Facile Synthesis of Surface-Clean Monodispersed CuOx Nanoparticles and Their Catalytic Properties for Oxidative Coupling of Alkynes

Article abstract of DOI:10.1021/acscatal.5b02935

We show a facile method to prepare surface-clean monodispersed small and stable CuOx nanoparticles with controllable average sizes from below 1 nm up to ～5 nm without using bulk capping agent. Structural and surface characterizations show that the chemica

Full text of DOI:10.1021/acscatal.5b02935

Subscriber access provided by UNIV OF YORK
Article
Facile Synthesis of Surface-Clean Monodispersed CuOx Nanoparticles
and Their Catalytic Properties for Oxidative Coupling of Alkynes
Lichen Liu, Toshiyuki Matsushita, Patricia Concepción, Avelino Corma, and Antonio Leyva-Perez
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b02935 • Publication Date (Web): 05 Feb 2016
Downloaded from http://pubs.acs.org on February 17, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 29
ACS Catalysis
1
2
3
4
5
6
Facile Synthesis of SurfaceꢀClean Monodispersed CuOx Nanoparticles and Their
Catalytic Properties for Oxidative Coupling of Alkynes
7
8
9
Lichen Liua, Toshiyuki Matsushitab, Patricia Concepciona, LeyvaꢀPerez, Antonio and Avelino
Cormaa*
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
aInstituto de Tecnología Química, Universidad Politécnica de Valencia-Consejo Superior de Investigaciones
Científicas (UPV-CSIC), Av. de los Naranjos s/n, 46022 Valencia, Spain
bUbe Industries, Ltd., 1978-5, Kogushi, Ube, Yamaguchi 755-8633, Japan
*Email: acorma@itq.upv.es
Abstract
We show a facile method to prepare surfaceꢀclean monodispersed small and stable CuOx
nanoparticles with controllable average sizes from below 1 nm up to ~5 nm without using bulk
capping agent. Structural and surface characterizations show that the chemical states of CuOx
nanoparticles and their interactions with O2 are dependent on the particle size. To show their
relevance to catalysis, the wellꢀdefined monodispersed CuOx nanoparticles have been used for
oxidative coupling of alkynes. While the generally used CuCl catalysts presents a reaction induction
period and agglomerate into CuOx nanoparticles during the reaction, the induction period disappears
when monodispersed CuOx nanoparticles (~2 nm) were used as catalyst. Supported CuOx
nanoparticles on TiO2 behave in the same way as monodispersed CuOx nanoparticles. Kinetic,
spectroscopic and isotopic studies show that O2 activation is the rateꢀcontrolling step and the nature
of the oxygen species formed on supported CuOx nanoparticles are dependent on the size of CuOx
and determine the catalytic properties for oxidative coupling of alkynes.
Keywords: CuOx nanoparticles, size effect, oxidative coupling, terminal alkyne, O2 activation,
unsymmetric 1,3ꢀdiynes
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 29
1
2
3
4
5
1. Introduction
CuOx (Cu2O and CuO) are important semiconductors that present wide applications in catalysis,
gas sensing and photocatalysis. There is much work on controllable synthesis of Cu2O and CuO
nanostructures and their applications in energy conversion, gas sensing and catalysis.1,2 However,
most CuOx nanomaterials prepared in the literatures are larger than 5 nm with low surface areas.
Since the catalytic properties of nanoparticles (NPs) are obviously related with their size, developing
methodologies for synthesis of small CuOx NPs with diameter below 3 nm can be of much
importance for catalytic applications.
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
In 2005, Yin et al. presented a method for the synthesis of uniform monodispersed Cu2O
nanocrystals with sizes between 2 and 10 nm,3 in where oleic acid was used as capping agent and
showed a negative effects on catalytic properties of Cu2O NPs. Meanwhile, highꢀtemperature
o
calcination (> 200 C) is usually required to remove the ligands with long carbon chains.4 Others
have also prepared CuOx NPs but strongly interacting capping agents were also needed.5 So far, the
preparation of small “surface clean” CuOx NPs without bulky capping agent still remains a
challenge.
Copper compounds (like CuCl, CuCl2, Cu(acac)2 et al.) are widely used in organic synthesis as
catalysts for oxidation reactions.6,7 There are many mechanistic studies on the role of Cu for
oxidation reactions, and most of those works have been performed on the basis that mononuclear Cu
or, in some cases, double Cu centers stabilized by ligands are thought to be active centers.8,9
Although there are some reports on the application of Cu NPs for aerobic oxidation reaction, the
relationships between Cu compounds and Cu NPs are still not clear. In a recent work, we have seen
that the Cu salts used initially as the catalyst for CꢀN coupling reactions are not necessarily the active
species, but they agglomerate during the reaction to form Cu clusters with a few atoms.10 This can be
particularly relevant for oxidation reactions in where O2 activation may require Cu species with a
certain number of Cu atoms. In the case of organic reactions that take place at low temperature, O2
dissociation or, in a more general way, O2 activation can become the rateꢀcontrolling step. To achieve
the activation of the organic compounds and O2 simultaneously on isolated Cu cations seems like a
difficult task, and in natural Cuꢀcontaining enzymes, binuclear Cu centers are thought to be the active
sites for activation of O2.11,12 In the case of Cu NPs, the activation of O2 could be easier on small Cu
2
ACS Paragon Plus Environment

