C-C Coupling on Single-Atom-Based Heterogeneous Catalyst

Article abstract of DOI:10.1021/jacs.7b09314

Compared to homogeneous catalysis, heterogeneous catalysis allows for ready separation of products from the catalyst and thus reuse of the catalyst. C-C coupling is typically performed on a molecular catalyst which is mixed with reactants in liquid phase

Full text of DOI:10.1021/jacs.7b09314

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Article
C-C coupling on single-atom based heterogeneous catalyst
Xiaoyan Zhang, Zaicheng Sun, Bin Wang, Yu Tang, Luan Nguyen, Yuting Li, and Franklin (Feng) Tao
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b09314 • Publication Date (Web): 20 Dec 2017
Downloaded from http://pubs.acs.org on December 20, 2017
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.
Journal of the American Chemical Society 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 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
C-C coupling on single-atom based heterogeneous catalyst
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Xiaoyan Zhang,abc Zaicheng Sun,b Bin Wang,d Yu Tang,a Luan Nguyen,a Yuting Li,a
Franklin (Feng) Taoa*
aDepartment of Chemical Engineering and Department of Chemistry, University of Kansas, Lawrence, KS,
66045, USA
bSchool of Chemistry, Beijing University of Technology, Beijing, 100080, China
cState Key Laboratory of Photocatalysis on Energy and Environment and College of Chemistry, Fuzhou
University, Fuzhou, 350116, China
dSchool of Chemical, Biological and Materials Engineering, Oklahoma University, Norman, OK 73019, USA
KEYWORDS: Catalysis• C-C coupling • Single atom catalysis • Palladium
ABSTRACT: Compared to homogeneous catalysis, heterogeneous catalysis allows for ready separation of products from its
catalyst and thus reuse of the catalyst. C-C coupling is typically performed on a molecular catalyst which is mixed with
reactants in liquid phase during catalysis. This homogenous mixing at a molecular level in the same phase makes separation
of the molecular catalyst extremely challenging and costly. Here we demonstrated that a TiO2-based nanoparticle catalyst
anchored singly dispersed Pd atoms (Pd1/TiO2) is selective and highly active for more than ten Sonograshira C-C coupling
reactions (R-≡H+R′-X→R-≡-R′; X=Br, I; R′=Aryl or Vinyl). The coupling between iodobenzene and phenylacetylene on
Pd1/TiO2 exhibits a turn-over rate of 51.0 diphenylacetylene molecules per anchored Pd atom per minute at 60 oC with a low
apparent activation barrier of 28.9 kJ/mol and no cost of catalyst separation. DFT calculations suggest that the single Pd
atom bonded to surface lattice oxygen atoms of TiO2 acts as a site to dissociatively chemisorb iodobenzene to generate an
intermediate, phenyl which then couples with phenylacetylenyl bound to a surface oxygen atom. This coupling of phenyl
adsorbed on Pd1 and phenylacetylenyl bound to Oad of TiO2 forms the product molecule, dipenylacetylene.
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
INTRODUCTION
for different C-C couplings3-7, 9-14, 16-19, 21, 23-24 in contrast to
these conventional Pd-based molecular catalysts.
Mechanistic understanding of C-C couplings on these
heterogeneous catalysts, mainly Pd NPs protected with
surfactant such as polyvinylpyrrolidone (PVP) or
supported on metal oxide NPs is still an on-going topic.
Carbon-carbon couplings including Suzuki, Heck,
Sonograshira, Suzuki-Miyaura, and Stille C-C couplings
have been considered as the cornerstones of organic
synthesis.1-22 For example, Suzuki-Miyaura cross-coupling
of aryl halides with arylboronic acids has been the key step
in production of arylalkynes and conjugated enynes which
are important precursors for chemical production of
2
15
industries.
complex-based
Typically, these C-C couplings use Pd(II)
homogeneous catalysts such as
Pd(PPh3)2Cl2 (Figure 1a), called molecular catalysts.
However, one obvious limit of using molecular catalysts is
the technical challenge and quite high cost in separating a
used molecular catalyst from the left reactants and formed
products before reusing the catalyst or recycling the
precious metal. Pd nanoparticles (NPs), a heterogeneous
catalyst (Figure 1b) have been demonstrated to be active
Figure 1. Schematic showing the structural models of Pd-
based molecular catalyst (a) Pd(PPh3)2Cl2, (b) Pd nanoparticle
catalyst, and (c) Pd single atom catalyst (Pd1/TiO2).
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Whether these metal nanoparticles are active phases of C-
Page 2 of 11
1
2
3
4
5
6
7
8
C coupling or not is still being debated.
General procedure for performing Sonogashira C-C
As schematically shown in Figure 1b, metal atoms are
continuously packed on surface of a metal nanoparticle.
Atoms on a metal nanoparticle typically exhibits quite
different coordination environments with their
neighboring atoms. For instance, metal atoms at corners
and edges have definitely different coordination numbers
(CN) compared to those on facets. 25-27 On the other hand,
the fraction of atoms with a specific coordination number
such as atoms at corners with CN (M-M)=7 strongly
depends on size and shape of metal nanoparticles. Size-
and shape-dependent catalytic performances on metal
nanoparticles have been widely reported in literatures.25-27
Compared to the ubiquitous coexistence of multiple metal
atoms with different coordination and chemical
environments on surface of a metal nanoparticle (Figure
1b), a single atom of metal such as a Pd atom anchored on
an nonmetallic substrate such as TiO2 (Figure 1c) are
expected to have a minimized number of choices of
binding sites for reactants or intermediates. Thus, a
catalyst with these singly dispersed metal atoms is
expected to exhibit selectivity higher than a catalyst
surface with multiple types of reactive sites such as Pd
nanoparticles supported on TiO2. More importantly, by
anchoring precious metal atoms on a solid support, the
catalyst can be readily separated. Such a catalyst is
expected to exhibit high activity and selectivity and can be
readily separated for reuse. Here we chosen the
Sonogashira C-C cross-coupling of phenylacetylene and
iodobenzene as a probe reaction and tested its activity and
selectivity on Pd atoms singly anchored on TiO2 toward
both efficient catalysis and ready separation for high
reusability of precious metals.
coupling on catalysts. Under nitrogen atmosphere, an
oven-dried round-bottomed flask was charged with 0.20
wt% Pd/TiO2 (50mg), K2CO3 (2.00 mmol), an aryl halide
such as iodobenzene (1.00 mmol), a terminal alkyne such
as phenylacetylene (1.00 mmol), 0.30 mmol dodecane and
certain amount of ethanol, making the total volume of the
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
EXPERIMENTAL
Synthesis of catalyst. Pd(NO3)2·2H2O (Aldrich) was
used as the source of Pd cations for the preparation of
singly dispersed Pd/TiO2 catalyst. Anatase TiO2 (99.7%,
Aldrich) was used as the support. To anchor Pd cations.
