A New Molybdenum Nitride Catalyst with Rhombohedral MoS2 Structure for Hydrogenation Applications

Article abstract of DOI:10.1021/jacs.5b01446

Nitrogen-rich transition-metal nitrides hold great promise to be the next-generation catalysts for clean and renewable energy applications. However, incorporation of nitrogen into the crystalline lattices of transition metals is thermodynamically unfavorable at atmospheric pressure; most of the known transition metal nitrides are nitrogen-deficient with molar ratios of N:metal less than a unity. In this work, we have formulated a high-pressure route for the synthesis of a nitrogen-rich molybdenum nitride through a solid-state ion-exchange reaction. The newly discovered nitride, 3R-MoN₂, adopts a rhombohedral R3m structure, isotypic with MoS₂. This new nitride exhibits catalytic activities that are three times more active than the traditional catalyst MoS₂ for the hydrodesulfurization of dibenzothiophene and more than twice as high in the selectivity to hydrogenation. The nitride is also catalytically active in sour methanation of syngas with >80% CO and H₂ conversion at 723 K. Our formulated route for the synthesis of 3R-MoN₂ is at a moderate pressure of 3.5 GPa and, thus, is feasible for industrial-scale catalyst production.

Full text of DOI:10.1021/jacs.5b01446

Subscriber access provided by NEW YORK MED COLL
Article
A New Molybdenum Nitride Catalyst with Rhombohedral
MoS2 Structure for Hydrogenation Applications
Shanmin Wang, Hui Ge, Shouli Sun, Jianzhong Zhang, Fangming Liu, Xiaodong Wen,
Xiaohui Yu, Liping Wang, Yi Zhang, Hongwu Xu, Joerg C. Neuefeind, Zhangfeng Qin,
Changfeng Chen, Changqing Jin, Yongwang Li, Duanwei He, and Yusheng Zhao
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b01446 • Publication Date (Web): 23 Mar 2015
Downloaded from http://pubs.acs.org on March 26, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
A New Molybdenum Nitride Catalyst with Rhombohedral MoS
2
Structure for
Hydrogenation Applications
1
,3,4,*
2,*
6
4
3
2,7
4,5
Shanmin Wang,
Hui Ge, Shouli Sun, Jianzhong Zhang, Fangming Liu, Xiaodong Wen, Xiaohui Yu,
1 4 8 2 1 5
1
Liping Wang, Yi Zhang, Hongwu Xu, Joerg C. Neuefeind, Zhangfeng Qin, Changfeng Chen, Changqin Jin,
2,7
3
1,4,5,*
Yongwang Li, Duanwei He, and Yusheng Zhao
1
HiPSEC & Physics Department, University of Nevada, Las Vegas, Nevada 89154, USA
Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan, Shanxi 030001, China
Institute of Atomic & Molecular Physics, Sichuan University, Chengdu 610065, China
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA
Institute of Physics, Chinese Academy of Sciences, Beijing, 100190, China
National Institute of Clean– & Low–Carbon Energy (NICE), Beijing 102209, China
Synfuels China, Beijing, 100195, China
2
3
4
5
6
7
8
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
Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA
Abstract
Nitrogen–rich transition–metal nitrides hold great promise to be the next–
generation catalysts for clean and renewable energy applications. However,
incorporation of nitrogen into the crystalline lattices of transition metals is
thermodynamically unfavorable at atmospheric pressure; most of the known
transition metal nitrides are nitrogen–deficient with molar ratios of N : metal less
than a unity. In this work, we have formulated a high–pressure route for the
synthesis of a nitrogen–rich molybdenum nitride through a solid–state ion–
exchange reaction. The newly discovered nitride, 3R–MoN
rhombohedral R3m structure, isotypic with MoS . This new nitride exhibits
catalytic activities that are three times more active than the traditional catalyst
MoS for the hydrodesulfurization of dibenzothiophene and more than twice
higher in the selectivity to hydrogenation. The nitride is also catalytically active
in sour methanation of syngas with >80% CO and H conversion at 723 K. Our
formulated route for the synthesis of 3R–MoN is at a moderate pressure of 3.5
GPa and is thus feasible for industrial–scale catalyst production.
2
, adopts a
2
2
2
2
Keywords: molybdenum nitride; MoN
2
; highꢀpressure synthesis; catalyst; hydrodesulfurization;
sour mathanation.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
Heterogeneous catalysis plays a central role in key chemical processes associated
with petrochemical and renewable energy generation. Since the first success in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
catalytic liquefaction of coal in 1920s, MoS –based catalysts have been extensively
2
1
2
studied and used for hydrotreating of fossil fuels, sour methanation of syngas,
3
4
electrocatalysis for water splitting, and solar energy conversion. The excellent
catalytic performances of this family of catalysts are primarily attributed to the
2
layered structure of MoS in which the interlayers of Mo and S are held together by
5
,6
weak van der Waals forces. This structural characteristic allows producing a variety
of nano–clusters with highly exposed active edge sites on the S–Mo–S sandwich layer.
7
The catalytic activity of the exfoliated MoS
2
thin layers, for example, is three times
8
that of the dispersed MoS
2
on a silica mesoporous substrate. In addition, the edge
sites of the S–Mo–S layer can absorb promoter species, such as Co and Ni,
1,8
significantly improving the catalytic performances.
Despite several decades of searches for other catalysts, the transition–metal (TM)
disulfides remain the most studied and used catalysts in deep hydroprocessing of
1
,9
fossil fuels including hydrodesulfurization (HDS) and hydrodenitrogenation (HDN).
Over the past decade, however, TM nitrides have attracted considerable attention
because they exhibit a number of unique and advanced catalytic properties for
10
hydrotreating (e.g., Mo
2
N and W
2
N) and photo– and electrochemical–catalysis (e.g.,
1
1ꢀ13
δ–MoN, Mo N, and Ta
2
3
N
5
),
although none of them adopts a layered structure.
One important property is their excellent resistance to corrosion by sulfur, a common
2
ACS Paragon Plus Environment

