Investigation of active metal species formation in Pd-promoted sulfated zirconia isomerization catalyst

Article abstract of DOI:10.1016/j.apcata.2010.07.052

The state of palladium in Pd/SO₄^2--ZrO₂ (Pd/SZ) isomerization catalyst was investigated by the temperature-programmed reduction (TPR), chemisorption technique, infrared spectroscopy of adsorbed carbon monoxide (FTIRS), X-ray photoelectron spectroscopy (XPS), diffuse-reflectance UV-vis spectroscopy (UV-vis DRS), and benzene hydrogenation as a test reaction. It has been stated that reduction temperature has a great impact on the metal function of Pd-promoted sulfated zirconia catalyst. Metal centers are formed at about 30-70 °C and characterized by high palladium dispersion and activity in benzene hydrogenation. At temperatures above 200 °C, intensive sulfate decomposition occurs and products of sulfate reduction poison the metal function of the catalyst. According to XPS and FTIRS study, palladium particles in the poisoned samples are only partly oxidized, but the main part is presented by metallic phase without large amount of PdS. Reduction of Pd-containing catalyst at 150 °C (instead of 250 °C) leads to higher conversion and 2,2-dimethylbutane yield in acid-catalyzed reaction of n-hexane isomerization. Higher isomerization activity in this case is provided by prevention of active sulfate species decomposition due to the capability of palladium metallic particles formation at low reduction temperatures.

Full text of DOI:10.1016/j.apcata.2010.07.052

Applied Catalysis A: General 387 (2010) 5–12
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Investigation of active metal species formation in Pd-promoted sulfated
zirconia isomerization catalyst
a,b,∗
c,d
c
a
a
c
Olga B. Belskaya
, Irina G. Danilova , Maxim O. Kazakov , Tatyana I. Gulyaeva , Lidiya S. Kibis ,
a
a,b
Andrey I. Boronin , Alexander V. Lavrenov , Vladimir A. Likholobov
a
Institute of Hydrocarbons Processing SB RAS, Neftezavodskaya 54, Omsk 644040, Russia
Omsk State Technical University, Pr. Mira 11, Omsk 644050, Russia
Boreskov Institute of Catalysis SB RAS, Pr. Akad. Lavrentieva 5, Novosibirsk 630090, Russia
Novosibirsk State University, Pirogova 2, Novosibirsk 630090, Russia
b
c
d
a r t i c l e i n f o
a b s t r a c t
The state of palladium in Pd/SO42 –ZrO2 (Pd/SZ) isomerization catalyst was investigated by the
temperature-programmed reduction (TPR), chemisorption technique, infrared spectroscopy of adsorbed
carbon monoxide (FTIRS), X-ray photoelectron spectroscopy (XPS), diffuse-reflectance UV–vis spec-
troscopy (UV–vis DRS), and benzene hydrogenation as a test reaction. It has been stated that reduction
temperature has a great impact on the metal function of Pd-promoted sulfated zirconia catalyst. Metal
−
Article history:
Received 5 April 2010
Received in revised form 9 July 2010
Accepted 27 July 2010
Available online 6 August 2010
◦
centers are formed at about 30–70 C and characterized by high palladium dispersion and activity in ben-
zene hydrogenation. At temperatures above 200 C, intensive sulfate decomposition occurs and products
Keywords:
Palladium
Metal function
Sulfated Zirconia
Isomerization catalyst
◦
of sulfate reduction poison the metal function of the catalyst. According to XPS and FTIRS study, palla-
dium particles in the poisoned samples are only partly oxidized, but the main part is presented by metallic
phase without large amount of PdS. Reduction of Pd-containing catalyst at 150 C (instead of 250 C) leads
to higher conversion and 2,2-dimethylbutane yield in acid-catalyzed reaction of n-hexane isomerization.
Higher isomerization activity in this case is provided by prevention of active sulfate species decompo-
sition due to the capability of palladium metallic particles formation at low reduction temperatures.
◦
◦
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
However, Paal et al. [12] suggested that platinum is present as a
bulk metal phase covered with a layer of platinum sulfide. Sayari
et al. [13,14] found from TPR, XRD, and XPS experiments that plat-
inum supported on sulfated zirconia is present in a metallic state
Sulfated zirconia (SZ) is an attractive isomerization catalyst due
to its high activity at low temperatures and high selectivity to iso-
mers. Promotion of SZ with noble metals improves the activity
and diminishes coke formation. The most important works con-
cerning hydroconversion of n-alkanes use platinum supported on
SZ [1,2]. In spite of numerous studies, metal state and its role
in alkanes isomerization mechanism are still a matter of con-
troversy. In particular, platinum promotion of sulfated zirconia
is suggested to occur via increasing the catalyst acidity [3–6],
enhancing the hydride transfer to reaction intermediates [7], or
decreasing the catalyst deactivation by prevention of coke forma-
tion [8,9].
◦
already after calcination in air at 600 C. These results were sup-
ported by X-ray absorption spectroscopy [15]. It should be noted
that in the works [10–15] Pt/SZ catalysts were prepared by impreg-
nation of sulfated hydrous zirconium oxide (SO42 –ZrO ·xH O)
−
2
2
with Pt precursor solution, followed by drying and calcination. This
preparation method is considered as a conventional for Pt/SZ cata-
lysts [1,16,17]. In the subsequent works Paal et al. [18,19] showed
that in the case of Pt/SO42 –ZrO2 catalysts prepared in such way
the Pt particles are buried under zirconia layers. Above-mentioned
phenomenon along with the poisoning by sulfur species during
−
oxidation of Pt/SO42 –ZrO ·xH O or reduction of Pt/SO
−
2−
–ZrO2
4
Chemical state of the metal is also a source of debate. From
XPS studies, Ebitani et al. [10,11] concluded that Pt on sulfated
zirconia exists in the cationic state, possibly as platinum oxide.
2
2
[3,7,14] are the reasons for anomalous state of Pt in these catalysts.
In order to avoid metal particles encapsulation Pt/SZ isomerization
catalysts may be obtained by impregnation of the precalcined sul-
fated zirconia [16,17,20]. However, this preparation method does
not prevent Pt poisoning by H S, which is formed on the reduc-
tion stage. As a result, Pt supported on sulfated zirconia has a
poor hydrogenation/dehydrogenation activity [3,16] and a negli-
∗ Corresponding author at: Institute of Hydrocarbons Processing SB RAS, Nefteza-
2
vodskaya 54, Omsk 644040, Russia. Tel.: +7 3812 670474; fax: +7 3812 646156.
