Pt supported nitrogen doped hollow carbon spheres for the catalysed reduction of cinnamaldehyde

Article abstract of DOI:10.1016/j.apcata.2016.02.025

Carbon and nitrogen coated St?ber silica spheres were prepared by a standard procedure using acetonitrile at different CVD temperatures (800-1000 °C) and as a function of time (1-4 h). HF treatment was used to remove the silica from the carbon coated silica to give monodispersed nitrogen doped hollow carbon spheres (N-HCSs; shell thickness = 50 nm). The surface areas of the N-CSs increased significantly after removal of the SiO₂ core (N-HCS = 47 m²/g). Platinum nanoparticles, prepared by a [Pt(COD)Cl₂] deposition method were dispersed (d = 2.8 ± 0.4 nm) on the N-HCS supports. Thermal treatment of the Pt/N-HCS900-4 (prepared at 900°C for 4 h) in a 5% H₂/Ar atmosphere at 120 °C (Pt/N-HCS900-4-120) and 300 °C (Pt/N-HCS900-4-300) gave metallic Pt (2.48%) which was confirmed by XRD and XPS studies. The XPS N1s spectra of Pt/N-HCS900-4-120 showed the presence of quaternary, pyridinic and pyrollic nitrogen atoms. These nitrogen doped carbon supported Pt nanoparticle catalysts were observed to reduce cinnamaldehyde to the unsaturated alcohol, cinnamyl alcohol (isopropanol/30 bar H₂/80 °C) with >99% conversion and selectivity. The catalyst used in the reduction of cinnamaldehyde to cinnamyl alcohol could be recycled 12 times with minimal loss in conversion and this correlated with the small amount of Pt leaching or physical loss of catalyst (<2%). The corresponding Pt/catalyst made on undoped HCSs showed a much greater loss of activity with recycling. It is seen that N doping of carbon is an effective method to bind Pt to carbon.

Full text of DOI:10.1016/j.apcata.2016.02.025

Applied Catalysis A: General 517 (2016) 30–38
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Pt supported nitrogen doped hollow carbon spheres for the catalysed
reduction of cinnamaldehyde
Isaac Nongwea,c, Vilas Ravata, Reinout Meijboomb, Neil J. Covillea,∗
a DST-NRF Centre of Excellence in Strong Materials and Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, Johannesburg,
WITS 2050, South Africa
b Department of Chemistry, University of Johannesburg, Johannesburg, Auckland Park 2006, South Africa
c Department of Civil and Chemical Engineering, University of South Africa, P/Bag X6, FL, Johannesburg 1710, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
Carbon and nitrogen coated Stöber silica spheres were prepared by a standard procedure using acetoni-
trile at different CVD temperatures (800–1000 ◦C) and as a function of time (1–4 h). HF treatment was
used to remove the silica from the carbon coated silica to give monodispersed nitrogen doped hollow
carbon spheres (N-HCSs; shell thickness = 50 nm). The surface areas of the N-CSs increased significantly
after removal of the SiO2 core (N-HCS = 47 m2/g). Platinum nanoparticles, prepared by a [Pt(COD)Cl2]
deposition method were dispersed (d = 2.8 0.4 nm) on the N-HCS supports. Thermal treatment of the
Pt/N-HCS900-4 (prepared at 900 ◦C for 4 h) in a 5% H2/Ar atmosphere at 120 ◦C (Pt/N-HCS900-4-120) and
300 ◦C (Pt/N-HCS900-4-300) gave metallic Pt (2.48%) which was confirmed by XRD and XPS studies. The
XPS N1s spectra of Pt/N-HCS900-4-120 showed the presence of quaternary, pyridinic and pyrollic nitro-
gen atoms. These nitrogen doped carbon supported Pt nanoparticle catalysts were observed to reduce
cinnamaldehyde to the unsaturated alcohol, cinnamyl alcohol (isopropanol/30 bar H2/80 ◦C) with >99%
conversion and selectivity. The catalyst used in the reduction of cinnamaldehyde to cinnamyl alcohol
could be recycled 12 times with minimal loss in conversion and this correlated with the small amount
of Pt leaching or physical loss of catalyst (<2%). The corresponding Pt/catalyst made on undoped HCSs
showed a much greater loss of activity with recycling. It is seen that N doping of carbon is an effective
Received 23 October 2015
Received in revised form 23 February 2016
Accepted 25 February 2016
Available online 2 March 2016
Keywords:
Nitrogen doped hollow carbon spheres
Silica template
Pt catalyst
Cinnamaldehyde reduction
method to bind Pt to carbon.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
resulting carbon material is an inverse replica of the silica template
and the morphology of the silica template is retained in the carbon
Spherically shaped solid carbon materials have been known for
decades. They have been well studied and characterised and their
properties have been exploited for use in many areas (tyres, batter-
ies, printer inks, supports for catalysis, medical probes etc) [1–7].
More recently a new class of spheres called hollow carbon spheres
(HCSs) have been developed [8–11].
In most procedures the HCSs are made from a core shell struc-
ture in which the core comprised of a removable template material
and the shell is made of carbon. Silica spheres are often used as the
template, and the deposition of a carbon reactant on the template
followed by subsequent removal of the template (HF [12], NaOH
[13] etc.) produces the HCSs. Partial removal of the interior has led
to egg/yolk or rattle materials [14]. In general, the structure of the
material [15]. Thus, by varying the size of the template, the size of
the HCS can be varied. The carbon layer deposited on the template
will be affected by the carbon containing reagent used to prepare
the layer. By choice of the reaction conditions and the type of the
carbon used the thickness of the layer can be controlled.
The surface of a carbon can be modified by functionalization
or by doping. Dopants such as N, B, P and O when incorporated
into a carbon structure will also tune the physicochemical prop-
erties of the carbon material i.e. doping on any shaped carbon
material should give materials with properties different from their
undoped analogues. Doping of HCSs can be expected to show sim-
ilar behaviour [15].
