Bi modified Pd/support (SiO2, Al2O3) catalysts for hydrodechlorination of 2,4-dichlorophenol

Article abstract of DOI:10.1016/j.molcata.2010.07.011

The influence of bismuth addition on the activity and selectivity of home-made supported palladium catalysts in the reaction of hydrodechlorination of 2,4-dichlorophenol to phenol was studied. Bimetallic Pd-Bi/Al ₂O₃ and Pd-Bi/SiO

Full text of DOI:10.1016/j.molcata.2010.07.011

Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Bi modified Pd/support (SiO , Al O ) catalysts for
2
2
3
hydrodechlorination of 2,4-dichlorophenol
∗
Izabela Wito n´ ska, Aleksandra Królak, Stanisław Karski
Institute of General and Ecological Chemistry, Technical University of Lodz, ul. Z˙ wirki 36, 90-924 Lodz, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 May 2010
Received in revised form 15 July 2010
Accepted 21 July 2010
Available online 30 July 2010
The influence of bismuth addition on the activity and selectivity of home-made supported palladium
catalysts in the reaction of hydrodechlorination of 2,4-dichlorophenol to phenol was studied. Bimetallic
Pd–Bi/Al2O3 and Pd–Bi/SiO2 catalysts, containing 5 wt.% of Pd and 1–5 wt.% of Bi, are characterized by high
conversion of 2,4-dichlorophenol, which reaches about 100% at the end of the reaction run. However,
the catalysts 5%Pd–8%Bi/SiO2 and 5%Pd–8%Bi/Al2O3 show considerably lower conversion (respectively,
about 25% and 60%), but they are characterized by the highest selectivity towards phenol (respectively,
about 95% and 60%). The stability of the chosen bimetallic catalyst 5%Pd–1%Bi/Al2O3 is higher than that of
the monometallic palladium catalyst. The XRD and ToF-SIMS studies proved the presence of intermetallic
compounds BiPd and Bi2Pd, which probably modify the catalytic properties of Pd–Bi/support catalysts in
hydrodechlorination of 2,4-dichlorophenol.
Keywords:
Hydrodechlorination
2
,4-Dichlorophenol
Pd–Bi/support catalysts
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
Although the study of catalytic HDC of chlorophenols was ini-
tiated in the 1960s, the first literature reported the destruction
of chlorophenol in the liquid phase in 1992 [13]. The authors of
this work describe a catalytic process of destruction of chlorinated
phenols at 35 C using supported monometallic palladium catalyst,
where hydrogen was used as a reducing agent. The only organic
product of hydrodechlorination reaction was phenol.
A lot of research about hydrodechlorination of chlorophenols
in liquid phase with the usage of supported palladium catalysts
has been conducted in recent years [14–27]. The papers show that
activity and selectivity of studied catalysts depend on dispersion,
the amount of active phase, methods of preparation, the type of
support used, temperature of reaction, etc.
Chlorophenols belong to an important group of organic com-
pounds. They are commonly used as chemical feedstocks in the
manufacture of pesticides, herbicides, biocides, wood preser-
vatives, pharmaceuticals and dyes. The wide application of
chlorophenols in different branches of chemical industry may be
the reason for their easy penetration to the natural environment.
The presence of a great amount of chlorophenols in air, water and
soil poses a serious risk to human health and it negatively influences
all ecosystem functions. These compounds have toxic, mutagenic
and possibly carcinogenic properties and they are characterized by
high stability.
◦
From the environmental point of view, the development of
an efficient method of elimination of chlorophenols is necessary.
Several techniques were investigated for the treatment of waste
streams containing toxic chlorinated organic contaminants. The
widely used chemical method of incineration often leads to the
formation of highly toxic by-products such as dioxins [1,2]. The
methods based on electrochemistry [3,4], photochemistry [5,6]
and also on biotechnology [7–10] result in low conversion. On
the other hand, the thermal methods, which include pyrolysis and
hydrogenolysis, require high energy. Catalytic hydrodechlorination
Recently, in many carbon–halogen bond hydrogenolysis pro-
cesses the research on palladium bimetallic systems characterized
by higher activity and stability than in the case of monometallic
systems has started. However, the influence of the addition of a
second metal to Pd/support systems on catalytic properties in the
reaction of hydrodechlorination of 2,4-dichlorophenol in the liquid
phase has not been studied thoroughly yet.
The data gathered in the literature [28] concern mainly
hydrodechlorination of organohalogen compounds in the gas
phase, which is conducted at relatively high temperatures (above
◦
(
HDC) is now viewed as a promising emerging technology [11–14].
200 C). The influence of the addition of such metals as Fe [29–31],
Re [32,33], Ag [30,34], Au [35], Co [30], K [36] and La, Bi, Sb, Sn, Ba
and Zn [37,38] to palladium catalysts on their activity and selectiv-
ity in gas-phase hydrodechlorination of CFCs has been studied.
The changes in catalytic properties of bimetallic systems were
explained by different effects. Malinowski et al. [35] reported
∗ Corresponding author. Tel.: +48 42 631 30 94; fax: +48 42 631 31 28.
E-mail address: karski@p.lodz.pl (S. Karski).
