Noble metal promoted CeO2-ZrO2-supported Ni catalysts for liquid-phase hydrogenation of cinnamaldehyde

Article abstract of DOI:10.1007/s10562-013-1064-9

Pd or Pt promoted CeO₂-ZrO₂-supported Ni catalysts exhibited superior catalytic activity to the hitherto known Ni catalysts for liquid-phase hydrogenation of cinnamaldehyde at moderate conditions. Under similar experimental conditions, the unpromoted catalyst was selective for hydrocinnamaldehyde product (C=C hydrogenation) whereas the promoted catalyst yielded 3-phenyl propanol (C=C and C=O hydrogenation product). Enhanced dispersion of Ni was the cause for higher activity of the promoted Ni catalysts.

Full text of DOI:10.1007/s10562-013-1064-9

Catal Lett
DOI 10.1007/s10562-013-1064-9
Noble Metal Promoted CeO2–ZrO2-Supported Ni Catalysts
for Liquid-Phase Hydrogenation of Cinnamaldehyde
•
S. Bhogeswararao V. Pavan Kumar
•
•
K. V. R. Chary D. Srinivas
Received: 18 April 2013 / Accepted: 5 July 2013
Ó Springer Science+Business Media New York 2013
Abstract Pd or Pt promoted CeO2–ZrO2-supported Ni
catalysts exhibited superior catalytic activity to the hitherto
known Ni catalysts for liquid-phase hydrogenation of cin-
namaldehyde at moderate conditions. Under similar
experimental conditions, the unpromoted catalyst was
selective for hydrocinnamaldehyde product (C=C hydro-
genation) whereas the promoted catalyst yielded 3-phenyl
propanol (C=C and C=O hydrogenation product).
Enhanced dispersion of Ni was the cause for higher activity
of the promoted Ni catalysts.
flavours [4, 5]. Hydrocinnamaldehyde was found as an
important intermediate in the preparation of a drug used in
the treatment of HIV [4, 5]. Supported Ni catalysts are
selective for C=C hydrogenation in CA. Although, hydro-
genation of C=C compared to C=O is easy, development of
still more efficient catalysts that operate at moderate con-
ditions is desirable. We report here, for the first time, the
catalytic application CeO2–ZrO2-supported Ni for the
liquid-phase hydrogenation of CA. The influence of noble
metal (Pd and Pt) and alkali (NaOH) promoters on catalytic
activity and product selectivity of the catalysts is investi-
gated. Enhancement in hydrogenation activity of some
catalysts when a small amount of noble metal was added
(Cu–Pt, Co–Pt, Ru–Pt, Fe–Pt and Sn–Pt) has already been
reported [6–11]. The present study confirms such an
enhancement even in the case of supported Ni catalysts.
The objective of this study is to establish the effect of
promoters on the activity and selectivity of Ni in liquid
phase hydrogenation of CA. Detailed characterization
studies revealed that increased metal dispersion (MD) is
the cause for the enhanced activity of the promoted cata-
lysts. The catalysts of this work are more active than the
hitherto known Ni catalysts for C=C hydrogenation.
Keywords Selective hydrogenation ꢀ Cinnamaldehyde ꢀ
Supported Ni catalyst ꢀ Ceria–zirconia ꢀ Noble metal
promoted catalyst
1 Introduction
Selective hydrogenation of a,b-unsaturated aldehydes is an
organic transformation of fundamental and industrial
importance (Scheme 1) [1–3]. In such compounds, hydro-
genation occurs at both C=C and C=O functional groups.
While hydrogenation of the olefinic group leads to satu-
rated aldehydes, the carbonyl group hydrogenation yields
unsaturated alcohols. Both these hydrogenation products of
cinnamaldehyde (CA)—a representative a,b-unsaturated
aldehyde are widely used in pharmaceuticals, perfumes and
2 Experimental
2.1 Catalyst Preparation
2.1.1 Ni(5 wt%)/CeO2–ZrO2
S. Bhogeswararao ꢀ D. Srinivas (&)
Catalysis Division, CSIR-National Chemical Laboratory,
Pune 411 008, India
e-mail: d.srinivas@ncl.res.in
The supported Ni catalysts were prepared by impregnation
method. 0.2597 g of Ni(NO3)2ꢀ6H2O (Thomas Baker) dis-
solved in 10 ml of deionized water was added to 1 g of
CeO2–ZrO2 prepared as reported by us earlier [12, 13]. It was
V. Pavan Kumar ꢀ K. V. R. Chary
Inorganic and Physical Chemistry Division, CSIR-Indian
Institute of Chemical Technology, Hyderabad 500 607, India
123

S. Bhogeswararao et al.
Scheme 1 Hydrogenation
products of cinnamaldehyde
O
Hydrocinnamal-
dehyde(HCA)
OH
O
OH
3-Phenylpropanol(PPL)
Cinnamaldehyde(CA)
Cinnamylalcohol(CAL)
mixed thoroughly. Solvent was removed using a rotary
evaporator. The solid obtained was recovered and dried at
120 °C for 6 h in an oven. Then, the material was calcined at
500 °C for 2 h. Prior to use, the material was reduced in a
flow of H2 (30 ml/min) at 400 °C for 2.5 h. Supported metal
catalysts with Ni loading of 3, 10 and 15 wt% were prepared
in the same manner taking appropriate amounts of the nickel
precursor. Also CeO2- and ZrO2-supported Ni catalysts were
prepared by the same method.
Specific surface area (SBET) of the samples was determined
using a NOVA 1200 Quanta Chrome instrument. Ni dispersion
was examined using high-resolution transmission electron
microscope (HRTEM, JEOL, model 1200 EX; operating at
100 kV) and H2 chemisorption (Auto Chem 2910, Microm-
eritics, USA) techniques. X-ray photoelectron spectra (XPS)
were acquired using a VG Microtech Multilab ESCA 3000
instrument; Mg Ka radiation (hm = 1253.6 eV) was
employed.
Specific surface area, dispersion and average particle
size of the metal were determined from hydrogen chemi-
sorption measurements. Prior to chemisorption, about 0.1 g
of the catalyst sample was reduced in a flow of hydrogen
(50 ml/min) at 300 °C for 3 h and flushed with pure He gas
for an hour at 300 °C. Then, the sample was cooled to
ambient temperature in the same He stream. Hydrogen
uptake was determined by injecting pulses of 9.96 % H2
(balance He) from a calibrated on-line sampling valve into
He stream passing over the reduced samples at 30 °C.
Nickel surface area, percentage dispersion and average
particle size were calculated assuming a stoichiometric
factor (Sf = number of moles of hydrogen per surface
metal) of 0.5. In the average particle size determination, the
particle shape factor (F) was considered as six assuming
spherical or near spherical shaped particles like cubo-
octahedral crystallites or a collection of cubic particles.