E-mail address: obelska@ihcp.oscsbras.ru (O.B. Belskaya).
0
926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.07.052

6
O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
gible chemisorption capacity for hydrogen [7,20] and CO [10] at
room temperature.
tion on the influence of activation temperature on the metal
function of Pd/SZ catalyst and sulfate species decomposition.
Finally, we tested the acid function of Pd/SZ systems reduced at
different temperatures by reaction of n-hexane isomerization.
Obtaining a promoted SZ isomerization catalyst with improved
metal function could be important for several reasons [16,21]:
isomerization activity could be increased by improving the dehy-
drogenation capacity of M/SZ; removal of coke precursors could
be enhanced by an increase in the production of atomic hydro-
gen; the isomerization process could be performed at a lower
2. Experimental
2.1. Catalysts preparation
H /hydrocarbon ratio, thus leading to lower operating costs. Unfor-
2
tunately, there is a mutual antagonism between the conditions
for optimal activity of the acid and metal functions of Pt/SZ
catalysts [21]. Synthesis and activation conditions that lead to
better metal properties decrease the level of isomerization activ-
ity.
In this respect palladium may be a promising metal promoter
for sulfated zirconia isomerization catalyst with improved metal
function. The main reason is that supported Pd can be reduced
even at temperatures close to ambient [22]. In order to prevent
possible encapsulation and poisoning of metal function during
calcination Pd should be added to the precalcined sulfated zirco-
nia. Capability of reduction at low temperatures can be beneficial
Hydrous zirconium oxide was obtained by slow precipitation
from an aqueous solution of 1 M ZrO(NO3)2 with a molar excess
of NH OH to produce final pH of approximately 10. After washing
4
◦
in distilled water, the precipitate was then dried at 120 C for 12 h
and treated with 2 M sulfuric acid solution by incipient wetness
◦
impregnation. Thus prepared sulfated material was dried at 120 C
◦
overnight and calcined at 650 C for 2 h in an air stream.
Palladium (0.3 and 1.0 wt%) and platinum (0.3 wt%) were added
by incipient wetness technique to the calcined SO42 –ZrO support
−
2
using the appropriate amount of an aqueous solutions of H PdCl4
2
and H PtCl6 respectively. The obtained solids were then dried for
2
◦
12 h, calcined in an air stream at 400 C for 3 h and then reduced
for preventing Pd poisoning by H S, which is formed due to
for 2 h in a hydrogen stream. Reduction temperatures were varied
from 75 to 300 C. It should be noted that for TPR, CO chemisorption,
FTIRS, XPS, UV–vis DRS and benzene hydrogenation we used sam-
ples with 1.0 wt% of Pd content. Samples with 0.3 wt% metal loading
were used only for the test reaction of n-hexane isomerization.
2
◦
sulfate species reduction. Besides, Pd in comparison with Pt is
more resistant to sulfur compounds [23] that can form due to
reduction of SO4 groups of SZ in a hydrogen-hydrocarbon media.
According to Watanabe et al. [24,25], Pd-loaded sulfated zirco-
nia catalysts exhibit good performance in both the isomerization
and hydrodesulfurization reactions, thus allowing the operation
with a sulfur-containing feed. From a practical standpoint, it is
important to understand whether transition metals less expen-
sive than platinum can also provide effective performance of
M/SZ.
2.2. Characterization
2.2.1. TPR
Temperature-programmed reduction (TPR) was performed on
an AutoChem II 2920 Micromeritics TPD/TPR instrument equipped
with a thermal conductivity detector (TCD), using an Ar stream con-
taining 10% H2 at a flow-rate of 50 ml/min and at a heating rate
Only few studies on Pd-promoted SZ isomerization cata-
lysts are available in the literature [5,24–32]. Pd-containing
sulfated zirconia catalysts were used for conversion of n-butane
◦
of 10 C/min. Before TPR measurements, samples were calcined ex
◦
[
5,26], n-heptane [29–31], n-octane [29,32], n-nonane [29], sulfur-
containing n-pentane [24] and sulfur-containing light naphtha
25]. According to Larsen et al. [26], reduction of sulfated zirco-
situ at 400 C in air. Palladium dispersion was measured by carbon
monoxide chemisorption (the pulse method, 10% CO in He).
[
nia containing 2 wt% of Pd (Pd was added to hydrous zirconia) at
2.2.2. FTIR
◦
2
40 C resulted in the formation of the metal phase with particle
The IR spectra were recorded on a Shimadzu FTIR-8300 spec-
trometer in the range of 700–6000 cm with 4 cm resolution
and 100 scans for signal accumulation. For IR measurements,
−
1
−1
size about 2.5–4.5 nm, but the surface of this phase had negligi-
ble hydrogen chemisorption capacity due to sulfur poisoning. It
should be mentioned that Pd/SZ catalysts studied in the works
◦
the samples calcined at 400 C in air were used. Samples were
[
5,24–26,28–32] were reduced at fixed temperatures in the range
placed in a quartz homemade IR cell, which consists of two
sections. First section is designed for thermal treatment of the
sample including reduction in a hydrogen stream. Second section
is equipped with KBr windows and allows recording the spectra
at room temperature. Before recording the FTIR spectra, pow-
dered samples were pressed into self-supported wafers (surface
◦
of 200–350 C.
The state of palladium in Pd/SZ reduced at different tempera-
◦
tures (100, 200, and 300 C) was investigated by diffuse-reflectance
IR spectroscopy using CO as a probe molecule [27]. The authors
studied Pd-containing systems in which the metal was sup-
ported on precalcined sulfated zirconia. They concluded that
on the SO4 -promoted surface metal particles with the charge
of +1 (Pd ) and partial positive charge (Pd
However, there is no information concerning the influence of
reduction temperature on the sulfate species decomposition
and the properties of Pd/SZ in metal- or acid-catalyzed reac-
tions.