Page 3 of 29
ACS Catalysis
1
2
3
4
5
6
NPs and the size as well as redox properties of Cu NPs, will be a key issue for establishing
structureꢀactivity relationships for Cuꢀcatalyzed oxidation reactions.13
7
8
9
In the present work, we will show a facile method to prepare monodispersed small CuOx NPs with
controllable average sizes from below 1 nm up to ~5 nm. By kinetic experiments, we will show how
the size of the CuOx NPs controls the adsorption and activation of organic reactants and O2, as well
as their facility to go into redox cycles. To show their relevance to catalysis, we have used these
wellꢀdefined monodispersed CuOx NPs for oxidative coupling of alkynes. Then, by comparing with
CuCl (the widely used catalyst for oxidative coupling of alkynes), it has been proved that a reaction
induction period exists when CuCl is used as the starting catalyst. We have found that CuCl will
evolve and agglomerate during the reaction. On the other hand, the reaction induction period
disappears when monodispersed CuOx NPs (~2 nm) were used as catalyst. After identifying the
important role of CuOx NPs in oxidative coupling of alkynes, supported CuOx NPs on TiO2 were
prepared as heterogeneous catalysts. The CuOx/TiO2 catalysts show the same particle size effects,
that with wellꢀdefined isolated nanoparticles. By means of kinetic, spectroscopic and isotopic studies,
it has been found that O2 activation is the rateꢀcontrolling step and the nature of the oxygen species
formed on supported CuOx NPs are dependent on the size of CuOx NPs.
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. Experiments
2.1 Synthesis of monodispersed CuOx NPs and supported CuOx NPs.
The CuOx NPs with different sizes were prepared as following. CuCl powder was added into 20
o
mL dimethylsulfoxide (DMSO). The suspension was heated to 90 C under stirring. The CuCl
powder became dissolved into DMSO when the suspension was kept at 90 oC for 20 minutes, leading
to formation of a homogeneous solution and the size of CuOx NPs was modulated through the
amount of CuCl in the solution. Samples will be labeled as CuOxꢀxx with xx corresponding to the
average particle size determined from TEM images.
Supported CuOx NPs were prepared by a wet impregnation method. 2 g TiO2 nanopowder (from
Aldrich, 21 nm average particle size), 30 mL H2O and the desired Cu(CH3COO)2 were mixed at
room temperature and kept under stirring at room temperature for 1 h. Then water is evaporated by
o
heating in an oil bath at 120 C. Finally, CuOx/TiO2 samples were obtained after calcination in N2 at
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 29
1
2
450 oC for 2 h.
3
4
5
6
7
8
Metallic Cu clusters with 8ꢀ16 atoms were prepared by an electrochemical method according to
reference.14
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
2.2 Characterizations
The sizes of copper particles on supports were analyzed by TEM microscopy. Samples for electron
microscopy studies were prepared by dropping the suspension of CuOx NPs directly onto
holeyꢀcarbon coated Nickel grids. Studies have been performed in a JEOL 2100F microscope
operating at 200 kV both in transmission (TEM) and scanningꢀtransmission modes (STEM). STEM
images were obtained using a High Angle Annular Dark Field detector (HAADF), which allows
Zꢀcontrast imaging.
Xꢀray photoelectron spectra of the catalysts were carried out with a SPECS spectrometer equipped
with a Phoibos 150MCD multichannel analyzer using MgKα (1253.6 eV) irradiation. The spectra
were recorded at ꢀ175ºC and with an Xꢀray power of 50 mW in order to avoid photoꢀreduction of the
copper species. The residual pressure in the analytical chamber was maintained below 10ꢀ9 mbar
during data acquisition. For spectra acquisition, a drop of the monodispersed CuOx NP solution was
deposited onto a Molibdenum sample holder and dried under N2 atmosphere inside the load lock of
the XPS spectrometer. On the other hand, supported CuOx NP were prepared for analysis by
depositing the powder onto a molybdenum sample holder. The binding energies of Cu 2p were
corrected for surface charging by referencing them to the energy of C1s peak of adventitious carbon,
set at 284.5 eV. Peak intensities have been calculated after nonlinear Shirleyꢀtype background
subtraction and corrected by the transmission function of the spectrometer. The CasaXPS software
has been used for spectra deconvolution.
Raman spectra were obtained with an “in via Reflex” Renishaw spectrometer, equipped with an
Olympus microscope. The exciting wavelength was 785 nm from a Renishaw HPNIR laser with a
power of ca. 15 mW on the sample. A micro quartz reactor located in a furnace has been used for the
in situ Raman studies. The furnance is provided with a small hole in order to allow focalization on
the sample with the laser spot. The reactor was positioned below the microscope objective, using a
x50 long working distance objective (WD=10.6 nm). For the in situ studies the sample (10mgr) has
4
ACS Paragon Plus Environment

Page 5 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
been in situ activated in N2 flow for 1h at 25 ºC. Then the sample has been exposed to a
phenylacetylene + O2 flow at 90 ºC for 1h. Afterwards the temperature was decrease to 25 ºC, and
the gas flow switch to only O2. Temperature was increased from 25º to 60º and 90 ºC, collecting
several spectra at each temperature.
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
FTIR spectra have been collected with a Nexus spectrometer from Thermo equipped with a DTGS
detector (4 cm–1 resolution, 32 scans).The samples were pressed into selfꢀsupported pellets (ca. 10mg
cmꢀ2) and activated in vacuum (10ꢀ4 mbar) for 1h at 25 ºC prior to the IR studies. In the
phenylacetylene adsorption studies, deuterated phenylacetylene (PhꢀC≡CꢀD) was adsorbed at 25ºC
at increasing dosing from 0.5 to 6.5 mbar, followed by further increasing the temperature to 60 and
90 ºC. Spectra were acquired at each temperature and after each dose. After 90 ºC, the sample was
evacuated and the temperature lowered to 25 ºC where 35 mbar CO was adsorbed for surface
titration experiments. After CO evacuation the sample was exposed to 30 mbar O2 at 90 ºC for 45min,
followed by evacuation and further decreasing temperature to 25 ºC were 35 mbar CO was again
reꢀadsorbed for surface titration. In the in situ IR studies 6.5 mbar phenylacetylene was coadsorbed
with 30 mbar O2 at 25ºC and kept for 45 min at that temperature. Afterwards the temperature was
raised to 90 ºC for 45 min followed by sample evacuation at 10ꢀ4 mbar.
18O Isotopic exchange of surface oxygen species followed by H2 titration, were carried out in a
quartz flow reactor coupled to a quadrupole mass spectrometer (Omnistar, Balzers) for onꢀline
monitoring of the exit gas composition. In each experiment, 100 mg of sample diluted in SiC (1:1
ratio), has been activated in N2 flow for 1 h at 25 ºC. In the TPR of the 18O2 oxidized samples, the
sample, after activation, was exposed to a 20% 18O2 in Ar flow for 1 h at 90 ºC, followed by
decreasing the temperature to 40 ºC in an Ar flow and kept at that temperature for 20 min.
Afterwards the gas feed was switch to 10% H2 in Ar, and the temperature raised to 450 ºC, with a
heating rate of 2 ºC/min. In the TPR of the phenylacetylene+18O2 experiments the sample, after
activation, was exposed to a phenylacetylen+18O2 flow at 90 ºC for 1h, followed by decreasing the
temperature to 40 ºC in an Ar flow and kept at that temperature for 20 min. Afterwards the gas feed
was switch to 10% H2 in Ar, and the temperature raised to 450 ºC, with a heating rate of 2 ºC/min.
Powder Xꢀray diffraction (XRD) was performed with a HTPhilips X’Pert MPD diffractometer
equipped with a PW3050 goniometer using CuKα radiation and a multisampling handler.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 29
1
2
3
4
5
6
Sample compositions of the DMSO solution of CuOx NPs were measured after acidꢀdigestion by
ICPꢀOES on a Varian 715ꢀES instrument.
7
8
9
The redox properties of copper particles on supports are evaluated by Temperatureꢀprogrammed
reduction (TPR). Micromeritics AutoChem 2910 catalyst characterization system with a thermal
conductivity detector (TCD) was used. Prior to each experiment, about 100 mg of sample was
pretreated at room temperature in flowing He (10 mL/min) for 20 min. The sample was treated by
heating from 25 °C to 600 °C at a rate of 5 °C min–1 in a flow of 10 vol.% H2 in Ar. The total gas
flow rate was 50 mL/min–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
Fluorescence spectra were obtained with a LP Sꢀ220B (Photon Technology International)
equipped with 75 W Xe lamp.
2.3 Oxidative homoꢀcoupling of alkynes.
2.3.1 Oxidative homoꢀcoupling of alkynes by monodispersed CuOx NPs
110 ꢁL of alkyne (phenylacetylene) and 50 ꢁL nitrobenzene (as internal standard) and a given
amount of copper catalyst (Cu amount ranges from 0.5 mol% to 10.0 mol%) were added to 2 ml of
solvent (DMSO) in 10 mL roundꢀbottom test tube with stirrer. The reaction flask was introduced in a
preheated silicon oil bath at 90 oC and stirred at around 1000 rpm under oxygen atmosphere (1 bar).
Gas chromatography (Bruker with FID detector) was used for quantitative analysis of substrate and
product.
2.3.2 Oxidative homoꢀcoupling of alkynes by supported CuOx NPs
55 ꢁL of alkyne (phenylacetylene) and 50 ꢁL nitrobenzene (as internal standard) and supported
copper catalyst (Cu amount ranges from 1.0 mol% to 8.0 mol%) were added to 2 ml of solvent
(Toluene) in 10 mL roundꢀbottom test tube with stirrer. The reaction flask was introduced in a
preheated silicon oil bath at 90 oC and stirred at around 1000 rpm under oxygen atmosphere (1 bar).
Gas chromatography (Bruker with FID detector) was used for quantitative analysis of substrate and
product.
2.3.3 Oxidative heteroꢀcoupling of alkynes by supported CuOx NPs
0.5 mmol alkyne A, 0.25 mmol alkyne B and 40 ꢁL dodecane (as internal standard) and supported
copper catalyst (CuOx/TiO2ꢀ2.5%, 100 mg) were added to 2 ml of solvent (Toluene) in 10 mL
6
ACS Paragon Plus Environment