The following equations [14, 15] were used in determining
the metal specific surface area (MSS; m2/g), percentage
MD and average metal particle size (Da; in units of
nanometers).
2.1.2 Ni(5 wt%)–M(0.5 wt%)/CeO2–ZrO2, M = Pt or Pd
These catalysts were prepared by adding 0.005 g of
Pt(NH3)4(NO3)2 (Aldrich Co.) or 0.0704 g of Pd(NH3)4(NO3)2
(10 wt% aqueous solution; Aldrich Co.) to CeO2–ZrO2 along
with the nickel source. Then, it was treated and reduced in the
same way as detailed above.
2.1.3 M(0.5 wt%)/CeO2–ZrO2, M = Pt or Pd
0.005 g of Pt(NH3)4(NO3)2 or 0.0704 g of Pd(NH3)4
(NO3)2 (10 wt% aqueous solution) were taken in beaker
containing 10 ml of deionized water. This solution was
added to 1 g of CeO2–ZrO2. It was thoroughly mixed.
Water was removed over a rotary evaporator. The solid
obtained was recovered, dried at 120 °C for 6 h and cal-
cined at 500 °C for 2 h. Prior to use, the material was
reduced in a flow of H2 (30 ml/min) at 350 °C (for Pt) or
250 °C (for Pd) for 2.5 h.
MSS ¼ Vm ꢀ Na=Sf ꢀ Sd ꢀ L
MD ¼ Vm ꢀ AW ꢀ 104=%W ꢀ Sf
Da ¼ 103 ꢀ F=MSS ꢀ Dm
Here, Vm is hydrogen gas adsorbed at monolayer
coverage (moles/g-catalyst), Na is Avagadro number, Sd
is metal surface density i.e., number of metal atoms per
square meter of Ni which is 0.154 9 1020, AW is atomic
weight of nickel (58.693), L is grams of metal per gram of
ð1Þ
ð2Þ
ð3Þ
2.2 Characterization Techniques
X-ray diffraction (XRD) profiles of the reduced catalyst sam-
pleswere recorded in the 2hrangeof 10–90°(scanrate = 3.3°/
min) using an Xpert Pro PANalytical diffractometer equipped
with a Ni-filtered Cu-Ka radiation source (40 kV, 30 mA).
Nickel content in the catalyst was estimated by atomic
absorption spectroscopy (AAS; Varian Spectra AA220FS).
123

Selective Hydrogenation of Cinnamaldehyde
catalyst, %W is weight percent of metal in the catalyst
(which is equal to 100L) and Dm is metal density i.e.,
grams per metal unit volume which is 8.9 for nickel.
estimation of Ni crystallite size. Dotted traces in Fig. 1 show
the blown-up XRD patterns. The crystallite size of Ni on
CeO2–ZrO2 was found larger (24.6 nm) than that on ZrO2
(17.8 nm) and CeO2 (3.7 nm; Table 1). This variation of Ni
crystallite size agrees well with the available metal content on
the surface of support and enhanced MD property of CeO2.
The crystallite sizes of Ni for 5, 10 and 15 wt% Ni/ZrO2 were
estimated to be 17.8, 28.0 and 25.6 nm, respectively. The size
of metal crystals on Ni(3 wt%)/ZrO2 was below the X-ray
detection limit. Co-presence of Pt and Pd decreased the mean
crystallite sizeofNifrom24.5[for Ni(5 wt%)/CeO2–ZrO2] to
21.5 nm [for Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2] and
10.7 nm [for Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2]. It may
be noted that the Ni contents (nominal value—5 wt% and
actual value determined by AAS—4.5, 4.3 and 4.4 wt%;
Table 1) of these three compositions—Ni(5 wt%)/CeO2–
ZrO2, Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2 and Ni(5 wt%)–
Pd(0.5 wt%)/CeO2–ZrO2, respectively, are nearly the same.
TheactualvaluesofPtorPdcontentsestimatedbyinductively
coupled plasma-atomic emission spectroscopy (Spectro Ar-
cos ICP-OES) for Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2,
Pt(0.5 wt%)/CeO2–ZrO2, Ni(5 wt%)–Pd(0.5 wt%)/CeO2–
ZrO2 and Pd(0.5 wt%)/CeO2–ZrO2 were found to be 0.39,
0.42, 0.35 and 0.38 wt%, respectively. H2-temperature-pro-
grammed reduction experiments showed individual reduction
peaks of Ni and noble metal suggesting that Ni–Pt and Ni–Pd
exist as bimetallic components and not as alloys. Raab and
Lercher [17] reported a similar observation for Pt promoted
dispersion of Ni supported on TiO2 used for gas phase
hydrogenation of crotonaldehyde.
2.3 Reaction Procedure: Hydrogenation
of Cinnamaldehyde
Prior to hydrogenation reactions, the catalyst was reduced
under a flow of hydrogen. The color of the catalyst changed
from dark brown to black. Without exposing to atmo-
sphere, 0.05 g of the reduced catalyst was transferred into a
100 ml stainless-steel Parr micro bench-top reactor (Paar
4875) containing CA (1 g) and iso-propanol (25 ml). The
reactor was initially flushed and then, pressurized with
hydrogen to a desired value (5–30 bar). Temperature of the
reactor was raised to 75–150 °C and the reaction was
conducted while stirring (600 revolutions per minute) for
0.5–8 h. The liquid product was analyzed and quantified by
gas chromatography (GC; Varian 3400; CP-SIL8CB col-
umn; 30 m-long and 0.53 mm-i.d.). The products were
identified by GC–MS (Shimadzu GCMS-QP5050A; HP-5
column; 30 m-long 9 0.25 mm i.d. 9 0.25 lm thickness).
3 Results and Discussion
3.1 Structural and Textural Properties
Catalyst samples reduced in hydrogen atmosphere showed
two characteristic XRD peaks at 44.5° and 52° arising from
(111) and (200) planes of metallic Ni having a face-centered
cubic structure (JCPDS Card No. 04-0850). Intensities of
these peaks increased with increasing Ni content (Fig. 1). In
the case of Ni(5 wt%)/CeO2 and Ni(5 wt%)/ZrO2, a
decrease in unit cell parameter of the support (aSupport) was
noted when Ni was present, revealing that a portion of Ni is
possibly substituted in the lattice sites of CeO2 and ZrO2
{aSupport: 0.542 nm (for ‘‘neat’’ CeO2), 0.512 nm (for
‘‘neat’’ ZrO2), 0.540 nm [for Ni(5 wt%)/CeO2] and
0.508 nm [Ni(5 wt%)/ZrO2]} [16]. No such decrease in the
unit cell parameter (aSupport) was observed for Ni supported
on CeO2–ZrO2 solid solutions suggesting that most of the
Ni in those samples is located on the surface of the support
only. This surface-rich Ni on cubic fluorite structured
CeO2–ZrO2 thereby showed more intense metallic Ni peaks
than Ni/CeO2 and Ni/ZrO2 (Fig. 1).