It should be emphasized that in most cases metal state
was analyzed for the finished catalyst prepared under spe-
cific conditions, and the process of metal centers formation
including the influence of activation temperature was not con-
sidered. We used the temperature-programmed reduction (TPR),
chemisorption technique, infrared spectroscopy of adsorbed car-
bon monoxide (FTIRS), X-ray photoelectron spectroscopy (XPS),
diffuse-reflectance UV–vis spectroscopy (UV–vis DRS), and ben-
zene hydrogenation as a test reaction to characterize the state of
palladium in Pd/SZ catalysts. Furthermore, we focused our atten-
2
density 0.012–0.017 g/cm ). The wafer was placed in the IR cell
2
−
◦
and pretreated at first in air (400 C, 1 h) and then under vac-
+
␦+
◦
−3
)
are present.
uum (25 C, p < 10 mbar). After the activation, Pd/SZ catalyst was
reduced in H2 at a selected temperature (1 h) and outgassed at
the same temperature. Finally all the samples were evacuated
(300 C, p < 10 mbar) for residual adsorbed water removal [33].
CO adsorption was performed stepwise from 0.1 to 10 mbar at room
temperature. The presented spectra were normalized to wafer den-
sity.
◦
−3
2.2.3. XPS
The X-ray photoelectron spectroscopy (XPS) analyses were car-
ried out with a Kratos Analytical electron spectrometer ES300 using
non-monochromatic Mg K␣ source (hꢀ = 1253.6 eV). Survey scan
spectra were recorded using pass energy of 50 eV and a 1 eV step.
High resolution scans of the S2p, Zr3p, C1s, Pd3d and O1s lines were
acquired using a pass energy of 25 and a 0.1 eV step. The analysis

O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
7
was performed using Shirley background subtraction procedure.
The data were processed and analyzed using the XPS-Calc pro-
gram, which has been tested on a number of systems [34,35]. The
experimental curves were fitted with a combination of Gaussian
and Lorentzian peaks. The Gauss/Lorentz ratios were determined
by the XPS-Calc program and varied from 0.9 (for S2p spectra) to
0
.7 (for Zr3p and Pd3d lines). The resulted values of binding energy
splitting (area ratios) for doublets Pd3d, Zr3p and S2p were 5.3 eV
0.7), 13.6 eV (0.6) and 1.2 eV (0.3), correspondingly. These values
(
are in a good agreement with the literature data [36]. All spectra
were calibrated with respect to C1s core-level line at 284.8 eV.
2.2.4. UV–vis DRS
Diffuse reflectance UV–vis spectra (UV–vis DRS) of the
catalysts were recorded using a UV-2501 PC Shimadzu spec-
trometer with IRS-250A diffusion reflection attachment in the
−1
1
1,000–54,000 cm range. Barium sulfate was used as a reference
sample. The powder samples were placed in a 2 mm quartz cell and
◦
◦
Fig. 1. TPR profiles of SZ after calcination at 650 ◦C (a) and Pd/SZ after oxidation in
air at 400 C (b).
pretreated in air (400 C, 1 h) or then under vacuum (300 C, 0.5 h,
◦
−3
p < 10 mbar). Measurements were performed at room tempera-
ture. The UV–vis spectra were transformed into the Kubelka–Munk
2
function (F(R)) calculated as F(R) = (1 − R) /2R, where R is the exper-
It is well known that the large PdO particles obtained after
◦
imentally measured reflectance of the samples [37].
oxidative treatment can be reduced already at −15 C with follow-
ing Pd ␤-hydride formation [40,41]. In agreement with the above,
◦
2.3. Benzene hydrogenation
the sharp negative peak at about 70 C is bound to be in TPR pro-
file assigned to Pd ␤-hydride decomposition [42–44]. Our Pd/SZ
sample did not show low temperature-reducible Pd (II) species
and possessed only PdO reducible at temperatures in the range of
Benzene hydrogenation was carried out in gradientless condi-
tions under atmospheric pressure and at temperatures of 50–90 C
◦
◦
and H /benzene molar ratio 20. Benzene conversion was between
30–70 C. This result evidences high dispersity of supported palla-
2
2
0% and 90%. Catalytic performance measurement and analysis
dium species.
methods used were reported elsewhere [38].
The presence of a broad and small desorption peak indicates that
hydrogen adsorption on the forming palladium metal particles can
occur along with the reduction of oxidized palladium. Subsequent
2.4. Hexane isomerization
◦
hydrogen desorption was observed at 70–90 C.
Skeletal isomerization of n-hexane was carried out in a fixed bed
Pd2 + 1.5H2 = Pd − H + 2H
+
0
+
(1)
(2)
◦
flow reactor at 120 C, under 3.0 MPa total pressure, with weight
−1
hourly space velocity 3 h and H /n-hexane molar ratio 3. Prod-
0
0
2
Pd − H = Pd + 0.5H
2
ucts of reaction were analyzed with on line gas chromatography
using capillary column and flame ionization detector. Prior to each
experiment catalysts were reduced at desired temperature for 2 h
in a hydrogen stream.
The amount of consumed hydrogen in the temperature range of
30–70 C (the H2 to Pd molar ratio was about 0.9) is not in agree-
ment with the stoichiometry of Eq. (1). It is very likely that partial
reduction of palladium may proceed during calcination, by analogy
with Pt supported on sulfated zirconia [13–15].
◦
3
. Results and discussion
◦
Besides, a big broad signal starting at about 150 C was observed
3
.1. Metal centers formation
in the TPR plot of palladium-containing sample. This signal
appeared due to a major part of sulfate reduction. Therefore, the
promotion of SZ with palladium shifts the sulfate reduction pro-
cess to lower temperatures. The result is similar to works [43–45],
demonstrating that the presence of the noble metal in sulfated
zirconia catalyzes sulfate reduction.
Reduction of Pd- and Pt-promoted SZ catalysts is usually carried
◦
out in H2 flow at temperatures from 200 to 300 C [24–26,28,30].
This temperature range is considered as optimal because it is not
only sufficient for the metal reduction, but also restores acidic activ-
ity of SZ after contact with an aqueous solution of metal precursor.
Besides, it is assumed that a noticeable decrease of sulfur content
◦
Unlike palladium, which is reduced below 70 C, platinum is
characterized by higher reduction temperature. According to [20]
reduction of Pt supported on sulfated zirconia is observed in the
resulting in H S evolution does not occur under these conditions.
