Effects of NaCl on Pt/ZrO2 catalysts for selective hydrogenation of crotonaldehyde

Article abstract of DOI:10.1016/j.apcata.2010.08.044

A series of NaCl-modified Pt/ZrO₂ catalysts were prepared and tested for vapor phase selective hydrogenation of crotonaldehyde. It was found that the reactivities of the catalysts increased with NaCl contents, with the highest selectivity to crotyl alcohol (60%) at a NaCl content of 1%. Ammonia temperature programmed desorption and Fourier transmission infrared spectra of pyridine adsorption results indicated that the strengths of surface Lewis acids of the catalysts were weakened with increasing NaCl contents, such weakening was responsible for the enhancement of the reactivity. Diffuse reflectance infrared transform spectra of CO adsorption and X-ray photoelectron spectrometer results suggested that increased electron density on Pt species with the NaCl modification was responsible for the enhanced selectivity. However, high NaCl content (>3%) in the catalyst resulted in the growth of Pt particles and fewer interfacial Pt sites contacting with the surface Lewis acid sites, such changes were unfavorable for the selectivity.

Full text of DOI:10.1016/j.apcata.2010.08.044

Applied Catalysis A: General 388 (2010) 134–140
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Effects of NaCl on Pt/ZrO2 catalysts for selective hydrogenation of crotonaldehyde
Xiao-Xia Wang, Hai-Ying Zheng, Xi-Jing Liu, Guan-Qun Xie, Ji-Qing Lu, Ling-Yun Jin, Meng-Fei Luo∗
Key Laboratory of the Ministry of Education for Advanced Catalysis Materials, Institute of Physical Chemistry, Zhejiang Normal University, Jinhua 321004, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of NaCl-modified Pt/ZrO2 catalysts were prepared and tested for vapor phase selective hydrogena-
tion of crotonaldehyde. It was found that the reactivities of the catalysts increased with NaCl contents,
with the highest selectivity to crotyl alcohol (60%) at a NaCl content of 1%. Ammonia temperature pro-
grammed desorption and Fourier transmission infrared spectra of pyridine adsorption results indicated
that the strengths of surface Lewis acids of the catalysts were weakened with increasing NaCl contents,
such weakening was responsible for the enhancement of the reactivity. Diffuse reflectance infrared trans-
form spectra of CO adsorption and X-ray photoelectron spectrometer results suggested that increased
electron density on Pt species with the NaCl modification was responsible for the enhanced selectivity.
However, high NaCl content (>3%) in the catalyst resulted in the growth of Pt particles and fewer interfa-
cial Pt sites contacting with the surface Lewis acid sites, such changes were unfavorable for the selectivity.
Received 18 February 2010
Received in revised form 18 August 2010
Accepted 18 August 2010
Available online 26 August 2010
Keywords:
Pt/NaCl/ZrO2 catalyst
NaCl
Crotonaldehyde hydrogenation
Crotyl alcohol
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
ification could enhance the catalytic performance, probably by
changing the electronic structure of the active Pt metal. Riguetto
Unsaturated alcohols are very important intermediates for
pharmaceutical, agrochemical, and fragrance productions [1–3].
Selective hydrogenation of ␣,␤-unsaturated aldehydes using het-
erogeneous catalysts is of great interest because it can facilitate
the reaction process and lower the cost. More importantly, the het-
erogeneous method can avoid the waste that is produced in the
present process using NaBH4 or AlLiH4 [4]. However, the challenge
for heterogeneous processes lies in the fact that the reduction of
et al. [11] demonstrated that the electronic properties of the active
metal exerted important impacts on the selectivity and concluded
that the surface electron density of the active metal particles was
crucial to improving the adsorption and reactivity of the C
O
bond. On the other hand, the structure of the Pt metal is also
essential to the reaction, especially the selectivity. Abid et al. [12]
reported high selectivity to unsaturated alcohol on Pt/CeO2 cat-
alysts. They concluded that the selectivity of Pt/CeO2 catalyst is
mainly determined by the metal particle size. The large parti-
cles of Pt containing high fractions of Pt(1 1 1) surfaces favored
the adsorption of crotonaldehyde via the C O bond, leading to
a higher selectivity of unsaturated alcohol. Similar conclusions
were also drawn by Plomp et al. for the selective hydrogenation of
cinnamaldehyde [13].
Surface properties of the support are also reported to be crucial
to the reactivities of the Pt catalysts. Englisch et al. [14] found that
Lewis acid sites enhanced the adsorption strength of the C O bond,
thus resulting in the enhanced selectivity to crotyl alcohol. So the
appropriate strength of Lewis acid sites is critical for the reaction.
In this work, the role of NaCl as a promoter for the Pt/ZrO2 cata-
lysts for gas phase hydrogenation of crotonaldehyde is investigated
and reported. The addition of NaCl in the catalysts not only affects
the electronegativity and particle size of Pt metal, but also the sur-
face acidity of the support, which greatly influences the catalytic
behavior of the catalysts for selective hydrogenation of crotonalde-
hyde.
the C C bond is usually preferred over the reduction of the C
O
bond on monometallic catalysts due to both thermodynamic and
kinetic factors [5,6].
