Identification of an Active NiCu Catalyst for Nitrile Synthesis from Alcohol

Article abstract of DOI:10.1021/acscatal.9b00043

Development of heterogeneous catalysts for alcohol transformation into nitriles under oxidant-free conditions is a challenge. Considering the C-H activation on α-carbon of primary alcohols is the rate-determining step, decreasing the activation energy of C-H activation is critical in order to enhance the catalytic activity. Several NiM/Al₂O₃ bimetallic catalysts were synthesized and scrutinized in catalytic transformation of 1-butanol to butyronitrile. Ni-Cu was identified as a suitable combination with the optimized Ni_0.5Cu_0.5/Al₂O₃ catalyst exhibiting 10 times higher turnover frequency than Ni/Al₂O₃ catalyst. X-ray absorption spectroscopy (XAS) and high angle annular dark field scanning transmission electron microscopy (HAADF-STEM) revealed that the NiCu particles in the catalyst exist in the form of homogeneous alloys with an average size of 8.3 nm, providing an experimental foundation to build up a catalyst model for further density functional theory (DFT) calculations. Calculations were done over a series of NiM catalysts, and the experimentally observed activity trend could be rationalized by the Br?nsted-Evans-Polanyi (BEP) principle, i.e., catalysts that afford reduced reaction energy also feature lower activation barriers. The calculated activation energy (E_a) for C-H activation with coadsorbed NH₃ dropped from 63.4 kJ/mol on pure Ni catalyst to 49.9 kJ/mol on the most active NiCu-2 site in NiCu bimetallic catalyst, in good agreement with the experimentally measured activation energy values. The Ni_0.5Cu_0.5/Al₂O₃ catalyst was further employed to convert 11 primary alcohols into nitriles with high to near-quantitative yields, at a Ni loading 10 times less than that of the conventional Ni/Al₂O₃ catalyst.

Full text of DOI:10.1021/acscatal.9b00043

Subscriber access provided by ALBRIGHT COLLEGE
Article
Identification of an Active NiCu Catalyst for Nitrile Synthesis from Alcohol
Yunzhu Wang, Shinya Furukawa, and Ning Yan
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b00043 • Publication Date (Web): 17 Jun 2019
Downloaded from http://pubs.acs.org on June 18, 2019
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 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
Identification of an Active NiCu Catalyst for Nitrile
Synthesis from Alcohol
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
Yunzhu Wang†, Shinya Furukawa*‡§, and Ning Yan*†
† Department of Chemical and Biomolecular Engineering, National University of
Singapore, 4 Engineering Drive 4, Singapore 117585, Singapore.
‡ Institute for Catalysis, Hokkaido University, N-21, W-10, Sapporo 001-0021, Japan.
§ Elementary Strategy Initiative for Catalysis and Battery, Kyoto University, Kyoto
Daigaku Katsura, Nishikyo-ku, Kyoto, Japan, 615-8510.
ABSTRACT: Development of heterogeneous catalysts for alcohol transformation into
nitriles under oxidant free conditions is a challenge. Considering the C–H activation on α-
carbon of primary alcohols is the rate-determining step, decreasing the activation energy
of C–H activation is critical to enhance the catalytic activity. Several NiM/Al2O3 bimetallic
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 56
1
2
3
4
5
6
7
8
9
catalysts were synthesized and scrutinized in catalytic transformation of 1-butanol to
butyronitrile. Ni–Cu was identified as a suitable combination with the optimized
Ni0.5Cu0.5/Al2O3 catalyst exhibiting 10 times higher turnover frequency than Ni/Al2O3
catalyst. X-ray absorption spectroscopy (XAS) and high angle annular dark field scanning
transmission electron microscopy (HAADF-STEM) revealed that the NiCu particles in the
catalyst exist in the form of homogeneous alloys with an average size of 8.3 nm, providing
experimental foundation to build up a catalyst model for further density functional theory
(DFT) calculations. Calculations were done over a series of NiM catalysts, and the
experimentally observed activity trend could be rationalized by the Brønsted–Evans–
Polanyi (BEP) principle, i.e., catalysts afford reduced reaction energy also feature lower
activation barriers. The calculated activation energy (Ea) for C–H activation with co-
adsorbed NH3 dropped from 63.4 kJ/mol on pure Ni catalyst to 49.9 kJ/mol on the most
active NiCu-2 site in NiCu bimetallic catalyst, in good agreement with the experimentally
measured activation energy values. The Ni0.5Cu0.5/Al2O3 catalyst was further employed to
convert 11 primary alcohols into nitriles with high to near quantitative yields, at a Ni
loading ten times less than the conventional Ni/Al2O3 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
2
ACS Paragon Plus Environment