2
◦
In our study, the process of supported palladium particles for-
temperature range of 100–250 C. It should be emphasized that in
◦
the case of Pd/SZ the treatment in hydrogen may be carried out at
mation in the temperature range of 25–300 C was monitored by
TPR method. TPR curves for SZ and palladium-supported SZ are
shown in Fig. 1. It was found that the plot of SZ support had a small
◦
temperatures up to 150 C without marked decomposition of sul-
fate species. Sulfate reduction becomes more pronounced starting
◦
from 180 ◦C and reaches maximum rate above 250 ◦C. Therefore,
reduction peak at about 180 C. This peak can be attributed to the
elimination of a small amount of weakly bound sulfur during reduc-
tion. Reduction of the remaining sulfur with stronger bonds starts
the advantage of Pd/SZ is the capability of metal particles forma-
tion at low temperature. It leads to decrease of probability of active
sulfate species removal at the reduction stage.
◦
◦
at about 480 C and has a maximum at about 600 C [13,39].
Palladium oxidic species formed on the SZ surface after calcina-
tion in an oxidizing atmosphere were reduced readily in a hydrogen
flow. A low-temperature peak of hydrogen consumption in the
3.2. Sulfate groups decomposition
◦
range of 30–70 C on the TPR curve is associated with the reduction
Decomposition of sulfate species during reduction by hydrogen
was studied using FTIRS and XPS. Fig. 2 shows the FTIR spectra
of palladium (II) species.

8
O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
◦
Fig. 2. FTIR spectra of Pd/SZ catalyst: after oxidation in air at 400 C (a), after reduc-
◦
◦
◦
◦
tion in hydrogen at 70 C (b), 200 C (c), 300 C (d) and 350 C (e). Inset shows the
difference between spectra (c) and (d).
of Pd/SZ catalyst after the activation in air and after subsequent
treatment in H2 at different temperatures in the IR cell. Several
◦
Fig. 3. S2p core-level spectra after oxidation in air at 400 C (a) and after reduction
2−
in hydrogen at 300 ◦C (b).
types of isolated surface sulfate SO
(absorption bands (a.b.) at
4
−1
−1
␯S O 1350–1400 cm and ␯S–O 900–1100 cm ) and pyrosulfate
be seen in the spectra. This peak can be assigned to the sulfide S2
species [52,53]. Sulfide species are observed in the spectrum of oxi-
dized sample, so it can be supposed that exposure to X-rays under
vacuum conditions leads to a minor reduction of surface sulfates to
−
S2O72 groups (a.b. at ␯S O 1405 cm ) [46,47] are observed for
−
−1
the air pretreated sample. IR spectroscopic analysis of the S
O
stretching region for the samples reduced at different tempera-
tures revealed that decomposition of the surface sulfate species in
◦
◦
sulfide. After the action of H2 at 300 C, the quantity of sulfide com-
hydrogen atmosphere takes place even at 70 C (Table 1). According
ponent slightly increases (from 4% to 6% of total sulfur intensity).
Probably, a thin film of sulfide forms on the palladium surface after
reduction.
According to XPS-data, the S/Zr ratio in the catalysts decreased
from 0.21 for the sample pretreated in air to 0.18 for the sample
to IR data, about 34% of the surface sulfate groups are eliminated
◦
during Pd/SZ catalyst reduction at 350 C. The majority of pyrosul-
fate groups, to which the greatest catalytic isomerization activity
is usually attributed [48], decompose during hydrogen treatment
◦
◦
at 300 C (Fig. 2, inset). After reduction of Pd/SZ catalyst at 350 C,
a maximum of the residual ␯ mode shifts to a lower frequency
(
◦
pretreated in hydrogen at 300 C.
S
O
−
1
1360 cm ) (Fig. 2), which is attributed to the removal of active
3.3. State of Pd particles in Pd/SZ
surface sulfate species. Decomposition of sulfate groups of Pd/SZ
catalyst in hydrogen atmosphere leads to the formation of H S,
which is found in the pumped out gases by darkening of the paper
2
The state of Pd particles in Pd/SZ catalyst was studied by the
FTIR spectroscopy of adsorbed CO molecules, X-ray photoelectron
spectroscopy, and diffuse reflectance UV–vis spectroscopy.The IR
spectra of CO adsorbed at room temperature on differently pre-
filter impregnated with Pb(CH COO) . This result is in agreement
3
2
with TPR-MS data for the Pt-promoted sulfated zirconia reduction
process leading to the formation of H S without SO2 [45].
2
◦
treated Pd/SZ catalysts (oxidized and reduced at 150–350 C) are
Fig. 3 presents the S2p core-level spectra of the samples oxi-
◦
◦
shown in Fig. 4. The spectrum of CO adsorbed on the sample that
was activated in air shows several a.b. located at 1935, 2030, 2090,
dized in air at 400 C (a) and reduced in H2 at 300 C (b). As it
can be seen, the S2p spectra consist of a high-intensity broad peak
−
1
2
2
125, 2150, 2170 and 2198 cm
090 cm
(Fig. 4(a)). The sharp band at
with the binding energy E (S2p) ∼169 eV and a low-intensity peak
b
−
1
−1
and the broad band around 1935 cm
are ascribed
at ∼162 eV. The first peak characterizes sulfur as a component of
sulfate species [49–51]. From the broad asymmetric line shape of
sulfate peak in S2p spectra we can propose that several different
sulfate species seem to be present on the sample surface. Along with
0
to terminal and bridge-coordinated CO on Pd , respectively [54].
−
1
Bands at 2030, 2125 and 2150 cm can be attributed, respectively,
+
+
to bridged CO complex with Pd , and linear CO complexes with Pd
and Pd2+ isolated ions and the band at 2170 cm can be assigned to
−1
sulfates, another low-intensity component with E ∼ 161.6 eV can
b
2+
CO linearly adsorbed on Pd in PdO species [55]. Besides, the band
−
1
at 2196–2198 cm is present in all spectra and may be assigned
Table 1
to CO adsorption on Zr4+ ions [47]. The highest intensity of this a.b.
Variation of surface sulfate groups concentration as a function of reduction temper-
ature of Pd/SZ.