Intensive studies have been performed on the Pt based catalysts,
the most commonly used catalysts for selective hydrogenation of
␣,␤-unsaturated aldehydes. Usually, the selectivity to the desired
unsaturated alcohol is low on the monometallic Pt catalysts sup-
ported on non-reducible support, due to the preferential reduction
of C C bond. In order to obtain high selectivity, people have tried
to modify the Pt catalysts. For example, it was reported that the
addition of electropositive metals such as Fe, Sn and Ni could
enhance the selectivity [7–9]. Also, Cl species in the catalysts, which
originated from the H2PtCl6 were considered important for the
reaction [10]. These findings indicated that such electronic mod-
∗
Corresponding author. Tel.: +86 579 82283910; fax: +86 579 82282595.
E-mail address: mengfeiluo@zjnu.cn (M.-F. Luo).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.08.044

X.-X. Wang et al. / Applied Catalysis A: General 388 (2010) 134–140
135
Table 1
syringe. Pt dispersion was evaluated from the consumption of CO,
assuming that CO:Pt = 1:1.
Contents of Pt, Cl and Na in Pt/NaCl/ZrO2 catalysts.
The acidic properties of the catalysts were studied by ammo-
nia temperature programmed desorption (NH3-TPD) and Fourier
transmission infrared spectra (FTIR) analysis of pyridine adsorp-
tion. For the NH3-TPD experiment, the catalyst was pretreated
in a flow of N2 (30 ml/min) at 500 ◦C for 1 h, and then cooled to
room temperature. Afterwards, a flow of NH3 (30 ml/min) was
introduced into the reactor for 15 min, followed by purging with
a N2 flow (30 ml/min at 100 ◦C) to remove the physisorbed NH3.
The NH3-TPD profile was measured using a TECHTEMP GC 7890
II with a thermal conductivity detector (TCD) when the sample
was heated at a rate of 10 ◦C/min from 200 to 560 ◦C. Mean-
while, the above process at the same conditions was repeated
and the total amount of NH3 desorbed was determined by reac-
tion with an excess of dilute hydrochloric acid and back titration
with sodium hydroxide solution. The sensitive indicator contains
a mixture of 0.1% brom–cresol green ethanol solution and 0.2%
methyl red ethanol solution with a volume ratio of 3:1. For the
FTIR experiment, the spectra were recorded at room temperature
on a NEXUS 670 Fourier transform infrared spectroscope equipped
with a diffuse reflectance accessory. The self-supported sample
wafer was dried in an oven for 1 h at 100 ◦C prior to pyridine
treatment for IR measurements. The catalyst was exposed first to
pyridine directly, and then kept in a vacuum oven at 120 ◦C for 1 h to
remove physisorbed pyridine. After the sample was cooled down
to room temperature, each IR spectrum was recorded in a spec-
tral range of 1200–2250 cm−1 with 64 scans and at a resolution of
Catalyst
Content (wt.%)
Pta
Cla
Nab
Pt/ZrO2
2.99
2.92
2.83
2.79
2.81
2.49
2.53
2.69
3.07
3.84
–
Pt/NaCl(0.5)/ZrO2
Pt/NaCl(1)/ZrO2
Pt/NaCl(3)/ZrO2
Pt/NaCl(6)/ZrO2
0.12
0.23
0.66
1.20
a
Detected by XRF.
Detected by ICP-AES.
b
2. Experimental
2.1. Catalyst preparation
All reagents were analytically pure and were purchased
from Sinopharm Chemical Reagent Co. ZrO2 was prepared from
hydrochloride precursors by a precipitation method. An NH3 aque-
ous solution was added to a ZrOCl2 solution under constant stirring.
The pH of the resulting mixture was controlled at 10 0.5. The mix-
ture was stirred at room temperature for 24 h, and the precipitate
was then filtered after aging at 90 ◦C for 5 days and washed with
distilled water several times. The achieved white powder was dried
at 70 ◦C for 4 h under vacuum, followed by calcination at 400 ◦C
for 4 h, leading to the final support with a specific surface area of
263 m2/g.
The NaCl/ZrO2 supports were prepared by impregnating ZrO2
with a NaCl aqueous solution (0.1 mol/L). The NaCl-modified sup-
port was dried at 120 ◦C overnight and calcined at 400 ◦C for 4 h.
The modified supports were denoted as NaCl(x)/ZrO2, where x (0,
0.5, 1, 3 and 6) refers to the molar ratio of NaCl/ZrO2 (%).
The supported Pt catalysts were prepared by impregnating the
supports with aqueous solution of H2PtCl6, with a nominal Pt con-
tent of 3 wt.%. Excess solution was removed by mild evaporation.
Finally, the catalysts were dried overnight at 120 ◦C. The catalysts
were denoted as Pt/NaCl(x)/ZrO2.
4 cm−1
.
In order to study the oxidation state of Pt, we performed in situ
diffuse reflectance infrared transform spectra (DRIFTS) measure-
ment of CO-chemisorption. Prior to the measurement, the sample
was in situ reduced in a H2–N2 mixture (30 ml/min, 5 vol% H2)
at 500 ◦C for 1 h. After the pretreatment, the sample was cooled
down to 25 ◦C and a flow of He (30 ml/min) was fed to remove
the residual H2. Then the sample was exposed to CO for 30 min.
