Facet-dependent catalytic activity of Cu2O nanocrystals in the one-pot synthesis of 1,2,3-triazoles by multicomponent click reactions

Article abstract of DOI:10.1002/chem.201302065

We report the highly facet-dependent catalytic activity of Cu₂O nanocubes, octahedra, and rhombic dodecahedra for the multicomponent direct synthesis of 1,2,3-triazoles from the reaction of alkynes, organic halides, and NaN₃. The catalytic activities of clean surfactant-removed Cu ₂O nanocrystals with the same total surface area were compared. Rhombic dodecahedral Cu₂O nanocrystals bounded by {110} facets were much more catalytically active than Cu₂O octahedra exposing {111} facets, whereas Cu₂O nanocubes displayed the slowest catalytic activity. The superior catalytic activity of Cu₂O rhombic dodecahedra is attributed to the fully exposed surface Cu atoms on the {110} facet. A large series of 1,4-disubstituted 1,2,3-triazoles have been synthesized in excellent yields with high regioselectivity under green conditions by using these rhombic dodecahedral Cu₂O catalysts, including the synthesis of rufinamide, an antiepileptic drug, demonstrating the potential of these nanocrystals as promising heterogeneous catalysts for other important coupling reactions.

Full text of DOI:10.1002/chem.201302065

DOI: 10.1002/chem.201302065
Facet-Dependent Catalytic Activity of Cu2O Nanocrystals in the One-Pot
Synthesis of 1,2,3-Triazoles by Multicomponent Click Reactions
Kaushik Chanda, Sourav Rej, and Michael H. Huang*[a]
Abstract: We report the highly facet-
dependent catalytic activity of Cu2O
nanocubes, octahedra, and rhombic do-
decahedra for the multicomponent
direct synthesis of 1,2,3-triazoles from
the reaction of alkynes, organic halides,
and NaN3. The catalytic activities of
clean surfactant-removed Cu2O nano-
crystals with the same total surface
area were compared. Rhombic dodeca-
hedral Cu2O nanocrystals bounded by
{110} facets were much more catalyti-
cally active than Cu2O octahedra ex-
posing {111} facets, whereas Cu2O
nanocubes displayed the slowest cata-
lytic activity. The superior catalytic ac-
tivity of Cu2O rhombic dodecahedra is
attributed to the fully exposed surface
Cu atoms on the {110} facet. A large
series of 1,4-disubstituted 1,2,3-tria-
zoles have been synthesized in excel-
lent yields with high regioselectivity
under green conditions by using these
rhombic dodecahedral Cu2O catalysts,
including the synthesis of rufinamide,
an antiepileptic drug, demonstrating
the potential of these nanocrystals as
promising heterogeneous catalysts for
other important coupling reactions.
Keywords: click chemistry · copper ·
multicomponent reactions · nano-
crystals · synthetic methods
Introduction
arylboronic acid and phenylacetylene.[15] Cu2O nanocubes,
octahedra, nanospheres, and etched nanocubes can catalyze
the N-arylation of imidazole with iodobenzene.[16] A reac-
tion time of 10–15 h at 1108C was found to be generally nec-
essary to reach a product yield of more than 90%. Since the
pioneering work by Sharpless[17] and co-workers in 2001, the
copper(I)- catalyzed 1,3-dipolar cycloaddition reaction[18] of
organic azides and terminal alkynes also known as “click”
chemistry has a powerful impact on drug discovery, organic
synthesis, and biological and electrochemical applications.[19]
The most practical aspect of click chemistry is the ability to
attach diverse structures with a wide range of functional
groups. Click reactions generating a diverse array of 1,2,3-
triazoles represent highly desirable reactions that may be
catalyzed by Cu2O nanocrystals, since the CuI state has been
known to be active for click reactions. Copper nanoparticles
have been reported to catalyze cycloaddition of azides with
The recent development in the syntheses of cuprous oxide
(Cu2O) and silver oxide (Ag2O) nanocrystals in aqueous sol-
utions with systematic shape evolution has yielded sharp-
faced oxide nanocubes, octahedra, and rhombic dodecahe-
dra bounded exclusively by the {100}, {111}, and {110} surfa-
ces, respectively.[1–6] Their facet-dependent photocatalytic,
electrical, and molecular adsorption properties can be inves-
tigated with greater certainty.[2,3,5–11] The relative stability of
these surface planes toward face-selective chemical etching
has been studied using particles exposing two or more of
these surface facets.[12,13] Comparative catalytic activity of
these nanostructures represents another important surface-
related property to study. Examinations of facet-dependent
cross-coupling reactions catalyzed by Cu2O and other oxide
nanocrystals exposing these low-index facets are highly in-
teresting and attractive, but such studies are not quite possi-
ble due to the availability of rhombic dodecahedra until re-
cently. Bulky Cu2O and nanoparticles have been reported to
terminal alkynes.[20] A click reaction performed on a Cu
ACHTUNGTRENNUNG(111)
surface under ultrahigh vacuum conditions was also demon-
strated.[21] Recently, irregularly-shaped and polyvinylpyrroli-
done (PVP)-coated Cu2O nanoparticles were used to cata-
lyze cycloaddition reactions between azides and alkynes in
water under aerobic conditions.[22] A reaction time of 24 h at
378C was needed to reach product yields of 68–92% for
a range of azides and alkynes examined. Clearly the use of
polyhedral Cu2O nanocrystals as efficient catalysts for click
reactions is essentially not studied. Facet-dependent catalyt-
ic activity of Cu2O nanocrystals for the triazole-forming cy-
cloaddition reactions remains unexplored.
