Shape-controlled synthesis and catalytic behavior of supported platinum nanoparticles

Article abstract of DOI:10.1055/s-0028-1087921

A 5 wt% Pt/C catalyst with around 20 nm cubic platinum particles was prepared through a conventional preparation method (i.e., precursor impregnation, reduction, and calcination) by choosing to use hydrophobic solvent in the impregnation procedure. The pr

Full text of DOI:10.1055/s-0028-1087921

LETTER
595
Shape-Controlled Synthesis and Catalytic Behavior of Supported Platinum
Nanoparticles
S
ynthesis of Supp
h
orted
P
latin
a
um
N
anopa
n
rticles gkun Liu,* Zhenhua Zhou, Zhihua Wu, Martin Fransson, Bing Zhou
Headwaters Technology Innovation LLC 1501, New York Avenue, Lawrenceville, NJ 08648, USA
Fax +1(609)3949602; E-mail: cliu@htinj.com
Received 7 April 2008
metal loading. For higher metal loadings of supported cat-
alyst large volumes of water need to be carefully removed
from the system. This process in not only time consuming,
but also it increases the risk of nanoparticle aggregation
Abstract: A 5 wt% Pt/C catalyst with around 20 nm cubic platinum
particles was prepared through a conventional preparation method
(i.e., precursor impregnation, reduction, and calcination) by choos-
ing to use hydrophobic solvent in the impregnation procedure. The
precise nanoparticles shape control can be achieved by using this and loss of morphology control. This method has been
preparation method as compared to the commonly used precolloid
widely used in recent studies, but it is not suitable for
method. These 20 nm cubic Pt particles showed high selectivity (ca.
99.4%) for the hydrogenation of o-chloronitrobenzene.
large-scale supported-catalyst production. Conventional
methods, such as impregnation of supports with metal
precursors with subsequent reduction and calcination are
much preferred for a large-scale production and widely
Key words: platinum nanoparticles, shape control, supported cata-
lyst, hydrogenation, chloronitrobenzene
applied in chemical industry for preparation of supported
metal catalysts. However, because of the aforesaid rea-
The physical and chemical properties of nanosized parti-
cles are greatly affected not only by their size, but also by
their shape. Great efforts have been made to better under-
stand the relationship between the structure of a supported
metal and its catalytic reactivity.1 It is clear that signifi-
cant improvements in catalytic activity and selectivity can
be achieved by controlling the morphological properties
of supported metal particles.2
sons it was thought that shape control would be mini-
mal.3,10 This study first shows that through the
conventional method, shape control can be accomplished
on high metal-loading supported nanoparticles by careful-
ly adjusting the reaction parameters.
Platinum nanoparticles with a 5 wt% loading supported on
carbon black was prepared using the following steps: (i)
carbon was soaked in methanol, (ii) the soaked carbon
was impregnated with a toluene solution of Pt acetylaceto-
nate, (iii) the excess toluene was removed using a rotary
evaporator, (iv) the resulting carbon material was dried
and reduced in hydrogen at 100 °C. The HRTEM charac-
terization of this carbon-supported catalyst (Figure 1)
shows that very uniform Pt microcrystals are embedded in
the support whereas a closer look at the nanoparticles
shows high defined cubic microcrystals. The size of the
supported Pt crystallites was estimated to be 14.5 nm from
the broadening of the Pt(111) peak in X-ray diffraction at
2q = ca. 40° (Figure 2). As can be seen in Figure 1, the
dTEM of 20 nm is in fairly good agreement with dXRD of
14.5 nm.
Many studies have focused on shape-controlled synthesis
of metal nanoparticles. However, morphological control
is still difficult to obtain, especially for supported metal
particles, since the preparation process governs metal
nanocrystal formation, growth, and complex interaction
between metal particles and supports. Therefore, the final
characteristic of the material is extremely sensitive to the
selected preparation method.3 A convenient method to
minimize preparation and supports effects on the metal
morphology is to first prepare a well-defined metal nano-
particle colloid, and then deposit the metal nanoparticles
on supports.4
Normally, well-defined metal nanoparticle colloids are
prepared by reduction of a Pt source in aqueous solution
in the presence of protective agents, such as small organic
compounds with multifunctional groups (ethylene glycol,
citric acid, oleic acid, etc.) or linear hydrophilic functional
polymers [such as poly (vinylpyrrolidone), poly (N-iso-
propylacrylamide), poly acrylic acid, etc.].5–9 In order to
obtain a stable metal colloid solution, the concentration of
metal nanoparticles is normally very low (10–3 to 10–6 M).
After the impregnation of the support with the colloidal
solution, the resulting supported catalyst will have a low
It is commonly believed that the final morphology of the
fcc nanocrystals is dependent upon the R value, defined as
the relative growth rate along the <100> direction with
respect to that of the <111> (R valve for cubes, cubocta-
hedra, and octahedral is 0.58, 0.87, and 1.73, respective-
ly).11 In <100> crystal facet, Pt atom has less adjacent
atoms than that in facet <111> so that surface Pt atom in
crystal facet <100> is more electron deficient than that in
<111>. We assume that hydrogen molecules seem to be
preferentially adsorbed on the more electron-deficient
<100> Pt nuclei than on facets <111> during reduction.
The Pt(II) ions approached to <100> Pt-nuclei facet will
be reduced first and so that the growth rate along <100>
SYNLETT 2009, No. 4, pp 0595–0598
x
x.
x
x.
2
0
0
9
Advanced online publication: 16.02.2009
DOI: 10.1055/s-0028-1087921; Art ID: S03108ST
© Georg Thieme Verlag Stuttgart · New York