Page 7 of 29
ACS Catalysis
1
2
3
4
5
6
roundꢀbottom test tube with stirrer. The reaction flask was introduced in a preheated silicon oil bath
o
at 100 C and stirred at 1000 rpm under oxygen atmosphere (5 bar). Gas chromatography (Bruker
7
8
with FID detector) was used for quantitative analysis of substrate and product.
9
3. Results and Discussions
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
Figure 1. (aꢀd) STEM images of small Cu2O NPs with different sizes, a) CuOxꢀNC, b) CuOxꢀ1.8, c)
CuOxꢀ2.2 and d) CuOxꢀ2.5. (eꢀh) size distributions of small CuOx NPs with different sizes, e)
CuOxꢀNC, f) CuOxꢀ1.8, g) CuOxꢀ2.2 and h) CuOxꢀ2.5.
3.1 Synthesis and characterizations of monodispersed CuOx nanoparticles
Monodispersed CuOx NPs could be easily synthesized by a thermal decomposition of CuCl in
DMSO in presence of air. CuCl powder is not soluble in DMSO at room temperature. However, after
o
heating the suspension at 90 C for 20~30 min in air, a transparent solution was obtained. If
CuClꢀDMSO mixture is heated in N2 atmosphere, only small amount of CuCl can be dissolved in
DMSO, suggesting that O2 participate in the thermal decomposition process (Figure S1). When the
concentration of Cu is as low as 0.08 mg/mL, CuOx NPs (denoted in this case as CuOxꢀNC) with
sizes below 2 nm can be seen in the STEM images. According to the STEM images, the average size
of those small CuOx NPs is about 1.5 nm. However, it should be noticed that a large number of Cu
clusters with size below 1 nm appear to exist but, due to the limitation of our STEM, they cannot be
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 29
1
2
3
4
5
6
7
8
9
seen clearly. Indeed, photoemission studies show the presence of Cu clusters with 10ꢀ20 atoms (see
the following discussion). So, the ~1.5 nm average size given is certainly an over estimation of the
average particle size. Increasing the concentration of Cu species leads to increasing the size of CuOx
NPs. Then by controlling the concentration of Cu, CuOx NPs with different sizes and narrow size
distributions can be obtained ranging from clusters (< 1 nm) up to ~5 nm, as shown in Figure 1 and
Figure S2.
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
In a deeper structural characterization, the crystal lattice information of the CuOx NPs was
acquired with highꢀresolution TEM. As shown in Figure S3, lattice fringes corresponding to Cu2O
and CuO can be seen in the NPs of those four samples (From CuOxꢀNC to CuOxꢀ2.5), suggesting
that they should consist of both Cu(I) and Cu(II).15,16 The elemental mapping results also confirm
that those nanoparticles consist of both Cu and O (Figure S4a). Very small amounts of Cl have also
been detected in several analyzed areas (Figure S4b). Compared with Cu, the amount of Cl is much
lower, suggesting that part of Cl is still residual in the DMSO solution after the thermal
decomposition of CuCl. In the HRTEM images of CuOx NPs with larger size (CuOxꢀ4.5), most of
the particles show lattice fringes of CuO (Figure S2). Thus, based on these HRTEM images, it is
implied that the chemical states of Cu species may be different in different samples, depending on
the particle size of CuOx. In addition, the formation of CuOX NP due to thermal decomposition of
the CuCl precursor in air, has been supported by Raman spectroscopy where the absence of CuꢀCl
vibrations (~200cmꢀ1)17 while the presence of CuꢀOꢀCu stretching modes (~434, 493 and 582cmꢀ1)
18,19has been identified in all three samples (Figure 2d).
8
ACS Paragon Plus Environment

Page 9 of 29
ACS Catalysis
1
2
3
4
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
Figure 2. Spectroscopic characterizations of monodispersed CuOx NPs of different sizes. Cu 2p 3/2
XPS peak (a), CuL3VV Auger line (b) and fluorescence spectra at 350 nm excitation (c) and Raman
spectra (d).
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 29
1
2
3
4
Table 1 Fitting results of XPS spectra of CuOxꢀ4.5, CuOxꢀ2.2 and CuOxꢀNC. a) Cu(II)/Cu(I)
5
6
7
8
atomic ratio determined from XPS peak decovolution b) Satellite to main peak area ratio
Cu(II) (934.6eV)/Cu(I) (932.6eV)a
Sample
CuOxꢀNC
CuOxꢀ1.8
CuOxꢀ4.5
s/mb
0
0.16
2.8
5
9
0.35
0.66
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 Cu 2p3/2 Xꢀray photoꢀelectron spectra (XPS) and Cu L3VV Auger spectra of the above
samples are shown in Figure 2A and Figure 2B respectively. Two components at binding energy (BE)
of 934.6eV and 932.6eV are observed in all samples but in different proportion. The higher BE of
934.6eV is due to Cu2+, while the lower BE of 932.6eV can be related either to Cu(I) or metallic
Cu0.20 Assignation to Cu0 can be excluded in samples CuOxꢀ1.8 and CuOxꢀ4.5 according to the peak
position and band shape of the auger Cu L3VV peak. An anomalous Cu L3VV KE value is observed
on sample CuOxꢀNC (915.5eV). Similar behavior, due to a small particle size effect, has already
been observed by others authors,21 hindering true identification of copper oxidation state. Following
the above assignment, it can be seen that from the values given in Table 1, the relative amount of
Cu(II) to Cu(I) increases when increasing particle size (from clusters to 5 nm), which fits with the
increase in the shakeꢀup satellite intensity. Moreover, the Cu L3VV peak position agrees with a
higher contribution of Cu(II) in sample CuOxꢀ4.5 (Cu L3VV of 917.8 eV) and higher contribution of
Cu(I) in sample CuOxꢀ1.8 (Cu L3VV of 916.8 eV).22 CuOxꢀNC is a mixture of clusters and
nanoparticles, where the nanoparticles may exist in an oxidized form, according to their raman
spectra, and mayoritary as Cu(I) due to the absence of satellite structure in the XPS spectra. Clꢀ ions
have also been detected in the XPS analysis, which unavoidable comes from the thermal
decomposition of the CuCl precursor in air. The existing form of Clꢀ is difficult to identify according
to XPS BE data, while it may most probably remain dissolved in small amount of water present in
the air or in the solvent.
The existence of Cu clusters in CuOxꢀNC sample is also confirmed by fluorescence spectroscopy.
As shown in Figure 2c, a strong emission peak can be found around 390 nm, which corresponds to
Cu clusters with 10~20 atoms.14 Pure CuCl solid shows no fluorescence emission, implying that
CuCl is transformed into CuOx clusters in DMSO. The fluorescence intensity decreases dramatically
when the particle size increases to 1.8 nm, due to the high sensitivity of fluorescence with particle
10
ACS Paragon Plus Environment

