Porous single-crystalline palladium nanoparticles with high catalytic activities

Article abstract of DOI:10.1002/anie.201107376

Palladium's pore cousin: A facile approach is described for the size-controlled preparation of porous single-crystalline Pd nanoparticles. These porous Pd nanoparticles exhibit size-independent catalytic activities for the Suzuki coupling and are more active than commercial Pd/C catalysts. Copyright

Full text of DOI:10.1002/anie.201107376

Angewandte
Chemie
DOI: 10.1002/anie.201107376
Palladium Catalysts
Porous Single-Crystalline Palladium Nanoparticles with High Catalytic
Activities**
Feng Wang, Chuanhao Li, Ling-Dong Sun,* Chun-Hu Xu, Jianfang Wang,* Jimmy C. Yu, and
Chun-Hua Yan*
Catalysis is essential in chemical production, solar energy
harvesting, pollution control, and food processing. Industrial
metal-based catalysts are usually ultrasmall nanoparticles
(NPs) dispersed on supports. Owing to the presence of
multiple components and size scales, investigating the
dependence of the catalytic performance on the composition,
size, and structure is often difficult. In the past 15 years,
advances in nanotechnology have enabled the preparation of
noble-metal NPs with controllable sizes, facets, and compo-
sitions. Applying these NPs for catalysis can not only benefit
our understanding of the structure-determined catalytic
activity and selectivity, but also help in designing better
catalysts.[1–8] For example, judicious control of the preparation
conditions can produce metal NPs with different facets.[2,3,6,7]
Metal NPs with high-index facets are usually more catalyti-
cally active than those with low-index facets. Coating metal
NP catalysts with porous silica shells can improve their high-
temperature stability.[5] To achieve economical catalysis,
catalytic NPs can be integrated with magnetic components,
thus allowing easy purification and recycling.[4] Bilayers of
metal and oxide NPs can be utilized for tandem reactions.[8]
Furthermore, the catalytic performance of metal NPs can
also be improved by control of their morphologies. Many
efforts have been devoted to the control of metal NP
shapes.[9–11] Dendrites have recently received much attention
because they are porous, their branches provide many high-
index facets,[12] and they can be readily separated and
recycled. In the past few years, pure Pt,[13–18] Pt-on-Pd,[12,19–21]
Au/Pd core–Pt shell,[22] and Rh[23] dendritic NPs have been
prepared chemically. Dendritic Pt NPs have been shown to be
more active in ethylene hydrogenation than Pt cubes and
cuboctahedrons.[13] Pt-on-Pd nanodendrites have been found
to be 2.5 times more active than C-supported Pt catalysts and
5 times more active than Pt-black catalysts for O2 reduc-
tion.[12] These results suggest that engineering noble metals
into porous NPs is an attractive approach for achieving high
catalytic performance.
Herein, we describe the facile preparation of porous
single-crystalline Pd NPs of controllable sizes. Rapid growth
was found to be crucial for their formation. The porous Pd
NPs are more catalytically active for the Suzuki coupling
reaction than Pd nanospheres and commercial C-supported
Pd NPs (Pd/C) at both 858C and room temperature. More-
over, they also exhibit remarkable substrate selectivity. We
note that there have been no reports on the room-temper-
ature catalytic activities of Pd nanostructures for the Suzuki
coupling reaction.
