size
Hydrodechlorination
4-Chlorophenol
Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction
Thus, HDC can be used as a detoxifying stage prior to a biological
treatment.
In the last years, catalytic hydrodechlorination (HDC) has
received increasing attention for the treatment of wastewaters
containing chlorinated organic compounds since it shows some
advantages when compared to other techniques of potential appli-
cation for these hazardous pollutants. HDC can operate at mild
conditions [1–3], so that it does not require high temperatures
and/or pressures as in the case of incineration or wet oxidation
processes. Large amounts of reagents, such as in Fenton oxidation,
are not needed and can be applied within a wide range of concen-
trations of chlorinated compounds, which is not the case of biolog-
ical methods [4]. Though HDC does not provide the complete
destruction of the pollutants, it can lead to a convenient transfor-
mation of them into substantially less harmful species [2,4,5].
Transition and noble metal-based catalysts have been com-
monly used in liquid-phase HDC [6–14]. Among those metals, Pd
has been the most extensively used because of its singular catalytic
properties [15] that lead to high dechlorination efficiencies [16–
18]. These specific properties are affected by a number of factors
such as the metallic particle size, shape, oxidation state and inter-
action with the support, among others [19–21]. Particularly, metal
particle size is crucial for the performance of the catalyst in struc-
ture-sensitive reactions. A number of authors have concluded that
HDC reaction is structure sensitive [22–28], although there is no
general agreement with respect to this topic [29]. Thus, in order
to design efficient catalysts, it is necessary a previous in-depth
knowledge about how this factor affects the catalytic properties.
In general, it is widely accepted that decreasing the particle size
of the metallic phase increases the catalytic activity because of a
higher surface-to-volume ratio although some limit has been
claimed for this effect [30,31]. Moreover, the nature and the
strength of bonding between the molecules adsorbed and the me-
tal particles is affected by the number of particles surrounding a
surface metal atom (electronic or ligand effect). Thus, adsorption
⇑
Corresponding author. Fax: +34 914973516.
E-mail addresses: josealberto.baeza@uam.es (J.A. Baeza), luisa.calvo@uam.es
(L. Calvo), miguel.gilarranz@uam.es (M.A. Gilarranz), angelf.mohedano@uam.es
(A.F. Mohedano), jose.casas@uam.es (J.A. Casas), juanjo.rodriguez@uam.es
(J.J. Rodriguez).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.06.009

86
J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
Nomenclature
a
catalytic activity (mmol/gcat min)
4-CP molar concentration (mmol/L)
ꢁr4-cp
Sa
t
4-CP conversion rate (mmol/L min)
C4-cp
ds
k
specific surface area (m2/g)
time (min)
surface-area-weighted average mean diameter (m)
rate constant (minꢁ1
)
qpd
Pd density (g/m3)
mpd
Pd nanoparticles mass (g)
energies could change with particle size, leading to poisoning of
the catalyst or activation of the substrate, depending on the nature
of the adsorption of the molecule on the particle surface [15]. It is
important to remark that the catalysts commonly studied have
their metallic phase dispersed at nanometric scale, but there is
no precise control on the nanoparticle size upon the synthesis.
Therefore, there is a remarkable interest in the literature in design-
ing catalysts with controlled nanoparticle size and size distribution
[32–35]. However, it must be emphasized that the influence of
nanoparticle size on the catalytic activity is not well defined so
far, since decreasing the particle size not always increases the reac-
tion rates evaluated per unit mass of nanoparticle. Thus, when a
certain minimum particle size is reached, an additional decrease
produces an increase in the ionization energy and, as a result, an
increase in the adsorption energy. The resulting stronger adsorp-
tion of intermediate species on the particle surface can provoke
poisoning of active sites of the catalyst, which diminishes the
catalytic activity [15,36].
Some research efforts have been focused on developing meth-
ods of synthesis of nanoparticles with the objective of controlling
their size within a narrow distribution in order to investigate
chemical and physical size-dependent properties [37–39]. Among
the methods of synthesis of noble metals nanoparticles that have
been reported in the literature, chemical [40,41], electrochemical
[42], sonochemical [43] and microbial-based methods [44] are
the most relevant. Chemical reduction is considered as a simple
method based on the reduction of a metal precursor salt in the
presence of a reducing agent (hydrazine, pyrogallol, potassium
borohydride, ethanol, methanol, propanol) and a protective or cap-
ping agent (poly(N-vinyl-2-pyrrolidone), polyvinyl alcohol, dode-
cylamine). The last inhibits aggregation and is responsible for the
controlled growth and shape of the particles. This method has been
claimed in some works [45–48] for the design of size-controlled
nanoparticles whose catalytic activities have been analyzed in dif-
ferent reactions. For example, Narayanan and El-Sayed [36] syn-
thesized size-controlled Pd nanoparticles that were supported on
activated carbon and tested them for the Suzuki reaction between
phenylboronic acid and iodobenzene for producing biphenyl. In an-
other related work, Joo et al. [48] studied the effect of particle size
on CO oxidation over Ru nanoparticles catalysts. Nevertheless,
investigation of size-controlled nanoparticles applications in other
types of catalytic reactions is scarce, and thus, so far there is no still
a comprehensive knowledge on how particle size influences the
catalytic activity in reactions such as liquid-phase HDC.
40.000, Sigma–Aldrich Co.) was used as capping agent to avoid
uncontrolled growing and inhibit aggregation of Pd nanoparticles.
Methanol (>99%) and ethanol (>99%), used as reducing agents,
were purchased from Panreac Quimica S.A.U. and Merck KGaA,
respectively. Hydrogen (>99%) was supplied by Praxair. All the
reagents were used as received without further purification.