Upon Ni loading, the specific surface area (SBET) of
CeO2–ZrO2 decreased from 115 to 65 m2/g. While plati-
num had little effect, co-presence of Pd increased the
surface area of Ni/CeO2–ZrO2 from 65 to 77 m2/g
(Table 1). Reason for this increase is however, not clear at
this point of time.
HRTEM images (Fig. 2) revealed that Ni is uniformly
distributed on the support. From the XRD data (Table 1) it
appeared that in some cases, the size of Ni crystallites is
larger than that of the support crystallites. However, it
should be noted that XRD provides crystallite size infor-
mation of only those crystals which are above the X-ray
detection limit. HRTEM images showed that Ni is uni-
formly distributed on the support surface. Hence, it is likely
that the crystallites of Ni which are bigger in size are only a
few in numbers, while a majority of Ni crystals are smaller
than the size of the support crystallites and are finely dis-
persed and located on the support surface.
The XRD peaks corresponding to Ni were much smaller
compared to those of the supports (Fig. 1). However, by using
ORIGIN-8 software and blowing up the XRD patterns, we
managed to estimated the full-width at half height of Ni(111)
reflection, which in turn enabled determination of the average
crystallite size of Ni using Debye–Scherrer formula. As this
signal is weak there is possibility of some error in the
Binding energy (BE) values of Ce (3d5/2), Zr (3d5/2) and Ni
(2p3/2) determined from XPS are listed in Table 1. The BE
values of Ni 2p3/2 for different catalysts decreased in the
order: Ni(5 wt%)/CeO2 (852.5 eV) [Ni(5 wt%)/CeO2–ZrO2
(852.1 eV) [Ni(5 wt%)/ZrO2 (852.0 eV). Noble metal
123

S. Bhogeswararao et al.
Ni(111)
Ni(111)
(a)
(b)
Ni(5 wt%)-Pt/CeO-ZrO2
2
Ni(15 wt%)/ZrO
2
Ni(5 wt%)-Pd/CeO-ZrO2
2
Ni(10 wt%)/ZrO
2
Ni(5 wt%)/CeO -ZrO2
2
Ni(5 wt%)/ZrO
2
Ni(5 wt%)/ZrO
2
Ni(3 wt%)/ZrO
2
Ni(5 wt%)/CeO
2
30
40
50
60
70
80
90
30
40
50
60
70
80
90
2θ (degree)
2θ (degree)
Fig. 1 XRD profiles: a CeO2–ZrO2-supported Ni catalysts. Nominal values of Pd and Pt are 0.5 wt%. b Different loadings of Ni supported on
ZrO2. Dotted curves show the blown-up of the XRD profiles
Table 1 Physicochemical characteristics of CeO2–ZrO2-supported Ni catalysts
Samplea
Ni content
(output, wt%; (XRD)
Structural properties
Specific
surface
area (SBET
m2/g)
Acidity
(mmol/g; NH3- (mmol/g; CO2- values (eV, XPS)
Basicity
Binding energy
AAS)
;
TPD)
TPD)
Unit cell Average
parameter crystallite size
Ce
(3d5/2
Zr
(3d5/2
Ni
(2p3/2)
)
)
(aSupport
nm)
;
(nm)
Support Ni
CeO2–ZrO2
–
0.537
0.537
0.540
0.508
0.533
7.7
7.8
–
115
65
0.180
0.208
0.175
0.302
–
0.220
0.248
0.234
0.255
–
–
–
–
Ni(5 wt%)/CeO2–ZrO2
Ni(5 wt%)/CeO2
Ni(5 wt%)/ZrO2
4.5
4.3
4.4
4.3
24.6
3.7
881.7
881.3
–
181.3
–
852.4
852.1
852.0
851.2
12.9
10.7
8.4
63
17.8
21.5
88
181.8
181.4
Ni(5 wt%)–Pt(0.5 wt%)/
CeO2–ZrO2
67
881.1
Ni(5 wt%)-Pd(0.5 wt%)/ 4.4
CeO2–ZrO2
0.540
13.5
10.7
77
–
–
880.9
181.2
851.4
a
Input values of metals used in synthesis are denoted in parentheses
Fig. 2 HRTEM images of a Ni(5 wt%)/CeO2, b Ni(5 wt%)/ZrO2 and c Ni(5 wt%)/CeO2–ZrO2
123

Selective Hydrogenation of Cinnamaldehyde
incorporation lowered the BE value still further [Ni(5 wt%)/
CeO2–ZrO2 (852.1 eV) [ Ni(5 wt%)–Pd(0.5 wt%)/CeO2–
ZrO2 (851.5 eV) [ Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2
?4 oxidations states with their compositions being 65.8,
15.8 and 18.4 %, respectively. The 4f7/2 peak of Pt0 was
observed at 70.5 eV. The main line of Pd2? and Pd4?
appeared at 71.9 and 75.6 eV, respectively. Pd was present
in zero and ?2 oxidation states with composition being
88.2 and 11.8 %, respectively. The main line for metallic
Pd (3d5/2) appeared at 335.5 eV and of Pd2? (as PdO)
appeared at 337.0 eV. The BE values of Pt0 and Pd0 are
well within the expected range [12, 13, 22].
¨
¨
(851.4 eV)]. Croy et al. [18] and Domok et al. [19] reported
that BE values of photoelectrons ejected from metal particles
depend on the metal particle size. The smaller the particle,
the higher would be the binding energy. Hence, the average
particlesizeofNi ondifferent supportsisexpectedtodecrease
in the order: Ni(5 wt%)/ZrO2 [ Ni(5 wt%)/CeO2–ZrO2 [
Ni(5 wt%)/CeO2. Pt and Pd enhanced the dispersion of Ni
in Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2 and Ni(5 wt%)–
Pd(0.5 wt%)/CeO2–ZrO2, respectively. This agrees well with
the XRD and TEM results. Ni on ZrO2 support is more
metallic in nature (lower BE value) than that on CeO2 and
CeO2–ZrO2. Noble metals (especially Pd) enhanced the
metallic nature of Ni. The observed shift in BE values is also
because of weaker metal-support interactions in the case of
ZrO2 than in CeO2–ZrO2 and CeO2. Shifts in energy can be
explained in terms of charge transfer tothe metal particlefrom
the support due to delocalized electron distributions arising
from oxygen vacancies [20, 21]. Shift to higher binding
energy values is an indication of strong interaction between
support and metal particle. In the case of Ni(5 wt%)/ZrO2, the
Ni 2p3/2 peak appeared at lower BE value of 852.0 eV com-
pared to that of Ni(5 wt%)/CeO2–ZrO2 (at 852.1 eV) and
Ni(5 wt%)/CeO2 (at 852.5 eV). Hence, it can be concluded
that support-metal interactions are weaker in the case of
Ni(5 wt%)/ZrO2 than in Ni(5 wt%)/CeO2–ZrO2 and
Ni(5 wt%)/CeO2. Similar observations were found by Croy
et al. [21] for Pt supported on several oxides. ‘‘Neat’’ CeO2
shows a main line for 3d5/2 at 882.2 eV. In the case of CeO2-
and CeO2–ZrO2-supported Ni(5 wt%) catalysts, this line
appeared at 881.8 and 882.0 eV, respectively. Shift in the line
position of 3d5/2 is due to changes in the electronic structure of
Ce4? as a consequence of support-metal interactions. XPS
revealed that a part of Ni in the catalysts is in ?2 oxidation
states showing a main 2p3/2 line at around 855 eV and its
corresponding satellite line at 859.8 eV (Fig. 3). Spectral
deconvolution enabled the relative percentage concentrations
of zero and ?2 valent Ni species (sensitivity factor of Ni
in concentration estimations was considered as 4.5), which
were (69, 31), (43.2, 56.8), (28.1, 71.9), (18, 72) and (63, 37)
for Ni(5 wt%)/ZrO2, Ni(5 wt%)/CeO2–ZrO2, Ni(5 wt%)/
CeO2, Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2 and Ni(5 wt%)–
Pd(0.5 wt%)/CeO2–ZrO2, respectively. This indicates that
ZrO2 and CeO2–ZrO2 supports, in presence of Pd, keep most
of the Ni in metallic state. Such effects of support on the
electronic properties of Ni are found to have profound influ-
ence on the activity of these catalysts.