À
À
be efficient catalysts for C C and C O bond-formation re-
actions.[14] Cu2O nanocubes, rhombic dodecahedra, and octa-
decahedra were used for the aerobic oxidative arylation of
[a] Dr. K. Chanda,+ S. Rej,+ Prof. Dr. M. H. Huang
Department of Chemistry and Frontier Research Center
on Fundamental and Applied Sciences of Matters
National Tsing Hua University, Hsinchu 30013 (Taiwan)
E-mail: hyhuang@mx.nthu.edu.tw
In this study, we have used sharp-faced Cu2O nanocubes,
octahedra, and rhombic dodecahedra synthesized in aqueous
solutions as catalysts for a range of Huisgen [3+2] cycload-
dition reactions, or 1,3-dipolar cycloaddition, to examine the
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201302065.
16036
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16036 – 16043

FULL PAPER
facet-dependent catalytic activity of these Cu2O nanocrystals
for the first time. Nanocrystals with approximately the same
total surface area were employed to precisely compare the
catalytic activity of different surface planes. Crystal models
of the different planes of Cu2O and zeta potential measure-
ments of the nanocrystals were used to explain the high cat-
alytic activity of rhombic dodecahedra. Rhombic dodecahe-
dra are highly efficient at coupling alkynes and organic hal-
ides in the presence of NaN3 to generate diversely 1,4-disub-
stituted 1,2,3-triazoles in excellent yields with high regio-
ACHTUNGTRENNUNGselectivity under green conditions. Remarkably, the
multicomponent coupling strategy can be easily extended to
the synthesis of rufinamide (A), an antiepileptic drug, dem-
onstrating the potential of Cu2O rhombic dodecahedra for
applications in the synthesis of pharmaceutically active com-
pounds.
Results and Discussion
The Cu2O nanocubes, octahedra, and rhombic dodecahedra
were synthesized in aqueous solutions following our previ-
ously reported procedures.[2,3] Precise amounts of CuCl2 so-
lution, SDS surfactant, NaOH solution, and NH2OH·HCl re-
ductant were mixed at room temperature and aged for 1–2 h
for particle growth. Figure 1 shows the SEM images of the
Cu2O nanocubes, octahedra, and rhombic dodecahedra syn-
thesized with sizes of a few hundreds of nanometers. Rhom-
bic dodecahedra are more uniformly-sized, whereas octahe-
dra and cubes have larger size distributions. All nanocrystals
possess sharp faces for their facet-dependent catalytic activi-
ty investigation. Figure S1 in the Supporting Information
shows the higher-magnification SEM images of the nano-
crystals and photographs of the particle solutions. Their size
distribution histograms and the average particle sizes are
given in Figure S2 and Table S1 (the Supporting Informa-
tion). The solutions change from orange for the rhombic do-
decahedra to dark-brown for the octahedra as the particle
size increases. X-ray diffraction (XRD) patterns of these
nanocrystals match to that of Cu2O (Figure S3, the Support-
ing Information). Rhombic dodecahedra showed enhanced
(110) and (220) peaks as a result of their {110} faces. Octah-
dra displayed an exceptionally strong (111) reflection peak
and very weak intensities for the other peaks possibly be-
cause most of the octahedra were oriented with their {111}
faces parallel to the substrate plane. The cubes were more
randomly oriented, so the relative peak intensities resemble
those in the standard XRD pattern of Cu2O.
Figure 1. SEM images of a) cubic, b) octahedral, and c) rhombic dodeca-
hedral Cu2O nanocrystals synthesized. Magnified SEM images of the dif-
ferent particle shapes are also provided. The scale bars are equal to
1 mm.
porting Information). Essentially, only two bands were ob-
served in their FTIR spectra. The sharp peak at 631 cmÀ1
[23]
À
corresponds to the Cu O lattice vibration.
The band at
around 3440 cmÀ1 is attributed to the OH stretching vibra-
tion, which comes from the surface-adsorbed H2O. The ab-
À
À
For catalytic activity comparison of different surface
facets, it is most desirable that capping surfactant is cleanly
removed from the surfaces of the nanocrystals. After synthe-
sis, the nanocrystals were washed with ethanol to remove
the sodium dodecyl sulfate (SDS) surfactant completely. To
confirm that the nanocrystal surfaces were clean and free
from SDS or any kind of adsorbed organic molecules, Fouri-
er transform infrared spectroscopy (FTIR) measurements
were carried out on these nanocrystals (Figure S4, the Sup-
sence of stretching frequency for C H bonds, which general-
ly appears in the range of 2850–2960 cmÀ1, proves that the
surfaces were free from surfactant. X-ray photoelectron
spectrum (XPS) of the Cu2O rhombic dodecahedra provides
further evidence for the absence of SDS from the particle
surfaces (Figure 2). First, oxidation state of Cu in the parti-
cles was analyzed. In the Cu2p region, peaks at 932.1 and
951.9 eV are identified as the Cu2p3/2 and Cu2p1/2 character-
istic peaks, respectively.[24] These peak positions are consis-
Chem. Eur. J. 2013, 19, 16036 – 16043
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16037

M. H. Huang et al.
Scheme 1. A typical click reaction involves first the preparation of an
alkyl or aromatic azide, followed by its reaction with an alkyne to obtain
the product. Depending on the reaction temperature, regioisomers can
form.