1
381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.07.011

2
2
I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
the increase in selectivity towards CH2F2 in the gas-phase
hydrodechlorination of CCl2F2 over Pd–Au/MgF2 catalysts which
showed two Pd–Au solid solutions [35]. Coq et al. [30] stud-
ied the behavior of palladium when alloyed with Fe, Co or K in
hydrodechlorination of the same compound. They found that the
addition of those metals causes the maintenance of a constant value
reaction was conducted at room temperature, pH = 7 and for the
steady amount of catalyst (mcat = 0.4 g). The solution NaOH with
assumed concentration was introduced to the system manually
from the burette. The mixture was stirred at 1300 rpm and hydro-
3
gen was bubbled through at 0.300 dm /min. No pressure control
was required as the total operating pressure in the system was
equal to atmospheric pressure. A further increase in the speed
of stirring and the speed of hydrogen flow did not cause further
changes in activity. The choice of the stirring speed and hydrogen
flow rate served to minimize transport limitations.
Samples of the reaction medium were taken systematically, fil-
tered and analyzed using a liquid chromatograph coupled with a
variable wavelength UV detector Varian 9050. The analytical wave-
length was 210 nm. The reaction products were separated on a
Zorbax SB-C18 column 250 mm × 4.6 mm ID, using water solution
of MeOH as a mobile phase. For the selected catalysts, the reac-
tion products were detected by GC (Hewlett Packard 5890 with FID
detection; packed column 8% Carbowax 1540 on Chromosorb W,
1500 mm × 8 mm ID) for the full analysis of the mixture of phe-
nol/cyclohexanone/cyclohexanol.
of selectivity to CH F , so it improves the stability of those systems
2
2
in the studied process. The presence of Fe, Co or K prevents the dif-
fusion of halide species into the bulk palladium, which is the reason
for lowering the selectivity value in the process over Pd/Graphite
catalyst. Shekar et al. [37,38] studied Bi–Pd/FCCA catalysts in the
reaction of gas-phase hydrodechlorination of CCl F in the temper-
2
2
◦
ature range 200–320 C and stated that the addition of bismuth to
palladium catalysts improves not only the thermal stability but also
the selectivity towards CH F . In the case of Bi–Pd/FCCA catalysts,
2
2
the presence of intermetallic compounds BiPd type was observed.
3
These compounds can modify the catalytic properties of bimetallic
catalysts.
On the basis of the facts presented above, two main aspects need
to be pointed out to understand the effect of adding a second metal
to palladium catalysts: the degree of interaction between both met-
als and the distribution of metal components on the catalyst. These
issues will be thoroughly studied in our work for Pd–Bi/support
systems. In the case of these bimetallic catalysts the formation of
Catalytic results are expressed as conversion (X, %) and selectiv-
ities (S, %). Those parameters were defined as:
ꢀ
ꢁ
ꢂꢃ
C
2
,4DCP
X = 1 −
× 100%
× 100%
C
0
2,4DCP
intermetallic compounds BiPd and Bi Pd [39,40], which probably
2
ꢀ
ꢃ
modify the catalytic properties in studied reaction was observed in
our earlier studies.
In this work, we present the results of activity and
selectivity of monometallic palladium supported catalysts and
bimetallic systems modified with Bi in hydrodechlorination of
CP
2,4DCP−C
S =
C
0
2,4DCP
where C
is a molar concentration of 2,4-dichlorophenol at
02,4DCP
the beginning of the hydrodechlorination process, C2,4DCP is a molar
concentration of 2,4-dichlorophenol after time t, CP is a molar con-
centration of phenol after time t. The tests of catalytic activity were
conducted for each catalyst three times and the presented results
constitute arithmetical mean of these three measurements.
2
,4-dichlorophenol. The catalysts were characterized by X-ray
diffraction (XRD), time-of-flight secondary ion mass spectrome-
try (ToF-SIMS) and temperature-programmed reduction (TPR). The
main focus will be put on mutual interactions between palla-
dium and bismuth and their influence on catalytic properties of
Pd–Bi/support systems.
2
2
.3. Characterization of catalysts
.3.1. Powder X-ray diffraction (XRD)
2
. Experimental
Room temperature powder X-ray diffraction patterns were
collected using a PANalytical X’Pert Pro MPD diffractometer. A
copper long fine focus X-ray diffraction tube operating at 40 kV
and 30 mA was the X-ray source. Divergent optics were used in
a Bragg–Brentano (flat-plate sample) geometry, with fixed diver-
2
.1. Catalyst preparation
Catalysts containing 5 wt.% of palladium supported on silica
2
2
(
Aldrich, 250 m /g) and alumina (Fluka Typ 507C, 143 m /g) were
◦
◦
gence (1/2 ) and antiscatter (1 ) slits. Incident and receiving
.04 rad Soller slits were used to limit axial divergence, and a nickel
filter on the receiving side was used to eliminate CuK␤ radiation.
prepared from water solution of PdCl2 (aq., POCh Gliwice) by
aqueous impregnation. The water was evaporated at an increased
temperature (353 K) under vacuum. The catalysts were dried in air
at 110 C for 6 h, calcined at 500 C for 4 h in oxygen atmosphere
and then reduced in hydrogen atmosphere for 2 h at 300 C.
