Atomically Dispersed Ru on Manganese Oxide Catalyst Boosts Oxidative Cyanation

Article abstract of DOI:10.1021/acscatal.0c00485

There is a strong incentive for environmentally benign and sustainable production of organic nitriles to avoid the use of toxic cyanides. Here we report that manganese oxide nanorod-supported single-site Ru catalysts are active, selective, and stable for oxidative cyanation of various alcohols to give the corresponding nitriles with molecular oxygen and ammonia as the reactants. The very low amount of Ru (0.1 wt %) with atomic dispersion boosts the catalytic performance of manganese oxides. Experimental and theoretical results show how the Ru sites enhance the ammonia resistance of the catalyst, bolstering its performance in alcohol dehydrogenation and oxygen activation, the key steps in the oxidative cyanation. This investigation demonstrates the high efficiency of a single-site Ru catalyst for nitrile production.

Full text of DOI:10.1021/acscatal.0c00485

Subscriber access provided by the University of Exeter
Article
Atomically Dispersed Ru on Manganese
Oxide Catalyst Boosts Oxidative Cyanation
Hai Wang, Dongyang Xu, Erjia Guan, Liang Wang, Jian Zhang, Chengtao Wang,
Sai Wang, Hua Xu, Xiangju Meng, Bo Yang, Bruce C. Gates, and Feng-Shou Xiao
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.0c00485 • Publication Date (Web): 07 May 2020
Downloaded from pubs.acs.org on May 7, 2020
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 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 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.
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 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
Atomically Dispersed Ru on Manganese Oxide
Catalyst Boosts Oxidative Cyanation
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
Hai Wanga,#, Dongyang Xub,#, Erjia Guanc,#, Liang Wanga,*, Jian Zhange, Chengtao
Wangd, Sai Wangd, Hua Xud, Xiangju Mengd, Bo Yangb,*, Bruce C. Gatesc, and Feng-
Shou Xiaoa,d,e,*
aKey Lab of Biomass Chemical Engineering of Ministry of Education, College of
Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, China.
bSchool of Physical Science and Technology, ShanghaiTech University, Shanghai
201210, China.
cDepartment of Chemical Engineering, University of California, Davis CA 95616,
United States.
dKey Lab of Applied Chemistry of Zhejiang Province, Department of Chemistry,
Zhejiang University, Hangzhou 310028, China.
eBeijing Advanced Innovation Center for Soft Matter, Science and Engineering,
Beijing University of Chemical Technology, Beijing 100029, China.
*Corresponding author: liangwang@zju.edu.cn; yangbo1@shanghaitech.edu.cn;
fsxiao@zju.edu.cn
# H. Wang, D. Xu, and E. Guan contributed equally.
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 39
1
2
3
4
5
ABSTRACT
6
7
8
9
There is a strong incentive for environmentally benign and sustainable production of
organic nitriles to avoid the use of toxic cyanides. Here we report that manganese oxide
nanorod-supported single-site Ru catalysts are active, selective, and stable for oxidative
cyanation of various alcohols to give the corresponding nitriles with molecular oxygen
and ammonia as the reactants. The very low amount of Ru (0.1 wt%) with atomic
dispersion boosts the catalytic performance of manganese oxides. Experimental and
theoretical results show how the Ru sites enhance the ammonia resistance of the
catalyst, bolstering its performance in alcohol dehydrogenation and oxygen activation,
the key steps in the oxidative cyanation. This investigation demonstrates the high
efficiency of a single-site Ru catalyst for nitrile production.
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
Keywords: single-site Ru; nanorod manganese oxide; oxidative cyanation; ammonia;
nitrile
2
ACS Paragon Plus Environment

Page 3 of 39
ACS Catalysis
1
2
3
4
5
1. INTRODUCTION
6
7
8
9
Organic nitriles play essential roles in chemical technology, in applications for
producing pharmaceuticals, agrochemicals, fine chemicals, and building blocks for
high-performance polymers, rubbers, and molecular electronics.1-6 The classical routes
for production of organic nitriles employ toxic HCN or metal cyanides (e.g., KCN,
NaCN, and CuCN),5 and these are associated with the tragic environmental disasters at
Bhopal in 1984 and Baia Mare, Romania, in 2000. There is a need for sustainable
technology to avoid these toxic chemicals.
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
Several routes have been developed for nitrile synthesis without cyanides,
including, for example, the hydroxylamine route using a clay catalyst, whereby the
hydroxylamine hydrochloride reactant was used under microwave conditions.7
Catalytic ammoxidation has emerged as a preferred route nitrile synthesis, with
molecular ammonia being used as the nitrogen source reacting with hydrocarbons,3
alcohols,2,4,8 or aldehydes8. The ammoxidation processes generally takes place with
solid catalysts and gaseous reactants at high temperatures (300–550 °C), being limited
by over-oxidation to form CO2 and thus insufficient selectivities to nitriles.5,9 Moreover,
the strongly corrosive nature of gaseous ammonia at high temperatures is a serious issue.
Liquid-phase ammoxidation can be carried out to effectively hinder the over-oxidation
under mild reaction reactions, with high activities offered by copper complex
catalysts.10,11 However, these processes are limited by the catalyst separation and
regeneration challenges that are intrinsic to homogeneous catalysis. Seeking improved
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 39
1
2
3
4
5
6
7
8
9
solid catalysts, researchers have reported successes with Ru(OH)3,8 ferric and cobalt
carbides,2,12 Au-13 and Pt-containing catalysts,14 and manganese oxide.4,15,16 However,
correlations between catalyst structure and performance are still lacking, and we posit
that new insights would help to accelerate the development of new catalysts with
enhanced performance for oxidative cyanations.
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
Now we report a manganese oxide-supported catalyst that is highly efficient for
the oxidative cyanation. Key to the success of this catalyst is atomically dispersed
ruthenium on the support, used in the form of nanorods (denoted as MnO2-r). The
catalyst, xRu/MnO2-r (x is the Ru loading in wt percent), was made by a wet-chemical
method with Ru loadings as low as 0.1 wt%. The data presented herein show that the
single-site Ru remarkably boosts the catalytic activity for oxidative cyanation of
aliphatic, benzylic, allylic, and heterocyclic alcohols into the corresponding nitriles by
enhancing the ammonia resistance of the catalyst and accelerating the alcohol
dehydrogenation and oxygen activation. These are new characteristics of atomically
dispersed (single-site) catalysts.
2. EXPERIMENAL SECTION
2.1. Catalyst Preparation. Synthesis of nanorod α-MnO2 (MnO2-r). In a typical
synthesis, an aqueous solution of KMnO4 (0.37 M, 50 mL) was added dropwise into a
stirred aqueous solution of MnSO4 (1.73 M, 15 mL). After stirring at room temperature
for another 0.5 h, the resultant precipitate was collected by filtration, washed with a
4
ACS Paragon Plus Environment