Page 3 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
,2
catalyst poison; hence, they can be used for sour catalysis applications. In some
reactions such as dehydrogenation and hydrogenolysis, molybdenum nitride shows
1
0,12,14,15
similar catalytic activities as platinum–group catalysts.
Thus, nitrogen–rich
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
TM nitrides, particularly those of MoS –type, may hold great promise as the
2
nextꢀgeneration catalysts for a wide range of applications.
Preparation of nitride–rich TM nitrides is challenging because incorporation of
nitrogen into the crystal lattice of TMs is often thermodynamically unfavorable at
16
atmospheric pressure. As a result, most of the known TM nitrides are nitrogen–
x
deficient with x ≤ 1 in TMN , in spite of the fact that TMs often exhibit oxidation
states exceeding +3. To date, reports on TM nitrides with x > 1 are rather fragmentary,
sometimes even contradictory, especially on their crystal structures. In the binary Mo–
N system, three different phases with varying nitrogen concentrations have been
17
reported: tetragonal β–MoN
x
with x ≤ 0.5 (e.g., Mo
2
N ), cubic
γ
–MoN with 0.5 ≤ x <
x
18
19
1
(e.g., Mo
Mo
Using high–pressure techniques, a number of novel nitrogen–rich TM nitrides have
3
N
2
), and hexagonal δ–MoN
x
with x ≥ 1 (e.g., MoN ). Also noted is that
1
9
5
6
N is the only nitride with x > 1 in the binary system Mo–N.
20
2
recently been synthesized from direct metal–gas (N ) reactions. These compounds
21
21,22
23
24
include new polymorphs of Zr
3
N
4
, Hf
3
N
4
,
Ta
2 3 3 5
N , and Ta N , and noble metal
25,26
dinitrides (OsN , IrN , and PtN ).
Successful high–P synthesis of nitrides with
2
2
2
higher oxidation states demonstrates that pressure can effectively promote the role of
d–electrons in chemical bonding with nitrogen. However, the required pressures in
most of the previous synthetic routes are relatively high, in a range of 11 – 50 GPa,
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
which is beyond the current technological capability for massive, industrial–scale
production. Hexagonal ReN
2
has recently been synthesized through a solid–state
–type
reaction at 7.7 GPa, and is the only nitride possessing a layered MoS
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
27
structure. To date, no nitrogen–rich nitride has been synthesized in the Mo–N
system, although an oxidation state as high as +6 occurs in other chemical systems
(
e.g., MoO
Most recently, we successfully synthesized stoichiometric CrN and a series of
novel nitrogen–rich tungsten nitrides (e.g., W and W ) through a newly
formulated solid–state ion–exchange reactions between Na XO (M = Cr, W) and hBN
3
).
2
N
3
3 4
N
2
4
28,29
at pressures up to 5 GPa.
to the Mo–N system with a focus on nitrogen–rich phases, leading to the discovery of
a novel nitride, MoN . Here we report the high–P synthesis, structural
In this work, we have extended this high–P methodology
2
characterization and catalytic property measurements of this novel nitride compound.
Experimental Section
Starting materials, synthesis, and purification
High–purity Na
2
MoO
4
(> 99.5%, ~50 ꢁm) and hBN (> 99.9%, ~50 ꢁm) powders in
the molar ratio Na
2
MoO
4
: BN = 1 : 2 were homogeneously mixed for the synthesis of
molybdenum nitrides. Na
2 4
MoO was obtained by dehydration of commercially
available Na MoO ·2H O in a muffle furnace at 473 K for 12 hours.
2
4
2
High–P–T synthetic experiments were carried out using a DS 6 × 14 MN cubic
30
press at Sichuan University, China. In each experiment, the powder mixture was
compacted into a cylindrical pellet of 12 mm in diameter and 10 mm in height, which
4
ACS Paragon Plus Environment

Page 5 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
was then sealed within a molybdenum capsule to prevent possible contamination. The
30
detailed experimental procedures have been described elsewhere. The run products
were washed with distilled water to remove the byproduct NaBO and unreacted
Na MoO , followed by drying in an oven at 348 K. To obtain nitrogen–rich nitride,
2
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
the synthesis was performed at a relatively low temperature of 773 K to avoid
potential high–T degassing. For each experimental run, the reaction (1) was
incomplete even after a prolonged heating of 8–10 hours, and the insoluble hBN
impurity was identified in the purified sample (see Suppl Fig. S1). Therefore, a two–
step reaction route was adopted to eliminate hBN: the purified run products from the
first–step reaction, the nitride and unreacted hBN, were mixed with additional
2 4
Na MoO in a weight ratio of ~ 1 : 2 for the second–step synthesis.
Characterization
The final products were characterized by x–ray diffraction (XRD) with Cu Kα
radiation and time–of–flight powder neutron diffraction performed at the beamline of
Nanoscale–Ordered Materials Diffractometer (NOMAD) of SNS, Oak Ridge National
31
Laboratory. The crystal structure was determined from Rietveld analysis of the x–
32
ray and neutron diffraction data using the GSAS software.
Thermogravimetric mass spectrometer (TG–MS) measurements were carried out in
an Ar atmosphere on a Setaram TGA–92 instrument equipped with an OmniStar–200
quadrupole mass spectrometer, which was used to qualitatively monitor the outlet
gases at high temperatures. In air, a simultaneous TG–DTG (derivative
thermogravimetry) measurement was performed to study the oxidation of the new
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
nitride. For each experimental run, phase–pure nitride powders (~10 mg) were loaded,
ꢀ1
and automatically heated to a target temperature of 1073 K at a rate of 10 K·min and
soaked for 30 min before quenching the sample to room temperature. For the TG–MS
measurement, the mass–to–charge ratios, m/z, were qualitatively recorded to monitor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
2
and nitrogen oxide gases, such as N
nitrogen content, x, in the obtained MoN
oxidation of MoN
2
O, NO, and NO
2
, for possible oxidation. The
x
, was determined by the TGA–DTG via
x
.
Microstructures of the purified sample were characterized by a field emission
scanning electron microscope (SEM) and high–resolution transmission electron
microscopy (HRTEM). Energy–dispersive x–ray (EDX) was employed for
compositional analyses and the selected area electron diffraction (SAED) was carried
out to confirm the crystal structure. X–ray photoelectron spectroscopy (XPS)
measurements were performed to determine the electronic structure of the new phase.
Using the Archimedes method, the density of the sample was measured on a bulk
sample sintered at 3.5 GPa and 753 K for 8 hours.
The specific surface area and pore structures of the nitride catalyst were determined
from physisorption of N
2
by using the BET and BJH methods at 77 K and low relative
33,34
pressure (P/P ) of 0–0.3.
0
The measurement was performed in a Tristar–3000
instrument. Traditional MoS catalyst was also measured for comparison. Before
2
measurement, the two catalysts were treated at 473 K in a vacuum for 6–8 hours to
remove possible impurity gases adsorbed from the air.
The temperature–programmed desorption (TPD) measurement of the pre–
6
ACS Paragon Plus Environment