2.2. Synthesis and characterization of Pd nanoparticles
Pd (II) chloride (0.5 mmol), 0.2 N HCl (1 mmol) and distilled
water (250 mL) were mixed to obtain a H2PdCl4 (2 ꢀ 10ꢁ3 M)
pale-yellow aqueous solution. An aliquot of 30 mL of the
2 ꢀ 10ꢁ3 M H2PdCl4 solution was mixed with 70 mL of alcohol/
water solution and PVP. The mixture was refluxed in a 250-mL flask
connected to a Liebig condenser for 3 h under atmospheric pressure
[37]. The values of PVP/Pd ratio were varied between 5 and 40 (mol/
mol), and different mixtures of alcohol/water (10–70%, v/v) were
used in order to learn the effect of these variables on the size and
size distribution of Pd nanoparticles. Finally, the colloidal disper-
sion of nanoparticles (100 mL) was concentrated up to a total
volume of approximately 10 mL in a rotary evaporator (Büchi).
The Pd nanoparticles synthesized were characterized by trans-
mission electron microscopy (TEM) at 400 kV (JEOL, mod. JEM-
4000 EX). Samples were prepared placing a drop of colloidal Pd
nanoparticles onto a carbon-coated copper grid and lefting it dry
at room temperature. Software ‘ImageJ 1.44i’ was used for
counting and measuring particles on digital TEM images (more
than 200 nanoparticles were measured per image). Surface-
P
P
area-weighted mean diameters ðds ¼ nid3= nid2Þ and size
i
i
distribution, characterized by the standard deviation
ðrs ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
2
ðdi ꢁ dsÞ =nÞ, were calculated as described elsewhere [49].
Likewise, the catalysts were characterized by XPS (Physical Elec-
tronics, mod. ESCA 5701 equipped with a Mg K
a X-ray excitation
source, 1253.6 eV). Software ‘Multipak v8.2b’ was used for spectro-
grams deconvolution in order to determine both Pd electrodefi-
cient and zero-valent species in the particle surface. The analysis
procedure involved smoothing, a Shirley background subtraction
and mixed Gaussian–Lorentzian by a least-square method curve
fitting as described in a previous work [50]. C 1s peak (284.6 eV)
was used as internal standard for binding energies corrections
due to sample charging. Doublet separation for Pd 3d was
5.26 eV as described elsewhere [51]. Deconvoluted peaks showed
binding energies in the range of 334.8–335.8 eV and 336.2–
337.1 eV for Pd 3d5/2, which can be attributed to metallic Pd
(Pd0) and electrodeficient Pd (Pdn+), respectively. All these data val-
ues are in good agreement with those reported in NIST X-ray Pho-
toelectron Spectroscopy Database [51]. A probing depth of several
nanometers can be assumed; therefore, the surface Pdn+/Pd0 ratio
calculated should be essentially the same as the bulk ratio for
the particles studied.
The aim of this work is to address the effect of the synthesis
conditions and the size of Pd nanoparticles on their activity in 4-
CP HDC in aqueous phase. The Pd nanoparticles were prepared
with poly(N-vinyl-2-pyrrolidone) (PVP) as capping agent and eth-
anol or methanol as reducing agent.
2. Experimental
2.1. Materials
2.3. HDC experiments
Sigma–Aldrich PdCl2 (99%) was used as Pd nanoparticles pre-
cursor. Poly(N-vinyl-2-pyrrolidone) (average molecular weight:
Catalytic HDC runs were carried out during 4 h in a three-
necked jacketed glass reactor where H2 was continuously fed at

J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
87
50 N mL/min flow rate under vigorous stirring (600–800 rpm) in
order to facilitate hydrogen distribution through the 4-CP solution
(150 mL). The reaction temperature (303–323 K) was controlled by
a thermostatic bath connected to the reactor jacket. The system
was equipped with a cold trap at the vent to check any possible
stripping; no significant stripping was detected. The initial concen-
tration of 4-CP was 100 mg/L, and the catalyst load was
2.45 ꢀ 10ꢁ3 g/L of Pd.
It has been reported that the role of PVP is to act as a protecting
agent, allowing controlled particles growth in a reactor of a nano-
sized water pool wrapped with PVP, preventing particles aggrega-
tion [52,53]. Increasing the PVP/Pd molar ratios favors the forma-
tion of micellar structures that surround the particles and control
their growth.
3.1.3. Effect of the reducing agent
A sampling system consisting in a thin tube connected to a syr-
inge allowed taking liquid samples (1 mL), which were collected
The alcohol used as reducing agent showed a crucial effect on
the size of the nanoparticles, as can be seen in Table 1. Fig. 3 shows
that, for a PVP/Pd molar ratio of 5, the ethanol concentration did
not show a well-defined influence in the range of 10–40% where
mean diameters and size distributions within 3.6–4.7 nm and
1.0–1.9 nm, respectively, were observed. However, a significant in-
crease took place in 70% ethanol medium up to 22 nm mean parti-
cle diameter. A different behavior was observed with methanol as
reducing agent. For a PVP/Pd molar ratio of 20, the nanoparticle
mean diameter decreased when the methanol ratio increased from
10% to 40% where a minimum size of 3.8 nm was reached without
any further decrease at higher methanol concentration. As re-
ported in the literature [37], two main effects lead to the genera-
tion of smaller Pd nanoparticles in relation to the concentration
of reducing agent. Firstly, a higher concentration of the reducing
agent increases the reduction rate of [PdCl4]2ꢁ to Pd0, provoking
the generation of smaller nanoparticles because of the occurrence
of a higher number of Pd nuclei. This effect seems to prevail at eth-
anol and methanol concentrations within 10–40% and 10–55%,
respectively. Secondly, while PVP is dissolved in both alcohol and
water, H2PdCl4 shows decreasing solubility at high alcohol concen-
tration, so it does not disperse homogeneously in a mixed water–
alcohol solution at the highest alcohol ratios tested. Therefore,
nanoparticle mean diameter and size distribution become larger
and wider, respectively, at high alcohol ratios. This seems to be
the prevailing effect at alcohol concentrations beyond 40%.