with
a
calculated total particle surface area of
ꢀ0.0028 m2gÀ1 were used for all the catalytic reactions in
this study. Please note that the conventional practice of
loading amount and mole percent of catalyst is not suitable
for this study, because the total surface area, rather than the
weight of the catalyst, is more important for facet-depen-
dent catalytic activity investigation.
For the [3+2] cycloaddition reactions by click chemistry,
azides are typically obtained from the reaction of organic
halides and NaN3. The synthesized azides subsequently react
with aliphatic or aromatic alkynes to give the final products
(see Scheme 1). Such an approach has been adopted for
most nanoparticle-catalyzed click reactions.[20a,b,22] In this
work, we found that a one-pot multicomponent click reac-
tion can be successfully carried out using the Cu2O nano-
crystal catalysts, thus a single reaction directly gives the cy-
cloaddition product. For the initial reaction, benzyl bromide
and NaN3 were subjected to a [3+2] cycloaddition reaction
with phenyl acetylene for the one-pot synthesis of 1-benzyl-
4-phenyl-1H-1,2,3-triazole in ethanol at 558C under nitrogen
atmosphere (see Table 1). Upon completion of the reaction,
Figure 2. a) XPS spectrum of rhombic dodecahedral Cu2O nanocrystals.
b, c) Magnified XPS spectra showing the exact positions of the Cu 2p and
O 1s peaks.
tent with the CuI oxidation state. The O1s peak appearing
at 530.0 eV is comparable with the literature value at
530.2 eV for Cu2O.[25] Sulfonic acid has been reported to
show O1s peak at 531.7 eV.[26] Thus, the XPS results indicate
that the oxygen peak comes from Cu2O, not from oxygen
atoms on SDS. The absence of surfactant on the surface of
the nanocrystals was also con-
1
the H NMR spectra of the as-synthesized product indicate
firmed by the absence of the
S2p3/2 and S2p2/1 peaks at 167.8
and 168.9 eV, respectively, from
the sulfate species of SDS.[24]
The surfactant-free nanocrystals
can be directly used for their
catalytic activity comparison
provided that the amounts of
nanocrystal samples used have
the same total surface area.[27]
Specific surface areas of the
nanocrystals were obtained
from the nitrogen adsorption/
desorption curves and BET sur-
face area calculations (Fig-
ure S5, the Supporting Informa-
tion). Cu2O nanocubes, octahe-
dra, and rhombic dodecahedra
weighing 22.4, 25.1, 23.0 mg, re-
Table 1. Comparison of the catalytic activity of different Cu2O nanocrystals for different 1,3-dipolar cycloaddi-
tion reactions.[a]
Alkynes
Organic
halides
Product
t
RD Yield
t
OC Yield
t
Cube Yield
[h] [%][b]
[h] [%][b]
[h] [%][b]
1
1
96
4.5 90
7
88
2
3
1.5 92
90
5
88
7
8
80
77
2
5.5 90
[a] Reagents and conditions: 1 (0.25 mmol), 2 (0.25 mmol), and NaN3 (0.38 mmol) in EtOH (3 mL) at 558C.
RD=rhombic dodecahedra; OC=octahedra. [b] Yield of the isolated product.
spectively, were used for the surface area measurements.
The surface areas of Cu2O nanocubes, octahedra, rhombic
dodecahedra, and commercially available Cu2O powder
from Aldrich were found to be 2.84, 0.56, 1.35, and
1.18 m2 gÀ1, respectively. From these numbers, 1 mg of nano-
cubes, 5 mg of octahedra, and 2 mg of rhombic dodecahedra
regioselective formation of pure 1,4-disubstituted triazoles
without any by-products. Cu2O nanocubes gave a yield of
88% after 7 h of reaction, whereas octahedra delivered
a 90% yield in 4.5 h. Therefore, octahedra are catalytically
more efficient than the nanocubes for this reaction. Re-
markably, rhombic dodecahedra achieved a 96% yield after
16038
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16036 – 16043

Facet-Dependent Catalytic Activity
FULL PAPER
just 1 h of reaction. Rhombic dodecahedra are clearly the
best and most efficient catalysts among these three nano-
crystal morphologies. By comparison, the commercially
available Cu2O powder with larger grain sizes showed 80%
yield in 5 h (see Table S2, the Supporting Information).