0
◦
◦
Data were collected in the range 20–90 2ꢀ with step 0.0167 and
exposition per step of 20 s. A PANalytical X’Celerator detector based
on Real Time Multiple Strip technology capable of measuring the
◦
◦
◦
Bimetallic Pd–Bi/support (SiO , Al O ) catalysts containing
wt.% Pd and 1, 3, 5, 8 wt.% Bi were obtained from monometal-
◦
2
2
3
intensities simultaneously in the 2ꢀ range of 2.122 was used. Crys-
5
talline phases were identified by referring to ICDD PDF-2 (ver. 2004)
data base.
lic palladium catalysts by repeated impregnation of these systems
with water solution of Bi(NO ) ·5H O (aq., POCh Gliwice) according
3
3
2
to the procedure described above.
2.3.2. Time of fly secondary mass spectrometry (ToF-SIMS)
The secondary ions mass spectra were recorded with a TOF-SIMS
IV mass spectrometer manufactured by Ion-Tof GmbH, Muenster,
Germany. The instrument was equipped with Bi liquid metal ion
gun and high mass resolution time-of-flight mass analyzer. During
the measurement, the analyzed area was irradiated with the pulses
2
.2. Catalytic measurements
The catalysts were activated in the flow reactor in hydrogen
◦
atmosphere at 300 C directly before the measurements of activity.
Having put the catalyst in the reactor in water, it was saturated
with hydrogen for 0.5 h before introducing 2,4-dichlorophenol to
the system. It was done to ensure complete palladium reduction.
+
of 25 keV Bi3 ions at 10 kHz repetition rate and an average ion
current 0.6 pA. The analysis time was 30 s for both positive and
negative secondary ions giving an ion dose below the static limit of
1
3
2
1
× 10 ions/cm . Secondary ion mass spectra were recorded from
The hydrodechlorination of 2,4-dichlorophenol solution
3
an approximately 100 ␮m × 100 ␮m area of the spot surface, ion
images were obtained for 500 ␮m × 500 ␮m surface area.
The catalyst samples were prepared by pressing pellets.
(
100 mg/dm ) was performed in a thermostated glass reactor
of 400 ml equipped with a stirrer, a hydrogen supply system, a
burette containing NaOH (0.01 mol/dm ) and a pH electrode. The
3

I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
23
Fig. 1. (a) Conversion of 2,4-dichlorophenol and (b) selectivity towards phenol as a function of time of hydrodechlorination in liquid phase over ꢀ—5%Pd/Al2O3 and
ꢁ
—5%Pd/SiO2 catalysts. Reaction was conducted at room temperature for the steady amount of catalyst (mcat = 0.4 g). 400 ml solution of 2,4-dichlorophenol (0.04 g) in
water was used. Reaction mixture was stirred at 1300 rpm and hydrogen was bubbled through at 0.300 dm /min.
3
2
.3.3. Temperature-programmed reduction (TPR) measurements.
The measurements were carried out in AMI-1 (Altamira) appa-
clusive. In our study of hydrodechlorination of 2,4-dichlorophenol
in the liquid phase we observed the influence of palladium dis-
persion on both activity and selectivity in the studied reaction
(Fig. 1.)
ratus. The catalyst samples (0,05 g), prior to the TPR measurements
were calcined at 500 C for 2 h in the mixture of oxygen–argon
◦
3
(
10 vol.% of O ) at a flow rate of 30 cm /min. TPR runs were
The conversion of 2,4-dichlorophenol for 5%Pd/SiO2 catalysts
2
◦
performed in the temperature range: TPR(1): 20–300 C; TPR(2):
2
is higher than for 5%Pd/Al O3 systems. For the former catalyst
2
◦
◦
0–300 C; TPR(3): 20–650 C, using a mixture of hydrogen–argon
almost complete conversion of 2,4-dichlorophenol was observed
after 20 min. The physicochemical characteristics of the palla-
dium catalysts in question are presented in Table 1. According to
expectations, the dispersion of palladium supported on silica is
smaller than for palladium supported on alumina. This fact con-
firms the influence of the dispersion of the active phase on the
catalytic activity, which was also observed by Gomez-Quero et al.
[41].
The dispersion of palladium has also an impact on selectivity.
For the systems with lower dispersion, higher selectivity towards
phenol is observed. On the other hand, for the systems with higher
dispersion (5%Pd/Al2O3), the concentration of phenol decreases
considerably and the reaction proceeds to the formation of cycko-
hexanone and cyclohexanol (Fig. 2 and Table 2).
3
(
10 vol.% of H ) at a flow rate of 30 cm /min and a linear tempera-
2
◦
−1
.
ture growth of 10 C min
2.3.4. CO sorption measurements
The measurements of CO sorption were carried out with the use
of an impulse method at room temperature applying argon as car-
rier gas. The catalyst samples prior the sorption step were oxidized
at 500 C for 4 h using O2 and reduced at 300 C for 2 h using H .
After this operation the samples of catalysts were cooled in argon
to room temperature. The volume of dosed carbon monoxide equals
◦
◦
2
3
0
.57 cm .
3
. Results and discussion
In order to prove the hypothesis the reaction of hydrodechlori-
nation of 2,4-dichlorophenol proceeds with better yield of phenol
on bigger crystallites, the samples of 5%Pd/Al2O3 catalyst were
In the process of hydrodechlorination of 2,4-dichlorophenol in
the liquid phase, supported palladium catalysts are characterized
by good activity [14–27]. In the open literature, the influence of
many factors such as the kind of support, type of salt, method
of catalysts preparation and amount of active phase on the cat-
alytic activity in this process have been considered. The role of
palladium dispersion in the hydrodechlorination process in the
liquid phase has been investigated only in few papers [41–46].