Page 3 of 56
ACS Catalysis
1
2
3
4
5
Keywords: alcohol, nitrile, ammonia, BEP, heterogeneous catalyst, NiCu
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
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 56
1
2
3
4
5
1. INTRODUCTION
6
7
8
9
Nitriles are widely used to manufacture fine chemicals, pharmaceuticals, agrochemicals
and polymers. Despite the commercial significance of nitriles, their traditional preparation
methods, including Sandmeyer reaction1, Rosenmund-von Braun reaction,2 and the
nucleophilic substitution of cyanides to alkyl and aryl halides,3 generally require toxic
starting materials and harsh reaction conditions.4-8 In addition, stoichiometric amount of
chemical wastes are often produced. Primary alcohols represent a class of alternative,
greener starting material,9-10 which can be converted to nitriles through oxygen-involved
ammoxidation11-20 or amination-dehydrogenation21-26 reactions generating only H2O and
H2 as by-products. Limited selectivity due to over-oxidation and/or high energy
consumption associated with high reaction temperatures (280-500 °C) are the significant
disadvantages in the current transformation of alcohols to nitriles using heterogeneous
catalysts. In addition, despite the fact that various aromatic alcohols have been converted
into corresponding nitriles on supported catalysts, transformation of aliphatic alcohols
has been less successful.13, 18, 20, 26-27
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
ACS Paragon Plus Environment

Page 5 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
Ni-based catalysts28-38 have been extensively investigated in the hydrogen-free
conversion of alcohols to primary amines, based on a “hydrogen-borrowing” strategy.39-
42 In this case, the amination reaction follows a dehydrogenation-imination-hydrogenation
pathway, where alcohols are dehydrogenated to aldehydes/ketones and then react with
ammonia to afford imines and finally to amines via hydrogenation. Recently, a Ni/Al2O3
catalyst was reported that converts a series of aliphatic and aromatic primary alcohols to
nitriles under oxidant-free conditions.43 The reaction followed a three-step
dehydrogenation-imination-dehydrogenation pathway, similar to the amination reaction in
the first two steps. Kinetics study revealed that C–H bond cleavage at α-carbon was the
rate-determining step. A high amount of Ni (40 mg Ni per 1µ/min substrate) catalyst was
used to achieve satisfactory yields of nitriles, and the nitrile selectivity for some substrates
was below 90%. To further increase the catalytic performance, an improved catalyst to
reduce the reaction barrier of the rate-determining step is essential.
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
Due to the unique electronic and geometric characteristics in the alloy or core-shell
structure, bimetallic catalysts often show better catalytic performance than their single-
component metal catalysts.44 With its high tunability to form various bimetallic systems,
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 56
1
2
3
4
5
6
7
8
9
Ni is commonly employed as a base metal to construct bimetallic catalysts.45-52 Of
particular significance is the combination of Ni with other 3-d metal elements to form noble
metal free bimetallic catalyst that are affordable and easily scalable. These catalysts have
been used in a wide range of applications, and the superior catalytic activity has often
been correlated to improved morphologies, electronic state and surface properties.44 To
our knowledge, however, bimetallic Ni based catalyst has not yet been systematically
studied in alcohol transformation into amine and nitriles.
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
Recently, density functional theory (DFT) calculations become increasingly important
to rationalize the performance of catalysts and to help catalyst design. Several groups
have utilized a scaling relationship for catalysis research,53-57 for example, to identify
improved catalysts for propane dehydrogenation55 and for ammonia synthesis.57
Considering a major task of rational catalyst design is to reduce the reaction barrier, it is
critical to predict the activation energy of a catalyst of given structure with reasonable
accuracy. In physical chemistry, the BEP principle correlates the activation energies to
the reaction energies of a series of reactions in a family with a linear relationship.58-59 This
empirical model helps predict the kinetic property (activation energy) of a reaction with its
6
ACS Paragon Plus Environment