Doping of HCSs can in principle be achieved by two major
routes—in situ doping or post addition of an element to a HCS;
typically with N as the dopant. Studies have been reported in
which in situ N doping has been achieved from a variety of N
sources. For example, NHCSs have been synthesized by hydrother-
∗
Corresponding author.
E-mail address: neil.coville@wits.ac.za (N.J. Coville).
http://dx.doi.org/10.1016/j.apcata.2016.02.025
0926-860X/© 2016 Elsevier B.V. All rights reserved.

I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
31
mal carbonization of poly(o-phenylenediamine), [16] pyrolysis of
poly(o-phenylenediamine) [17], aminated tannins [18], ionic liq-
uids [19] and pyrroles [20] as the precursors. The N-HCSs produced
by hydrothermal methods result in amorphous carbon walls with
low thermal stability and a low level of graphitization. Alternatively
N-HCSs made by a high temperature CVD process e.g. using CH3CN
as N (and C) source gives HCSs with a more graphitic structure
[13,21].
HCSs with a controlled size, shape, morphology and chemi-
cal composition, have received growing interest because of their
potential applications in catalysis [3,16,22], gas storage [23],
adsorption [24], drug delivery [25], and energy conversion [26–28].
As expected, N-HCSs can also be exploited in similar areas to those
of HCSs [29]. Our own studies in this area relate to the use of doped
HCSs as catalyst supports [30,31].
In a further exploration of these materials we wish to report on
our studies using Pt/NHCS as a catalyst for the reduction of cin-
namaldehyde. We have chosen to prepare our NHCSs using a CVD
method as this method yields a more graphitic carbon with less
oxygenated surface groups. The binding interaction between the
carbon and the Pt should thus be predominantly influenced by the
effect of the N dopant atoms.
Selective catalytic hydrogenation of ␣,␤-unsaturated aldehy-
des is an important step in the industrial preparation of many
fine chemicals and attracts much interest [32]. Reduction of cin-
namaldehyde can lead to three different products: a saturated
aldehyde, a saturated alcohol and the desired unsaturated alcohol
(see reaction Scheme 1). When the reaction is performed in iso-
propanol as solvent other side products involving interaction with
the solvent can form [33]. Cinnamyl alcohol plays an important role
in the perfume and flavouring industries [34]. The hydrogenation of
the C C bond in ␣,␤-unsaturated aldehydes is both kinetically and
thermodynamically more favourable than the hydrogenation of the
washed with ethanol and distilled water 4 times and dried at 80 ◦C
for 12 h. Finally, they were sonicated for 15 min in a water and
ethanol mixture to give well separated monodisperse silica spheres.
2.2. Preparation of hollow carbon spheres and preparation of
catalysts
Deposition of carbon onto the large SiO2 spheres (1100 nm) was
carried out using a CVD bubbling method. A range of N containing
reagents (acetonitrile, aniline, triethylamine and trimethylamine)
were initially used as a carbon and nitrogen source to make the
N-HCSs respectively but the acetonitrile gave the best results
(see Supplementary material) [13]. Briefly, the silica spheres were
placed in a quartz tube and were heated to 800, 850 or 900 ◦C for
4 h or 900 ◦C for 1–3 h at a heating rate of 10 ◦C/min under a N2
flow (90 mL min−1) at atmospheric pressure. During this time the
silica spheres shrank and had diameters of 800 nm (800 ◦C), 750 nm
(850 ◦C) and 705 nm (900 ◦C) respectively. Then, N2, was bubbled
through the acetonitrile and the vapours passed through a quartz
tube reactor (20 cm × 2 cm) containing the silica spheres in order to
allow carbon deposition to occur on the surface of the silica spheres.
After a given period of time, a black composite consisting of car-
bon/silica spheres was obtained. Removal of the silica using 20 wt%
HF solution (15 mL HF was added to 1.5 g carbon/silica composite
and the mixture aged for 48 h) yielded a range of hollow carbon
spheres: N-HCS800-4, N-HCS850-4 and N-HCS900-4 for samples
made at 800, 850 and 900 ◦C for 4 h respectively, and N-HCS900-1,
N-HCS900-2 and N-HCS900-3, for the samples made when carbon
was deposited after 1–3 h at 900 ◦C.
A HCS sample without N was also prepared by depositing carbon
from toluene onto silica spheres (HCS900-4).
2.3. Synthesis of [Pt(COD)Cl2] and Pt/N-HCS catalyst
C
O group [35]. The most significant factors determining the selec-
tivity are the presence of a catalyst, the type and the structure of the
active metal, the characteristics of the catalyst particle (size/shape),
the type of the support and the conditions of the reaction, e.g. the
hydrogen pressure and the reaction temperature. The cinnamalde-
hyde reduction reaction thus provides an excellent model reaction
to study new catalyst formulations.
Many studies have been reported on the use of catalysts for cin-
namaldehyde reduction. Of the metals investigated Pt has proven
to give excellent results and reported studies using Pt, especially on
other carbon supports, provide good reference studies to evaluate
the use of N-HCSs [36–49].
Here we use the CVD synthesis route to make N-HCSs. The ability
of the N-HCS to act as a catalyst support was studied and the per-
formance of a supported Pt catalyst was evaluated for the selective
reduction of cinnamaldehyde to cinnamyl alcohol in isopropanol at
different temperatures and hydrogen pressures. No promoter was
used in the study.
The complex [Pt(COD)Cl2] was prepared by a method reported
in the literature [51,52]. A solution of K2PtCl4 (2.5 g, 6.0 mmol)
was prepared in H2O (40 mL). Glacial acetic acid (60 mL) and 1,3-
cyclo-octadiene (COD; 2.5 mL, 20 mmol) were added to the above
solution. The reaction mixture was stirred in air and heated at 90 ◦C
on a steam bath. After 30 min, the red solution turned pale yellow.