Page 7 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
adsorbed CO gas was used to determine the concentration of active Mo atoms in the
catalysts. Experiment was conducted in a TP–5080 quartz microreactor. To remove
the possible adsorbed impurity gases, the samples (~100 mg) were treated at 623 K
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
for 30 min in a He atmosphere with a flow rate of 30 mL·min , followed by cooling
the sample to 323 K over a period of 60 min. Before desorption, chemisorption of CO
was carried out at 323 K in a He atmosphere with a 0.4% volume fraction of CO. The
ꢀ1
mixed gases were flowed for 60 min at a flow rate of 50 mL·min to have a sufficient
adsorption. In order to eliminate the physisorbed CO, the catalyst was then flushed in
ꢀ1
a He flow at 323 K for 60 min with a flow rate of 30 mL·min . The subsequent
desorption process of CO was monitored at temperatures up to 673 K with a heating
ꢀ
1
rate of 10 K·min . In all these processes, the volume of the desorbed CO was
determined by a quadrupole mass spectrometer, OmniStar GSD–320, which was pre–
calibrated by using standard mixed gases of He and CO (see ref. 35 for detail) and the
calibrated parameter is ~13.147 mole CO per unit area.
The catalytic hydrodesulfurization (HDS) of dibenzothiophene (DBT) using
MoN
2
and MoS
2
catalysts was carried out in a high–P batch reactor/autoclave at 623
atmosphere. Before the experiment, DBT was dissolved into
K and 6 MPa in H
2
decalin as the feedstock in a weight fraction of 500 ppm and the total feed oil is 90.0 g
with a volume of 101.1 mL. For each experimental run, 0.3 g catalyst was placed in
the reactor with a stirring speed of 300 rpm. At 623 K, the reaction was followed for
48 hours by sampling the products every 8 hours. Sour methanation of the syngas was
performed in a continuous–flow and fixed–bed reactor. The nitride sample used in the
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
experiment was 0.25 g, which was sandwiched between quartz fibers in the middle of
the reactor. A syngas mixture of H
reactants in a molar ratio of H : CO : H
experiment was conducted at 3 MPa and 723 K with an hourly space velocity of 5000
2 2 2
, CO, H S, and N was used as the starting
2
2
S : N
2
= 47.3 : 47.3 : 1.2 : 5.2. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ1
h . For both HDS and methanation experiments, the outlet gases were quantitatively
analyzed by an online gas chromatograph (see refs 36 and 37 for details).
Results and discussion
Fig. 1a shows an XRD pattern of the purified product synthesized at 3.5 GPa and
7
53 K for 20 hours. As discussed below, MoN was formed through the following
2
high–P solid–state reaction,
Na MoO
which may simply be viewed as an ion–exchange process between Mo and B . We
2
4
+ 2BN = MoN
2
+ 2NaBO
2
(1).
6+
3+
2 3
have also prepared W N and CrN at high P–T conditions through similar
28,29,38
reactions.
In all the experimental runs performed, reaction (1) was incomplete
even after a prolonged heating of 8–10 hours (see Suppl Fig. S1), indicating that the
reaction is kinetically sluggish. The nitride sample can be purified by washing the
2 4 2
unreacted Na MoO and reaction by–product, NaBO , with water, leaving behind
hBN as the only impurity phase in the run product. The unwanted hBN can further be
eliminated in a two–step reaction route, described in detail in the Experimental
Section, leading to a phase–pure nitride after a second washing. The obtained run
product displays a number of Bragg reflections that do not belong to any of the known
molybdenum nitrides such as Mo
2
N, Mo
3
N
2
, MoN, Mo
5
N
6
, and their solid solutions
8
ACS Paragon Plus Environment

Page 9 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
7,19,39
(Fig. 1a).
Instead, the diffraction peaks can be indexed by a rhombohedral
structure with the lattice parameters a = 2.854 (1) Å and c = 15.938 (2) Å. Further, the
40
5
XRD pattern of this new nitride resembles those of WN
2
, 3R–MoS
2 2
, and LiXN (X
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
1,42
ꢀ
=
W and Mo)
with a symmetry of either R3m (No. 166) or R3m (No. 160). In most
6
molybdenum and tungsten nitrides, trigonal–prismatic coordination [XN ] (X = Mo
38,41,42
ꢀ
and W) is preferably adopted for X atoms.
On the other hand, the R3m structure
only contains the octahedral coordination [XN
6
]. Therefore, the space group R3m (No.
1
60), in which X atoms reside in the trigonal prismatic sites, was used for Rietveld
analysis of the XRD data.
In the R3m structure, three crystallographically distinct positions are allowed for
43
Mo with multiplicities of 3, 9, and 18. For most reported binary TM–N compounds,
as a constraint, their densities typically cannot exceed those of their parent metals .
Thus the Wyckoff site 3a (0, 0, z), which has the lowest number of multiplicity, is the
only possible position for Mo in this structure. Because the XRD intensities of
molybdenum nitride primarily originate from x–ray scattering of Mo, N was omitted
from the structure model during the initial Rietveld analysis. The thus–refined
position for Mo is 3a (0, 0, z) with z = 0. Since further treatment of the as–purified
nitride at 3.5 GPa and 1523 K led to its decomposition into a known hexagonal δ–
MoN (see Suppl Fig. S6), the new nitride phase must be nitrogen–rich with y : x > 1
x y
in Mo N . However, as an upper limit, the x : y ratio should not be greater than 2
because the highest possible oxidation state for Mo is +6. In addition, two
independent 3a (0, 0, z) sites are allowed for N to achieve the experimental density of
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ꢀ
3
ꢁꢂꢃ≈ 4.97 g·cm , determined using a well–sintered bulk sample (see Suppl Fig. S8
and Table S2). Based on these considerations, the stoichiometry of this new nitride
compound is determined to be Mo
with the nitrogen concentration obtained from thermogravimetric measurement, x ≈
.2 (3) in MoN , which will be discussed latter.
To further resolve the crystal structure, we conducted powder neutron diffraction
3 6 2
N (i.e., MoN ). This formula is in good agreement
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
x
experiment on the purified run product. Unlike x–rays, the N atom is an excellent
neutron scatterer with a coherent scattering length of ~9.36 fm, which is ~40% larger
4
4
than that of Mo (~6.72 fm). Therefore, the atomic positions of nitrogen can be well
determined from Rietveld analysis of neutron diffraction data (Fig. 1b). The thus–
obtained coordinates for nitrogen are 3a (0, 0, z) with z = 0.258(1) and z = 0.402(2),
indicating that Mo ions are in trigonal–prismatic coordination, [MoN
Fig. 1). The refined structural parameters of MoN are summarized in Table 1. As
shown in Fig. 1, the calculated x–ray and neutron patterns using this structure model
6
] (see inset of
2
2
agrees well with the observed patterns. Because the resolved structure for MoN is
5
isotypic with that of rhombohedral 3R–MoS
2
, we denote it as 3R–MoN
2
. Evidently, a
subset of diffraction peaks in Fig. 1, such as 104, 015, 107, and 018, exhibit
significant broadening, whereas other peaks (such as 003, 101, and 110) remain sharp
(see Suppl Fig. S3). In addition, some reflections in the low d–spacing (i.e., large 2θ)
region either are too weak to be seen (e.g., 009 and 116) or show significant reduction
in the diffraction intensity (e.g., 107 and 018). These phenomena, frequently occurred
in layer–structured systems in nanocrystalline forms such as nanocrystalline
10
ACS Paragon Plus Environment