Page 7 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
thermodynamic property (reaction enthalpy). Applied in catalysis, DFT methods were
generally used to demonstrate the BEP correlation of a surface-catalyzed elementary
step. Many efforts have been devoted to reveal this correlation for molecule dissociation
over transition-metal surfaces by varying the nature of the catalyst.60-63 Bimetallic alloys
differ from their parent metals in surface composition and electronic property, thus
affecting the adsorption energies of substrates and products, and consequently the
reaction energy.
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
Herein, we prepared a series of five NiM bimetallic catalysts for the nitrile synthesis
from alcohols. NiCu was identified as the best combination affording the highest product
yield and selectivity. The NiCu/Al2O3 catalysts with varied metal ratio were prepared, fully
characterized and extensively evaluated in alcohol conversion into nitriles. The DFT
calculations explicitly suggested that NiCu-bimetallic catalyst affords one of the lowest
reaction energies and activation energies all NiM catalysts, providing a rational
understanding of why NiCu is superior to single component Ni and other bimetallic NiM
catalysts. The optimized NiCu/Al2O3 catalyst was applied to transform more than 10
primary alcohols into nitriles with high to near quantitative yields.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 56
1
2
3
4
5
6
7
8
2. EXPERIMENTAL
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.1 Catalysts preparation
γ-Al2O3 was prepared by calcination of boehmite (γ-AlOOH, supplied by SASOL) at 900
°C for 3 h. Ni/Al2O3, Cu/Al2O3 and Ni–M/Al2O3 catalysts (M = Fe, Co, Cu, and Zn) were
prepared by deposition-precipitation (DP) method or wet-impregnation (WI) method with
loading amount of 6 wt%. Metal precursors included Ni(NO3)2·6H2O (99%, Wako),
Cu(NO3)2·3H2O (99%, Sigma Aldrich), Co(NO3)2·6H2O (98%, Wako), Zn(NO3)2·6H2O
(99%, Kanto) and Fe(NO3)3·9H2O (98%, Sigma Aldrich). For DP method, aqueous
solution of urea was dropwise added to a vigorously stirred mixture of γ-Al2O3 and
aqueous solution of the metal precursors in a glass beaker. The molar ratio of Ni and M
(Ni–M/Al2O3) was set to 1. The mixture was sealed tightly with a plastic film and heated
with stirring on a hot stirrer. The temperature of the mixture was kept at 70 °C for 10 h
(Cu and Ni–Cu) or 90 °C for 5 h (Ni, Ni–Co, Ni–Fe, and Ni–Zn). Note that, for Cu-
containing catalysts, the temperature must not exceed 70 °C because dehydration of the
8
ACS Paragon Plus Environment

Page 9 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
precipitated Cu(OH)2 occurs to form aggregated CuO, typically resulting in fatal lowering
of dispersion. Molar ratio of urea and the total amount of metals was set as 36 (Cu, Ni–
Cu, Ni–Zn and Ni–Fe) or 15 (Ni, Ni–Co). After deposition, the colourless supernatant was
removed and the resulting solid was washed with deionize water three times, followed by
drying under reduced pressure. The resulting powder was calcined at 500 °C for 1 h, and
then reduced at 400 °C (Cu/Al2O3) or 600 °C (Ni/Al2O3 and Ni–M/Al2O3) under 50 ml/min
H2 flow for 1 h. For WI method, γ-Al2O3 was added into an aqueous solution of metal
precursors followed by stirring for 2 h at room temperature and drying under a reduced
pressure. The obtained powder was calcined and reduced in a similar fashion as used in
DP method.
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
Unsupported Ni, Cu, and NiCu particles were obtained by a precipitation method. A
diluted solution of sodium hydroxide (NaOH, ≥ 98.5%, Sigma Aldrich, 0.05 M) was slowly
added into a Ni(NO3)2 solution, a Cu(NO3)2 solution, or a mixture of Ni(NO3)2 and
Cu(NO3)2 solution with molar ratio 1:1. The obtained slurry was washed and filtered with
DI water and the solid was dried in an oven at 70 °C. The dried sample was calcined at
400 °C for 3 h, and reduced at 600 °C under 40 ml/min H2 flow for 1 h.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 56
1
2
3
4
5
6
7
8
2.2 Catalyst characterization
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
XAS was conducted at the BL01B1 station in SPring-8, using a Si(111) double-crystal
monochromator. Energy calibration was under taken by using Cu foil. The spectra were
recorded at room temperature at Ni and Cu K-edges in a transmission mode. For static
measurements, the pelletized sample was pre-reduced by H2 at 600 °C for 0.5 h, followed
by sealing into a plastic pack together with an ISO A500-HS oxygen absorber (Fe powder)
under N2 atmosphere. Details for XAS data processing, as well as for X-ray diffraction
(XRD), X-ray fluorescence (XRF), HAADF-STEM, energy-dispersive X-ray spectroscopy
(EDS) and diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) analysis
are provided in the supporting information.
2.3 General procedure for catalytic reactions
Catalytic evaluation experiments were carried out in a stainless steel, fixed-bed tube
reactor (SS316, length 0.3 m, diameter 3/8 inch or 9.5 mm). The reactor was heated by
10
ACS Paragon Plus Environment