MSS, MD and average metal particle size (Da) were
determined from hydrogen chemisorption studies (Table 2).
The values of MSS and MD for different supported
Ni catalysts decreased as follows: Ni(5 wt%)/CeO2 [
Ni(5 wt%)/CeO2–ZrO2 [ Ni(5 wt%)/ZrO2. The value of
Da varied in the reverse order. In other words, support CeO2
facilitated dispersion of nickel. An enhancement in MD was
observed also when Pt or Pd (0.5 wt%) was present with
Ni(5 wt%) in the catalyst (Table 2). It should, however, be
noted that under chemisorption conditions, Pt and Pd too ad/
absorb hydrogen and thereby enhanced hydrogen uptake
would be observed. In other words, in those cases, the
estimated dispersion is more than the actual value. Dis-
crepancy in the values of crystallite size determined from
XRD and H2 adsorption may noticed. This was due to the
fact that XRD detects only those crystallites which are
above its detection limit of ca., 3–5 nm whereas, crystals of
all dimensions contribute to H2 adsorption and hence, dis-
crepancy is inevitable whenever there is a mix of large and
smaller crystallites of Ni present in the system.
All these physicochemical studies reveal that support
and noble metal co-deposition have effects on crystallite/
particle sizes, MD and electronic properties of Ni. These
effects, in turn, have bearing on the activities of these
catalysts.
3.2 Catalytic Activity
Controlled, blank experiments revealed that CA conversion
is negligible (1.5 wt% in 8 h) in the absence of a catalyst
(Table 3; run no. 1). The supported Ni catalysts of the
present study are highly active even at 100 °C and at a
hydrogen pressure of 20 bar. A small quantity of the cat-
alyst (0.05 g; Table 3) is enough for quantitative conver-
sion of CA (1 g). In general, HCA is the selective product
(89.6–100 wt%). Support showed significant effect on the
catalytic activity of Ni. Conversion of CA and turnover
frequency [TOF = moles of CA converted per mole of
exposed Ni (determined from chemisorption) per hour] for
different supported Ni catalysts decreased in the following
order: Ni(5 wt%)/ZrO2 (100 %, 573) [ Ni(5 wt%)/CeO2
(68.3 %, 456) [ Ni(5 wt%)/CeO2–ZrO2 (45.3 %, 219)
(Table 3; run nos. 2, 3 and 4). Although it appears that the
The chemical states of Pt and Pd in reduced catalysts,
Pt(0.5 wt%)/CeO2–ZrO2 and Pd(0.5 wt%)/CeO2–ZrO2,
respectively were determined from their corresponding
core level spectra (Fig. 4). Pt was present in zero, ?2 and
123

S. Bhogeswararao et al.
854.8
2
(c) Ni(5 wt%)/CeO -ZrO
854.7
(a)Ni(5 wt%)/CeO
(b) Ni(5 wt%)/ZrO
852.0
2
2
2
2+
855 (Ni
)
859.8 (Ni2+
)
852.5 (Ni0)
Sat
852.1
Ni 2p
3/2
849
852
855
858
861
864
849
852
855
858
849
852
855
858
Binding Energy (eV)
Binding Energy (eV)
Binding Energy (eV)
Fig. 3 XPS of Ni(5 wt%) supported on ZrO2, CeO2 and CeO2–ZrO2
Fig. 4 XPS core level spectra
of Pt(0.5 wt%)/CeO2–ZrO2 and
Pd(0.5 wt%)/CeO2–ZrO2
Pd(0.5 wt%)/CeO-ZrO2
Pt(0.5 wt%)/CeO -ZrO2
2
2
Pd0
Pt0
Pt4+
Pd2+
Pt2+
80
78
76
74
72
70
68
66
344
340
336
332
Binding Energy (eV)
Binding Energy (eV)
Table 2 H2 chemisorption
data of CeO2–ZrO2-supported
Ni catalysts
Catalyst
H2 chemisorbed Metal surface area Metal
(m2/g)
Av. metal particle
dispersion (%) size (nm)
Ni(5 wt%)/CeO2
Ni(5 wt%)/ZrO2
0.716
0.366
0.440
58.1
29.0
34.2
8.7
4.4
5.1
11.6
23.2
19.7
Ni(5 wt%)/
CeO2–ZrO2
Ni(5 wt%)–Pt(0.5 wt%)/
CeO2–ZrO2
0.763
61.9
98.0
9.3
10.9
6.9
Ni(5 wt%)–Pd(0.5 wt%)/ 1.235
CeO2–ZrO2
14.7
selectivity for HCA at these conversions decreased in
the order: Ni(5 wt%)/CeO2–ZrO2 (97.3 %) [ Ni(5 wt%)/
CeO2 (93.7 %) [ Ni(5 wt%)/ZrO2 (89.6 %), it should be
noted that HCA selectivity is nearly the same at similar
conversion levels over all these different supported Ni
catalysts (Fig. 5). Formation of PPL through further
hydrogenation of HCA was more predominant on
Ni(5 wt%)/ZrO2. The catalyst with Pt(0.5 wt%) supported
on CeO2–ZrO2 showed weaker catalytic activity (CA
conversion) than Ni(5 wt%)/CeO2–ZrO2 (Table 3; com-
pare run no. 7 with 4). The Pt catalyst, unlike others,
yielded CAL with a selectivity of 8.2 wt%. Pd(0.5 wt%)/
CeO2–ZrO2 was highly active (run no. 8). HCA was the
main product, with a small quantity of other products
(acetals and higher molecular weight products; 2.5 wt%)
formed over this catalyst.