Catalyst precursors were synthesized through
a
deposition-precipitation method. In a typical preparation,
500 mg TiO2 was mixed with 100 mL deionized water
through a vigorous stirring to form a white suspension. 4.0
mL Pd(NO3)2·2H2O aqueous solution (contain 2.6 mg
Pd(NO3)2·2H2O) was introduced to the TiO2 suspension
through controlled injection with a syringe pump while the
solution was vigorously stirred. The new suspension was
continuously stirred for 2-3 hrs to allow sufficient natural
adsorption of metal ions. Then, the pH value of the mixture
was carefully adjusted to 9.5 by gradually introducing
ammonium hydroxide solution, followed by vigorously
stirring for another 5 hrs in order to having a complete
equilibrium in solution. Then the slurry was centrifuged
Figure 2. X-ray absorption fine structure of Pd K-edge and
TEM studies of used 0.20 wt% Pd/TiO2. (a) normalized
absorption coefficient as the function of photon energy, (b)
near edge absorption fine structure, (c) Fourier Transformed
r-space data of the catalyst compared with Pd metal foil and
PdO as references and (d) analysis and fitting of the r-space
data of the 0.20 wt% Pd/TiO2 catalyst, (e) fitting perameters
of 0.20 wt% Pd/TiO2 catalyst. The Pd foil and PdO reference
are rescaled by the factors of 0.4 and 0.85 respectively for
comparison in (c). (f) Large scale TEM image of catalyst
particles .(g) High-resolution TEM of catlayst particles.
o
and dried the solution at 60 C in oven overnight; the
reaction mixture to 10.0 ml. All other substrates and
corresponding catalytic performance can be found in
Supporting Information. The catalysis was performed by
heating the reaction mixture to 60 oC and then
immediately stirring the solution vigorously at 60 oC for 3
o
catalyst precursor was calcinated at 400 C in air for 4h.
The catalyst was termed as 0.20 wt% Pd/TiO2. The actual
loading of Pd is 0.056 wt% based on inductively coupled
plasma-atomic emission spectrometry (ICP-AES).
2
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
hrs under the protection of nitrogen atmosphere. After fast
The reaction rate, r for C-C coupling on either 5.0 mg
0.20 wt% Pd/ TiO2, or 0.50 mg Pd(PPh3)2Cl2 was calculated
1
2
3
4
5
6
7
8
cooling the refluxed solution to the room temperature with
ice bath, the solution was centrifuged and filtered before
GC-MS measurements of the amount of the format
products and the amount of left reactants.
with
equation,
ꢀ =
ꢁꢂꢃꢄꢅꢆꢇꢂꢈꢉꢊ ꢋꢇꢊꢇꢌꢉꢄꢍꢉꢄꢉ (ꢈꢇꢎ)
when
the
(
)
ꢏꢊꢐꢌꢃꢅꢉꢊ ꢂꢉꢃꢑꢁꢒꢓꢉ ꢅꢒꢁꢉꢅ ꢈꢇꢎ ×ꢂꢉꢃꢑꢁꢒꢇꢄ ꢁꢒꢈꢉ ꢒꢄ ꢈꢒꢄ
catalysis was under kinetics controlled regime. The unit of
r here is molecules of product per Pd site per min. In the
kinetics studies, 2.00 mmol iodobenzene and 2.00 mol
phenylacetylene were used as reactants; 5.0 mg 0.20 wt%
Pd/TiO2 or 1.0 mg Pd(PPH3)2Cl2 (99%, Aldrich) was used as
catalysts. The number of Pd-based sites of 5.0 mg 0.20 wt%
Pd/TiO2 was calculated to be about 1×10-7 mol. The number
of Pd-based sites of 1.0 mg Pd(PPh3)2Cl2 is 1.4×10-6 mol. For
Measurements of formed products and left
reactants. GC-MS is the main experimental technique to
quantify the amounts of the left reactant and the formed
products. Here the reaction system of C-C coupling
between phenylacetylene and iodobenzene to form
diphenylacetylene is used as one example to illustrate the
method to quantify the reactants and products. The ideal
product of this C-C coupling is diphenylacetylene. To
accurately measure the amount of formed product and
unreacted reactants for calculating conversion and
selectivity of the C-C coupling, standard curves of the two
reactants (iodobenzene and phenylacetylene) and the
ideal product (dipenylacetylene) were established through
preparing five solutions with different concentration for
each of them. The relative intensity of GC peak of a
reactant or a product to the added reference (0.30 mmol
dodecane) was taken as Y-axis value in Figure S2. The
corresponding amount of the chemical (iodobenzene,
phenylacetylene, or dipenylacetylene) in unit of mmol was
marked as the X-axis. By plotting this relative intensity in
terms of area ratio of a chemical to 0.30 mmol dodecane as
a function of the known amount of the chemical such as
phenylacetylene, a linear regression was readily achieved
as shown in Figure S2a. With the same method, the
standard curves of iodobenzene and diphenylacetylene
were established as shown in Figures S2b and S2c,
respectively. They were used as standard curves in the
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
o
instance, the reaction rate on 0.20 wt% Pd/TiO2 at 60 C
under kinetics controlled regime is 51.0 diphenylacetylene
molecules produced from a Pd atom-based site per minute.
It is 32.1 diphenylacetylene molecules produced from a
o
Pd(PPh3)2Cl2 molecule per minute at 60 C under kinetics
controlled regime.
Characterization. Powder X-ray diffraction patterns of
the samples were measured on a Bruker SMART APEX
single-crystal diffractometer. The size, shape and lattice
fringe of the catalyst were identified using high-resolution
JEM-2100F Transmission Electron Microscope operated at
an accelerating voltage of 200 KV. X-ray photoelectron
spectroscopy was conducted by using a lab-based ambient-
pressure X-ray photoelectron spectroscopy system (AP-
XPS) designed and built in Tao group. Amount of reactants
or/and product in solutions before or after catalysis were
analyzed with a GC-MS system containing GC (Quattro
Micro) and mass spectrometer (LCT Premier) Samples are
introduced into the Quattro Micro GC via the Agilent
6890N gas chromatograph equipped with an auto sampler;
the LCT is equipped with a lock mass and is set to a
resolution high enough to perform exact mass experiments
when ions of the proper intensity are selected.
calculations
of
conversions
of
iodobenzene,
phenylacetylene and selectivity for production of
diphenylacetylene.
Kinetic studies for C-C coupling on heterogeneous
catalyst 0.20 wt% Pd/TiO2 and homogeneous catalyst
Pd(PPh3)2Cl2. Under nitrogen atmosphere, an oven-dried
round-bottomed flask was charged with 0.20 wt% Pd/ TiO2
(5.0 mg), K2CO3 (2.0 mmol), iodobenzene (2.00 mmol),
phenylacetylene (2.00 mmol), 0.30 mmol dodecane and
ethanol, total volume of the solution is 10.0 ml. The
reaction mixture was refluxed for one hour at different
temperature (55 -75 oC). After cooling to the room
temperature, the solution as centrifuged and filtered for
GC-MS measurement.