Page 11 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
a furnace (Yuanbang Furnace, max 1000 °C) and its temperature was measured by a
temperature probe located close to catalyst bed. A temperature controller was used to
control the temperature of middle part of the tube reactor, where the catalysts were
supported by a sieve and quartz wool. A syringe pump (Harvard PHD 2000 Infusion) was
used to control the flow rate of liquid substrates in a glass syringe (Hamilton 81620), and
the liquid substrates were supplied from the top of the reactor. The flow rates of NH3 and
N2 are controlled by gas flow meters before they were mixed to carry the substrate
through the catalyst bed.
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
All the catalysts were in-situ reduced with H2 (50 mL/min) at 600 °C for 0.5 h before
experiments. The reactor was cooled down to the reaction temperature under H2 flow
after reduction and N2 was switched on to purge the system and remove physically
adsorbed H2. After that, the substrate, NH3 and N2 were supplied to the reactor, while the
products and unconverted substrate were detected online with GC-FID using an Agilent
HP-5 capillary column and He as the carrier gas. The organic products were enriched by
passing through ethanol kept at 0 °C before GC-MS analysis.
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 56
1
2
3
4
5
2.4 Density functional theory (DFT) calculations
6
7
8
9
Periodic DFT calculations were performed using the CASTEP code64 with Vanderbilt-
type ultrasoft pseudopotentials65 and the revised version of Perdew−Burke−Ernzerhof
exchange−correlation functional based on the generalized gradient approximation.66
Other details could be referred to an earlier publication67 and are also available 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
3. RESULTS AND DISCUSSION
3.1 Experimental identification of NiCu as an improved catalyst for butyronitrile synthesis
We prepared a series of NiM (M = Fe, Co, Cu, and Zn) alloys and pure Ni nanoparticles
supported on Al2O3 by DP method, with 6 wt% Ni loading and 1:1 Ni:M ratio. The metallic
phase of each catalyst was determined by the XRD patterns (Figure S1), and details of
the catalysts were summarized in Table S1. Catalytic transformation of 1-butanol to
butyronitrile was selected as a model reaction to compare the activity of the catalysts.
12
ACS Paragon Plus Environment

Page 13 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
Pure Ni catalyst provided 13% butyronitrile. The Ni0.5Co0.5 catalyst was slightly better than
pure Ni (14% yield), while the Ni0.5Cu0.5 catalyst was substantially better, affording a
butyronitrile yield of 47% (Figure 1a). On the other side, incorporating Fe and Zn into Ni
displayed a negative effect: Ni0.5Fe0.5 and Ni0.5Zn0.5 catalysts afforded only 3.5% and 4.7%
butyronitrile, respectively. The Ni0.5Co0.5 and Ni0.5Cu0.5 alloy particles supported on Al2O3
with the same Ni loading were further prepared by WI method. The activity order follows:
Ni0.5Cu0.5 > Ni0.5Co0.5 > Ni, which is in agreement with activity trend for catalysts prepared
by DP method (data shown in Figure S2). From above, we confirm that the incorporation
of a second 3-D metal element into Ni has a significant effect in transformation of alcohols
to nitriles. Combination of Ni with its two adjacent elements, i.e., Co and Cu, results in an
enhanced catalytic activity, in particular in the case of NiCu. The DP method in general
provide more active catalyst than the WI method with the same metal composition,
probably due to the fact that the DP method normally enables more uniform distribution
of metal nanoparticles on the support.68
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
Subsequently, various NiCu catalysts with different metal loading and alloy phase were
prepared by DP method. Alloy phase and particle size were easily modified via altering
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 56
1
2
3
4
5
6
7
8
9
preparation conditions. Ni0.2Cu0.8 in Figure 1b was prepared with 20 wt% Ni, while all the
others contained 6 wt% Ni. For comparison purpose, pure Cu catalyst was prepared by
the same method with 6 wt% metal loading. As shown in Figure 1b, the alloy phase is a
critical factor affecting activity. Incorporation of Cu to Ni significantly enhanced the
catalytic activity compared to pure Ni catalyst, the butyronitrile yield increased with
decreasing Cu ratio. As Cu ratio decreased from 0.8 to 0.55, the butyronitrile yield
increased from 22% to 46%. However, further decreasing the ratio of Cu induced a
reduced butyronitrile yield, with 39% over Ni0.75Cu0.25 catalyst. Pure Cu catalyst caused
high level of side reactions to produce 1-butylamine (Table S2), and resulted in lower
selectivity towards butyronitrile, which is unfavored in nitrile production. This is probably
because of the significant difference in C-H (and/or N-H) activating ability between Ni and
Cu (Ni >> Cu, as demonstrated later by DFT calculation). Dehydrogenation of imine to
nitrile seems to require high C-H (N-H) activation ability, while hydrogenation of imine to
amine is preferable thermodynamically and kinetically. Imine intermediate on Cu prefers
to be amine rather than nitrile due to the insufficient C-H (N-H) activation ability. 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
14
ACS Paragon Plus Environment