Page 11 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
size. For CuOxꢀ4.5 sample, almost no emission signal can be observed under those conditions.
Accordingly with the characterization results observed up to now, it is possible to say that CuOx NPs
with controllable and very narrow size distribution, ranging from clusters to ~ 5 nm NPs, can be
easily obtained by decomposing CuCl in DMSO in the presence of air, and the size of CuOx NPs can
be tuned by changing the Cu concentration. Moreover, we can also say that the chemical state of Cu
in the CuOx NPs changes with the particle size, increasing the Cu(II) to Cu(I) ratio when increasing
the size of CuOx NPs.
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
3.2 Catalytic properties of monodispersed CuOx NPs
CuCl, 0.5-10mol%
2 mL DMSO, 90 oC, O2, 1 bar
1.0 mmol
Figure 3. (a) The kinetic curves of CuClꢀcatalyzed homoꢀcoupling of phenylacetylene using
different amounts of CuCl. (b) graphical zoom in the first 2.5 h.
Cu compounds are widely used as superior catalysts for the oxidative homoꢀcoupling of alkynes.
Cu salts in pyridine media or in alcohol media can give high conversion for oxidative coupling of
alkynes.23,24 It is not infrequent to find in the literatures that the reaction mechanism is thought to
involve the mononuclear Cuꢀalkyne complex. However, since Cu species can evolve during the
reaction, one has to investigate the dynamic evolution of Cu species in the catalytic process to
determine which are the catalytically active species. Therefore, we have used CuCl as the starting
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 29
1
2
3
4
5
6
7
8
9
catalyst for performing the homoꢀcoupling of phenylacetylene. Different amounts of CuCl were used
as catalyst and the evolution of the reactants and catalyst were followed with time. As shown in
Figure 3a and Figure 3b, it can be seen that when the amount of CuCl was below 5 mol%, the
reaction goes through an induction period. During that time, conversion was not observed in the first
10ꢀ20 min. Furthermore, the induction period decreases when increasing the amount of CuCl. These
kinetic results indicate that the initial CuCl may not be the real active catalyst, and some in situ
transformation of CuCl may occur during reaction that give rise to the catalytically active Cu species.
In order to identify the evolution process of CuCl during the homoꢀcoupling of the alkyne, the
following experiment was carried out. A reaction mixture was prepared in one reactor with 10 mol%
CuCl and phenylacetylene. An aliquot of the reaction mixture was taken at different reaction times
and introduced into a second reactor in where only the phenylacetylene was present. The amount of
Cu in the second reactor was kept as 0.5 mol%. It should be remembered that when the oxidative
coupling of phenylacetylene was catalyzed by 0.5 mol% of CuCl, an induction period of about 1 h
was observed (see Figure 3). As shown in Figure S5, an induction period was observed in the
second reaction when the aliquot from the first reactor was taken at time zero with the still no
evolved original CuCl. However, if the aliquot of the reaction mixture was taken from the first
reactor after 0.5 h, and it was added into the second reactor, the reactor started immediately without
induction period (see Figure S5). The above experiment shows that, whatever the Cu active species
for the oxidative coupling of alkynes are, they are not the CuCl but are formed in situ during the
reaction. Therefore, we follow the evolution of the Cu species by taking samples at different reaction
times. We choose the homoꢀcoupling reaction catalyzed by 1.0 mol% of CuCl and the Cu species in
solution were analyzed by with STEM. STEM results (see Figure S6) show that the reaction starts
when nanoparticles with size between 1 and 3 nm are present. Elemental mapping and
highꢀresolution TEM images (see Figure S7) confirm that those nanoparticles should be CuOx NPs
with lattice fringes corresponding to Cu2O. The results imply that, the in situ transformation of CuCl
into CuOx NPs may play an important role in the oxidative coupling of alkynes.
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
Similar phenomenon of the in situ transformation of Cu species during the oxidative coupling of
alkynes have also been observed when other starting Cu compounds were used as the starting
catalysts. For example, when 5.0 mol% of Cu(Ac)2 was used as catalyst, an induction period of about
12
ACS Paragon Plus Environment

Page 13 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
2 h occurs (see Figure S8) and small CuOx NPs can also be observed with STEM after the induction
period, as shown in Figure S9. For different Cu compounds, the induction period may be different,
but when the reaction takes off, the in situ formed CuOx NPs show similar average sizes of ~2 nm.
This common point implies that small CuOx NPs may play an important role in the oxidative
homoꢀcoupling of alkynes. Therefore, since CuOx NPs of controllable sizes can now be prepared, we
performed the reaction using the separately prepared monodispersed CuOx NPs described before.
Moreover, we have also used Cu clusters with 8ꢀ16 atoms prepared by electrochemical method (see
experimental section) and their catalytic property was measured. Results given in Figure S10 clearly
show that metal clusters give no activity for the oxidative homoꢀcoupling of phenylacetylene.
Moreover, the monodispersed CuOx samples (CuOxꢀNC and CuOxꢀ2.2) are not only active but also
give no induction period. Nevertheless, CuOxꢀ2.2 show higher activity than CuOxꢀNC. Then,
considering the presence of both small clusters and NPs in the CuOxꢀNC sample, we can speculate
that only CuOx NPs are responsive for the catalytic activity.
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
Figure 4. Kinetic studies of small CuOx NPs for homoꢀcoupling of phenylacetylene. The initial TOF
values of different CuOx NPs are calculated based on considering all the Cu species used in the
reaction.
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 29
1
2
3
4
5
6
7
8
9
The catalytic properties of the other CuOx NPs with different sizes have also been tested, and the
results given in Figure 4 show a maximum activity for CuOx NPs around ~2 nm. Moreover, when
the initial reaction rate is plotted versus the amount of CuOxꢀ2.2 NPs added as catalyst (shown in
Figure S11a and Figure S11b), a linear relationship is observed, indicating that the CuOx NPs are
the active species for the oxidative coupling of phenylacetylene. The morphology of the CuOxꢀ2.2
NPs was again checked after the homoꢀcoupling reaction (after reaction time of 4 h) and the size
distribution is very similar to the fresh sample (see Figure S11c and Figure S11d). It should be
remarked here that, the CuOx NPs with larger size (CuOxꢀ4.5) show a short but visible induction
period (about 10 min) for the homoꢀcoupling of phenylacetylene (see Figure S12). STEM and
HRTEM images of CuOxꢀ4.5 NPs after the homoꢀcoupling reaction indicate that they have been
transformed into CuOx NPs of around 2 nm during the reaction (see Figure S13). These results are
again in line with the conclusion that CuOx NPs of ~2 nm are the active species for the reaction.
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 Preparation and characterizations of supported CuOx NPs
From the work on isolated CuOx NPs, we have learnt that oxidative coupling of phenylacetylene
is better catalyzed by CuOx NPs of ~2 nm. Thus, before studying why this occurs, we thought on the
possibility to extend the knowledge acquired to prepare active supported Cu catalyst for the above
reaction, by generating the adequate CuOx NPs on a support, to prepare an optimized and recyclable
solid catalyst. Herein, a series of CuOx/TiO2 samples (denoted as CuOx/TiO2ꢀx%, where x is the
loading weight percentage of Cu) were prepared through an impregnation method (see experimental
section). Their XRD patterns are shown in Figure S14. Only diffraction peaks of anatase TiO2 and
rutile TiO2 can be seen when the loading amount of Cu is lower than 5.0 %. For CuOx/TiO2 samples
with 10.0 wt% of Cu, diffraction peaks corresponding to CuO can be observed, indicating the
formation of big CuO NPs in this sample. As shown in Figure S15, the sizes of the different CuOx
NPs supported on TiO2 were also investigated by STEM. According to the STEM images, their sizes
range from below 1 nm to ca. 10 nm. The elemental mapping results (shown in Figure S16 to Figure
S20) also confirm the uniform dispersion of CuOx NPs on TiO2 support. The size distributions of
CuOx NPs are also presented in Figure S21, indicating that the size of CuOx NPs increases when
increasing the loading of CuOx on TiO2.
14
ACS Paragon Plus Environment