The smallest nanoparticles were obtained at different alcohol
ratios and PVP/Pd molar ratios for each of the alcohols used. As
shown in Table 1, under identical synthesis conditions (alcohol
concentration and PVP/Pd ratios), Pd nanoparticles synthesized
with ethanol as reducing agent were smaller. The effect of the
reducing agent on the particle size has been discussed in the liter-
ature. Teranishi et al. [37] suggested a faster reduction of [PdCl4]2ꢁ
by alcohols with higher boiling point, providing a higher number of
Pd nuclei. Thus, alcohols of higher boiling point lead to Pd nanopar-
ticles of lower size. Likewise, they indicated that ethanol can be
oxidized more easily than methanol according to the hydrogena-
tion enthalpies of the corresponding aldehydes. Our results are
consistent with this explanation, since all of the nanoparticles
through previous filtering (PTFE filter, pore size 0.45 lm). The sam-
ples were analyzed by high-performance liquid chromatography
(Varian Prostar equipped with a UV–VIS detector) using a C18 col-
umn as stationary phase and a mixture of acetonitrile and water
(1:1, v/v) as mobile phase. Cyclohexanone and cyclohexanol were
analyzed by a GC/FID device (GC 3900 Varian) using a 30 m
long ꢀ 0.25 mm i.d. capillary column (CP-Wax 52 CB, Varian) and
nitrogen as carrier gas. The quantification of chloride ion was per-
formed by ion chromatography (Metrohm 790 Personal IC). The
carbon and chlorine balances matched always between 90% and
95%, respectively, at the least.
3. Results and discussion
3.1. Pd Particle size and size distribution
3.1.1. TEM characterization
Table 1 shows the mean diameters and size distributions of Pd
nanoparticles prepared under different conditions. Fig. 1 shows, as
representative examples, TEM images of four selected series. The
nanoparticles appeared mostly spherical although some differ-
ences in shape were found in the case of the nanoparticles synthe-
sized in 70% ethanol (series 4), with polyhedral shapes prevailing
over spheres. These nanoparticles showed the largest size with a
mean diameter of 22.1 nm.
3.1.2. Effect of the capping agent on nanoparticles size
The influence of the capping agent on the particle size can be
observed in Fig. 2. When an intermediate concentration of 25% eth-
anol was used, an increase in the PVP/Pd molar ratio from 5 to 20
led to a reduction in both particle size and size distribution. The
same trend was observed with a concentration of 10% methanol
but within a wider range of the PVP/Pd molar ratio (5–35). At high-
er PVP/Pd molar ratios, no additional decrease in nanoparticle size
and size distribution was found.
Table 1
Mean diameter (ds), size distribution (
r
s), Pdn+/Pd0 ratio, specific surface area and activity (a) of the Pd nanoparticles prepared under different conditions
Vol alcohol (%) PVP/Pd (mol/mol) ds (nm) rs (nm) Pdn+/Pd0 ratio Specific surface area (m2/gcat
Pd nanoparticles series Alcohol
)
a (mmol/gcat min)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Methanol 10
Methanol 10
Methanol 10
Methanol 10
Methanol 25
Methanol 40
Methanol 55
10
25
40
70
10
25
25
40
55
5
5
5
4.7
3.6
4.0
22.1
3.5
2.7
2.7
2.8
2.8
12.9
10.9
3.1
3.6
8.5
1.0
1.2
1.9
10.4
0.9
0.7
0.8
0.8
0.4
7.0
2.9
0.6
0.7
4.6
0.8
1.2
106
138
125
23
143
184
181
178
178
39
7.4
10.1
4.8
0.67
0.35
2.79
0.48
5
3.5
20
20
35
20
20
5
20
35
40
20
20
20
14.6
19.4
39.1
16.7
17.8
16.8
17.2
25.3
20.1
19.8
25.7
23.7
0.54
1.46
1.05
0.48
0.51
1.03
0.56
0.39
46
161
139
59
131
119
3.8
4.2

88
J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
40
(a)
Series 1
35
30
25
20
15
10
5
0
0
3
6
9 12 15 18 21 24 27 30
Size (nm)
40
35
30
25
20
15
10
5
(b)
Series 7
0
0
3
6
9 12 15 18 21 24 27 30
Size (nm)
40
35
30
25
20
15
10
5
Series 10
(c)
0
0
3
6
9 12 15 18 21 24 27 30
Size (nm)
40
35
30
25
20
15
10
5
Series 11
(d)
0
0
3
6
9 12 15 18 21 24 27 30
Size (nm)
Fig. 1. TEM images and size distributions of four samples of Pd nanoparticles prepared in ethanol (a and b) and methanol (c and d) water media.