Page 11 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
4
5ꢀ48
MoS
2
,
can mainly be attributed to the layer stacking disorder and atomic disorder
5
,45,49
on the surface of nanocrystals.
MoN is also in a nanocrystalline form with layer stacking disorder. It is worthwhile
to mention that 3R–MoN and MoS have almost the same core–level Mo3d and
As will be discussed below, as–prepared 3R–
2
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
Mo3p binding energy (see x–ray photoelectron spectroscopy measurement in Suppl
Figs S4–S5). The determined valence state for Mo is +3.5 in both 3R–MoN and
2
2
.5
MoS
2
, inferring a peculiar 4d electronic structure. Note that a non–integral valence
50
state (< +4) has previously been reported in MoS
2
. Further computational work is
warranted to quantify and understand the interesting electronic structures in 3R–
MoN
2
.
The microstructures obtained from high–resolution transmission electron
microscopy (HRTEM) are shown in Fig. 2. As expected for a layered compound, the
grains of 3R–MoN
2
exhibit a laminated morphology with a crystallite size of ~100–
2
00 nm and an edge thickness of ~10–25 nm (see Fig. 2a). Fig. 2b shows the atomic–
scale crystalline edge. Along the c–axis, the Mo atoms show a fringe pattern with a
distance of ~0.53 nm, which corresponds to the (003) plane (also see Table 1). The
crystalline edge is also nano–twinned and nano–twisted, corresponding to layer
stacking disorder of the N–Mo–N interlayers. As shown in Fig. 2c, (101) lattice
fringes with a d–spacing of ~0.25 nm are clearly revealed; this dꢀspacing is close to
the refined value of 2.4716 Å (i.e., ꢄ√3/2, see Table 1). Consistent with the resolved
2
structure for 3R–MoN , the HRTEM image (inset in Fig. 2c) shows a clear hexagon
formed by Mo atoms. Also noted is that the crystals along (101) plane show a ribbon–
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
like morphology with distinct grain boundaries. Fig. 2d shows a selectedꢀarea electron
diffraction pattern; four diffraction rings are identified, and they correspond to the
planes (003), (101), (012), and (104).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Thermal stabilities of the new nitride were studied by simultaneous
thermogravimetry  derivative thermogravimetry (TG–DTG), as shown in Fig. 3. In
air, there is an obvious weight increase at ~610 K, indicating the onset of oxidation,
which is comparable to that of MoS
complete at ~780 K when a plateau value is reached, and the final product is MoO
see Suppl Fig. S7). The nitrogen concentration in the nitride can be readily
2
(~623 K). Evidently, the oxidation process is
3
(
38
determined from thermogravimetric data, and the obtained value is x = 2.2(3) in
MoN . The sublimation of MoO happens at ~1000 K as the weight decreases steeply.
As shown in Fig. 3b, in an Ar atmosphere, the nitrogen degassing of 3R–MoN starts
x
3
2
at ~850 K, which gives rise to a more stable product of hexagonal δ–MoN (see Suppl
51
Fig. S6) via the reaction
MoN
The desorption of NO and N
which results from the reaction between MoN
indicates that the NO chemisorption analysis, a common method used for the
2
= MoN + 1/2 N
O occurs in the temperature region of ~400–450 K,
and O adsorbed on the surface. This
2
(2).
2
2
2
5
2
determination of coordinately unsaturated Mo sites of MoS –based catalysts, cannot
2
be used for 3R–MoN
2
, which is discussed below in more detail.
The catalytic properties of 3R–MoN
2
were measured in hydrodesulfurization (HDS)
of dibenzothiophene (DBT), a common sulfur–bearing compound in diesel fuels that
12
ACS Paragon Plus Environment

Page 13 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
is more difficult to be hydrodesulfurized compared with thiols and thiophenes. For
comparison, traditional catalyst MoS and 3R–MoN containing BN impurity were
2
2
also tested. For Mo–based catalysts, the surface area with the coordinately unsaturated
Mo sites is catalytically active centers. Equally important is the pore structure of the
catalyst, which also affects the surface area. For a given catalyst, the distribution of
pore sizes may be such that the internal surface area is completely inaccessible for
large reactant molecules and, which may restrict the rate of conversion to products by
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
53,54
impeding the diffusion of reactants and products throughout the porous medium.
We examined the surface area and pore size for 3R–MoN and MoS using the BET
2
2
and BJH analyses of N
(P/P ) of 0–0.3 (see Suppl Fig. S11). The obtained data are listed in Table 2. For
both catalysts, the measured specific surface area is ~3–10 m ·gcat , ~10–50 times
smaller than those reported for the promoted and supported MoS –based
Similar differences in the pore sizes are also found. Such small
2 2
physisorption performed at 77 K and low relative N pressure
3
4
0
2
ꢀ1
2
7,36,55,56
55
catalysts.
values are unfavorable for achieving a high catalytic activity as already reported in
7,36,55,56
well–designed catalysts.
2
Compared with MoS used in our comparative study,
3R–MoN
2
possesses a larger specific surface area, although the pore size is nearly
identical (see Table 2). Based on the measured specific surface area, the catalytic
activity for MoN can be normalized and, to a first approximation, assessed relative to
2
2
the traditional MoS . For a more accurate assessment, we determined the active Mo
sites per mole of each of the two catalysts based on CO uptake measurement, i.e.,
34,52,57
chemisorption–desorption of CO molecules on the catalyst’s surface,
and
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
experiments were performed at similar conditions (see Experimental Section for
detail). Because of the structural similarity, the two catalysts show similar desorption
spectra characterized by a main peak and a weaker satellite peak (Fig. 4a),
corresponding to two different adsorption sites. These characteristics have
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ
ꢀ
theoretically been attributed to coordinately unsaturated (1010) and (1010) Mo
57
sites on the edge of hexagonally packed slabs. For 3R–MoN
2
, the main peak occurs
at a higher temperature (~440 K) than that of MoS (~420 K), indicating that CO
2
ꢀ
molecules are more tightly bounded at unsaturated (1010) Mo sites. By analysis of
the CO uptake data, the molar volume for the absorbed CO can be quantified, and
they are listed in Table 2 (also see Suppl Fig. S12). The calculated ratio between 3R–
MoN and MoS is COꢅꢆ: COꢇꢈꢉꢊ= 1.35 : 1, which is used to normalize the catalytic
2
2
activity between two catalysts. It is worthwhile to point out that an NO uptake test
was initially attempted, but the gas reacted with MoN to form N O (also see Fig. 3b).
Fig. 4b shows the results of HDS of DBT at 623 K using 3R–MoN and MoS .
2
2
2
2
Strikingly, at the early stages of HDS, the DBT conversion using phase–pure 3R–
MoN is more than four times higher than the conversion using MoS . Overall, after
two days of the reaction, ~97% DBT conversion was achieved using 3R–MoN
whereas only 50% conversion was attained using MoS . The catalytic activity can
2
2
2
,
2
more generally be described by a pseudo–first order rate constant, k, for the HDS of
DBT, given by
ꢍ
ꢐ∙(ꢑꢒꢓ
ꢔꢓꢕꢖ
)
ꢎ
ꢋꢌ ꢍ = −ꢃ
∙ ꢗ
(3)
ꢏ
where c
t
/c
0
denotes the ratio of DBT concentration between reaction time t and t = 0,
14
ACS Paragon Plus Environment