Page 15 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
9
The chemical states of Cu species in (CuOx/TiO2ꢀ2.5% and CuOx/TiO2ꢀ10%) were measured by
XPS. From Figure S22, it can be deduced that both Cu+ and Cu2+ are present in the two samples.
Moreover, CuOx/TiO2ꢀ10% has a higher ratio of Cu2+/Cu+ than CuOx/TiO2ꢀ2.5. The reducibility of
CuOx NPs with different sizes was also measured by temperatureꢀprogrammed reduction (TPR). As
shown in Figure S23, for CuOx/TiO2ꢀ0.25% the H2 consumption is very low compared to
CuOx/TiO2ꢀ2.5% and to CuOx/TiO2ꢀ10% with lager particle sizes, implying the low redox reactivity
of the subnanometric CuOx species in CuOx/TiO2ꢀ0.25%.13 For CuOx/TiO2ꢀ2.5%, a big sharp peak
(α peak) can be observed in the TPR profile, which is associated to the reduction of small CuOx NPs
of around 2 nm. In the case of CuOx/TiO2ꢀ10%, two reduction peaks (α and β peak) can be observed,
which can be ascribed to small CuOx NPs and big CuOx NPs, respectively.25 Considering that the H2
consumption corresponding to the β peak is much larger than that for the α peak, it could be deduced
that most CuOx NPs in CuOx/TiO2ꢀ10% are big CuOx NPs, which is consistent with the STEM
characterization.
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
Figure 5 Kinetic studies(a) and the initial reaction rate (b) of CuOx/TiO2 samples with different sizes
of CuOx NPs in the homoꢀcoupling reaction. The amount of Cu in all the tests are kept as 8.0 mol%.
3.4 Catalytic properties of supported CuOx NPs
The activities of CuOx/TiO2 samples in the homoꢀcoupling of phenylacetylene were measured
under the experimental conditions described in the experimental section. As shown in Figure 5a, no
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 29
1
2
3
4
5
6
7
8
9
induction period was observed and the activity of CuOx/TiO2 samples is related with the sizes of the
CuOx NPs formed on TiO2. CuOx/TiO2ꢀ2.5% sample with CuOx NPs around 2 nm gives the best
activity compared with smaller or larger CuOx NPs (see Figure 5b), and the reaction can be finished
in 40 min when CuOx/TiO2ꢀ2.5% is used as catalyst.
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
Combining the results on monodispersed and supported CuOx NPs, we can propose that the
suitable size of CuOx NPs in CuOx/TiO2ꢀ2.5% should be the reason accounting for its high activity.
Furthermore, the sizeꢀdependent catalytic properties of CuOx NPs follow a similar trend on both
monodispersed and supported NPs.
The activity CuOx/TiO2ꢀ2.5% sample was also tested for homoꢀcoupling of phenylacetylene using
different amounts of Cu as the catalyst. As presented in Figure S24a, the reaction will boost once the
catalyst is added into the reaction mixture, indicating that CuOx/TiO2ꢀ2.5% sample can catalyze the
reaction directly. When the catalyst is separated from the reaction mixture, the formation of the
coupling product stops, indicating that the active Cu species are those supported on TiO2 and the
process is heterogeneous. The linear relationship between the initial reaction rate and amount of Cu
species (shown in Figure S24b) also confirms that small CuOx NPs supported on TiO2 are the active
species for the oxidative coupling of alkynes.
As discussed previously, Cu clusters seem not to be active for the oxidative coupling of
phenylacetylene. Theoretical calculations have shown that the activation energy of O2 on Cu clusters
is quite high,13 which may explain the very low activity of Cu clusters. In the CuOx/TiO2ꢀ0.25%
catalyst, Cu exists as subnanometric species dispersed on TiO2, namely supported Cu clusters. As can
be seen in Figure 5a, CuOx/TiO2ꢀ0.25% sample shows negligible activity and, in any case, the
activity per Cu atom is much lower than that for CuOx/TiO2ꢀ2.5% with CuOx NPs around 2 nm. On
the other hand, when the size of CuOx NPs in CuOx/TiO2 sample is larger than 5 nm, the activity
drops dramatically, indicating that only CuOx NPs around 2 nm are active. It should be taken into
account that, due to the stabilization effect of TiO2 support and, specially, to the use of toluene
instead of DMSO as solvent, large CuOx NPs (≥ 5 nm) cannot undergo in situ transformation into
smaller NPs during the oxidative coupling reaction, what results in their low activity. At this point, it
has become clear that regardless how the CuOx NPs are obtained, a maximum activity for the
oxidative coupling of phenylacetylene occurs with CuOx NPs of ~2 nm. Thus, now we will attempt
16
ACS Paragon Plus Environment

Page 17 of 29
ACS Catalysis
1
2
3
4
to explain the sizeꢀdependent activity of CuOx NPs for the oxidative coupling of alkynes.
5
6
7
8
3.5 Kinetic and in situ spectroscopic studies
9
The global process for oxidative coupling of alkynes can be decomposed into the following
elementary steps: activation of the CꢀH bond in the alkyne group, coupling reaction, O2 activation
and reaction of the adsorbed H with the activated O2.26 To find which one is the reaction controlling
step, CuOx/TiO2ꢀ2.5% was used as catalysts and the initial rate of the reaction was measured by
keeping the O2 pressure (1 bar) constant and changing the initial concentration of phenylacetylene.
Results in Figure S25 indicate that under these reaction conditions, the rate of the reaction does not
depend on the concentration of alkyne, indicating that neither the phenylacetylene activation nor the
homoꢀcoupling can be the rateꢀcontrolling step. On the other hand, the initial rate of the reaction
clearly increases when increasing the oxygen pressure while keeping the concentration of
phenylacetylene constant. From the kinetic results, one should then conclude that the controlling step
during the oxidative coupling of alkynes corresponds to O2 activation, or to the reaction between
activated O2 and the abstracted hydrogen.
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
With respect to the molecular interaction of the reactants with the catalyst, the surface interaction
of phenylacetylene was clearly seen from the UVꢀvis spectra of the alkyne adsorbed on the
monodispersed CuOx NPs (see Figure S26). When phenylacetylene is interacted with CuOx NPs at
room temperature, no obvious change occurs in the UVꢀvis spectra. However, when the system is
heated to the reaction temperature (90 oC), a new peak at ~475 nm is observed that can be associated
to the formation of C≡CꢀCu bond, suggesting that the adsorption and dissociation of the CꢀH bond of
phenylacetylene on the CuOx NPs has occurred.27 After one hour, when the phenylacetylene has
been consumed, the peak at ~475 nm disappears.
Alkyne activation on supported CuOx NPs has also been studied by in situ IR spectroscopy. When
deuterated phenylacetylene is adsorbed on TiO2, CuOx/TiO2ꢀ2.5% (active sample) and
CuOx/TiO2ꢀ10% (nonꢀactive sample), a band corresponding to OꢀD bond has been identified in all
samples at room temperature (see Figure S27). It was not possible to identify whether the ꢀOD
groups come from the TiO2 support or from the CuOX NP, due to the fact that a fast isotopic scrambling
with the ꢀOH groups of the support can also occur. On the other hand, the formation of C
≡CꢀCu bond
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 29
1
2
3
4
5
6
7
8
9
at the surface is consistent with the presence of a rather asymmetric C≡C IR bond in both CuOx/TiO2
samples (see Figure S28) when compared with that formed on pure TiO2. The asymmetric vibration
mode of C≡C bond in IR spectra can be due to the interaction of the PhꢀC≡Cꢀ* adduct with surface
copper sites on both CuOx/TiO2ꢀ2.5% and CuOx/TiO2ꢀ10%.28 According to the data from UVꢀvis
and IR spectra, C≡CꢀH activation really occur on the surface of the CuOx nanoparticles.
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
XPS results of CuOx/TiO2ꢀ2.5% and CuOx/TiO2ꢀ10% presented before showed that the chemical
states of Cu species in the two samples are different, which may be a reason for their different
catalytic activity. To further investigate the nature of the surface Cu species on CuOx/TiO2ꢀ2.5% and
CuOx/TiO2ꢀ10% in the presence of phenylacetylene and O2, CO has been used as probe molecule for
in situ IR spectroscopy. Indeed, CO is very sensitive to Cu(I) species while Cu(II) cannot be detected
by CO even at low temperature.29,30 As depicted in Figure S29, Cu(I) has been observed on both
o
samples after the adsorption of phenylacetylene at 90 C for 1 hour due to the surface reduction
o
reaction. After O2 addition at 90 C, interestingly, the amount of Cu(I) remains stable on the
CuOx/TiO2ꢀ2.5% sample while the amount of Cu(I) clearly decreases on CuOx/TiO2ꢀ10% sample as
a result of reꢀoxidation of Cu(I) into Cu(II) by O2. In other words, these results indicate that the type
of interaction between O2 and the CuOx NPs and the nature of the active oxygen species are different
in the two samples. Therefore, since the rateꢀcontrolling step in the oxidative coupling of alkynes is
O2 activation, the observed size effects of CuOx NPs on the catalytic properties for oxidative
coupling of alkynes should be related to their influence on the activation of O2.
18
ACS Paragon Plus Environment