However, the presented results are not only scant but also incon-
◦
reduced in hydrogen at the temperature of 300, 500 and 700 C.
After that, the samples underwent catalytic tests (Fig. 3) and XRD
studies (Fig. 4).
◦
The catalysts which were reduced at 700 C were character-
ized by bigger crystallites (Fig. 4) and showed higher activity and
selectivity towards phenol than the systems reduced at a lower
◦
temperature (300 and 500 C).
Table 1
Specific surface area and diameter of the Pd crystallites estimated for monometallic 5%Pd/support and bimetallic 5%Pd–X%Bi/support catalysts.
Catalysts
Activation procedure
Specific surface area
SPd (m /gcat)
Diameter of the Pd
crystallites (nm)b
2
a
◦
◦
5
5
5
5
5
5
%Pd/Al2O3
%Pd–3%Bi/Al2O3
%Pd/SiO2
%Pd–1%Bi/SiO2
%Pd–3%Bi/SiO2
%Pd–8%Bi/SiO2
O2, 500 C, 2h/H2, 300 C, 2h
4.61
6.95
0.28
0.45
0.54
0.52
7
–
47
21
8
8
◦
◦
O2, 500 C, 2h/H2, 300 C, 2h
◦
◦
◦
◦
◦
◦
◦
◦
O2, 500 C, 2h/H2, 300 C, 2h
O2, 500 C, 2h/H2, 300 C, 2h
O2, 500 C, 2h/H2, 300 C, 2h
O2, 500 C, 2h/H2, 300 C, 2h
a
Specific surface area determined on the basis of CO chemisorption, with the assumption that 0.84 × 10 of CO particles takes up the surface of 1 cm2 Pd [47].
15
b
Diameter of the Pd crystallites estimated from XRD analysis, using the Scherrer equation.

2
4
I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
Fig. 2. Scheme of the reaction of hydrodechlorination of 2,4-dichlorophenol.
Table 2
Initial specific rate of reaction and selectivities towards phenol, 2-chlorophenol and 4-chlorophenol for the chosen catalytic systems.
Ra × 10−6 (mol min−1
m
−2
)
X2,4-dichlorophenol (%)
t20 t45
Sphenol (%)
t20
S2-chlorophenol (%)
t20 t45
S4-chlorophenol (%)
t20 t45
Catalyst
t45
5
5
5
5
5
5
5
5
5
%Pd/Al2O3
1.24
–
0.88
–
20.54
13.67
10.46
–
82.08
93.90
100
100
48.16
56.03
64.69
78.61
66.51
79.06
94.44
85.30
99.98
28.71
37.36
40.71
56.16
53.99
68.99
85.09
77.72
95.09
0.84
3,91
–
–
8.21
18,67
–
–
–
–
–
–
–
–
–
%Pd–1%Bi/Al2O3
%Pd–3%Bi/Al2O3
%Pd–8%Bi/Al2O3
%Pd/SiO2
%Pd–1%Bi/SiO2
%Pd–3%Bi/SiO2
%Pd–5%Bi/SiO2
%Pd–8%Bi/SiO2
98.59
94.97
33.12
–
–
–
–
–
–
–
–
–
5.62
63.24
–
100
100
100
100
98.49
23.29
0.10
1.57
–
–
–
92.00
92.26
89.19
4.29
12.25
–
–
–
0.48
a
R—initial specific rate calculated after the first 20 min of reaction.
An addition of bismuth to supported palladium catalysts leads
Al O ) catalysts in the amount of up to 5% wt. practically causes
2 3
to the change of both activity and selectivity in the studied reac-
tion of hydrodechlorination of 2,4-dichlorophenol in the liquid
phase (Table 2). The introduction of bismuth to 5%Pd/support (SiO2,
only a slight change in the activity, however, a significant decrease
in the activity after the introduction of high amounts of bismuth (8-
wt.% Bi) is observed. Additionally, bimetallic Pd–Bi/SiO2 catalysts
Fig. 3. (a) Conversion of 2,4-dichlorophenol and (b) selectivity towards phenol as a function of time of hydrodechlorination in liquid phase over 5%Pd/Al2O3 catalysts reduced
◦
◦
◦
at ꢀ—300 C, ꢂ—500 C and ꢀ—700 C. Reaction was conducted at room temperature for the steady amount of catalyst (mcat = 0.4 g). 400 ml solution of 2,4-dichlorophenol
3
(
0.04 g) in water was used. Reaction mixture was stirred at 1300 rpm and hydrogen was bubbled through at 0.300 dm /min.

I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
25
ence of BiPd and Pi Pd intermetallic compounds is observed. BiPd
2
intermetallic compounds are observed for all the studied systems
and Bi Pd compounds are detected only for catalysts with higher
2
amounts of bismuth (5%Pd–8%Bi/SiO ). However, in the case of
2
Pd–Bi/Al O3 catalysts, intermetallic BiPd and Bi Pd compounds
2
2
were not detected by XRD measurements.
On the basis of the XRD studies, it was stated that bimetal-
lic systems supported on alumina and silica significantly differed
from each other as regards the degree of crystallinity. A low degree
of crystallinity of these systems supported on Al O impeded the
2
3
identification of metallic phases in the case of these catalysts.