Page 5 of 39
ACS Catalysis
1
2
3
4
5
large excess of deionized water, and dried at 150 °C overnight to give the MnO2-r
6
7
sample.
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
Synthesis of Ru/MnO2-r. RuCl3 (2.1 mg) was dissolved in 85 mL of deionized water,
followed by addition of 10 mL of aqueous NaOH (0.2 M) solution with stirring. Then
1.0 g of as-synthesized MnO2-r was added and the mixture stirred at 50 °C for 12 h.
The slurry was separated by centrifugation, and the resultant solid was washed with an
excess of deionized water, then dried at 150 °C for 2 h to give the Ru-loaded sample.
Analysis of the product by ICP-OES determined a Ru loading of 0.1 wt%, with the
sample denoted 0.1Ru/MnO2-r. Ru/MnO2-r catalysts with different Ru contents were
synthesized similarly but with different masses of RuCl3 in the starting solution: 0.20,
6.16, and 63.47 mg of RuCl3 for the synthesis of 0.01Ru/MnO2-r, 0.3Ru/MnO2-r, and
3.0Ru/MnO2-r, respectively.
Synthesis of Pd/MnO2-r, Pt/MnO2-r, Rh/MnO2-r, Au/MnO2-r, Ag/MnO2-r, and
Fe/MnO2-r. These catalysts were prepared by prodecures similar to that stated above
for the synthesis of Ru/MnO2-r, except that Na2PdCl4 (6.0 mgPd mL-1, 0.5 mL), H2PtCl6
(7.50 mgPt mL-1, 0.40 mL), RhCl3 (7.92 mgRh mL-1, 0.38 mL), HAuCl4 (9.56 mgAu mL-1,
0.32 mL) aqueous solutions and solid AgNO3 (5.00 mg), FeCl3·6H2O (15.0 mg) were
used as precursors, respectively. The Pd, Pt, Rh, Au, Ag, and Fe loadings were
controlled to be ~0.3 wt% for these samples.
Synthesis of Ru/CeO2, Ru/MnO, Ru/A-MnO2, Ru/Mn2O3, Ru/Mn3O4, Ru/Co3O4,
Ru/SiO2, Ru/C, and Ru/Al2O3. In a typical synthesis, of Ru/CeO2, 2.1 mg of RuCl3 was
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 39
1
2
3
4
5
6
7
8
9
dissolved in 85 mL of deionized water with stirring, followed by the addition of an
aqueous NaOH solution (0.2 M, 10 mL). Then 1.0 g of CeO2 was added and the mixture
stirred at 50 °C for another 12 h; the resultant solid product in the slurry was separated
by centrifugation, washed with deionized water, and dried at 150 °C for 2 h, giving the
Ru/CeO2 sample with a Ru loading of 0.1 wt%. The Ru/MnO, Ru/A-MnO2, Ru/Mn2O3,
Ru/Mn3O4, Ru/Co3O4, Ru/SiO2, Ru/C, and Ru/Al2O3 catalysts were synthesized
similarly, but with the supports being MnO, A-MnO2, Mn2O3, Mn3O4, Co3O4, SiO2,
carbon, and Al2O3, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2. Characterization of Catalysts. X-ray diffraction (XRD) patterns were obtained
with a Rigaku D/MAX 2550 diffractometer with Cu Kα radiation (λ = 1.5418 Å). Metal
loadings were determined by inductively coupled plasma optical emission spectrometry
(ICP-OES) analysis (Perkin-Elmer 3300DV). Transmission electron microscopy
(TEM), scanning transmission electron microscopy (STEM), and energy dispersive
spectroscopy were carried out with a Cs-corrected JEM-ARM300F electron
microscope (JEOL, Japan) with an acceleration voltage of 300 kV. EXAFS data at the
ruthenium K-edge were recorded at beamline 8-ID at NSLS-II. In these experiments,
each sample (approximately 50 mg) was pressed into a wafer and sealed with Kapton
tape. Each sample was tested in air at room temperature. The spectra were collected in
fluorescence mode, and at least 10 scans were collected for each sample to optimize
data quality. Analysis of the EXAFS data was carried out with the software ATHENA
and ARTEMIS of the IFEFFIT package.17 Theoretical phases and amplitudes of single
scattering paths were obtained in FEFF for a ruthenium oxide reference,18 and a
6
ACS Paragon Plus Environment

Page 7 of 39
ACS Catalysis
1
2
3
4
5
theoretical model was built with single-site Ru located on Mn defect sites in MnO2.
6
7
8
9
2.3. Catalyst Performance Evaluation. Oxidative Cyanation of Alcohols. Oxidative
cyanation reactions were carried out in a high-pressure steel autoclave reactor equipped
with a PTFE bottle, magnetic stirrer (900 rpm), and an explosion-proof pressure sensor.
In a typical run, the as-synthesized catalyst, aqueous ammonia (28–30% NH3), and
substrate alcohols were dispersed within t-amyl alcohol solvent in the reactor, and then
the autoclave was purged with oxygen for three times. Then the reactor, equipped with
an internal thermocouple, was quickly heated to the desired temperature and controlled
at that temperature (the reaction temperature was measured by a thermocouple in the
autoclave), and the autogenous pressure was measured and maintained. After the
desired reaction time, the reactor was placed in an ice bath to quench the reaction. After
separation of the solid catalyst, the liquid products and unconverted reactants were
analyzed with a Shimadzu GC-2014 gas chromatograph equipped with a flame
ionization detector (FID) and an Agilent HP-6890 gas chromatograph-mass
spectrometer, with bromobenzene and hexadecane used as internal standards. The
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
13
1
products with high boiling points were analyzed by C and H NMR spectroscopies.
The gas-phase products (e.g., CO and CO2) were analyzed with a Fu Li-9790 gas
chromatograph equipped with a thermal conductivity detector (TCD). (Safety Note:
The high-pressure oxygen or air are widely used in the aerobic oxidations,19,20 and the
reaction systems in this work were out of the explosion limits of the reactants. For
example, the explosion limit of benzyl alcohol is 1.3-13.0% in oxygen, and the
concentration of benzyl alcohol in the gaseous phase in the reactor is in a 0-0.4% region.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 39
1
2
3
4
5
The same trends also occur for n-heptanol. Furthermore, the fire and static electricity
6
7
are still not allowed to access the internal reactor for safety reasons.
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
Catalytic Dehydrogenation of Methanol. The dehydrogenation of methanol was
carried out in a continuous-flow vertical fixed-bed glass reactor with length of 0.5 m
and an internal diameter of 6 mm. Methanol was injected into the reactor with a pump,
and the flow rate of NH3 was controlled with a mass flow controller. Typically, 100 mg
of catalyst particles (40-60 mesh) mixed with 300 mg of quartz sand (40-60 mesh) were
localized within the reactor tube, with the flow rate of the 1.0 volume% NH3 in N2 being
10 sccm; the injection rate of the methanol was 1.2 mL/h, and the reaction temperature
was 300 °C. The gas-phase products were analyzed on-line with a Fu Li-9790 gas
chromatograph equipped with a TCD.
Recycling Test and Hot Filtration Experiment. After each reaction experiment, the
catalyst was removed from the reactor and washed with ethanol/deionized water, dried
overnight at 150 °C, and then used in the next experiment to evaluate its recyclability.
Hot filtration was carried out to evaluate the catalyst leaching in the reaction liquor.
After oxidative cyanation of n-heptanol for 4 h (0.2 mmol of n-heptanol, 100 mg of
0.1Ru/MnO2-r, 4 mL of t-amyl alcohol solvent, 100 μL of aqueous NH3 (28-30% NH3),
2.0 MPa of oxygen, 100 °C), the gas produced in the reaction was released after cooling
of the reactor to room temperature. Then the resultant product mixture was filtered at
80 °C, and a known mass of t-amyl alcohol solvent and another 100 μL of aqueous
ammonia were added to the liquor to make the total volume 4 mL, which was then used
8
ACS Paragon Plus Environment