Page 19 of 29
ACS Catalysis
1
2
3
4
5
3.6 Relation between particle size of CuOx and O2 activation
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
Figure 6 In situ Raman spectra of CuOx/TiO2ꢀ2.5% (a) and CuOx/TiO2ꢀ10% (b) in O2 atmosphere
after preꢀtreatments in reactant mixture of phenylacetylene and O2. In (a) and (b), spectra were
o
o
o
obtained at different temperature. (i) 25 C, (ii) 60 C and (iii) 90 C. For comparison, the fresh
CuOx/TiO2 samples and phenylacetylene adsorbed on CuOx/TiO2 are also presented.
To further identify the nature of surface species generated under reaction conditions, in situ Raman
and in situ IR studies have been performed on both CuOx/TiO2ꢀ2.5% and CuOx/TiO2ꢀ10% samples.
Figure 6a shows the in situ Raman spectra acquired on the active CuOx/TiO2ꢀ2.5% sample in the
presence of phenylacetylene and O2 at different reaction temperatures. A Raman band at 998 cmꢀ1,
ascribed to peroxo species, is clearly detected at 25 ºC, 60 ºC and at the reaction temperature used in
the catalytic studies (90 ºC).31 No peroxo species have been detected on pure TiO2. Notice that a
decrease in the intensity of the 998 cmꢀ1 band at 90 ºC occurs, which could be due to thermal
desorption or to dissociation into atomic oxygen, and regeneration of surface lattice oxygen species.
Moreover, peroxo species (IR active vibration mode at 1227 and 1217 cmꢀ1) have also been detected
by in situ IR studies performed at different reaction temperatures (Figure S30a).32 These results
indicate that peroxo species can be formed on the surface of CuOx/TiO2ꢀ2.5 sample. In opposite, on
the much less active CuOx/TiO2ꢀ10% sample, peroxo species have not been observed during the in
situ Raman studies (Figure 6b), and peroxo species (1227 and 1217 cmꢀ1) are detected by IR
o
spectroscopy only at 25 ºC and disappear after increasing temperatures to 60 C (Figure S30b).
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 29
1
2
3
4
5
6
7
8
9
Based on these data, it can be speculated that oxygen as peroxo species is preferentially stabilized on
the CuOX/TiO2ꢀ2.5% sample. In contrast, the peroxo species and less favorable on CuOx/TiO2ꢀ10%
at the reaction temperature. Considering the high reactivity of peroxo species in selective oxidation
reaction, the formation of peroxo species could be the reason accounting for the higher activity of
CuOx/TiO2ꢀ2.5%.33
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
Figure 7 TPR profile on 18O2 oxidized CuOX/TiO2ꢀ10% and CuOX/TiO2ꢀ2.5% samples (a, b), and
after in situ reaction in the presence of phenylacetlyene and 18O2 (c, d).
It has been presented in the literature that oxygen activation can also be performed with the help of
surface lattice defects.34 Since we cannot observe surface adsorbed active oxygen species on
CuOx/TiO2ꢀ10%, the activation of O2 on CuOx/TiO2ꢀ10% may be performed through the
reꢀoxidation of Cu(I) species on big CuOx NPs as observed in the IR spectra (see Figure S29). In
order to distinguish whether the different reactivity is associated to peroxo species or lattice oxygen
species, 18O2 has been used for the in situ spectroscopic studies. By analyzing the composition of
20
ACS Paragon Plus Environment

Page 21 of 29
ACS Catalysis
1
2
16
18
3
water (H2 O or H2 O) formed after reaction by H2 titration of the catalyst (TPR), it could be
distinguished whether the O in the water comes from the 18O2 feed or from the lattice oxygen of
CuOx/TiO2 catalyst (the corresponding TPR profile are shown in Figure 7). For comparison, the
fresh CuOx/TiO2ꢀ2.5% and CuOx/TiO2ꢀ10% were oxidized by 18O2 at 90 oC and then reduced by H2.
4
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
The TPR profile of the 18O2 oxidized CuOx/TiO2ꢀ10% sample shows both H2 O and H2 O
16
18
16
formation with an onset temperature of 136 ºC in both cases. H2 O formation takes place in a very
broad temperature range of 136ꢀ355 ºC, corresponding to lattice oxygen species of different
reactivity. A small amount of H2 O is formed, indicating the oxygen exchange between 18O2 and the
18
CuOx NPs in CuOx/TiO2ꢀ10% (see Figure 7a). In opposite the TPR profile of the 18O2 oxidized
16
CuOx/TiO2ꢀ2.5% sample, shows only one peak corresponding to H2 O with an onset temperature of
18
150 ºC, corresponding to lattice oxygen species. Interestingly H2 O is not detected in this sample,
indicating that no oxygen exchange between the O2 and the CuOx lattice takes place on
CuOx/TiO2ꢀ2.5% (Figure 7b).
18
When the two samples have been exposed to a mixture of phenylacetylene and O2 at reaction
conditions (90 ºC, 1h), different TPR profiles have been observed in both cases. On the
16
18
CuOx/TiO2ꢀ2.5% sample, neither H2 O nor H2 O are detected (see Figure 7d). It means, the
surface has been reduced under reaction conditions, but it is not reꢀoxidized by 18O2. With
16
18
CuOx/TiO2ꢀ10% sample, both H2 O and H2 O are observed with an onset temperature of 203 ºC
18
(Figure 7c), which indicates the surface has been partially reduced and reꢀoxidized by O2 under
reaction conditions.
From the above presented results combined with the in situ Raman and IR studies, we speculate
that oxygen is activated by CuOx NPs in the CuOx/TiO2ꢀ2.5% sample as peroxo species on the
catalyst surface. Meanwhile, in the case of CuOx/TiO2ꢀ10% sample, O2 dissociation occurs,
transforming into lattice oxygen species. This hypothesis would also be supported by the IR results
where the catalysts surface has been titrated, using CO as probe molecule, after phenylacetylene
adsorption at 90 ºC and after subsequent O2 insertion at the same temperature (see Figure S29).
Thus we can conclude that the different reactivity observed between the CuOx/TiO2ꢀ2.5% and
CuOx/TiO2ꢀ10% samples is related to a different nature of the oxygen species on the catalyst surface.
Stabilization of molecular oxygen as peroxo species or further dissociation into atomic oxygen and
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 29
1
2
3
4
5
6
7
8
9
generation of lattice oxygen species is strongly related to the different particle size of CuOx NP in
both samples. On CuOx NPs around 2 nm, O2 molecules are activated to form peroxo species, which
is more active than surface lattice oxygen formed on big CuOx NPs in CuOx/TiO2ꢀ10%.35,36 As a
consequence, CuOx/TiO2ꢀ2.5% with ~2 nm particle size shows much higher activity than
CuOx/TiO2ꢀ10% with ~10 nm particle size in the oxidative coupling of alkynes. Notice that this is
the case regardless that monodispersed isolated CuOx NPs or supported NPs are used as catalysts.
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
22
ACS Paragon Plus Environment