Page 15 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
Ni0.5Cu0.5/Al2O3 and Ni/Al2O3 mentioned in Figure 1a, and Cu/Al2O3 mentioned in Figure
1b prepared with DP method were selected for further study.
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 1. Catalytic activities of converting 1-butanol to butyronitrile with (a) various
NiM/Al2O3 and Ni/Al2O3 catalysts prepared with DP method, and (b) various NiCu/Al2O3
and Cu/Al2O3 catalysts prepared with DP method. Reaction conditions: 1 µL/min 1-
butanol, 8 mL/min NH3, 20 mL/min N2, 200 mg catalyst (or 60 mg Ni0.2Cu0.8), 160 °C,
NH3:1-butanol = 30:1.
3.2 Alloy structure of the NiCu catalyst
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 56
1
2
3
4
5
6
7
8
9
X-ray absorption near edge structure (XANES) spectra (Figure 2a-b) show that both Ni
and Cu in Ni0.5Cu0.5/Al2O3 catalyst are close to the metallic states with a small amount of
oxidized Ni species perhaps on the surface. Ni0.5Cu0.5/Al2O3 shows similar extended X-
ray absorption fine structure (EXAFS) oscillations for Ni and Cu in the catalyst (Figure
2c). The periods are almost identical to each other for both Ni and Cu edges, and are
intermediate between those of Ni foil and Cu foil. These are well consistent with the
formation of 1:1 alloy confirmed by the XRD patterns (Figure S1). Fourier-transformed
EXAFS (FT-EXAFS) spectra (Figure 2d-e) shows that only a small contribution of Ni–O
bond to the spectrum of Ni0.5Cu0.5/Al2O3 is suggested, and Cu is almost in full metallic
state. However, no feature of Ni–Ni shell of NiO was observed, indicating that the NiOx
species, if any, are insignificant. EXAFS curve fitting parameters are summarized in Table
S3, and the M–M bond lengths of Ni0.5Cu0.5/Al2O3 for both edges were very close to the
intermediate value (2.513 Å) of those for Ni foil (2.484 Å) and Cu foil (2.542 Å), which is
also consistent with the formation of 1:1 alloy. No apparent difference in the coordination
numbers was observed for Ni–Ni(Cu) and Cu–Cu(Ni) shells (9.1 and 8.8, respectively),
supporting homogeneous alloying in the nanoparticles. The coordination number of Ni–O
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
16
ACS Paragon Plus Environment

Page 17 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
shell in Ni0.5Cu0.5/Al2O3 was 1.1, indicating the existence of a small amount of Ni oxide
species not participated in alloy formation.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. (a-b) XANES spectra, (c) EXAFS oscillations and (d-e) FT-EXAFS spectra of
Ni0.5Cu0.5/Al2O3, Ni/Al2O3 and Cu/Al2O3 catalysts.
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 56
1
2
3
4
5
6
7
8
9
The STEM images of Ni0.5Cu0.5/Al2O3 show that nanoparticles are well dispersed, with
an average particle size of 8.3 nm (Figure 3a-b). A high-resolution STEM image revealed
that the nanoparticle had an fcc crystal structure (Figure 3c). EDS mapping images
pronouncedly demonstrate that Ni and Cu are uniformly mixed (Figure 3d-f). These
results strongly suggest that the nanoparticles are solid-solution alloy of Ni and Cu, which
is consistent with the result of EXAFS. Surprisingly, Ni/Al2O3 and Cu/Al2O3 have much
smaller particle sizes of 0.9 and 1.7 nm, respectively (Figure S3). According to idealized
cubic close-packed full-shell metal clusters,69-70 the surface atom percentage of the
Ni0.5Cu0.5/Al2O3, Ni/Al2O3 and Cu/Al2O3 is estimated to be 17.1%, 87.3 % and 65.6%
(Table S4), based on the average particle size.
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
18
ACS Paragon Plus Environment