Page 15 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
V
Oil is the volume of the feed solution (i.e.,101.1 mL DBT and decalin oil), and nCO is
the number of active Mo sites, which can be calculated from analysis of the CO
uptake data (see Table 2) and also serves as a normalization parameter the k value.
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
ꢀ1
The obtained value for MoN is k = 0.141 L ·(mol ) ·s , which is more than three
2
Oil
CO
ꢀ1
ꢀ1
times higher than that of MoS
2
(~0.044 LOil·(molCO) ·s ). Therefore, 3R–MoN
2
is
substantially more catalytically active than the traditional MoS
due to unsaturated Mo atoms that are more energetically enhanced at (1010) sites in
R–MoN (see Fig. 4a). It is worth mentioning that, to determine the intrinsic
catalytic activity, neither promoter nor support was used for 3R–MoN and MoS .
2
catalyst, presumably
ꢀ
3
2
2
2
2
However, the catalytic activity of 3R–MoN is significantly deteriorated in the
presence of hBN impurity (see XRD in Suppl Fig. S9), indicating that non–metallic
impurities would lead to the reduction of active sites for the catalytic reaction. On the
2
other hand, as shown in Fig. 4d, the BN–MoN catalyst exhibits an enhanced catalytic
activity for the hydrogenation (HYD) of DBT to form sulfur–free products of
cyclohexylbenzene (CHB) and bicyclohexyl (BCH) (see Suppl Fig. S10). For direct
desulfurization (DDS), biphenyl (BP) is the only product. Compared with MoS
BN–MoN catalyst is 6–12 times more selectivity to HYD. Besides, the phase–pure
R–MoN exhibits selectivity to HYD that is 2–7 times higher than MoS , close to the
case of the promoted and supported Ni–MoS /Al O , a widely used commercial
2
, the
2
3
2
2
2
2
3
58
2 2
catalyst. For both BN–MoN and phase–pure MoN , the reported ranges of
selectivity reflect their sensitivity to reaction time (Fig. 4d), which may be associated
with the DBT concentrations, as the latter decrease with increasing reaction time.
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Clearly, 3R–MoN
2
is an intrinsic supercatalyst for the hydrogenation, which is highly
attractive from the perspectives of ultra–deep hydrodesulfurization of fossil fuels.
The catalytic activities were also evaluated using MoN as a catalyst in the sour
methanation of syngas, a synthesis gas consisting primarily of CO, H , and sour gases
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
such as H
2
S, which often causes catalyst poisoning when traditional Ni–based
2
catalysts were used. Our preliminary results of sour methanation of syngas are shown
in Fig. 5a. At the early stages of the reaction, more than 80% conversion was achieved
for both H
2 2
and CO at 723 K in a H S atmosphere. As the reaction products, the outlet
gases of CH and CO were detected in a molar ratio CH : CO ≈ 1 : 1, indicating a
4
2
4
2
sour methanation process given by,
2H + 2CO = CH
The observed high catalytic activity of 3R–MoN
2
4
+ CO
2
(4).
2
for sour methanation is comparable
2
to that obtained recently using a Zr–promoted MoS
2
catalyst, and is significantly
37
higher than the catalytic activities of Co–MoS /γ–Al O . However, the reaction (4)
2
2
3
lasted for only 1.4 hours after which the conversion for both H
drastically. Concurrently, as shown in Fig. 5a, the excessive N
reaction proceeds. Most likely, this is due to the degassing of 3R–MoN
which converts 3R–MoN to hexagonal δ–MoN. As the N output is stabilized at the
input level, H and CO are converted into CH in a molar ratio of H : CO ≈ 3 : 1,
2
and CO was lowered
was released as the
at 723 K,
2
2
2
2
2
4
2
which is typical of a traditional methanation reaction,
3H + CO = CH + H (5).
2
4
2
O
Obviously, 3R–MoN
2
is catalytically more active and efficient than δ–MoN in the
16
ACS Paragon Plus Environment

Page 17 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
methanation of syngas. The 3R–MoN
2
and δ–MoN catalysts, however, show similar
selectivity for CH
4
as seen in Fig. 5b. In practice, the syngas can be produced by
gasifying coal and biomass. The sour methanation is an essential process for
upgrading such low–grade energy sources. Search for new catalysts with high
poisoning resistance is highly demanded from the perspectives of enhanced process
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
and energy efficiency. As a candidate catalyst, the 3R–MoN
future sour methanation applications.
2
holds great promise for
To the best of our knowledge, 3R–MoN
2
is the first binary nitride having a
rhombohedral MoS structure and is only the fifth binary nitride in bulk forms
2
exhibiting the stoichiometry of XN
2
, after noble metal nitrides ReN
As expected, the layer–structured 3R–MoN is intrinsically an excellent
catalyst. It is sulfur–poisoning tolerant and exhibits superior catalytic activities over
the traditional MoS –based catalysts for the HDS of DBT, even without the presence
of a promoter or a supporting substrate. It is conceivable that the discovery of 3R–
MoN may open new opportunities for developments of highly active, MoN –based
catalysts through the introduction of promoter species (e.g., Co, Ni, and Zr) and
2 2 2
, OsN , IrN , and
2
5,27,59
PtN
2
.
2
2
2
2
supporting substrates (e.g., Al
2
O
3
and SiO
2
), as already demonstrated for the MoS
2
1,2
catalysts. As another nitride of the group 6 metals, WN
2
may structurally be isotypic
with 3R–MoN and hence a promising catalyst candidate, although an R3 structure
2
33
has previously been reported. The crystal structure of WN
2
, however, has not been
well resolved because the sample used in ref. 33, prepared by evaporating W
2
filaments in atmospheres of N , is poorly crystalline. More studies are required to
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2 2 2
resolve the structure of WN . The ternary nitrides of LiMoN and LiWN can be
another group of materials with good catalytic activities. They also crystallize in a
R3m structure, which can simply be viewed as the intercalation of Li ions into the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
2,60
interlayer of XN to form octahedrally coordinated [LiN ].
These ternary nitrides
are thermodynamically more favorable than binary 3R–MoN . For example, at
atmospheric pressure the degassing temperature of LiMoN (~1023 K) is ~20%
higher than that of 3R–MoN (~850 K). This high thermal stability may make them
2
6
2
4
1
2
2
robust catalyst in high–temperature catalytic application.
Conclusions
In summary, we have successfully synthesized a novel molybdenum nitride
2
compound, 3R–MoN , which has the rhombohedral MoS
2
structure, through a high
2.5
P–T route of solid–state ion–exchange. 3R–MoN
2
adopts a peculiar 4d electron
configuration with a non–integral valence of +3.5 for Mo ions. Most importantly, 3R–
MoN exhibits superior catalytic activities and high hydrogenation selectivity over
2
MoS
selectivity of this nitride can be greatly improved, when the the hBN impurity is
present. 3R–MoN is also catalytically more active and efficient than δ–MoN in the
2
in the hydrodesulfurization of dibenzothiophene. Further, the hydrogenation
2
sour methanation of syngas. Hence, this compound holds great promise as the next–
generation catalyst for a wide range of applications. The new nitride can be
synthesized at a moderately high pressure of 3.5 GPa, making it feasible for massive
and industrial–scale production.
Supporting Information Available
18
ACS Paragon Plus Environment