Page 9 of 39
ACS Catalysis
1
2
3
4
5
in the next experiment without additional catalyst.
6
7
8
9
Analysis of Catalyst Performance Data. The conversion, product yields, and
selectivity in the oxidative cyanation were calculated according to the following
equations:
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
Moles of substrate converted
Conversion =
100%

(
)
Moles of substrate fed
Moles of product
Yield of product =
100%

(
)
Moles of substrate fed
Moles of product
Selectivity to product =
100%

(
)
Moles of substrate converted
In the kinetics study, the average reaction rates were calculated from the moles of
-1
substrate converted per gram of catalyst in one hour (mmol gcat h-1), with the
conversion of substrate controlled to be lower than 20.0%.
Carbon balances. The carbon balances before and after reaction were based on the
number of carbon atoms in all the reactants and products in the liquor before and after
reaction. The CO2 formed by over-oxidation was not included in calculating the carbon
balances. The carbon balance values were calculated according to the following
equation:
Mf * 7 + M1 * n1 + M2 * n2 + M3 * n3 + … + Mx * nx
C% =
100%

Me * 7
Here C% is the carbon balance closure percentage, Mf is the final number of moles of
n-heptanol in the reactor after reaction; 7 is the number of carbon atoms in a single n-
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 39
1
2
3
4
5
6
7
8
9
heptanol molecule; M1 (M2, M3, … Mx) are the numbers of moles of liquid product 1
(product 2, 3,……x) in the reactor after reaction; n1 (n2, n3,…..nx) is the number of
carbon atoms in a single molecule of product 1 (product 2, 3,……x); Me is the moles
of n-heptanol in the feed before 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
2.4. Computational Details. All the density functional theory (DFT) calculations were
performed with the VASP package21 using the PBE functional and the projector
augmented wave (PAW) pseudopotential.22 The van der Waals interactions were
described with Grimme’s empirical three-body dispersion correction in the scheme of
Becke-Johnson damping, namely, PBE-D3(BJ).23 Hubbard U corrections were used,
and we employed U - J = 5.2 eV, for the spherical part of the interaction, and J = 1.0
eV, according to the values reported.24 The calculated lattice parameters for α-MnO2
with the PBE + U approach were a = b = 9.814 Å and c = 2.895 Å, which agree well
with the experimental parameters (a = b = 9.750 Å and c = 2.861 Å).25 The (100) surface
dominates and represents 54% of the α-MnO2 surface area with low energies of 0.64 J
m-2 according to a previous report.24 This present work built the optimized MnO2 (100)
slab with a dimension of 9.81  8.69  23 Å, consisting of 48 O and 24 Mn atoms. The
bottom halves of the slabs were frozen at their bulk positions, and the rest of the atoms
were relaxed. A value of the cut-off energy of 400 eV and a k-point mesh of 3  3  1
were employed. For Al2O3, a plane-wave basis set with a cutoff energy of 400 eV and
a 6 × 6 × 2 k-point grid generated with the Monkhorst−Pack scheme were found to give
converged results. A surface slab consisted of 48 O atoms and 32 Al atoms, and the
slabs were separated by a vacuum with a spacing of 15 Å in the z-direction to stabilize
10
ACS Paragon Plus Environment