Page 19 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. HAADF-STEM images of Ni0.5Cu0.5/Al2O3 and the corresponding elemental
maps of Ni and Cu acquired by EDS: (a) STEM image, (b) size distribution, (c) high-
resolution image of an fcc crystal viewed along [100] direction, (d) close-up of some
nanoparticles, and element sensitive maps of (e) Ni and (f) Cu.
3.3 Reaction kinetics
Intrinsic activities of the three catalysts were measured when the conversions were kept
low (< 30%). Ni0.5Cu0.5/Al2O3 afforded a high turnover frequency (TOF, measured by
amount of butyronitrile produced per surface metal per hour) of 46 h-1, an order of
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 56
1
2
3
4
5
6
7
8
9
magnitude higher than Ni/Al2O3 (4.9 h-1) and Cu/Al2O3 (5.5 h-1), as shown in Figure 4a.
The conspicuously enhanced activity of Ni0.5Cu0.5/Al2O3 may be resulted from the lowered
activation energy. Indeed, experimentally measured apparent Ea for Ni0.5Cu0.5/Al2O3 (51.6
kJ/mol) was much lower than Ni/Al2O3 (72.0 kJ/mol) and Cu/Al2O3 (78.2 kJ/mol) as shown
in Figure 4b-d, which verifies our hypothesis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. (a) TOFs towards production of butyronitrile at 210 °C. Reaction conditions: 4
ul/min 1-butanol, 8 ml/min NH3, 92 ml/min N2, 40 mg catalyst, 210 °C, NH3:1-butanol =
7.5:1. (b-d) Activation energies of Ni0.5Cu0.5/Al2O3, Ni/Al2O3 and Cu/Al2O3 in the
20
ACS Paragon Plus Environment

Page 21 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
conversion of 1-butanol to butyronitrile. Reaction conditions: 1 ul/min 1-butanol, 8 ml/min
NH3, 92 ml/min N2, 10 mg catalyst, NH3:1-butanol = 30:1.
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
The reaction orders of 1-butanol and NH3 were measured at 210 C to avoid
thermodynamic limitation of the reaction. Since 1-butanol is easily vaporized under this
condition, the reaction is regarded as two-phasic between gas and solid catalyst surface,
and the reaction rate is assumed to follow the equation:
r = k * [1-butanol]α * [NH3]β
The reaction orders of 1-butanol and NH3 were measured with constant concentration
of the other reactant. As shown in Figure 5, NH3 was adsorbed much stronger on the
catalysts than 1-butanol resulting in negative orders, and competitive adsorption existed
between NH3 and 1-butanol. The positive order of 1-butanol indicated its activation may
be involved in the rate determining step. The reaction order of alcohol increases from
0.63 to 0.83 after Cu incorporation, suggesting adsorption of 1-butanol on Ni was
weakened in the NiCu alloy. On the other hand, incorporation of Cu to Ni did not change
the NH3 adsorption to an appreciable level (order of -0.14 to -0.13), which indicates that
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 56
1
2
3
4
5
6
7
8
9
NH3 was probably adsorbed on Ni (or Ni rich) sites. For pure Cu catalyst, NH3 adsorption
afforded a close to 0 order (Figure S4), suggesting weaker adsorption of NH3 than on Ni
surface. After the kinetics study, we collected the spent Ni0.5Cu0.5/Al2O3 and Cu/Al2O3
catalysts for further characterizations. X-ray fluorescence (XRF) results (Table S5)
revealed that the Ni and Cu contents in the spent catalyst were almost identical to those
in the fresh catalyst, indicating that leaching of metal species was negligible. XRD
patterns (Figure S5) confirmed the original 1:1 phase was retained in the Ni0.5Cu0.5/Al2O3
catalyst after reaction. The peak for Ni0.5Cu0.5 alloy did not become sharper, suggesting
insignificant growth of the crystalline particle size.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
ACS Paragon Plus Environment

