Highly efficient CO: x-free hydrogen evolution activity on rod Fe2N catalysts for ammonia decomposition

Article abstract of DOI:10.1039/c9nj04273c

Ammonia decomposition is a critical method for CO_x-free hydrogen production, and many metal nitrides have been studied as excellent catalysts for NH₃ decomposition recently. Here, Fe₂N catalysts with rod, cubical and discoidal structures were prepared via nitrogenization of the corresponding morphologies of Fe₂O₃ precursors, which were obtained by a hydrothermal synthesis method. These Fe₂N catalysts were characterized by XRD, TEM, BET, XPS and M?ssbauer spectroscopy, and the results showed that the morphologies of Fe₂O₃ precursors had a significant effect on the crystallite size and yield of the nitride phase of Fe₂N catalysts. The rod Fe₂N had a small crystallite size and a high relative nitrogen content among the three samples. The catalytic activity for hydrogen production via ammonia decomposition was measured using a micro-fixed bed reactor under conditions of conventional pressure and a GHSV of 6000 mL g_cat^-1 h^-1. The rod Fe₂N exhibited a higher NH₃ conversion of 90% at 823 K compared with Fe₂N derived from cubical and discoidal Fe₂O₃, and maintained high stability for 40 h.

Full text of DOI:10.1039/c9nj04273c

NJC
View Article Online
View Journal | View Issue
PAPER
Highly efficient CO -free hydrogen evolution
x
activity on rod Fe N catalysts for ammonia
2
Cite this: New J. Chem., 2019,
decomposition†
43, 18277
a
a
b
a
a
Bowen Lei, Jie Wen, * Shan Ren, Lianhong Zhang and Hui Zhang
*
Ammonia decomposition is a critical method for CO
have been studied as excellent catalysts for NH decomposition recently. Here, Fe
cubical and discoidal structures were prepared via nitrogenization of the corresponding morphologies of
Fe precursors, which were obtained by a hydrothermal synthesis method. These Fe N catalysts were
characterized by XRD, TEM, BET, XPS and Mossbauer spectroscopy, and the results showed that the
x
-free hydrogen production, and many metal nitrides
3
2
N catalysts with rod,
O
2 3
2
¨
morphologies of Fe
2
O
3
precursors had a significant effect on the crystallite size and yield of the nitride
N had a small crystallite size and a high relative nitrogen content
Received 18th August 2019,
Accepted 16th October 2019
phase of Fe N catalysts. The rod Fe
2
2
among the three samples. The catalytic activity for hydrogen production via ammonia decomposition was
measured using a micro-fixed bed reactor under conditions of conventional pressure and a GHSV of
DOI: 10.1039/c9nj04273c
catꢀ
1
ꢀ1
2 3
. The rod Fe N exhibited a higher NH conversion of 90% at 823 K compared with
6000 mL g
h
rsc.li/njc
Fe N derived from cubical and discoidal Fe O
, and maintained high stability for 40 h.
2
2 3
To date, transition metal nitrides have drawn much attention in
hydrodesulfurization (HDS), Fischer–Tropsch (F–T) processes, hydro-
1
. Introduction
18
The global community is being confronted with more and more denitrogenation (HDN) and NH
3
9–23
decomposition. Monometallic
and bimetallic nitrides (Ni–Mo–N
1
complicated environmental pollution issues due to the large con- nitrides (VN
sumption of fossil fuels. It is urgent to search for renewable and and Co–Mo–N)
clean energy sources to decrease the current level of dependence on NH decomposition. Our previous reports have shown that Fe
fossil fuels. Recently, hydrogen production, which is recognized as played an active role in the enhanced catalytic ability of Fe O /CNTs
x x x
, MoN and FeN )
24–31
all have demonstrated excellent performance for
3
2
N
3
4
1–3
one of the cleanest energy sources, has received great attention.
in ammonia decomposition and the key was the effective control of
32
Considering the high hydrogen densities (17.6 wt%) and huge the density of the Fe N active sites. Recently, Leybo et al. have
2
3 3 2
production of NH , the use of NH to produce H is a superior synthesized various Ni–Mo nitride catalysts derived from different
choice in comparison with currently used reforming processes of precursors. These precursors had a significant effect on the compo-
2
6
methane or methanol, which usually ends with potential CO
x
sition and specific surface area of Ni–Mo nitride.