Page 23 of 29
ACS Catalysis
1
2
3
4
5
3.7 Proposed reaction mechanism
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
Figure 8 (a) Schematic illustration of the dynamic transformation of monodispersed Cu species.
Sizeꢀdependent catalytic mechanism of supported CuOx NPs, (b) CuOx/TiO2ꢀ0.25%, (c)
CuOx/TiO2ꢀ2.5% and (d) CuOx/TiO2ꢀ10%.
Based on the above results and discussions, a proposed reaction mechanism for CuOx NPs with
different sizes is shown in Figure 8. For monodispersed Cu species (see Figure 8a), small CuOx
NPs around 2 nm are formed from the in situ transformation of bulk CuCl, as a result of which an
induction period can be observed. These in situ formed small CuOx NPs are the active species for the
oxidative coupling of alkynes when DMSO was chosen as solvent. When a large amount of CuCl is
used as the catalyst, small CuOx NPs will further grow into big CuOx NPs around 5 nm, which can
transform back with time to small CuOx NPs in the oxidative coupling of alkynes. In the case of
supported CuOx NPs (see Figure 8b to Figure 8d), the activity will also be dependent on the size of
CuOx NPs, being the dynamic transformation between different sizes of CuOx NPs restricted under
the reaction condition used here. For Cu clusters supported on TiO2, according to theoretical
calculation, O2 activation is not favorable, resulting in their very low activity. When the size of CuOx
NPs increases to around 2 nm, O2 can be activated into active peroxo species on the surface of
CuOx/TiO2ꢀ2.5%, which can serve as the active intermediate for the oxidative coupling of alkynes.
However, a further increase in the size of CuOx NPs will change the nature of the oxygen species
formed. O2 will form lattice oxygen species after activation by large CuOx NPs, which is less active
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 29
1
2
3
4
for the oxidative coupling reaction.
5
6
7
8
3.8 Synthesis of unsymmetric 1,3ꢀdiynes with CuOx/TiO2 catalyst
9
Compared with oxidative homoꢀcoupling of alkynes, the coupling reaction between two different
alkyne molecules is more difficult due to their different reactivity. The yield of unsymmetric
1,3ꢀdiynes is much lower than the oxidative homoꢀcoupling of alkynes as a result of the competitive
reaction pathways between homoꢀcoupling and heteroꢀcoupling reactions.37 Therefore, in some cases,
only trace amount of 1,3ꢀdiynes can be obtained when Cuꢀbased catalyst is used, even though one
alkyne substrate is in a large excess (the molar ratio of two alkynes is 5.).38,39 Since the CuOx NPs
supported TiO2 in the CuOx/TiO2ꢀ2.5% sample can catalyze the oxidative coupling of alkynes
directly, the oxidative heteroꢀcoupling of different alkynes could be processed smoothly with higher
yield of unsymmetric 1,3ꢀdiynes. The catalytic results in the synthesis of unsymmetric 1,3ꢀdiynes
with CuOx/TiO2 catalyst are shown in Table 2. It should be noted that, in this work, the molar ratio
of two alkynes is 2. And higher temperature and O2 pressure is used to accelerate the heteroꢀcoupling
between different alkynes. As can be seen in Table 2, unsymmetric 1,3ꢀdiynes can be obtained with
moderate to good yield, ranging from 61% to 92%. The coupling between phenylacetylene and
hexyne can be realized with 61% yield although the amount of phenylacetylene is only twice that of
hexyne. It should be noticed that only trace amount of the unsymmetric coupling product was
obtained in previous work (but five times excess of hexyne).40 These results suggest that supported
small CuOx NPs can work as efficient catalyst for synthesis of unsymmetric 1,3ꢀdiynes.
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. Synthesis of unsymmetric 1,3ꢀdiynes with CuOx/TiO2 catalysta
Entry
Yield/%
1
87
24
ACS Paragon Plus Environment

Page 25 of 29
ACS Catalysis
1
2
3
4
5
6
7
8
2
3
4
5
6
7
92
76
61
88
82
92
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
a
40 ꢁL dodecane was added as internal standard and the reaction time was 8 h. The yield is
calculated based on
.
4. Conclusions
In this work, a facile method for synthesis of monodispersed small CuOx NPs has been presented.
Through kinetic analysis and following the dynamic transformation of Cu species, small CuOx NPs
around 2 nm are proved to be the active species in Cuꢀcatalyzed oxidative coupling of alkynes.
Furthermore, the sizeꢀdependent catalytic properties of supported CuOx NPs on TiO2 was also
studied. Similar activity trend is obtained on supported CuOx NPs and monodispersed CuOx NPs.
By spectroscopic characterizations, it has been clarified that the nature of the active oxygen species
are dependent on the size of CuOx NPs. Active peroxo species are preferentially formed on small
CuOx NPs (~2 nm) while O2 prefers to form lattice oxygen species after dissociation on large CuOx
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 29
1
2
3
4
5
6
7
8
NPs (~10 nm). Therefore, the catalytic properties in oxidative coupling of alkynes are greatly
dependent on the size of CuOx. Finally, monodispersed CuOx NPs are able to catalyze the
unsymmetric coupling of alkynes.
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
ASSOCIATED CONTENT
Supporting Information
This material is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
*Corresponding author: acorma@itq.upv.es (A.C.)
Notes
The authors declare no competing financial interests.
ACKNOWLEDGEMENT
L. L. thanks ITQ for providing a contract. The authors also thank Microscopy Service of UVP for
kind help on TEM and STEM measurements. Financial supports from ConsoliderꢀIngenio 2010
(project MULTICAT) and “Severo Ochoa” program are also gratefully acknowledged. The European
Union is also acknowledged by ERCꢀAdGꢀ2014ꢀ671093 — SynCatMatch.
References
(1) Kuo, C.ꢀH.; Huang, M. H. Nano Today 2010, 5, 106ꢀ116.
(2) Zhang, Q.; Zhang, K.; Xu, D.; Yang, G.; Huang, H.; Nie, F.; Liu, C.; Yang, S. Prog. Mater. Sci.
2014, 60, 208ꢀ337.
(3) Yin, M.; Wu, C. K.; Lou, Y.; Burda, C.; Koberstein, J. T.; Zhu, Y.; O'Brien, S. J. Am. Chem. Soc.
2005, 127, 9506ꢀ9511.
(4) Hua, Q.; Cao, T.; Gu, X. K.; Lu, J.; Jiang, Z.; Pan, X.; Luo, L.; Li, W. X.; Huang, W. Angew.
Chem. Int. Ed. 2014, 53, 4856ꢀ4861.
(5) Liu, Y.; Shi, J.; Peng, Q.; Li, Y. Chem. Eur. J. 2013, 19, 4319ꢀ4326.
26
ACS Paragon Plus Environment