Page 19 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Detailed sample characterizations (including XRD, SEM, EDX, NPD, and XPS
measurements) and measurements on phase stability, density, and catalytic properties .
This information is available free of charge via the internet at http://pubs.acs.org.
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
Acknowledgments
This work was partially supported by UNLV High Pressure Science and Engineering Center
(HiPSEC), which is a DOE NNSA Center of Excellence operated under Cooperative Agreement
DE–FC52–06NA27684, and UNLV start–up funding to Y. Zhao. The work was also partially
supported by the NSF of China (No. 21473231). Initial work was supported by Los Alamos
National Laboratory, which is operated by Los Alamos National Security LLC under DOE
Contract DE–AC52–06NA25396. Use of the ORNL’s Spallation Neutron Source at B1 beamline
(NOMAD) was sponsored by the Scientific User Facilities Division, DOE–BES.
Corresponding Authors
*
Eꢀmails: ShanminWang@gmail.com (S. Wang), gehui@sxicc.ac.cn (H. Ge), and Yusheng.Zhao@UNLV.edu
Y. Zhao)
(
The authors declare that they have no competing financial interests.
References:
(
1) Song, C. Catal. Today 2003, 86, 211.
(2) Anand, M.; Sughrue, E. L.; Yao, J.; Patents: 2012.
3) Voiry, D.; Yamaguchi, H.; Li, J.; Silva, R.; Alves, D. C. B.; Fujita, T.; Chen, M.; Asefa, T.; Shenoy, V.
B.; Eda, G.; Chhowalla, M. Nature Mater. 2013, 12, 850.
4) Bernardi, M.; Palummo, M.; Grossman, J. C. Nano Lett. 2013, 13, 3664.
(5) Wang, S.; Zhang, J.; He, D.; Zhang, Y.; Wang, L.; Xu, H.; Wen, X.; Ge, H.; Zhao, Y. J. Phys. Chem.
Solids 2014, 75, 100.
6) Jaramillo, T. F.; Jørgensen, K. P.; Bonde, J.; Nielsen, J. H.; Horch, S.; Chorkendorff, I. Science 2007,
17, 100.
(7) Tye, C. T.; Smith, K. J. Top. Catal. 2006, 37, 129.
(
(
(
3
(
8) Huang, Z. D.; Bensch, W.; Kienle, L.; Fuentes, S.; Alonso, G.; Ornelas, C. Catal. Lett. 2008, 122, 57.
9) Song, C.; Ma, X. In Practical Advances in Petroleum Processing; Hsu, C., Robinson, P., Eds.; Springer
(
New York: 2006, p 317.
(10) Abe, H.; Cheung, T. K.; Bell, A. T. Catal. Lett. 1993, 21, 11.
(
(
(
11) Tabata, M.; Maeda, K.; Higashi, M.; Lu, D.; Takata, T.; Abe, R.; Domen, K. Langmuir 2010, 26, 9161.
12) Chen, W.ꢀF.; Muckerman, J. T.; Fujita, E. Chem. Commun. 2013, 49, 8896.
13) Zhao, Y.; Kamiya, K.; Hashimoto, K.; Nakanishi, S. J. Am. Chem. Soc. 2015, 137, 110.
(14) Neylon, M. K.; Choi, S.; Kwon, H.; Curry, K. E.; Thompson, L. T. Appl. Catal., A 1999, 183, 253.
(
9186.
(
15) Cao, B.; Veith, G. M.; Neuefeind, J. C.; Adzic, R. R.; Khalifah, P. G. J. Am. Chem. Soc. 2013, 135,
1
16) Pierson, H. O. In Handbook of Refractory Carbides & Nitrides : Properties, Characteristics,
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Processing and Applications; William Andrew: Burlington, 1996.
17) Machon, D.; Daisenberger, D.; Soignard, E.; Shen, G.; Kawashima, T.; TakayamaꢀMuromachi, E.;
McMillan, P. F. Phys. Status Solidi A 2006, 203, 831.
18) Troitskaya, N. V.; Pinsker, Z. G. Sov. Phys. Crystallogr. 1959, 4, 33.
(19) Ganin, A. Y.; Kienle, L.; Vajenine, G. V. J. Solid State Chem. 2006, 179, 2339.
(
(
(
(
(
20) Salamat, A.; Hector, A. L.; Kroll, P.; McMillan, P. F. Coord. Chem. Rev. 2013, 257, 2063.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
21) Zerr, A.; Miehe, G.; Riedel, R. Nature Mater. 2003, 2, 185.
22) Salamat, A.; Hector, A. L.; Gray, B. M.; Kimber, S. A. J.; Bouvier, P.; McMillan, P. F. J. Am. Chem. Soc.
2013, 135, 9503.
(23) Zerr, A.; Miehe, G.; Li, J.; Dzivenko, D. A.; Bulatov, V. K.; Hoefer, H.; BolfanꢀCasanova, N.; Fialin,
M.; Brey, G.; Watanabe, T.; Yoshimura, M. Adv. Funct. Mater. 2009, 19, 2282.
(24) Salamat, A.; Woodhead, K.; Shah, S. I. U.; Hector, A. L.; McMillan, P. F. Chem. Commun. 2014, 50,
10041.
(25) Crowhurst, J. C.; Goncharov, A. F.; Sadigh, B.; Evans, C. L.; Morrall, P. G.; Ferreira, J. L.; Nelson, A. J.
Science 2006, 311, 1275.
(26) Young, A. F.; Sanloup, C.; Gregoryanz, E.; Scandolo, S.; Hemley, R. J.; Mao, H.ꢀk. Phys. Rev. Lett.
2
006, 96, 155501.
(
27) Kawamura, F.; Yusa, H.; Taniguchi, T. Appl. Phys. Lett. 2012, 100, 251910
28) Wang, S.; Yu, X.; Zhang, J.; Chen, M.; Zhu, J.; Wang, L.; He, D.; Lin, Z.; Zhang, R.; Leinenweber, K.;
(
Zhao, Y. Phys. Rev. B 2012, 86, 064111.
(
(
(
29) Chen, M.; Wang, S.; Zhang, J.; He, D.; Zhao, Y. Chem. Eur. J. 2012, 18, 15459.
30) Wang, S.; He, D.; Wang, W.; Lei, L. High Pressure Res. 2009, 29, 806.
31) Neuefeind, J.; Feygenson, M.; Carruth, J.; Hoffmann, R.; Chipley, K. K. Nucl. Instrum. Methods Phys.
Res., Sect. A 2012, 287, 68.
(
(
(
32) Toby, B. H. J. Appl. Cryst. 2001, 34, 210.
33) Langmuir, I. J. Am. Chem. Soc. 1913, 35, 931.
34) Brunauer, S.; Emmett, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309.
(35) Burch, R.; Collins, A. Applied Catalysis 1985, 17, 273.
36) Alvarez, L.; Espino, J.; Ornelas, C.; Rico, J. L.; Cortez, M. T.; Berhault, G.; Alonso, G. J. Mol. Catal. A:
Chem. 2004, 210, 105.
(
(37) Jiang, M.; Wang, B.; Yao, Y.; Li, Z.; Ma, X.; Qin, S.; Sun, Q. Catal. Sci. Technol. 2013, 3, 2793.
(38) Wang, S.; Yu, X.; Lin, Z.; Zhang, R.; He, D.; Qin, J.; Zhu, J.; Han, J.; Wang, L.; Mao, H.ꢀK.; Zhang, J.;
Zhao, Y. Chem. Mater. 2012, 24, 3023.
(39) Karam, R.; Ward, R. Inorg. Chem. 1970, 9, 1385.
(40) Wriedt, H. A. Bull. Alloy Phase Diagrams 1989, 10, 358.
(41) Elder, S. H.; Doerrer, L. H.; Disalvo, F. J.; Parise, J. B.; Guyomard, D.; Tarascon, J. M. Chem. Mater.
992, 4, 928.
42) Herle, P. S.; Hegde, M. S.; Vasanthacharya, N. Y.; Gopalakrishnan, J.; Subbanna, G. N. J. Solid State
Chem. 1994, 112, 208.
(43) Hahn, T. International Tables for Crystallography, Brief Teaching Edition of Volume A: Space Group
Symmetry; Springer, 2002.
1
(
(
44) Sears, V. F. Neutron News 1992, 3, 26.
45) Liang, K. S.; Chianelli, R. R.; Chien, F. Z.; Moss, S. C. J. NonꢀCryst. Solids 1986, 79, 251.
(
(46) Peng, Y.; Meng, Z.; Zhong, C.; Lu, J.; Yu, W.; Yang, Z.; Qian, Y. J. Solid State Chem. 2001, 159, 170.
20
ACS Paragon Plus Environment