◦
is observed for the sample reduced in H2 at 300 C. This fact con-
firms earlier studies concerning the necessity of high-temperature
reduction of SZ to obtain the greatest Lewis acidity of the catalyst
after metal incorporation.
Sample
Pd/SZ
Treatmenta
I (1350–1405 cm−1)b, %
◦
Ox 400
Red 70
C
100
92
87
80
◦
C
As it was shown by TPR, a negligible reduction of metal dur-
ing thermal treatment in air is possible. According to FTIR data,
◦
Red 200
Red 300
Red 350
C
◦
◦
C
C
0
66
in oxidized Pd/SZ sample palladium is presented mostly as Pd
−
1
a
(high intensities of a.b. at 1935 and 2090 cm ). Supposedly, the
formation of metallic palladium can be caused by evacuation at
high temperatures during the pretreatment in IR cell.
Ox: oxidation; Red: reduction.
Integrated IR intensity of a.b. at 1350–1405 cm , %. The intensity of this a.b. for
b
−1
◦
Pd/SZ oxidized at 400 C was assumed as 100%.

O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
9
2+
decreases the concentration of Pd ions in PdO (a decrease in the
−
1
intensity of a.b. at 20,500 cm was observed). It should be noted
that changes of the absorption background in the visible range for
vacuum pretreated Pd/SZ catalyst can be related to strong struc-
tureless absorption due to the presence of large palladium metal
particles.
In the IR spectra of CO adsorbed on Pd/SZ samples that were
◦
reduced at 150–200 C (Figs. 4(b), (c) and 6(a)), the a.b. at 1930 and
−
1
2
090–2095 cm attributed to metallic palladium dominate. Bands
−
1
at 2125–2170 cm , assigned to CO adsorbed on oxidized Pd ions,
almost vanish.
The differences between spectra of adsorbed CO on Pd/SZ sam-
◦
ples reduced at 200 and 300–350 C are dramatic (Fig. 4(c)–(e)).
An increase in the reduction temperature suppresses the bridge-
coordinated CO IR bands, decreases the intensity and slightly
shifts the stretching frequency of linearly adsorbed CO to higher
wavenumbers. Such shift occurs when CO is chemisorbed on the
◦
Fig. 4. FTIR spectra of CO adsorbed on Pd/SZ (25 C; 10 mbar) after oxidation in air at
4
(
sulfur-saturated Pd surface [57,58]. So, appearance of the band at
◦
◦
◦
◦
◦
C
00 C (a), after reduction in hydrogen at 150 C (b), 200 C (c), 300 C (d) and 350
e). All spectra are background subtracted. Spectra were offset for clarity. Dashed
lines show computer deconvolution of the spectrum (a).
−1
2
105–2108 cm can be attributed to linear CO complexes with pal-
ladium in an electron-deficient state with high S-coverage. In [27]
it was supposed that a possible reason for the decrease in concen-
tration of bridging CO is the partial covering of Pd surface by the
products of interaction with surface sulfur species. According to
[
58], increasing sulfur content on a single-crystal face of Pd (1 0 0)
decreases the CO saturation coverage on account of palladium site
blocking. Thus, we suppose that low CO chemisorption capacity of
0
Pd atoms in Pd/SZ samples reduced at high temperatures is due to
partial sulfur coating of the metal surface.
−
1
Therewith, new a.b. at 2135 and 2160 cm
are observed in
the spectra of CO adsorbed on Pd/SZ samples that were reduced
◦
at 300–350 C (Figs. 4 and 6). These bands appear on admission
of 0.1 mbar CO and grow in intensity with increasing CO pressure
+
(
Fig. 6(b)); they are attributed to CO complexes with oxidized Pd
2+
n+
2−
and Pd species, probably in Pd –S sites. In [59] it was shown
that upon direct sulfidation with H S at 250 C, Pd particles trans-
◦
2
formed into PdS . It can be supposed that during the reduction of
2
◦
Pd/SZ at 300 C metallic particles are poisoned by H S, which is
2
produced by reducing the sulfate species.
Fig. 5. UV–vis DR spectra of SZ (a) and Pd/SZ after successive pretreatment: oxida-
tion in air at 400 C (b), evacuation at 300 C (c).
◦
◦
Fig. 7 shows the X-ray photoelectron spectra in the spectral
region of Zr3p and Pd3d core-level lines. The difference curves
between experimental spectrum and the envelope of the fit are
presented under each spectrum.
The effect of evacuation on palladium reduction was studied by
diffuse reflectance UV–vis spectroscopy. Pd/SZ catalyst after the
oxidation is characterized by a.b. at 20,500, 34,000, 39,500 and
◦
After calcination in air at 400 C (Fig. 7(a)), palladium spectrum
can be described by one doublet line with Eb (Pd3d5/2) 336.5 eV.
Such value of Eb is close to that for the oxidized palladium species
in palladium oxide PdO [60–62]. A small decrease in Eb (0.2–0.3 eV)
can be attributed to the formation of non-crystallized oxide palla-
−1
−1
4
6,000 cm (Fig. 5(b)). The a.b. at 20,500 and 39,500 cm can be
attributed, respectively, to d–d transition and L → M charge transfer
of Pd2+ ions in D oxygen environment [56]. Evacuation at 300 C
◦
4
h
◦
◦
◦
Fig. 6. FTIR spectra of CO adsorbed on Pd/SZ (25 C; 0.1, 0.2, 3 and 10 mbar) after reduction in hydrogen at 200 C (a) and 300 C (b). All spectra are background subtracted.

1
0
O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
Fig. 8. Effect of the reduction temperature on palladium dispersion in Pd/SZ.
a typical model reaction for the testing of metal function [66,67].
This reaction occurs on metal sites and the deactivation of the
catalyst does not take place under low temperature conditions.
