Fabrication of Ultrafine Palladium Phosphide Nanoparticles as Highly Active Catalyst for Chemoselective Hydrogenation of Alkynes

Article abstract of DOI:10.1002/asia.201500939

Monodisperse palladium phosphide nanoparticles (Pd-P NPs) with a smallest size ever reported of 3.9nm were fabricated using cheap and stable triphenylphosphine as phosphorous source. After the deposition and calcination at 300 °C and 400 °C, the resulting Pd-P NPs increased in size to 4.0nm and 4.8nm, respectively. Notably, the latter NPs probably crystallized with a single phase of Pd₃P_0.95, which acted as a highly active catalyst in semi-and stereoselective hydrogenation of alkynes. X-ray photoelectron spectroscopy analysis determined a positive shift of binding energy for Pd(3d) in Pd-P NPs compared to that in Pd on carbon. It indicated the electron flow from metal to phosphorus and the larger electron deficiency of Pd in Pd-P NPs, which suppressed palladium hydride formation and subsequently increased the selectivity. Thus, this result may also indicate the applications of Pd-P and other metal-P NPs in various selective hydrogenation reactions.

Full text of DOI:10.1002/asia.201500939

DOI: 10.1002/asia.201500939
Communication
Nanoparticle Catalysis
Fabrication of Ultrafine Palladium Phosphide Nanoparticles as
Highly Active Catalyst for Chemoselective Hydrogenation of
Alkynes
[a]
Ming Zhao*
electronic deficient and also formed a high energy barrier for
Abstract: Monodisperse palladium phosphide nanoparti-
cles (Pd–P NPs) with a smallest size ever reported of
subsurface chemistry, segregation, and metal hydride forma-
[11,14,15]
tion.
Nickel phosphide (Ni–P) nanocatalysts were once
3
.9 nm were fabricated using cheap and stable triphenyl-
[16]
reported for the chemoselective hydrogenation of alkynes;
however, low reactivity and low Z/E ratio of internal alkenes
were observed. Although metal phosphide nanomaterials have
phosphine as phosphorous source. After the deposition
and calcination at 3008C and 4008C, the resulting Pd–P
NPs increased in size to 4.0 nm and 4.8 nm, respectively.
Notably, the latter NPs probably crystallized with a single
[17]
been extensively studied in many areas, the application of
Pd–P NPs in selective catalysis is rare. To date, the fabrication
of Pd–P NPs was usually carried out using white phosphorus
phase of Pd3P0.95, which acted as a highly active catalyst in
semi- and stereoselective hydrogenation of alkynes. X-ray
[17]
or trioctylphosphine (TOP) as phosphorus source. Between
photoelectron spectroscopy analysis determined a positive
shift of binding energy for Pd(3d) in Pd–P NPs compared
to that in Pd on carbon. It indicated the electron flow
from metal to phosphorus and the larger electron defi-
ciency of Pd in Pd–P NPs, which suppressed palladium hy-
dride formation and subsequently increased the selectivi-
ty. Thus, this result may also indicate the applications of
Pd–P and other metal–P NPs in various selective hydroge-
nation reactions.
the two, the latter was more likely to form monodisperse NPs
[18]
with fine particle size. However, TOP is very sensitive and is
easily oxidized in air. It is critical to find a more easily handled
phosphine as an alternative. Moreover, there have been few re-
[18]
ports on Pd–P nanocatalysts for selective reactions. To the
best of our knowledge, Pd–P NPs have not yet been investigat-
ed in the selective hydrogenation of alkynes. Herein, we re-
ported the first fabrication of Pd–P NPs using inexpensive and
stable triphenylphosphine (PPh ) as phosphorous source. The
3
monodisperse NPs were obtained with the smallest size ever
reported (3.9 nm) and a novel single phase (Pd3P0.95). After an-
nealing at 3008C and 4008C, the NPs grow to 4.0 and 4.8 nm,
respectively. Notably, the well-crystallized 4.8 nm Pd–P NPs
showed very high catalytic selectivity in semi- and stereoselec-
tive hydrogenation of alkynes. Without any additive, high se-
lectivity of terminal alkene (e.g. styrene, 96%) and high Z/E
ratio of internal alkene (e.g. ethyl cinnamate, 97:3) were ob-
tained. With the help of quinoline, various terminal and inter-
nal alkenes were obtained in high yield (and Z/E selectivity).