Page 11 of 39
ACS Catalysis
1
2
3
4
5
the species in the slabs. The dimensions of the Al2O3(0001) surface slab used in the
6
7
calculations were: a = 8.253 Å, b = 9.530 Å, and c = 20.660 Å.
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
3. RESULTS AND DISCUSSION
3.1. Evaluation of Data Characterizing n-Heptanol Cyanation.
Figure 1 shows data characterizing the catalytic performance in a challenging
reaction: oxidative cyanation of n-heptanol. The data provide a comparison of the
performance of the supported catalyst containing 0.3 wt% Ru with that of the support
MnO2-r alone, which itself is highly active (n-heptanol conversion of 40.7% with 94.3%
selectivity to heptanonitrile under our conditions) (Table S1, Entry 2). Addition of Ru
to the support led to marked increases in both the n-heptanol conversion and
heptanonitrile selectivity, illustrated, for example, by the conversion of 66.9%
with >99.5% selectivity to heptanonitrile under these conditions (Figure 1, Table S1).
Even with a decreased Ru loading of 0.1 wt%, the catalyst gave an n-heptanol
conversion of 58.3% with unchanged selectivity. Increasing the reaction time from 2 to
5 h led to an improved n-heptanol conversion of 82.4% with a nitrile selectivity of
98.3%.
When metals other than Ru were incorporated in the catalyst (Au, Pt, Pd, Rh, Ag,
or Fe), however, the conversion and selectivity in the n-heptanol oxidative cyanation
were markedly less, than with the MnO2-r catalyst (Figure 1, Table S1), which might
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 39
1
2
3
4
5
6
7
be an indication that these additives blocked part of the MnO2-r surface sites. These
results indicate that the Ru sites play a critical role in the n-heptanol oxidative cyanation.
8
9
[Cat]
OH
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
CN
NH3, O2
100
90
80
70
60
50
40
30
20
100
90
80
70
60
50
40
30
20
Yield of heptanonitrile
Conv. of n-heptanol
Sel. for heptanonitrile
]
[ar
2
4
2
r
r
r
O
O
3
O
r
r
r
-
r
r
-
2
r
r
r
-
-
-
-
-
-
-
-
-
e
n
-
o
2
2
2
2
2
2
2
C
/
2
2
2
M
O
O
O
O
O
O
C
O
R
O
u
2
O
O
O
-
/
n
O
n
n
u
n
n
n
n
n
n
n
n
A
u
n
/
R
M
M
M
/
M
0
M
M
/
M
M
M
M
/
M
/
R
/
/
/
u
/
/
/
1
M
t
.
d
/
1
.
h
g
u
u
u
e
u
u
R
0
P
P
A
F
A
R
1
0
R
R
R
.
3
3
3
3
R
3
3
.
1
3
.
0
1
.
.
.
.
0
.
.
.
1
.
0
.
0
0
0
0
0
0
3
0
0
0
Figure 1. Performance data characterizing oxidative cyanation of n-heptanol with
various catalysts. Reaction conditions: 0.2 mmol of substrate, 4 mL of t-amyl alcohol,
100 μL of aqueous NH3 (28–30 wt%), 100 mg of catalyst, 2.0 MPa of oxygen, 100 °C,
2 h. [a]Reaction time: 5 h. The carbon balance data are shown in Table S1 in the
Supporting Information.
It is well known that Ru(OH)3 present at a high loading (>2.0 wt%) on a support
is active for the oxidative cyanation,8 but Ru at a low loading (0.1 wt%) on the supports
CeO2, commercial activated MnO2 (A-MnO2), Co3O4, MnO, Mn3O4, and Mn2O3 does
not catalyze this oxidative cyanation (Figures 1, S1, and S2, Tables S2 and S3 in the
Supporting Information)-with the exception of MnO2-r. These results demonstrate that
both the Ru and MnO2-r are necessary for the catalysis. Increasing the loading of Ru
(to 3.0 wt%) increased the activity for the n-heptanol transformation, but at the cost of
lower heptanonitrile selectivity, as more COx formed by over-oxidation(Figure 1, Table
12
ACS Paragon Plus Environment

Page 13 of 39
ACS Catalysis
1
2
3
4
5
S1 in the Supporting Information).
6
7
8
9
3.2. Investigation of a Range of Substrates. Figure 2 includes data characterizing the
synthesis of a group of substituted and functionalized benzonitriles in the presence of
the 0.1Ru/MnO2-r catalyst. These substrates include aliphatic, allylic, alkynyl, and
other, structurally diverse, heterocyclic nitriles. Under mild reaction conditions (0.5
MPa of oxygen, 100 °C, 0.025 mol% of Ru with an S/C ratio of ~4042, where this is
the molar ratio of substrate to Ru in the reaction system), the functionalized benzylic
nitriles with electron-donating or electron-withdrawing substituents were efficiently
synthesized with high yields (Figure 2, 1-17). In particular, the halogenated
benzonitriles, which can be used for the preparation of dyes, pharmaceuticals, potential
high-performance pesticides, and specialty engineering plastics, were also synthesized
with yields exceeding 71.7%, giving the corresponding by-product amides from the
side-reaction nitrile hydration. Considering that the amides are also a class of valuable
product and could be transformed into nitriles by simple dehydration,26 we also include
them in the evaluation of the product yields characterizing the various substrates (the
total yields of nitriles and amides are given in parentheses in Figure 2) that gave greater
than 99.0% yields of nitriles and amides for these substrates (Figure 2, 1-17). The
sterically hindered substrate, such as 2-methoxybenzyl alcohol, were also successfully
transformed into the corresponding nitriles with yields >99.0% (Figures 2, 9).
Significantly, the relatively complex substrates, such as quinoline-6-carbonitrile,
piperonyl alcohol, biphenyl-4-methanol, 4-phenoxyphenyl methanol, 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
benzyloxybenzyl alcohol, and 3,5-dibenzyloxybenzyl alcohol, were also selectively
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 39
1
2
3
4
5
converted into the corresponding nitriles in >90% yields (Figure 2, 18-23).
6
7
8
9
Compared with the benzylic alcohols, aliphatic alcohols usually have lower
reactivities.2,13 Nonetheless, the 0.1Ru/MnO2-r catalyst still exhibited high efficacy for
the ammoxidation of aliphatic alcohols under our reaction conditions (0.5 MPa of
oxygen, 100 °C, S/C at ~1010). For example, the alcohols 1-hexanol, n-heptanol, 1-
nonanol, 1-decanol, and 4-phenylbutan-1-ol were almost fully transformed on
0.1Ru/MnO2-r (Figure 2, 24-28), giving >90% yields to the desired nitriles and amides.
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 the transformation of allylic and alkynyl alcohols, the side reactions of
isomerization, hydration, and over oxidation usually occur.8 But the 0.1Ru/MnO2-r
catalyst still gave nitrile yields higher than 99.0% from the substrates cinnamyl alcohol,
geraniol, and 4-ethynylphenyl methanol (Figure 2, 29-31). In particular, the structurally
complex 3,7-dimethyl-2,6-octadienenitrile was effectively synthesized with a
yield >99.0%, showing that the catalyst clearly outperformed Cu11 and Co/Fe catalysts,2
which have been reported to be highly efficient. Furthermore, in the oxidative
cyanations of heterocyclic alcohols including 3-pyridinemethanol, furfuryl alcohol, and
3-thiophenemethanol, the 0.1Ru/MnO2-r catalyst still gave >99% yields to the nitriles
(Figure 2, 32-34). Such performances are even comparable with those in the
conventional heterocyclic nitrile synthesis route via the cyanation of heteroaryl halide.5
14
ACS Paragon Plus Environment