The solution was allowed to cool and pale yellow crystals were
isolated. The crystals were washed with H2O, ethanol and diethyl
ether. All spectroscopic data were in agreement with the literature
values [51,52].
The [Pt(COD)Cl2] (0.037 g) was dispersed in a solution of sodium
citrate (4 mL) in deionized water (50 mL) containing hollow carbon
spheres (1 g) and sonicated for 1 h. Then, 2 mL of 0.5 M NaBH4 was
introduced dropwise to the above suspension with vigorous stir-
ring at room temperature. The mixture was left stirring overnight
and the solid phase was recovered by filtration. It was washed with
water and then thermally treated in an Ar/H2 (5%) atmosphere
at 120 and 300 ◦C (these samples were called Pt/N-HCS900-4-120
and Pt/N-HCS900-4-300). A Pt/HCS900-4-120 catalyst was also
prepared to allow for comparison with the Pt/N-HCS900-4-120 cat-
alyst. Four other catalysts were prepared by the same procedure as
described above but using N-HCS800-4, N-HCS850-4, N-HCS900-1
and N-HCS900-2 as supports. The materials generated from the dif-
ferent N-HCSs heated at 120 ◦C in an Ar/H2 (5%) atmosphere were
called Pt/N-HCS800-4-120, Pt/N-HCS850-4-120, Pt/N-HCS900-1-
120 and Pt/N-HCS900-2-120.
2. Experimental
2.1. Synthesis of silica spheres
Monodisperse silica spheres with a diameter d = 1100 100 nm
were synthesized following a two-step modified Stöber procedure
described elsewhere [50]. In this reaction, 22.4 mL of TEOS, 15.4 mL
of 25% NH3 solution, and 18 mL of deionised water were added
into 130 mL of isopropanol and the mixture was aged for 1 h under
mild stirring at 40 ◦C. A further 22.4 mL of TEOS was then added
to the above reaction mixture which was stirred at room tempera-
ture overnight, resulting in the formation of a white silica colloidal
suspension. The silica particles were centrifuged, separated and
2.4. Catalytic reduction reaction
Liquid phase hydrogenation of cinnamaldehyde (98%, Aldrich)
was carried out in a 100 mL stirred batch reactor (autoclave) under
temperature control. The total volume in the teflon container

32
I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
3-Phenyl propionaldehyde
CHO
OH
CHO
3-Phenyl 1-propan-1-ol
OH
Cinnamldedyde
Cinnamyl alcohol
Scheme 1. Reaction network for the cinnamaldehyde hydrogenation reaction.
placed inside the autoclave was 35 mL. Reactions were performed
under different pressures (10, 20 and 30 bar) and at different
temperatures (20, 40, 60 and 80 ◦C). Prior to hydrogenation, the
catalysts were activated under a hydrogen atmosphere at 120 ◦C
(heating rate: 5 ◦C/min) for a period of 2 h. The reaction mix-
ture containing cinnamaldehyde (5 mL), catalyst (0.2 g) and solvent
(isopropanol, 30 mL) was placed in the teflon lined autoclave. To
determine if any Pt leaching had occurred, a hot filtration test
was used i.e. at the end of a catalytic reaction, the product mix-
ture was filtered off and the solid catalyst remaining was washed
with acetone and reused (or analyzed). The reaction products and
unconverted reactants from the filtrate were analyzed by gas chro-
matography with an FID, using a capillary column (Phenomenex
Zebron, 30 m × 0.53 mm I.D) and N2 as the carrier gas. Appropriate
standards were used to establish the identity of the GC peaks.
5 nm) coating of gold was deposited onto the samples to reduce
the effects of charging. Raman spectra of the spent catalysts were
obtained on a T64000 Raman spectrometer (Jobin Yvon triple spec-
trometer) under ambient conditions. A 514.5 nm Ar laser was used
as the exciting source with a power density of 1 mW cm−2 on the
sample surface and a power of 2 mW. The measurements were ref-
erenced to Si at 521 cm−1 with 16 data acquisitions in 180 s. A low
laser power was used to prevent burning of the samples.
3. Results and discussions
3.1. Structure and morphology of the nitrogen doped N-HCS
samples
3.1.1. Effect of CVD temperature
The silica spheres were prepared using a modified Stöber syn-
thesis and after deposition of carbon and nitrogen, the silica was
removed using HF. The SEM images (Fig. S1a) show monodisperse
silica spheres with a narrow size distribution (d = 1100 100 nm)
and homogeneous shape. The silica spheres were placed in a CVD
oven that was heated from room temperature to 800, 850 and
900 ◦C at 10 ◦C min−1 during which time the spheres shrank. N2
(100 mL min−1) was then bubbled through acetonitrile and this
mixture was passed over the spheres at 800, 850 and 900 ◦C for
4 h. The material was shown by SEM and TEM analysis (Fig. S1a)
to consist of accreted spheres with diameters of 825, 800 and
750 50 nm respectively with a single shell and thickness between
25 and 50 5 nm (see Table 1). The results indicate that a higher
CVD temperature enhanced the carbon deposition rate, increased
the shell thickness and the extent of silica shrinkage and promoted
the formation of a smooth carbon surface. The N content was found
to be 6.7, 7.2 and 8.6 wt% N for the N-HCS800-4, N-HCS850-4 and
N-HCS900-4 spheres. The data shows that the nitrogen content
increases with reaction temperature.
2.5. Catalyst characterization
Chemical analysis of the catalysts was performed using ICP-
OES to measure the Pt concentration (Spectro Genesis instrument).