596
C. Liu et al.
LETTER
Figure 2 The XRD characterization of Cubic-Pt
The key factors in controlling high-loading Pt-size parti-
cles are first the choice of solvent pair (methanol–toluene)
and second the extent of the hydrogenation reduction step.
Toluene, a hydrophobic solvent, is not usually chosen for
carbon impregnation because of the hydrophilic groups on
the carbon black surface. However, in our synthesis, the
repulsion between toluene and the carbon black surface
was thought to help the formation of isolated Pt acetyl-
acetonate microcrystals on the carbon surface upon slow
evaporation off the excess toluene. These platinum acetyl-
acetonate microcrystals behaved as the isolated Pt source
during hydrogenation reduction step. As comparison, ex-
periments by using hydrophilic solvents such as ethanol or
THF indicated that the dominant Pt species were amor-
phous with a 2–5nm particle size. Similar results were re-
ported by G. Jacobs who claims that supported amorphous
Pt nanoparticles were obtained by immersing calcined po-
rous materials in an aqeous solution of Pt acetylaceto-
nate.12 In addition to the choice of solvent, the
hydrogenation step is also a key step to product morphol-
ogy control. Slow reduction of platinum acetylacetonate
in a hydrogen–nitrogen (5% H2) gas mixture at 100 °C is
beneficial for growth of uniform crystal shape and avoid-
ing Pt acetylacetonate heat decomposition. Fast reduction
of platinum acetylacetonate by use of pure hydrogen or at
high temperature will increase ununiformity of crystal
size and shape. We assume that when Pt(II) ion is reduced
in pure hydrogen or at high temperature, too much excess
hydrogen and high temperature will distort the H2 prefer-
entially <100> facet absorption.
These cubic Pt nanocrystals supported on carbon, marked
as Cubic-Pt, exhibit extra performance in catalytic hydro-
genation of halonitrobenzene when compared to two other
5% Pt/C catalysts Amorp-Pt and DG-Pt. Amorp-Pt was
prepared as followed: the first step involved the prepara-
tion of a Pt colloidal solution in which EDTA was used as
a stabilizing agent. The second step involved the impreg-
nation of the support with the colloid solution followed by
the final step which was reduction. The TEM character-
ization for the catalyst Amorp-Pt showed almost no Pt
crystal lattice with loose aggregate size around 13–20 nm.
These results were corroborated by the XRD results in
which no apparent Pt diffraction pattern was observed.
Figure 1 TEM image of 5 wt% Pt /C, showing a uniform size dis-
tribution of the supported particles. High-magnification image high-
lighting the Cubic shape of the Pt nanocrystallites.
will be enhanced and/or the growth along <111> will be
suppressed.
Synlett 2009, No. 4, 595–598 © Thieme Stuttgart · New York