Page 15 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
General reaction scheme
a
b
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
Ru/MnO2-r
(0.025 - 0.099 mol% Ru)
R CH2OH
R CN
R CONH2
NH3
+
+
O2, t-amyl alcohol, 100-130 C
Substituted and functionalized benzonitriles
CN
CN
CN
CN
CN
CN
F
F
Br
F
Cl
F
4: 93.5% (>99.0) 4h
2: >99.0% 5h
3: >99.0% 4h
1: >99.0% 5h
CN
CN
O
O
8: >99.0% 7h
7: >99.0% 5h
6: >99.0% 3h
O
5: >99.0% 4h
CN
CN
CN
CN
F
F
F
O
O
O
9: >99.0% 12h
10: >99.0% 10h
11: >99.0% 8h
12: 95.5% (>99.0) 4h
Cl
CN
Br
CN
Cl
CN
CN
O
Cl
Br
O
14: 71.7% (>99.0) 3h⁋ 15: 77.3% (>99.0) 5.5h
13: 86.4% (>99.0) 1.5h‡
16: >99.0% 12h
CN
CN
CN
CN
O
O
O
N
20: >99.0% 6h
18: >99.0% 4h
19: >99.0% 4h
17: >99.0% 12h
CN
O
O
CN
O
CN
O
21: 98.3% 10h
23: 90.0% (>99.0%) 14h
22: >99.0% 14h
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 39
1
2
3
4
5
6
7
8
9
Aliphatic nitriles
c
d
CN
CN
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
CN
24: 91.3% 12h†
25: 37.8% (>99.0%) 14h⁋
26: 35.7% (>99.0%) 24h†
CN
CN
28: 23.2% (90.2%) 24h†
27: 25.6% (>99.0%) 30h†
Allylic and alkynyl nitriles
CN
Me
CN
CN
Me
Me
30: >99.0% 24h‡
31: >99.0% 5h
29: >99.0% 5h
Structurally diverse heterocyclic nitriles
e
CN
CN
O
CN
N
S
34: >99.0 5h
33: 97.3 (>99.0) 14h
32: >99.0 4h
Figure 2. Synthesis of different types of nitriles catalyzed by 0.1Ru/MnO2-r.
Reaction conditions: 0.4 mmol of substrate, 4 mL of t-amyl alcohol, 100 μL of
aqueous NH3 (28–30 wt%), 10 mg of catalyst (0.025 mol% Ru), 0.5 MPa of oxygen,
100 °C. The values in parentheses are yields of the nitriles and amides. ⁋20 mg of
catalyst (0.05 mol% Ru). †0.2 mmol of substrate, 200 μL of aqueous NH3 (28–30
wt%), 20 mg of catalyst (0.099 mol% Ru), 120 °C. ⁋0.2 mmol of substrate, 200 μL
of aqueous NH3 (28–30 wt%), 20 mg of catalyst (0.099 mol% Ru), 130 °C. ‡30 mg
of catalyst (0.074 mol% Ru).
Moreover, the catalyst 0.1Ru/MnO2-r was used in the gram-scale synthesis of
16
ACS Paragon Plus Environment

Page 17 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
various nitriles, and high yields were still obtained (Figure 3)—results that might
suggest the potential practicability of the 0.1Ru/MnO2-r catalyst. These data confirm
the catalytic performances of 0.1Ru/MnO2-r catalyst in the oxidative cyanation of
various substrates (Tables S4 and S5 in the Supporting Information).
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
Representative gram-scale nitrile syntheses
CN
CN
CN
O
O
O
O
97.7% (95.3%) 8h
93.0% (88.7%) 4.5h
97.7% (93.6%) 6h
CN
CN
Me
CN
N
Me
Me
O
97.9% (92.0%) 26h‡
98.5% (96.0%) 6.5h
98.0% (96.2%) 16h
Figure 3. Gram-scale syntheses of nitriles catalyzed by 0.1Ru/MnO2-r. Reaction
conditions: 1 g of substrate, 60 mL of t-amyl alcohol, 100 μL of aqueous NH3 (28–30
wt%) for each 0.4 mmol substrate, 0.1Ru/MnO2-r (0.025 mol% Ru), 0.5 MPa of oxygen,
100 °C. ‡0.074 mol% Ru. The values in parentheses are the yields of isolated products.
The photographs show the products isolated from the reaction system.
3.3. Catalyst Recyclability. We further emphasize that the 0.1Ru/MnO2-r catalyst is
stable in the reaction solution, as confirmed by the results of a hot filtration experiment
(Figure S3 in the Supporting Information). After each reaction run, the catalyst was
easily separated from the products, washed with ethanol and water, dried, and reused.
In the continuous recycling tests (Figure S4 in the Supporting Information), the catalyst
0.1Ru/MnO2-r exhibited stable performance with undetectable deactivation. Moreover,
we did not observe Ru leaching by characterization of the post-reaction solutions by
ICP analyses, a further indication of the excellent recyclability of the catalyst.
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 39
1
2
3
4
5
6
7
8
9
3.4. Catalyst Structure Characterization. To indentify the catalyst structure and
function of Ru species, multiple characterizations, catalyst performance measurements,
and computational simulations were performed. Figure S5 in the Supporting
Information shows XRD patterns of MnO2-rand 0.1Ru/MnO2-r, including weak, broad
peaks associated with α-MnO2 crystals (Figure S6 in the Supporting Information). In
contrast, XRD patterns indicative of metallic Ru, RuOx, or Ru(OH)3 species were
undetectable, pointing to a high dispersion of the Ru species on the support.
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
4.2 ±1.1 nm
6
5
4
3
2
1
0
b
a
0.3Ru/MnO2-r
0.1Ru/MnO2-r
0.01Ru/MnO2-r
c
Ru-O
0 1 2 3 4 5 6 7 8
Diameter (nm)
Ru-Mn
50 nm
b
1 nm
Ru-Ru
0
1
2
3
4
5
5 nm
Radial Distance (Å)
Figure 4. (a) TEM and (b) high-resolution HAADF-STEM images of the catalyst
0.1Ru/MnO2-r. Inset in (a), distribution of diameters of the nanorods; inset in (b),
enlarged view of the area in the red square in the HAADF-STEM image. The atomically
dispersed Ru is highlighted by yellow circles. (c) Ru K-edge EXAFS spectra in R-space
of various catalysts. The dotted lines are as follows: black, Ru-Ru and blue, Ru-O single
scattering paths generated from a Ru oxide reference, and red, Ru-Mn single scattering
18
ACS Paragon Plus Environment