Finally, the sample was purged by He for another 30 min and
the spectra were recorded. In all cases the spectra were taken at
25 ◦C, with a resolution of 4 cm−1 and a cumulative averaging of
64 scans.
The crotonaldehyde chemisorption on catalyst surface was per-
formed by DRIFTS measurement. The pretreatment condition was
the same as that in the in situ DRIFTS of CO adsorption. When
the sample had cooled down to 50 ◦C, a flow of crotonaldehyde
was introduced by pure He for 15 min. The crotonaldehyde was
introduced in a trap set maintained at 0 ◦C to achieve a constant
crotonaldehyde partial pressure (1062 Pa). Finally, the sample was
purged by He for 120 min and a spectrum was recorded. In all cases,
the spectra were taken at 50 ◦C with a resolution of 4 cm−1 and with
a cumulative averaging of 64 scans.
For the Pt/ZrO2 and Pt/NaCl(3)/ZrO2 catalysts, after the ultra-
pure H2 reduction at 500 ◦C treatment, the oxidation states of Pt
were measured on a VG ESCALAB MK-2 X-ray photoelectron spec-
trometer (XPS) with Al K␣ monochromatic X-rays. The voltage and
power for the measurements were 12.5 kV and 250 W, respectively.
The vacuum in the test chamber during the collection of spectra
was maintained at 2 × 10−10 mbar. The binding energies were cal-
ibrated for the surface charge by referencing to the C1S peak of the
contaminant carbon at 284.6 eV.
For comparison, a chlorine free catalyst (denoted as Pt(N)/ZrO2)
was also prepared. The ZrO2 was prepared using ZrO(NO3)2 as the
precursor, and the synthesis process was the same as that for the
chlorine-containing ZrO2. Then an aqueous solution of Pt(NO3)2
was impregnated to the support. The nominal Pt content was 3 wt.%.
2.2. Catalyst characterizations
All the catalysts were reduced in H2 at 500 ◦C for 1 h prior to
characterizations as well as prior to the reactions.
X-ray diffraction (XRD) patterns were recorded with a PANalytic
X’Pert PW3040 diffractiometer using Cu K␣ radiation. The working
voltage was 40 kV and the working current was 40 mA. The patterns
were collected in a 2ꢀ range from 20◦ to 110◦, with scanning steps
of 0.15◦ s−1
.
Elemental compositions of the reduced catalysts were deter-
mined by X-ray fluorescence (XRF) analysis, with an ARL ADVANT’X
Intelli Power 4200 scanning X-ray fluorescence spectrometer. The
results were analyzed using UniQuant non-standard sample quan-
titative analysis software. The sodium content was determined by
inductively coupled plasma atomic emission spectrometry (ICP-
AES) technique. The results are listed in Table 1.
CO-chemisorption experiments were carried out on a Quan-
tachrame CHEMBET-3000 instrument. The sample was placed in a
quartz U-tube, and purified He was used as the carrier gas. The sam-
ple was reduced in a H2–N2 mixture (5 vol% H2) stream at 500 ◦C for
1 h and cooled down to 30 ◦C in a pure He flow. Then pulses of CO
were fed into the stream of carrier gas with a precision analytical
2.3. Catalytic test
The gas phase crotonaldehyde hydrogenation was carried out
in a fixed bed reaction system at atmospheric pressure, using a
quartz tube (10 mm i.d.) reactor. One hundred mg of catalyst was
loaded in the reactor with a thermocouple placed in the middle of

136
X.-X. Wang et al. / Applied Catalysis A: General 388 (2010) 134–140
Table 2
Dispersion and particle size of Pt in Pt/NaCl/ZrO2 catalysts.
Catalyst
Dispersion (%)
Particle size (nm)
Pt/ZrO2
50
43
37
31
19
2.3
2.8
3.1
3.7
6.0
Pt/NaCl(0.5)/ZrO2
Pt/NaCl(1)/ZrO2
Pt/NaCl(3)/ZrO2
Pt/NaCl(6)/ZrO2
3.2. Surface acidity of the Pt/NaCl/ZrO2 catalysts
Fig. 2 shows the NH3-TPD profiles and the amount of acid of
the Pt/NaCl/ZrO2 catalysts. The NH3-TPD results (Fig. 2) indicate
that the amount of acid sites decreased with NaCl content, which
suggests that the acidic sites decrease with NaCl content. Consid-
ering that NaCl is not a basic compound, it is more likely that NaCl
blocked the acidic sites of the ZrO2, instead of neutralizing the acidic
sites. Meanwhile, we found that the initial temperature of NH3 des-
orption decreases, from 196 ◦C for the Pt/ZrO2 catalyst down to
179 ◦C for the Pt/NaCl(6)/ZrO2. At the same time, the NH3 desorp-
tion peak temperatures shift from 304 and 337 ◦C for the Pt/ZrO2
catalyst to 292 and 322 ◦C for the Pt/NaCl(6)/ZrO2, respectively. The
results here indicate that the addition of NaCl has weakened the
acid strength of the catalysts.
Fig. 1. XRD pattern of Pt/NaCl/ZrO2 catalysts.