Results of low temperature benzene hydrogenation are presented
as a kinetic curve of reaction rate versus benzene partial pressure
Fig. 7. Zr3p and Pd3d core-level spectra of the samples after oxidation in air at
◦
◦
4
00 C (a) and after reduction in hydrogen at 300 C (b). For clear identification,
Pd3d spectra are multiplied by 3.
dium species. After the action of H (Fig. 7(b)), Pd3d spectrum gives
2
(Fig. 9).
a peak with E (Pd3d5/2) 335.6 eV, which is assigned to small metal
b
All the catalysts were reduced for 1 h in an H stream at different
2
particles. Bulk metal palladium is known to have Eb of Pd3d5/2 at
temperatures (similar to dispersion test) prior to the reaction. Anal-
ysis of experimental data showed that there were no significant
changes of the reaction active sites when reduction temperature
was varied. For all the catalysts, the shape of kinetic curves did
not change in the entire range of conversions, and zero order
was observed at high benzene partial pressures. Activation ener-
gies were similar for all studied samples, their values being about
3
35.2 eV [60,62,63]. In our case, a small increase of Eb is likely to
originate from the size effect [64]. The formation of bulk palladium
sulfide was not observed because the corresponding value of Eb
of Pd3d5/2, which is about 337.2 eV [65], was not detected. How-
ever, Chaplin et al. [65] showed that low-sulfide-fouled palladium
sample (with the molar ratio S/Pd about 0.2 as determined by XPS)
gives a Pd3d5/2 peak at 335.4 eV. One can see small increase of rel-
ative intensity of Pd signal after reduction (the Pd/Zr ratio is 0.01
after oxidation and 0.02 after reduction). This growth of Pd3d inten-
sity can be explained by the morphological changes of the sample
surface under the action of H2.
1
4–16 kcal/mol.
Fig. 10 presents a comparison of hydrogenation activity of Pd/SZ
catalysts reduced at different temperatures. The rate of reaction at
zero order was taken as a measure of activity.
There was a clear effect of pretreatment conditions on the con-
version of benzene to cyclohexane: a decrease in hydrogenation
activity correlated with a higher reduction temperature. Pd/SZ cata-
lyst, reduced at the temperature corresponding to metal formation
Thus, changes of the metal surface after high temperature reduc-
tion in the H2 atmosphere, which are demonstrated by the FTIR
spectra, have no significant effect on the electronic state of metal
particles: according to XPS data, Pd remains in the metallic state.
(
according to TPR data), demonstrated the highest conversion. In
H S resulting from sulfate reduction blocks the active metal surface,
2
◦
the case of reduction temperature above 120 C, before the cat-
most likely without formation of PdS in large amounts. However, it
cannot be unambiguously concluded from Pd3d and S2p core-level
spectra whether the sulfide film forms on the palladium surface
after reduction or not.
alytic test H S was detected in the exhaust gases, and poisoning of
2
the metal function was observed.
3.4. Chemisorption capacity and hydrogenation activity of the
supported Pd
Pulse chemisorption of CO at room temperature for the samples
◦
reduced at 70–120 C showed a relatively high palladium disper-
sion (more than 60%). However, as the reduction temperature was
increased, CO uptake declined (Fig. 8). Only a small amount of CO
chemisorbed on the metal surface atoms in Pd/SZ reduced at tem-
◦
peratures above 250 C due to poisoning of the metal by sulfur.
An important point is that the deactivation was reversible at
◦
least at the reduction temperatures below 300 C, and regenera-
◦
tion stage, including calcination at 300 C in air and subsequent
◦
reduction at 120 C, led to nearly full recovery of metal dispersion.
Hydrogenation activity of supported Pd was assessed with the
catalytic test of benzene hydrogenation. Benzene hydrogenation is
◦
C
Fig. 9. Effect of benzene partial pressure on hydrogenation reaction rate at 90
over Pd/SZ reduced at 120 C. The H2/benzene molar ratio was 20.
◦

O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
11
Table 2
Isomerization of n-hexane over Pd/SZ and Pt/SZ catalysts.
Samplea
tRedb
,
◦
C
Xc, %
Yield , wt%
d
C1–C5
2,2-DMB
2,3-DMB
2-MP
3-MP
n-C6
Others
Pd/SZ
Pd/SZ
Pt/SZ
150
250
250
83.8
74.6
72.5
0.8
0.5
0.6
25.6
19.2
19.3
8.9
7.7
7.4
31.7
30.8
29.4
16.5
16.1
15.5
16.1
25.2
27.3
0.4
0.5
0.5
a
Metal content: 0.3 wt%.
Reduction temperature.
Conversion of n-hexane.
b
c
d
2
,2-DMB: 2,2-dimethylbutane; 2,3-DMB: 2,3-dimethylbutane; 2-MP: 2-methylpentane; 3-MP: 3-methylpentane; n-C6: n-hexane; others: methylcyclopentane, cyclo-
hexane, benzene and C7+.
Thus, in Pd/SZ catalysts, after the reduction treatment under
conventional conditions at 250–300 C, hydrogenation activity of
and stable isomerization of alkanes. However, it should be taken
into account that the reductive treatment at 250–300 C is essen-
◦
◦
palladium is strongly inhibited. This might be the reason why Pd/SZ
isomerization catalysts need to be operated at high hydrogen par-
tial pressures. In the case of reaction at high pressures, partial
desorption of sulfur species from the metal surface and recovery
of hydrogen activation capacity occur. Activation of H2 is required
for the hydrogenation of coke precursors and the prevention of
deactivation [32,39].
tial for the formation of metal particles in Pt-containing catalyst
[20]. At the same time Pd on the surface of sulfated zirconia can be
completely reduced even at 70 C. In the case of Pd/SZ a capability
◦
of obtaining active metal sites at lower reduction temperature pre-
determines the prevention of sulfate species decomposition during
reduction. It means that there is a possibility for retaining the acid-
ity of Pd/SZ, which is provided by the surface sulfate species. The
latter is very important for alkanes isomerization because acid
sites of the catalyst provide generation and skeletal rearrange-
ments of the corresponding carbenium ions [7]. Test reaction of
n-hexane isomerization confirmed higher acidity of Pd/SZ catalyst
It should be noted that regeneration of Pd/SZ catalyst deac-
◦
tivated during reduction at temperatures higher than 300 C not
only increased CO chemisorption capacity of palladium, but also
enhanced hydrogenation activity of the catalyst. Oxidative treat-
◦
◦
ment of deactivated catalyst with successive reduction at 120 C
reduced at 150 C. The catalyst showed higher n-hexane conver-
−
1
−1
increased benzene hydrogenation rate to 0.011 l g
h . This value
sion and 2,2-dimethylbutane yield (Table 2). It can be supposed
that higher acidity in this case is provided by prevention of active
sulfate species decomposition due to the capability of palladium
metallic particles formation at low reduction temperatures.
is comparable with the rate of benzene hydrogenation on the most
active Pd/SZ catalysts reduced at low temperatures.