J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
330 332 334 336 338 340 342 344 346 348
89
(a)
4
3
2
1
0
Pd0
3d5/2
series 12
Pd0
3d3/2
Pdn+
3d5/2
Pdn+
3d3/2
16 (b)
series 11
14
12
10
8
Pd0
Pdn+
3d5/2
3d5/2
Pd0
3d3/2
Pdn+
3d3/2
6
4
2
0
0
5
10 15 20 25 30 35 40 45 50
PVP/Pd ratio
330 332 334 336 338 340 342 344 346 348
Binding Energy (eV)
Fig. 2. Mean diameter (d) and size distribution (s) of Pd nanoparticles obtained at
different PVP/Pd ratios. (a) Ethanol 25 vol.% and (b) methanol 10 vol.%.
Fig. 4. XPS spectra of Pd nanoparticles of series 11 and 12 (Table 1).
(a)
25
3.2. XPS characterization
XPS was used to establish the oxidation state of Pd. The Pdn+/Pd0
ratios obtained from the deconvolution of the XPS Pd 3d region
spectra are summarized in Table 1. The presence of both Pdn+
and Pd0 was observed in all the Pd nanoparticles series analyzed.
Gomez-Sainero et al. [54] reported that both electrodeficient and
zero-valent species are required for carbon tetrachloride HDC on
Pd/AC in gas phase, whereas catalysts with only one of the species
(Pdn+ or Pd0) remained inactive. In this sense, they proposed a dual
nature of Pd active sites by the association of both species, reaching
a maximum of activity at a Pdn+/Pd0 ratio close to 1. Other authors
were also in agreement with the proposed dual nature of the Pd ac-
tive sites in HDC in gas phase [50]. Fig. 4 shows the Pd 3d decon-
volved spectra for the nanoparticles of series 11 and 12. Zero-
valent Pd is the prevailing species in series 12, whereas quite sim-
ilar proportions of electrodeficient and zero-valent Pd were found
for series 11. As can be seen in Fig. 5, there is a relationship be-
tween the Pdn+/Pd0 ratio values and the mean particle diameter.
Three different patterns can be distinguished independently of
the alcohol used as reducing agent. The smaller particles (2.7–
4.2 nm) show Pdn+/Pd0 ratios between 0.35 and 0.67, whereas for
larger particles (8.5–12.9 nm), that ratio is between 1.03 and
1.46. Oversized particles (22.1 nm) show Pdn+/Pd0 ratio value of
2.79. This observation is consistent with the previous results re-
ported in the literature, where small particles had a lower Pdn+/Pd0
ratio than large ones [55]. The differences found in Pdn+/Pd0 ratios
can be related with the higher surface/volume ratio of smaller par-
ticles, which provides a higher surface per unit mass of Pd to be re-
duced by the reducing agent, thus leading to a lower Pdn+/Pd0 ratio
of small particles compared to larger ones. However, due to the dif-
ficulty to address where Pdn+ and Pd0 species are predominantly
located in Pd nanoparticles, further research is needed.
20
15
10
5
0
(b)
12
10
8
6
4
2
0
0
5
10 15 20 25 30 35 40 45 50 55 60 65 70 75
Alcohol Volume %
Fig. 3. Mean diameter (j) and size distribution (h) of Pd nanoparticles prepared
with (a) ethanol (PVP/Pd ratio = 5 mol/mol) and (b) methanol (PVP/Pd ratio = 20 -
mol/mol).
synthesized in ethanol medium were of lower size than those
synthesized in methanol medium under the same synthesis
conditions.

90
J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
3.0
2.8
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
(a)
100
80
60
40
20
Series 1
Series 2
Series 6
Series 4
0
(b)
100
80
60
40
20
0
2
4
6
8
10 12 14 16 18 20 22 24
dS(nm)
Series 10
Series 13
Series 12
Series 14
Fig. 5. Pdn+/Pd0 ratio versus nanoparticle mean diameter: methanol (j) and
ethanol series (s).
3.3. HDC activity of Pd nanoparticles
The HDC runs were carried out at variable concentrations of PVP
in the reactor. Preliminary experiments carried out in the absence
of catalyst proved that there was no reaction between PVP, 4-CP
and H2. Moreover, a control experiment was carried out adding
to the reaction medium a dose of PVP ten times higher than that
needed for the synthesis of the Pd nanoparticles with the aim of
addressing the effect of the excess of PVP on the catalytic activity.
The results showed no significant differences in activity between
the series 10 of Pd nanoparticles (16.8 mmol/gcat min) and the
same particles series in the control experiment (16.5 mmol/
gcat min). Thus, no synergistic or inhibitory effect of PVP was
observed. These results are in good agreement with those reported
by Hirai and Yakura [56] on the study of activity of Pd
nanoparticles of different size protected by PVP layers of different
thickness. These authors concluded that activity depends mainly
on specific surface area and not on PVP concentration. Chemical
control of the reaction has been checked experimentally by using
different concentrations of catalyst in 4-CP HDC reaction; no signif-
icant differences in the HDC rates per unit mass of catalyst were
observed.
Fig. 6 shows the evolution of 4-CP conversion upon reaction
time with some Pd nanoparticles series. High 4-CP conversion val-
ues were achieved (80–100%) in all cases. Two different behaviors
were observed in relation with the reducing agent used. In the case
of Pd nanoparticles from the methanol series, changes in particle
size were accompanied by small differences in activity. In contrast,
in the ethanol series, the activity increased when decreasing the
particle size.