LETTER
Synthesis of Supported Platinum Nanoparticles
597
This indicated that Pt was in an amorphous form without tration of hydroxylamine-chlorobenzene intermediates
definite particle shape. DG-Pt is a commercial product during the hydrogenation of o-chloronitrobenzene. In the
from Degussa AG which composition is 5% Pt on carbon case of this cubic Pt catalyst, we believe that the Pt parti-
with bismuth doped as promoter.
cles located at lateral boundaries has less adjacent atoms
so that it is more electron deficient than the atoms in the
cubic faces and prone to absorb hydrogen atoms easily.
Therefore, we believe that most hydrogenation may hap-
pen at the lateral boundary area, thus results the highest
selectivity.
Hydrogenation of o-chloronitrobenzene to o-chloro-
aniline is a tedious process and often yields many dechlo-
ride byproducts during the reaction.13 In this study,
hydrogenation catalyzed by these three catalysts yielded
aniline as the detectable dechloride byproduct and o-
hydroxylamine-chlorobenzene as the main intermediate Efforts to prepare smaller size cubic Pt crystals on carbon
of an incomplete reaction (Scheme 1). The results as as well as to optimize reaction parameters are ongoing.
shown in Table 1 indicated that catalyst Cubic-Pt took 10 We believe that not only smaller size cubic Pt will in-
hours to almost complete the reaction, meanwhile cata- crease hydrogenation rates but also more lateral boundary
lysts Amorp-Pt and DG-Pt only needed about 6 hours. Pt atoms will result in a higher selectivity.
The product analysis showed that Cubic-Pt yielded the
lowest concentration of dechloride byproduct aniline
(0.28% in 10 h) as compared to catalyst Amorp-Pt
(4.83% in 6 h) and DG-Pt (0.48% in 6 h).
Preparation of 20 nm Cubic 5% Pt/C Catalyst (Cubic-Pt)
Activated carbon (6.0 g) was soaked with methanol (30 mL) for 12
h. After removal of excess MeOH, the carbon solid was dispersed
in toluene (60 mL) and platinum(II) acetylacetonate (0.605 g) in tol-
uene (20 mL) was added. The total suspension was kept rotated on
rotary evaporator. Toluene was slowly evaporated out in 0.03 bar
vacuum and at 85 °C. The activated carbon with platinum complex
was dried in vacuum oven at 80 °C for 6 h and then reduced in 5%
H2 in N2 flow at 100 °C for 4 h. After that, gas flow was switched
to nitrogen and let the system cool to r.t.
Cl
Cl
Cl
Pt/C
+
+
H2
+
EtOH
NO2
NH2
NH2
NHOH
AN
o-CAN
o-HOCAN
Scheme 1 Hydrogenation of o-chloronitrobenzene
Table 1 Hydrogenation of o-CNB with Pt/C Catalystsa
Preparation of Amorphous 5% Pt/C Catalyst (Amorp-Pt)
Aqueous solution of H2PtCl6 (1.96 g of 25.52% H2PtCl6) was dilut-
ed to 25 mL with distilled H2O. To this solution of EDTA (0.119 g)
in H2O (25 mL) was added. The mixed solution was heated up to
refluxed for 1 h, then cooled down to r.t. with continuous stirring.
Activated carbon (4.75 g) presoaked with MeOH was added to the
solution. The resulted mixture was heated up to 80 °C until all the
free liquid evaporated. The solid was dried at 80 °C for 3 h, and re-
duced in 5% H2 in N2 flow at 100 °C for 4 h. After cooling to r.t.,
the reduced catalyst was washed with warm distilled H2O (4 × 200
mL) to remove of Cl anions and then dried at 80 °C for 3 h.
Catalystsb Time Selectivityc
(h)
Conversion
(%)
o-CAN
AN
o-HOCAN
Cubic-Pt
4.5 66.53
0.13
0.19
0.23
0.28
33.33
13.65
2.67
100
100
100
100
6
86.16
8.5 97.10
10 99.41
0.31
Amorp-Pt 4.5 92.98
94.23
4.33
4.83
2.69
0.94
100
100
6
General Procedure of Hydrogenation of 2-Chloronitrobenzene
A typical hydrogenation procedure was conducted as follows: 2-
chloronitrobenzene catalyst (4.04 g) and proper amount of catalyst
was dispersed in EtOH (60 mL). The suspension was added into a
300 mL stainless-steel autoclave with glass liner equipped with a
mechanically stir blade, a pressure gauge, a gas inlet tube attached
to a hydrogen source, and a cooling circler connected to the temper-
ature controller. The autoclave was purged by nitrogen for three
times. When the temperature stabilized at 25 °C, the vessel was
pressurized to 150 psi with hydrogen. During the reaction mixture
was vigorously stirred and was temporarily stopped for sample col-
lection when no pressure decrease was observed. The product was
analyzed on gas chromatography Agilent 6890 equipped with a FID
detector and Rtx-5 Amine column.
DG-Pt
4.5 90.73
98.94
0.39
0.48
8.88
0.59
100
100
6
a Reaction conditions: 25 °C, 10 bar (H2), o-chloronitrobenzene (4.04
g) in EtOH (60 mL), 5% Pt/C (0.02 g, substrate/catalyst = 5050:1).
b Catalyst, Cubic-Pt: 20 nm cubic 5% Pt/C; Amorp-Pt: amorphous
5% Pt/C; DG-Pt: 5% Pt/C with Bi doping bought from Degussa AG.
d o-CAN = o-chloroaniline, AN = aniline, o-HOCAN = o-hydroxy-
lamine-chlorobenzene. Other byproducts or intermediates with con-
centration lower than 0.10% are exclusive from calculation.
Most researchers agree that the mechanism of hydrode-
halogenation involves an electrophilic attack of cleaved
hydrogen on the adsorbed aromatic halides.14–16 It is be-
lieved that the electron-deficient state of the Pt particles
would weaken the extent of electron feedback from the Pt
particles to the aromatic ring in o-CAN, which would fur-
ther suppress the hydrodechlorination of o-CAN.17 Based
on this principle, efforts to change the electronic proper-
ties of the Pt particle led to the use of Cr, Mn, Fe, Bi, etc.
as the doping metals. This change in the nature of the cat-
alyst gave an increased selectivity or decreased concen-
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett.
Acknowledgment
We gratefully acknowledge the assistance of Dr. Douglas Yates
from Pennsylvania State University for HRTEM characterization
and discussion.
Synlett 2009, No. 4, 595–598 © Thieme Stuttgart · New York