To get a better understanding of the nature of interactions
between palladium and bismuth on the surface of alumina and
silica, the samples of bimetallic catalysts were characterized by
ToF-SIMS. This technique made it possible to observe the changes
of catalyst surface, which are invisible for XRD, for example, the
application of ToF-SIMS allowed us to observe the presence of inter-
metallic BixPdy compounds for bimetallic catalysts supported on
alumina.
◦
◦
Fig. 4. XRD patterns for 5%Pd/Al2O3 catalysts reduced at 1–300 C, 2–500 C and
◦
3
–700 C. Diffractogram 4 was detected for pure support Al2O3. Patterns obtained
◦
with CuK␣ radiation, step scanned 20 s per step of 0.0167 of 2ꢀ.
Fig. 5 shows an ionic view of microarea of palladium catalyst
supported on alumina and modified with bismuth after oxidation
are characterized by higher selectivity towards phenol in compari-
son with monometallic systems. Studies of CO sorption performed
at room temperature and XRD studies show an increase in the dis-
persion of palladium after the introduction of bismuth (Table 1). For
the monometallic and bimetallic systems the increase in the disper-
sion resulted in the decrease in the selectivity to phenol. However,
comparing the selectivity to phenol for monometallic palladium
catalyst supported on silica with the selectivity of bimetallic cata-
lysts on the same support, one can see that despite the increase in
the dispersion of palladium after introducing bismuth to the sys-
tems, the selectivity of these systems is higher (Table 2). Strong
metal–promotor interactions (SMPI) leading to the formation of
intermetallic compounds could probably be responsible for such
behavior of these bimetallic systems.
◦
in oxygen atmosphere for 4 h at 500 C and reduction in hydrogen
◦
atmosphere for 2 h at 300 C. The brightness of particular places in
the images corresponds to the intensity of secondary ions emission.
On the basis of images of bimetallic palladium–bismuth cata-
lysts, it is visible that Pd and Bi atoms are distributed homogenously
on the surface of alumina. The obtained images of those cat-
alysts show that the secondary ions BiPd+ and BiPdO+ are
emitted from the surface, which demonstrated strong interactions
between those metals. Slightly different images were obtained for
palladium–bismuth systems supported on silica. In the case of this
carrier, both metals were not located homogenously on the sur-
+
face, forming explicit clusters with increased emission of both Pd
and Bi+ ions and BiPd+ and Bi2Pd ions [39,40]. What is more, the
intensity of emission of BiPd and Bi2Pd+ ions is about 10 times
higher for the systems supported on alumina than it was observed
for the catalysts supported on silica (Fig. 6). Emission of those ions
from the surface of catalysts Pd–Bi/Al2O3 and Pd–Bi/SiO2 proves
the occurrence of BiPd and Bi2Pd phases in these systems, which
was shown using XRD only for silica supported catalysts.
In the light of performed studies, it can be assumed that
the presence of intermetallic compounds of BixPdy on the sur-
face of Pd–Bi/support catalysts may have an influence on their
activity and selectivity in the reaction in question. For the cat-
alysts containing small amounts of bismuth 5%Pd–1%Bi/support
and 5%Pd–3%Bi/support, the formation of mainly BiPd-type inter-
metallic compounds is observed (Table 3). These catalysts are
characterized by high activity and selectivity towards phenol
(Table 2). However, in the catalysts containing higher amounts of
bismuth (>5% weight), we observed the presence of both Bi2Pd-type
and BiPd-type intermetallic compounds. Despite high selectivity
towards phenol, these catalysts showed surprisingly low activity
in the studied reaction.
+
+
The studies of Pd–Bi/C catalysts in oxidation of glucose per-
formed by Wenkin et al. [48] show the presence of two binary alloys
with the compositions of BiPd and BiPd . Small amounts of Pd metal
3
and Bi O oxide were simultaneously detected, and another phase
2
3
which most probably corresponds to Bi Pd5 was present. The same
2
authors [49,50] performed catalytic oxidation of glucose over pure
intermetallics Bi Pd, BiPd and BiPd . They noticed that catalytic
2
3
behavior of those phases was very different.
Also, our studies performed for Pd–Bi/SiO catalysts in oxidation
2
of glucose and lactose show the presence of intermetallic com-
pounds BiPd and Bi Pd, which determine the catalytic properties
2
of these systems [39,40].
Therefore, we assume that the selectivity of palladium–bismuth
catalysts in the reaction of hydrodechlorination of 2,4-
dichlorophenol in the liquid phase is probably connected with the
presence of intermetallic compounds on the surface.
The surface structure of Pd–Bi/SiO2 and Pd–Bi/Al O3 was ana-
2
lyzed using ToF-SIMS and XRD techniques.
The results of the XRD study of Pd–Bi/SiO2 and Pd–Bi/Al O3
2
reduced catalysts are presented in Table 3. The conducted XRD
studies show that in the case of Pd–Bi/SiO2 catalysts the pres-
Similarly, temperature-programmed reduction studies prove an
existence of strong interactions between Pd and Bi in Pd–Bi/Al2O3
systems.
Fig.
7 presents TPR1 spectra obtained for 5%Pd/Al O ,
Table 3
2 3
The results of XRD analysis of reduced Pd–Bi/support catalysts. Crystalline phases
were identified by references to ICDD PDF-2 (ver. 2004) data base.