Notably, the deactivation of catalyst in this material by doping
phosphorus was not observed at all in comparison with mono-
metallic Pd/C.
Selective hydrogenation of carbon–carbon triple bonds is an
important route to achieve olefin chemicals, especially for ste-
reospecific products. Among others, noble metals are one type
of the most active and versatile heterogeneous catalysts for
hydrogenation reactions. Once modified, they can act as ideal
selective catalysts. For example, lead poisoned Pd/CaCO
3
[1,2]
named as the conventional Lindlar catalyst,
decorated by self-assembled monolayer (SAM),
noble metal
and Pd–M
[3–6]
[7–10]
(
M=Cu, Ag, Zn, et al) bimetallic systems
have been exten-
sively studied as selective hydrogenation catalysts. However,
the use of toxic lead compounds, high cost, and/or decreased
activity with respect to these catalysts blocked their practical
applications. On the other hand, metal–metalloid alloy nano-
materials were promising catalyst candidates for hydrogena-
The fabrication of colloidal Pd–P NPs was modified from our
[
19]
previous report. Briefly, palladium acetylacetonate (Pd(acac)2,
0.1 mmol) and triphenylphosphine (TPP, 0.88 mmol) as Pd and
P sources, trioctylphosphine oxide (TOPO, 1.16 g) as a surfac-
tant, borane tert-butylamine complex (BTB, 1 mmol) as a reduc-
ing reagent, and oleylamine (OLA, 70%, 6.5 mL) as a co-reduc-
ing reagent as well as a stabilizer were used. A mixture of the
above-mentioned chemicals was first deaerated at 508C using
a nitrogen flow for 30 min, stirred at 2208C for 30 min, and in-
cubated at 2708C for 15 min. A dark suspension finally formed,
from which the colloidal Pd–P NPs were collected by centrifu-
gation. The monodisperse NPs were then dispersed in hexane
and characterized carefully. Transmission electron microscopy
[11–13]
tion of acetylene in gas-phase conditions.
The interaction
between noble metal and metalloid element made the metal
[
a] Dr. M. Zhao
School of Chemical Engineering
China University of Mining and Technology
Xuzhou 221116 (People’s Republic of China)
E-mail: ming815zhao@163.com
Supporting information for this article, including additional preparations
details and characterizations of nanomaterials, is available on the WWW
under http://dx.doi.org/10.1002/asia.201500939.
Chem. Asian J. 2016, 11, 461 – 464
461
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
particle sizes increased to 4.0 nm and 4.8 nm (Figure 1D,F), re-
spectively, yet still with a narrow distribution each. When Fig-
ure 1G (and Figure S2A in the Supporting Information) is com-
pared with Figure 1H (and Figure S2B in the Supporting Infor-
mation), XRD patterns (and HRTEM images) indicated the crys-
talline structure of Pd–P NPs became much clearer. Both induc-
tively coupled plasma mass spectrometry (ICP-MS, which
determined the composition of Pd–P NPs as Pd P ) and a ref-
75 25
erence XRD pattern (Figure 1I) suggested a single phase in the
annealed Pd–P NPs was Pd3P0.95. This is the first time this phase
is reported in nanoparticles.
The catalytic performance of the supported Pd–P NPs was
evaluated in semi-hydrogenation reaction. The catalysts an-
nealed at 3008C and 4008C are denoted Pd–P/C-300 and Pd–
P/C-400, respectively. The reaction was performed using
0
.13 mol% catalyst under one atmosphere of hydrogen gas in
the presence of quinoline. As shown in Table 1, an impressive
Table 1. Optimization of the Conditions for Semi-Hydrogenation of
Phenyl Acetylene.