Turnover frequencies (TOF) have been determined for
these three particle shapes. TOF numbers for the Cu2O
nanocubes, octahedra, and rhombic dodecahedra are 616,
754, and 6652 hÀ1, respectively, again showing the rhombic
dodecahedra as highly efficient catalysts (see the Supporting
Information for the calculations). To show the rhombic do-
decahedra as consistently the best catalysts, two more reac-
tions using quite different alkynes and organic halides as re-
agents were performed (entries 2 and 3, Table 1). Again
rhombic dodecahedra achieved high yields in substantially
less time than the nanocubes and octahedra, and are the
best catalysts for a broad range of click reactions. Since the
reaction takes place on the particle surfaces, the catalytic ac-
tivity differences are ascribed to the different facets exposed
(i.e., {110}>{111}>{100}).
tion of CH3CN to the surface copper atoms.[28] The results
show that the click reactions can be carried out in green sol-
vents. Interestingly, around 20% of the rhombic dodecahe-
dra became slightly etched during reaction in water, so etha-
nol is a better solvent to use. Since the solution pH is nearly
neutral, dissolved oxygen may be a possible source of etch-
ant in water.
The recyclability of the Cu2O rhombic dodecahedra was
also examined. After finishing one run of the reaction, an-
other cycle of the reaction was carried out using the same
nanocrystals. It has been observed that the rhombic dodeca-
hedra are effective and recyclable catalysts for 1,3-dipolar
cycloaddition leading to the regioselective synthesis of 1,4-
disubstituted triazole (3a) with excellent product yields of
92–96% after 1 h of reaction in the second cycle. SEM
images of the Cu2O rhombic dodecahedra after completing
two cycles of 1,3-dipolar cycloaddition reaction show no no-
ticeable changes in morphology, supporting their use as re-
cyclable catalysts (Figure 3).
To confirm that the observed catalytic activity of Cu2O
nanocrystals is not due to dissolved CuI species in the solu-
tion, control experiments were performed. First, we took
3 mg of Cu2O rhombic dodecahedra in ethanol and stirred
the particles for 2.5 h under a nitrogen atmosphere at 558C
to simulate the actual conditions used for the click reactions.
After centrifugation at 5000 rpm for 3 min, the supernatant
liquid was carefully collected. Using the supernatant liquid,
click reaction was performed under the same reaction condi-
tion as stated in the experimental section. After 8 h, we did
not obtain any trace of click product from the reaction mix-
ture, indicating the absence of Cu+ ions in the supernatant
liquid. Furthermore, solutions of CuCl and CuCl2 in ethanol
were prepared and their UV/Vis spectra were taken. Con-
centrated HCl was added to the CuCl solution to improve
CuCl solubility. To the supernatant liquid, 0.1m NaCl and
20 mL of concentrated HCl were added to test the presence
of any copper species by forming CuCl or CuCl2. The UV/
Vis spectrum of the resulting supernatant liquid did not
show any peak in the 200–800 nm range, but several absorp-
tion bands were recorded for CuCl and CuCl2 in this spec-
tral range (Figure S6, the Supporting Information). The re-
sults further verify the absence of dissolved CuI species in
the solution, and the observed catalytic activity is attributed
to the catalytic action of the Cu2O nanocrystals.
Next, the effect of solvents on the coupling reaction was
investigated using rhombic dodecahedra as the catalysts
(Table S3, the Supporting Information). When the reactions
were conducted in EtOH, EtOH/H2O (1:1 volume ratio),
and H2O, the products were obtained in excellent yields
(92–96%). The use of CH3CN and THF as solvent did not
yield the desired product. SEM images of Cu2O rhombic do-
decahedra after 1.5 h of reaction in THF show the particles
are coated with a thick layer of substance, thus blocking the
catalytic sites (Figure S7, the Supporting Information).
When CH3CN was used as the solvent, the particles were
observed to be highly etched possible due to the coordina-
Figure 3. SEM images of the rhombic dodecahedral Cu2O nanocrystals
a) before and b) after two cycles of 1,3-dipolar cycloaddition reaction.
Scale bar is equal to 500 nm.
A simple analysis of the different crystal planes of Cu2O
can assist the explanation of the experimental observations.
Figure 4 presents crystal models of the (100), (110), and
(111) planes of Cu2O. The (100) planes comprise the surface
planes of a body-centered cubic unit cell of Cu2O with
oxygen atoms forming the crystal lattice and copper atoms
occupying half of the tetrahedral sites.[9] However, the (100)
plane can also be presented to expose terminal Cu atoms.[29]
For consistency with experimental observations of the low
reactivity of nanocubes, the surface Cu atoms are considered
to lie just below the uppermost layer of oxygen atoms. The
(111) plane contains terminal copper and oxygen atoms.
However, many of the surface Cu atoms reside below the
plane of surface oxygen atoms (see Figure 4c).[30] The (110)
plane is terminated with copper and oxygen atoms lying es-
sentially on the same plane, and so all the surface Cu atoms
are fully exposed (see Figure 4 f). An area density analysis
of surface Cu atoms reveals that the (110) plane actually has
the lowest surface Cu atom density (10.98, 14.27, and 7.76
Cu atomsnmÀ2 for the (100), (111), and (110) planes of
Chem. Eur. J. 2013, 19, 16036 – 16043
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16039

M. H. Huang et al.
tion were carried out (Figure S8, the Supporting Informa-
tion). The measured z potentials for Cu2O nanocubes, octa-
hedra, and rhombic dodecahedra are +0.37, +6.9, and
+14.5 mV respectively, which agree well with the crystal
plane analysis. The surface charge of Cu2O nanocubes is
close to neutral, whereas the positive surface charge of
rhombic dodecahedra is 2.1 times higher than octahedra.