The nanoparticle size affects the nature and strength of the
interaction of 4-CP with the Pd surface since it determines the
number of particles surrounding each surface metal atom. More-
over, the coordination of surface atoms depends on the particle
size. Thus, smaller particles contain more surface atoms residing
on the edges of the crystallographic planes or on edge junctions,
increasing the number of active sites which could be accessible
to the 4-CP molecules [15]. Additionally, Diaz et al. [57] indicated
that structure sensitive appears as a determining factor on the
HDC activity. The results reported suggested that a metal particle
size within the 3–4 nm range can be considered as the optimum
for Pd-based catalysts. Contradictory results have been found in
the literature about the most favorable catalyst site (terrace vs.
edges/corners) in HDC. A number of works have considered the
edges/corners as the most favorable catalyst sites in HDC since a
high activity seems to be related with the number of surface Pd
0
50
100
150
Time (min)
200
300
Fig. 6. HDC of 4-CP with Pd nanoparticles (2.45 ꢀ 10ꢁ3 g/L) of: ethanol (a) and
methanol series (b) (Table 1).
atoms of low coordination placed at the edges and corners of Pd
particles [58–60], being such atoms more abundant in smaller par-
ticles. This interpretation has also been reported for other catalytic
reactions such as Suzuki–Heck reaction [61,62]. Nevertheless,
other authors have indicated that large ensembles or single atoms
can be considered as the most favorable catalytic sites for HDC
[63,64]. Although studies about the mechanism of liquid-phase
HDC of 4-CP in the presence of Pd are not available, Orlov et al.
[65] reported structure-dependent adsorption of 4-CP on a poly-
crystalline TiO2 surface. Likewise, Solomon et al. [66] found differ-
ences in phenol adsorption varying from physical adsorption on
clean Ag surface to chemisorption on oxygenated Ag surface. Thus,
the adsorption pattern resulting from the different Pd arrange-
ments may have an important role in catalytic activity.
A relevant result is that phenol was the only HDC product de-
tected. Previous works in the literature showed high catalytic
activities for supported Pd catalysts in the HDC of chlorophenols
[67]. Previous works on the use of Pd on Al2O3 under mild condi-
tions [68,69] and Pd on activated carbon at higher temperature
[4] reported complete conversion of 4-CP, yielding cyclohexanone
and cyclohexanol as reaction products in addition to phenol. The
concentration of Pd used in those works was several times higher
than ours. Thus, it is remarkable that, in the operating conditions
tested, the Pd nanoparticles used in the current work did not show
activity for phenol hydrogenation in contrast to the behavior re-
ported for supported Pd catalysts. To learn more on this fact, some
phenol hydrogenation experiments were carried out at the same
experimental conditions than for 4-CP HDC, using a 100 mg/L phe-
nol aqueous solution, and no detectable phenol hydrogenation was
observed after 4 h reaction time. Thus, the support must play a key
role in supported Pd catalysts, resulting in phenol hydrogenation to
by-products, such as cyclohexanone and/or cyclohexanol. The sup-
port can affect the selectivity of reactions due to its interaction
with the metallic active phase, provoking changes in the electronic
density of Pd nanoparticles that can affect the strength of
adsorption on the metal sites [69].

J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
91
Table 2
0.8
(a)
Pseudo-first-order rate constant and apparent activation energies values (Ea) for 4-CP
HDC with Pd nanoparticles
0.6
0.4
0.2
0.0
0.8
0.6
0.4
0.2
0.0
0.8
0.6
0.4
0.2
0.0
Pd
Reaction
temperature
(K)
Rate constant r2
(minꢁ1
Ea
(kJ/mol)
r2
nanoparticles
series
)
303
308
313
0.026 0.004 0.990
0.057 0.007 0.996 100 15 0.953
0.090 0.013 0.995
2
3
303
310
313
0.011 0.002 0.989
0.030 0.002 0.997 105
0.042 0.004 0.997
(b)
(c)
6
3
1
3
0.994
0.998
0.999
0.999
303
310
313
0.072 0.008 0.995
0.181 0.020 0.998 106
0.278 0.038 0.998
12
11
15
303
313
323
0.026 0.004 0.990
0.093 0.010 0.999 100
0.310 0.014 0.998
303
308
310
0.058 0.013 0.985
0.129 0.025 0.992 118
0.261 0.020 0.999
A simple pseudo-first-order kinetic equation was checked to
describe the rate of 4-CP disappearance:
0
50
100
150
200
250
300
ꢁdC4-cp
ðꢁr4-cpÞ ¼
¼ k ꢀ C4-cp
ð1Þ
Time (min)
dt
Fig. 7. Time evolution of (j) 4-CP and (s) phenol in 4-CP HDC with Pd
nanoparticles series 11. Reaction temperature: (a) 303 K, (b) 313 K and (c) 323 K.
Predicted curves in dashed lines.
t ¼ 0; C4-cp ¼ C0
Table 2 summarizes the results obtained by fitting the experi-
mental 4-CP concentration–time curves. Besides the rate constants
values, Table 2 includes the correlation coefficients and the appar-
ent activation energy as calculated from the Arrhenius equation.
Fig. 7 depicts, as representative example, the validation of the rate
curves for Pd nanoparticles of series 11. Apparent activation energy
values within the range of 100–118 kJ/mol were obtained and did
not show dependence on nanoparticle size. These activation energy
values are significantly higher than those reported in the literature
for the HDC of chlorophenols with supported Pd catalysts (21–
58 kJ/mol) [17,68,70,71]. This fact was also reported by Schwab
[72], who indicated that activation energies for formic acid decom-
position were significantly higher for bulk metals than for sup-
ported metals (Cu/MgO, Ag/MgO and Ni/Al2O3) due to an effect
of metal-support electronic movement, even in supports with
low conductivity. Therefore, the supports seem to play an impor-
tant role in the HDC pathway that modifies not only the selectivity
but also the activation energy.