Approximately 50 mg of catalyst was weighed and the carbon oxi-
dised at 700 ◦C. Then the samples were dissolved in 5 mL HF, 4 mL
HNO3 and few drop of HClO4. This procedure was repeated twice
after the acid evaporation step. Finally, the residue was dissolved
in aqua regia and heated until dryness and then an aqueous solu-
tion of HNO3 (1% v/v) was added. The solution was transferred to
a 50 mL volumetric flask for analysis. CN elemental analyses were
performed with an Elementar Vario Micro cube analyser (Rhodes
University, SA). A JEM 100s, an FEI Tecnai G2 Spirit and an FEI Tecnai
F20 X-Twin at 200 kV FEG with an Oxford EDS system were used
for transmission electron microscopy (TEM) studies. All samples
were ultrasonically suspended in methanol and a drop of the sus-
pension was transferred to a copper grid and allowed to dry before
TEM analysis. The number and diameters of the Pt particles and
the NHCSs were determined by standard counting procedures from
TEM images. The phase composition of the support and catalysts
were determined by means of XRD analysis on a Bruker D2 phaser
in Bragg Brentano geometry with a Lynxeye detector using Cu-K␣
radiation at 30 kV and 10 mA. The scan range was 10◦ < 2ꢀ < 90◦ in
0.040 steps, using a standard speed with an equivalent counting
time of 1 s per step. The diffraction peaks were then compared with
those of standard compounds reported in the Diffracplus evalua-
tion package using the EVA (V11.0, rev.0, 2005) software package.
A PerkinElmer TG/DTA Thermogravimetric analyzer was used to
measure weight changes of samples heated in air or nitrogen at a
constant heating rate of 10 ◦C/min. The sample mass used was var-
ied between 0.005 and 0.01 g. Scanning electron microscopy (SEM)
images were recorded using a Philips XL-30 instrument coupled to
an energy dispersion unit using an EDX Link Analytical QX-20000
system at an accelerator voltage of 5 kV. The samples were mounted
on a copper stub with conductive carbon sticky tape. A thin (ca.
The N-HCS samples obtained at 800, 850 and 900 ◦C were stud-
ied by thermogravimetric analysis under air. Fig. S1b displays the
TGA curves of these N-HCS samples. The sample prepared at the
higher temperature showed the highest stability (Fig. S1). This data
also correlates with the graphitic nature of the carbon as deter-
mined by Raman spectroscopy (Fig. S2).
3.1.2. Effect of CVD reaction time on the shell thickness of N-HCSs
The silica spheres (d = 1100 nm 100) were placed in a CVD oven
heated at 900 ◦C (under N2) while acetonitrile was passed over the
spheres for different time periods to study the effect of time on
the sphere size. The shell thickness and size as determined from
SEM and TEM images increased continuously and the resultant car-
bon/silica composites after 1–4 h CVD time gave N-HCSs with shell
thicknesses of 14, 20, 35 and 45 5 nm respectively (Fig. 1). The
samples studied at the different times had a N content of 8.4 (1 h),
8.5 (2 h), 8.6 (3 h) and 8.6 (4 h) wt% N respectively. Thus, the N con-

I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
33
Table 1
BET surface areas and pore parameters of the N-HCS materials produced at different temperatures.
Sample
Temperature ◦
C
Spheres size nmb
SBETm2/g
Pore volume cm3g−1
SiO2
SiO2
SiO2
SiO2@ C
SiO2@ C
SiO2@ C
N-HCS800-4
N-HCS850-4
N-HCS900-4
Pt/N-HCS900−4-120a
800
850
900
800
850
900
800
850
900
–
800
755
700
825
800
750
–
800
750
750
2.0
1.4
1.6
1.4
1.7
0.7
59
55
47
35
0.03
0.01
0.02
0.01
0.02
0.01
0.30
0.28
0.22
0.18
a
Pt/N-HCS900-4-120: pre-treated in H2 (5%)/Ar (100 mL) at 120 ◦C for 2 h.
50 nm.
b
Fig. 1. TEM images of N-HCSs obtained from 1100 nm silica spheres after different CVD times for (a) N-HCS900-1, (b) N-HCS900-2, (c) N-HCS900-3 and (d) N-HCS900-4.
centration did not vary significantly with reaction time suggesting
that the N was deposited in a homogeneous process into the carbon
sphere.
core-shell structures to the hollow structures leads to a successful
enhancement in the carbon surface area. Thus the N-HCSs should
provide a better catalyst support material (allowing for enhanced
metal particle dispersion) relative to the SiO2@C material.
3.1.3. Structural parameters for the N-HCS material
Fig. 2 depicts the wide-angle diffraction pattern of samples that
were carbonized at different temperatures. The diffraction peaks
at 2␪ = 26 and 43◦ (JCPDS Card File, No. 41–1487) corresponded
to (0 0 2) and (1 0 0) reflections from the graphite phase. For the
N-HCS800 sample, the (0 0 2) diffraction peak is relatively low in
intensity and broad in shape. This suggests that the sample pos-
sesses a low degree of graphitization. This result is consistent with
the TEM images (Fig. 1a) where broken N-HCSs were observed. As
the temperature of carbonization increased from 800 to 900 ◦C, the
sharpness of the carbon 002 diffraction peak was enhanced, indi-
cating the positive effect of increasing carbonization temperature
The nitrogen sorption data for silica spheres and silica/carbon
composites after 4 h CVD at 800–900 ◦C gave low BET surface area
values between 1 and 2 m2/g. After silica etching from the sil-
ica/carbon composite, an increase in surface area was noted. The
Brunauer-Emmett-Teller (BET) surface areas of N-HCS800-4, N-
HCS850-4 and N-HCS900-4 materials were 59, 55 and 47 m2/g
respectively (Table 1). The decrease in BET value with increasing
CVD temperature was due to the formation of a more porous car-
bon (N-doped) shell. Further, the carbon pieces and broken spheres
formed at the lower synthesis reaction temperatures would also
give a higher surface area material. Thus the conversion of the

34
I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
Fig. 2. XRD patterns for N-HCSs (prepared at different temperatures) and the Pt/N-HCS catalysts.
on the formation of a more highly graphitic carbon structure. The
shell thickness also increased with reaction temperature.