The higher adsorption capability of anionic deprotonated al-
kynes on the positively charged and more accessible Cu
atoms of Cu2O should favor the formation of the Cu-acety-
lide active species and increase the reaction rate. Thus,
Cu2O rhombic dodecahedra with the fully exposed terminal
Cu atoms show the best catalytic activity.
Cu2O rhombic dodecahedra have been subsequently em-
ployed as catalysts to a wide variety of terminal alkynes and
organic halides using the optimized reaction conditions and
the results are summarized in Table 2. The corresponding
1,4-disubstituted triazoles were obtained as highly regiose-
lective products with excellent yields after a simple work-up
involving centrifugation to remove the nanocrystals, washing
and solvent evaporation. Finally the crude products were
purified by column chromatography followed by spectro-
1
scopic characterization using H/13C NMR spectroscopy, and
high-resolution mass spectroscopy (HR-MS). Substituent ef-
fects of both azide precursors and alkynes have been ob-
served. The use of benzyl bromides or substituted benzyl
bromides containing electron donating groups reacted nicely
to obtain the corresponding triazoles in excellent yields (en-
tries 1–11, Table 2). Interestingly, deactivated halides such as
n-butyl bromides and n-hexyl bromides reacted with termi-
nal alkynes in 1.5–2 h to obtain the 1,4-disubstituted tria-
zoles in excellent yields (entries 12–15, Table 2). Apart from
organic halides, it has been observed that aromatic alkynes
containing electron-withdrawing groups such as NO2, F, or
sterically hindered substituents have no effect on the out-
come of the reaction (entries 5, 8, and 15, Table 2). Hy-
droxy-substituted alkynes such as propargyl alcohol reacted
nicely with both activated and unactivated organic halides
to obtain the corresponding triazoles as single regioisomers
in good to high yields (entries 2, 10, and 14, Table 2). Fur-
thermore, alkynes bearing electron-withdrawing substituents
such as ethyl propiolate and propiolic acid underwent tria-
zole-forming reactions with high yields (entries 4 and 7,
Table 2) compared with the slugging nature of the aliphatic
alkynes (entry 3, Table 2). In addition, disubstituted acety-
lenes such as dimethylacetylene dicarboxylate (DMAD) re-
acted with benzyl bromides and NaN3 under these optimized
reaction conditions to obtain the 1,4-disubstituted triazoles
in excellent yields (entry 11, Table 2).
Figure 4. Crystal structure models of cuprite Cu2O showing the a,
b) {100}, c, d) {111}, and e, f) {110} surfaces. Oxygen atoms are shown in
a dark color, whereas copper atoms are shown in white. Two different
viewing angles have been presented for each surface to more clearly
show the number of surface atoms. For panel (b), the uppermost layer of
oxygen atoms have been removed to clearly show the number of surface
copper atoms. Panel (d) shows the {111} planes rotated 638 with respect
to the model shown in panel (c). The triangle in panel (d) encloses the
same area as that shown in panel (c). The cubic unit cell parameter is a.
Cu2O, respectively). However, all of the surface Cu atoms
on the (110) planes are fully exposed for interaction with li-
gands, whereas many of the surface Cu atoms are partially
blocked for the (111) planes and only partially exposed Cu
atoms are available for the (100) plane to hinder the ligand
interaction. These differences explain the observed relative
catalytic activity of these surfaces. It is known that the first
step of the [3+2] cycloaddition reaction involves the forma-
tion of Cu-acetylide through the cationic displacement of
terminal protons of alkynes by the positively charged Cu
atoms (Scheme S1, the Supporting Information).[31] The for-
mation of Cu-acetylide happens on the Cu2O nanocrystal
surface. Mixing terminal alkynes with Cu2O rhombic dodec-
ahedra immediately form Cu-acetylide because of the fully
exposed Cu atoms on the (110) planes, which reacts with in
situ-generated azides to form 1,4-disubstituted triazoles in
a short time. To further substantiate the greater catalytic ac-
tivity of the {110} faces, z potential measurements on the
freshly prepared Cu2O nanocrystals in pure methanol solu-
After the successful regioselective syntheses of a wide va-
riety of 1,4-disubstituted triazoles, we considered the synthe-
sis of rufinamide (A), an antiepileptic drug, through this
multicomponent coupling reaction. This drug is usually pre-
pared by multistep procedures, and its synthesis requires the
use of a high temperature (1358C) and long reaction times
(more than 1 day).[32] The synthetic conditions are time- and
energy-consuming. We recognized that rufinamide (A)
16040
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16036 – 16043

Facet-Dependent Catalytic Activity
FULL PAPER
Table 2. A list of three-component 1,3-dipolar cycloaddition reactions catalyzed by Cu2O rhombic dodecahe-
dra using organic halides, alkynes, and NaN3.[a]
Conclusion
We have employed surfactant-
free Cu2O nanocubes, octahe-
dra, and rhombic dodecahedra
as catalysts for the multicompo-
nent synthesis of 1,2,3-triazoles
in ethanol by 1,3-dipolar cyclo-
addition reactions. All these
nanocrystals can serve as effi-
cient catalysts for the one-pot
click reactions. However, the
Cu2O rhombic dodecahedra
bounded by the {110} facets are
the best catalysts with the high-
est yields and shortest reaction
times for a broad range of re-
agents used. They have also
been demonstrated to be
highly efficient for the one-pot
synthesis of rufinamide. The
Cu2O rhombic dodecahedra are
promising efficient and green
heterogeneous catalysts for
other types of coupling reac-
tions.