Page 23 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
Figure 5. Plots of reaction orders for the conversion of 1-butanol to butyronitrile, of (a)
NH3, reaction conditions: 4 ul/min 1-butanol with 40 mg Ni0.5Cu0.5/Al2O3, or 3 ul/min 1-
butanol with 80 mg Ni/Al2O3, 100 ml/min total gas flow rate with supplementing N2, 210 °C;
(b) 1-butanol, reaction conditions: 1 to 4 ul/min 1-butanol, 8 ml/min NH3, 92 ml/min N2, 40
mg Ni0.5Cu0.5/Al2O3 or 100 mg Ni/Al2O3, 210 °C.
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
Since product desorption and reactant adsorption may affect the reaction rate, the
desorption rate of 1-butanol, butanal, 1-butylamine, and butyronitrile on metals were
measured with in-situ DRIFTS. To avoid the effect of Al2O3, the DRIFTS analysis was
measured on unsupported metal particles, which were prepared by co-deposition method
and reduced under H2 flow before measurement. The desorption behaviour was analysed
at low temperature that no chemical reactions were observed. The desorption rates of the
organic compounds are assumed to follow:
rdes = kdes * [organic compound]γ
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 56
1
2
3
4
5
6
7
8
9
The adsorption spectra of the various compounds on NiCu, Ni, and Cu are shown in
Figure S6-S8. The peak intensity of C–H stretching at around 2970 cm-1 was selected as
an indicator of adsorbed amount, and the percentage of compounds which cannot be
desorbed after stabilization was regarded as an indicator of adsorption strength. The
desorption is assumed to follow first-order law, while the desorption rate constants were
summarized in Table 1. R square numbers were also provided to highlight the fitting error.
1-butanol and butanal presented the slowest desorption rate constants on pure Ni, while
incorporation of Cu increased their desorption rate constants, indicating their less stable
adsorption on NiCu. This observation was consistent with the 1-butanol order obtained.
In contrast, 1-butylamine had the slowest desorption rate constant on NiCu, and its stable
adsorption on NiCu surface favoured further dehydrogenation to butyronitrile, promoting
the reaction selectivity towards butyronitrile to some extent. Interestingly, the desorption
of final product butyronitrile on three metal surfaces did not show much difference,
indicating nitrile desorption is not the rate limiting step. In addition, close to 100% 1-
butanol desorbed from NiCu surface under the applied temperature (Table S6), implying
the weakened adsorption strength after Cu incorporation into Ni. In contrast, 82% 1-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24
ACS Paragon Plus Environment

Page 25 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
butylamine remained on NiCu surface after stabilization for more than 30 mins, which was
favourable as 1-butylamine could be further converted to butyronitrile. It is worth pointing
out, however, that the values of all rate constants are within the same order of magnitude,
suggesting the key differences of the catalysts do not lie in their ability to adsorb the
starting material, or the product and side-product, or the aldehyde intermediate. On the
other hand, the adsorption of 1-butanol on the surface of Ni0.5Cu0.5/Al2O3 catalyst was
much slower, indicating the strong adsorption of 1-butanol or 1-butoxide on the real
catalyst surface (Figure S9).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Desorption rate constant based on first-order desorption ratea
Desorption Rate
Constant (min-1)
NiCu
Ni
Cu
0.38 (R2 =
0.985)
0.27 (R2 =
0.981)
0.47 (R2 =
0.990)
1-Butanol
0.42 (R2 =
0.969)
0.20 (R2 =
0.874)
0.36 (R2 =
0.948)
Butanal
0.33 (R2 =
0.974)
0.68 (R2 =
0.987)
0.63 (R2 =
0.963)
1-Butylamine
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 56
1
2
3
4
5
6
0.33 (R2 =
0.970)
0.34 (R2 =
0.989)
Butyronitrile
0.34 (R2 = 0.990)
7
8
9
a. At t = 0, 10 µl of organic compounds were injected under N2 flow, 130 °C.
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
Our previous work identified that the rate-determining step of the reaction is the C–H
bond cleavage at α-carbon over pure Ni catalyst.43 To verify whether this still applies on
the NiCu bimetallic catalyst, a kinetic isotope effect (KIE) study was conducted using the
Ni0.5Cu0.5/Al2O3 catalyst. 1-Butanol and isotope labelled substrate, 1-butanol-1,1-d2
(deuteration of both H at the α-carbon of 1-butanol), was converted to butyronitrile at 160
°C. TOF obtained with CH3(CH2)2–CD2–OH was smaller than with 1-butanol, affording a
KIE value of 1.89 (Table 2), which was slightly higher than that obtained over pure Ni
catalyst in the previous work.43 The modest positive value suggested that C–H bond
breakage at α-carbon remain as a slow step in the overall reaction kinetics, establishing
a starting point of the following DFT calculations.
26
ACS Paragon Plus Environment

