Shape and ligand effect of palladium nanocrystals on furan hydrogenation

Article abstract of DOI:10.1039/C8NJ05152F

The Pd nanocrystals, including cubes, octahedra and wire, were prepared by shape-controlled solution phase reduction. The shape-dependent effect of Pd, and the effect of residual halogen ions and PVP, were investigated in selective hydrogenation of furan to tetrahydrofuran (THF). It was found that the residual halogen ions and PVP on the surface of Pd nanocrystals possibly reduced the hydrogenation activity and in turn it prevented the further reaction such as ring opening, so high selectivity towards THF was achieved even at high temperature. The 5-fold twinned wire displayed poor activity in furan hydrogenation due to a large amount of strongly adsorbed iodide ion residues covering most of the Pd active sites. An appropriate PVP residue is necessary, which can effectively maintain the shape and size stability of the Pd nanocube and octahedron, although the residual PVP partially blocks active Pd sites and reduces the activity for furan hydrogenation. The Pd nanocube enclosed by {100} facets exhibited about two times higher turnover frequency and lower apparent activation energy compared to the octahedron enclosed by {111} facets, suggesting a significant shape-dependent effect.

Full text of DOI:10.1039/C8NJ05152F

View Article Online
View Journal
NJC
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: C. SUN, Z. Cao, J.
Wan, L. Lin and X. Xie, New J. Chem., 2019, DOI: 10.1039/C8NJ05152F.
This is an Accepted Manuscript, which has been through the
Volume 40 Number
1 January 2016 Pages 1–846
Royal Society of Chemistry peer review process and has been
accepted for publication.
NJC
Accepted Manuscripts are published online shortly after
New Journal of Chemistry
www.rsc.org/njc
A journal for new directions in chemistry
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1144-0546
PAPER
Jason B. Benedict et al.
The role of atropisomers on the photo-reactivity and fatigue of
diarylethene-based metal–organic frameworks
rsc.li/njc

Page 1 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
Shape and ligand effect of palladium nanocrystals on furan
9
hydrogenation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Changyong Sun, Zhou Cao, Jiandian Wang, Liangbiao Lin, Xiaowei Xie 
Abstract: The Pd nanocrystals, including cube, octahedron and wire, were prepared by
shape-controlled solution phase reduction. The shape-dependent effect of Pd, together with
the effect of residual halogen ions and PVP, were investigated in selective hydrogenation
of furan to tetrahydrofuran (THF). It was found that the residual halogen ions and PVP on
the surface of Pd nanocrystals possibly reduced the hydrogenation activity in turn it
prevented the further reaction such as ring opening, so high selectivity towards THF was
achieved even at high temperature. The 5-fold twinned wire displayed poor activity in furan
hydrogenation due to a large amount of strongly adsorbed iodide ion residues covering
most of the Pd active sites. An appropriate PVP residue is necessary, which can effectively
maintain the shape and size stability of Pd nanocube and octahedron, although the residual
PVP partially block active Pd sites and reduce the activity for furan hydrogenation. The Pd
nanocube enclosed by {100} facets exhibited about two times higher turnover frequency
and lower apparent activation energy compared to the octahedron enclosed by {111} facets,
suggesting a significant shape-dependent effect.
Key words: Palladium nanocrystal; Furan; Catalytic hydrogenation; Shape sensitivity;
Ligand effect
Key Laboratory of Clean Chemical Technology, School of Chemical Engineering and Light Industry,
Guangdong University of Technology, Guangzhou 510006, China.
E-mail: xwxie@gdut.edu.cn; Tel. & fax: +86-20-39337595
† Electronic supplementary information (ESI) available.
1