In order to determine the nature of acidic sites, we conducted
pyridine adsorption experiments. Fig. 3 shows the FTIR spectra
of pyridine adsorption of the Pt/NaCl/ZrO2 catalysts. The band at
1445 cm−1 is characteristic of Lewis acid sites, and the broad band
near 1629 cm−1 is assigned either to Lewis acid centers or to the
vibrations of some remaining water molecules [15]. The band at
1542 cm−1 attributed to Brønsted acid sites [16] is very weak. The
ratio of Lewis acid sites to Brønsted acid sites (L/B) can be calcu-
lated by comparing the peak area of the band at 1445 cm−1 and
that of the band at 1542 cm−1. The ratio is about 5/1, indicating
that the surfaces of the catalyst are mainly occupied by Lewis acid
sites.
the catalyst bed to monitor the reaction temperature. Before the
catalytic test was run, the catalyst was reduced at 500 ◦C for 1 h in
ultra-pure H2 (26 ml/min) and then it was cooled down to 50 ◦C. The
crotonaldehyde was introduced in a trap set before the reactor tube
and was maintained at 0 ◦C to achieve a constant crotonaldehyde
partial pressure (1062 Pa). The gas line was kept at about 60 ◦C to
avoid any condensation.
The reaction products and reactant were analyzed on line using
a gas chromatography (Shimazu GC-2014) equipped with a flame
ionization detector (FID) and a DB-Wax column (0.32 mm × 30 m)
capillary column.
3.3. In situ DRIFTS of CO adsorption on Pt/NaCl/ZrO2 catalysts
3. Results and discussion
Fig. 4 shows the in situ DRIFTS spectra of CO adsorption on
Pt/NaCl/ZrO2 catalysts. It can be seen that the areas of IR peaks
decrease with increasing of NaCl content, that is because the parti-
3.1. Phase and composition of Pt/NaCl/ZrO2 catalysts
Table 1 lists the contents of Pt, Cl and Na elements analyzed by
XRF and ICP-AES in the reduced catalysts. Note that the Na content
was measured by ICP-AES in order to avoid any disturbance of ZrO2
due to using the XRF technique. It is found that the Pt content in all
catalysts is about 2.8%, which is close to the nominal Pt loading (3%).
Also, Cl is present in the catalysts even after H2 reduction, which
is due to the fact that ZrO2 support was prepared from ZrOCl2 and
H2PtCl6 precursors. It is also found that the contents of Na and Cl
increase with the increasing of NaCl contents.
Fig. 1 shows the XRD patterns of the Pt/NaCl/ZrO2 catalysts.
It can be seen that the diffraction peaks of the ZrO2 support are
very weak, indicating that the support is amorphous. No diffrac-
tion peaks due to Pt or PtOx are observed when the NaCl content
is lower than 3%, indicating the presence of very small Pt particles
in the catalysts. However, for the Pt/NaCl(6)/ZrO2 catalyst, weak
diffraction peaks assigned to metal Pt are observed at 2ꢀ = 39.9◦
(1 1 1), 46.7◦ (2 0 0), 68.1◦ (2 2 0) and 81.8◦ (3 1 1). This suggests that
the Pt particles grow larger with the increasing of NaCl contents.
Table 2 lists the Pt dispersion and particle size in Pt/NaCl/ZrO2
catalysts, based on the CO-chemisorption analysis. The dispersion
ofPtparticles decreaseswithincreasingNaCl content, andtherefore
the average particle size of Pt grows from 2.3 to 6.0 nm. This is in
agreement with the XRD results, further suggesting the growth of
Pt particles with the addition of NaCl.
Fig. 2. NH3-TPD profiles of Pt/NaCl/ZrO2 catalysts.

X.-X. Wang et al. / Applied Catalysis A: General 388 (2010) 134–140
137
Fig. 3. FTIR spectra of pyridine adsorbed on Pt/NaCl/ZrO2 catalysts.
Fig. 5. XPS Pt 4f spectra of reduced Pt catalysts.
cle size of Pt increases, which leads to the reduction of the number
of Pt particles [17]. For the Pt/ZrO2 catalyst without NaCl, the spec-
trum showed a band centered at 2065 cm−1, which is assigned to
CO linearly adsorbed on Pt particles [18]. However, no bridge CO
species (∼1855 cm−1) [19] are observed for any of the catalysts, sug-
gesting that the CO adsorption on the catalysts is in linear mode.
The adsorptions of linear CO bands shift to lower wavenumbers
with the increasing of NaCl contents. The variation of wavenum-
bers for CO adsorption band has been extensively studied [19,20].
For instance, Rades et al. [19] concluded that the reason for the
shift to a lower frequency in Pt/NaY catalysts was due to nega-
tive charging of supported metal particles, which resulted from the
transfer of electrons from the basic oxygen anions of the carrier to
the metal. Panagiotopoulou and Kondarides [20] studied Pt/TiO2
catalysts modified with Na and Cs; they found that Pt particles on
Pt/TiO2 modified with alkali metals had higher electron density val-
ues than these on Pt/TiO2. From the above-mentioned findings, it
can be concluded that the addition of NaCl in the Pt/ZrO2 catalysts
leads to an electronic modification of the metal phase. The addi-
tion of NaCl brings an increase in the electron density of Pt and is
considered to be the reason for the shift [20].