3.5. Isomerization activity of the catalysts
◦
◦
Acid function of Pd/SZ catalyst reduced at 150 and 250 C
4. Conclusions
was tested by reaction of n-hexane isomerization at 120 C.
Pd-containing catalysts were also compared with Pt-promoted sul-
fated zirconia. The results are shown in Table 2.
It has been established that during hydrogen reduction of Pd/SZ
◦
catalyst the metal centers are formed at about 30–70 C. They are
characterized by high palladium dispersion and activity in benzene
hydrogenation.
It can be seen that both Pd- and Pt-containing sulfated zirconia
◦
catalysts reduced in a hydrogen stream at 250 C showed similar n-
hexane conversions (74.6% and 72.5% respectively) and isohexanes
yields (73.8 and 71.6 wt% respectively). Yields of the most valu-
able isomerization product 2,2-dimethylbutane were 19.2 wt% for
Pd/SZ and 19.3 wt% for Pt/SZ. This fact indicates that after this acti-
vation procedure both catalysts are characterized by close relation
between metal and acid functions, which are required for effective
At the same time, palladium reduction initiates sulfate decom-
◦
position, which was observed from 70 C, and poisoning of the
◦
metal function takes place. At 250 C and higher temperatures,
products of sulfate reduction poison the metal particles completely.
As a result, the hydrogenation capacity and CO adsorption are
strongly inhibited. According to XPS and FTIRS study, palladium
particles in the poisoned samples are oxidized only partly, but the
main part is represented by metallic phase probably covered with
sulfur species.
The results of acid-catalyzed reaction of n-hexane isomerization
◦
at 120 C show that Pd/SZ catalyst reduced at lower tempera-
◦
ture (150 instead of 250 C) has higher conversion (83.8% versus
7
4.6%), yields of isohexanes (82.7 wt% versus 73.8 wt%) and 2,2-
dimethylbutane (25.6 wt% versus 19.2 wt%). Higher isomerization
activity in this case is provided by prevention of active sulfate
species decomposition due to the capability of palladium metallic
particles formation at low reduction temperatures. Thus, the effec-
tive SZ-based isomerization catalyst with improved metal function
may be obtained with the use of palladium as a promoter.
Acknowledgements
This work was supported by the President of Russian Federation
grant under the program for young Russian scientists and leading
research schools (project no. 8552.2006.3).
Fig. 10. Dependence of benzene hydrogenation reaction rate on the reduction tem-
perature. Reaction temperature 70 C, benzene pressure 0.04 atm.
◦

1
2
O.B. Belskaya et al. / Applied Catalysis A: General 387 (2010) 5–12
References
[36] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photo-
electron Spectroscopy, Physical Electronics Inc., Eden Prairie, MN, 1992.
37] H.-P. Boehm, H. Knozinger, in: J.R. Anderson, M. Boudart (Eds.), Catalysis Science
and Technology, vol. 4, Springer-Verlag, Berlin-Heidelberg-New York, 1983, pp.
[
[
[
[
[
[
[
[
[
[
1] X. Song, A. Sayari, Catal. Rev. Sci. Eng. 38 (1996) 329–412.
2] G.D. Yadav, J.J. Nair, Micropor. Mesopor. Mater. 33 (1999) 1–48.
3] K. Ebitani, J. Konishi, H. Hattori, J. Catal. 130 (1991) 257–267.
4] K. Ebitani, J. Tsuji, H. Hattori, H. Kita, J. Catal. 135 (1992) 609–617.
5] M. Hino, K. Arata, Catal. Lett. 30 (1995) 25–30.
6] M. Hino, K. Arata, J. Chem. Soc. Chem. Commun. (1995) 789–790.
7] E. Iglesia, S.L. Soled, G.M. Kramer, J. Catal. 144 (1993) 238–253.
8] F. Garin, D. Andriamasinoro, A. Abdulsamad, J. Sommer, J. Catal. 131 (1991)
39–209.
38] O.B. Bogomolova, N.M. Ostrovskii, M.D. Smolikov, A.S. Belyi, V.K. Duplyakin,
React. Kinet. Catal. Lett. 40 (1989) 131–136.
[
[
39] J.M. Grau, C.R. Vera, J.M. Parera, Appl. Catal. A 227 (2002) 217–230.
40] W. Palczewska, in: Z. Paal, P.G. Menon (Eds.), Hydrogen Effects in Catalysis,
Dekker, New York, 1988, pp. 373–386.
[
[
41] Z. Karpinski, Adv. Catal. 37 (1990) 45–100.
42] S. Melada, R. Rioda, F. Menegazzo, F. Pinna, G. Strukul, J. Catal. 239 (2006)
1
99–203.
[9] J.C. Yori, M.A. Damato, G. Costa, J.M. Parera, J. Catal. 153 (1995) 218–223.
422–430.
[10] K. Ebitani, H. Konno, T. Tanaka, H. Hattori, J. Catal. 135 (1992)
[
[
[
[
43] F. Menegazzo, P. Burti, M. Signoretto, M. Manzoli, S. Vankova, F. Boccuzzi, F.
Pinna, G. Strukul, J. Catal. 257 (2008) 369–381.
44] G. Bernardotto, F. Menegazzo, F. Pinna, M. Signoretto, G. Cruciani, G. Strukul,
Appl. Catal. A 358 (2009) 129–135.
6
0–67.
[
[
[
[
[
[
11] K. Ebitani, H. Konno, T. Tanaka, H. Hattori, J. Catal. 143 (1993) 322–323.
12] Z. Paal, M. Muhler, R. Schlogl, J. Catal. 143 (1993) 318–321.
13] A. Sayari, A. Dicko, J. Catal. 145 (1994) 561–564.
14] A. Dicko, X.M. Song, A. Adnot, A. Sayari, J. Catal. 150 (1994) 254–261.
15] J. Zhao, G.P. Huffman, B.H. Davis, Catal. Lett. 24 (1994) 385–389.