Alkynes
Organic
halides
Product
Mass
t
Yield
[%][b]
[h]
1
2
3
235
189
201
1
96
93
92
1
1.5
4
5
231
307
1
89
91
2.5
6
7
199
203
1.5
1
93
98
8
280
1
90
9
231
203
1
1
92
94
Experimental Section
10
Chemicals: Anhydrous copper(II)
chloride (CuCl2; 97%) and hydroxyla-
mine hydrochloride (NH2OH·HCl;
99%) were purchased from Aldrich.
Sodium hydroxide (98.2%) and
sodium dodecyl sulfate (SDS; 100%)
were acquired from Mallinckrodt. All
chemicals were used as received with-
out further purification along with de-
ionized water for all solution prepara-
tions. Commercially available reagents
were used for the 1,3-dipolar cycload-
dition reactions.
11
12
275
201
1
97
91
1.5
13
14
229
155
1.5
2
94
90
Synthesis of Cu2O nanocubes and
rhombic dodecahedra: The synthetic
method described here is adopted
from our previously reported proce-
dure.[2] For the synthesis of Cu2O
nanocrystals with cubic and rhombic
dodecahedral shapes, 8.92 and 6.95 mL
of deionized water were added in turn
15
273
2
92
[a] Reagents and conditions: 1 (0.25 mmol), 2 (0.25 mmol), and NaN3 (0.38 mmol) in EtOH (3 mL) at 558C.
[b] Yield of the isolated product.
could be synthesized through a multicomponent coupling re-
action by employing 2,6-difluorobenzyl bromide (2e), NaN3,
and propiolamide (1k) (Scheme 2). Cu2O rhombic dodeca-
hedra were selected as the catalysts. Gratifyingly, this multi-
component cycloaddition reaction produced rufinamide (A)
in 2.5 h at 558C with 95% yield. The huge advantage of
using these nanocrystal catalysts for this important reaction
is demonstrated.
Scheme 2. One-pot synthesis of rufinamide (A). RD=rhombic dodecahe-
dra.
Chem. Eur. J. 2013, 19, 16036 – 16043
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16041

M. H. Huang et al.
to sample vials. The volume of water added to each vials was adjusted in
such a manner that after the addition of NH2OH·HCl, the total volume
of the final solution is 10 mL. The sample vials were placed in a water
bath set at 30–328C. Then, a solution of CuCl2 (0.1m, 0.5 mL) and SDS
powder (0.087 g) were added to the vials with vigorous stirring. When
the solution became clear, a solution of NaOH (1.0m, 0.18 mL) was
added and shaken for ꢀ10 s. The solution turned light-blue immediately,
due to the formation of threadlike Cu(OH)2 precipitate. Finally, 0.40 and
2.37 mL of 0.1m NH2OH·HCl were quickly injected in 5 s for the synthe-
sis of nanocubes and rhombic dodecahedra, respectively. After stirring
for 20 s, the solutions were kept in the water bath for 1 h for nanocrystal
growth. The concentrations of Cu2+ ions and SDS surfactant in the final
solution are 1.0ꢁ10À3 and 3.0ꢁ10À2 m, respectively. The reaction mixtures
were centrifuged at 5000 rpm for 3 min. After decanting the top solution,
the precipitate was washed with 6 mL of 1:1 volume ratio of water and
ethanol for three times to remove unreacted chemicals and SDS surfac-
tant. The final washing step used 5 mL of ethanol, and the precipitate
was dispersed in 0.6 mL of ethanol for storage and analysis.
Instrumentation: SEM images of the samples were obtained using
a JEOL JSM-7000F electron microscope. XRD patterns were recorded
on a Shimadzu XRD-6000 diffractometer with CuKa radiation. One drop
of the Cu2O nanocrystal solution was added to an optical microscope
cover slide and dried naturally. The slide was mounted onto a sample
holder for the XRD measurements. FT-IR spectra were taken on
a Perkin–Elmer 55855 spectrometer. XPS characterization was carried
out on a ULVAC-PHI Quantera SXM high-resolution XPS spectrometer.
Data were recorded using a monochromatized Al anode as the excitation
source. The C1s peak was chosen as the reference line. Surface areas of
the three samples of Cu2O nanocrystals were determined through nitro-
gen adsorption–desorption isotherms at 77 K using a Quantachrome
Nova 2000e analyzer. z potential measurements were carried out on
a
Malvern particle size analyzer MS2000. 1H NMR (600 MHz) and
13C NMR (150 MHz) spectra were recorded with a Bruker DRX600 spec-
trometer. Chemical shifts are reported in ppm relative to the internal sol-
vent peak (d=7.24 and 77.0 ppm, respectively, for CDCl3). Coupling con-
stants, J, are given in Hz. Multiplicities of peaks are given as d (doublet),
m (multiplet), s (singlet), and t (triplet). TLC plates were Merck silica
gel 60 F254 on aluminum. Flash column chromatography was performed
with silica gel (60–100 mesh). All reagents required for the 1,3-dipolar cy-
cloaddition reactions are available commercially.