3.2.1 Influence of Noble Metal
A significant enhancement in CA conversion of CeO2–
ZrO2-supported Ni was observed when Pd or Pt (0.5 wt%)
was also part of the catalyst composition (Table 3; com-
pare run nos. 5 and 6 with 4). At 100 wt% CA conversion,
HCA and PPL formed with equal weight percentage over
Ni–Pt catalyst while they formed in the ratio of 20:80 wt%
over Ni–Pd catalyst. Metal dispersion was higher in the
case of Ni–Pd than ‘‘neat’’ Ni supported on CeO2–ZrO2.
Due to this, 100 % conversion of CA (with HCA
123

Selective Hydrogenation of Cinnamaldehyde
Table 3 Hydrogenation of cinnamaldehyde over supported Ni (5 wt%) catalysts: effect of support, noble metal and alkali promoter
Run No.
Catalyst
CA conversion (wt%)
TOF (h-1
)
Product selectivity (%)
HCA
PPL
Others
Without alkali
1
–
1.5
–
100
0
0
2
Ni(5 wt%)/ZrO2
100
79.4
68.3
8.8
573
3643
456
470
219
58
89.6
96.1
93.7
95.6
97.3
100
10.5
3.9
6.3
4.4
2.7
0
0
2a
Ni(5 wt%)/ZrO2
0
3
Ni(5 wt%)/CeO2
0
3a
Ni(5 wt%)/CeO2
0
4
Ni(5 wt%)/CeO2–ZrO2
Ni(5 wt%)/CeO2–ZrO2
Ni(5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2
Pt(0.5 wt%)/CeO2–ZrO2
Pt(0.5 wt%)/CeO2–ZrO2
Pd(0.5 wt%)/CeO2–ZrO2
Pd(0.5 wt%)/CeO2–ZrO2
45.3
1.5
0
4a
0
4b
14.2
100
64.1
100
100
100
17.4
1.2
275
277
1424
171
1373
2746
–
100
0
0
5
45.1
96.0
20.0
85.0
92.0
77.5
74.2
88.3
94.7
54.9
4.0
80.0
15.0
8.0
5.8
13.5
9.3
4.2
0
5a
0
6
0
6a
0
6b
0
7
8.5(8.2)b
12.3
2.5
7a
–
8
100
41.5
–
8a
With alkalia
–
1.1
9
–
8.7
–
99.2
7.4
0.8
0.0
10
10a
11
12
13
13a
14
15
16
Ni(5 wt%)/ZrO2
100
96.7
100
100
100
100
100
78.5
100
573
4436
668
484
277
2221
171
–
89.9
6.1
2.8
Ni(5 wt%)/ZrO2
93.9
7.7
0
Ni(5 wt%)/CeO2
77.7
65.3
80.1
31.8
99.9
11.2
20.3
14.6
11.1
4.2
Ni(5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2
Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2
Pt(0.5 wt%)/CeO2–ZrO2
Pd(0.5 wt%)/CeO2–ZrO2
23.6
15.7
68.2
0.03
71.8
70.3
0
0.0
-(17.0)b
–
9.4
Reaction conditions catalyst (reduced at 400 °C for 2.5 h) = 0.05 g, cinnamaldehyde = 1 g, solvent (iso-propanol) = 25 ml, H2 pressur-
e = 20 bar, reaction temperature = 100 °C, reaction time = 8 h, stirring speed = 600 rpm. Turnover frequency (TOF) = moles of CA con-
verted per mole of exposed Ni (determined from chemisorption) per hour. Run nos. 2a, 3a, 4a, 5a, 6a, 7a, 8a, 10a and 13a are those with reaction
time of 1 h and 4b is with reaction time of 2 h and 6b is with reaction time of 0.5 h, rest of the conditions are the same for all
CA cinnamaldehyde, HCA hydrocinnamaldehyde, PPL 3-phenyl propanal, others include higher molecular weight condensation products
a
Alkali = NaOH (0.01 g) in 5 ml water
b
Values in parentheses corresponds to the selectivity of cinnamyl alcohol
selectivity of 85 % and PPL selectivity of 15 %) was
achieved in just 0.5 h itself over Ni–Pd catalyst. For sup-
ported ‘‘neat’’ Ni catalyst, CA conversion was only 45.3 %
even after 8 h of reaction. TOF value of Ni–Pd at 1 h
(1373 h-1, run number 6a) is higher than that of Ni
(58 h-1, run no. 4a). While Pd(0.5 wt%)/CeO2–ZrO2 and
Ni(5 wt%)/CeO2–ZrO2 showed CA conversions of 41.5
and 1.5 wt% in 1 h (run nos. 8a and 4a), Ni(5 wt%)–Pd(0.5
wt%)/CeO2–ZrO2 showed a conversion of 100 wt% (run
no. 6a). Similar observations were found also in the case of
Pt promoted Ni catalyst (compare run nos. 4a, 7a and 5a).
All these observations point out that enhanced Ni disper-
sion due to the presence of noble metal is the cause for the
higher catalytic activity of supported-noble metal promoted
Ni catalysts. H2 chemisorption values reported in Table 2
have contribution from both the metals as we could not
separate out the individual contributions of Ni and noble
metal in bimetallic systems,. So dispersion of Ni in bime-
tallic catalysts is actually lower than what is reported in the
table. Taking this into consideration, TOF for run nos. 5–6
and 13–14 in Table 3 are the lowest values reported than
the actual ones.
123

S. Bhogeswararao et al.
carbonyl group in CA and enhances the rate of conversion
of HCA to PPL [12, 13]. Comparison of data for runs 10
and 10a and 13 and 13a, reveals that near 100 % conver-
sion of CA is achieved in just 1 h itself when alkali pro-
moter is added (Table 3). HCA is the selective product at
1 h. This product gets further hydrogenated in the due
course of time and PPL becomes the main product at 8 h.