The Pd NP samples were prepared by using a seed-
mediated method in aqueous solutions with cetyltrimethyl-
ammonium chloride (CTAC) as the stabilizing agent (see the
Supporting Information for the experimental details). The
sizes of the Pd NPs were relatively uniform, as revealed by
transmission electron microscopy (TEM) imaging (Fig-
ure 1a). High-angle annular dark-field scanning TEM
(HAADF–STEM) imaging shows that the Pd NPs assume
a roughly spherical shape and are porous (Figure 1b). N2-
sorption measurements were performed to determine the
surface areas of the Pd NP samples, but were unsuccessful,
owing to the aggregation and coalescence caused by drying
[*] F. Wang, Prof. L.-D. Sun, C.-H. Xu, Prof. C.-H. Yan
State Key Lab of Rare Earth Materials Chemistry and Applications
Peking University, Beijing 100871 (China)
E-mail: sun@pku.edu.cn
yan@pku.edu.cn
F. Wang, Prof. J. F. Wang
Department of Physics, The Chinese University of Hong Kong
Shatin, Hong Kong SAR (China)
E-mail: jfwang@phy.cuhk.edu.hk
C. H. Li, Prof. J. C. Yu
Department of Chemistry, The Chinese University of Hong Kong
Shatin, Hong Kong SAR (China)
[**] This work was supported by the National Natural Science
Foundation of China (Project Code: 20931160429) and the Research
Grants Council of Hong Kong (NSFC/RGC Joint Scheme, Ref.:
N_CUHK465/09, Project Code: 2900339; Direct Allocation, Project
Code: 2060417).
Figure 1. a) TEM and b) HAADF–STEM images of porous Pd NPs with
size of 34 nm. c) Plot of the average NP size versus the seed volume.
The standard deviation in the size for the largest NP sample is 4 nm,
and those for the other smaller NP samples are 1 nm. d) XRD pattern
of porous Pd NPs with size of 53 nm.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
during the measurements. The NP sizes, which were obtained
by measuring the NP diameter on the TEM images, is readily
controlled by varying the seed volume. The average size
decreases from 53 to 14 nm as the seed amount is increased
(Figure 1c and Figure S1 in the Supporting Information).
Figure 1d shows the X-ray diffraction (XRD) pattern of the
sample sized at 53 nm. All of the peak positions and
intensities accord well with the face-centered-cubic structure
of palladium (JCPDS No. 46-1043), indicating high crystal-
linity. The average grain size estimated using the Scherrer
formula from all the peaks is 49 nm. This value is close to the
average size obtained from the TEM images, implying that
each porous Pd NP might be single-crystalline instead of
being composed of randomly oriented small nanocrystals.
We characterized the crystalline structure of the individ-
ual NPs with high-resolution (HR) TEM imaging and
electron diffraction (ED) (Figure 2). For the large NPs, the
HRTEM images recorded at different positions on a single
NP show atomic lattices with the same orientation (Fig-
ure 2a–c). For the small NPs, the atomic lattice of an entire
NP can be observed in one HRTEM image (Figure 2d). In
addition, the imaging contrast on the small NPs is non-
uniform. The center of the NP shown in Figure 2d is brighter
than the exterior. The non-uniform contrast also suggests the
porous nature of the single-crystalline NPs. The HRTEM
images in Figure 2a–d were taken along the [011] lattice
direction. We also observed the atomic lattices of the porous
NPs along the [112] lattice direction (Figure S2 in the
Supporting Information).
ED patterns acquired on the Pd NPs of varying sizes
under different lattice orientations all display single sets of
sharp diffraction spots (Figure 2e–j). Taken together, the
HRTEM imaging and ED results unambiguously corroborate
that the porous Pd NPs are single-crystalline. Each Pd NP
contains many small domains that are 3–7 nm large and
oriented in the same crystalline direction. Moreover, the
porous NPs exhibit many atomic steps on their surfaces, as
revealed by the HRTEM images. The Pd atoms on these
atomic steps can serve as catalytically active sites because of
their coordinative unsaturation. We note that there have been
only two reports on branched, single-crystalline metal NPs,
which are Pt[15] and Pt-on-Pd.[12] In those two studies, the
single-crystalline nature is evidenced only by the HRTEM
images and Fourier transforms along the [011] direction.
The formation of the porous Pd NPs might undergo two
mechanisms. One is the formation of very small Pd nano-
crystals and the subsequent oriented integration. The other is
seed-mediated growth. Our TEM results strongly support the
second mechanism from three aspects. First, no single small
Pd nanocrystals are seen under TEM imaging. Second, the
sizes of the porous Pd NPs are well controlled by the seed
amount. Third, the size distributions of the porous NPs are
narrow. To understand better the growth mechanism, we
studied their time-dependent morphological evolution by
TEM imaging (Figure S3 in the Supporting Information). The
major growth step was found to be completed rapidly in the
first few seconds after the injection of the reducing agent. The
sizes of the products collected subsequently at varying periods
of time remain nearly the same. Within 1 min of growth, the
number and size of the pores in the NPs are small. More pores
are seen with increasing growth periods of time, and their
sizes increase. In addition, no small Pd nanocrystals were
observed. This evolution suggests that the porous NPs are
produced through direct homoepitaxial growth on the Pd
seeds.