Synthesis of Cu2O octahedra: The synthetic procedure used for making
octahedral Cu2O nanocrystals is based on our reported procedure with
a slight modification in the volume of NH2OH·HCl solution added.[3]
First, deionized water (9.02 mL) was added to a sample vial. The sample
vial was placed in a water bath set at 30–328C. Next, CuCl2 (0.1 mL of
0.1m) and a solution of NaOH (0.2 mL, 1.0m) were added and the vial
was shaken for ꢀ10 s. Then, SDS powder (0.087 g) was introduced with
vigorous stirring. Finally, NH2OH·HCl (0.68 mL, 0.2m) was quickly in-
jected. After stirring for 20 s, the solution was kept in the water bath for
2 h for nanocrystal growth. The concentrations of Cu2+ ions and SDS sur-
factant in the final solution are 1.0ꢁ10À3 and 3.0ꢁ10À2 m. The reaction
mixture was centrifuged at 3500 rpm for 2 min. After decanting the top
solution, the precipitate was washed with 6 mL of 1:1 volume ratio of
water and ethanol for three times to remove unreacted chemicals and
SDS surfactant. The final washing step used 5 mL of ethanol, and the
precipitate was dispersed in ethanol (0.6 mL) for storage and analysis.
Acknowledgements
We thank the National Science Council of Taiwan for the financial sup-
port of this work (NSC 98–2113-M-007–005-MY3 and NSC 101–2811-M-
007–033). We also thank Dr. Lain-Ming Lyu for assistance in the presen-
tation of the crystal models.
[1] C. H. Kuo, M. H. Huang, Nano Today 2010, 5, 106–116.
[2] W.-C. Huang, L.-M. Lyu, Y.-C. Yang, M. H. Huang, J. Am. Chem.
Soc. 2012, 134, 1261–1267.
[3] J.-Y. Ho, M. H. Huang, J. Phys. Chem. C 2009, 113, 14159–14164.
[4] X. Wang, H. F. Wu, Q. Kuang, R. B. Huang, Z. X. Xie, L. S. Zheng,
Langmuir 2010, 26, 2774–2778.
[5] L.-M. Lyu, W.-C. Wang, M. H. Huang, Chem. Eur. J. 2010, 16,
14167–14174.
[6] P. Yang, Nature 2012, 482, 41–42.
[7] M. H. Huang, P.-H. Lin, Adv. Funct. Mater. 2012, 22, 14–24.
[8] C.-H. Kuo, M. H. Huang, J. Phys. Chem. C 2008, 112, 18355–18360.
[9] Y. Zhang, B. Deng, T. Zhang, D. Gao, A. W. Xu, J. Phys. Chem. C
2010, 114, 5073–5079.
[10] C.-H. Kuo, Y.-C. Yang, S. Gwo, M. H. Huang, J. Am. Chem. Soc.
2011, 133, 1052–1057.
[11] W.-C. Wang, L.-M. Lyu, M. H. Huang, Chem. Mater. 2011, 23, 2677–
2684.
[12] L.-M. Lyu, M. H. Huang, J. Phys. Chem. C 2011, 115, 17768–17773.
[13] C.-H. Kuo, M. H. Huang, J. Am. Chem. Soc. 2008, 130, 12815–
12820.
[14] a) B. X. Tang, F. Wang, J. H. Li, Y. X. Xie, M. B. Zhang, J. Org.
Chem. 2007, 72, 6294–6297; b) H. Cao, H. Jiang, G. Yuan, Z. Chen,
C. Qi, H. Huang, Chem. Eur. J. 2010, 16, 10553–10559.
[15] L. Li, C. Nan, Q. Peng, Y. Li, Chem. Eur. J. 2012, 18, 10491–10496.
[16] Y. Xu, H. Wang, Y. Yu, L. Tian, W. Zhao, B. Zhang, J. Phys. Chem.
C 2011, 115, 15288–15296.
[17] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056–2075; Angew. Chem. Int. Ed. 2001, 40, 2004–2021.
[18] a) R. Huisgen, Pure Appl. Chem. 1989, 61, 613–628; b) C. W.
Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67, 3057–
3064; c) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708–2711; Angew. Chem. Int. Ed. 2002,
41, 2596–2599.