40
35
30
25
20
15
10
5
0
0
5
10
15
20
25
d (nm)
s
The activity (a) of Pd nanoparticles was evaluated from the
pseudo-first-order rate constant of 4-CP disappearance and was
normalized taking into account the Pd dose used in each
experiment:
Fig. 8. Catalytic activity of Pd nanoparticles versus mean diameter: (j) methanol
and (s) ethanol series.
media. The particles synthesized in methanol medium showed
a higher activity for equivalent particle size and also a less pro-
nounced decrease with increasing size. Therefore, not only the
particle size but also the synthesis conditions are relevant for
activity. In this sense, the type of reducing agent seems to be
an important factor.
Other additional effects reported in the literature in relation to
particle size and catalytic activity are the differences in hydrogen
solubility between larger (better solubility) and smaller (worse sol-
ubility) particles [73–75], and the better resistance of larger Pd
particles to inhibition/poisoning by chlorine [30,64,76]. Thus, the
higher activity per unit surface area of larger particles found in
the present work could be related to a combination of these
factors. Changes in coordination number with size should also be
taken into account since larger particles could present higher coor-
dination number, thus favoring the bonding strength of reactants
0:150L
mpdðgÞ
a ¼ k ꢀ
ð2Þ
The available catalytic surface was estimated from the specific
surface area and the Pd dose used in each experiment. The specific
surface area of Pd nanoparticles was calculated assuming spherical
shape from the equation [3].
6
Sa ¼
ð3Þ
ds ꢀ qpd
Table 1 shows the specific surface area and catalytic activity
values of the Pd nanoparticles tested in the HDC runs. A decrease
in nanoparticle size is associated with a higher surface area giv-
ing rise to a higher activity. Fig. 8 shows this relationship for the
two series of nanoparticles prepared in ethanol and methanol

92
J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
In the case of ethanol series, the influence of Pdn+/Pd0 ratio was
not so clearly observed, which can be attributed to the narrower
Pdn+/Pd0 ratio range provided by the Pd nanoparticles synthesis
method based in the reduction with ethanol.
4. Conclusions
Pd nanoparticles have been synthesized and tested as catalysts
in HDC of 4-CP at mild conditions (303–323 K, 1 atm). The load of
PVP, the alcohol used as reducing agent and its concentration affect
the particle size. Moreover, it seems that there is also a relationship
between that particle size and Pdn+/Pd0 ratio, the smaller particles
showing Pdn+/Pd0 ratios around 0.5. The Pd nanoparticles are fairly
active for 4-CP HDC, but further phenol hydrogenation was not ob-
served opposite to the reported for supported Pd catalysts.
Decreasing particle size increases catalytic activity, which can be
mostly attributed to the higher surface area. However, the larger
particles showed a more active surface looking at the activity per
unit surface area in the case of the methanol series. The higher
activity per unit surface area of larger particles may be related to
their Pdn+/Pd0 ratio and to coordination of Pd atoms and lower
strength adsorption of reactants and products onto them. At equal
others synthesis conditions, the nanoparticles synthesized using
methanol as reducing agent were larger than those prepared with
ethanol and at similar particle size, and the former showed a high-
er activity. 4-CP disappearance upon HDC can be well described by
a simple pseudo-first-order rate equation. Apparent activation en-
ergy values within the range of 100–118 kJ/mol were obtained for
4-CP disappearance, being significantly higher than the reported
for supported Pd catalyst, indicating some important role of the
supports on the reaction pathway.
0
20 40 60 80 100 120 140 160 180 200 220
Catalyst total available surface (m2/gcat
)
Fig. 9. Catalytic activity per unit surface area versus catalysts total available
surface: (j) methanol and (s) ethanol series.
and products, avoiding the inhibition/poisoning by chlorine of the
catalysts and taking advantage of the higher solubility of H2 as
well.
Fig. 9 shows that as a general trend, the nanoparticles of lower
diameter, whose total available surface is higher, exhibit a lower
catalytic activity per unit of surface area, the trend being more
clear in the case of methanol series. Therefore, the nanoparticles
of higher diameter seem to have a more active surface for HDC.
The ethanol series did not show significant differences in catalytic
activity per unit surface area; besides, such activity values were
lower than for the methanol series. Moreover, as can be seen in
Fig. 10, in the case of the methanol series, a positive effect of the
increase in Pdn+/Pd0 ratio on the catalytic activity per unit surface
area can be observed. The values of Pdn+/Pd0 ratio of the particles
with a higher activity per unit surface area were close to the opti-
mum reported in the literature [5,54,77,78]. In the ethanol series,
the synthesis method used provided Pdn+/Pd0 ratios far from that
optimum. Thus, Pd nanoparticles with extreme values of Pdn+/Pd0
ratio within the range studied presented a low activity per unit
surface area. Although activity should peak in Fig. 10 according
to the literature [54], the synthesis method used in this work did
not allow obtaining a Pdn+/Pd0 ratio within the range suitable to
verify this relationship.
Acknowledgments
We greatly appreciate financial support from the Spanish MCYT
(CTQ2009-09983 and SD2006-44) and CAM (REMTAVARES S-2009/
AMB-1588 and CCG10-UAM/AMB-5521). J.A. Baeza thanks the
Spanish MICINN a research Grant (BES-2010-030059).
References
Therefore, the increase in activity per unit surface with increas-
ing nanoparticle diameter in the case of the methanol series may
be related to changes in the coordination of Pd atoms and to a
lower interaction potential with reactants and products, which
reduces adsorption strength.
[1] G. Yuan, M.A. Keane, Catal. Today 88 (2003) 27.
[2] C. Xia, Y. Liu, S. Zhou, C. Yang, S. Liu, J. Xu, J. Yu, J. Chen, X. Liang, J. Hazard.