DFT calculation and optimization of TiO2(101)
surface. Density functional calculations were carried out
using the VASP package28. The Perdew-Burke-Ernzerh of
generalized gradient approximation exchange-correlation
potential (PBE-GGA)29 was used, and the electron-core
interactions were treated in the projector augmented wave
(PAW) method30-31. Structures were optimized until the
atomic forces were smaller than 0.02 eV Å-1 with a kinetic
cut off energy of 300 eV. Van der Waals interactions were
taken into account by incorporating the DFT-D3 semi-
empirical method32. Reaction barriers were determined
with the Nudged Elastic Band method33-34. The optimized
lattice constants of the anatase TiO2 unit cell is 3.814 Å ×
3.814 Å × 9.628 Å, which is in good agreement with
experimental results35. We modeled the TiO2 (101) surface
using a slab with four TiO2 layers and a rectangular surface
cell with dimensions of 20.7 Å × 19.1 Å. The vacuum layer
is at least 16 Å in all the calculations to eliminate
interactions between adjacent cells. The bottom 3 layers
were fixed at their bulk positions while the top TiO2 layer
is fully relaxed.
Similarly, an oven-dried round-bottomed flask was
charged with Pd(PPh3)2Cl2 (0.50 mg), 99μl piperidine,
iodobenzene (2.00 mmol), phenylacetylene (2.00 mmol),
0.30 mmol dodecane and acetonitrile, total volume of the
solution is 10.0 mL. The reaction mixture was refluxed at a
specific temperature under nitrogen atmosphere for 10
min. The kinetics of C-C coupling on Pd(PPh3)2Cl2 was
done in the temperature of 45-65 oC After cooling to room
temperature, the solution was centrifuged and filtered for
GC-MS measurement.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
RESULTS AND DISSCUSSIONS
Pd-O bond. Thus, these Pd atoms on the used 0.20 wt%
1
2
3
4
5
6
7
8
Formation single-atom catalytic site Pd1O4 on TiO2.
The catalyst 0.20 wt% Pd/TiO2 was synthesized through a
well-controlled deposition-precipitation method with a
following drying and calcination. The coordination
environment of Pd atoms anchored on TiO2 of a used 0.20
wt% Pd/TiO2 was studied with X-ray absorption near edge
structure (XANES) and extended X-ray absorption fine
structure (EXAFS). Figures 2a and 2b present the energy
space of Pd K-edge of both metallic Pd foil taken as a
reference and the used 0.20 wt% Pd/TiO2 after catalysis.
Clearly, the edge position of Pd K-edge of the used 0.20
wt% Pd/TiO2 in Figures 2a and 2b is different from metallic
Pd foil. Compared to the Pd foil, the K-edge of Pd atoms in
our catalyst has up-shift. It shows that the Pd atoms of our
catalyst is at oxidizing state instead of the metallic state of
Pd atoms of the Pd foil. Figure 2c presents r-space of Pd-K
5.26
1.97
1.16
2.87
2.87
1.97
2
4
6
8
10
2
4
6
8
10
Time (min)
Time (min)
(a)
(b)
9
4.5
4.2
3.9
3.6
3.3
3.0
2.7
2.4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
178.0223
r
151.9552
150
PdCl2(PPh3)2
0.2wt% Pd/TiO2
2.85 2.90 2.95 3.00 3.05 3.10 3.15
1000/T
30
60
90
120
180
m/z
(d)
(c)
Figure 4. Catalytic performance of C-C coupling on Pd1/TiO2
and Pd(PPh3)2Cl2. (a) GC spectra of reactants
(phenylacetylene and Iodobenzene) and internal standard
(dodecane) before reaction. (b) GC spectra of solution after
catalysis on 50 mg Pd1/TiO2, containing left reactants
(phenylacetylene and Iodobenzene), formed product
(diphenylacetylene), and internal standard (dodecane);
retention time for phenylacetylene, iodobenzene, dodecane,
and diphenylacetylene is 1.16 min, 1.97 min, 2.87 min, and
5.26 min, respectively. (c) Mass spectrum of the product,
diphenylacetylene separated by GC at 5.26 min. (d)
Arrhenius plots of heterogeneous catalyst Pd1/TiO2 (Ea=28.9
KJ/mol) and homogeneous catalyst Pd(PPh3)2Cl2 (Ea=51.7
KJ/mol); r is the reaction rate of phenylacetylene (molecule
per Pd atom per min).
Pd/TiO2 bonded with oxygen atoms of TiO2. As shown in
Figure 2c, PhO nanoparticles have very obvious second
shell of Pd atoms in terms of Pd-O-Pd structure at 3.0 Å on
r-space spectrum; the lack of this peak in the r-space
spectrum of Pd K-edge of the used 0.20 wt% Pd/TiO2 shows
that our catalyst does not have Pd-O-Pd species. The lack
of peaks of both Pd-Pd and Pd-O-Pd in r-space spectrum
of the used 0.20 wt% Pd/TiO2 shows Pd atoms of the used
0.20 wt% Pd/TiO2 are obviously singly dispersed on surface
of TiO2 through bonding with oxygen atoms of TiO2.
Fitting the peak of the first shell of Pd-O (black line in
Figure 2c) gives the average coordination number of
oxygen atoms around a Pd atom, CN (Pd-O) =3.63. It shows
that on average each Pd atom bonds with about four
oxygen atoms of the TiO2 surface. The DFT calculations to
be discussed later suggested that such a coordination in
fact gives a stable structure as shown in Figure 3c. Pd1/TiO2
will be used to denote 0.20 wt% Pd/TiO2 in some cases
since we confirmed that Pd atoms are singly dispersed on
Figure 3. DFT calculations of the structures of the
supported Pd1 catalyst on anatase TiO2 (101) surface and the
reaction profile for the C-C coupling reaction. (a-e) five
tested structures with varied coordination number and
averaged Pd-O distance. (f) Reaction pathway proposed by
DFT calculations and the corresponding energy profile. The
Ti, O, Pd, C, H, I atoms are colored blue, red, blue, black,
green, and pink, respectively.
edge of 0.20 wt% Pd/TiO2 (black line), Pd foil (red line),
and PdO nanocrystals (blue line). Obviously, the peak
position of Pd K-edge of 0.20 wt% Pd/TiO2 at ~1.45 Å in r-
space is very different from the the Pd foil, showing the
lack of metallic Pd nanoclusters in the used 0.20 wt%
Pd/TiO2. It does overlap the peak of the first shell of PdO,
suggesting that Pd atoms of the used 0.20 wt% Pd/TiO2 and
PdO reference sample have a similar first coordination
shell. The first shell of Pd K-edge of PdO was assigned to
TiO2 through four Pd-O bonds.