New Journal of Chemistry
Page 2 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
1. Introduction
8
9
Extensive studies over the past few decades have demonstrated that the size and shape
of a catalyst particle on the nanoscale profoundly affect its reaction performance. In
particular, controlling the catalyst particle shape allows a selective exposure of the reactive
facets with desirable surface coordination, on which the active sites can be enriched and
tuned.1-5 Zhou et al. reported that the CeO2 nanorods with exposed {100} and {110} planes
showed a higher CO oxidation activity than the {111}-exposed irregular nanoparticles.2
Xie et al. demonstrated that the Co3O4 nanorods with predominantly exposed {110} planes
catalyzed CO oxidation at temperatures as low as -77 ºC even in a moist feed gas.3 In fact,
besides the activity enhancement, exploring and understanding shape-dependent effect can
also facilitate the catalytic selectivity improvement.4 Bratlie et al. discovered that the
exposed {100} surface on Pt cubic nanoparticles showed exclusive selectivity towards
cyclohexane in benzene hydrogenation, whereas the {100} and {111} planes on
cuboctahedral nanoparticles produced both cyclohexane and cyclohexene.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Palladium is a common noble metal active component with a broad range of industrial
applications in catalysis.6,7 Therefore, investigating its shape-dependent effect is of great
scientific importance and also of industrial interest.8 Shuai et al. demonstrated that the
catalytic reduction activity for waterborne contaminant removal varies with the Pd shape
and size using nitrite, N-nitrosodimethylamine and diatrizoate hydrogenation as the probe
reaction.9 Recently, Choi et al. reported shape-dependent effect in formic acid oxidation
reaction over a variety of Pd nanocrystals, including cubes, right bipyramids, octahedra,
tetrahedra, decahedra, and icosahedra. They discovered that the Pd nanocrystals enclosed
by {100} facets have higher specific activities than those enclosed by {111} facets and the
twin defects can greatly enhance formic acid oxidation activities.10 Nevertheless, till now,
the research for shape-dependent effect of palladium is scarce in contrast to its extensive
applications.
Furan and its derivatives can be massively obtained from biomass and therefore they are
2

Page 3 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
important platform molecules in hydrogenation and hydrodeoxygenation catalyzed by Pd,
Pt, Ni, Cu-based catalysts to produce biofuels and chemicals.11-20 Although furan and its
derivatives hydrogenation over Pd-based catalysts has been extensively studied through
experimental12-15 and theoretic methods21-25, the shape-dependent effect of palladium has
not been involved and discussed. Accordingly, in this work, we conducted shape-controlled
synthesis of monodispersed Pd nanocrystals and tested their catalytic performance to gain
insight into the shape-dependent effect of palladium nanocrystals on furan hydrogenation.
On the other hand, the residual surfactants and other surface contamination originating
from the nanocrystal synthesis process often affect the catalytic performance, even mask
or mislead the shape effect. Thus the ligand-effect of palladium nanocrystals was also
investigated.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. Experimental
2.1. Pd nanocrystals preparation
The Pd nanocrystals, including cube, octahedron and wire, were prepared by solution
phase reduction referenced to previous reports.26-28 In a typical synthesis of Pd cubes, 5.5
mL of an aqueous solution containing polyvinylpyrrolidone (PVP, MW=55000, 105 mg),
L-ascorbic acid (60 mg), KBr (75 mg) and KCl (141 mg) was pre-heated at 80 ºC for 5 min,
then 5.5 mL of Na2PdCl4 aqueous solution (50 mM) was added. The mixture was heated
at 80 ºC under magnetic stirring for 3 h and cooled down to room temperature. The products
were precipitated by acetone, separated via centrifugation at 12000 rpm, further purified
by thorough washing with ethanol, water or ethanol-acetone mixture, vacuum desiccated,
and finally collected. The procedure of Pd octahedra synthesis was similar to that of Pd
cubes except that PVP (105 mg), citric acid monohydrate (197.5 mg) and Na2PdCl4 (81
mg) were dissolved in a mixture solution containing 3 mL of ethanol and 8 mL of water.
In the synthesis of Pd wires, 1.0 mL of aqueous solution containing PVP (388.5 mg), KI
(332 mg), and 1.5 mL of NaOH aqueous solution (24 mM) and 2.0 mL of Na2PdCl4
3

New Journal of Chemistry
Page 4 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
aqueous solution (50 mM) were added into a 20 mL Teflon-lined stainless-steel autoclave,
heated at 200 ºC for 2 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2. Characterization
The Pd nanocrystals were characterized by TEM, HRTEM, XRD, FTIR, XPS, N2
physisorption, CO pulse chemisorption and TPO techniques. The experimental details of
characterization were shown in the ESI.
2.3. Catalytic test
The liquid-phase selective hydrogenation of furan was carried out in a 100 ml autoclave
equipped with a magnetic stirrer. The mass transfer limitation was eliminated when an
agitating speed of 1000 rpm was used. In a typical reaction, 13.5 mmol of furan, 19.0 ml
of ethanol, 0.18 g of n-tetradecane and 5 mg of Pd catalyst without any pretreatment were
charged into the reactor, which was flushed with N2 and further pressurized with H2 to the
desired pressure. Then the reaction temperature was raised to the desired value and
agitation was started. After different reaction periods, the products were analyzed by an
Agilent 7890B gas chromatograph equipped with a flame ionizing detector (FID) and a HP
Innowax or PlotAl2O3/S capillary column. The conversion of furan and the selectivity of
products were calculated on the basis of the mass balance of carbon using n-tetradecane as
internal standard. In some kinetics measurements, the constant H2 pressure was kept. All
the experiments were conducted in triplicate for reproducibility of the data and the results
were within an error of ±10 %.
3. Results and discussion
3.1. Characterization of as-prepared Pd nanocrystals
4