N and Pt are shown in Fig. S4. The XPS N1s spectra can be decon-
voluted into three peaks centered at 401.5 (22%), 400.1 (48%) and
398.7 (30%) eV respectively assigned to quaternary, pyrollic and
pyridinic-N species [54,55]. It is not known if all, or some, of these
N atoms are the key species that lead to the N-Pt interactions that
are responsible for the enhanced Pt dispersion on the N-HCSs.
3.1.4. Pt loaded N-HCSs
Pt was loaded onto the N-HCSs and the Pt/N-HCS materials
were analyzed. The Pt metal content was determined by ICP-OES
for Pt/N-HCS900-4-120 and shown to be 2.48%. A decrease in sur-
face area was observed for Pt/N-HCS as compared with the N-HCS
samples (Table 1).
The XRD patterns were recorded on the Pt supported materials.
The data revealed that the Pt metallic phase (JCPDS file No-01-1194)
was observed for Pt/N-HCSs. Thus the reduction process in Ar (5%
H2) atmosphere at 120 ◦C (Pt/N-HCS900-4-120) and 300 ◦C (Pt/N-
HCS900-4 300) respectively was sufficient to generate the Pt metal
particles (Fig. 2). The average crystallite size of the Pt nanoparti-
cles in the Pt/N-HCS900-4-120 and Pt/N-HCS900-4-300 catalysts
were 2.8 nm and 4.8 nm as estimated by using the Scherrer equation
(Fig. 2) [53].
TEM images of the samples show spherical hollow carbon
spheres with Pt particles with controlled size and distribution;
a TEM image of Pt/NHSC900-4-120 is shown in Fig. 3. The well
dispersed Pt nanoparticles had a narrow particle-size distribution
centered at d = 2.8 0.4 nm (consistent with data determined from
the XRD analysis).
Many studies have been performed in which Pt has been bound
to a range of C supports and in these studies functionalization of the
surface with acid, hydroxyl and ketonic groups has been reported
to provide the binding sites for Pt [10,36–48]. The functionalised
support generated small particles and this impacted on the selec-
tivity and activity of the Pt. Here we have shown (as seen by others)
that doping of carbon can also give a good metal dispersion for the
Pt metal particles.
Pt XPS data shows two distinct sharp Pt (4f7/2) and Pt (4f5/2
)
peaks around 71.7 and 76.2 eV respectively for Pt/N-HCS900-4-
120 (Fig. S4). These are typical values for Pt(0), indicating that the
deposited Pt (after H2 treatment) is in the metallic state [56].
3.2. Catalytic activity tests
The various Pt loaded N-HCS supports (prepared at different
temperatures and after different synthesis times) were studied as
catalysts for the cinnamaldehyde reduction reaction at 80 ◦C and
30 bar. The results of the catalytic studies (Table 2) show that Pt on
the different N-HCS supports leads to similar high conversions with
similar good selectivities [Pt/N-HCS800-4-120 (90%/75%), Pt/N-
HCS850-4-120 (96%/76%), Pt/N-HCS900-1-120 (91%/78%)] and
Pt/N-HCS900-2-120 (98%/79%)] that increased slightly with the
N-HCS carbon deposition temperature and time. These results sug-
gested the use of a high carbon deposition temperature and a long
reaction time to investigate a set of catalysts in more detail i.e. the
Pt/N-HCS900-4-120 catalyst.
Data
were
collected
at
different
temperatures
(25 ◦C/40 ◦C/60 ◦C/80 ◦C) and pressures (10 bar/20 bar/30 bar)
and are plotted in Table 3 (see also Fig. 4). It is clearly seen that
increasing both the pressure and reaction temperature shifts
the cinnamaldehyde transformation reaction to higher conver-
sion/selectivity. Pressure has less effect on the selectivity to
cinnamyl alcohol than temperature. For example at 60 ◦C the con-
version changed from 65 to 87% and the selectivity from 71% to 87%
as the pressure changed from 10 bar to 30 bar. In contrast as the T
The XPS survey scan of Pt/N-HCS900-4-120 showed signals for
the elements C, N, O and Pt (Fig. S3) and the expanded spectra for

I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
35
Fig. 3. TEM images of platinum nanoparticles on the Pt/N-HCS900-4-120 catalyst.
Table 2
The effect of catalyst synthesis parameters on the conversion of cinnamaldehyde and formation of cinnamyl alcohol a
.
Catalysts
Temperature ◦
C
H2(30 bar)
Select (%)
Conver(%)
A
B
C
Pt/N-HCS900-4-120
Pt/N-HCS900-4-300
Pt/N-HCS800-4
80
80
80
80
80
80
80
99
85
90
96
91
98
80
97
95
75
76
78
79
93
1
3
2
2
9
10
10
11
2
16
14
12
10
5
Pt/N-HCS850-4
Pt/N-HCS900-1-120
Pt/N-HCS900-2-120
Pt/HCS900-4-120
a
Reaction conditions: cinnamaldehyde = 5 mL, catalyst weight = 0.2 g, solvent = isopropanol (30 mL), P = 30 bar; A = cinnamyl alcohol; B = 3-phenylprionaldehyde; C = 3-
phenyl-1-propan-1-ol).
Table 3
The effect of temperature and pressure on the conversion of cinnamaldehyde and formation of cinnamyl alcohola.