3.4. XPS analysis of Pt/NaCl(x)/ZrO2 catalysts
Fig. 5 shows the XPS spectra of Pt 4f for the Pt/ZrO2 and
Pt/NaCl(3)/ZrO2 catalysts after in situ reduction treatment. For all
the catalysts, the spectra show two broad bands, corresponding to
the Pt 4f7/2 (at low binding energies) and Pt 4f5/2 levels (at high
binding energies). For the Pt/ZrO2 catalyst, the binding energies
of Pt 4f5/2 and Pt 4f7/2 are 74.7 and 71.4 eV, respectively; while
for the Pt/NaCl(3)/ZrO2 catalyst, these values are 74.4 and 71.1 eV.
These results clearly show that the Pt species in the reduced cat-
alysts are metallic Pt [21]. However, compared with the Pt/ZrO2
catalyst, a shift of about 0.3 eV toward lower binding energies for
the Pt/NaCl(3)/ZrO2 catalyst is observed, which can be attributed
to an increase in the electron density of Pt due to the addition of
NaCl [22]. This is in agreement with the in situ DRIFTS spectra of
CO adsorption (Fig. 4), further demonstrating that the addition of
NaCl enhances the electronegativity of Pt.
3.5. Reactivities of Pt/NaCl/ZrO2 catalysts
Fig. 6 shows crotonaldehyde conversion and selectivities to
hydrogenation products as a function of reaction time on the
Pt/NaCl(1)/ZrO2 catalyst for selective hydrogenation of crotonalde-
hyde. The products are crotyl alcohol (formed by hydrogenation of
the carbonyl C O bond), butanal (formed by hydrogenation of the
C
C bond), butanol (formed by hydrogenation of the C O and C
C
bands) and hydrocarbons (formed by polymerization).
Fig. 6 clearly shows that the catalyst gives a conversion of 20%
during the first 30 min of reaction, but it is deactivated and reaches
a quasi-steady state after 1.5 h. Such deactivation is commonly
observed for this reaction, and it has been explained as due to
polymerization and decarbonylation reactions that lead to a loss
of active sites [23]. In the quasi-steady state regime of the catalyst,
a conversion of 7% is obtained after 5 h of reaction. The selectivity to
crotyl alcohol increases quickly during the first 30 min and reaches
a stable level (about 60%) afterwards.
In order to explain the loss of the activity of the catalyst with the
reaction time, we performed an FTIR examination of used catalyst.
Fig. 7 shows the FTIR spectra of the Pt/NaCl(1)/ZrO2 catalyst. Com-
Fig. 4. In situ DRIFTS spectra of CO adsorbed on Pt/NaCl/ZrO2 catalysts.

138
X.-X. Wang et al. / Applied Catalysis A: General 388 (2010) 134–140
Scheme 1. Proposed adsorption model of crotonaldehyde on Pt/NaCl/ZrO2 catalysts.
at the NaCl content of 1% (60%) and decreases with the further
increasing of NaCl contents. In contrast, the selectivity to butanal
first decreases but then increases when the NaCl content is higher
than 1%. In addition, the TOFs of the catalysts based on particle
size of Pt show a more pronounced enhancement with the increas-
ing of NaCl contents, from 1.6 to 11.1 × 10−3 s−1. Adding excess
NaCl is disadvantageous to C O hydrogenation. The reason is that
the reaction rate of C O hydrogenation continued to increase with
NaCl content; however, the C C hydrogenation rate increased more
intensively, so the selectivity to crotyl alcohol decreased.
Fig. 6. Crotonaldehyde conversion and selectivities to hydrogenation products as
functions of time on stream for Pt/NaCl(1)/ZrO2 catalyst.
3.6. Active sites of Pt/NaCl(1)/ZrO2 catalysts
pared with the fresh catalyst, it is found that the adsorption bands
are very different for the used one. Used catalyst exhibits bands at
1171, 1082, 799, 788, 674 cm−1. The adsorption bands at 1171 and
1082 cm−1 are assigned to ꢁC–O [24,25], while the bands at 799, 788,
674 cm−1 are assigned to ıC–H. This result suggests that during the
reaction, organic compounds are formed on the catalyst surface by
polymerization of reaction products; such formation may lead to a
partial loss of active sites and thus suppress the reactivity.
Table 3 shows the reactivities of the Pt/NaCl/ZrO2 catalysts for
selective hydrogenation of crotonaldehyde. For comparison, we
give the reactivities of the catalysts to which other additives such
as KOH, KCl, NaOH and Na2CO3 are added. It can be seen that the
catalytic activities of Pt/NaOH(3)/ZrO2 and Pt/Na2CO3(3)/ZrO2 are
similar to that of Pt/NaCl(3)/ZrO2 catalyst. Considering that many
works pointed out that the presence of Cl has great influence on
the reaction [10], NaCl was selected as an additive to be inves-
tigated in detail. From Table 3, one can see that the conversion
of the catalyst increases with increasing NaCl content, from 6% to
14%. However, the selectivity to crotyl alcohol reaches a maximum
The activity of the supported Pt catalysts for selective hydro-
genation of crotonaldehyde is affected by various factors, such
as the surface acidity and electronic property of the Pt metal, as
described in Section 1. In this work, the NH3-TPD (Fig. 2) and
the pyridine adsorption (Fig. 3) results indicate that the support
surface contains mainly Lewis acid sites, and that the strength
of the Lewis acid sites weakens with the increasing of NaCl con-
tents. The CO-chemisorption (Fig. 4) and XPS (Fig. 5) results suggest
that the electron density of Pt increases with the increasing of
NaCl contents. According to the adsorption model proposed by
Englisch et al. [14] and based on the co-existence of the surface
Lewis acid sites and the Pt active sites in this system, a proposed
model for crotonaldehyde adsorption on Pt/NaCl/ZrO2 catalyst is
shown in Scheme 1. The carbonyl oxygen is suggested to strongly
interact with Lewis acid sites (electron pair donor sites) and the
Pt particle is the adsorption center of C C band and carbonyl
carbon.