Morphology of the
catalyst particles of Pd1/TiO2 were examined with TEM
(Figure 2f). The average size of these nanoparticles is 20-25
nm. Lattice fringe of TiO2 nanocrystals was measured. The
4
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
preferentially exposed surface of TiO2 nanoparticles is (101) GC-MS was used to separate different reactants in the
1
2
3
4
5
6
7
8
as shown in a representative image in Figure 2g.
solution before catalysis such as Figure 4a and separate
reactants and products of a solution after catalysis such as
Figure 4b. In GC of a solution before catalysis, the peaks at
retention time at 1.16, 1.97, and 2.87 minutes are
contributed from phenylacetylene, iodobenzene, and
dodecane, respectively. In the GC of solution after catalysis
on Pd1/TiO2, the peak intensity of phenylacetylene at 1.16
minutes attributed to phenylacetylene was below the
detection limit, suggesting that most phenylacetylene was
reacted. A new peak observed at 5.26 minutes is attributed
to the product of diphenylacetylene. The recognition of a
specific chemical (a reactant or product) in GC
chromatography was clearly confirmed by its mass spectra
in Figure 5. After reaction, the ratio of peak area of a
reactant or product to the reference dodecane was
calculated. For instance, with this ratio the amount of yield
of diphenylacetylene after catalysis was readily calculated
from Figure 5c.
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 5. Mass Spectrum of (a) phenylacetylene, (b)
iodobenzene, (c) dodecane, (d) diphenylacetylene
separated by GC column as exampled in Figures 4a and
4b.
The Sonograshira C-C coupling catalysis was performed
o
at 60 C for three hours. As shown in GC of the solution
after catalysis (Figure 4b), no phenylacetylene was found.
It shows phenylacetylene was completely transformed to
products through C-C coupling on 50 mg 0.20 wt%
C-C coupling on Pd1 single atom sites anchored on
Figure 6. Control experiment for elucidating whether Pd atoms detached from surface of TiO2 nanoparticles. (a) Schematic
showing Pd1/TiO2 catalyst. (b) Schematic showing coexistence of Pd1/TiO2 and the detached Pd atoms during catalysis. (c)
Schematic showing coexistence of Pd1/TiO2 nanoparticles and Pd nanocluster (after aggregation of Pd atoms) during catalysis.
(d) Schematic showing coexistence of Pd1/TiO2 nanoparticles and PdO nanocluster formed drying solution containing calcinating
Pd1/TiO2 nanoparticles and Pd nanocluster and then calcination in air. (e) XPS spectra of the Pd 3d of 0.20 wt% Pd/TiO2 before
catalysis. (f) Ti 2p of 0.2 wt% Pd/TiO2 before catalysis. (g) XPS spectra of the Pd 3d of 0.20 wt% Pd/TiO2 after catalysis, dissolution
of base, drying and calcination. (h) XPS spectra of the Ti 2p of 0.20 wt% Pd/TiO2 after catalysis, dissolution of base, drying and
calcination. (i) Area ratio of peak of Pd 3d to peak of Ti of 0.20 wt% Pd/TiO2 before catalysis (marked as “fresh” in the figure) and
of 0.20 wt% Pd/TiO2 after catalysis, dissolution of base, drying and calcination (marked as “after reaction” in the figure).
o
TiO2. The catalyst was used for C-C coupling in a liquid
phase which consists of 1.00 mmol iodobenzene and 1.00
mmol phenylacetylene as reactants, 2.0 mmol K2CO3 as a
base, 0.30 mmol of dodecane, and certain volume of
solvent (ethanol, dimethyl sulfoxide or benzene) which
makes the total volume of the solution to 10.0 ml before
Pd/TiO2 and the conversion at 60 C reached 99% under
this catalytic condition. By calculating the area ratio of
diphenylacetylene to dodecane and referring it to the
standard curve of diphenylacetylene (Figure S2c), the yield
of diphenylacetylene is >90%. Thus, the selectivity for
production of diphenylacetylene is ≥91%. The reusability
of the catalyst was tested by performing catalysis on the
regenerated catalyst under a condition exactly same as the
first cycle. The regeneration of the catalyst was done by
drying the solution (after catalysis) in oven and then
catalysis.
Dodecane is used as a reference for
measurements of the concentrations of reactants and
products with GC-MS. The details of measurements of
product and reactants were described in experimental
section.
o
calcinating the powder in air at 500 C for 4 hrs. The
conversion of phenylacetylene and the selectivity for
production of diphenylacetylene on the recycled catalysts
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
of the second and third cycles nearly remained little could detach from Pt1/TiO2 during C-C coupling. To clarify
1
2
3
4
5
6
7
8
change as shown in Table S1. Thus, this catalyst exhibits
the high activity, selectivity and reusability of 0.20 wt%
Pd/TiO2 for Sonograshria C-C coupling.
it, we performed a few series of control experiments to
check whether Pd nanoparticles formed or not.
A simple way to identify whether Pd atoms anchored on
TiO2 have been detached from surface of TiO2 during C-C
coupling, is to measure the concentrations of Ti and Pd in
the liquid solution after precipitating the used catalyst
nanoparticles Pt1/TiO2 by centrifugation. Here we call it
the simple method. Through centrifuging the solution after
catalysis, K2CO3 and the used catalyst were readily
precipitated. Concentrations of Ti and Pd of the obtained
solution after precipitation were measured with ICP-AES.
Small portions of Pd and Ti atoms were found in the
solution; the observation of Ti in the solution in fact results
from the small TiO2 nanoparticles of catalyst which could
not be precipitated in the current centrifugation. About
7% of all Pd atoms of 50 mg 0.20 wt% Pd/TiO2 catalyst
remained in the solution after precipitation. Very likely,
these Pd atoms detected by ICP after centrifugation are
these Pd atoms anchored on these small catalyst
nanoparticles (0.20 wt% Pd/TiO2) whose sizes are too
small to be centrifuged.