The catalyst, Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2 gave
100 wt% conversion of CA and 99.9 wt% selectivity for
PPL (run no. 14). Thus, the remarkable promotional
activity of Pd and alkali is worth noting in the hydroge-
nation of CA over supported Ni catalysts. While
Ni(5 wt%)/ZrO2 yielded 100 wt% CA conversion and
89.6 wt% HCA selectivity, Ni(5 wt%)–Pd(0.5 wt%)/
CeO2–ZrO2 gave 100 wt% CA conversion and 99.9 wt%
PPL selectivity in a reaction at 100 °C, 20 bar of H2
pressure and reaction time of 8 h. By a small modification
in catalyst composition, the product selectivity could be
altered significantly. Upon alkali addition, even in the case
of monometallic Pt catalysts, an enhancement in CA con-
version from 17.4 to 78.5 wt% and increase in CAL
selectivity from 8.2 to 17 wt% was observed. However,
unlike with monometallic Ni, HCA is the major product on
the monometallic Pt and Pd catalysts even after addition of
alkali. This difference is, perhaps, due to difference in the
amounts of Ni (5 wt%) and Pt/Pd (0.5 wt%) present on the
catalyst surface and also due to their differences in crys-
tallite sizes and strengths of adsorption of the product HCA
molecules for them to react in further hydrogenation. Run
nos. (2 and 10) and (5 and 13) for Ni(5 wt%)/ZrO2 and
Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2 without and with
alkali addition showed similar CA conversion at the end of
8 h. So, for demonstrating the influences of support and Pt
on the catalytic activity of Ni (CA conversion and TOF),
we have carried out reactions for shorter period of time
(1 h) and the data are included in run nos. (2a and 10a) and
(5a and 13a), respectively.
100
80
60
40
20
0
Ni(5 wt%)/CeO2 -ZrO
2
(a) Without alkali
(b) With alkali
CA conversion
HCA selectivity
PPL selectivity
0
2
4
6
8
0
2
4
6
8
Reaction time (h)
Ni(5 wt%)-Pt(0.5 wt%)/CeO -ZrO
2
2
100
80
60
40
20
0
(b) With alkali
(a) Without alkali
CA conversion
HCA selectivity
PPL selectivity
0
2
4
6
8
0
2
4
6
8
Reaction time (h)
Fig. 5 Hydrogenation of cinnamaldehyde (CA) with and without
alkali over Ni(5 wt%)/CeO2–ZrO2 and Ni(5 wt%)–Pt(0.5 wt%)/
CeO2–ZrO2. Reaction conditions catalyst (reduced at 400 °C for
2.5 h) = 0.05 g, CA = 1 g, iso-propanol = 25 ml, alkali = NaOH
(0.01 g) in 5 ml water, H2 pressure = 20 bar, reaction tempera-
ture = 100 °C, stirring speed = 600 rpm
As seen from Table 3, in the absence of alkali, the
system is clean with only HCA and PPL as products [the Pt
catalyst (run no. 7)] makes the alcohol as expected. How-
ever, when alkali was added, base-catalyzed aldol con-
densation products (others) formed in good amount.
Supported Ni–Pd (run no. 14) is highly active so that
hydrogenation is more dominant than aldol condensation
thus, enabled high activity and PPL selectivity. As seen
from the catalytic data, addition of Pt has less important
effect on the activity of Ni catalyst while Pd has more
significant effects. This could be because of differences in
the d-band width of Pt and Pd [23]. Based on theoretical
calculations for adsorption of ethylene on Pt(111) and
Pd(111) surfaces, Sautet and co-workers [24, 25] reported
that Pd metal in bulk state has a slightly higher Fermi level
than Pt metal and more importantly a d-band width
3.2.2 Influence of Alkali
When a small of quantify of alkali (NaOH) was added to
the reaction mixture, catalytic activity (TOF) increased still
further (Table 3, run nos. 10–16). PPL was the selective
product in case of the supported Ni catalysts. Alkali alone
couldn’t catalyze this reaction to a significant extent (CA
conversion = 8.7 wt%; run no. 9). When run 4a was
conducted adding additionally 5 ml of water, increase in
CA conversion from 14.2 to 25.4 wt% was noted [product
selectivity: 97.3 % (HCA) and 2.7 wt% (PPL)]. These
studies, point out that both water and alkali have a role in
enhancing catalytic activity of these catalysts. We believe
that while water and OH- (in alkali) alter the electronic
properties of the metal, Na? ion facilitates polarization of
123

Selective Hydrogenation of Cinnamaldehyde
Table 4 Influence of amount of alkali on hydrogenation of cinnamaldehyde over Ni(5 wt%)/CeO2–ZrO2
Run no.
Amount of NaOH (g)
CA conv. (wt%)
TOF (h-1
)
Product selectivity (%)
HCA
PPL
Others
1
2
3
4
0
14.2
274
748
801
955
97.2
2.8
0
0.0025
0.005
0.01
38.7
96.2
3.9
0
41.4
49.4 (100)a
95.4
98.5 (23.6)a
4.7
1.5 (65.3)a
0
0 (11.1)a
Reaction conditions Catalyst (reduced at 400 °C for 2.5 h) = 0.05 g, cinnamaldehyde (CA) = 1 g, H2 pressure = 20 bar, solvent (iso-
propanol) = 25 ml, alkali = 0.0025–0.01 g of NaOH in 5 ml of water, reaction temperature = 100 °C, reaction time = 2 h, stirring
speed = 600 rpm. Turnover frequency (TOF) = moles of CA converted per mole of exposed Ni (determined from chemisorption) per hour
HCA hydrocinnamaldehyde, PPL 3-phenylpropanal, others include higher molecular weight condensation products
a
Values in parentheses are the data at 8 h of reaction
significantly reduced compared to that of Pt. Hence, the
radial expansion of the Pd d orbitals is smaller than the Pt d
orbitals resulting in reduced overlap of the d orbitals with
the orbitals of the adsorbate on Pd (i.e., reduced 4e-
interactions between adsorbate and surface and reduced
stabilizing 2e- interactions). The direct result of this is
stronger adsorbate interactions on Pd than on Pt surfaces.
As noted above alkali had a positive effect on the rate of
hydrogenation. While product selectivity is nearly the
same, CA conversion and TOF increased with increasing
amount of alkali–NaOH (Table 4). This increase was found
marginal beyond an alkali amount of 0.005 g. Alkali ions
modify the electronic structure of the active metal [12].
They facilitate adsorption of CA via. polarizing the C=O
group. This enhanced adsorption and effects on electronic
structure of Ni (XPS) are the possible causes for the high
convesion of CA in the presence of alkali ions. A strong
alkali-support interaction exists [12] and this effect could
also play a role on the overall activity. Ni(5 wt%)/CeO2–
ZrO2 has an acidity of 0.208 mmol/g (NH3-TPD). The
added alkali (0.005 g for 50 mg of catalyst which is
equivalent to 2.5 mmol/g catalyst) neutralizes these acidic
sites in the catalyst, modifies the electronic structure of the
metal [12] and facilitates adsorption polarizing the car-
bonyl group of CA molecule. The promotional effects
observed could be explained in terms of the catalyst
modifications. However, more detailed studies are needed.
Ni(5 wt%)/CeO2–ZrO2 is weakly active below 100 °C.
TOF increased with increasing temperature; 100 wt%
conversion of CA was achieved at 125 °C. Further rise in
temperature changed product selectivity from HCA to PPL.