The choice of the precursor greatly determined the physico-
4
3
release. However, the rate of the NH decomposition reaction is
33,34
too slow to proceed, even at a very high temperature (B923 K). chemical and catalytic properties of the final nitride product.
Hence, searching for suitable catalysts to promote NH decom- In this work, we obtained rod, cubical and discoidal structure
3
position is extremely imperative. Though noble metal catalysts Fe N nanocatalysts through nitrogenization of regular Fe O with
2
2 3
have exhibited satisfactory catalytic performance in NH
corresponding morphologies. The as-prepared nano-rod Fe N
2
3
5
–11
decomposition,
their extensive use in practical applications,
to find appropriate non-noble-metal candidates as NH
their high cost and scarcity strictly limited exhibited a higher NH
3
decomposition activity in comparison
propelling us with both cubical and discoidal Fe N. The high catalytic activity of
decom- rod Fe N was due to its small crystallite size and relatively high
nitrogen content.
1
2–14
2
3
2
1
5–17
position catalysts, such as transition metals Ni, Co, and Fe.
a
The Centre of New Energy Materials and Technology, College of Chemistry and
Chemical Engineering, Southwest Petroleum University, Chengdu 610500, China.
E-mail: huizhang@swpu.edu.cn, wenjie@swpu.edu.cn
2
. Experimental methods
b
The Institute of Functional Nano & Soft Materials (FUNSOM), Jiangsu Key
Laboratory for Carbon-Based Functional Materials & Devices, Soochow University,
Suzhou 215123, China
2.1 Chemical materials
Analytical grade aluminium subacetate (Al(OH)C
4
6 4
H O ) was
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/c9nj04273c purchased from Aldrich. Sodium hydroxide (NaOH), ferric
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18277--18284 | 18277

View Article Online
Paper
NJC
nitrate (Fe(NO
3
)
3
ꢁ9H
2
O), ammonium hydroxide solution (28%)
and ethyl alcohol were all obtained from Chengdu Kelong
Chemical Reagents Co., Ltd. All chemicals used in this work
were not further purified. Deionized water was employed in all
experiments.
2.2 Synthesis of catalysts
2
.2.1 Synthesis of rod Fe N. In a typical synthesis proce-
2
3
5
dure for a rod Fe
dissolved in 20 mL deionized water (DI) to form a homogeneous
2
O
3
precursor, Fe(NO
3
)
3
ꢁ9H
2
O (5 mmol) was
solution and the pH was gradually adjusted to 4 13 using 10 mL
3.0 M NaOH under stirring. The obtained solution was trans-
ferred to 50 mL Teflon-lined stainless-steel autoclaves and kept
at 453 K for 6 h. Then the above product was separated via
centrifugation and was washed several times with ethanol and
water. The obtained solids were vacuum dried at 333 K for 12 h
Fig. 1 TEM images of (a) Fe
e) Fe N-C and (f) Fe N-D.
2 3 2 3 2 3 2
O -R, (b) Fe O -C, (c) Fe O -D, (d) Fe N-R,
(
2
2
2 3
and then were calcined at 623 K for 4 h in air to obtain rod Fe O
(
denoted as Fe
2
O
3
ꢀ
-R). Finally, the Fe
2
O
3
-R was heated to 773 K at resolution transmission electronic microscopy (HRTEM) image
1
ꢀ1
a rate of 5 1C min and kept at 773 K for 2 h in a 10 mL min
and energy dispersive X-ray spectroscopy (EDS) mapping
NH flow to obtain rod Fe N (denoted as Fe N-R).