Page 21 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(47) Chianelli, R. R.; Prestridge, E. B.; Pecoraro, T. A.; Deneufville, J. P. Science 1979, 203, 1105.
(
(
(
48) Wang, S.; Wang, Z.; Qin, J.; Wang, W.; Li, W.; He, D. Mater. Chem. Phys. 2011, 130, 170.
49) Warren, B. E. Phys. Rev. 1941, 59, 693.
50) Bowden, F. L.; Heveldt, P. F.; Watson, D. J. In Inorganic Chemistry of the Transition Elements: Volume
6; Johnson, B. F. G., Ed.; The Royal Society of Chemistry: 1978; Vol. 6, p 1.
(
51) Bull, C. L.; McMillan, P. F.; Soignard, E.; Leinenweber, K. J. Solid State Chem. 2004, 177, 1488.
52) Maugé, F.; Lamotte, J.; Nesterenko, N. S.; Manoilova, O.; Tsyganenko, A. A. Catal. Today 2001, 70,
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
71.
(53) Song, C.; Nihonmatsu, T.; Nomura, M. Ind. Eng. Chem. Res. 1991, 30, 1726.
(
54) Shimura, M.; Shiroto, Y.; Takeuchi, C. Ind. Eng. Chem. Fundam. 1986, 25, 330.
55) Valencia, D.; Klimova, T. Appl. Catal. B: Environ. 2013, 129, 137.
(
(56) Huang, Z. D.; Bensch, W.; Kienle, L.; Fuentes, S.; Alonso, G.; Ornelas, C. Catal. Lett. 2009, 127, 132.
(
(
(
57) Zeng, T.; Wen, X.ꢀD.; Li, Y.ꢀW.; Jiao, H. J. Phys. Chem. B 2005, 109, 13704.
58) Cho, A.; Lee, J. J.; Koh, J. H.; Wang, A.; Moon, S. H. Green Chem. 2007, 9, 620.
59) Young, A. F.; Sanloup, C.; Gregoryanz, E.; Scandolo, S.; Hemley, R. J.; Mao, H. K. Phys. Rev. Lett.
2006, 96.
60) Maoujoud, M.; JardinierꢀOffergeld, M.; Bouillon, F. Appl. Surf. Sci. 1993, 64, 81.
(
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
FIGURES
a
XRD
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Obs.
Calc.
Diff.
Peaks
10
20
30
40
50
60
70
80
2theta (deg)
b
NPD
0.5
1.0
1.5
2.0
2.5
dꢀspacing (Å)
1
2
3
4
5
6
dꢀspacing (Å)
Fig. 1. Refined XRD and NPD patterns of 3R–MoN . (a) XRD data collected using a
2
2
Cu Kα radiation. Inset is a polyhedral view of the 3R–MoN structure. (b) NPD data
collected at a Bragg scattering angle of 30º. A higher–resolution NPD pattern is
plotted in the inset, which was collected at a 66º scattering angle. The observed and
calculated patterns are shown in black and red, respectively. The positions of allowed
Bragg reflections are denoted by cyan tick marks. The sample synthesis was
performed at 3.5 GPa and 753 K for 20 hours through a two–step reaction (see
Experimental Section). The misfit in the d–spacing range of 1.5–2.1 Å is mainly
22
ACS Paragon Plus Environment