Page 27 of 56
ACS Catalysis
1
2
3
4
5
Table 2. Kinetic isotopic effects for converting 1-butanol to butyronitrile
6
7
8
9
Substrate
1-butanol
TOF (h-1) KIE
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
2
3
5.94
3.15
1.89
1-butanol-1,1-d2
k1-butanol / k1-butanol-1,1-d2
Reaction conditions: 1 ul/min substrate, 2 ml/min NH3, 23 ml/min N2, 50 mg
Ni0.5Cu0.5/Al2O3, 160 °C, NH3:1-butanol = 7.5:1.
3.4 Energetics on the catalyst surface based on DFT
Based on the alloy structure revealed by XANES, EXAFS, XRD and STEM studies, we
performed DFT calculations for the C–H activation of methoxide at various Ni-Cu hollow
sites (Ni3, Ni2Cu, NiCu2, and Cu3, Figure 6a) with or without ammonia co-adsorbed, to
better understand the energetics on the catalyst surface. Table 3 summarizes the
adsorption (Ead) and C–H activation energy of methoxide (Ea) and ΔE for various reaction
sites of Ni and NiCu (entries 1~7). Ead increased as the number of Cu atoms in the hollow
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 56
1
2
3
4
5
6
7
8
9
site (OMe site) increased, suggesting that Cu makes the adsorption of methoxide
unfavorable. A good linear correlation was observed between Ead and the number of Ni
atoms included in the hollow site (0~3, Figure S10). Correction considering the
contribution of Ni atoms in the second surface layer improved the linearity (R2 = 0.993,
Figure S10b). These results suggest that the adsorption strength depends dominantly on
the surface atomic arrangement, but little on the bulk structure. Ea also varied widely
depending on the Ni–Cu combination of the reaction site (Table 3). C–H cleavage
occurring at Cu atom (conformation NiCu-4 and -6, entries 5 and 7, respectively) gave
high Ea and ΔE, which reflects that Cu itself is less active for C–H activation of methoxide
than Ni. The conformation NiCu-2, where the reaction occurs over Ni2Cu hollow site,
showed lower Ea and ΔE values than pure Ni (entries 1 and 3). The energetics of C–H
activation in the presence of co-adsorbed ammonia was also studied, which is closer to
the real situation. We focused on the effect of ammonia on the adsorption strength of
methoxide and formaldehyde. A strong correlation was observed between the adjacency
of the oxygen atom of formaldehyde and ammonia’s hydrogen: Ead became more
negative as the O–H distance decreased (3.1 → 1.9 Å, Figure S11). This result strongly
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
28
ACS Paragon Plus Environment

Page 29 of 56
ACS Catalysis
1
2
3
4
5
6
7
8
9
suggests the formation of hydrogen bond (HB) between the carbonyl oxygen and
ammonia’s hydrogen, which stabilizes the adsorbates. On the contrary, for methoxide,
Ead was almost constant and independent on the distance to NH3. This is probably
because the oxygen atom of methoxide has already been fully coordinated (3 Ni and 1
C), hence the close contact of NH3 is hindered sterically and electronically. Consequently,
aldehyde is preferentially stabilized in the presence of NH3, which significantly decreases
ΔE. This trend was more prominent when the number of hydrogen bonding increased,
while there was no effect when NH3 was too far from the carbonyl oxygen (Figure S11
and Table 3 entries 10~12). The positive effect of NH3 was also observed for the C–H
activation at Cu sites (NiCu-4~NiCu-6: Table 3 entries 13~15). Thus, the co-adsorption
of ammonia facilitates the C–H activation of methoxide kinetically and thermodynamically
with the aid of hydrogen bonding. It should be noted that the calculated Ea for Ni (63.4
kJ/mol) and the most active NiCu-2 (49.9 kJ/mol) with ammonia are consistent with the
corresponding experimental values obtained in this study (Ni: 72.0 kJ/mol, NiCu: 51.6
kJ/mol), hinting at the validity of DFT calculation. Figure 6b shows the plot of all the
calculated Ea and ΔE, where a strong linear correlation was observed. This clearly
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
29
ACS Paragon Plus Environment