We performed that C-C coupling experiments on this
catalyst by using seven arkyl halides and three different
terminal alkanes. The conversion and selectivity for C-C
couplings of different substrates were listed in Table S2. As
shown in Table S2, the catalyst Pd1/TiO2 exhibits high
activity and selectivity for C-C couplings between Aryl
halide (X=I and Br) and different terminal 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
We tested the chlorobenzene with pheylacetlyene and
found that the activity for forming diphenylacetylene is
low although couplings between terminal alkynes and
bromobenzene and between terminal alkynes and
iodobenzene on the Pd1/TiO2 catalyst are very active. The
reason for the low catalytic activity of chlorobenzene is
that C-Cl of chlorobenzen bond is stronger than C-Br of
bromobenzene and C-I of iodobenzene. Thus, breaking C-
Cl bond of chlorobenzene needs a higher temperature,
leading to a low activity between chlorobenzene and
terminal alkynes at 60-100oC. Notably, similar to the
observed high activity for bromobenzene and iodobenzene
and quite low activity for chlorobenzene, the quite low
activity in C-C coupling between chlorobenzene and
phenylacetylene on molecular catalysts was reported by
Mery et al,36 with the reported molecular catalyst, only 4%
chlorobenzene can be converted although 100%
iodobenzene can be converted under the same catalytic
condition. A similar report on high activity on aryl halide
(X=I and Br) and very low activity on aryl chloride (X-Cl)
was published by Leadbeater et al37. Thus, the high activity
of Pd1/TiO2 to these C-C coupling reactions between aryl
halides (X=I and Br) and terminal alkynes clearly can
justify that the Pd1/TiO2 is highly active for Sonogashira C-
C coupling.
As mentioned above, if Pd cations could have detached
from TiO2 during catalysis, they should have been
immediately reduced by the solvent such as ethanol to
form Pd nanoparticles in the solution. This is because Pd2+
in solution can be immediately reduced by methanol at 25
oC to form Pd nanoparticles based on reference.41 To test
the reducibility of ethanol to Pd2+, 1.0 μmol of Pd(NO3)2
(mass of Pd, about 1.0×10-4 gram) which is about 3.5 times
of Pd cations of 50.0 mg of 0.20 wt% Pd/TiO2 (about 2.8×10-
5 gram of Pd) and the same amount of base 2.0 mmol K2CO3
were added to 9.8 ml ethanol at 25 oC, being a test solution.
As shown in Figure S3a1, the solution was immediately
o
turned to gray at 25 C. It suggested the high reducibility
of ethanol to Pd2+ at 25 oC. Upon the solution of Pd(NO3)2,
o
ethanol and K2CO3 was heated to 60 C and remained at
Catalyst with lower loading, 0.10 wt% Pd/TiO2 was
prepared with the same method as 0.20 wt% Pd/TiO2. It
shows low conversion (55%) but the same selectivity for C-
C coupling. It suggests the sites on 0.10 wt% Pd/TiO2 is very
similar to 0.20 wt% Pd/TiO2. In addition, 2.0 wt% Pd/TiO2
was prepared as well. Catalysis on 2.0 wt% Pd/TiO2 was
performed under the same condition as 0.10 wt% Pd/TiO2
and 0.20 wt% Pd/TiO2. 50.0 mg catalyst of 2.0 wt% Pd/TiO2
exhibits nearly 100% conversion but low selectivity of <75%
due to the formation of Pd nanoparticles supported on
TiO2 nanoparticles.
this temperature, the color became dark. Figures S3a2-
S3a4 show sequential change of color of the Pd(NO3)2
solution at 60 oC as a function of time. As reduction of Pd2+
of the Pd(NO3)2 solution to Pd atoms to form Pd
nanoparticles is the only reduction reaction being
performed in this test solution, the evolution of gray to
dark of the test solution results from the formation of more
Pd nanoparticles. The formation of Pd nanoparticles
through reduction of Pd2+ by ethanol at room temperature
were confirmed by TEM studies. An experiment using 0.28
μmol of Pd(NO3)2 (mass of Pd, about 2.8×10-5 gram) was
performed under the same condition as the above
experiment using 1.0 μmol of Pd(NO3)2; Similar to Figure
S3a1, darkness of the solution at 25 oC was observed.
Preservation of anchored Pd1-based sites on TiO2
during catalysis. One concern is that the activity of
Pd1/TiO2 could have resulted from Pd nanoparticles if Pd
cations could potentially detach from Pd1/TiO2 catalyst
nanoparticles and then could have been reduced by
ethanol to form Pd nanoparticles. It is reported that Pd
nanoparticles are active for the C-C coupling although
whether Pd nanoparticle is the active phase for C-C
coupling or not has been debated.38-40 To explore the origin
of the catalytic activity of Pd1/TiO2 catalyst, a key question
is whether the Pd atoms of Pd1O4 sites anchored on TiO2
For comparison, the evolution of color of a typical
catalysis solution was studied. The catalysis solution
consisted of 50.0 mg catalyst nanoparticles 0.20 wt%
Pd/TiO2, 2.0 mmol K2CO3, 1.00 mmol phenylacetylene, 1.00
mmol iodobenzene, 0.30 mmol dodecane, and about 9.73
ml ethanol for making the total volume to 10.0 ml. Photos
of the catalysis solution were taken at 0th hour at 25 oC, 1st
hour at 60 oC, 2nd hour at 60 oC, and 3rd hour at 60 oC during
o
catalysis (Figures S3b1-S3b4). At 25 C, it was colorless
6
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
(Figure S3b1). Along the catalysis at 60oC, it turned to light
the catalytic cycle, it is reductive removal and Pd (II) is
yellow at 1st hour (Figure S3b2) and then orange at 2nd hour
(Figure S3b3) and remained orange at 3rd hour (Figure
S3b4). The evolution of color of the catalysis solution in
Figures S3b1-S3b4 is distinctly different from the evolution
of color of test solution consisting of 1.00 μmol Pd(NO3)2,
3.0 mmol K2CO3, and 10 ml ethanol in Figures S3a1-S3a4.
TEM studies of the solution after C-C coupling catalysis of
3 hrs (Figure 3b4) showed that no Pd nanoparticles were
formed in the solution during C-C coupling, consistent
with the lack of dark color. This distinct difference of the
two series of experiments in Figure S3a and S3b clearly
suggest that the detected Pd in the solution after
centrifuging the solution obtained after catalysis in the
simple method was not in the format of free standing Pd
cations since free standing Pd cations can be reduced to
thus changed to Pd (0). As each catalytic cycle is performed
fast, XPS cannot track the change of oxidation state of Pd
atoms in a catalytic cycle. In the case of C-C coupling
performed on surface of Pd1/TiO2, the formation of
intermediate state (see structure III in Figure 3) can be
considered as a step similar to oxidative addition of
homogeneous catalysis; the bonding of a C atom of phenyl
group to a Pd atom anchored on Pd1O4/TiO2 should vary
the electronic density; the desorption of diphenyl
acetylene makes Pd atom of Pd1O4 restore to its original
electronic state on TiO2 and be ready for next catalytic
cycle. These decrease of electron density of Pd atoms by
bonding to carbon atom of phenyl and the increase of
electron density by desorption of product molecule are
performed in one catalytic cycle. As these elementary steps
are performed fast; XPS cannot catch the change of
oxidation state of Pd atom. Within one catalytic cycle, the
binding of one C atom of a phenyl group to Pd atom
changes the electronic state of Pd atoms; however, this
change is not significant since Pd atom of Pd1O4 always
bonds with four oxygen atoms of surface of TiO2
nanoparticle. Photoelectron spectroscopy may not be able
to identify this subtle change even if it could perform
ultrafast scan at the timescale of finishing a catalytic cycle.