3%Bi/Al2O3, 5%Pd–3%Bi/Al2O3, 5%Pd–8%Bi/Al2O3 systems after cal-
cination in O2 at 500 C for 2 h.The calcined samples of the system
◦
containing 5%Pd/Al O3 show two low temperature peaks located
in the temperature range of 25–150 C. This seems to be connected
with the formation of two different forms of palladium oxide in
the first stage of contact between the samples and oxygen. For
Catalysts
Pd
Bi
BiPd (C)
BiPd(Sob)
Bi2Pd
2
◦
5%Pd–1%Bi/SiO2
5%Pd–3%Bi/SiO2
5%Pd–8%Bi/SiO2
5%Pd–1%Bi/Al2O3
5%Pd–5%Bi/Al2O3
5%Pd–8%Bi/Al2O3
+
+
+
+
+
+
−
−
−
−
−
−
+
+
+
−
−
−
+
+
+
−
−
−
−
−
+
−
−
−
5%Pd–3%Bi/Al O3 catalyst two maxima of the reduction rate are
2
observed, which are probably connected with the reduction of dif-
ferent oxide forms. An introduction of bismuth in the amount of

2
6
I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
Fig. 5. Positive secondary ions ToF-SIMS image of Pd–Bi/Al2O3 catalysts surfaces after single oxidation–reduction cycle. The analyzed area of the surface 500 ␮m × 500 ␮m.
+
During measurement the analyzed area was irradiated with pulses of 25 keV Bi3 ions at 10 kHz repetition rate and an average ion current 0.6 pA. Secondary ions emitted
from the bombarded surface were mass separated and counted in time-of-flight (ToF) analyzer.
◦
8
wt.% to 5%Pd/Al O catalyst causes the disappearance of the first
surements were taken up to the temperature of 650 C and their
results are presented in Fig. 8. The TPR2 and TPR3 images are prac-
tically the same, which proves that the surface of the systems is
2
3
peak connected with hydrogen adsorption.
More interesting results have been obtained from the successive
measurement cycles. The catalyst samples which underwent pre-
stabilized during the initial reduction process (TPR ).
1
◦
liminary calcination in O at 500 C and were treated with hydrogen
In the case of catalysts 5%Pd/Al O3 and 5%Pd–1%Bi/Al O ,
2
2 2 3
◦
in TPR1 process were reoxidized at 500 C for 2 h and reduced in
TPR2 up to 300 C. Finally, after the reoxidation step, TPR3 mea-
instead of hydrogen adsorption peaks only desorption peaks were
observed. It leads to the conclusion that the total reduction of all
◦
◦
◦
Fig. 6. The ToF-SIMS (+) spectra of (a) 5%Pd–8%Bi/Al2O3 and (b) 5%Pd–8%Bi/SiO2 catalysts after oxidation at 500 C in O2 and reduction at 300 C in H2. The analysis time was
1
3
2
3
0 s for positive secondary ions giving an ion dose below the static limit of 1 × 10 ions/cm .

I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
27
Table 4
Stability of 5%Pd/Al2O3 and 5%Pd–1%Bi/Al2O3 in the reaction of 2,4-dichlorophenol hydrodechlorination.
Catalyst
X2,4-dichlorophenol (%)
1st cycle
3rd cycle
5th cycle
7th cycle
t5
t20
t45
t5
t20
t45
t5
t20
t45
t5
t20
t45
5
5
%Pd/Al2O3
%Pd–1%Bi/Al2O3
33.02
70.33
82.08
98.59
93.90
100
14.47
50.01
63.16
94.11
90.80
100
7.06
45.98
47.54
80.47
77.05
100
4.50
18.72
32.41
73.36
57.39
97.08
oxide forms took place at room temperature. Moreover, Pd has the
ability to form hydride phases under normal conditions. The des-
orption peaks observed for 5%Pd/Al O3 and for bimetallic systems
2
◦
in the TPR3 process at a temperature of around 80 C probably cor-
respond to ␤-PdH decomposition. Similar patterns are obtained
for supported palladium catalysts promoted with Cu, Sn, Pb and
Ag [51–53]. However, upon the incorporation of a second metal
the suppression of ␤-PdH phase was postulated for the supported
bimetallic catalyst. The introduction of bismuth into supported pal-
ladium catalysts did not lead to the disappearance of the desorption
peak suggesting that the ␤-PdH formation was not inhibited.
Moreover, the appearance of two reduction peaks at the tem-
◦
perature of about 180 and 300 C for the 5%Pd–3%Bi/Al O and
2
3
5
%Pd–8%Bi/Al O catalysts might indicate the formation of a mixed
2 3
oxide in those bimetallic systems. The existence of this form could
lead to the formation of intermetallic compounds in the reduction
atmosphere. For those systems the formation of BiPd and Bi Pd
2
phases was indicated by ToF-SIMS study.
In Table 4 the stability of mono- and bimetallic systems in
the reaction of hydrodechlorination of 2,4-dichlorophenol in an
aqueous phase was shown. The stability of catalysts was esti-
mated on the basis of activity changes during seven 45-min-long
measurement cycles. The process of reduction was conducted for
the steady amount of catalyst (mcat = 0.4 g) without its removal
Fig. 8. Temperature-programmed reduction (TPR3) of (1) 3%Bi/Al2O3; (2)
5
%Pd/Al2O3; (3) 5%Pd–1%Bi/Al2O3; (4) 5%Pd–3%Bi/Al2O3 and (5) 5%Pd–8%Bi/Al2O3
◦
catalysts. Samples which underwent preliminary calcination in O2 at 500 C and
were treated with hydrogen in TPR1 and TPR2 processes, and reoxidized each time
at 500 C for 2 h.