[a]
Entry Catalyst
Solvent
Time
Selectivity 2a:3a
1
2
3
4
5
Pd/C
EtOAc
EtOAc
EtOAc
1.9 h
4.0 h
48 min 98:2
88:12
95:5
[
b]
Pd–P/C-300
Pd–P/C-400
Pd–P/C-400
Pd–P/C-400
Pd–P/C-400
Pd–P/C-400
Pd–P/C-400
EtOAc (dehydrated) 37 min 98:2
CH CN (dehydrated) 30 min 98:2
MeOH/H O (25:1) 56 min 99:1
O (10:1) 1.1 h 98:2
35 min 96:4
1.0 h 91:9
3
6
7
8
2
MeOH/H
2
[
c]
MeOH/H
2
O (25:1)
[
c,d]
9
Lindlar catalyst MeOH/H
2
O (25:1)
Reaction conditions: 1a (0.5 mmol), solvent (5 mL), room temperature;
the catalyst (0.13 mol%) was pretreated with additive quinoline
(1.1 equiv) for 3 h. Conversion was >99% for alkynes in each case. [a] De-
termined by GC-MS. [b] Reaction temperature was 508C. [c] In the ab-
Figure 1. TEM image (A) and size distribution analysis (B) of OLA coated Pd–
P NPs; TEM images, size distribution analysis, and XRD patterns of Pd–P/C
after annealing at 3008C (C, D, and G) or 4008C (E, F, and H) for 1 h; Refer-
sence of quinoline. [d] 0.62 mol% of catalyst was used.
enced XRD pattern of Pd
3
P0.95 alloy with powder diffraction file (PDF)
number #01-089-3046 (I).
selectivity for alkene was achieved using Pd–P/C-300 (entry 2)
compared with commercial Pd/C (entry 1, TEM image shown in
Figure S3 in the Supporting Information). To our surprise, when
Pd–P/C-400 was used, both the activity and selectivity was fur-
ther enhanced (entry 3). Dehydrated solvents accelerated the
reaction but did not change the selectivity (entries 4 and 5). In
(
TEM, Figure 1A) and size distribution analysis (Figure 1B)
showed the uniform NPs formed with an average size of
.9 nm in diameter. In high resolution TEM (HRTEM, Figure S1A
3
in the Supporting Information), a lattice fringe of the NPs were
measured as 0.235 nm, which is larger than that of the face-
centered cubic Pd (111) plane (0.224 nm), indicating that the
Pd–P alloy may have formed. Energy-dispersive X-ray spectros-
copy (EDX) analysis further determined the coexistence of ele-
mental Pd and P in one particle (Figure S1B in the Supporting
Information).
entry 6, we found that a mixed solvent of MeOH and H O with
2
a volume ratio of 25:1 was very special for the hydrogenation,
and the alkene/alkane ratio reached up to 99:1. In entry 7,
using larger amount of H O did not help. It is worth mention-
2
ing that a high selectivity for alkene (96%) was achieved even
without additive (entry 8). In contrast, the traditional Lindlar
catalyst (Pd-CaCO -PbO/PbAc ) showed both low activity
3
2
Next, the as-synthesized Pd–P NPs were deposited on acti-
vated carbon (Vulcan XC-72, Pd loading: 6.5 wt%) for catalysis
use. Figure 1C,E show TEM images of the supported NPs after
annealing at 3008C and 4008C under a nitrogen flow. Their
(0.62 mol% of Pd, 1.0 h) and low selectivity of 91% (entry 9).
Various terminal alkynes were investigated under the opti-
mized conditions (Table 2). In entry 2, the bromide and in en-
tries 3 and 4, other easily reduced groups were tolerated in
Chem. Asian J. 2016, 11, 461 – 464
www.chemasianj.org
462
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
even in the absence of additive, although over-reduction of
alkene was observed. The roles of quinoline may include two
points: i) coordination to Pd to change the active sites of the
catalyst; ii) reaction with hydrogen dissociated on Pd to decel-
Table 2. Semi-Hydrogenation of Terminal Alkynes.
[
20]
erate the hydrogenation process. Therefore, the Pd–P/C-400
hybrid catalyst itself probably contributed to the stereospecific
hydrogenation, while quinoline is important to suppress the
total hydrogenation in the case of internal alkynes.