We performed additional control experiments to unravel
the growth mechanism further. First, cetyltrimethylammo-
nium bromide (CTAB) was used for the growth (Figure S4 in
the Supporting Information). Second, the growth was carried
out in a mixture of CTAC and CTAB at varying molar ratios
(Figure S5 in the Supporting Information). Third, an excess
amount of halide anions (FÀ, ClÀ, and BrÀ) was introduced in
the growth solution in addition to CTAC or CTAB (Figure S6
in the Supporting Information). The binding of ammonium
surfactants to noble metals is assisted by the adsorption of
halide anions to the metal surface.[24] The binding strengths of
halide ions to the Pd surface increase in the order of FÀ, ClÀ,
Figure 2. a) TEM image of a single porous Pd NP with size of 53 nm.
b,c) HRTEM images recorded from the regions indicated by the upper
and lower boxes in (a), respectively. d) HRTEM image of a single
porous Pd NP with size of 18 nm. e–j) TEM images and corresponding
ED patterns of the porous Pd NPs. The NP in (e,f) is from the sample
sized at 53 nm, and those in (g–j) are from the sample sized at 18 nm.
The numbers in the ED patterns indicate the zone axes.
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
and BrÀ. Porous Pd NPs are obtained only from the growth
with FÀ and CTAC. When BrÀ ions are present or the
concentration of ClÀ is very high, the growth is much slower
and produces solid Pd NPs. Fourth, the growth solution pH
was adjusted from 1–12 (Figure S7 in the Supporting Infor-
mation). A distinct morphological transition occurs in the pH
range 6.5–7.5, where the products change from solid NPs into
porous ones. In addition, the growth becomes faster as the pH
value is raised. The fast growth and morphological change is
ascribed to the fact that the reducing capability of the
reducing agent, ascorbic acid, increases with the solution pH
value.
The above results suggest that rapid growth is essential for
the formation of the porous single-crystalline Pd NPs
(Figure 3). It produces Pd NPs with many small crevices.
Figure 4. TOF values of the Pd-based catalysts for the coupling
between iodobenzene and phenylboronic acid. a) Reaction equation.
b) TOF values at 858C for 15 min. For each catalyst, the left column
shows the TOF value in the first run, and the right one indicates the
TOF value in the second run. c) TOF values at room temperature for
120 min.
Figure 3. Schematic illustration of the proposed growth mechanism of
the porous single-crystalline Pd NPs. The TEM images show the
products obtained after growth for 30 s and 30 min, respectively. The
volume of the seed solution was 0.010 mL.
The initially formed Pd NPs then undergo oxidation owing to
the presence of ClÀ and O2 in the solution.[25] The oxidative
etching occurs preferentially at the highly active sites in the
crevices, giving the porous structure. If the growth is slow,
solid NPs will be produced. Pores are difficult to form in these
solid NPs through the oxidative etching.
nanobranches,[27] and icosahedrons[28] in terms of the Pd
amount and reaction time. In those studies, the molar
percentages of Pd relative to iodobenzene are 3.0%, 1.6%,
1.8%, and 1.7%, respectively. The reaction time ranges from
4 to 8 h. The reaction yields are around 90%. In our
experiments, the molar percentage of Pd relative to iodoben-
zene is 0.04%, and the reaction time is only 15 min.
We tested the catalytic activities of the porous Pd NPs
with the Suzuki coupling reaction, a widely used model
reaction. The amounts of Pd contained in the catalysts were
adjusted to be the same by inductively coupled plasma atomic
emission spectroscopy (ICP–AES) (Table S1 in the Support-
ing Information). The turnover frequencies (TOFs) were
calculated according to the total amounts of Pd. For the
coupling reaction between phenylboronic acid and iodoben-
zene (Figure 4a) at 858C, the reaction yields of biphenyl
range from 89 to 99% for the differently sized porous Pd NPs
(Table S2 in the Supporting Information). The TOF values
are around 2.5 sÀ1 (Figure 4b). The size-independence of the
catalytic activity arises from the porous nature of these NPs,
which makes the NPs have approximately the same specific
surface area. In comparison, the TOF values are 0.92 and
1.75 sÀ1 for a Pd nanosphere sample with a size of 24 Æ 2 nm
(Figure S8 in the Supporting Information) and the commer-
cial Pd/C containing less-than-5-nm Pd NPs, respectively. The
TOFs of the porous Pd NPs are approximately 3 and 1.5 times
those of the Pd nanospheres and Pd/C. Moreover, the
catalytic efficiencies of the porous Pd NPs are much higher
than those reported for Pd nanospheres,[26] nanorods,[27]
The recycling capability of the porous Pd NPs was
examined by removing the supernatant after the reaction
and leaving the precipitated Pd NPs for another run of the
reaction. Product yields above 80% were obtained in the
second run (Figure 4b and Table S3 in the Supporting
Information). The morphologies of the Pd NPs remain
nearly unchanged after the first run (Figure S9 in the
Supporting Information). In addition, reaction yields of 80–
90% were obtained in the first four cycles for the 53-nm- and
18-nm-sized Pd NP samples (Table S4 in the Supporting
Information). The reaction yields dropped to 10.0% at the
fifth cycle, as a result of aggregation of the Pd NPs at the
bottom of the reaction flask. To avoid aggregation, we further
deposited the 53-nm-sized Pd NPs on a piece of cotton fabric
with a size of 2 cm ꢀ 2 cm and immersed it in the solution for
catalyzing the coupling reaction between iodobenzene and
phenylboronic acid. The reaction was carried out at 858C for
30 min. The reaction yields of biphenyl were above 80% even
after 11 cycles (Figure S10 and Table S5 in the Supporting
Information). We are currently carrying out more catalytic
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
studies by depositing various metal nanostructures on differ-
ent types of fabrics. The amounts of the leached Pd were
measured with ICP-AES to be 1.0–1.5% after the reaction
mixture containing the 34-nm-sized Pd NPs was kept at 858C
for 15 min. The leached Pd was found to show no catalytic
activity under our reaction conditions.
The TOF values of the vaiously sized Pd NPs at room
temperature were 0.16 sÀ1, which is 30 times that of the Pd/C
(Figure 4c and Table S6 in the Supporting Information).
When bromobenzene, which is less reactive than iodobenzene
Pd NPs is ascribed to a rapid-growth process. The porous Pd
NPs display high catalytic activities toward the Suzuki
coupling reaction at both raised and room temperatures.
The TOF values of the porous Pd NPs are around 30 times
that of the commercial Pd/C for iodobenzene, and 6 times for
bromobenzene. The Pd NPs also exhibit good substrate
selectivity. The high catalytic activities of the porous Pd NPs is
attributed to their large surface area and high concentration
of coordinatively unsaturated surface atoms. Our results
demonstrate that porous metal NPs could provide an
excellent catalytic platform for organic reactions.
À
because of the larger C Br bond energy, is used, the TOF
values of the 34-nm-sized Pd NP sample were 0.24 sÀ1 at 858C
and 39.3 hÀ1 at room temperature. Both of these values are
around 6 times those of the Pd/C (Table S7 in the Supporting
Information). So far, the Suzuki coupling reactions at room
temperature have been reported only with Pd complexes as
the catalysts.[29,30] There have been no reports with Pd
nanostructures as the catalysts at room temperature.
Received: October 19, 2011
Revised: March 9, 2012
Published online: && &&, &&&&
À
Keywords: C C coupling · crystal growth ·
mesoporous materials · nanoparticles · palladium
.
To help in understanding the higher catalytic activities of
the porous Pd NPs than that of the Pd/C, we measured the
cyclic voltammetry (CV) curves for O-atom sorption on three
Pd NP samples and the Pd/C (Figure S11 in the Supporting
Information). The average electrochemical surface areas
obtained from O atom sorption were (5.1 Æ 0.1), (5.7 Æ 0.4),
(5.6 Æ 0.6), and (13.3 Æ 0.7) ꢀ 104 cm2 gÀ1 for the 34-nm-, 28-
nm-, 18-nm-sized Pd NP samples, and the Pd/C, respectively
(Table S8 in the Supporting Information). The average
electrochemical surface areas of the three differently sized
Pd NP samples are very similar, indicating that the differently
sized Pd NPs have approximately the same specific surface
area. Although the electrochemical surface area of the Pd/C is
larger than those of the porous Pd NC samples, the Pd NP
samples exhibit higher TOF values. This behavior might be
caused by the different adsorption abilities of the species
involved in the catalytic reaction and electrochemical mea-
surement.
[1] K. M. Bratlie, H. Lee, K. Komvopoulos, P. D. Yang, G. A.
Somorjai, Nano Lett. 2007, 7, 3097 – 3101.
[2] N. Tian, Z.-Y. Zhou, S.-G. Sun, Y. Ding, Z. L. Wang, Science
2007, 316, 732 – 735.
[3] Y. Y. Ma, Q. Kuang, Z. Y. Jiang, Z. X. Xie, R. B. Huang, L. S.
Zheng, Angew. Chem. 2008, 120, 9033 – 9036; Angew. Chem. Int.
Ed. 2008, 47, 8901 – 8904.
[4] J. P. Ge, T. Huynh, Y. X. Hu, Y. D. Yin, Nano Lett. 2008, 8, 931 –
934.
[5] S. H. Joo, J. Y. Park, C.-K. Tsung, Y. Yamada, P. D. Yang, G. A.
Somorjai, Nat. Mater. 2009, 8, 126 – 131.
[6] C.-L. Lu, K. S. Prasad, H.-L. Wu, J. A. Ho, M. H. Huang, J. Am.
Chem. Soc. 2010, 132, 14546 – 14553.
[7] F. Wang, C. H. Li, L.-D. Sun, H. S. Wu, T. Ming, J. F. Wang, J. C.
Yu, C.-H. Yan, J. Am. Chem. Soc. 2011, 133, 1106 – 1111.
[8] Y. Yamada, C.-K. Tsung, W. Y. Huang, Z. Y. Huo, S. E. Habas, T.
Soejima, C. E. Aliaga, G. A. Somorjai, P. D. Yang, Nat. Chem.
2011, 3, 372 – 376.
Finally, we investigated the catalytic activities of the 34-
nm-sized Pd NPs for the reactions between the respective
position isomers of bromoanisole and tolylboronic acid. A
total of nine reactions were performed. At 858C, all of the
reactions involving m-bromoanisole show relatively high
TOF values (Figure S12 and Tables S9 and S10 in the
Supporting Information). The reactions involving o-tolylbor-
onic acid are less efficient than those involving m- and p-
tolylboronic acid. The TOF value of the reaction between o-
bromoanisole and m-tolylboronic acid is the highest. In
comparison, the average TOF value at room temperature is
about 1/20 of that at 858C. The overall trend in the TOFs is
also different from that at 858C. These results indicate that
the porous Pd NPs are also highly active for the reactions
involving complicated molecules. More studies will be needed
to compare the substrate selectivities between the porous Pd
NPs and Pd complexes and better understand the catalytic
properties of the porous Pd NPs. However, we also tested the
catalytic activities of the 34-nm-sized Pd NPs for the reaction
between chlorobenzene and phenylboronic acid both at 858C
and room temperature. No reaction activities were obtained.
In summary, we have shown the size-controlled growth of
porous single-crystalline Pd NPs. The formation of the porous
[9] A. R. Tao, S. Habas, P. D. Yang, Small 2008, 4, 310 – 325.
[10] Y. N. Xia, Y. J. Xiong, B. Lim, S. E. Skrabalak, Angew. Chem.
2009, 121, 62 – 108; Angew. Chem. Int. Ed. 2009, 48, 60 – 103.
[11] D. S. Wang, Y. D. Li, Adv. Mater. 2011, 23, 1044 – 1060.
[12] B. Lim, M. J. Jiang, P. H. C. Camargo, E. C. Cho, J. Tao, X. M.
Lu, Y. M. Zhu, Y. N. Xia, Science 2009, 324, 1302 – 1305.
[13] H. Lee, S. E. Habas, S. Kweskin, D. Butcher, G. A. Somorjai,
P. D. Yang, Angew. Chem. 2006, 118, 7988 – 7992; Angew. Chem.
Int. Ed. 2006, 45, 7824 – 7828.
[14] X. W. Teng, X. Y. Liang, S. Maksimuk, H. Yang, Small 2006, 2,
249 – 253.
[15] M. A. Mahmoud, C. E. Tabor, M. A. El-Sayed, Y. Ding, Z. L.
Wang, J. Am. Chem. Soc. 2008, 130, 4590 – 4591.
[16] L. Wang, Y. Yamauchi, J. Am. Chem. Soc. 2009, 131, 9152 – 9153.
[17] A. Mohanty, N. Garg, R. C. Jin, Angew. Chem. 2010, 122, 5082 –
5086; Angew. Chem. Int. Ed. 2010, 49, 4962 – 4966.
[18] S. H. Sun, G. X. Zhang, D. S. Geng, Y. G. Chen, R. Y. Li, M. Cai,
X. L. Sun, Angew. Chem. 2011, 123, 442 – 446; Angew. Chem. Int.
Ed. 2011, 50, 422 – 426.
[19] Z. M. Peng, H. Yang, J. Am. Chem. Soc. 2009, 131, 7542 – 7543.
[20] B. Lim, Y. N. Xia, Angew. Chem. 2011, 123, 78 – 87; Angew.
Chem. Int. Ed. 2011, 50, 76 – 85.
[21] L. Wang, Y. Nemoto, Y. Yamauchi, J. Am. Chem. Soc. 2011, 133,
9674 – 9677.
[22] L. Wang, Y. Yamauchi, J. Am. Chem. Soc. 2010, 132, 13636 –
13638.
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[23] H. Zhang, W. Y. Li, M. S. Jin, J. Zeng, T. Yu, D. R. Yang, Y. N.
Xia, Nano Lett. 2011, 11, 898 – 903.
[27] Y.-H. Chen, H.-H. Hung, M. H. Huang, J. Am. Chem. Soc. 2009,
131, 9114 – 9121.
[24] H. A. Keul, M. Mçller, M. R. Bockstaller, Langmuir 2007, 23,
10307 – 10315.
[25] B. Lim, M. J. Jiang, J. Tao, P. H. C. Camargo, Y. M. Zhu, Y. N.
Xia, Adv. Funct. Mater. 2009, 19, 189 – 200.
[26] S.-W. Kim, M. Kim, W. Y. Lee, T. Hyeon, J. Am. Chem. Soc.
2002, 124, 7642 – 7643.
[28] C. C. Li, R. Sato, M. Kanehara, H. B. Zeng, Y. Bando, T.
Teranishi, Angew. Chem. 2009, 121, 7015 – 7019; Angew. Chem.
Int. Ed. 2009, 48, 6883 – 6887.
[29] B. Saito, G. C. Fu, J. Am. Chem. Soc. 2007, 129, 9602 – 9603.
[30] T. Nishikata, A. R. Abela, S. L. Huang, B. H. Lipshutz, J. Am.
Chem. Soc. 2010, 132, 4978 – 4979.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Palladium Catalysts
F. Wang, C. H. Li, L.-D. Sun,* C.-H. Xu,
J. F. Wang,* J. C. Yu,
C.-H. Yan*
&&&& — &&&&
Porous Single-Crystalline Palladium
Nanoparticles with High Catalytic
Activities
Palladium’s pore cousin: A facile
nanoparticles exhibit size-independent
catalytic activities for the Suzuki coupling
and are more active than commercial Pd/
C catalysts.
approach is described for the size-con-
trolled preparation of porous single-crys-
talline Pd nanoparticles. These porous Pd
6
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!