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
o
metal Pd atoms immediately even at 25 C. Thus, the
detected Pd atoms in ICP were the Pd atoms attached to
small TiO2 nanoparticles which cannot be centrifuged due
to their small size.
Other than those control experiments, the following
experiment was performed. 50 mg of 0.20 wt% Pd/TiO2
was added to 10 ml ethanol; then was heat to 60oC and then
o
remained at 60 C under reflux condition; picture of the
reaction system was taken at end of 1st, 2nd, and 3rd hour by
quickly elevating the flask from oil heating bath to take a
picture and then immediately putting it back to the 60 oC
oil bath. As shown in Figure S3d1-S3d4, there is no change
of the color of the solution. It is quite different from the
obvious color change of 0.26 mg Pd(NO3)2·2H2O in 10 ml
ethanol (Figure S3a1-a4) and 0.026 mg Pd(NO3)2·2H2O in 10
ml ethanol (Figure S3c1-c4). Thus, the lack of change of
color of the solutions in the control experiment of Figure
S3d1-d4 shows that the anchored Pd atoms on TiO2 in
ethanol of 3 hrs at 60 oC do not detach.
Another set of control experiments was performed to
check whether the anchored Pd atoms on TiO2 are active
o
sites for C-C coupling at 60 C. Different from all above
series of catalysis reactions using ethanol as solvent,
benzene instead of ethanol was used as a solvent in this
series of control experiments. It is well known that
benzene cannot reduce Pd cations to metallic Pd atoms to
form Pd nanoparticles. Thus, the activity of 0.20 wt%
Pd/TiO2 in benzene must contribute from the anchored Pd
atoms of 0.20 wt% Pd/TiO2 if product molecules
diphenylacetylene could be formed when benzene is used
the solvent. The detailed description of this control
experiment was given in Section 2 in Supporting
Information. This control experiment showed that 50.0 mg
0.20 wt% Pd/TiO2 is active for production of
diphenylacetylene from C-C coupling of iodobenzene (1.0
mmol), phenylacetylene (1.0 mmol) at 60oC when benzene
was used as the solvent. As benzene does not reduce Pd
cations of 0.20 wt% Pd/TiO2 to metallic Pd, the production
of diphenylacetylene confirmed that the singly dispersed
Pd atoms anchored on TiO2 are intrinsically active for C-C
coupling of iodobenzene and phenylacetylene.
The preservation of anchored Pd atoms on 0.20 wt%
Pd/TiO2 nanoparticles left in solution in the simple method
was further confirmed by XPS studies for measuring Pd/Ti
atomic ratio of surfaces of fresh catalyst and used catalyst
(Figures 6e and 6g). The 0.20 wt% Pd/TiO2 nanoparticles
left in solution was prepared by drying a mixture of ten
experiments of the simple method through vaporizing
water and solvent at 120 oC for 5 hrs. After drying, the left
powder in beaker was collected and used as the sample of
Figure 6g. XPS studies show Pd/Ti ratio of surface of the
used catalyst remained the same as the fresh catalyst. The
detailed description of this series of control experiments
was provided in Section 1 of Supporting Information.
Although the low signal to noise ratio of Pd 3d of the
collected 0.20 wt% Pd/TiO2 nanoparticles after drying
(Figure 6g) makes the spectrum of Pd 3d in Figure 6g
appear different from that before catalysis (Figure 6e),
peak positions of Pd 3d5/2 in Figures 6e and 6g are the
same. Thus, Pd atoms on TiO2 before and after catalysis
have very similar chemical state.
Two hot filtration studies were performed. One was for
measurement of the concertation of Pd atoms left in
solution. In this study, C-C coupling between
phenylacetylene and iodobenzene was performed at 60 oC
for 1 hour and then immediately transferred the hot liquid
to syringe filter to separate the reactant solution from the
catalyst particles when the reaction system (the solution
and the catalyst particles) was still hot. The Pd
concentration of the filtered liquid was analyzed by ICP-
AES. Upon removal of catalyst particles, the concentration
of Pd in 10 ml solution (10.0 g) is 0.020 ppm; thus, the total
amount of Pd in the solution is 2.0×10-7 gram. Compared to
the simple method discussed early, much less Pd atoms
XPS cannot track the change of oxidation state of Pd
atom in a catalytic cycle. In the case of homogeneous
catalysis, the oxidation state does change from (0) to (II)
when oxidative addition is performed in the first
elementary step of a catalytic cycle; in the second step of
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
were left in the solution due to the use a fine filter paper in optimization with DFT shows that the average Pd-O
1
2
3
4
5
6
7
8
the syringe filtration. As mentioned in the preparation of
catalysts of the experimental section, the actual
concentration of 50 mg 0.20 wt% Pd/TiO2 is 0.056 wt%
Pd/TiO2; thus, the total Pd in 50 mg 0.20 wt% Pd/TiO2 is
2.8×10-5 gram. After reaction of one hour, 0.71% of all Pd
atoms anchored on TiO2 was detached. Similar studies
were performed after catalysis of 3 hrs; the total amount of
Pd in the solution is 0.57×10-6 gram after three hour
reaction; 2.02% of all Pd atoms anchored on catalyst was
detached.
distance in structure of Figure 3c is 2.06 Å, in good
agreement with the experimental values, 2.00 Å measured
with EXAFS (Figure 2e). As shown in Table 1 in Supporting
Information, formation of this structure (Figure 3c) is
highly energetically favorable with an energy gain of 128
kJ/mol. Although structure of Figure 3d has a Pd atom
coordinating with four oxygen atoms (Pd1O4), it has
definitely higher energy than Figure 3c. Thus, Figure 3d is
not a favorable structure. As the coordination number of
O atoms to a Pd atom and the optimized distance between
Pd and O in the structure of Figure 3c are quite consistent
with the measured values of catalyst 0.20 wt% Pd/TiO2
(Figure 2f), we take the structure in Figure 3c as active site
of Pd1/TiO2 in the following simulation of reaction pathway
of C-C coupling reaction on Pd1/TiO2.