16] C.R. Vera, C.L. Pieck, K. Shimizu, J.C. Yori, J.M. Parera, Appl. Catal. A 232 (2002)
45] C. Morterra, G. Cerrato, S. Di Ciero, M. Signoretto, F. Pinna, G. Strukul, J. Catal.
1
65 (1997) 172–183.
46] M. Bensitel, O. Saur, J.-C. Lavalley, B.A. Morrow, Mater. Chem. Phys. 19 (1988)
47–156.
1
1
69–180.
[
[
[
[
47] C. Morterra, G. Cerrato, C. Emanuel, V. Bolis, J. Catal. 142 (1993) 349–367.
48] X. Li, K. Nagaoka, J.A. Lercher, J. Catal. 227 (2004) 130–137.
49] L. Portela, P. Grange, B. Delmon, J. Catal. 156 (1995) 243–254.
50] J. Baltrusaitis, D.M. Cwiertny, V.H. Grassian, Phys. Chem. Chem. Phys. 9 (2007)
[
[
17] S. Vijay, E.E. Wolf, Appl. Catal. A 264 (2004) 117–124.
18] J.-M. Manoli, C. Potvin, M. Muhler, U. Wild, G. Resofszki, T. Buchholz, Z. Paal, J.
Catal. 178 (1998) 338–351.
[
[
[
19] Z. Paal, U. Wild, M. Muhler, J.-M. Manoli, C. Potvin, T. Buchholz, S. Sprenger, G.
Resofszki, Appl. Catal. A 188 (1999) 257–266.
20] J.M. Grau, J.C. Yori, C.R. Vera, F.C. Lovey, A.M. Condo, J.M. Parera, Appl. Catal. A
5542–5554.
[
[
[
51] Y. Li, R.A. van Santen, Th. Weber, J. Solid State Chem. 181 (2008) 3151–3162.
52] E. Puello-Polo, J.L. Brito, J. Mol. Catal. A 281 (2008) 85–92.
53] V. Yufit, D. Golodnitsky, L. Burstein, M. Nathan, E. Peled, J. Solid State Elec-
trochem. 12 (2008) 273–285.
2
65 (2004) 141–152.
21] C.R. Vera, J.C. Yori, C.L. Pieck, S. Irusta, J.M. Parera, Appl. Catal. A 240 (2003)
61–176.
1
[
54] N. Sheppard, T.T. Nguyen, in: R.J.H. Clark, R.E. Hester (Eds.), Advances in Infrared
and Raman Spectroscopy, vol. 5, Heyden and Sons, Philadelphia, 1978, pp.
[
[
[
22] K. Moller, D.C. Koningsberger, T. Bein, J. Phys. Chem. 93 (1989) 6116–6120.
23] H. Yasuda, N. Matsubayashi, T. Sato, Y. Yoshimura, Catal. Lett. 54 (1998) 23–27.
24] K. Watanabe, N. Oshio, T. Kawakami, T. Kimura, Appl. Catal. A 272 (2004)
67–147.
[
[
[
[
[
[
55] K.I. Hadjiivanov, G.N. Vayssilov, Adv. Catal. 47 (2002) 307–511.
56] A. Rakai, D. Tessier, F. Bozon-Verduraz, New J. Chem. 16 (1992) 869–875.
57] C.R. Guerra, J. Colloid Interface Sci. 29 (1969) 229–234.
58] S.W. Jorgensen, R.J. Madix, Surf. Sci. 163 (1985) 19–38.
59] A. Vazquez-Zavala, S. Fuentes, F. Pedraza, Appl. Surf. Sci. 78 (1994) 211–218.
60] Th. Pillo, R. Zimmermann, P. Steiner, S. Hufner, J. Phys.: Condens. Matter 9
2
81–287.
25] K. Watanabe, T. Kawakami, K. Baba, N. Oshio, T. Kimura, Appl. Catal. A 276
2004) 145–153.
26] G. Larsen, E. Lotero, R.D. Parra, L.M. Petkovic, H.S. Silva, S. Raghavan, Appl. Catal.
A 130 (1995) 213–226.
[
[
(
[
[
27] A.V. Ivanov, L.M. Kustov, Russ. Chem. Bull. 47 (1998) 55–59.
28] A.V. Ivanov, A.Yu. Stakheev, L.M. Kustov, Stud. Surf. Sci. Catal. 130 (2000)
(
1997) 3987–3999.
61] M. Brun, A. Berthet, J.C. Bertolini, J. Electron Spectrosc. Relat. Phenom. 104
1999) 55–60.
[
[
2
63–268.
(
[
[
[
[
[
[
29] U.B. Demirci, F. Garin, Catal. Lett. 76 (2001) 45–51.
62] A.M. Venezia, L.F. Liotta, G. Pantaleo, V. La Parola, G. Deganello, A. Beck, Zs.
Koppany, K. Frey, D. Horvath, L. Guczi, Appl. Catal. A 251 (2003) 359–368.
63] K. Otto, L.P. Haack, J.E. de Vries, Appl. Catal. B 1 (1992) 1–12.
64] M.G. Mason, Phys. Rev. B 27 (1983) 748–762.
30] N. Bouchenafa-Saib, R. Issaadi, P. Grange, Appl. Catal. A 259 (2004) 9–15.
31] U.B. Demirci, F. Garin, J. Mol. Catal. A 271 (2007) 216–220.
32] J.M. Grau, J.M. Parera, Appl. Catal. A 162 (1997) 17–27.
33] F. Babou, G. Coudurier, J.C. Vedrine, J. Catal. 152 (1995) 341–349.
34] O.V. Magaev, A.S. Knyazev, O.V. Vodyankina, N.V. Dorofeeva, A.N. Salanov, A.I.
Boronin, Appl. Catal. A 344 (2008) 142–149.
[
[
[
65] B.P. Chaplin, J.R. Shapley, C.J. Werth, Environ. Sci. Technol. 41 (2007)
5491–5497.
[
[
66] T.A. Dorling, R.L. Moss, J. Catal. 5 (1966) 111–115.
67] J.M. Basset, G. Dalmai-Imelik, M. Primet, R. Mutin, J. Catal. 37 (1975) 22–36.
[35] L.S. Kibis, A.I. Titkov, A.I. Stadnichenko, S.V. Koscheev, A.I. Boronin, Appl. Surf.
Sci. 255 (2009) 9248–9254.