◦
from the reaction mixture in the subsequent cycles. After each 45-
min-long measurement cycle, a new portion of 2,4-dichlorophenol
(m2,4-dichlopophenol = 0.04 g) was introduced to the reaction mixture.
In the case of monometallic catalyst 5%Pd/Al O , a systematical
2
3
decrease in its activity was observed in each measurement cycle. In
the 7th measurement cycle, the conversion of 2,4-dichlorophenol
falls to the value of about 50%. This phenomenon is not observed
for the bimetallic system 5%Pd–1%Bi/Al O . In this case, even in
2
3
the 7th measurement cycle, the conversion of 2,4-dichlorophenol
is practically complete. Therefore, the addition of bismuth to pal-
ladium catalysts, which lead to the formation of intermetallic BiPd
compounds, causes a significant improvement of their stability in
the process.
4. Conclusions
Bimetallic Pd–Bi/Al O3 and Pd–Bi/SiO2 catalysts, containing
2
5
wt.% of Pd and 1–5 wt.% of Bi, are characterized by high activity
in the hydrodechlorination of 2,4-dichlorophenol. However, sys-
tems containing a larger amount of bismuth (5%Pd–8%Bi/Al O ,
2
3
5
%Pd–8%Bi/SiO ) show poorer activity in the studied reaction. The
2
addition of bismuth into supported palladium leads to the improve-
ment of selectivity towards phenol and stability in the studied
process.
The ToF-SIMS and XRD measurements of bimetallic
Pd–Bi/support (SiO2, Al2O3) catalysts showed the presence of
Fig. 7. Temperature-programmed reduction (TPR1) of (1) 3%Bi/Al2O3; (2)
5
%Pd/Al2O3; (3) 5%Pd–3%Bi/Al2O3 and (4) 5%Pd–8%Bi/Al2O3 catalysts. Samples
◦
which underwent preliminary calcination in O2 at 500 C.

2
8
I. Wito n´ ska et al. / Journal of Molecular Catalysis A: Chemical 331 (2010) 21–28
intermetallic compounds on their surfaces. The presence of BiPd-
type compounds was observed independently of the carrier used.
[21] J. Morales, R. Hutcheson, I.F. Cheng, J. Hazard. Mater. B 90 (2002) 97–108.
[
[
22] Y. Ukisu, T. Miyadera, Appl. Catal. A 271 (2004) 165–170.
23] D. Fritsch, K. Kuhr, K. Mackenzie, F.-D. Kopinke, Catal. Today 82 (2003) 105–118.
However, additionally the creation of Bi Pd compound on the
2
[24] S. Gómez-Quero, F. Cárdenas-Lizana, M.A. Keane, Ind. Eng. Chem. Res. 47 (2008)
6841–6853.
surface was observed for the systems containing high amounts
of bismuth. At the same time, these catalysts were characterized
by low conversion of 2,4-dichlorophenol in the studied process.
[
[
25] G. Yuan, M.A. Keane, Ind. Eng. Chem. Res. 46 (2007) 705–715.
26] L. Calvo, M.A. Gilarranz, J.A. Casas, A.F. Mohedano, J.J. Rodríguez, J. Hazard.
Mater. 161 (2009) 842–847.
Therefore, the presence of intermetallic Bi Pd compound on the
[27] E. Diaz, J.A. Casas, A.F. Mohedano, L. Calvo, M.A. Gilarranz, J.J. Rodríguez, Ind.
Eng. Chem. Res. 47 (2007) 3840–3846.
2
surface of Pd–Bi/support systems may be responsible for low
activity of these catalysts in the reaction of hydrodechlorination of
[
[
28] F.J. Urbano, J.M. Marinas, J. Mol. Catal. A. Chem. 173 (2001) 329–345.
29] N. Lingaiah, M.A. Uddin, A. Muto, Y. Sakata, Chem. Commun. 17 (1999)
1657–1658.
2
,4-dichlorophenol.
The high activity, selectivity and stability of Pd–Bi/support sys-
tems containing up to 5 wt.% of bismuth constitute a premiss to
their possible application in the chemical industry.
[30] B. Coq, S. Hub, F. Figueras, D. Tournigant, Appl. Catal. A 101 (1993) 41–50.
[
[
31] W.X. Zhang, C.B. Wang, H.L. Lien, Catal. Today 40 (1998) 387–395.
32] A. Malinowski, W. Juszczyk, M. Baronowska, J. Pieluszek, Z. Karpinski, J. Catal.
177 (1998) 153–163.
[
[
33] M. Baronowska, A. Malinowski, Z. Karpinski, Appl. Catal. A 188 (1999) 145–154.
34] B. Heinrichs, P. Delhez, J.P. Schoebrechts, J.P. Pirard, J. Catal. 172 (1997)
322–335.
35] A. Malinowski, W. Juszczyk, J. Pieluszek, M. Baronowska, M. Wojciechowska, Z.
Karpinski, Chem. Commun. 8 (1999) 685–686.
References
[
[
[
[
1] N. Graham, W. Chu, C. Lau, Chemosphere 51 (2003) 237–243.
2] W. Chu, C.K. Law, Water Res. 37 (2003) 2339–2346.