Temperature ◦
C
H2 (10 bar)
(%)
Convera select (%)
B
H2 (20 bar)
(%)
Conver select (%)
B
H2 (30 bar)
(%)
Conver select (%)
B
A
C
A
C
A
C
25
40
60
80
15
41
65
70
40
60
71
78
38
30
18
13
22
15
11
7
20
50
72
83
45
68
78
85
35
20
16
10
20
12
6
30
61
82
99
50
76
87
97
26
8
5
23
6
8
5
1
2
a
Reaction conditions: cinnamaldehyde = 5 mL, catalyst weight = 0.2 g, solvent = isopropanol (30 mL), A = cinnamyl alcohol; B = 3-phenylprionaldehyde; C = 3-phenyl-1-
propan-1-ol).
changed from 25 ◦C to 80 ◦C the conversion changed from 15 to 70%
i) The catalyst Pt/N-HCS900-4-300 (i.e. the same catalyst but
heated at 300 ◦C for 2 h) gave a conversion of 85% cinnamalde-
hyde with 95% selectivity to cinnamyl alcohol. The lower activity
is related to the presence of the larger (4.8 nm vs 2.8 nm) Pt
nanoparticles.
and the selectivity from 40% to 78% (at 10 bar). Importantly, at the
higher pressures/temperatures little over reaction occurs to give
3-phenyl-1-propan-1-ol. Thus the formation of the by-products,
3-phenyl-1-propan-1-ol and 3-phenylpropionaldehyde was only
favoured at low temperature and pressure. This is consistent with
the model discussed by Durndell et al. [37].
ii) A catalytic reaction was performed with Pt/HCS900-4-120 (80 ◦C
and 30 bar) and the data compared with Pt/N-HCS900-4-120
(both had similar particle sizes from TEM analysis). A lower con-
version (80%) and selectivity (93%) for the undoped supported
catalyst are noted (Fig. 5) indicating a measurable effect due
To understand the origin of the activity/selectivity data further,
other catalysts were tested.

36
I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
Fig. 4. Selectivity to cinnamyl alcohol (CA) and other (Oth) products as a function of temperature and pressure in the presence of Pt/N-HCS900-4-300 [reaction conditions:
cinnamaldehyde = 5 mL, catalyst weight = 0.2 g, solvent = isopropanol (30 mL), Oth = 3-phenylprionaldehyde and3-phenyl-1-propan-1-ol].
to the presence of the N atoms. This could be due to a surface
‘polarity’ effect associated with the N atoms [37].
A comparison with other reports on Pt supported on various
other types of carbon reveals that excellent activity/selectivity
results can also be achieved when the carbons are surface func-
tionalized [36–49]. Our study reveals that N doping of the carbon
can also lead to the selectivities/conversions described by others.
Studies have been reported that suggest that the binding energies of
Pt increase when Pt is supported on nitrogen doped carbon as com-
pared with Pt supported on undoped carbon [57,58]. Thus, provided
the surface area of the carbon is large enough (achieved in our case
by removing the silica from the NHCSs) N doping provides an alter-
native to functionalization to generate good Pt-carbon interactions
and resulting good catalyst conversions.
3.3. Catalyst stability test
Deactivation is an important factor when studying the long-
term performance of a catalyst. Deactivation can occur by a number
of processes including fouling, coking and leaching. At low temper-
atures, hydrocarbon coking reactions do not occur and any deposits
Fig. 5. The conversion of cinnamaldehyde and selectivity to cinnamyl alcohol (CA)
are mainly derived from adsorbed reaction products and reactants.
Catalyst recycling experiments were performed with the Pt/N-
HCS900-4-120 catalyst that was studied at 80 ◦C and 30 bar for
2 h. Fig. 5 shows that after reuse of the Pt/N-HCS900-4-120 twelve
times, the conversion of cinnamaldehyde decreased from >99% to
90%, while the selectivity to the formation of cinnamyl alcohol
over Pt/N-HCS900-4-120 and Pt/HCS900-4-120 catalysts before use (fresh), and over
12 recycles [reaction conditions: cinnamaldehyde = 5 mL, catalyst weight = 0.2 g, sol-
vent = isopropanol (30 mL), reducing agent = H2; P = 30 bar; temperature = 80 ◦C].

I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
37
doping the HCSs, the advantage is readily seen in the catalyst re-
use experiments where leaching is seen to be more pronounced
from the undoped carbon.
v) The effect of pressure and temperature on the catalyst conver-
sion/selectivity is seen to be consistent with the many other
studies reported by others. The model proposed by Durndell
et al. [37] involving ‘surface crowding’ is consistent with our
data.
vi) The actual data compare with the best data reported to date for
the selective conversion of cinnamaldehyde to cinnamyl alcohol
The study reveals that the Pt catalyst deposited on Pt/N-HCS900-
4-120 showed good stability for the cinnamaldehyde reduction.
Thus, the key factor revealed in these studies is that the Pt is
well dispersed (small particles) and shows minimal leaching from
Pt/N-HCS900-4-120 and is not sensitive to the formation of any
carbonaceous residues deposited on the surface of the active metal
species.
Fig. 6. The conversion of cinnamaldehyde and selectivity to cinnamlyl alco-
hol (CA) over Pt/N-HCS900-4-120 before use (fresh), and after 3rd recycle.
4. Conclusion
a
The filtered solution of the catalyst (hot filtrate; 3rd recycle) is also shown.
The conversions are shown afer different time intervals [reaction condi-
tions: cinnamaldehyde = 5 mL, catalyst weight = 0.2 g, solvent = isopropanol (30 mL);
P = 30 bar; temperature = 80 ◦C].
The data from the various studies indicate that the carbon shell
deposited on the silica core is influenced by the reaction tempera-
ture and the reaction time. It appears that the reaction time has little
influence on the C/N ratio. Indeed once the reactant has deposited
on the silica the C/N ratio is not influenced further by the template.
In the CVD reaction when using low reaction temperatures or short
reaction times carbon pieces, broken spheres and deformed thin
carbon shells were observed. At higher CVD temperatures (900 ◦C)
and longer reaction times (4 h) the shell thickness increased and
the N-HCSs had a superior graphitic nature.
XPS data recorded on Pt/HCS900-4-120 revealed that the N was
inserted into the carbon with a high nitrogen content (8.6 wt%) and
in the form of a range of species (quaternary, pyrollic, pyridinic).