In order to further explain the influence of Pt active sites and
the Lewis acid sites on selective hydrogenation of crotonalde-
hyde, we studied in situ DRIFTS of crotonaldehyde adsorption, as
shown in Fig. 8. The bands due to free crotonaldehyde molecules
(ꢁC at 1694 cm−1, ꢁC at 1639 cm−1, ıasCH3 at 1445 cm−1, ıCH3
O
C
at 1395 cm−1, ıCH at 1376 cm−1) can be clearly identified [26,27].
However, when the crotonaldehyde molecules adsorb on the
NaCl(1)/ZrO2 support and on the Pt/NaCl(1)/ZrO2 catalyst, the
bands undergo remarkable shifts. The band at 1677 cm−1 is close
to the main band of crotonaldehyde molecules, and its inten-
sity becomes weaker with purging He. Therefore, the band at
1677 cm−1 is attributed to ꢁC O of physisorbed crotonaldehyde. The
band at 1586 cm−1 for the support and the band at 1576 cm−1 for
the catalyst are attributed to ꢁC of chemisorbed of crotonalde-
O
hyde. The relevant down shift of the crotonaldehyde ꢁC
(from
O
1694 to 1586 cm−1) indicates that crotonaldehyde molecules par-
tially adsorb on Lewis acid sites through carbonyl oxygen, which
is consistent with the previous study [27]. Moreover, due to the
stronger surface acidity of the NaCl(1)/ZrO2 support compared to
that of the TiO2 in Ref. [27], the shift is more pronounced. For the
Pt/NaCl(1)/ZrO2 catalyst, the band of ꢁC O is observed at 1576 cm−1
,
which is 10 cm−1 lower than that on the support. This indicates that
the interaction of carbonyl group with the catalyst is even stronger
than that with the support. Moreover, this result further validates
Fig. 7. FTIR spectra of Pt/NaCl(1)/ZrO2 (A: fresh, B: reacted).

X.-X. Wang et al. / Applied Catalysis A: General 388 (2010) 134–140
139
Table 3
Effect of NaCl content on reactivities of Pt/NaCl/ZrO2 catalysts.
Catalysta
Conversion (%)
TOFb (10−3 s−1
)
Selectivity (%)
Crotyl alcohol
Butanal
Pt/ZrO2
5.9
4.8
7.2
1.6
1.6
2.6
4.1
11.1
–
–
–
–
31
41
60
50
46
53
43
52
47
35
28
21
24
35
27
28
25
30
Pt/NaCl(0.5)/ZrO2
Pt/NaCl(1)/ZrO2
Pt/NaCl(3)/ZrO2
Pt/NaCl(6)/ZrO2
Pt/NaOH(3)/ZrO2
Pt/KOH(3)/ZrO2
Pt/Na2CO3(3)/ZrO2
Pt/KCl(3)/ZrO2
9.1
14.6
13.5
6.4
10.2
6.2
a
Data obtained at 5 h of time on stream.
Turnover frequency (TOF) in hydrogenation is calculated on the basis of conversion and the amount of CO adsorption at room temperature.
b
the proposed adsorption model, that is, carbonyl carbon adsorbs
on the Pt sites and carbonyl oxygen adsorbs on the Lewis acid
sites.
higher than that for the Pt(N)/ZrO2, indicating that the acid strength
of Pt/ZrO2 is stronger than that of the Pt(N)/ZrO2. This result tells us
that the Lewis acid sites are mainly generated by chlorine species in
the catalyst. Fig. 9 also presents the conversion of crotonaldehyde
on the two catalysts (obtained at 5 h reaction). We found that the
conversion of crotonaldehyde on the Pt/ZrO2 (6%) is lower than that
on the Pt(N)/ZrO2 (11%). This result is consistent with the hypoth-
esis mentioned above, that is, a Lewis acid with too great strength
is disadvantageous to the catalytic activity [28]. Therefore, for the
Pt/NaCl/ZrO2 catalysts, the addition of NaCl leads to the weakened
strength of Lewis acid (Fig. 2), which could be one reason of the
improvement of the reactivity.