598
C. Liu et al.
LETTER
(10) Yoo, J. W.; Hathcock, D.; El-Sayed, M. A. J. Phys. Chem. A
2002, 106, 2049.
(11) Song, H.; Kim, F.; Connor, S.; Somorjai, G. A.; Yang, P.
J. Phys. Chem. B 2005, 109, 188.
(12) Jacobs, G.; Ghadiali, F.; Pisanu, A.; Borgan, A.; Alvarez,
W. E.; Resasco, D. E. Appl. Catal., A 1999, 188, 79.
(13) Han, X. X.; Zhou, R. X.; Zheng, X. M.; Jiang, H. J. Mol.
Catal. A: Chem. 2003, 193, 103.
References
(1) Gates, B. C. Chem. Rev. 1995, 95, 511.
(2) Balint, I.; Miyazaki, A.; Aika, K.-i. Chem. Commun. 2002,
1044.
(3) Rickard, J. M.; Genovese, L.; Moata, A.; Nitsche, S.
J. Catal. 1990, 121, 141.
(4) Miyazaki, A.; Balint, I.; Aika, K.-i. J. Catal. 2001, 204, 364.
(5) Puntes, V. F.; Krishnan, K. M.; Alivisatos, A. P. Science
2001, 291, 2115.
(6) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.;
El-Sayed, M. A. Science 1996, 272, 1924.
(7) Sun, Y. G.; Xia, Y. N. Science 2002, 298, 2176.
(8) Teranishi, T.; Kurita, R.; Miyake, M. J. Inorg. Organomet.
Polym. 2000, 10, 145.
(14) Coq, B.; Tijani, A.; Dutartre, R.; Figuéras, F. J. Mol. Catal.
1993, 79, 253.
(15) Menini, C.; Park, C.; Shin, E.; Tavoularis, G.; Keane, M. A.
Catal. Today 2000, 62, 355.
(16) Chon, S.; Allen, D. AIChE J. 1991, 37, 1730.
(17) Zhang, J. L.; Wang, Y.; Ji, H.; Wei, Y. G.; Zhang, J.; Wang,
Y.; Ji, H.; Wei, Y.; Wu, N.; Zuo, B.; Wang, Q. J. Catal.
2005, 229, 114.
(9) Miyazaki, A.; Nakano, Y. Langmuir 2000, 16, 7109.
Synlett 2009, No. 4, 595–598 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.