Mater. 169 (2009) 1029.
[3] G. Yuan, M.A. Keane, Chem. Eng. Sci. 58 (2003) 257.
[4] L. Calvo, M.A. Gilarranz, J.A. Casas, A.F. Mohedano, J.J. Rodríguez, Appl. Catal., B
67 (2006) 68.
[5] K.V.R. Chary, P.V.R. Rao, V. Vishwanathan, Catal. Commun. 7 (2006) 974.
[6] S. Kovenklioglu, Z. Cao, D. Shah, R.J. Farrauto, E.N. Balko, AIChEm J. 38 (1992)
1003.
[7] J.B. Hoke, G.A. Gramiccioni, E.N. Balko, Appl. Catal., B 1 (1992) 285.
[8] R. Muftikian, Q. Fernando, N. Korte, Water Res. 29 (1995) 2434.
[9] G.V. Lowry, M. Reinhard, Environ. Sci. Technol. 33 (1999) 1905.
[10] T.T. Bovkun, Y. Sasson, J. Blum, J. Mol. Catal. A: Chem. 242 (2005) 68.
[11] W. Wu, J. Xu, R. Ohnishi, Appl. Catal., B 60 (2005) 129.
[12] K. Mackenzie, H. Frenzel, F.D. Kopinke, Appl. Catal. B: Environ. 63 (2006) 161.
[13] L. Calvo, M.A. Gilarranz, J.A. Casas, A.F. Mohedano, J.J. Rodríguez, Appl. Catal., B
78 (2008) 259.
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
´
´
[14] M. Bonarowska, Z. Kaszkur, L. Ke˛pinski, Z. Karpinski, Appl. Catal., B 99 (2010)
248.
[15] A.Y. Stakheev, I.S. Mashkovskii, G.N. Baeva, N.S. Telegina, Russ. J. Gen. Chem. 80
(2010) 618.
[16] F. Murena, F. Gioia, Catal. Today 75 (2002) 57.
[17] G. Yuan, M.A. Keane, Catal. Commun. 4 (2003) 195.
[18] L. Calvo, A.F. Mohedano, J.A. Casas, M.A. Gilarranz, J.J. Rodríguez, Carbon 42
(2004) 1377.
[19] B.R. Cuenya, Thin Solid Films 518 (2010) 3127.
[20] J. Jolivet, S. Cassaignon, C. Chanéac, D. Chiche, O. Durupthy, D. Portehault, C. R.
Chim. 13 (2010) 40.
0.0
0.5
1.0
1.5
Pdn+/Pd0
2.0
2.5
3.0
[21] J. Park, J. Joo, S.G. Kwon, Y. Jang, T. Hyeon, Angew. Chem., Int. Ed. 46 (2007)
4630.
[22] M.A. Keane, G. Pina, G. Tavoularis, Appl. Catal., B 48 (2004) 275.
Fig. 10. Catalytic activity per unit surface area vs. Pdn+/Pd0 ratio: (j) methanol and
(s) ethanol series.

J.A. Baeza et al. / Journal of Catalysis 293 (2012) 85–93
93
[23] S. Gómez-Quero, F. Cárdenas-Lizana, M.A. Keane, Chem. Eng. J. 166 (2011)
1044.
[50] M.A. Álvarez-Montero, L.M. Gómez-Sainero, M. Martín-Martínez, F. Heras, J.J.
Rodriguez, Appl. Catal. B 96 (2010) 148.
[24] E. Díaz, L. Faba, S. Ordóñez, Appl. Catal., B 104 (2011) 415.
[25] X. Liu, J. Chen, J. Zhang, Ind. Eng. Chem. Res. 47 (2008) 5362.
[26] J.W. Bae, I.G. Kim, J.S. Lee, K.H. Lee, E.J. Jang, Appl. Catal., A 240 (2003) 129.
[27] M.A. Keane, C. Park, C. Menini, Catal. Lett. 88 (2003) 89–94.
[28] S. Ordóñez, E. Díaz, R.F. Bueres, E. Asedegbega-Nieto, H. Sastre, J. Catal. 272
(2010) 158.
[29] T. Janiak, J. Okal, Appl. Catal., B 92 (2009) 384.
[30] M.A. Aramendía, V. Boráu, I.M. García, C. Jiménez, F. Lafont, A. Marinas, J.M.
Marinas, F.J. Urbano, J. Catal. 187 (1999) 392.
[51] http://srdata.nist.gov/xps/.
[52] M. Králik, A. Biffis, J. Mol. Catal. A: Chem. 177 (2001) 113.
[53] T. Ashida, K. Miura, T. Nomoto, S. Yagi, H. Sumida, G. Kutluk, K. Soda, H.
Namatame, M. Taniguchi, Surf. Sci. 601 (2007) 3898.
[54] L. Gomez-Sainero, X. Seoane, J. Fierro, A. Arcoya, J. Catal. 209 (2002) 279.
[55] F. Chen, Z. Zong, X. Xu, J. Luo, J. Mater. Sci. Lett. 40 (2005) 1517.
[56] H. Hirai, N. Yakura, Polym. Adv. Technol. 12 (2001) 724.
[57] E. Diaz, A.F. Mohedano, J.A. Casas, L. Calvo, M.A. Gilarranz, J.J. Rodriguez, Appl.
Catal., B 106 (2011) 469.
[31] E. Diaz, A.F. Mohedano, J.A. Casas, L. Calvo, M.A. Gilarranz, J.J. Rodriguez, Appl.
Catal., B 106 (2011) 469.
[58] M. Nutt, J. Hughes, M. Wong, Environ. Sci. Technol. 39 (2005) 1346.