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
Another hot filtration for checking whether Pd atoms
detached from surface of TiO2 or not was performed. After
the hot filtration, the hot solution obtained after catalysis
of 3 hrs was transferred back to a clean flask and 1.00 mmol
iodobenzene and 1.00 mmol phenylacetylene, 2.0 mmol
K2CO3 and 0.30 mmol dodecane were added to the
Our DFT calculations suggested that the C-C coupling
performs on Pd1/TiO2 through four steps on a Pd-Oad-Ti5c
site (Figure 3f). In the first step iodobenzene adsorbs on
the anchored Pd1 atom with an adsorption energy of 56
kJ/mol (Figure 3f-I) and then C-I bond is broken (Figure 3f-
III). Then, the iodine atom adsorbs on the adjacent Ti5c
atom with an apparent barrier of 40 kJ/mol (Figure 3f-II)
referenced to the energy of gas phase iodobenzene. The
remaining phenyl group binds to the Pd atom (Figure 3f-
III). The second step is physisorption of phenylacetylene
molecule (Figure 3f-IV). The third step is the dissociative
adsorption of C-H bond of phenylacetylene. The C-H bond
of acetylenyl of HC≡C-C6H5 interacts with Oad of TiO2,
resulting in breakage of this C-H bond (Figure 3f-V); this is
due to the nature of high activity of the Oad atom. Upon
this dissociation, phenylacetylenyl binds to the Oad atom;
the H atom couples with adsorbed iodine atom, forming
HI which is one product. The fourth step is the interaction
between the adsorbed C6H5 and C6H5C≡C through the
transition state (Figure 3f-VI) to from product molecule
C6H5-C≡C-C6H5.
As shown in the energy profile (Figure 3f), the step of
dehydrogenation of phenylacetylene (from IV to V in
Figure 3f) is barrier-less and highly exothermic with a large
energy gain of 170 kJ/mol. The last step is the coupling
between the phenyl group adsorbed on the Pd atom and
the phenylacetylenyl bound on the Oad site due to their
close proximity as shown in Figure 3f-V. Although this step
has an activation barrier of 72 kJ/mol, the gained energy,
170 kJ/mol in the C-H dissociation of phenylacetylene
(from IV to V) makes the intermediate (Figure 3f-V) have
enough energy to cross the barrier of 72 kJ/mol for
coupling between phenyl and phenylacetylenyl (Figure 3f-
VI). Thus, we can consider that the step from adsorption
of phenylacetylene (Figure 3f-IV) to formation of product
diphenylacetylene (Figure 3f-VII) is barrierless. From this
point of view, the rate-limiting step of this pathway is thus
the dissociative chemisorption of iodobenzene; activation
barrier of this step is 52 kJ/mol (Figure 3f-II) if referenced
to energy of gaseous iodobenzene (the black dashed line
aligned to 0 kJ/mol). It agrees reasonably with the
experimental value derived from the Arrhenius plot of
kinetics studies of C-C coupling on Pd1/TiO2, 28.9 kJ/mol
(Figure 4d). This difference between the calculated value
o
solution; the solution was heat to 60 C and remained at
o
60 C for 3 hrs under reflux. After that, the solution was
cooled to room temperature for GC-MS analysis. No
detectable products were found in the GC-MS analysis. It
shows that the contribution of Pd atoms left in the solution
for the production of diphenyl acetylene is negligible.
Kinetics studies of C-C coupling on Pd1/TiO2
catalyst. Kinetics study of the C-C coupling on Pd1/TiO2 in
o
the temperature range of 55 – 75 C was studied in a
kinetics controlled regime. Reaction rate, r was calculated
by dividing the number of the reacted phenylacetylene
molecules by the reaction time and the number of the Pd
atoms anchored on TiO2. By plotting ln r as a function of
1/T, an Arrhenius plot was achieved (Figure 4d). With the
slope of Arrhenius plot, the apparent activation barrier for
C-C coupling on 0.20 wt% Pd/TiO2 was calculated. It is 28.9
kJ/mol. With the same method, the activation barrier of
C-C coupling of phenylacetylene and iodobenzene by a
molecular catalyst Pd(PPh3)2Cl2 was measured in the
o
temperature range, 45-65 C. As shown in Figure 4d, the
apparent activation barriers for the same C-C coupling
catalyzed by Pd(PPh3)2Cl2 is 51.7 kJ/mol.
Computational studies of C-C coupling on Pd1/TiO2.
To understand catalytic mechanism of C-C coupling on
Pd1/TiO2, we performed DFT calculations to optimize
surface structure of Pd1/TiO2 and then simulate reaction
pathway by using the anatase (101) surface which has a low
formation energy. As shown in Figure 3a, a Pd atom bond
with two oxygen atoms of TiO2 surface; the Pd atom in
Figure 3a is highly unstable. Compared to the structure in
Figure 3a, we find that one or two oxygen atoms (Oad) can
be added between Pd1 and one 5-coordinted Ti atoms (Ti5c)
near to the Pd1 atom, forming a different structure (Figures
3b) in which three O atoms coordinate with the Pd atom.
Adding two oxygen atoms between Pd1 and two Ti5c can
form two different structures (Figures 3c and 3d) in which
four oxygen atoms bond with a Pt atom. The structure in
Figure 3c is more stable than the one in Figure 3b by about
30 kJ/mol. Table S3 lists the calculated formation energy of
oxygen adatoms for structures in Figure 3. The structural
8
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
This work was mainly supported by the U.S. Department of
and the experimental one likely results from the solvent
1
2
3
4
5
6
7
8
Energy, Office of Science, Office of Basic Energy Sciences,
Chemical Sciences, Geosciences, and Biosciences Division
under Award Number DE-SC0014561.
effect that was not included in the computational studies.
It is expected that the solvent could lower the activation
barrier by stabilizing the tarnation state (structure II in
Figure 3). In summary, the theoretical simulation of C-C
coupling on Pd1/TiO2 suggested that the unique site Pd1-
Oad-Ti5c provides multifunctional atoms Pd, Oad, and Ti5c
for activating C-I of iodobenzene, activating C-H of
phenylacetylene and then performing C-C coupling to
form dipenylacetylene.
REFERENCES
1.
Rev. 2000, 100 (8), 3009-3066.
2. Chinchilla, R.; Najera, C. Chem. Rev.
2007, 107 (3), 874-922.
Beletskaya, I. P.; Cheprakov, A. V. Chem.
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
SUMMARY
We have shown for the first time the efficient C-C
coupling on single atom sites anchored on an oxide
support. Spectroscopic evidence confirmed the
preservation of single Pd atom (Pd1O4) sites anchored on
TiO2 of used catalyst after catalysis. This single-atom
catalyst exhibits high activity and selectivity. Systematic
control experiments confirmed the durability of these
single atoms sites anchored to TiO2. Kinetic studies show
that the apparent activation barrier of C-C coupling on
Pd1/TiO2, 28.9 kJ/mol is lower than 51.7 kJ/mol on
molecular catalyst Pd(PPh3)2Cl2. Computational studies
suggest that Pd1 atom anchored to four oxygen atoms of
TiO2 through Pd-O-Ti bonds is a thermodynamically
favorable structure. A catalytic cycle was proposed, in
which the rate-determining step is the dissociative
chemisorption of iodobenzene with an activation barrier of
this step is 40 kJ/mol. This activation barrier agrees
basically with the experimental value (28.9 kJ/mol) derived
from the Arrhenius plot of kinetics studies of C-C coupling
on Pd1/TiO2. The high activity and selectivity of C-C
coupling on single Pd atom sites anchored on TiO2 suggest
a new chemical route to perform C-C coupling which are
mostly catalyzed by Pd-based molecular catalysts; in this
route products can be readily separated from catalyst.