Under similar conditions, conversion increased with
increasing CA concentration from 0.29 to 0.9 mol/l and
above that, it was unaltered. In agreement with the reports
by others [26], this experiment reveals that the reaction is
1st order with respect to CA up to a molar concentration of
0.9 and 0th order above that concentration. CA conversion
increased with increasing hydrogen pressure. HCA
Table 5 Influence of Ni loading on the hydrogenation of cinnamal-
dehyde over Ni/ZrO2 catalyst
Ni loading (wt%) CA conversion (wt%) Product selectivity (%)
HCA
PPL
3
70.0
97.4
96.5
98.8
94.1
92.5
92.8
88.2
5.9
5
7.5
10
15
7.2
10.6 (1.2)a
Reaction conditions Catalyst (reduced at 400 °C for 2.5 h) = 0.05 g,
cinnamaldehyde (CA) = 1 g, H2 pressure = 20 bar, solvent (iso-
propanol) = 25 ml, alkali = nil, reaction temperature = 100 °C,
reaction time = 2 h, stirring speed = 600 rpm
HCA hydrocinnamaldehyde, PPL 3-phenylpropanal
a
Value in parentheses corresponds to that for higher molecular
weight condensation (other) products
decreased with a growth in PPL selectivity. Based on initial
rates, energy of activation was estimated and it was found
to be 116 kJ/mol. This value is higher than that for Pd/SiO2
(30.1 kJ/mol) [26] and Co–B amorphous alloy (18.0 kJ/
mol) [27, 28] indicating weaker hydrogenation activity of
Ni. As such, to the best of our knowledge, there are no
reports for comparison on the energy of activation of Ni
catalysts for this reaction. The amount of Ni loading
affected the catalytic activity. CA conversion increased
with increasing Ni loading indicating that there is no mass
transfer limitation on the kinetics of the reaction (Table 5).
On ZrO2 support, nickel loading of 5 wt% was found
sufficient to achieve 100 wt% CA conversion, in just, 2 h
at 100 °C. At still higher loadings, more and more amount
of PPL formed due to hydrogenation of C=O on the larger
crystallites of Ni formed at higher Ni loadings (XRD).
CA conversion increased with increasing reaction time.
About 4 h was need for quantitative conversion of CA over
Ni(5 wt%)/ZrO2. This upon alkali addition could be
achieved in 2 h. CA conversion was only 45.3 % even after
8 h over Ni(5 wt%)/CeO2–ZrO2. However, on adding alkali
123

S. Bhogeswararao et al.
Fig. 6 Variation of product
selectivity as a function of CA
conversion in the absence of
alkali addition. (Left) over Ni(5
wt%)/CeO2–ZrO2, and (right)
over all the three supports
100
98
96
94
92
90
100
95
90
85
80
Ni(5 wt%)/CeO -ZrO
HCA
PPL
HCA
PPL
2
2
Without alkali addition
20
15
10
5
8
6
4
2
0
0
0
20
40
60
80
100
0
20
40
60
80
100
CA conversion (wt. %)
CA conversion (wt. %)
Table 6 Comparative catalytic activity data of supported Ni catalysts
Entry Catalyst
no.
Reaction conditions
CA HCA
conversion selectivity
Ref.
(wt%)
(wt%)
1
2
3
4
5
6
7
8
Ni(5 wt%)/ZrO2
Liquid phase; catalyst = 0.05 g, CA = 1 g, iso- 100
propanol = 25 ml, H2 = 20 bar, T = 100 °C,
t = 8 h
89.6
This work
Ni(5 wt%)–Pd(0.5 wt%)/CeO2–ZrO2 Liquid phase; catalyst = 0.05 g, CA = 1 g, iso- 100
propanol = 25 ml, NaOH (0.01 g) in 5 ml
0.03 (PPL = 99.9) This work
water, H2 = 20 bar, T = 100 °C, t = 8 h
Ni(5 wt%)–Pt(0.5 wt%)/CeO2–ZrO2 Liquid phase; catalyst = 0.05 g, CA = 1 g, iso- 100
propanol = 25 ml, NaOH (0.01 g) in 5 ml
92.8
90
This work
water, H2 = 20 bar, T = 100 °C, t = 0.5 h
Ni12P5/SiO2
Liquid phase; catalyst = 0.5 g, CA = 3 ml,
cyclohexane = 115 ml, H2 = 30 bar,
T = 120 °C, t = 8 h
Liquid phase; catalyst = 0.36 g, CA = 1 ml,
ethanol = 19 ml, H2 = 20 bar, T = 80 °C,
t = 2 h
90
96
20
23
22
Pt(0.5 wt%)–Ni(0.34 wt%)/CNT
Ni(10 wt%)/c-Al2O3
88
Vapor phase; H2/CA = 10 ml/ml, W/F = 136 s, 87.5
T = 300 °C (initial activity; catalyst
deactivated with time)
73.7
Ni(15 wt%)/MCM-41
Ni(10 wt%)–Cu(10 wt%)/SiO2
Vapor phase; H2/CA = 2.8, W/F = 0.19 gNi h/ 99.5
mol, T = 200 °C, P = 1 bar (initial activity;
catalyst deactivated with time)
21.3 (PPL = 26.9) 24
Vapor phase; H2 = 20 ml/min, CA = 1–3 ml/h, 80
T = 200 °C, P = 1 bar
70
21
the conversion reached 100 % at the end of 4 h (Fig. 3). The
CeO2–ZrO2-supported Ni catalyst was less active than the
ZrO2-supported Ni catalyst. When both Pt and alkali were
present, the catalytic activity of Ni was promoted and
100 wt% conversion of CA was achieved in just 0.5 h
(Fig. 5). HCA transformed into PPL on prolonging the
reaction time. The S-shaped curves for HCA suggest that it
is an intermediate product and two consecutive reactions (1:
CA ? HCA and 2: HCA ? PPL) are operative at our
reaction conditions. In order to confirm this further, based on
the data reported in Tables 3 and 4, the conversion verses
product selectivity plots for Ni(5 wt%)/CeO2–ZrO2 and Ni
supported on all the supports were drawn (Fig. 5). It could
be noted that irrespective of support and promoters, the
selectivity for HCA decreased and PPL increased system-
atically with a rise in CA conversion (Fig. 6). This reveals
that support and promoters have influenced the catalytic
activity of Ni without affecting its inherent selectivity for
hydrogenating C=C functionality. Although Pt alone has
affinity for C=O hydrogenation, its addition did not influ-
ence the affinity of Ni for C=O over C=C hydrogenation.
The CeO2–ZrO2-supported Ni promoted with noble metal
and alkali was found superior to most of the reported Ni cat-
alysts (Table 6) [29–33]. Both CA conversion and HCA
123

Selective Hydrogenation of Cinnamaldehyde
selectivity are higher over the catalyst of the present study.