[59] J. Xu, D. Bhattacharyya, Ind. Eng. Chem. Res. 46 (2007) 2348.
[60] Z. Jin, C. Yu, X. Wang, Y. Wan, D. Li, G. Lu, J. Hazard. Mater. 186 (2011) 1726.
[61] J. Le Bars, U. Specht, J.S. Bradley, D.G. Blackmond, Langmuir 15 (1999) 7621.
[62] Y. Li, E. Boone, M. El-Sayed, Langmuir 18 (2002) 4921.
[63] Z. Karpinski, K. Early, J.L. d’Itri, J. Catal. 164 (1996) 378.
[64] C. Thompson, R. Rioux, N. Chen, F. Ribeiro, J. Phys. Chem. B 104 (2000) 3067.
[65] A. Orlov, D.J. Watson, F.J. Williams, M. Tikhov, R.M. Lambert, Langmuir 23
(2007) 9551.
[32] Y. Yang, L.D. Unsworth, N. Semagina, J. Catal. 281 (2011) 137.
[33] A. Binder, M. Seipenbusch, M. Muhler, G. Kasper, J. Catal. 268 (2009) 150.
[34] L. Kiwi-Minsker, N. Semagina, A. Renken, B. Corain, G. Schmid, N. Toshima
(Eds.), Metal Nanoclusters in Catalysis and Materials Science, Elsevier,
Amsterdam, 2008. p. 293.
[35] J.W. Kim, B. Lim, H. Jang, S.J. Hwang, S.J. Yoo, J.S. Ha, E.A. Cho, T. Lim, S.W. Nam,
S. Kim, Int. J. Hydrogen Energy 36 (2011) 706.
[36] R. Narayanan, M.A. El-Sayed, Top. Catal. 47 (2008) 15.
[37] T. Teranishi, M. Miyake, Chem. Mater. 10 (1998) 594.
[38] J. Durand, E. Teuma, M. Gómez, Eur. J. Inorg. Chem. 2008 (2008) 3577.
[39] S. Shen, X. Wang, Chem. Commun. 46 (2010) 6891.
[40] C. Luo, Y. Zhang, Y. Wang, J. Mol. Catal. A: Chem. 229 (2005) 7.
[66] J.L. Solomon, R.J. Madix, J. Stöhr, Surf. Sci. 255 (1991) 12.
[67] M.A. Keane, ChemCatChem 3 (2011) 800.
[68] E. Diaz, J.A. Casas, A.F. Mohedano, L. Calvo, M.A. Gilarranz, J.J. Rodriguez, Ind.
Eng. Chem. Res. 47 (2008) 3840.
[69] H.M. Roy, C.M. Wai, T. Yuan, J. Kim, W.D. Marshall, Appl. Catal., A 271 (2004)
137.
´
[41] A. Gniewek, A.M. Trzeciak, J.J. Ziółkowski, L. Ke˛pinski, J. Wrzyszcz, W. Tylus, J.
Catal. 229 (2005) 332.
[42] J. Cha, K. Kim, S. Choi, S. Yeon, H. Lee, C. Lee, J. Shim, Korean J. Chem. Eng. 24
(2007) 1089.
[43] K. Okitsu, H. Bandow, Y. Maeda, Y. Nagata, Chem. Mater. 8 (1996) 315.
[44] P. Yong, N.A. Rowson, J.P.G. Farr, I.R. Harris, L.E. Macaskie, Biotechnol. Bioeng.
80 (2002) 369.
[70] E. Diaz, J.A. Casas, A.F. Mohedano, L. Calvo, M.A. Gilarranz, J.J. Rodriguez, Ind.
Eng. Chem. Res. 48 (2009) 3351.
[71] Y. Shindler, Y. Matatov-Meytal, M. Sheintuch, Ind. Eng. Chem. Res. 40 (2001)
3301.
[72] G.M. Schwab, Adv. Catal. 27 (1978) 1.
´
[73] M.A. Aramendıa, V. Boráu, I.M. Garc´ıa, C. Jiménez, J.M. Marinas, F.J. Urbano,
[45] S. Domínguez-Domínguez, Á. Berenguer-Murcia, D. Cazorla-Amorós, Á.
Linares-Solano, J. Catal. 243 (2006) 74.
[46] R. Narayanan, M.A. El-Sayed, J. Catal. 234 (2005) 348.
Appl. Catal., B 20 (1999) 101.
[74] C.E. Gigola, H.R. Aduriz, P. Bodnariuk, Appl. Catal. 27 (1986) 133.
[75] J. Estelle, J. Ruz, Y. Cesteros, R. Fernandez, P. Salagre, F. Medina, J. Sueiras, J.
Chem. Soc., Faraday Trans. 92 (1996) 2811.
´
[47] A. Gniewek, J.J. Ziółkowski, A.M. Trzeciak, L. Ke˛pinski, J. Catal. 239 (2006)
272.
[76] R.M. Rioux, C.D. Thompson, N. Chen, F.H. Ribeiro, Catal. Today 62 (2000) 269.
[77] F. Alonso, I.P. Beletskaya, M. Yus, ChemInform 34 (2003).
[78] J.W. Bae, E.J. Jang, D.H. Jo, J.S. Lee, K.H. Lee, J. Mol. Catal. A: Chem. 206 (2003)
225.
[48] S.H. Joo, J.Y. Park, J.R. Renzas, D.R. Butcher, W. Huang, G.A. Somorjai, Nano Lett.
10 (2010) 2709.
[49] N. Krishnankutty, M.A. Vannice, J. Catal. 155 (1995) 312.