Well dispersed Pt nanoparticles with an average size of 2.8 nm
were deposited on the N-HCS900-4 support. XRD and XPS stud-
ies revealed Pt was present in the metallic phase. Temperature
and pressure studies were performed on Pt/HCS900-4-120 and the
temperature effect was found to be more important than pressure
in controlling the conversion and selectivity. The catalyst showed
excellent conversion, recyclability and selective formation to the
unsaturated cinnamyl alcohol. At 30 bar and a temperature of 80 ◦C
excellent cinnamaldehyde conversion and selectivity to the unsat-
urated alcohol was achieved.
remained constant (>97%) (Fig. 5). Chemical analysis of the used
catalysts revealed that very little (if any) Pt (<2%) leached from Pt/N-
HCS900-4-120. Indeed the Pt content was reduced from 2.49 wt%
to ca. 2.42 wt% after use.
Furthermore, a catalytic reaction was carried out on the filtered
solution (2 h at 30 bar and 80 ◦C) after separation of the catalyst
from the reaction mixture (Fig. 6). The Pt/N-HCS900-4-120 filtrate
gave < 4% conversion to cinnamyl alcohol. The data again suggests
minimal, if any, Pt leaching. Indeed the main reason for the drop in
conversion relates to the physical loss of the catalyst with repeated
use.
Catalyst recycling studies were also performed on Pt/HCS900-4-
120. The data reveal that while the selectivity remained constant at
ca. 90% the conversion dropped with time. This drop is associated
with catalyst leaching suggesting that the N doped support binds
more strongly to the Pt particles.
3.4. Summary: N-HCSs as a support for the Pt catalyzed reaction
The study shows that N doping provides a good alternative to the
use of carbon functionalization for binding Pt to carbon surfaces.
Many studies have appeared on the use of Pt supported catalysts
for the cinnamaldehyde hydrogenation reaction. Our results are
consistent with the many studies but provide added information
on the process/reaction.
Acknowledgements
We thank the NRF, the Universities of the Witswatersrand and
Johanesburg, and the DST-NRF Centre of Excellence in Strong Mate-
rials for financial support.
i) The use of carbon allows for excellent TEM/SEM contrast studies
to evaluate the Pt particle size. While this was not the focus
of this study, the particle size/conversion data are consistent
with the reaction being dependent on the number of Pt atoms
available for reaction. The selectivity is not affected by a change
in particle size from 2.8 to 4.8 nm
ii) As the synthesis of the N-HCSs does not require the use of a
catalyst, no contamination by residual metals (e.g. used to make
CNTs) occurs and acid treatment to functionalize the C surface
is not needed.
iii) The hollow sphere provides a means of making a carbon sur-
face with a higher surface area than found for a solid sphere
(SiO2@C). This should lead to a better Pt dispersion.
iv) The role of the N doping of the HCS was to provide stronger
support binding sites to the Pt. While the catalytic data shows
some enhancement in activity (less relating to selectivity) on
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.apcata.2016.02.
025.
References
[1] P. Chaturbedy, S. Chatterjee, R.B. Selvi, A. Bhat, M. Kavitha, V. Tiwari, A.B. Patel,
T.K. Kundu, T.K. Maji, M. Eswaramoothy, J. Mater. Chem. B 1 (2013) 939–945.
[2] A.A. Deshmukh, S.D. Mhlanga, N.J. Coville, Mater. Sci. Eng. R Rep. 70 (2010)
1–28.
[3] X. Feng, C. Mao, G. Yang, W. Hou, J.-J. Zhu, Langmuir 22 (2006) 4384–4389.
[4] M. Inagaki, Carbon 35 (1997) 711–713.

38
I. Nongwe et al. / Applied Catalysis A: General 517 (2016) 30–38
[5] P. Serp, R. Feurer, P. Kalck, Y. Kihn, J. Faria, J. Figueiredo, Carbon 39 (2001)
621–626.
[34] C. Letizia, J. Cocchiara, J. Lalko, A. Lapczynski, A. Api, Food Chem. Toxicol. 43
(2005) 837–866.
[6] P. Serp, J.L. Figueiredo, Carbon Materials for Catalysis, John Wiley & Sons, 2009.
[7] Y. Xia, B. Gates, Y. Yin, Y. Lu, Adv. Mater. 12 (2000) 693–713.
[8] M.-M. Titirici, M. Antonietti, Chem. Soc. Rev. 39 (2010) 103–116.
[9] M.-M. Titirici, A. Thomas, M. Antonietti, Adv. Funct. Mater. 17 (2007) 1010.
[10] H.C. Zeng, J. Mater. Chem. 21 (2011) 7511–7526.
[35] M.A. Vannice, B. Sen, J. Catal. 115 (1989) 65–78.
[36] K. Chizari, I. Janowska, M. Houllé, I. Florea, O. Ersen, T. Romero, P. Bernhardt,
M.J. Ledoux, C. Pham-Huu, Appl. Catal. A 380 (2010) 72–80.
[37] L.J. Durndell, C.M. Parlett, N.S. Hondow, M.A. Isaacs, K. Wilson, A.F. Lee, Sci.
Rep. 5 (2015).
[11] Q. Zhang, W. Wang, J. Goebl, Y. Yin, Nano Today 4 (2009) 494–507.
[12] J. Liu, S.Z. Qiao, J.S. Chen, X.W.D. Lou, X. Xing, G.Q.M. Lu, Chem. Commun. 47
(2011) 12578–12591.
[13] F. Su, X. Zhao, Y. Wang, L. Wang, J.Y. Lee, J. Mater. Chem. 16 (2006) 4413–4419.
[14] P. Valle-Vigón, M. Sevilla, A.B. Fuertes, Mater. Lett. 64 (2010) 1587–1590.
[15] X.-H. Li, M. Antonietti, Chem. Soc. Rev. 42 (2013) 6593–6604.
[16] Y. Li, T. Li, M. Yao, S. Liu, J. Mater. Chem. 22 (2012) 10911–10917.
[17] Z. Zhang, R. Zhang, C. Li, L. Yuan, P. Li, L. Yao, S. Liu, Electroanalysis 24 (2012)
1424–1430.