Page 23 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
attributed to significant peak broadening in a subset of reflections, notably 104, 015,
1
07 and 018, which originates from the layer stacking disorder (refs 5 and 49).
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
Fig. 2. HRTEM images and electron diffraction pattern for 3R–MoN . (a) Plate–
shaped crystals with laminate edges. (b) Atomic–scale images of the N–Mo–N
interlayers, i.e., the (003) plane, and (c) the (101) plane. The inset in both (b) and (a)
are the enlarged portions of the regions highlighted in yellow. Dotted double red lines
indicate the d–spacing for the fingerprints of (003) and (101) planes. Dotted cyan
lines in (c) denote the grain boundaries. (d) Selected area electron diffraction, SAED.
The observed four diffraction rings correspond to the lattice planes of (003), (101),
(012), and (104).
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
DTG
TG
1000 K
0
.0
610 K
ꢀ
0.2
1000
800
7
MoO
3
MoN
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
x
6
5
4
6
4
00
00
In Air
3
Nꢀconcentration
x ≈ 2.2 (3)
0
20
40
60
80
100
120
b
Time (min)
N O
2
445 K
8
50 K
N
2
NO
1
8
6
4
000
00
100
MoN2
98
00
9
9
6
4
In Ar
00
0
20
40
Time (min)
60
80
Fig. 3. Thermal stability analysis. (a) Thermogravimetry/differential thermal analysis,
TG/DTG, performed in air. Oxidation of MoN starts at ~610 K, and the final product
is MoO , which starts sublimating at a temperature of ~1000 K. The derived nitrogen
2
3
concentration is x = 2.2(3) in MoN
degassing of N in an Ar atmosphere. The outlet gases are qualitatively traced during
heating. At 400 – 445 K, the surface absorbed O
The significant degassing of N happens at ~850 K.
x
. (b) TG measurement of high–temperature
2
2
is released in forms of NO and N
2
O.
2
24
ACS Paragon Plus Environment

Page 25 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
ꢀ
2
a
b
×
10
7
6
5
4
00
00
00
00
100 HDS of DBT at 623 K
Peak 1
~
440 K
6
8
0
~420 K
3
RꢀMoN
MoS
Peak 2
2
60
40
4
2
0
2
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
3R
~
670 K
3R + hBN
MoS
2
0
2
300
0
1000
Time on stream (s)
2000
0
10 20 30 40 50
Reaction time (hours)
c
d
1
00
80
3R
3R + hBN
3
MoS
DDS
2
6
0
0
3R
R + hBN
MoS
2
1
0
3
4
2
HYD
20
0
.0
4
4
0
0
10 20 30 40 50
Reaction time (hours)
0
10 20 30 40 50
Reaction time (hours)
Fig. 4. Catalytic activities of 3R–MoN
2
in the hydrodesulfurization (HDS) of
dibenzothiophene (DBT). Traditional catalyst MoS was also tested for comparison.
2
(
a) Temperature–programmed desorption (TPD) of CO to determine the amounts of
active Mo sites. Before experiment, the catalysts were pre–exposed to CO gas at 323
K for 60 min to absorb CO molecules and form Mo–CO bonds with active Mo sites,
ꢀ
ꢀ
including coordinately unsaturated (1010) and (1010) Mo sites. (b) HDS of DBT
at 623 K and 6 MPa in H atmosphere. (c) Relative DBT concentration, c /c , as a
2
t
0
function of reaction time t. The normalized pseudo–first order rate constant, k, was
deduced by linear fits. (d) The measured selectivity of direct desulfurization (DDS)
and hydrogenation (HYD) in the HDS of DBT (see Suppl Figs S9–S14).
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a 100
3
RꢀMoN
Sour methanation
At 723 K
2
80
20
60
δꢀMoN
10
N out
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
40
N in
2
H conversion
2
0
20
CO conversion
0
0
1
2
3
4
5
b 60
3
RꢀMoN
δ
ꢀMoN
2
4
0
0
0
CH
4
C H
2
6
2
C H
3 8
0
1
2
3
4
5
Time on stream (hours)
Fig. 5. Catalytic performance of 3R–MoN in sour methanation. (a) Conversion of H
and CO as a function of reaction time at 723 K. Volume fractions of the input and
output N gas are plotted to show the high–T degassing of MoN and the subsequent
formation of δ–MoN. (b) The measured selectivity of CH , C H and C H in the sour
2
2
2
2
4
2
6
3
8
methanation. No promoters or supporting substrates were used in the catalytic
measurements.
26
ACS Paragon Plus Environment

Page 27 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Table 1. Summary of structural parameters for MoN
2
phases determined by Rietveld
analyses of x–ray and neutron diffraction data taken at ambient conditions.
3
R-MoN
2
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
Formula
Mo
3 6
N
Space group
rhombohedral , R3m (No. 160)
a = b = 2.854 (1), c = 15.938 (2)
112.426
Cell parameters [Å]
Cell volume [Å3]
ꢀ3
Density [g•cm ]
Mo Wyckoff site
N Wyckoff sites
5.492
Mo1: 3a, (0, 0, 0.000 (1))
N1: 3a, (0, 0, 0.258 (1)), N2:3a, (0,0,0.402 (2))
d
MoꢀN [Å]
1.976, 2.037
5.313 (1)
3.017 (1)
3.2, 5.3
D
D
R
Interlayer [Å]a
Layer distance [Å]b
P
, wR [%] (NPD)
P
a,b
D
Interlayer is the distances between the two nearest neighboring Mo planes, and
D
Layer
distance corresponds to the distance between the N planes.
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 29
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. Summary of CO uptake data, BET surface area, and pore volume and width
2 2
for 3R–MoN and MoS catalysts. Also listed are the pseudo–first order rate constant,
k, which are normalized to the active Mo sites.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catalyst
3R–MoN
2
MoS
2
Ratio 3R : MoS
2
BET surface area [m ·gcatꢀ1]
Pore volume [ꢁ·m³·gcatꢀ1]
Ave. pore width [nm]
2
8.656(9)
0.036
16.54
47.19
5.850
0.141
3.699(5)
0.016
17.14
34.97
5.598
0.044
2.34
2.25
0.96
CO uptake [ꢁ·molCO·gcatꢀ1]
1.35
CO / Mo [×10 molCO·molMoꢀ1] a
ꢀ
3
1.05
ꢀ
1
ꢀ1
k [LOil—molCO ·s ]
3.20
a
CO in the CO / Mo ratio denotes the molar volume of CO adsorbed onto the Mo
sites (i.e., coordinately unsaturated Mo sites). The ratio CO/Mo represents the
percentage of catalytically active Mo sites.
28
ACS Paragon Plus Environment

Page 29 of 29
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
TOC Abstract (“For Table of Contents Only”)
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