3.
Li, Y. G.; Zhou, P.; Dai, Z. H.; Hu, Z. X.;
Sun, P. P.; Bao, J. C. New J. Chem. 2006, 30 (6),
832-837.
4.
Son, S. U.; Jang, Y.; Park, J.; Na, H. B.;
Park, H. M.; Yun, H. J.; Lee, J.; Hyeon, T. J. Am.
Chem. Soc. 2004, 126 (16), 5026-5027.
5.
Thathagar, M. B.; Kooyman, P. J.;
Boerleider, R.; Jansen, E.; Elsevier, C. J.;
Rothenberg, G. Adv. Synth. Catal. 2005, 347 (15),
1965-1968.
6.
Hu, J.; Liu, Y. B. Langmuir 2005, 21 (6),
2121-2123.
7.
Kogan, V.; Aizenshtat, Z.; Popovitz-Biro,
R.; Neumann, R. Org. Lett. 2002, 4 (20), 3529-
3532.
8.
Chem. Res. 2003, 36 (8), 638-643.
9. Thathagar, M. B.; Beckers, J.; Rothenberg,
Moreno-Manas, M.; Pleixats, R. Account.
G. J. Am. Chem. Soc. 2002, 124 (40), 11858-
11859.
10.
Langmuir 2002, 18 (12), 4921-4925.
11. Kim, S. W.; Kim, M.; Lee, W. Y.; Hyeon,
T. J. Am. Chem. Soc. 2002, 124 (26), 7642-7643.
12. De La Rosa, M. A.; Velarde, E.; Guzmán,
A. Syn. Commun. 1990, 20 (13), 2059-2064.
13. Kaneda, K.; Higuchi, M.; Imanaka, T. J.
Mol. Catal. 1990, 63 (3), L33-L36.
14. Marck, G.; Villiger, A.; Buchecker, R.
Tetrahedron Lett. 1994, 35 (20), 3277-3280.
15. Miyaura, N.; Suzuki, A. Chem. Rev. 1995,
95 (7), 2457-2483.
Li, Y.; Boone, E.; El-Sayed, M. A.
Supporting Information. Computational studies, control
experiment for checking whether Pd atoms of 0.20 wt%
Pd/TiO2 detached or not, catalytic activity for C-C coupling on
Pd1/TiO2 in solvent benzene, computational studies. This
material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
* To whom all correspondence should be addressed.
Email: franklin.feng.tao@ku.edu
16.
Roth, G. P.; Farina, V.; Liebeskind, L. S.;
Funding Sources
Peña-Cabrera, E. Tetrahedron Lett. 1995, 36 (13),
2191-2194.
This work was supported by Chemical Sciences, Geosciences
and Biosciences Division, Office of Basic Energy Sciences,
Office of Science, U.S. Department of Energy.
17.
Kang, S.-K.; Lee, H.-W.; Jang, S.-B.; Kim,
T.-H.; Kim, J.-S. Syn. Commun. 1996, 26 (23),
4311-4318.
ACKNOWLEDGMENT
18.
Suzuki, A. J. Organomet. Chem. 1999, 576
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
(1), 147-168.
1
2
3
4
19.
Welton, T. Chem. Rev. 1999, 99 (8), 2071-
2084.
20.
Biffis, A.; Zecca, M.; Basato, M. J. Mol.
Catal. A: Chem. 2001, 173 (1), 249-274.
21. Littke, A. F.; Fu, G. C. Angew. Chem. Int.
Ed. 2002, 41 (22), 4176-4211.
22. Yin, L.; Liebscher, J. Chem. Rev. 2007, 107
(1), 133-173.
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
23.
Li, Y.; Fan, X. B.; Qi, J. J.; Ji, J. Y.; Wang,
S. L.; Zhang, G. L.; Zhang, F. B. Nano Res. 2010,
3 (6), 429-437.
24.
Gao, Y. Chem. Commun. 2005, (35), 4435-4437.
25. Cao, S. W.; Tao, F.; Tang, Y.; Li, Y. T.; Yu,
J. G. Chem. Soc. Rev. 2016, 45 (17), 4747-4765.
26. Bratlie, K. M.; Lee, H.; Komvopoulos, K.;
Stevens, P. D.; Li, G.; Fan, J.; Yen, M.;
Yang, P. D.; Somorjai, G. A. Nano Lett. 2007, 7
(10), 3097-3101.
27.
Joo, S. H.; Park, J. Y.; Renzas, J. R.;
Butcher, D. R.; Huang, W. Y.; Somorjai, G. A.
Nano Lett. 2010, 10 (7), 2709-2713.
28.
Kresse, G.; Furthmuller, J. Phys. Rev. B
1996, 54 (16), 11169-11186.
29.
Perdew, J. P.; Burke, K.; Ernzerhof, M.
Phys. Rev. Lett. 1996, 77 (18), 3865-3868.
30.
Blöchl, P. E. Phys. Rev. B 1994, 50 (24),
17953-17979.
31.
Kresse, G.; Joubert, D. Phys. Rev. B 1999,
59 (3), 1758-1775.
32.
Grimme, S.; Antony, J.; Ehrlich, S.; Krieg,
H. J. Chem. Phys. 2010, 132 (15), 154104.
33.
H. J. Chem. Phys. 2000, 113 (22), 9901-9904.
34. Henkelman, G.; Jonsson, H. J. Chem.
Phys. 2000, 113 (22), 9978-9985.
35. Howard, C. J.; Sabine, T. M.; Dickson, F.
Acta. Crystallogr. B 1991, 47, 462-468.
36. Mery, D.; Heuze, K.; Astruc, D. Chem.
Commun. 2003, (15), 1934-1935.
37. Leadbeater, N. E.; Tominack, B. J.
Tetrahedron Lett. 2003, 44 (48), 8653-8656.
38. Király, Z.; Dékány, I.; Mastalir, Á.; Bartók,
M. J. Catal. 1996, 161 (1), 401-408.
39. Ayyappan, S.; Gopalan, R. S.; Subbanna,
G.; Rao, C. J. Mater. Res. 1997, 12 (02), 398-401.
40. Sarkany, A.; Papp, Z.; Sajó, I.; Schay, Z.
Solid State Ionics 2005, 176 (1), 209-215.
41. Burton, P. D.; Boyle, T. J.; Datye, A. K. J.
Catal. 2011, 280 (2), 145-149.
Henkelman, G.; Uberuaga, B. P.; Jonsson,
10
ACS Paragon Plus Environment

Page 11 of 11
Journal of the American Chemical Society
TABLE OF CONTENT
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
11
ACS Paragon Plus Environment