Considering the selectivity to the desired product crotyl alcohol,
we found that when NaCl content is 1%, the selectivity to crotyl alco-
hol is the highest, while that for butanal is the lowest. The factors
that influence the selectivity have been extensively discussed in the
literature. Ammari et al. [10] concluded that the Pt catalysts pre-
pared from platinum chloride precursor with strong Lewis acidity
have high selectivities toward ␣,␤-unsaturated alcohol. Englisch
et al. [14] reported that the terminal oxygen of crotonaldehyde
strongly interacted with the Lewis acid sites and this interaction
led to high selectivity to crotyl alcohol. Therefore, Lewis acid sites
are considered to be one of the key factors leading to high crotyl
alcohol selectivity. In this study, the strength of Lewis acid becomes
weaker with the increasing of NaCl contents (Fig. 2), which could
have led to a decrease for the selectivity to crotyl alcohol. But in fact,
According to the adsorption model of Scheme 1, it can be seen
that the catalytic activity is mainly affected by both the Pt active
sites and the Lewis acid sites. The increase of the electron density on
Pt is beneficial to the adsorption of carbonyl carbon, which leads to
the hydrogenation of C O bond. Furthermore, it was reported that
high acidity can result in high catalytic activity at the initial stage of
reaction, but it also can lead to high coking and easy deactivation
for the reaction [28]. In this work, the strength of Lewis acidity
weakens with the increasing of NaCl in the catalyst, which could
well explain the enhanced reactivity on the NaCl-modified Pt/ZrO2
catalysts in the quasi-steady state.
In order to further explain the influence of the Lewis acidity
on the reactivity, we prepared the chlorine free Pt(N)/ZrO2 cat-
alyst. The particle size of Pt in the Pt(N)/ZrO2 catalyst measured
by CO-chemisorption is 2.0 nm, which is very close to that of the
chlorine-containing Pt/ZrO2 catalyst (Table 2). The NH3-TPD pro-
files of the Pt/ZrO2 and Pt(N)/ZrO2 catalysts are shown in Fig. 9. It is
obvious that the amount of NH3 desorbed on the Pt/ZrO2 catalyst
is much larger than that on the Pt(N)ZrO2, with a relative peak area
ratio of 5. This indicates that the Lewis acidity of the ZrO2 support
prepared by ZrOCl2 precursor can be reinforced by chlorine [29].
The desorption temperature of NH3 for the Pt/ZrO2 catalyst also is
Fig. 8. DRIFTS analysis of crotonaldehyde adsorption over NaCl(1)/ZrO2 support and
Fig. 9. NH3-TPD profiles of Pt(N)/ZrO2 and Pt/ZrO2 catalysts, the inserted profile is
Pt/NaCl(1)/ZrO2 catalyst.
the conversion of crotonaldehyde on these catalysts.

140
X.-X. Wang et al. / Applied Catalysis A: General 388 (2010) 134–140
the selectivity to crotyl alcohol increases and reaches the maximum
when the NaCl content is 1% in the catalyst (Table 3).
addition of NaCl decreases the strength of the Lewis acid sites; such
lower strength is believed to be the main factor for the enhanced
activity. With a low NaCl content (≤1%), the increase in the elec-
tron density on Pt leads to the increase of the selectivity to crotyl
alcohol; however with a high NaCl content (>1%), the growth of Pt
particle size and the weakened strength of Lewis acid lead to the
decrease of the selectivity.
In order to explain the enhancement of selectivity on the NaCl-
modified catalysts, the electronic state of Pt must be taken into
account [30]. When the NaCl content is below 1%, the Pt particle
size does not change much (about 3 nm, Table 1), so the Pt particle
size is not considered to be the major factor that affects the selec-
tivity to crotyl alcohol. The DRIFTS results of CO adsorption (Fig. 4)
and XPS results (Fig. 5) clearly show that with the increasing of
NaCl contents, the electron density of Pt increases; such higher
density can easily attract the electron positive carbonyl carbon.
The strong interaction of the electronegative Pt particles with elec-
tropositive carbonyl carbon could certainly improve the selectivity
to the crotyl alcohol; such improvement probably plays a key role
as it compensates the opposite effect of the weak acidic strength of
the NaCl-containing catalysts on the decline of selectivity.
Acknowledgments
The authors acknowledge Prof. Shishan Sheng from the Dalian
Institute of Chemical Physics for his help on conducting the in situ
XPS experiments. Dr. Xiang Wang is thanked for his help with pol-
ishing the latest version of our manuscript.
References
When the NaCl content is higher than 1% in the catalyst, the par-
ticle size of Pt changes significantly and grows from 3.7 to 6.0 nm.
Thus, the influence of Pt particle size on the selectivity must be
taken into account. Previous studies pointed out that large Pt parti-
cles are favorable for the formation of crotyl alcohol [31], however,
the selectivity declines with the increase of Pt particles (Table 3)
in the present work. According to the crotonaldehyde adsorption
model (Scheme 1), the formation of crotyl alcohol takes place on
the interface region, which involves both Pt and Lewis acid sites.
The growth of Pt particle results in a decline in the amount of inter-
facial sites of Pt contacting with Lewis acid sites, while the sites
for the C C bond adsorption increase. Therefore, the competing
adsorption of C C bond on large Pt particles suppresses the selec-
tivity to crotyl alcohol. Moreover, the weakened Lewis acidity on
[1] H. Surburg, J. Panten, Common Fragrance and Flavour Materials: Preparation
Properties and Uses, 5th ed., Wiley-VCH, New York, 2006, pp. 27–49.
[2] P. Gallezot, D. Richard, Catal. Rev. Sci. Eng. 40 (1998) 81–126.
[3] A. Corma, L.T. Nemeth, M. Renz, S. Valencia, Nature 412 (2001) 423–425.