Page 27 of 29
ACS Catalysis
1
2
3
4
(6) Allen, S. E.; Walvoord, R. R.; PadillaꢀSalinas, R.; Kozlowski, M. C. Chem. Rev. 2013, 113
,
5
6
6234ꢀ6458.
7
8
9
(7) Wendlandt, A. E.; Suess, A. M.; Stahl, S. S. Angew. Chem. Int. Ed. 2011, 50, 11062ꢀ11087.
(8) Haack, P.; Limberg, C. Angew. Chem. Int. Ed. 2014, 53, 4282ꢀ4293.
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
(9) Zhang, G.; Yi, H.; Zhang, G.; Deng, Y.; Bai, R.; Zhang, H.; Miller, J. T.; Kropf, A. J.; Bunel, E.
E.; Lei, A. J. Am. Chem. Soc. 2014, 136, 924ꢀ926.
(10) OliverꢀMesseguer, J.; Liu, L.; GarciaꢀGarcia, S.; CanosꢀGimenez, C.; Dominguez, I.; Gavara,
R.; DomenechꢀCarbo, A.; Concepcion, P.; LeyvaꢀPerez, A.; Corma, A. J. Am. Chem. Soc. 2015,
137, 3894ꢀ3900.
(11) Chen, P.; Solomon, E. I. Proc. Natl. Acad. Sci. USA 2004, 101, 13105ꢀ13110.
(12) Solomon, E. I.; Chen, P.; Metz, M.; Lee, S.ꢀK.; Palmer, A. E. Angew. Chem. Int. Ed. 2001, 40
,
4570ꢀ4590.
(13) Fernández, E.; Boronat, M.; Corma, A. J. Phys. Chem. C 2015, 119, 19832ꢀ19846.
(14) VilarꢀVidal, N.; Rivas, J.; LópezꢀQuintela, M. A. ACS Catal. 2012, 2, 1693ꢀ1697.
(15) Zou, W.; Liu, L.; Zhang, L.; Li, L.; Cao, Y.; Wang, X.; Tang, C.; Gao, F.; Dong, L. Appl. Catal.
A: Gen. 2015, 505, 334ꢀ343.
(16) Bao, H.; Zhang, W.; Hua, Q.; Jiang, Z.; Yang, J.; Huang, W. Angew. Chem. Int. Ed. 2011, 50
,
12294ꢀ12298.
(17) Göbel, A.; Ruf, T.; Cardona, M.; Lin, C. T. Phys. B: Conden. Matter. 1996, 219-220, 511ꢀ513.
(18) Kliche, G.; Popovic, Z. V. Phys. Rew. B., 1990, 42, 10060ꢀ10066.
(19) Xu, J. F.; Ji, W.; Shen, Z. X.; Li, W. S.; Tang, S. H.; Ye, X. R.; Jia, D. Z.; Xin, X. Q. J. Raman
Spectrosc. 1999, 30, 413ꢀ415
(20) Strohmeier, B. R.; Leyden, D. E.; Field, R. S.; Hercules, D. M. J. Catal., 1985, 94, 514ꢀ530.
(21) VilarꢀVidal, N.; Blanco, M. C.; LópezꢀQuintela, M. A.; Rivas, J.; Serra, C. J. Phys. Chem. C
,
2010, 114, 15924ꢀ15930.
(22) Grünert, W.; Hayes, N.W.; Joyner, R.W.; Shpiro, E.S.; Rafiq, M.; Siddiqui, H.; Baeva, G. J.
Phys. Chem., 1994, 98, 10832ꢀ10846.
(23) Adimurthy, S., Malakar, C. C., Beifuss, U. J. Org. Chem., 2009, 74, 5648ꢀ5651.
(24) Oishi, T., Katayama, T., Yamaguchi, K., Mizuno, N. Chem. Eur. J., 2009, 15, 7539ꢀ7542.
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 29
1
2
3
4
(25) Chary, K. V.; Sagar, G. V.; Naresh, D.; Seela, K. K.; Sridhar, B. J. Phys. Chem. B 2005, 109
,
5
6
9437ꢀ9444.
7
8
(26) Fomina, L.; Vazquez, B.; Tkatchouk, E.; Fomine, S. Tetrahedron 2002, 58, 6741ꢀ6747.
9
(27) Sagadevan, A.; Ragupathi, A.; Lin, C.ꢀC.; Hwu, J. R.; Hwang, K. C. Green Chem. 2015, 17
,
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
1113ꢀ1119.
(28) Maity, P.; Takano, S.; Yamazoe, S.; Wakabayashi, T.; Tsukuda, T. J. Am. Chem. Soc. 2013, 135
,
9450ꢀ9457.
(29) Hadjiivanov, K. I.; Kantcheva, M. M.; Klissurski, D. G. J. Chem. Soc., Faraday Trans., 1996,
92, 4595ꢀ4600.
(30) Kannan, S.; Venkov, T.; Hadjiivanov, K.; Knözinger, H. Langmuir, 2004, 20, 730ꢀ736.
(31) Guzman, J.; Carrettin, S.; Corma, A. J. Am. Chem. Soc. 2005, 127, 3286ꢀ3287.
(32) Root, D. E.; MahroofꢀTahir, M.; Karlin, K. D.; Solomon, E. I. Inorg. Chem. 1998, 37
,
4838ꢀ4848.
(33) Metal-Oxo and Metal-Peroxo Species in Catalytic Oxidations; Meunier, B., Ed.; Springer:
Berlin Heidelberg, 2000; p 179ꢀ211.
(34) Palmer, M. S.; Neurock, M.; Olken, M. M. J. Am. Chem. Soc. 2002, 124, 8452ꢀ8461.
(35) Wu, Z.; Li, M.; Howe, J.; Meyer, H. M., III; Overbury, S. H. Langmuir 2010, 26, 16595ꢀ16605.
(36) Huang, M.; Fabris, S. Phys. Rev. B 2007, 75, 081404.
(37) Peng, H.; Xi, Y.; Ronaghi, N.; Dong, B.; Akhmedov, N. G.; Shi, X. J. Am. Chem. Soc. 2014,
136, 13174ꢀ13177.
(38) Yin, W.; He, C.; Chen, M.; Zhang, H.; Lei, A. Org. Lett. 2009, 11, 709ꢀ712.
(39) Siemsen, P.; Livingston, R. C.; Diederich, F. Angew. Chem. Int. Ed. 2000, 39, 2632ꢀ2657.
(40) Wang, D.; Li, J.; Li, N.; Gao, T.; Hou, S.; Chen, B. Green Chem. 2010, 12, 45ꢀ48.
28
ACS Paragon Plus Environment

Page 29 of 29
ACS Catalysis
1
2
3
4
5
TOC image
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
29
ACS Paragon Plus Environment