Page 19 of 39
ACS Catalysis
1
2
3
4
path generated with FEFF models of single-site Ru dispersed on MnO2.
5
6
7
8
9
Transmission electron microscopic images of the 0.1Ru/MnO2-r show a nanorod
morphology, with small diameters, 3–5 nm (Figure 4a, inset, Figure S7 in the
Supporting Information). Aberration-corrected high-angle annular dark-field scanning
transmission electron microscopy (HAADF-STEM) provided direct atomic-resolution
observations of the Ru atoms (Figure 4b). The elemental distributions of Mn and O are
shown by energy dispersive X-ray data (EDX), but the signal of Ru is much weaker,
consistent with its low loading and high dispersion (Figures S8 and S9 in the Supporting
Information). Significantly, there is no evidence of Ru nanoparticles/nanoclusters in the
HAADF-STEM images—evidently, all of the ruthenium was atomically dispersed. Not
only do the STEM images of the 0.1Ru/MnO2-r sample provide no evidence of Ru
nanoclusters/nanoparticles, they demonstrate that the isolated Ru atoms occupy exactly
the positions of missing Mn atoms at the support surface (inset, Figure 4b), in good
agreement with earlier reports showing the presence of catalytic metals at support
surface sites.27,28
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
Further characterization by extended X-ray absorption fine structure (EXAFS)
spectroscopy showed Ru–O (distance R = 1.48 Å, before phase correction) and Ru–Mn
(R = 2.58 Å, before phase correction) contributions characterizing 0.01Ru/MnO2-r,
0.1Ru/MnO2-r, 0.3Ru/MnO2-r, and 3.0Ru/MnO2-r samples, consistent with the
bonding of the isolated Ru atoms to MnO2-r via Ru–O bonds (Figure 4c, Figure S10 in
the Supporting Information). These data gave no evidence of edge features
characteristic of metallic Ru or Ru–O–Ru, in agreement with the atomic Ru dispersion.
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 39
1
2
3
4
5
6
7
8
9
The XPS data characterizing the 0.1Ru/MnO2-r catalyst give evidence of the Ru
3p3/2 binding energy at 464.4 eV (Figure S11a in the Supporting Information), which
has a significant shift to higher energy compared with metallic Ru0 and is assigned to
positively charged Ru.29,30 Such feature is associated with the interaction between a
single-site Ru and the MnO2-r matrix via an Ru–O–Mn linkage involving interfacial
oxygen species (Figure S11b in the Supporting Information), in good agreement with
many observations characterizing metal oxide-supported single-site species.28,31,32
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
X-ray absorption near-edge structure (XANES) spectra of Ru/ MnO2-r catalysts
(Figure S11c in the Supporting Information) show the Ru K-edge white lines with
higher energy and stronger intensities than in the spectra of Ru foil. This result confirms
the positive oxidation state of Ru in the sample,29,33,34 consistent with the XPS results.
3.5. Reaction Pathways and Kinetics. To better understand the role of the single-site
Ru on MnO2-r as a catalyst, we did further experiments characterizing oxidative
cyanation using a model reactant, benzyl alcohol. Figure S12 in the Supporting
Information shows data characterizing benzyl alcohol conversion as a function of time
as the reaction was catalyzed by 0.1Ru/MnO2-r, MnO2-r, and A-MnO2. MnO2-r gave
almost full conversion of benzyl alcohol after 90 min under our conditions, showing
that it is more active than A-MnO2 (17% conversion after 90 min), consistent with the
known high activity of nanorod-MnO2.35,36 Significantly, addition of 0.1 wt% Ru to this
catalyst led to complete conversion after just 45 min, giving a measure of the promotion
effect of the Ru sites on the catalytic performance. Moreover, the 0.1Ru/MnO2-r and
20
ACS Paragon Plus Environment

Page 21 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
MnO2-r was found to be characterized by apparent activation energy (Ea) values of 29.2
and 40.1 kJ/mol, respectively (Figure S13 in the Supporting Information). The marked
lowering of Ea upon addition of Ru to the catalyst further demonstrates the benefit of
the Ru. Also significant, catalysts consisting of of 0.1 wt% Ru on silica, carbon, or
alumina (Table S3 in the Supporting Information) were characterized by undetectable
benzyl alcohol conversion or nitrile formation under our conditions (Figure S14 in the
Supporting Information), further demonstrating that both the Ru and MnO2-r are
necessary for a highly efficient oxidative cyanation 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
It has been reported that benzaldehyde is an intermediate in the oxidative
cyanation to synthesize nitriles.2,8,36,37 This reaction proceeds via oxidative
dehydrogenation of alcohol to give aldehyde, aldehyde-ammonia condensation to
afford imine, and oxidative dehydrogenation to form nitrile (Figure 5a).2,8,37 As shown
by data characterizing the transformation of benzyl alcohol and benzaldehyde under
oxidative cyanation conditions, the reaction converting benzaldehyde is generally faster
than that converting benzyl alcohol for each catalyst (Figure 5b, Figure S15 in the
Supporting Information), a comparison that suggests that the alcohol dehydrogenation
(alcohol-to-aldehyde transformation) is the kinetically crucial reaction in the network
of the oxidative cyanation process, consistent with the imine dehydrogenation being
known as extremely fast.36-38
Further insight into the reaction scheme was gained from experiments to determine
H/D kinetic isotope effect (KIE). The kH/kD value was measured by using benzyl
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 39
1
2
3
4
5
6
7
8
9
alcohol (PhCH2OH) and for comparison isotopically labeled benzyl alcohol (PhCD2OH)
as substrates, giving the value kH/kD = 6.5 (determined from the pseudo first-order rate
constants kH = 5.1  10-2 min-1 and kD = 7.9  10-3 min-1, Figure S16 in the Supporting
Information). Such a high kH/kD value suggests that the cleavage of the C–H bond at the
benzylic position is rate-determining, with the dehydrogenation of OH species
occurring facilly.39-42
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
O2
H2O
NH3 H2O
O2
H2O
a
R
CH2OH
R CHNH
R
CHO
34.2
R
CN
OH
O
b35
c
d
33.3
Ru
Au
Pt
25.6
40
2.0
1.6
1.2
0.8
27.9
30
25
20
15
10
5
25.0
24.7
Ru/MnO2-r
30
20
10
0
18.5
Pt/MnO2-r
Au/MnO2-r
MnO2-r
11.5
12.3
11.1
0.1
0
0.1
r
0.4
0
0
60 120 180 240
TOS (min)
r
r
r
f
f
f
O 2
r
n
n
f
f
O 2
-
n
-
-
f
-
o
o
o
o
o
O 2
O 2
o
O 2
n
n
O 2
H 3
H 3
n
M
M
n
n
H 3
H 3
H 3
H 3
n
-
-
M
M
N
N
M
N
A
A
N
N
/
M
/
N
u
u
R
R
Figure 5. (a) Proposed sequence of reactions in the oxidative cyanation of alcohols to
nitriles. (b) Average reaction rate (r0) in the conversion of benzyl alcohol and
benzaldehyde catalyzed by A-MnO2, MnO2-r, and 0.1Ru/MnO2-r. (c) Average reaction
rate (r0) in the conversion of benzyl alcohol catalyzed by Au/MnO2-r, Pt/MnO2-r, and
Ru/MnO2-r. Reaction conditions for (b) and (c): 0.4 mmol of substrate, 4 mL of t-amyl
alcohol solvent, 100 μL of aqueous NH3 (28–30 wt%), 1.5 MPa of oxygen, 75 °C. The
conversions were controlled to be in the range of 0–20% for calculating the average
reaction rates (which, in the limit of zero conversion, approach true reaction rates). (d)
Average H2 production rate in the dehydrogenation of methanol with ammonia in the
fed gas with various catalysts. Reaction conditions: 100 mg of catalyst, 300 °C,
methanol feed rate, 1.2 mL/h; 1% NH3/N2 flow rate of 10 sccm.
Recognizing that supported Au and Pt catalysts are efficient for oxidative
22
ACS Paragon Plus Environment