[4] A.L.
5454–5459.
[5] J. Jenck, J.E. Germain, J. Catal. 65 (1980) 133–140.
Gemal,
J.L.
Luche,
J.
Am.
Chem.
Soc.
103
(1981)
[6] A. Giroir-Fendler, D. Richard, P. Gallezot, Stud. Surf. Sci. Catal. 41 (1988)
171–178.
[7] P. Beccat, J.C. Bertolini, Y. Gauthier, J. Massardier, P. Ruiz, J. Catal. 126 (1990)
451–456.
[8] D.I. Jerdev, A. Olivas, B.E. Koel, J. Catal. 205 (2002) 278–288.
[9] J. Silvestre-Albero, F. Rodriguez-Reinoso, A. Sepulveda-Escribano, J. Catal. 210
(2002) 127–136.
[10] F. Ammari, J. Lamotte, R. Touroude, J. Catal. 221 (2004) 32–42.
[11] B.A. Riguetto, C.E.C. Rodrigues, M.A. Morales, E. Baggio-Saitovitch, L. Gengem-
bre, E. Payen, C.M.P. Marques, J.M.C. Bueno, Appl. Catal. A: Gen. 318 (2007)
70–78.
[12] M. Abid, V. Paul-Boncour, R. Touroude, Appl. Catal. A: Gen. 297 (2006) 48–59.
[13] A.J. Plomp, H. Vuori, A.O.I. Krause, K.P. de Jong, J.H. Bitter, Appl. Catal. A: Gen.
351 (2008) 9–15.
[14] M. Englisch, A. Jentys, J.A. Lercher, J. Catal. 166 (1997) 25–35.
[15] L. Jankovicˇ, P. Komadel, J. Catal. 218 (2003) 227–233.
[16] G.V. Shanbhag, T. Joseph, S.B. Halligudi, J. Catal. 250 (2007) 274–282.
[17] Y.Q. Huang, A.Q. Wang, L. Li, X.D. Wang, D.S. Su, T. Zhang, J. Catal. 255 (2008)
144.
[18] A. Bourane, O. Dulaurent, D. Bianchi, J. Catal. 196 (2000) 115–125.
[19] T. Rades, V.Y. Borovkov, V.B. Kazansky, M. Polisset-Thfoin, J. Fraissard, J. Phys.
Chem. 100 (1996) 16238–16241.
[20] P. Panagiotopoulou, D.I. Kondarides, J. Catal. 260 (2008) 141–149.
[21] A. Huidobro, A. Sepulveda-Escribano, F. Rodriguez-Reinoso, J. Catal. 212 (2002)
94–103.
the high NaCl content catalyst may suppress the adsorption of C
O
bond, which would result in the lowered selectivity to crotyl alco-
hol. This is consistent with the reaction results (Table 3), as the
selectivity to butanal increases when the NaCl content is higher
than 1%, accompanied by the loss of selectivity to crotyl alcohol.
In summary, the selectivity to the desired product crotyl alcohol
is influenced by several factors: the strength of Lewis acid sites, the
electronegativity of Pt species and the Pt particle size. The weak
strength of Lewis acid sites will suppress the selectivity, while
highly electronegative Pt species are beneficial to the enhance-
ment of the selectivity. Moreover, when the Pt particles grow larger,
the interfacial sites for C O adsorption decline while the adsorp-
tion of C C bond becomes more pronounced, thus suppressing the
selectivity to crotyl alcohol.
[22] S¸ . Özkara-Aydınog˘lu, E. Özensoy, A.E. Aksoylu, Int. J. Hydrogen Energy 34 (2009)
9711–9722.
[23] E.V. Ramos-Fernández, A. Sepúlveda-Escribano, F. Rodríguez-Reinoso, Catal.
Commun. 9 (2008) 1243–1246.
[24] L.V. Mattos, F.B. Noronha, J. Catal. 233 (2005) 453–463.
[25] A. Yee, S.J. Morrison, H. Idriss, J. Catal. 186 (1999) 279–295.
[26] J. Raskó, J. Kiss, Appl. Catal. A: Gen. 287 (2005) 252–260.
[27] J. Raskó, T. Kecskés, J. Kiss, Appl. Catal. A: Gen. 287 (2005) 244–251.
[28] V.N. Kalevaru, A. Benhmid, J. Radnik, M.M. Pohl, U. Bentrup, A. Martin, J. Catal.
246 (2007) 399–412.
[29] F. Ammari, C. Milone, R. Touroude, J. Catal. 235 (2005) 1–9.
[30] Q.Y. Yang, Y. Zhu, L. Tian, S.H. Xie, Y. Pei, H. Li, H.X. Li, Mi.H. Qiao, K.N. Fan, Appl.
Catal. A: Gen. 369 (2009) 67–76.
4. Conclusions
In this work, we demonstrated that the selective hydrogena-
tion of crotonaldehyde can be enhanced by the modification of
the Pt/ZrO2 catalyst. By adding NaCl to Pt/ZrO2 catalysts, one can
remarkably enhance the reactivity of the catalyst and the selec-
tivity to crotyl alcohol. The characterization results show that the
[31] M.E. Grass, R.M. Rioux, G.A. Somorjai, Catal. Lett. 128 (2009) 1–8.