The reaction occurs at moderate temperature (100 °C) and
pressure (20 bar). The noble metal promoted Ni catalysts
showed enhanced catalytic activities than the corresponding
individual mono metal catalysts. So far, Pt(0.5 wt%)–
Ni(0.34 wt%)/CNT [34] (0.36 g) is known as the best Ni
catalyst exhibiting CA conversion of 96 % and HCA selec-
tivity of 88 % at 80 °C, 20 bar H2 pressure and reaction time
of2 h.But,then,thepresentcatalystNi(5 wt%)–Pt(0.5 wt%)/
CeO2–ZrO2 (0.05 g) at 100 °C, 20 bar H2 pressure and
reaction time of 0.5 h showed 100 wt% CA conversion and
HCA selectivity of 92.8 %. This clearly demonstrates that the
catalysts of the present study are more active and selective for
C=C hydrogenation than the hitherto known Ni catalysts. As
described by Kyriakou et al. [34] for Pd/Cu(111), it is likely
even in the present case that the Pd atoms on Ni act as sites for
facile dissociation of H2, followed by its spillover on Ni for
further hydrogenation of CA. XPS revealed that in the pres-
ence of noble metals, BE value of Ni 2p3/2 was lower indi-
cating that Ni associated with noble metal is richer in electron
density. The electron-rich Ni (in the presence of noble metals)
facilitates adsorption of CA through C=C for its further
hydrogenation. Likewise, alkali also is known to affect the
electronic structure. Alkali ions (Na?) facilitate adsorption of
CAthroughpolarizationofthecarbonylofCA. This enhanced
adsorption of CA molecules and electronic effects are the
causes for more efficient activity of Ni and Ni–Pt/Pd catalysts
in presence of alkali solution.
addition enhanced the catalytic activity of these supported
Ni catalysts. Quantitative conversion of CA was achieved
at moderate conditions (100 °C and 20 bar).
Acknowledgments S.B. acknowledges the Council of Scientific and
Industrial Research, New Delhi for the award of Senior Research
Fellowship.
References
1. Gallezot P, Richard D (1998) Catal Rev Sci Eng 40:81
2. Claus P (1998) Top Catal 5:51
3. Singh UK, Vannice MA (2001) Appl Catal A 213:1
´
4. Maki Arvela P, Hajek J, Salmi T, Murzin DY (2005) Appl Catal
¨
A 292:1
5. Mahmoud S, Hammoudeh A, Gharaibeh S, Melsheimer J (2002) J
Mol Catal A 178:161
6. Chen X, Li M, Guan J, Wang X, Williams CT, Liang C (2012)
Ind Eng Chem Res 51:3604
7. Wu B, Huang H, Yand J, Zheng N, Fu G (2012) Angew Chem Int
Ed 51:1
8. Zheng R, Porosoff MD, Weiner JL, Lu S, Zhu Y, Chen JG (2012)
Appl Catal A 419:126
9. Merio AB, Machado BF, Vetere V, Faria JL, Casella ML (2010)
Appl Catal A 383:43
¨
10. Plomp AJ, van Asten DMP, van der Eerden AMJ, Maki-Arvela P,
Murzin DY, de Jong KP, Bitter JH (2009) J Catal 263:145
11. Teddy J, Falqui A, Corrias A, Carta D, Lecante P, Gerber I, Serp
P (2011) J Catal 278:59
12. Bhogeswararao S, Srinivas D (2010) Catal Lett 140:55
13. Bhogeswararao S, Srinivas D (2012) J Catal 285:31
14. Che M, Bennett CO (1989) Adv Catal 36:55
15. Fadoni M, Lucarelli L (1999) Stud Surf Sci Catal 120A:177
16. Srinivas D, Satyanarayana CVV, Potdar HS, Ratnasamy P (2003)
Appl Catal A 246:323
Delbecq and Sautet [23] reported that metal selectivities
can be rationalized in terms of different radial expansion of
their d bands; the smaller the band, stronger will be the
four-electron attraction interactions with the C=C bond and
higher will be the probability of its adsorption. The d band
width of Ni is much smaller than that of Pt, Ir and Os. This
accounts well for the enhanced selectivities of HCA in the
hydrogenations over supported Ni catalysts. Presence of
noble metal enhanced the dispersion of Ni and alkali
facilitated adsorption and polarization of carbonyl group of
CA. All these features in turn resulted in higher catalytic
activity of ZrO2-supported Ni catalysts.
17. Raab CG, Lercher JA (1993) Catal Lett 18:99
18. Croy JR, Mostafa S, Hickman L, Heinrich H, Roldan Cuenya B
(2008) Appl Catal A 350:207
´ ´
19. Domok M, Oszko A, Baan K, Sarusi I, Erdohelyi A (2010) Appl
¨
Catal A 383:33
¨
¨
20. Laursen S, Linic S (2006) Phys Rev Lett 97:026101
21. Croy JR, Mostafa S, Lin J, Sohn Y, Heinrich H, Cuenya BR
(2007) Catal Lett 119:209
22. Shukla AK, Raman RK, Choudhury NA, Priolkar KR, Sarode PR,
Emura S, Kumashiro R (2004) J Electroanalyt Chem 563:181
23. Delbecq F, Sautet P (1995) J Catal 152:217
24. Sautet P, Paul JF (1991) Catal Lett 9:245
25. Delbecq F, Sautet P (1994) Catal Lett 28:89
26. Li H, Chen X, Wang M, Xu Y (2002) Appl Catal A 225:117
27. Hollis DP, Selwood P (1961) J Chem Phys 35:378
28. Bonneviot L, Che M, Olivier D, Martin GA, Freund E (1986) J
Phys Chem 90:2112
4 Conclusions
29. Wang H, Shu Y, Zheng M, Zhang T (2008) Catal Lett 124:219
30. Reddy BM, Kumar GM, Ganesh I, Khan A (2006) J Mol Catal A
247:80
31. Chin SY, Lin AF-J, Ko AA-N (2009) Catal Lett 132:389
32. Li Y, Ge CH, Zhao J, Zhou R (2008) Catal Lett 126:280
33. Jao KY, Liu KW, Yang YH, Ko AN (2009) J Chin Chem Soc 56:885
34. Kyriakou G, Boucher MB, Jewell AD, Lewis EA, Lawton TJ,
Baber AE, Tierney HL, Flytzani-Stephanopoulos M, Sykes ECH
(2012) Scienc: 335:1209
The catalytic activity of Ni supported on CeO2, ZrO2 and
CeO2–ZrO2 for liquid-phase hydrogenation of CA was
investigated for the first time. These catalysts showed
higher activity than the hither-to-known Ni catalysts for
selective hydrogenation of olefinic bonds. The catalytic
activity of Ni over different supports decreased in the
order: Ni(5 wt%)/ZrO2 [ Ni(5 wt%)/CeO2 [ Ni(5 wt%)/
CeO2–ZrO2. Noble metals (Pd and Pt; 0.5 wt%) and alkali
123