Page 5 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 1 TEM images, 3D models and size distributions of the as-prepared Pd nanocubes (a), octahedra
(b) and wires (c). The amplified inset of (b) shows that a small amount of tetrahedra and decahedra,
exposing {111} planes as well, exist in octahedral synthetic system.
Fig. 1 shows the TEM images of as-prepared Pd nanocrystals. It could be found that
monodispersed nanocubes (Fig. 1a) and wires (Fig. 1c) with uniform shape and size were
synthesized. In the octahedral synthetic system (Fig. 1b), a small amount of tetrahedra and
5

New Journal of Chemistry
Page 6 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
decahedra, exposing {111} planes as well, could be observed. The careful shape statistics
demonstrates that proportion of octahedra is more than 80% (Fig. S1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 2 HRTEM and FT images of Pd nanocubes (a, b), octahedra (c, d) and wires (e, f).
Fig. 2 shows the HRTEM images of as-prepared Pd nanocrystals. The fringes with lattice
spacing of 0.22 and 0.19 nm could be indexed to the {111} and {200} planes of metal Pd.
The detailed HRTEM analysis indicates that the cube and octahedron is Pd single crystal
enclosed by six {100} facets or eight {111} facets, whereas the wire is 5-fold twinned
crystal growing along <110> direction, exposing five {100} side planes.
6

Page 7 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
The size distributions of Pd nanocrystals are given in Fig. 1 a2, b2 and c2. Based on the
shape and size statistics, the structural features of the Pd nanocrystals could be estimated
and listed in Table 1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1 Structural features of the Pd nanocrystals derived from TEM statistics
Sample
Cube
Size (nm)
7.5 a
Calculated surface area (m2/g) Number of surface atoms (/g)
66.5
76.4
65.3
8.79×1020
1.17×1021
8.63×1020
Octahedron
Wire
8.1 a
6.0 b
a Edge length of cube or octahedron. b Diameter of wire, the corresponding side length of pentagon cross
section is 3.7 nm.
To further identify the structural features of the Pd nanocrystals, we tentatively
conducted N2 physisorption and CO pulse chemisorption to measure BET surface area and
metal surface area of the Pd nanocrystals. Unfortunately, all the results were not well
reproducible, fluctuated from 1 to 10 m2/g. The surface areas derived from adsorption
measurements are much smaller than the calculated values from TEM statistics, which is
due to the impact of surface residues. The surface residue analysis (see later sections) show
that PVP and halogen ions exist on the Pd surface. It is reasonable that microporous
condensation for N2 adsorption over aggregated samples was hindered by PVP, and halogen
ions was responsible for surface poisoning for CO adsorption. The fluctuation of measured
values is likely to be related to different degrees of aggregation caused by sample drying,
and PVP on the surface of aggregated samples seems to have an uncertain effect on
adsorption results. This limited us to obtain the specific surface area and active Pd surface
area through adsorption measurements. However, the TEM analysis showed that the dried
sample was easy to be re-dispersed into ethanol to monodisperse state. Considering the
same solvent used in the catalytic test, thus in later sections the discussion of catalytic
behavior and the calculation of TOF were based on the structural features of monodisperse
7

New Journal of Chemistry
Page 8 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
Pd nanocrystals, as shown in Table 1. The effect of surface residues will also be discussed.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(111)
(200)
(a)
(b)
Cube
(220)
(222)
(311)
Octahedron
(c) Wire
30
40
50
60
70
80
90
2 (degree)
Fig. 3 XRD patterns of Pd nanocubes (a), octahedra (b) and wires (c).
The XRD patterns of as-prepared Pd nanocrystals are depicted in Fig. 3. All the samples
exhibited nearly identical diffraction peaks at 39.8，46.3，67.9，81.9 and 86.3 degrees,
which could be indexed to {111}, {200}, {220}, {311}and {222} diffractions of face-
centered cubic Pd (PDF2 No.65-2867). No other diffraction peaks could be observed,
suggesting almost complete reduction of Pd species.
Briefly, three metal Pd nanocrystals with different shapes were successfully prepared.
3.2. Furan hydrogenation reaction
The liquid-phase selective hydrogenation of furan was conducted to test the as-prepared
metal Pd nanocrystals. Generally, furan hydrogenation over metal catalysts proceeds along
two pathways, ring hydrogenation to yield dihydrofuran (DHF) and then tetrahydrofuran
(THF) or ring opening reaction to form butanol, butene, butane, propene, propane etc.11,19,24
In all of our hydrogenation experiments operated at 25-150 ºC, only two kinds of ring
hydrogenation products formed, the major product THF and trace DHF. When the reaction
temperature was raised, furan conversion increased and selectivity changed little. No ring
opening products could be detected even at 150 ºC. This product distribution was quite
8

Page 9 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
different from the previous observation over Pt single-crystal surfaces and Pt nanoparticles,
where THF was dominant at low temperatures, butanol or propene was the main product
at high temperatures.19 The difference in catalytic performance between the two metals
partially depends on their intrinsic properties. Compared with Pt catalysts, Pd shows a weak
ability to break C-O and C-C bond, thus requiring higher temperatures to ring-opening
hydrogenation. Another possible explanation is that the residual PVP or halogen ions on
the surface of Pd nanocrystals reduced the hydrogenation activity in turn it prevented the
further reaction such as ring opening. Considering the stability of Pd nanocrystals, no
higher temperatures more than 150 ºC was attempted in our tests.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Furan
THF
100
75
50
25
0
100
75
50
25
0
Cube
wire
Octahedron
Fig. 4 Catalytic performance of different non-supported Pd nanocrystals for furan hydrogenation.
Reaction conditions: furan, 13.5 mmol; ethanol, 19 ml; n-tetradecane, 0.18 g; catalyst loading, 5 mg;
agitating speed, 1000rpm; H2 pressure, 4.0 MPa; temperature, 80 ºC; reaction period, 0.5 h.
Fig. 4 shows the catalytic performance of different Pd nanocrystals for furan
hydrogenation at 80 ºC under the H2 pressure of 4.0 MPa. After 0.5 h reaction, the furan
conversion over the nanocube, octahedron and wire was 100%, 61% and 11%, respectively.
The cube enclosed by {100} facets is more active for furan hydrogenation than the
octahedron enclosed by {111} facets. Similarly, the {100} facets of Pd nanocrystals were
reported to be more active for the nitrite hydrogenation at low initial nitrite concentration9
9

New Journal of Chemistry
Page 10 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
or for the formic acid oxidation10 than the other terrace facets, such as {111}. For Pd wire,
THF selectivity was 48% and no other product was detected. This carbon balance of far
below 100% is due to the positive deviation of conversion caused by adsorbed furan on
catalysts and volatilized furan in gas phase. This conversion deviation is significant at low
furan conversion, while at high furan conversion it can be neglected and the carbon balance
is close to 100%, as shown on the nanocube (98%) and octahedron (95%). Previous report10
also indicated that twin defects of Pd nanocrystals can greatly enhance formic acid
oxidation activities. However, the wire with twin defects displayed unexpected poor
activity in furan hydrogenation, although the 5-fold twinned wire and nanocube both
mainly expose {100} planes and have almost the same surface area (Table 1). It seems that
the twin defects on the wire do not enhance furan hydrogenation activity, but lower it.
Nevertheless, the presence of adsorbed species on the nanocrystal surface, which could
block the active sites or affect the electronic structures, i.e., the ligand effect, should be
considered.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.3. Effect of residual halogen ions
Pd (0)
(a)
(b)
Pd3d
Br3d
10000
500
Pd (II)
75
70
65
345
340
335
330
Binding Energy / eV
Binding Energy / eV
Pd (0)
(d)
(c)
Pd3d
Cl2p
200
10000
Pd (II)
345
340
335
330
210
205
200
195
190
Binding Energy / eV
Binding Energy / eV
Pd (0)
(e)
(f)
Pd3d
I3d
10000
50000
Pd (II)
345
340
335
330
645
640
635
630
625
620
615
610
Binding Energy / eV
Binding Energy / eV
10

Page 11 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
Fig. 5 Pd 3d and halogen XPS spectra of Pd nanocube (a, b), octahedron (c, d) and wire (e, f). The
corresponding surface atom ratio: Br/Pd = 0.13 for nanocube, Cl/Pd = 0.07 for octahedron, I/Pd = 0.22
for wire.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Halogen ions can be adsorbed on the surface of Pd crystal. In our preparations, Br- was
used for nanocube synthesis, I- was used for wire synthesis, and Cl- existed in all three
synthetic raw materials. As reported in literature, the adsorption strength to Pd surface
increases in the order of chloride < bromide < iodide.29 Thus it is more likely that a large
amount of strongly adsorbed iodide ion residues covering most of the Pd active sites is
responsible for the poor activity of the Pd wire. The XPS analysis (Fig. 5) supports the
above speculation. There is only one kind of halogen ion on the surface of each as-prepared
sample, Cl- on octahedron, Br- on nanocube and I- on wire. The corresponding surface atom
ratio is 0.07 (Cl/Pd) for octahedron, 0.13 (Br/Pd) for nanocube, and 0.22 (I/Pd) for wire,
indicating the wire has the most halogen residues. This large amount of strongly adsorbed
iodide ion residues modify the surface properties and mask the intrinsic activity of the Pd
wire. The nanocube and octahedron with a relatively small amount of halogen residues
displayed high enough catalytic activity, therefore the subsequent discussion would focus
on the nanocube and octahedron.
Considering that the residual bromide or chloride ions can also occupy the active sites,
the intrinsic activities of the nanocube and octahedron should be higher than the displayed
ones. Note, the surface Br/Pd ratio of the nanocube is larger than the Cl/Pd ratio of the
octahedron, if the bromide and chloride ions are assumed to exert the same influence on
Pd active sites, the activity decrease of the nanocube should be greater than that of the
octahedron. Even so, the nanocube is more active than the octahedron, as shown in Fig. 4.
Obviously, the nanocube has much higher intrinsic activity for furan hydrogenation than
the octahedron.
3.4. Effect of residual PVP
11

New Journal of Chemistry
Page 12 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
The shape and size stability of Pd nanocrystal catalysts is an essential prerequisite to
investigate the shape effect. In our synthesis of Pd nanocrystals, PVP was used as
stabilizers in colloidal synthesis to inhibit nanocrystal overgrowth and aggregation as well
as to control the structural characteristics of the resulted nanocrystals. On the other hand,
similar to the residual halogen ions, the residual PVP can modify the surface properties and
impact catalytic behavior.30-34 Thus it is necessary to explore the residual PVP.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.0
(a)
500
1.5
1.0
0.5
0.0
400
300
200
100
0
Cube
Octahedron
0
1000
2000
3000
Time (s)
2.0
1.5
1.0
0.5
0.0
(b)
500
400
300
200
100
0
Cube
Octahedron
0
1000
2000
3000
Time (s)
Fig. 6 TPO-MS analysis of surface carbon over as-prepared Pd nanocube (a) and octahedron (b).
Based on TPO-MS results of the as-prepared Pd nanocube and octahedron (Fig. 6), the
surface carbon amount, 3.81 mmol/g for cube and 0.93 mmol/g for octahedron, could be
obtained from the sum of CO2 and CO. Combined with the structural features of the Pd
nanocrystals showing in table 1, the ratio of residual PVP monomer to surface Pd atom was
estimated to be 0.43 for cube and 0.08 for octahedron. It seems that a considerable number
of active sites were occupied although the as-prepared Pd nanocrystals were thoroughly
12

Page 13 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
washed, and the cube should loss more active sites. Nevertheless, the nanocube with more
PVP residue displayed higher catalytic activity than the octahedron (shown in Fig. 4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
δ(C-H2)(C-H)C-H2
(a) Cube
(b) Octahedron
ν(C-H)
(c) Cube-NaBH4
ν(C=O)
ν(C-N)
(d) Octahedron-NaBH4
4000
3500
3000
2500
2000
1500
1000
500
Wavenumber(cm-1)
Fig. 7 FTIR spectra of the as-prepared Pd nanocubes (a), octahedra (b) and those subjected to NaBH4
washing (c, d).
FTIR measurements were performed to further explore the residual PVP over Pd
nanocrystals. As shown in Fig. 7 a and b, upon the as-prepared nanocube and octahedron,
the band of PVP at around 1660 cm−1 attributed to C=O stretch vibration, 1274, 1290, 1495
cm−1 attributed to N-C stretch vibrations linking with different carbon atoms, and 1320-
1463, 2850-2930 cm−1 attributed to C-H vibrations can be distinctly distinguished.
According to previous report, PVP molecule chemisorbs with its oxygen atom in the ring
on the surface of fine Pd nanocrystals capped with a large number of PVP molecules, while
PVP molecule chemisorbs with both the oxygen atom and nitrogen atom in the ring on the
surface of large Pd nanocrystals capped by a small number of PVP molecules.35 In our case,
the presence of all three N-C stretch vibrations suggests that no nitrogen atom chemisorbs
on the as-prepared Pd surface, and the red shift of the C=O stretch vibrational band can
hardly be observed, implying that most of the oxygen atoms do not chemisorb, i.e., most
of the PVP monomer just loosely deposit or suspend on Pd surface. Therefore, the actual
coverage of PVP on the as-prepared Pd nanocrystal surface is not as high as the TPO-MS
result, and the effect of residual PVP on Pd active sites is finite.
13

New Journal of Chemistry
Page 14 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 8 TEM images of the Pd nanocubes and octahedra after catalytic test (a, c) and NaBH4 washing (b,
d). The inset of (a) and (c) shows the size distribution diagram.
Pd nanocrystals with clean surface are highly expected in the study of shape effect.31-34
So an extra washing by NaBH4 were operated to completely remove the residual PVP on
the as-prepared Pd nanocrystals. As shown in Fig. 7 c and d, almost all the IR bands from
PVP disappear after NaBH4 washing. However, the PVP-free Pd nanocrystals are not stable
and tend to deform and aggregate in ethanol solution, as shown in Fig. 8 b and d, thus
cannot be used in catalytic test. This suggests that an appropriate PVP residue is necessary,
which can effectively maintain the shape and size stability of Pd nanocube and octahedron,
although it is possible that residual PVP partially block active Pd sites and reduce activity
for furan hydrogenation. TEM images and the corresponding size distributions (Fig. 8 a, c
and Fig. S2) further demonstrate that the nanocube and octahedron are stable enough, after
catalytic reaction they still have almost the same shape and size as the as-prepared ones.
3.5. Kinetics test and shape-dependent effect
14

Page 15 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
(b)
(a)
25℃
40℃
50℃
25℃
40℃
50℃
0.6
0.6
0.4
0.2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.4
0.2
0.0
0.5
1.0
0.0
0.5
1.0
t (h)
t (h)
2.0
1.5
1.0
0.5
0.0
2.0
1.5
1.0
0.5
0.0
(c)
(d)
25℃
40℃
50℃
25℃
40℃
50℃
0.0
0.5
1.0
0.0
0.5
1.0
t (h)
t (h)
Fig. 9 Catalytic performance of Pd nanocube (a, c) and octahedron (b, d) for furan hydrogenation. (a,
b), furan concentration vs. reaction time; (c, d), ln(C0/C) vs. reaction time. Reaction conditions: furan,
13.5 mmol; ethanol, 19 ml; n-tetradecane, 0.18 g; catalyst loading, 5 mg; agitating speed, 1000rpm;
constant H2 pressure, 4.0 MPa.
Fig. 9 shows the kinetics test result of furan hydrogenation over Pd nanocube and
octahedron under the constant H2 pressure of 4.0 MPa. In most cases, linear plots of ln(C0/C)
versus time could be obtained, indicating that the reaction under the constant H2 pressure
follows pseudo first order kinetics for furan. The only exception was furan hydrogenation
over Pd octahedron at 25 ºC, where the furan concentration versus time displayed better
linearity, showing a near zero order kinetics. Low temperatures resulted in the low surface
reaction rate, and thus the Pd octahedron surface was saturated by adsorbed furan. The
15

New Journal of Chemistry
Page 16 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
effect of hydrogen pressure was also investigated, as shown in Fig. S3. The reaction order
with respect to hydrogen is 0.16 and 0.19 over cube and octahedron, respectively, which is
far below the reaction order with respect to furan, indicating that furan hydrogenation
reaction is more affected by the concentration of furan.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2 Comparison of furan conversion, initial reaction rate, turnover frequency and apparent
activation energy for Pd cube and octahedron
Sample
Cube
Conv.(％) a
r0 (mol g-1 h-1) b
TOF (s-1)
1.22
Ea (kJ mol-1)
45.7
66
56
6.4
4.8
Octahedron
0.69
57.4
a Reaction conditions: furan, 13.5 mmol; ethanol, 19 ml; n-tetradecane, 0.18 g; catalyst loading, 5 mg;
agitating speed, 1000rpm; H2 pressure, 4.0 MPa; temperature, 50 ºC; reaction period, 0.5 h.
b Initial reaction rates were derived from kinetics test at 50 ºC.
4
cube
Slope = -5.5
2
0
octahedron Slope = -6.9
-2
-4
3.0
3.1
3.2
3.3
3.4
3.5
1000/T (1/K)
Fig. 10 Arrhenius plots for furan hydrogenation over Pd nanocube and octahedron.
To verify the shape-dependent effect of Pd nanocrystals on furan hydrogenation, the
intrinsic activity of Pd surface atom over nanocube and octahedron should be compared.
As shown in Table 2, the initial reaction rate of Pd nanocube was 6.4 mol g-1 h-1, 33%
higher than that of octahedron (4.8 mol g-1 h-1) under the selected reaction conditions. In
fact, this activity difference was based on catalyst mass, contained the contribution of
16

Page 17 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
different numbers of active sites. So turnover frequency (TOF) was calculated to describe
the activity of single active site. The TOF value of the nanocube was about two times higher
than that of the octahedron, as shown in Table 2, suggesting a significant shape-dependent
effect of Pd in furan hydrogenation. The Pd nanocube enclosed by {100} facets is more
active for furan hydrogenation than the octahedron enclosed by {111} facets. Moreover,
apparent activation energies for Pd nanocube and octahedron were estimated at a
temperature range of 25-50 ºC, and the corresponding Arrhenius plots were in Fig. 10. As
expected, the highly active Pd nanocube presented a lower Ea, 45.7 kJ mol-1, while the
octahedron had a higher one, 57.4 kJ mol-1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The above TOF calculations are based on the structural features of monodisperse Pd
nanocrystals from TEM statistics, not on the number of active sites from chemisorption
representing the effective atomic surface available in liquid phase reaction. As
aforementioned, the residual PVP and halogen ions partially blocked Pd surface sites and
reduced furan hydrogenation activity. Taking into account that more PVP and halogen ions
present on the nanocube than those on the octahedron, so the decrease of active site number
due to surface residues for nanocube should be more significant than that for octahedron,
thus the presence of residual PVP and halogen ions does not influence the validity of the
TOF comparison and the resulted shape-dependent effect. Compared to the Pd {111} facet,
the {100} facet showing higher furan hydrogenation activity is probably related to the
formation of stronger bonds with the reactants or intermediates, similar to the report in
nitrite hydrogenation.9
4. Conclusions
The Pd nanocrystals, including cube, octahedron and wire, were prepared by solution
phase reduction to explore the shape-dependent effect of Pd on furan hydrogenation. It was
found that an appropriate PVP residue is necessary, which can effectively maintain the
shape and size stability of Pd nanocube and octahedron, although it is possible that residual
PVP partially block active Pd sites and reduce activity for furan hydrogenation. The Pd
17

New Journal of Chemistry
Page 18 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
nanocube enclosed by {100} facets exhibited much higher activity for furan hydrogenation
than the octahedron enclosed by {111} facets, suggesting a significant shape-dependent
effect. Considering the complex surface species on nanocube and octahedron, the
electronic effects of residual PVP and halogen contributing to the difference in catalytic
activity could not be completely excluded. The 5-fold twinned wire displayed poor activity
in furan hydrogenation due to a large amount of strongly adsorbed iodide ion residues
covering most of the Pd active sites.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conflicts of interest
There are no conflicts to declare.
Acknowledgments
This work was financially supported by the National Natural Science Foundation of
China (21103182, 21273049), the Natural Science Foundation of Guangdong Province
(S2013050014127), and the Education Department Funding of Guangdong Province
(CGZHZD1104, 2013CXZDA016).
References
1 Y. Li and W. Shen, Chem. Soc. Rev., 2014, 43, 1543-1574.
2 K. B. Zhou, X. Wang, X. M. Sun, Q. Peng and Y. D. Li, J. Catal., 2005, 229, 206-212.
3 X. Xie, Y. Li, Z.-Q. Liu, M. Haruta and W. Shen, Nature, 2009, 458, 746-749.
4 K. An and G. A. Somorjai, ChemCatChem, 2012, 4, 1512-1524.
5 K. M. Bratlie, H. Lee, K. Komvopoulos, P. D. Yang and G. A. Somorjai, Nano Lett.,
2007, 7, 3097-3101.
6 Y. Nishihata, J. Mizuki, T. Akao, H. Tanaka, M. Uenishi, M. Kimura, T. Okamoto and
N. Hamada, Nature, 2002, 418, 164-167.
7 L. X. Yin and J. Liebscher, Chem. Rev., 2007, 107, 133-173.
8 H. Zhang, M. S. Jin, Y. J. Xiong, B. Lim and Y. N. Xia, Acc. Chem. Res., 2013, 46,
18

Page 19 of 21
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
1783-1794.
8
9
9 D. M. Shuai, D. C. McCalman, J. K. Choe, J. R. Shapley, W. F. Schneider and C. J.
Werth, Acs Catal., 2013, 3, 453-463.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10 S. I. Choi, J. A. Herron, J. Scaranto, H. W. Huang, Y. Wang, X. H. Xia, T. Lv, J. H. Park,
H. C. Peng, M. Mavrikakis and Y. N. Xia, Chemcatchem, 2015, 7, 2077-2084.
11 E. Furimsky, Appl. Catal. A-Gen., 2000, 199, 147-190.
12 S. D. Jackson, A. S. Canning, E. M. Vass and S. R. Watson, Ind. Eng. Chem. Res., 2003,
42, 5489-5494.
13 Z.-F. Jiao, X.-N. Guo, Z.-Y. Zhai, G.-Q. Jin, X.-M. Wang and X.-Y. Guo, Catal. Sci.
Technol., 2014, 4, 2494-2498.
14 C. Wang, L. Wang, J. Zhang, H. Wang, J. P. Lewis and F.-S. Xiao, J. Am. Chem. Soc.,
2016, 138, 7880-7883.
15 J. Yang, S. Li, L. Zhang, X. Liu, J. Wang, X. Pan, N. Li, A. Wang, Y. Cong, X. Wang
and T. Zhang, Appl. Catal. B-Environ., 2017, 201, 266-277.
16 S. Sitthisa, T. Pham, T. Prasomsri, T. Sooknoi, R. G. Mallinson and D. E. Resasco, J.
Catal., 2011, 280, 17-27.
17 S. Bhogeswararao and D. Srinivas, J. Catal., 2015, 327, 65-77.
18 H. Jeong, C. Kim, S. Yang and H. Lee, J. Catal., 2016, 344, 609-615.
19 C. J. Kliewer, C. Aliaga, M. Bieri, W. Huang, C.-K. Tsung, J. B. Wood, K. Komvopoulos
and G. A. Somorjai, J. Am. Chem. Soc., 2010, 132, 13088-13095.
20 C. Y. Sun, P. Zeng, M. H. He, X. J. He and X. W. Xie, Catal. Commun., 2016, 86, 5-8.
21 M. K. Bradley, D. A. Duncan, J. Robinson and D. P. Woodruff, Phys. Chem. Chem.
Phys., 2011, 13, 7975-7984.
22 V. Vorotnikov, G. Mpourmpakis and D. G. Vlachos, Acs Catal., 2012, 2, 2496-2504.
23 S. Wang, V. Vorotnikov and D. G. Vlachos, ACS Catal., 2015, 5, 104-112.
24 S. G. Wang, V. Vorotnikov and D. G. Vlachos, Green Chem., 2014, 16, 736-747.
25 Y. Xu, Top. Catal., 2012, 55, 290-299.
19

New Journal of Chemistry
Page 20 of 21
1
2
3
4
View Article Online
DOI: 10.1039/C8NJ05152F
5
6
7
8
9
26 H. C. Peng, S. F. Xie, J. Park, X. H. Xia and Y. N. Xia, J. Am. Chem. Soc., 2013, 135,
3780-3783.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27 M. Shao, J. Odell, M. Humbert, T. Yu and Y. Xia, J. Phys. Chem. C, 2013, 117, 4172-
4180.
28 X. Huang and N. Zheng, J. Am. Chem. Soc., 2009, 131, 4602-4603.
29 Y. Xiong, H. Cai, B. J. Wiley, J. Wang, M. J. Kim and Y. Xia, J. Am. Chem. Soc., 2007,
129, 3665-3675.
30 Z. Niu and Y. Li, Chem. Mater., 2014, 26, 72-83.
31 M. Crespo-Quesada, J.-M. Andanson, A. Yarulin, B. Lim, Y. Xia and L. Kiwi-Minsker,
Langmuir, 2011, 27, 7909-7916.
32 D. G. Li, C. Wang, D. Tripkovic, S. H. Sun, N. M. Markovic and V. R. Stamenkovic,
Acs Catal., 2012, 2, 1358-1362.
33 N. Naresh, F. G. S. Wasim, B. P. Ladewig and M. Neergat, J. Mater. Chem. A, 2013, 1,
8553-8559
34 N. Nalajala, W. F. G. Saleha, B. P. Ladewig and M. Neergat, Chem. Commun., 2014, 50,
9365-9368.
35 J. Xian, Q. Hua, Z. Jiang, Y. Ma and W. Huang, Langmuir, 2012, 28, 6736-6741.
20

Page 21 of 21
New Journal of Chemistry
1
2
3
4
5
View Article Online
DOI: 10.1039/C8NJ05152F
Shape and ligand effect of palladium nanocrystals on furan
6
7
hydrogenation
8
9
Changyong Sun, Zhou Cao, Jiandian Wang, Liangbiao Lin, Xiaowei Xie*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Guangdong University of Technology, Guangzhou 510006, China
Nanocube enclosed by {100} facets is the most active, residual PVP and halogen ions
occupy partial surface sites.