[a]
Entry
Alkyne
t [min]
56
Selectivity 2:3
1
2
3
1a
1b
1c
99:1
54
99:1
Considering that metal hydride formation is often the rate-
determined step for selective hydrogenation reactions, the
doping of P was crucial for the excellent catalytic performance
of Pd–P NPs. X-ray photoelectron spectroscopy (XPS) analysis
determined a positive shift of Pd(3d) in Pd–P NPs compared to
that in Pd/C (Figure 2A). The binding energies of Pd(3d) in Pd/
24
99:1
4
5
6
1d
1e
1 f
34
30
24
98:2
99:1
95:5
7
8
1g
1h
26
31
98:2
[
b]
b]
99:1 (96)
2
3
4
5
nd cycle
rd cycle
th cycle
th cycle
30
36
20
24
99:1
99:1 (95)
98:2
[
[b]
97:3 (95)
1
[
[
a] Determined by GC-MS or H NMR. Conversion was >99% for alkynes.
b] Yield of isolated 2h shown in parenthesis.
this transformation. The corresponding alkenes were obtained
in higher than 98% yields. In entries 5 and 7, the hydrogena-
tion of aliphatic alkyne and alcohol gave high selectivity for al-
kenes of 99% and 98%, respectively. In entry 6, diene can also
be obtained in high selectivity of 95%. Finally in entry 8, the
stability of the catalyst was investigated using 4-ethynylbi-
phenyl. After five cycles’ test, the ratio of between alkene and
alkane was remain as high as 97:3 with a 95% yield of isolated
alkene. During the reuse process, the reaction time was not
prolonged, which may indicate that active sites of catalyst
were barely lost. Moreover, the selective semi-hydrogenation
of several internal alkynes was also studied. As shown in
Table 3, typical substrates including diphenylactylene (entry 1),
[
a]
Table 3. Selective Semi-Hydrogenation of Internal Alkynes.
Entry
Alkyne
t
Alkene:Alkane
Z:E
1
2
3
47 min
43 min
1.0 h
95:5
97:3
96:4
40:60
>99:1
>99:1
99:1
Figure 2. XPS spectra on A) Pd(3d) and B) P(2p) of different Pd-based cata-
lysts.
[b]
4
26 min
97:3
The reactions were carried out under the above mentioned optimized
conditions. Conversion was >99% for alkynes. [a] Selectivity was deter-
1
mined by H NMR spectroscopy. [b] Without quinoline.
C, Pd–P/C-300, and Pd–P/C-400 increased gradually from
3
35.67 to 335.78 eV and 335.85 eV, respectively. Besides, the
appearance of Pd-P-O peaks of P(2p) in both Pd–P/C-300 and
Pd–P/C-400 (Figure 2B) also indicated the formation of PdÀP
covalent bond and the electron flow from Pd to P, which sup-
1
-phenyl-1-propyne (entry 2), and ethyl phenylpropiolate
(
entry 3) were employed in the hydrogenation. The corre-
[
11,12]
sponding alkenes were obtained all in high yield of >95%. No-
tably, besides the high double bond to single bond ratio, the
excellent Z:E selectivity in alkenes were also achieved as above
pressed the Pd hydride formation.
Besides, P isolated the
crystalline Pd atoms, which created a high barrier for the seg-
regation of Pd atoms and their subsequent hydriding behav-
[
11,12]
9
9:1. In entry 4, high Z:E ratio of cinnamic acid was obtained
ior.
Additionally, P atoms could poison the surface by
Chem. Asian J. 2016, 11, 461 – 464
www.chemasianj.org
463
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
blocking the most unsaturated sites, in the same way that lead
[5] S. H. Pang, C. A. Schoenbaum, D. K. Schwartz, J. W. Medlin, Nat.
Commun. 2013, 4, 2448–2453.
[21]
was used to poison Pd.
The hydrogenation would work
[6] K. R. Kahsar, D. K. Schwartz, J. W. Medlin, ACS Catal. 2013, 3, 2041–
through the Horiuti–Polanyi mechanism on moderately unsatu-
rated Pd sites. The excellent Z-selectivity observed experimen-
tally (Table 3) correlates well with this hypothesis.
2044.
[7] K. H. Lee, B. Lee, K. R. Lee, M. H. Yi, N. H. Hur, Chem. Commun. 2012, 48,
4414–4416.
[8] G. Kyriakou, M. B. Boucher, A. D. Jewell, E. A. Lewis, T. J. Lawton, A. E.