Page 23 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
dehydrogenation of alcohols,39-42 we compared the catalytic activities in oxidative
dehydrogenation of benzyl alcohol with and without ammonia, with the catalysts being
Au/MnO2-r, Pt/MnO2-r, Ru/MnO2-r, and some others (Figures 5c, Figures S14, S17,
and S18, Tables S6 and S7 in the Supporting Information). Notably, the Au/MnO2-r,
Pt/MnO2-r, and MnO2-r are characterized by average reaction rates of 34.2, 25.6, and
18.6 mmol gcat-1 h-1 in the absence of ammonia, and the reaction on catalysts with Au
and Pt was found to be faster than that on Ru/MnO2-r (average reaction rate, 24.7 mmol
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
-1
gcat h-1). After introduction of ammonia into the reacting liquid, the Au/MnO2-r,
Pt/MnO2-r, and MnO2-r catalysts were characterized by significantly reduced average
rates (12.3, 11.1, and 11.5 mmol gcat-1 h-1), whereas the Ru/MnO2-r catalyst still gave a
high benzyl alcohol oxidative dehydrogenation average rate of 25.0 mmol gcat-1 h-1. This
comparison suggests the possibility of strong inhibition of the Au and Pt by
coordination of ammonia.43,44
Comparisons of the various catalysts for dehydrogenation in the presence of
ammonia were performed with methanol reactant, which was chosen because methanol
is the simplest alcohol and usually employed for evaluations of the dehydrogenation
activity.45,46 In these tests without oxygen in the feed, deep dehydrogenation occurred
to form CO and H2 as the products, and the toxic products (e.g., HCN) did not. With
regard to a comparison of the dehydrogenation activities, the data show clearly that the
Ru/MnO2-r catalyst is much more active than the others, including Au/MnO2-r,
Pt/MnO2-r, and MnO2-r (Figure 5d). We stress that the Ru/MnO2-r is stable for
hydrogen production in the presence of ammonia, demonstrating its superior ammonia-
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 39
1
2
3
4
5
6
7
resistance. This result is in good agreement with the well-known function of Ru
catalysts, which are widely used in ammonia synthesis.47,48
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
Furthermore, the approximate experimental reaction orders, that is, the
dependence of average reaction rate on the concentrations of NH3 and alcohol in the
reaction catalyzed by Ru/MnO2-r and MnO2-r, also indicate the enhanced ammonia-
resistance of Ru/MnO2-r (Figure S19 in the Supporting Information). These results
might help to explain the superior catalytic performance of Ru/MnO2-r in the oxidative
cyanation of alcohols, which is strongly correlated with its high activity for oxidative
dehydrogenation of alcohols in the presence of ammonia.
3.6. DFT Simulations. To help elucidate the role of Ru in the catalysis, we did DFT
simulations by employing single-site Ru anchored on the surface of an α-MnO2 nanorod
as a model of the Ru/MnO2-r catalyst (Figures S20 and S21 in the Supporting
Information). For comparison, in the simulations we replaced Ru with Au and,
alternatively, with Pt (Figure S22 in the Supporting Information). Figure 6 shows
adsorption energies (Eads) of various substrate molecules and energy profiles
characterizing the steps during benzyl alcohol dehydrogenation on the Au-, Pt-, and Ru-
decorated α-MnO2 nanorod, along with the proposed steps for PhCH2OH
dehydrogenation on the Ru/MnO2-r catalyst with pre-adsorbed NH3. Multiple
structures and reaction pathways were considered, and only the most energetically
favorable ones are shown here.
24
ACS Paragon Plus Environment

Page 25 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
PhCHβHO+Hα
Ph-CHβHO+Hα
a
b
c
0
-1
-2
-3
-4
Top view
Side view
Hα
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
PhCH
NH
2
OH
3
O2
Ru
MnO2-r
Pt
Au
0.03
0
-1
-2
-3
-4
-5
-0.96
-1.86
-1.22
-1.94
-2.32
-2.30
-2.82
Ph-CHO+Hα+Hβ
-2.70
-2.71
Pt+NH
Ru+NH
Au
Ru
3
Hα
-3.63
-3.62
3
-4.17
-4.69
Hβ
-4.76
Pt
)
g
(
H
+
H
H
2
+
+
H
O
O
H2
O
O
H
C
H
O
N
Mn
Ru
H2
C
H
H
C
C
h
C
h
-
h
P
h
P
P
P
Figure 6. (a) Adsorption energies of PhCH2OH, NH3, and O2 on MnO2-r, Au/MnO2-r,
Pt/MnO2-r, and Ru/MnO2-r. (b) Energy profiles for PhCH2OH dehydrogenation on
Au/MnO2-r, Pt/MnO2-r, Ru/MnO2-r, Pt+NH3/MnO2-r, and Ru+NH3/MnO2-r. (c) The
proposed PhCH2OH dehydrogenation steps on Ru/MnO2-r catalyst with pre-adsorbed
NH3 (Ru+NH3/MnO2-r).
The results show that benzyl alcohol can be efficiently adsorbed on MnO2-r, with
an Eads of -2.19 eV. With Au, Pt, or Ru sites on the MnO2-r, the benzyl alcohol tends
to adsorb on these sites, as confirmed by the Eads values of -2.32, -3.62, and -3.63 eV,
respectively (Figure 6a). Furthermore, these metal sites also enhance the oxygen
adsorption on the catalyst relative to what occurs on bare MnO2-r, particularly for the
Pt- and Ru-containing catalysts, with Eads values of -1.14 and -1.40 eV, respectively.
Thus, these sites should benefit the oxidation reactions.40,49
To elucidate the differences between Au, Pt, and Ru sites, we performed
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 39
1
2
3
4
5
6
7
8
9
calculations to determine the energetics of the dehydrogenation process converting
benzyl alcohol to benzaldehyde. When a benzyl alcohol molecule adsorbs on one of the
metal atoms (Au, Pt, or Ru), the hydrogen atom of the hydroxyl group (denoted as Hα,
Figures S23-25 in the Supporting Information) transfers to a nearby surface oxygen
atom (denoted as Oα) to form an adsorbed HαOα* species, resulting the benzyl alkoxide
(PhCH2O-) intermediate on the noble metal. This step is regarded as the most favorable
first step in alcohol oxidation, consistent with the reported results.40-42,50 Thereafter, the
adsorbed benzyl alkoxide (PhCH2O-) undergoes the second deprotonation process,
whereby the C–H bond at the benzylic position undergoes dissociation, giving the
hydrogen atom (denoted as Hβ) that adsorbs at the adjacent oxygen atom on the support
(denoted as Oβ) to produce the adsorbed benzaldehyde (Figures S23-25 in the
Supporting Information).40-42,50
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 the C–H cleavage step, the Au-, Pt-, and Ru-containing catalysts are
characterized by the energy barriers 0.46, 0.91, and 0.93 eV, respectively (Figure 6b,
Table S8 in the Supporting Information). This result suggests that the C–H cleavage is
the kinetically crucial step in the benzyl alcohol dehydrogenation, which is also
confirmed experimentally by the KIE results (Figure S16 in the Supporting
Information). The lower energy barrier for reaction on Au- than on Ru- and Pt-
containing catalysts is in agreement with the higher activity of Au/MnO2-r in the benzyl
alcohol dehydrogenation indicated by the results of the kinetics experiments (Figure
5c).
26
ACS Paragon Plus Environment