3] M.S. Ureta-Zanartu, P. Bustos, C. Berrios, M.C. Diez, M.L. Mora, C. Gutierrez,
Electrochim. Acta 47 (2002) 2399–2406.
[36] B. Coq, J.M. Cognion, F. Figueras, D. Tournigant, J. Catal. 141 (1993) 21–33.
[37] S.Ch. Shekar, J.K. Murthy, P.K. Rao, K.S.R. Rao, E. Kemnitz, Appl. Catal. A 244
(2003) 39–48.
[
[
4] H. Cheng, K. Scott, P.A. Christensen, J. Electrochem. Soc. 150 (2003) D17–D24.
5] S.G. Poulopoulos, M. Nikolaki, D. Karampetsos, C.J. Philippopoulos, J. Hazard.
Mater. 153 (2008) 582–587.
[38] J.K. Murthy, S.Ch. Shekar, K.S.R. Rao, G. Kishan, J.W. Niemantsverdriet, Appl.
Catal. A 259 (2004) 169–178.
[39] S. Karski, I. Wito n´ ska, J. Mol. Catal. A: Chem. 191 (2003) 87–92.
[40] S. Karski, J. Mol. Catal. A: Chem. 253 (2006) 147–154.
[41] S. Gomez-Quero, F. Cardenas-Lizana, M.A. Keane, Ind. Eng. Chem. Res. 47 (2008)
6841–6853.
[42] M.A. Armendia, V. Borau, I.M. Garcia, C. Jimenez, F. Lafont, A. Marinas, J.M.
Marinas, F.J. Urbano, J. Catal. 187 (1999) 392–399.
[43] L.M. Gomez-Sainero, X.L. Seoane, J.L.G. Fierro, A. Arcoya, J. Catal. 209 (2002)
279–288.
[44] S. Lambert, C. Cellier, P. Grange, J.P. Pirrard, B. Heinrichs, J. Catal. 221 (2004)
335–346.
[45] S. Lambert, J.F. Polard, J.P. Pirrard, B. Heinrichs, Appl. Catal. B 50 (2004) 127–140.
[46] Y.H. Choi, W.Y. Lee, Catal. Lett. 67 (2000) 155–161.
[47] R.L. Moss, D. Pope, B.J. Devis, D.H. Edwards, J. Catal. 58 (1979) 206–211.
[48] M. Wenkin, P. Ruiz, B. Delmon, M. Devillers, J. Mol. Catal. A 180 (2002) 141–159.
[49] M. Wenkin, C. Renard, P. Ruiz, B. Delmon, M. Devillers, Stud. Surf. Sci. Catal. 108
(1997) 391–398.
[6] G. Waldner, M. Pourmodjib, R. Bauer, M. Neumann-Spallart, Chemosphere 50
(
2003) 989–998.
[
[
7] S.K. Banerji, R.K. Bajpai, J. Hazard. Mater. 39 (1994) 19–31.
8] C.M. Kao, C.T. Chai, J.K. Liu, T.Y. Yeh, K.F. Chen, S.C. Chen, Water Res. 38 (2004)
6
63–672.
[9] B.-V. Chang, C.-W. Chiang, S.-Y. Yuan, Chemosphere 36 (1998) 537–545.
[10] D.-S. Shen, X.-W. Liu, H.-J. Feng, J. Hazard. Mater. B119 (2005) 239–243.
[11] E.J. Shin, M.A. Keane, Chem. Eng. Sci. 54 (1999) 1109–1120.
[12] P.N. Rylander, Catalytic Hydrogenation over Platinum Metals, Academic Press,
New York, 1967.
[
[
13] J.B. Hoke, G.A. Gramiccioni, E.N. Balko, Appl. Catal. B 1 (1992) 285–296.
14] V. Felis, P. Fouilloux, C.D. Bellefon, D. Schweich, Rec. Prog. Gen. Proc. 13 (1999)
3
03–309.
[15] M.A. Aramendia, V. Boráu, I.M. García, C. Jiménez, F. Lafont, A. Marinas, J.M.
Marinas, F.J. Urbano, J. Catal. 187 (1999) 392–399.
[
[
[
16] J.L. Benitez, G. del Angel, React. Kinet. Catal. Lett. 70 (2000) 67–72.
17] G. Yuan, M.A. Keane, Chem. Eng. Sci. 58 (2003) 257–267.
18] L. Calvo, M.A. Gilarranz, J.A. Casas, A.F. Mohedano, J.J. Rodríguez, Appl. Catal. B
[50] M. Wenkin, R. Touillaux, P. Ruiz, B. Delmon, M. Devillers, Appl. Catal. A 148
(1996) 181–199.
[51] S. Karski, I. Wito n´ ska, J. Rogowski, J. Gołuchowska, J. Mol. Catal. A: Chem. 240
(2005) 155–163.
[52] H.R. Aduriz, P. Bondariuk, B. Coq, F. Figueras, J. Catal. 129 (1991) 47.
[53] Q. Zhang, J. Li, X. Liu, Q. Zhu, Appl. Catal. 197 (2000) 221.
6
7 (2006) 68–76.
[19] H. Cheng, K. Scott, P.A. Christensen, Appl. Catal. A 261 (2004) 1–6.
[20] S.S. Zinovyev, N.A. Shinkova, A. Perosa, P. Tundo, Appl. Catal. B 55 (2005) 49–56.