[18] F. Braghiroli, V. Fierro, M. Izquierdo, J. Parmentier, A. Pizzi, A. Celzard, Carbon
50 (2012) 5411–5420.
[19] A. Chen, Y. Yu, H. Lv, Y. Wang, S. Shen, Y. Hu, B. Li, Y. Zhang, J. Zhang, J. Mater.
Chem. A 1 (2013) 1045–1047.
[38] Z. Guo, Y. Chen, L. Li, X. Wang, G.L. Haller, Y. Yang, J. Catal. 276 (2010) 314–326.
[39] M.S. Ide, B. Hao, M. Neurock, R.J. Davis, ACS Catal. 2 (2012) 671–683.
[40] B.F. Machado, S. Morales-Torres, A.F. Pérez-Cadenas, F.J. Maldonado-Hódar, F.
Carrasco-Marín, A.M. Silva, J.L. Figueiredo, J.L. Faria, Appl. Catal. A 425 (2012)
161–169.
[41] W.O. Oduro, N. Cailuo, K.M.K. Yu, H. Yang, S.C. Tsang, Phys. Chem. Chem. Phys.
13 (2011) 2590–2602.
[42] J. Serrano-Ruiz, A. López-Cudero, J. Solla-Gullón, A. Sepúlveda-Escribano, A.
Aldaz, F. Rodríguez-Reinoso, J. Catal. 253 (2008) 159–166.
[43] Z. Sun, Z. Rong, Y. Wang, Y. Xia, W. Du, Y. Wang, RSC Adv. 4 (2014) 1874–1878.
[44] Z. Tian, C. Liu, Q. Li, J. Hou, Y. Li, S. Ai, Appl. Catal. A 506 (2015) 134–142.
[45] M.L. Toebes, T.A. Nijhuis, J. Hájek, J.H. Bitter, A.J. Van Dillen, D.Y. Murzin, K.P.
de Jong, Chem. Eng. Sci. 60 (2005) 5682–5695.
[20] Y. Liao, L. Gao, X. Zhang, J. Chen, Mater. Res. Bull. 47 (2012) 1625–1629.
[21] Y. Xia, Z. Yang, R. Mokaya, J. Phys. Chem. B 108 (2004) 19293–19298.
[22] Y.-M. Lu, H.-Z. Zhu, W.-G. Li, B. Hu, S.-H. Yu, J. Mater. Chem. A 1 (2013)
3783–3788.
[23] J. Jiang, Q. Gao, Z. Zheng, K. Xia, J. Hu, Int. J. Hydrogen Energy 35 (2010)
210–216.
[24] L. Guo, J. Zhang, Q. He, L. Zhang, J. Zhao, Z. Zhu, W. Wu, J. Zhang, J. Shi, Chem.
Commun. 46 (2010) 7127–7129.
[25] X. Zhang, W. Zhang, Z. Yang, Z. Zhang, Front. Chem. Sci. Eng. 6 (2012) 246–252.
[26] S. Han, Y. Yun, K.-W. Park, Y.-E. Sung, T. Hyeon, Adv. Mater. 15 (2003)
1922–1925.
[27] X. Lai, J.E. Halpert, D. Wang, Energy Environ. Sci. 5 (2012) 5604–5618.
[28] Y. Wang, F. Su, J.Y. Lee, X. Zhao, Chem. Mater. 18 (2006) 1347–1353.
[29] K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Science 323 (2009) 760–764.
[30] V. Ravat, I. Nongwe, N.J. Coville, ChemCatChem 4 (2012) 1930–1934.
[31] V. Ravat, I. Nongwe, R. Meijboom, G. Bepete, N.J. Coville, J. Catal. 305 (2013)
36–45.
[32] P. Claus, Top. Catal. 5 (1998) 51–62.
[33] M.J. Taylor, L.J. Durndell, M.A. Isaacs, C.M. Parlett, K. Wilson, A.F. Lee, G.
Kyriakou, Appl. Catal. B 180 (2016) 580–585.
[46] H. Vu, F. Gonc¸ alves, R. Philippe, E. Lamouroux, M. Corrias, Y. Kihn, D. Plee, P.
Kalck, P. Serp, J. Catal. 240 (2006) 18–22.
[47] Y. Wan, Y.-L. Min, S.-H. Yu, Langmuir 24 (2008) 5024–5028.
[48] B.-H. Zhao, J.-G. Chen, X. Liu, Z.-W. Liu, Z. Hao, J. Xiao, Z.-T. Liu, Ind. Eng. Chem.
Res. 51 (2012) 11112–11121.
[49] R. Zheng, M.D. Porosoff, J.L. Weiner, S. Lu, Y. Zhu, J.G. Chen, Appl. Catal. A 419
(2012) 126–132.
[50] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62–69.
[51] J.X. McDermott, J.F. White, G.M. Whitesides, J. Am. Chem. Soc. 98 (1976)
6521–6528.
[52] H.A. Tayim, M. Kharboush, Inorg. Chem. 10 (1971) 1827–1828.
[53] A. Patterson, Phys. Rev. 56 (1939) 978.
[54] M.-C. Huang, H. Teng, Carbon 41 (2003) 951–957.
[55] X. Jin, V.V. Balasubramanian, S.T. Selvan, D.P. Sawant, M.A. Chari, G.e.Q. Lu, A.
Vinu, Angew. Chem. 121 (2009) 8024–8027.
[56] G. Zhang, D. Yang, E. Sacher, J. Phys. Chem. C 111 (2007) 565–570.
[57] L. Jia, D.A. Bulushev, O.Y. Podyacheva, A.I. Boronin, L.S. Kibis, E.Y. Gerasimov, S.
Beloshapkin, I.A. Seryak, Z.R. Ismagilov, J.R. Ross, J. Catal. 307 (2013) 94–102.
[58] X. Xie, J. Long, J. Xu, L. Chen, Y. Wang, Z. Zhang, X. Wang, RSC Adv. 2 (2012)
12438–12446.