Page 27 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
The calculations show that when ammonia is introduced onto the catalyst surface,
its adsorption by interaction of the lone pair electrons of the N atom is favored on the
metal sites rather than on the MnO2-r matrix (Figure 6a and Figure S26 in the
Supporting Information).29,43,51 Thus, there is a competition between benzyl alcohol and
ammonia for adsorption on Au sites, as is made evident by the comparable Eads values
characterizing ammonia and benzyl alcohol molecules (-2.32 and -2.13 eV,
respectively). The calculations, after structure optimization, show that desorption of
benzyl alcohol from the catalyst surface takes place when benzyl alcohol and ammonia
are co-adsorbed, and so the alcohol dehydrogenation on the Au sites is hindered (Figure
6a and Figure S27 in the Supporting Information).43
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 contrast, the Pt- and Ru-containing catalysts are characterized by stronger
adsorption of the alcohol than of ammonia, but their energy barriers for the cleavage of
the C–H bond in these substrates are quite different. For example, the energy barrier for
the C–H bond cleavage characterizing ammonia on the Pt-containing catalyst is 1.25
eV, which is much higher than that when ammonia is absent (0.91 eV, Figure 6b, and
Table S9 in the Supporting Information). The increased energy barrier for the C–H bond
cleavage in the presence of ammonia is ascribed to the lack of availability of the reactive
oxygen species for C–H bond cleavage, because the NH3 molecule occupies the site for
this neighboring oxygen species on the Pt atom (Figure S28 in the Supporting
Information).
To summarize: These results confirm the strong inhibition by ammonia in the
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 39
1
2
3
4
5
6
7
catalytic dehydrogenation on Au- and Pt-containing catalysts—inhibition caused by
hindering the alcohol adsorption on Au and enhancing the C–H cleavage barrier, on Pt.
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 contast, and of central importance to the work reported here, the Ru-containing
catalyst with coordinated ammonia is characterized by an energy barrier for C–H bond
cleavage of 0.98 eV, a value that is close to the value characterizing the ammonia-free
case (0.93 eV, Figure 6b, Table S9 in the Supporting Information).
We stress that the MnO2-r without Ru also shows activity for the reaction, which
motivated our further calculations characterizing the bare MnO2-r surface. The energy
barrier for C–H bond cleavage was found to be 0.45 eV with ammonia present, which
is even lower than that of 0.98 eV characterizing that on the Ru on the Ru/MnO2-r
surface. After the dehydrogenation step, the hydrogen atoms can reduce the reactive
oxygen to form water and oxygen vacancies (Ovac) on the MnO2-r surface. A complete
reaction cycle requires the regeneration of the reactive oxygen species preceding the
adsorption and dissociation of molecular oxygen. The calculated adsorption energy of
molecular oxygen over reduced MnO2-r (one oxygen vacancy near an Mn site, Mn-Ovac)
is -0.45 eV (Figure S29a in the Supporting Information) and the corresponding
activation barrier of oxygen dissociation is 2.35 eV (Figures S29b and S29c in the
Supporting Information). The high dissociation energy barrier of oxygen suggests that
the oxygen regeneration is unfavourable in the continuous dehydrogenation of more
alcohol molecules on the MnO2-r.
Further calculations demonstrated that more neighboring oxygen vacancies could
28
ACS Paragon Plus Environment

Page 29 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
lower the energy barrier. For example, the calculations show that the existence of two
neighboring oxygen vacancies around an Mn site leads to an oxygen dissociation barrier
of 1.76 eV (Figures S29b and S29d in the Supporting Information)— but the value is
still high.
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
It should be noted here that, under the practical reaction conditions, ammonia
might be competing with molecular oxygen at the vacancy sites over catalyst surface.
As shown in Figure S30 in the Supporting Information, the ammonia favors occupancy
of the Mn-Ovac site to suppress the oxygen adsorption, as is shown by the enhanced
adsorption energy (Figure S29a in the Supporting Information and Figure 6a).
Significantly, the Ru sites on MnO2-r efficiently decrease the oxygen dissociation
barrier, with a value of 0.56 eV on the oxygen vacancy close a Ru atom (Figures S31
and S32, Table S10 in the Supporting Information), which is much lower than what we
calculated for reaction on MnO2-r: values in the range of 1.76–2.35 eV. Even with the
presence of ammonia, the oxygen dissociation barrier still appeared at 0.69 eV on the
Ru/MnO2-r (Figure S33 in the Supporting Information). These values are also much
smaller than that characterizing the inert oxide-supported Ru (e.g., Ru/Al2O3 in Figures
S34 and S35 in the Supporting Information).
In summary, on the basis of these results characterizing substrate adsorption, C–
H cleavage, and regeneration of the reactive oxygen surface species, we suggest that
the Ru/MnO2-r catalyst appears to be optimal for continuous PhCH2OH
dehydrogenation in the presence of ammonia (Figure 6c), which might explain the high
29
ACS Paragon Plus Environment