Cu2O nanocrystal-catalyzed click reactions: Under an atmosphere of N2
gas, a 25 mL round-bottomed flask containing a stirrer bar was charged
with benzyl bromide (0.0428 g, 0.25 mmol, 1.0 equiv) and NaN3 (0.026 g,
0.40 mmol, 1.6 equiv) in ethanol (5 mL). After the mixture was stirred
for 10 min at room temperature, phenyl acetylene (0.0255 g, 0.25 mmol,
1.0 equiv) was added into the solution. A fixed amount of Cu2O nano-
crystals was immediately added to the reaction mixture (1 mg of nano-
cubes, 5 mg of octahedra, or 2 mg of rhombic dodecahedra with a calcu-
lated total particle surface area of 0.0028 m2 gÀ1). Subsequently, the reac-
tion mixture was heated to 558C with stirring. The progress of the reac-
tion was monitored by thin-layer chromatography (TLC). After comple-
tion, the reaction mixture was centrifuged at 5000 rpm for 3 min to
remove the nanocrystals. The solvent was then removed under reduced
pressure. Water (5 mL) was added to the resulting mixture, followed by
extraction with ethyl acetate (10 mL twice). The combined organic layer
was dried over anhydrous MgSO4. The combined filtrate was subjected to
evaporation to obtain the crude compound, which was purified over
silica gel column (60–120 mesh) using 1% ethyl acetate in hexane as
eluent to obtain the corresponding triazole as the product. To evaluate
the extent of versatility of these nanocrystal catalysts, the same cycload-
dition reaction was repeated with different substituents on organic hal-
ides and alkynes containing electron-donating and electron-withdrawing
groups. Cu2O rhombic dodecahedra were chosen as the catalysts in these
reactions.
Recyclability of the Cu2O rhombic dodecahedra for the click reactions:
The recyclability of the Cu2O rhombic dodecahedra as catalysts was also
surveyed. After completing one run of the reaction, the reaction mixture
was centrifuged. The solution containing the product was removed, leav-
ing behind the nanocrystal precipitate. To the freshly taken ethanol solu-
tion, freshly measured quantities of benzyl bromide and NaN3 were
added. After stirring for 10 min, phenyl acetylene and the used nanocys-
tals were added into the solution. Subsequently, the reaction mixtures
were heated to 558C with stirring. The progress of the reaction was also
monitored by TLC.
[19] a) K. B. Sharpless, H. C. Kolb, Drug Discovery Today 2003, 8, 1128–
1137; b) M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108, 2952–
16042
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16036 – 16043

Facet-Dependent Catalytic Activity
FULL PAPER
3015; c) C. O. Kappe, E. V. Eycken, Chem. Soc. Rev. 2010, 39, 1280–
1290; d) R. A. Decrꢂau, J. P. Collman, A. Hosseini, Chem. Soc. Rev.
2010, 39, 1291–1301.
[26] S. B. Idage, S. Badrinarayanan, S. P. Vernekar, S. Sivaram, Langmuir
1996, 12, 1018–1022.
[27] C.-Y. Chiu, P.-J. Chung, K.-U. Lao, C.-W. Liao, M. H. Huang, J.
Phys. Chem. C 2012, 116, 23757–23763.
[28] J. R. Black, W. Levason, M. Webster, Acta Crystallogr. C 1995, 51,
623–625.
[29] K. X. Yao, X. M. Yin, T. H. Wang, H. C. Zeng, J. Am. Chem. Soc.
2010, 132, 6131–6144.
[30] a) Q. Hua, D. Shang, W. Zhang, K. Chen, S. Chang, Y. Ma, Z. Jiang,
J. Yang, W. Huang, Langmuir 2011, 27, 665–671; b) M. Leng, C. Yu,
C. Wang, CrystEngComm 2012, 14, 8454–8461.
[31] a) C. Shao, X. Wang, J. Xu, J. Zhao, Q. Zhang, Y. Hu, J. Org. Chem.
2010, 75, 7002–7005; b) F. Himo, T. Lovell, R. Hilgraf, V. V. Ros-
tovtsev, L. Noodleman, K. B. Sharpless, V. V. Fokin, J. Am. Chem.
Soc. 2005, 127, 210–216.
[20] a) L. D. Pachꢃn, J. H. van Maarseveen, G. Rothenberg, Adv. Synth.
Catal. 2005, 347, 811–815; b) B. H. Lipshutz, B. R. Taft, Angew.
Chem. 2006, 118, 8415–8418; Angew. Chem. Int. Ed. 2006, 45, 8235–
8238; c) F. Alonso, Y. Moglie, G. Radivoy, M. Yus, Adv. Synth.
Catal. 2010, 352, 3208–3214; d) T. Jin, M. Yan, Y. Yamamoto,
ChemCatChem 2012, 4, 1217–1229.
[21] F. Bebensee, C. Bombis, S.-R. Vadapoo, J. R. Cramer, F. Besenbach-
er, K. V. Gothelf, T. R. Linderoth, J. Am. Chem. Soc. 2013, 135,
2136–2139.
[22] Z. Zhang, C. Dong, C. Yang, D. Hu, J. Long, L. Wang, H. Li, Y.
Chen, D. Kong, Adv. Synth. Catal. 2010, 352, 1600–1604.
ˇ
ˇ
ˇ ´
[23] Z. C. Orel, A. Anzlovar, G. Drazic, M. Zigon, Cryst. Growth Des.
2007, 7, 453–458.
[32] W. H. Mudd, E. P. Stevens, Tetrahedron Lett. 2010, 51, 3229–3231.
[24] C.-H. Kuo, Y.-T. Chu, Y.-F. Song, M. H. Huang, Adv. Funct. Mater.
2011, 21, 792–797.
[25] M. C. Biesingera, L. W. M. Laua, A. R. Gersonb, R. S. C. Smart,
Appl. Surf. Sci. 2010, 257, 887–898.
Received: May 30, 2013
Published online: October 14, 2013
Chem. Eur. J. 2013, 19, 16036 – 16043
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16043