In summary, we report the fabrication of monodisperse, ul-
Baber, H. L. Tierney, M. Flytzani-Stephanopoulos, E. C. H. Sykes, Science
trasmall, and single phase Pd–P NPs using PPh as a phospho-
3
2012, 335, 1209–1212.
rus source for the first time. After a simple annealing process,
the well-crystallized Pd3P0.95 NPs show excellent catalytic per-
formance in chemoselective hydrogenation of alkynes. Various
terminal and internal alkynes were investigated to produce
corresponding alkenes in high chemoselectivity. The electronic
deficiency of Pd in the Pd–P system probably contributed the
high selectivity of the catalyst. It is the first time that Pd–P NPs
have been used for selective semi-hydrogenation. The result is
expected to promote the design of Pd–P or other metal-P
nanocatalysts for various selective reactions.
[9] P. A. Sheth, M. Neurock, C. M. Smith, J. Phys. Chem. B 2005, 109, 12449–
12466.
10] M. W. Tew, H. Emerich, J. A. van Bokhoven, J. Phys. Chem. C 2011, 115,
[
[
8457–8465.
11] M. Armbrüster, K. Kovnir, M. Behrens, D. Teschner, Y. Grin, R. Schlçgl, J.
Am. Chem. Soc. 2010, 132, 14745–14747.
[12] A. Ota, M. Armbrüster, M. Behrens, D. Rosenthal, M. Friedrich, I. Kasatkin,
F. Girgsdies, W. Zhang, R. Wagner, R. Schlçgl, J. Phys. Chem. C 2011, 115,
1368–1374.
[
13] E. W. Shin, J. H. Kang, W. J. Kim, J. D. Park, S. H. Moon, Appl. Catal. A
2002, 223, 161–172.
[14] M. PØrez Jigato, B. Coussens, D. A. Kind, J. Chem. Phys. 2003, 118, 5623–
634.
5
[
[
15] J. A. Eastman, L. J. Thompson, B. Kestel, Phys. Rev. B 1993, 48, 84–92.
16] S. Carenco, A. Leyva-PØrez, P. Concepción, C. Boissi›re, N. MØzailles, C.
Sanchez, A. Corma, Nano Today 2012, 7, 21–28.
Acknowledgements
[
17] S. Carenco, D. Portehault, C. Boissi›re, N. MØzailles, C. Sanchez, Chem.
This work was supported by Fundamental Research Funds for
the Central Universities (no. 2015QNA22) and the Priority Aca-
demic Program Development of Jiangsu Higher Education In-
stitutions and Jiangsu Qinglan Project.
Rev. 2013, 113, 7981–8065.
[18] G. H. Layan Savithra, R. H. Bowker, B. A. Carrillo, M. E. Bussell, S. L. Brock,
ACS Appl. Mater. Interfaces 2013, 5, 5403–5407.
[
[
[
19] M. Zhao, K. Abe, S. Yamaura, Y. Yamamoto, N. Asao, Chem. Mater. 2014,
6, 1056–1061.
20] M. Yan, T. Jin, Y. Ishikawa, T. Minato, T. Fujita, L.-Y. Chen, M. Bao, N. Asao,
M.-W. Chen, Y. Yamamoto, J. Am. Chem. Soc. 2012, 134, 17536–17542.
21] . Molnµr, A. Sµrkµny, M. Varga, J. Mol. Catal. A 2001, 173, 185–221.
2
Keywords: alkynes · heterogeneous catalysis · hydrogenation ·
nanoparticles · palladium
[
[
1] J. Stachurskiand, J. M. Thomas, Catal. Lett. 1988, 1, 67–72.
2] G. A. Attard, J. A. Bennett, I. Mikheenko, P. Jenkins, S. Guan, L. E. Macas-
kie, J. Wood, A. J. Wain, Faraday Discuss. 2013, 162, 57–75.
3] S. T. Marshall, M. O’Brien, B. Oetter, A. Corpuz, R. M. Richards, D. K.
Schwartz, J. W. Medlin, Nat. Mater. 2010, 9, 853–858.
Manuscript received: September 5, 2015
Revised: November 4, 2015
[
[
Accepted Article published: November 17, 2015
Final Article published: November 25, 2015
4] C. A. Schoenbaum, D. K. Schwartz, J. W. Medlin, Acc. Chem. Res. 2014,
4
7, 1438–1445.
Chem. Asian J. 2016, 11, 461 – 464
www.chemasianj.org
464
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim