Polyhedral Cu2O to Cu pseudomorphic conversion for stereoselective alkyne semihydrogenation

Article abstract of DOI:10.1039/c7sc05232d

Cu₂O cubes, octahedra, and rhombic dodecahedra can be pseudomorphically converted into Cu crystals of corresponding shapes through reduction by ammonia borane in ethanol at 50 °C or below within 3 min, demonstrating the feasibility of making challenging polyhedral metal particles from metal oxide crystals. Hydrogen gas is also produced from ammonia borane in the process. The obtained Cu crystals have a slightly nanoporous interior. Addition of diphenylacetylene in the formation of Cu rhombic dodecahedra leads to complete stereoselective production of sterically hindered (Z)-stilbene. Semihydrogenation of other alkynes also gives pure (Z)-alkenes. Cu cubes and octahedra also showed considerable (Z)-stilbene selectivity along with minor formation of (E)-stilbene and bibenzyl as compared to CuCl₂ and commercial Cu₂O particles. Mechanistic studies reveal that the low binding affinity of alkenes on the rhombic dodecahedra surfaces leads to high product selectivity. These Cu crystals act as a green and low-cost catalyst for the synthesis of high-purity (Z)-alkenes.

Full text of DOI:10.1039/c7sc05232d

View Article Online
View Journal
Chemical
Science
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. Rej, M. Madasu,
C. Tan, C. Hsia and M. H. Huang, Chem. Sci., 2018, DOI: 10.1039/C7SC05232D.
This is an Accepted Manuscript, which has been through the
Volume
7 Number 1 January 2016 Pages 1–812
Royal Society of Chemistry peer review process and has been
accepted for publication.
Chemical
Science
Accepted Manuscripts are published online shortly after
www.rsc.org/chemicalscience
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 2041-6539
EDGE ARTICLE
Francesco Ricci et al.
Electronic control of DNA-based nanoswitches and nanodevices
rsc.li/chemical-science

Page 1 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
Polyhedral Cu2O to Cu pseudomorphic conversion for stereoselective alkyne
semihydrogenation†
Sourav Rej,‡ Mahesh Madasu,‡ ChihꢀShan Tan, ChiꢀFu Hsia, and Michael H. Huang*
Department of Chemistry, National Tsing Hua University, Hsinchu 30013, Taiwan
Eꢀmail: hyhuang@mx.nthu.edu.tw.
ORCID M. H. Huang: 0000ꢀ0002ꢀ5648ꢀ4345
†Electronic supplementary information (ESI) available. See DOI:
‡ These authors contributed equally to this work.
Abstract
Cu2O cubes, octahedra, and rhombic dodecahedra can be pseudomorphically converted into
Cu crystals of corresponding shapes through reduction by ammonia borane in ethanol at 50
ºC or below within 3 min, demonstrating the feasibility of making challenging polyhedral
metal particles from metal oxide crystals. Hydrogen gas is also produced from ammonia
borane in the process. The obtained Cu crystals have slightly nanoporous interior. Addition of
diphenylacetylene in the formation of Cu rhombic dodecahedra leads to complete
stereoselective production of sterically hindered (Z)ꢀstilbene. Semihydrogenation of other
alkynes also gives pure (Z)ꢀalkenes. Cu cubes and octahedra also showed considerable (Z)ꢀ
stilbene selectivity along with minor formation of (E)ꢀstilbene and bibenzyl as compared to
CuCl2 and commercial Cu2O particles. Mechanistic studies reveal low binding affinity of
alkenes on the rhombic dodecahedra surfaces leads to high product selectivity. These Cu
crystals act as green and lowꢀcost catalyst for the synthesis of highꢀpurity (Z)ꢀalkenes.
1

Chemical Science
Page 2 of 23
View Article Online
DOI: 10.1039/C7SC05232D
Introduction
Catalytic activity of inorganic nanocrystals has been shown to greatly depend on the exposed
surface facets.1‒5 Considering the lowꢀcost advantage of copper nanocrystals and their broad
applications in various catalytic reactions, it is highly desirable to make copper nanocrystals
with diverse shapes using benign reaction conditions.6 While cubic and octahedral copper
nanocrystals have been prepared, copper rhombic dodecahedra are unknown until
recently.7‒11 Cu2O crystals with these polyhedral shapes can be readily synthesized.1 Direct
conversion of Cu2O polyhedra to copper particles of corresponding morphologies is highly
useful, but such transformation has not been demonstrated until recently. Cu2O cubes,
octahedra, and rhombic dodecahedra were synthesized using different reagents, including the
addition of poly(vinylpyrrolidone) (PVP) to make octahedra.11 After a special heat treatment
to remove PVP and oleic acid, the Cu2O crystals were heated to 548 K under CO/N2 stream
for 2 h to obtain Cu crystals for corresponding shapes. However, the Cu surfaces are rough
and contain numerous pits. Sharp crystal faces are critically important for facetꢀdependent
examinations. It is also interesting to learn that examples of using Cuꢀbased nanomaterials for
alkyne semihydrogenation reactions are not available until recently.6,12 Stereoselective
semihydrogenation of internal alkynes to thermodynamically unstable and sterically hindered
pure (Z)ꢀalkenes is a powerful tool in industrial organic synthesis and necessary to
manufacture a wide variety of fine chemicals, drugs and bioactive natural products.13
Currently, the Lindlar catalyst has generally been used for (Z)ꢀalkene synthesis by
semihydrogenation of internal alkynes.14 This heterogeneous catalyst consists of costly
palladium deposited on calcium carbonate and poisoned with lead acetate.14 A large amount
of quinoline is also needed for higher product selectivity. Presence of toxic lead and excess
quinoline are the biggest drawbacks of the Lindlar catalyst. To overcome this problem, many
2

Page 3 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
modified nanocatalysts have been developed recently based on Pd,15−21 Ni,22,23 Au,24−26 and
other materials.27,28 In spite of all these advances, sometimes complex catalyst preparation
and harsh reaction conditions make them very hard to be reproduced by others even for
simple laboratoryꢀscale applications. Hence, a very simple, easily reproducible, low cost, and
green catalytic system is in high demand for synthesizing highꢀpurity (Z)ꢀalkenes.
In this work, we have shown that in the presence of ammonia borane (AB) and
ethanol, wellꢀdefined cubic, octahedral, and rhombic dodecahedral Cu2O crystals with
oxygen ions forming a bodyꢀcentered cubic crystal lattice and copper ions filling half of the
tetrahedral sites can be pseudomorphically converted into Cu crystals with a faceꢀcentered
cubic crystal structure maintaining the same particle shapes at room temperature.
Pseudomorphic transformations from Cu2O polyhedra to Cu2S, CdS, and ZnS cages have
been demonstrated.29,30 However, pseudomorphic conversion from a semiconductor to a
metal is not known until recently. The pseudomorphic transformation completes rapidly in a
few minutes with inꢀsitu generated hydrogen from AB. Hydrogen evolution from AB
decomposition in the presence of Au, Pd, and Cu nanocrystals has been reported.31,32 Cu
rhombic dodecahedra showed 100% stereoselectivity toward semihydrogenation of
diphenylacetylene (DPA) to (Z)ꢀstilbene in the presence of AB, although Cu cubes and
octahedra also displayed considerable product selectivity. Comparison of catalytic activity of
Cu crystals to those of CuCl2, commercial Cu2O, and Lindlar catalyst has been made. Timeꢀ
dependent formation of different products has also been studied. Mechanistic study reveals
that the lowest binding affinity for alkenes is the main reason for the highest stereoselectivity
of RD Cu crystals.
Results and discussion
Pseudomorphic formation of Cu polyhedra
3

Chemical Science
Page 4 of 23
View Article Online
DOI: 10.1039/C7SC05232D
Cu2O crystals with rhombic dodecahedral, cubic, and octahedral shapeswere synthesized at
room temperature by reacting an aqueoussolution of CuCl2 with NaOH and reducing
Cu(OH)2 or Cu(OH)42‒ with hydroxylamine hydrochloride (Fig. 1a–c).33,34 Highꢀangle
annular brightꢀfield scanning transmission electron microscopy (HAABFꢀSTEM) images
with energyꢀdispersive Xꢀray spectroscopy (EDS) line scans of rhombic dodecahedral, cubic,
and octahedral Cu2O crystals show perfect particle geometries and that the crystals are
composed of Cu and O atoms (Fig. 1d–f). To prepare Cu polyhedra, an ethanolic solution of
Cu2O crystals was treated with AB at room temperature for 5 to 9 minutes (Table S1, ESI†).
Addition of AB as a reducing agent leads to remarkable pseudomorphic conversion of Cu2O
crystals to Cu particles, showing the feasibility of metal polyhedra formation through direct
oneꢀpot pseudomorphic transformation of their corresponding metal oxides for the first time.
When we used hydrazine as the reducing agent, many Cu rhombic dodecahedra showed
etched faces. When sodium borohydride was used as the reducing agent, the resulting
particles have lost their original morphology (Fig. S1, ESI†). Scanning electron microscopy
(SEM) images confirm pseudomorphic formation of Cu rhombic dodecahedra, cubes, and
octahedra from Cu2O crystals of corresponding shapes (Fig. 1g–i). Size distribution
histograms show that the obtained Cu crystals retain the sizes of the original Cu2O particles
(Fig. S2, ESI†). The Cu rhombic dodecahedra, cubes, and octahedra have sizes of 262 ± 20,
183 ± 11, and 387 ± 45 nm, respectively. HAABFꢀSTEM images reveal that the resultant Cu
crystals contain some nanopores in the crystal interior (Fig. 1j–l). EDS line scans on these
crystals give only intense and continuous signals for Cu. Oxygen signals are very weak and
discontinuous and are considered only the background noise (Fig. 1m–o), proving essentially
complete loss of oxygen ions in the crystal lattice after the reduction process. Powder XRD
patterns (Fig. 1p) establish that the diffraction peaks for Cu2O cubes, octahedra, and rhombic
dodecahedra (JCPDS #77ꢀ0199) have been replaced entirely with those of Cu with a fcc
4

Page 5 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
lattice (JCPDS #85ꢀ1326). The compositional change can be readily observed visually, and it
is for this obvious solution color change from orange Cu2O crystals to reddish brown Cu
particles that further structural analysis of the resulting crystals was performed (Fig. 1q–v).
UVꢀvisible spectra of the three Cu2O particle shapes and their corresponding Cu products are
provided in Fig. S3, ESI†. Due to the large particle sizes, light scattering effects dominate the
absorption feature in the nearꢀinfrared region for Cu2O crystals. The synthesized Cu crystals
give a distinctive surface plasmon resonance peak in the range of 625 to 650 nm.35 Light
scattering effects in the nearꢀinfrared region are also obvious because of their large particle
sizes. Xꢀray photoelectron spectroscopy analysis (XPS) on Cu2O and Cu rhombic
dodecahedral crystals show only slight shifts in the Cu 2p peaks from 932.1 and 951.9 eV in
Cu2O crystals to 932.3 and 952.1 eV in Cu crystals with absence of satellite peaks (Fig. S4
and S5, ESI†).9,36 However, more distinct difference from the Cu LMM region can be used to
differentiate Cu2O crystals from Cu crystals.37
Largeꢀarea and singleꢀparticle transmission electron microscopy (TEM) images of Cu
rhombic dodecahedra, cubes, and octahedra reveal more details of the internal structures of
the Cu crystals (Fig. 2a–f). The particles appear lightly porous. However, compared to the
transformation from polyhedral Cu2O crystals to Cu2S hollow cages upon Na2S addition, the
Cu crystals retain most of their solid interior.28,29 Selectedꢀarea electron diffraction (SAED)
patterns (Fig. 2g–i) taken on a single Cu rhombic dodecahedron gives a ring pattern
indicating polycrystalline nature of the particle. However, the Cu cube and octahedron are
largely singleꢀcrystalline. An interesting 2ꢀfold symmetrical SAED pattern was recorded for
the Cu octahedron. The acidic solution condition in the preparation of Cu2O rhombic
dodecahedra naturally makes some particles slightly etched. It is unclear if this structural
feature leads to more local rather than uniform rate of the reduction process giving
polycrystalline Cu rhombic dodecahedra. The largeꢀarea and magnified HRꢀTEM images of a
5

Chemical Science
Page 6 of 23
View Article Online
DOI: 10.1039/C7SC05232D
Cu rhombic dodecahedron, cube, and octahedron reveal lattice fringes with interplanar
spacings of 1.27, 1.80, and 2.04 Å corresponding to the (220), (200), and (111) lattice planes
of fcc Cu (Fig. 2j–o). When lattice fringes close to the edge of the rhombic dodecahedron are
examined (Fig. S6, ESI†), the planes perpendicular to the crystal edge yield a dꢀspacing
consistent with the (110) planes of Cu, suggesting (110) planes are exposed in the Cu
rhombic dodecahedra. (111) planes have also been found to align perpendicular to the edge of
an octahedron (Fig. S7, ESI†). Fig. 2 also shows timeꢀdependent TEM images illustrating
how Cu2O rhombic dodecahedra are converted into Cu rhombic dodecahedra. The particle
surfaces become slightly rough 30 sec after the addition of AB. The reduction process occurs
so rapidly that after 60 sec a great extent of the particle interior appears porous, but the
central portion seems less affected. The reduction process is mostly completed within 2 min
after AB addition to give a more uniformly nanoporous appearance. Fig. S8, ESI† provides a
schematic diagram for the mechansim of this pseudomorphic transformation. In ethanol, BH3
unit from AB undergoes ethanolysis and releases hydride (H−) ions, which synchronously
reduces Cu+−NH3 units to Cu metal with liberating NH3 and hydrogen gas molecules.
Inductively coupled plasma mass spectrometry (ICPꢀMS) analysis gave a Cu concentration of
~ 0.039 mg/L from the top solution after the formation of Cu rhombic dodecahedra. This low
concentration of Cu in the solution indicates complete reduction of Cu+ ions in the crystals to
Cu0 atoms.
Stereoselective semihydrogenation of alkynes
In order to develop an easily reproducible, oneꢀpot, lowꢀcost green catalytic system for
stereoselective semihydrogenation of DPA to (Z)ꢀstilbene, Cu2O crystals exposing different
facets have been employed as the catalysts. DPA and anhydrous ethanolic solution of AB
6

Page 7 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
were added simultaneously to an ace pressure tube (see Fig. S9, ESI†). Within 3 min, the inꢀ
situ pseudomorphically formed Cu crystals then act as the active catalyst for this
semihydrogenation reaction without isolation (Fig. 3a). In this whole process, ammonia
borane, which is a highꢀdensity hydrogen storage material,38 acts as a reducing agent for Cu
crystal formation and then produces quantitative hydrogen gas required for the catalytic
reaction (Fig. 4). It should be mentioned that hydrogen is present in the form of hydrogen
atoms on the copper surface from the dissociation of ammonia borane, which initiate the
semihydrogenation reaction. So there is no need to dissociate H2 molecule, which generally
requires higher activation energy on Cu surface. This is the advantage of our reaction by
which one can easily overcome the higher activation energy for H2 dissociation on Cu surface.
Conversion of DPA and concentrations of three different products at different time intervals
were studied for Cu rhombic dodecahedra (Fig. 3b), cubes (Fig. 3c), and octahedra (Fig. 3d).
In all these studies, proton nuclear magnetic resonance (1HꢀNMR) spectrum of the crude
reaction product mixture was taken without performing any column chromatography to
determine the yields of different products at different time points. As evidenced by timeꢀ
dependent 1HꢀNMR spectra (Fig. S10, ESI†), Cu rhombic dodecahedra showed the highest
100% conversion of DPA in 45 min at 50 ºC and the highest product selectivity yielding
99.5% of (Z)ꢀstilbene using 0.4 mmol of ammonia borane. Lowering the amount of ammonia
borane added gave lower product conversion percentages (see Table S2). After the reaction,
the Cu rhombic dodecahedra kept their wellꢀdefined shape and nanoporous interior structure
(Fig. S11, ESI†). XRD patterns give only Cu reflection peaks (Fig. S12, ESI†). XPS
spectrum taken after 3 min of the semihydrogenation reaction shows the characteristic Cu
LMM peak feature (Fig. S13, ESI†). ICPꢀMS analysis gave a Cu concentration of ~ 0.139
mg/L from the clear reaction mixture after 45 min of reaction (as shown in Fig. S9, ESI†).
This low concentration of Cu in the solution further confirms the high stability of Cu RD
7

Chemical Science
Page 8 of 23
View Article Online
DOI: 10.1039/C7SC05232D
crystals and the heterogeneous nature of the catalytic reaction without any contribution from
dissolved Cu ions. In addition, hot filtration test was also performed. After 15 min of reaction,
we filtered the reaction solution and removed all the catalyst carefully. The filtered reaction
mixture was kept at 50 ºC for 30 min under constant stirring, but no increase in product
formation was observed, which clearly confirmed the heterogeneous nature of the reaction.
We also performed a blank test where only ammonia borane and DPA were used to carry out
the semihydrogenation reaction at 50 ºC for 45 min without using any Cu2O catalyst, but no
conversion of DPA was observed.
To test the recyclability of the catalyst, we have carried out one more catalytic cycle
under the standard condition using the same Cu rhombic dodecahedral catalyst isolated from
the first catalytic cycle (see Fig. S14, ESI†). The DPA conversion was 93.5%, whereas
selectivity for (Z)ꢀstilbene, (E)ꢀstilbene, and bibenzyl are 93.4%, 5.6% and 1%, respectively.
SEM image of the particles taken after the second catalytic cycle shows that the Cu RD
crystals maintain their shape. When the reaction was carried out at room temperature (25 ºC),
Cu rhombic dodecahedra took just 90 min to complete the reaction with 97.2% conversion of
DPA and 98.2% selectivity for (Z)ꢀstilbene (Fig. S15, ESI†). For Cu cubes at 50 ºC,
conversion of DPA dropped to 95.6% after 60 min of reaction with a relatively low (Z)ꢀ
stilbene selectivity of 92.1% (Fig. 3C and Fig. S16, ESI†). Cu octahedra showed 97% DPA
conversion and 95.1% selectivity for (Z)ꢀstilbene after 60 min of reaction (Fig. 3d and Fig.
S17, ESI†). Both Cu cubes and octahedra are also appeared as good catalysts considering the
simple and mild reaction conditions.
We also evaluated the catalytic activity and product selectivity of CuCl2, commercial
Cu2O and Lindlar catalyst for the same reaction (Fig. 3e). CuCl2 (see Fig. S18, ESI†) and
commercial Cu2O (Fig. S19, ESI†) showed conversion of DPA at 81.5% and 95.5% with
comparatively lower selectivity for (Z)ꢀstilbene at 89.2% and 90.9%, respectively.
8

Page 9 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
Commercial Lindlar catalyst (Fig. S20, ESI†) in the presence of hydrogen gas and quinoline
displayed 52% selectivity towards (Z)ꢀstilbene at 96% conversion of DPA with 48%
undesired bibenzyl formation (Fig. S21, ESI†). When AB was used to provide hydrogen gas
in the absence of quinoline, Lindlar catalyst produced only bibenzyl as the exclusive product
after 15 min of reaction and has lost its selectivity for (Z)ꢀstilbene (Fig. S22, ESI†).
Cu2O conversion to Cu crystals has been tested using different reducing agents such
as hydrazine (Fig. S23, ESI†) and sodium borohydride (Fig. S24, ESI†) with Cu2O RD
catalyst. The conversion of DPA for hydrazine and sodium borohydride dropped to only 7.8
% and 36.5 % with 50.2 % and 90.1 % selectivity for (Z)ꢀstilbene, respectively. The results
show the importance of AB in this reaction. Table 1 summarizes all the catalysis results. Cu
rhombic dodecahedra in conjunction with AB are the best catalytic system for the
semihydrogenation of internal alkynes with unsurpassed conversion and stereoselectivity. To
further demonstrate the general applicability of the Cu rhombic dodecahedra, stereoselective
semihydrogenation of 1ꢀmethoxyꢀ4ꢀ(phenylethynyl)benzene (Fig. 3f and Fig. S25, ESI†) and
1,4ꢀbutynediol (Fig. 3g and Fig. S26, ESI†), an aliphatic molecule, have been carried out. In
both cases, the DPA conversion was 100% with remarkably 100% selectivity for (Z)ꢀalkenes
in just 45 min.
Mechanism of high catalytic stereoselectivity
According to the HoriutiꢀPolanyi mechanism,22,39 initially DPA gets adsorbed on Cu surface
in the presence of hydrogen atoms (Fig. S27, ESI†). In the first hydrogenation step via a cisꢀ
addition of two hydrogen atoms, DPA molecule is converted to (Z)ꢀstilbene. To confirm the
source of both hydrogen atoms in the product alkene and their mode of addition,
semihydrogenation of DPA was carried out under our standard conditions but replacing
anhydrous EtOH with anhydrous EtOD. As evidenced by 1HꢀNMR spectrum (Fig. S28, ESI†),
9

Chemical Science
Page 10 of 23
View Article Online
DOI: 10.1039/C7SC05232D
semihydrogenation of DPA in EtOD afforded the nonꢀdeuterated (Z)ꢀstilbene, indicating that
both hydrogen atoms come from AB and get attached via cisꢀmode. Once (Z)ꢀstilbene is
formed on the surface, it can desorb or react further to the unwanted formation of bibenzyl
via a second hydrogenation step22 or (E)ꢀstilbene by cisꢀtrans isomerization (Fig. S27,
ESI†).39 An efficient selective catalyst for the semihydrogenation of alkynes is characterized
by having a high stability of adsorbed alkynes and a low stability of alkenes. The former
defines the overall conversion of alkynes, whereas the latter measures the product selectivity.
An inꢀdepth study for the adsorption behavior of acetylene (C2H2) and ethylene (C2H4) on Cu
(111) plane has been investigated.40 The results clearly indicate that acetylene gets
chemisorbed on Cu {111} surface with higher stability, whereas ethylene can be found as a
weakly perturbed physisorbed species on Cu (111) plane. In addition, there are some reports
showing that alkenes have the lowest binding affinity on Cu surface as compared to alkynes,
which makes Cu highly desirable candidate for industrial gasꢀphase semihydrogenation
reaction over Pd.40‒42 In fact, continued reaction in the presence of (Z)ꢀstilbene and Cu
rhombic dodecahedra does not lead to additional formation of bibenzyl (Fig. S29, ESI†). So
(Z)ꢀstilbene should preferentially desorb from the Cu {110} surface after their formation,
resulting in the highest product stereoselectivity for rhombic dodecahedral nanocrystals, and
this is another example of crystal facetꢀdependent organocatalysis.
Conclusions
The pseudomorphic formation of Cu rhombic dodecahedra, cubes, and octahedra at 30‒50 ºC
from Cu2O crystals of corresponding shapes through simple ammonia borane reduction in
ethanol has been demonstrated, opening up a new way for polyhedral metal nanocrystal
formation via direct metal oxide conversion. Hydrogen atoms derived from ammonia borane
are directly used for the Cu crystalꢀcatalyzed stereoselective semihydrogenation of alkynes.
10

Page 11 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
Cu rhombic dodecahedra showed 100% conversion and exceptional cisꢀproduct selectivity.
The high stereoselectivity of Cu rhombic dodecahedra is likely due to their exposure of {110}
faces. This lowꢀcost and highly efficient catalytic system should find broad applications.
Experimental
Chemicals
Anhydrous copper(II) chloride (CuCl2; 97%) and hydroxylamine hydrochloride
(NH2OH·HCl; 99%) were purchased from Aldrich. Sodium hydroxide (98.2%) and sodium
dodecyl sulfate (SDS; 100%) were acquired from Mallinckrodt. Diphenylacetylene (DPA;
98%), ammonia borane (H3N‒BH3; 97%), anhydrous ethanol (99.5%), ethanolꢀd (C2H5OD;
99 atom % D), Cu2O (99.99%), and Lindlar catalyst (5% palladium deposited on calcium
carbonate poisoned with lead) were purchased from Aldrich. All chemicals were used as
received without further purification. Ultrapure distilled and deionized water (18.3 Mꢁ) was
used for all solution preparations. Commercially available reagents were used for the
semihydrogenation reactions.
Synthesis of Cu2O rhombic dodecahedral and cubic nanocrystals
For the large scale synthesis, 0.348 g of SDS was added in two separate 50ꢀmL glass vials,
followed by the addition of 27.80 and 35.68 mL deionized water for the synthesis of Cu2O
rhombic dodecahedra and cubes, respectively. The sample vials were sonicated until SDS
was dissolved completely and then placed in a water bath set at 31 ºC. A solution of CuCl2
(0.1 M, 2 mL) was added to each of the vials with shaking for 1 min. For rhombic
dodecahedra, NaOH (1.0 M, 0.72 mL) and NH2OH·HCl solutions (0.1 M, 9.48 mL) were
added into the vial with 3 s interval, then shook for 10 s and kept in the water bath for 1 h for
crystal growth. Similarly, for cubic crystals, NaOH (1.0 M, 0.72 mL) and NH2OH·HCl
solutions (0.1 M, 0.4 mL) were added into the vial. The total volume of the final solution is
11

Chemical Science
Page 12 of 23
View Article Online
DOI: 10.1039/C7SC05232D
40 mL in each case. After 1 h, the reaction mixtures were centrifuged at 5000 rpm for 3 min.
After decanting the top solution, the precipitate was washed with 40 mL of ethanol for three
times to remove unreacted chemicals and SDS surfactant. The precipitate was dispersed in 4
mL of ethanol for conversion to Cu crystals. For semihydrogenation reaction, the final
precipitates were dried under high vacuum for overnight.
Synthesis of Cu2O octahedral nanocrystals
For the large scale synthesis, 0.348 g of SDS was added to a 50ꢀmL glass vial, followed by
the addition of 36.08 mL of deionized water. The sample vial was sonicated until SDS
became dissolved completely and then placed in a water bath set at 31 ºC. A solution of
CuCl2 (0.1 M, 0.4 mL) was added to the vial with shaking for 1 min. After that, NaOH (1.0
M, 0.8 mL) and NH2OH·HCl solutions (0.2 M, 2.72 mL) were added into the vial with 3 s
interval, then shook for 10 s and kept in the water bath for 2 h for crystal growth. The total
volume of the final solution is 40 mL. After 2 h, the reaction mixture was centrifuged at 5000
rpm for 3 min. After decanting the top solution, the precipitate was washed with 40 mL of
ethanol for three times to remove unreacted chemicals and SDS surfactant. The precipitate
was dispersed in 4 mL of ethanol. The final precipitate was dried under high vacuum for
overnight.
Synthesis of Cu polyhedra
For the pseudomorphic formation of Cu polyhedra from Cu2O crystals, 0.036 g of ammonia
borane was dissolved in 7.5 mL of ethanol with stirring in a 20ꢀmL glass vial, which was kept
in a water bath at 30 ºC. From the aboveꢀmentioned 4 mL concentrated Cu2O nanocrystal
solution, 2.4 mL of the solution was withdrawn and added to the ammonia borane solution
with continuous stirring. For rhombic dodecahedral, cubic, and octahedral nanocrystals, the
solution was stirred for 3, 5, and 7 min respectively until the solution completely changed to
reddish brown. After that, stirring stopped and the particles were aged for another 2 min to
12

Page 13 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
allow the particles to settle down. After carefully withdrawing the top solution, the precipitate
was washed with 10 mL of ethanol for three times just by shaking for 2 min to remove the
chemicals. The synthesized Cu polyhedra were dispersed in 1 mL of anhydrous ethanol by
sonication for further characterization.
Semihydrogenation catalysis
In a typical catalytic process, 4 mg of Cu2O nanocrystals (0.056 mmol Cu) and 1 mL of
anhydrous ethanol were added to a 25ꢀmL ace pressure tube and sonicated for 1 min to
disperse the catalyst homogeneously in the solution. 1 mL of anhydrous ethanolic solution of
diphenylacetylene (0.0178 g, 0.1 mmol) and 1 mL of anhydrous ethanolic solution of
ammonia borane (0.0123 g, 0.4 mmol) were added one by one and then sealed the tube. The
ace pressure tube was placed in a water bath kept at 50 ºC with continuous stirring. After 3
min, the solution color changed to reddish brown confirming the inꢀsitu pseudomorphic
formation of Cu crystals, which then acted as the active catalyst for the semihydrogenation of
alkynes. After completion of the reaction, the clear top anhydrous ethanolic reaction mixture
was carefully collected, leaving the Cu crystals behind, and the solvent was removed under
reduced pressure. 4 mL of dichloromethane was added to extract the products and subjected
to evaporation to obtain the crude compound, which was directly used for 1HꢀNMR analysis.
Timeꢀdependent study for the conversion of DPA and selectivity of different products
catalyzed by Cu RDs, cubes, and octahedra were also done by collecting 1HꢀNMR spectra of
the crude isolated product. In all cases, 4 mg of Cu2O nanocrystals were used, keeping a fixed
Cu molar mass (i.e. 0.056 mmol).We further used CuCl2 (7.5 mg, 0.056 mmol Cu),
commercial Cu2O (4 mg, 0.056 mmol Cu), and commercial Lindlar catalyst (121.7 mg, 0.056
mmol Pd) under the same reaction conditionsto evaluate their catalytic activities and product
selectivity relative to that of Cu rhombic dodecahedra. For Lindlar catalyst, we used
hydrogen gas in the presence of 100 µL of quinoline. When ammonia borane was used as a
13

Chemical Science
Page 14 of 23
View Article Online
DOI: 10.1039/C7SC05232D
source of hydrogen, no quinoline was added.To evaluate the generalapplicability,
semihydrogenation of internal alkyne was performed using different alkynes. Cu rhombic
dodecahedra were chosen as the catalyst.
Deuterium labelling experiments
Under the protection of nitrogen gas, 4 mg of Cu2O rhombic dodecahedra and 1 ml of
ethanolꢀd(C2H5OD) were added to a 25ꢀmL Schleck tube equipped with a magnetic stir bar,
sonicated for 1 min to disperse the catalyst homogeneously in the solution. Next, 1 mL of
ethanolꢀd solution of diphenylacetylene (0.0178 g, 0.1 mmol) and 1 mL of ethanolꢀd solution
of ammonia borane (0.0123 g, 0.4 mmol) were added. The Schleck tubewas then placed in a
water bath kept at 50 ºC with continuous stirring. After completion of the reaction, the clear
top ethanolꢀd solution was carefully collected and the solvent was removed under reduced
pressure. 4 mL of dichloromethane was added to extract the products and subjected to
evaporation to obtain the crude compound, which was directly used for 1HꢀNMR analysis.
Characterization
SEM images of the samples were obtained using a JEOL JSMꢀ7000F electron microscope.
TEM characterization was performed on a JEOL JEMꢀ2100 microscope with an operating
voltage of 200 kV. Elemental mapping images were acquired by EDS using the same JEOL
JEMꢀ2100 electron microscope equipped with a STEM unit and an Inca Energy 250 detector
from Oxford Instruments. XPS characterization was carried out on a ULVACꢀPHI Quantera
SXM highꢀresolution XPS spectrometer. A monochromatized Al anode was used as the
excitation source. The C 1s peak was chosen as the reference line. XRD patterns were
recorded on a Shimadzu XRDꢀ6000 diffractometer with Cu Kα radiation. UV−vis absorption
spectra were taken using a JASCO Vꢀ670 spectrophotometer.
Acknowledgments
14

Page 15 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
This work was funded by the Ministry of Science and Technology of Taiwan (Grants 104ꢀ
2119ꢀMꢀ007ꢀ013ꢀMY3, 104ꢀ2118ꢀMꢀ007ꢀ073, and 105ꢀ2811ꢀMꢀ007ꢀ017).
Notes and references
1
2
3
Y. Shang and L. Guo, L. Adv. Sci., 2015, 2, 1500140.
K. Chanda, S. Rej, S.ꢀY. Liu and M. H. Huang, ChemCatChem., 2015, 7, 1813‒1817.
M. Kim, Y. Kim, J. W. Hong, S. Ahn, W. Y. Kim and S. W. Han, Chem. Commun.,
2014, 50, 9454‒9457.
4
S. Rej, K. Chanda, C.ꢀY. Chiu and M. H. Huang, Chem.‒Eur. J., 2014, 20,
15991‒15997.
5
6
M. Madasu, C.ꢀF. Hsia and M. H. Huang, Nanoscale, 2017, 9, 6970‒6974.
M. B. Gawande, A. Goswami, F.ꢀX. Felpin, T. Asefa, X. Huang, R. Silva, X. Zou, R.
Zboril and R. S. Varma, Chem. Rev., 2016, 116, 3722‒3811.
H.ꢀJ. Yang, S.ꢀY. He, H.ꢀL. Chen and H.ꢀY. Tuan, Chem. Mater., 2014, 26, 1785‒1793.
S.ꢀC. Lu, M.ꢀC. Hsiao, M. Yorulmaz, L.ꢀY. Wang, P.ꢀY.; Yang, S. Link, W.ꢀS. Chang
and H.ꢀY. Tuan, Chem. Mater., 2015, 27, 8185‒8188.
7
8
9
C.ꢀF. Hsia, M. Madasu and M. H. Huang, Chem. Mater., 2016, 28, 3073‒3079.
T. Subashchandrabose, M. Madasu, C.ꢀF. Hsia, S.ꢀY. Liu and M. H. Huang, Chem.
Asian J., 2017, 12, 2318‒2322.
10
11
12
Z. Zhang, S.ꢀS. Wang, R. Song, T. Cao, L. Luo, X. Chen, Y. Gao, J. Lu, W.ꢀX. Li and
W. Huang, Nat. Commun., 2017, 8, 488.
A. Fedorov, H.ꢀJ. Liu, H.ꢀK. Lo and C. Copéret, J. Am. Chem. Soc., 2016, 138,
16502‒16507.
13
14
C. Oger, L. Balas, T. Durand and J.ꢀM. Galano, Chem. Rev., 2013, 113, 1313‒1350.
H. Lindlar and R. Dubuis, Org. Synth., 1966, 46, 89.
15

Chemical Science
Page 16 of 23
View Article Online
DOI: 10.1039/C7SC05232D
15
16
D. Teschner, J. Borsodi, A. Wootsch, Z. Révay, M. Hävecker, A. KnopꢀGericke, S. D.
Jackson and R. Schlögl, Science, 2008, 320, 86‒89.
C. W. A. Chan, A. H. Mahadi, M. M.ꢀJ. Li, E. C. Corbos, C. Tang, G. Jones, W. C. H.
Kuo, J. Cookson, C. M. Brown, P. T. Bishop and S. C. E. Tsang, Nat. Commun., 2014,
5, 5787.
17
18
19
20
E. D. Slack, C. M. Gabriel and B. H. Lipshutz, Angew. Chem., Int. Ed., 2014, 53,
14051‒14054.
T. Mitsudome, Y. Takahasi, S. Ichikawa, T. Mizugaki, K. Jitsukawa and K. Kaneda,
Angew. Chem., Int. Ed., 2013, 52, 1481‒1485.
S. Yun, S. Lee, S. Yook, H. A. Patel, C. T. Yavuz and M. Choi, ACS Catal., 2016, 6,
2435‒2442.
W. Long, N. A. Brunelli, S. A. Didas, E. W. Ping, C. W. Jones, ACS Catal., 2013, 3,
1700‒1708.
21
22
M. Laskar and S. E. Skrabalak, ACS Catal., 2014, 4, 1120‒1128.
F. Studt, F. AbildꢀPedersen, T. Bligaard, R. Z. Sørensen, C. H. Christensen,J. K.
Nørskov, Science, 2008, 320, 1320‒1322.
23
Y. Liu, X. Liu, Q. Feng, D. He, L. Zhang, C. Lian, R. Shen, G. Zhao, Y. Ji, D. Wang,
G. Zhou and Y. Li, Adv. Mater., 2016, 28, 4747‒4654.
24
25
G. Li and R. Jin, J. Am. Chem. Soc., 2014, 136, 11347‒11354.
T. Mitsudome, M. Yamamoto, Z. Maeno, T. Mizugaki, K. Jitsukawa and K. Kaneda, J.
Am. Chem. Soc., 2015, 137, 13452‒13455.
26
M. Yan, T. Jin, Y. Ishikawa, T. Minato, T. Fujita, L.ꢀY. Chen, M. Bao, N. Asao, M.ꢀW.
Chen and Y. Yamamoto, J. Am. Chem. Soc., 2012, 134, 17536‒17542.
16

Page 17 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
27
S. G. Kwon, G. Krylova, A. Sumer, M. M. Schwartz, E. E. Bunel, C. L. Marshall, S.
Chattopadhyay, B. Lee, J. Jellinek, E. V. Shevchenko, Nano Lett., 2012, 12,
5382‒5388.
28
29
30
K. Chernichenko, Á. Madarász, I. Pápai, M. Nieger, M. Leskelä and T. Repo, T. Nat.
Chem. 2013, 5, 718‒723.
C.ꢀH. Kuo, Y.ꢀT. Chu, Y.ꢀF. Song and M. H. Huang, Adv. Funct. Mater., 2011, 21,
792‒797.
H.ꢀL. Wu, R. Sato, A. Yamaguchi, M. Kimura, M. Haruta, H. Kurata and T. Teranishi,
Science 2016, 351, 1306‒1310.
31 S. Rej, C.ꢀF. Hsia, T.ꢀY. Chen, F.ꢀC. Lin, J.ꢀS. Huang and M. H. Huang, Angew. Chem.,
Int. Ed., 2016, 55, 7222‒7226.
32
33
34
M. Yurderi, A. Bulut, Đ. E. Ertas, M. Zahmakiran, M. Kaya, Appl. Catal. B: Environ.,
2015, 165, 169‒175.
W.ꢀC. Huang, L.ꢀM. Lyu, Y.ꢀC. Yang and M. H. Huang, J. Am. Chem. Soc. 2012, 134,
1261‒1267.
J.ꢀY. Ho and M. H. Huang, J. Phys. Chem. C, 2009, 113, 14159‒14164.
35 G. H. Chan, J. Zhao, E. M. Hicks, G. C. Schatz and R. P. Van Duyne, Nano Lett., 2007,
7, 1947‒1952.
36
G.ꢀZ. Yuan, C.ꢀF. Hsia, Z.ꢀW. Lin, C. Chiang, Y.ꢀW. Chiang and M. H. Huang,
Chem.‒Eur. J., 2016, 22, 12548‒12556.
37
38
39
C. W. Li, J. Ciston, M. W. Kanan, Nature, 2014, 508, 504‒507.
A. Staubitz, A. P. M. Robertson and I. Manners, Chem. Rev., 2010, 110, 4079‒4124.
I. Lee, F. Delbecq, R. Morales, M. A. Albiter and F. Zaera, Nat. Mater., 2009, 8,
132‒138.
17

Chemical Science
Page 18 of 23
View Article Online
DOI: 10.1039/C7SC05232D
40
D. Fuhrmann, D. Wacker, K. Weiß, K. Hermann, M. Witko and C. Wöll, J. Chem.
Phys., 1998, 108, 2651.
41
42
B. Bridier, N. Lopez, J. P. Ramirez, Dalton Trans., 2010, 39, 8412−8419.
R. A. Koeppel, J. T. Wehrlia, M. S. Wainwright, D. L. Trimm and N. W. Cant, Appl.
Catal., A, 1994, 120, 163−177.
18

Page 19 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
Fig. 1 Pseudomorphic formation of Cu polyhedra from Cu2O crystals. (a–c) SEM and (d–f)
HAABFꢀSTEM images with EDS elemental line scans for Cu2O RD, cubic, and octahedral
nanocrystals, respectively. The red and blue lines correspond to copper and oxygen,
respectively. (g–i) SEM images, (j–l) HAABFꢀSTEM images, and (m–o) EDS line scans
across the yellow line for Cu RD, cubic, and octahedral nanocrystals, respectively. The red
signals come from copper, while the blue signals are from oxygen. (p) XRD patterns of the
Cu2O and Cu crystals. (q–v) Photographs of (q, r) RD, (s, t) cubic, and (u, v) octahedral Cu2O
and Cu crystals. All scale bars are equal to 100 nm.
19

Chemical Science
Page 20 of 23
View Article Online
DOI: 10.1039/C7SC05232D
Fig. 2 TEM characterization of Cu crystals. (a–c) TEM images, (d–f) singleꢀparticle TEM
images with their orientations shown in white models, (g–i) corresponding SAED patterns,
(j–l) HRꢀTEM images, and (m–o) magnified HRꢀTEM images near edges of Cu RD, cubic,
and octahedral crystals, respectively. (p–s) Timeꢀdependent TEM images showing how a
Cu2O rhombic dodecahedron transforms into a Cu rhombic dodecahedron with intermediate
particles collected at specific time points indicated.
20

Page 21 of 23
Chemical Science
View Article Online
DOI: 10.1039/C7SC05232D
Fig. 3 Results of the catalytic semihydrogenation reaction. (a) Schematic diagram for the
stereoselective semihydrogenation of DPA by inꢀsitu formed Cu crystals. (b–d) Timeꢀ
dependent plots for the conversion of DPA for Cu (b) rhombic dodecahedra, (c) cubes, and (d)
octahedra. (e) Catalytic comparison of Cu rhombic dodecahedra with CuCl2, commercial
Cu2O, and Lindlar catalyst. (f, g) General applicability of Cu rhombic dodecahedra toward
different substituted alkynes.
20
15
10
5
0
0
5
10
15
20
25
30
35
Time (min)
Fig. 4 Representative plot for the volume of H2 generated from ammonia borane ethanolysis
as a function of reaction time in the presence of Cu2O rhombic dodecahedra. Reaction
21

Chemical Science
Page 22 of 23
View Article Online
DOI: 10.1039/C7SC05232D
conditions: catalyst = 4 mg, amount of ammonia borane used = 0.0123 g, temperature = 30 ºC.
Volume of hydrogen evolved is 20 mL in 31 min.
Table of Contents Graphic
Cu2O nanocrystals can be pseudomorphically converted into Cu crystals through ammonia
borane reduction releasing hydrogen for stereoselective semihydrogenation of
diphenylacetylene.
22

Page 23 of 23
Chemical Science
Reducing
agent
Time
(min)
Conversion of
DPA
TOF
(min-1)
Entry
Catalyst
Temperature
50 oC
Z-stilbene E-stilbene Bibenzyl
Cu2O RD crystals
Cu2O cubic
H3N-BH3
1
2
45
100%
99.5%
92.1%
0%
0.5%
5.2%
0.039
50 oC
H3N-BH3
H3N-BH3
60
95.6%
97%
2.7%
0.028
crystals
Cu2O octahedral
50 oC
3
60
95.1%
3.8%
1.1%
0.028
crystals
Cu2O RD crystals
25 oC
50 oC
50 oC
H3N-BH3
H3N-BH3
H3N-BH3
4
5
6
90
60
45
97.2%
81.5%
95.5%
98.2%
89.2%
90.9%
1.0%
8.4%
4.1%
0.8%
2.4%
5%
0.019
0.024
0.037
CuCl2
Commercial Cu2O
Lindlar catalyst +
Quinoline (100µL)
50 oC
H2
7
30
96%
52%
0%
48%
0.057
50 oC
50 oC
50 oC
H3N-BH3
N2H4
8
9
Lindlar catalyst
Cu2O RD crystals
Cu2O RD crystals
15
45
45
100%
7.8%
0%
0%
100%
8.9%
1.4%
0.119
0.003
0.014
50.2%
90.1%
40.9%
8.5%
NaBH4
10
36.5%