images were obtained using a JEM-2100F with a spherical
3
2
2
2
.2.2 Synthesis of cubical Fe N. In a typical synthesis aberration corrector operating at 200 kV. N
2
adsorption/
desorption isotherms of the samples were obtained using a
10 mmol) was dissolved in 20 mL DI and then 10 mL 0.6 M Micromeritics 2020 at 77 K. The specific surface area was
NaOH was gradually added to form a homogeneous solution calculated using a multipoint Brunauer–Emmett–Teller (BET)
under stirring, and the final pH of the solution was close to 2. method. The total pore volume was estimated from the N
The obtained solution was transferred to 50 mL Teflon-lined adsorption capacity at P/P = 0.995. The Barrett–Joyner–
2
3
5
procedure for a cubical Fe
2
O
3
precursor, Fe(NO
3
)
3
ꢁ9H
2
O
(
2
0
stainless-steel autoclaves and kept at 453 K for 12 h. Then the Halenda (BJH) method was used to measure the average pore
above product was separated via centrifugation and was washed size. X-ray photoelectron spectroscopy (XPS) was performed
several times with ethanol and water. The obtained solids were using a PHI5000 Verasa and an aluminum target as the excita-
vacuum dried at 333 K for 12 h and then were calcined at 623 K tion source. A constant acceleration spectrometer with the Co
for 4 h in air to obtain cubical Fe
Finally, the Fe -C was heated to 773 K at a rate of 5 1C min
and kept at 773 K for 2 h in a 10 mL min NH
cubical Fe N (denoted as Fe N-C).
2 3 2 3
O (denoted as Fe O
-C). g-quantum source in the Rh matrix was used to collect the
ꢀ1
2
O
3
Fe M o¨ ssbauer spectrum at room temperature. The spectra
ꢀ1
flow to obtain were fitted by the least squares fitting procedure through the
Lorentzian curve. All the isomer shifts (IS) were referenced to
3
2
2
2
.2.3 Synthesis of discoidal Fe N. In a typical synthesis the room temperature a-Fe centroid.
2
3
6
procedure for a discoidal Fe O precursor, Fe(NO ) ꢁ9H O
2
3
3 3
2
(
2 mmol) was dissolved in 10 mL DI to form a homogeneous 2.4 Catalytic performance measurement
4 6 4
solution, and the required 20 mL of 0.05 M Al(OH)C H O was
The activity measurement of catalysts for ammonia decomposi-
tion was carried out using a micro-fixed bed reactor of quartz
glass (inner diameter = 4 mm). Typically, 100 mg of catalysts
was charged into the reactor and pure ammonia gas (99.999%
added dropwise under stirring. After this, 10 mL ammonium
hydroxide solution was added dropwise. The obtained solution
was transferred to 50 mL Teflon-lined stainless-steel autoclaves
and kept at 433 K for 16 h. Then the above product was
separated via centrifugation and was washed several times with
ethanol and water. The obtained solids were vacuum dried at
NH , Jinkexing Gas Co., Ltd) was passed into the reactor with
3
ꢀ1
ꢀ1
6
000 mL gcat
h
gas hourly space velocity (GHSV). The
ꢀ1
temperature was gradually increased at a rate of 5 K min
and kept for 30 min at each reaction temperature. The reaction
products were analyzed using a gas chromatograph (SC-200)
with a Porapak Q column and a thermal conductivity detector
with Ar as the carrier gas. The ammonia conversion was
calculated using eqn (1):
3
33 K for 12 h and then were calcined at 623 K for 4 h in air
to obtain discoidal Fe (denoted as Fe -D). Finally, the
Fe -D was heated to 773 K at a rate of 5 1C min and kept at
73 K for 2 h in a 10 mL min NH
Fe N (denoted as Fe N-D).
2
O
3
2 3
O
ꢀ
1
2 3
O
ꢀ1
7
3
flow to obtain discoidal
2
2
2
.3 Characterization
À
Á
ANH ;in ꢀ ANH ;out
3
3
The crystallite structure was analyzed by Powder X-ray diffrac-
tion (XRD, Rigaku MiniFlex-600) using a copper target as the
radiation source (l = 0.15406 nm). A transmission electron
XNH3
¼
ꢂ 100%
(1)
ANH ;in
3
microscope (TEM, Hitachi-7700) was used to characterize the where ANH ,in is the total concentration of ammonia in feed gas,
3
morphologies of the samples operating at 100 kV. The high and ANH ,out is the unconverted concentration of ammonia.
3
1
8278 | New J. Chem., 2019, 43, 18277--18284 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
3. Results and discussion
Table 1 The calculation results by Scherrer’s equation
3.1 Characterization of catalysts
Samples
2y/1
Hkl
FWHM/rad
D/nm
The TEM images of the synthesized Fe O samples are shown Fe
2
2
2
N-R
N-C
N-D
42.9
42.9
43.2
101
101
101
0.195
0.127
0.230
50.8
4100.0
41.3
2
3
Fe
Fe
in Fig. 1. Fig. 1a shows that uniform Fe
structure was obtained. The average length of Fe
27 nm as seen from Fig. S1a (ESI†). The Fe -R showed large
numbers of pores, which was attributed to the dehydration
2
O
3
-R with a rod
2 3
O
-R was
2
2 3
O
2
Table 2 Structural properties of Fe N with different morphologies
3
5
effect during calcination. Fig. 1b and c show that cubical and
discoidal Fe O were obtained with slight modifications in the
Specific surface
area (m g
Pore volume
Average pore
size (nm)
2
3
2
ꢀ1
3
ꢀ1
Samples
)
(cm g
)
experimental procedure. Fig. S1b and c (ESI†) show that the
Fe
Fe
Fe
2
N-R
2
N-C
2
N-D
7.25
2.13
21.96
0.010
0.005
0.080
14.8
26.3
18.2
average sizes of Fe O -C and Fe O -D were 83 nm and 450 nm,
2
3
2 3
respectively. And then the three Fe
at 773 K under a NH flow for 2 h and Fe
formed. This conversion of Fe to Fe N was believed to be
2
O
3
precursors were heated
3
2
N species were
2
O
3
2
3
7
depicted using eqn (2).
Fe O + 4NH - 2Fe N + 6H O + N
2
2
(2)
2
3
3
2
2
The TEM images of the Fe
2
N samples are shown in Fig. 1d–f.
They indicated that the corresponding Fe O3 morphologies
2
were well maintained after the conversion, although a slight
agglomeration of Fe N particles was observed.
2
2 3
The XRD patterns of Fe O with rod, cubic and discoidal
shapes are shown in Fig. 2a. They indicated that the diffraction
peaks of the three samples were basically the same and
corresponded to the hexagonal phase Fe
The XRD patterns of Fe N with different shapes are displayed in
Fig. 2b. They clearly indicated that the diffraction peaks at 2y of
7.61, 40.81, 43.01, 56.81, 67.91 and 76.11 were attributed to the
100), (002), (101), (102), (110) and (103) facets of the hexagonal
phase Fe N (PDF No. 02-1206). Additionally, the diffraction peak
at 2y of 36.11 corresponded to the (311) facet of Fe (PDF
No. 75-0449) in Fe N-D. The crystallite sizes of the Fe N samples
were calculated using Scherrer’s equation and the results are
listed in Table 1. The crystallite sizes of Fe N-R and Fe
N-D were size of as-prepared Fe
0.8 nm and 41.3 nm, respectively. It can be noted that the obtained from N adsorption–desorption isotherms at 77 K.
crystallite size of Fe N-C was much larger than those of Fe N-R The results are listed in Table 2. The average pore sizes of Fe
2 3
O (PDF No. 02-0919).
2
3
(
Fig. 3 X-ray photoelectron survey spectra of Fe
morphologies.
2
N
with different
2
3 4
O
2
2
The BET specific surface area, pore volume and average pore
N with different morphologies were
2
2
2
5
2
N
2
2
2
and Fe N-D. The results indicated that the crystallite size of the were in the range of 15 to 26 nm. The pores should be ascribed
2
Fe
Fe
2
N-C was the largest, and the crystallite sizes of Fe
N-D were similar in the three samples.
2
N-R and to the voids originating from the porous Fe
2
O
3
. Fig. S2 (ESI†)
gas at 77 K
shows that the Fe samples adsorbed/desorbed N
2
O
3
2
2
2 3 2
Fig. 2 XRD patterns of (a) Fe O and (b) Fe N with different morphologies.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18277--18284 | 18279

View Article Online
Paper
NJC
displaying IV isotherms were characterized as mesoporous could be attributed to the slight agglomeration of particles after
materials, according to the IUPAC definition. Moreover, the nitrogenization, as can be seen from the TEM results.
specific surface areas of the Fe N samples were in the order of
To further characterize the electronic structure and elemental
2
Fe N-D 4 Fe N-R 4 Fe N-C, which was the same as that of Fe O
composition of Fe N with different morphologies, XPS analysis
2
2
2
2
3
2
samples (Table S1, ESI†). However, the Fe N samples exhibited was carried out. As shown in Fig. 3, the wide-scan XPS spectrum
2
2 3 2
lower specific surface areas compared with Fe O samples, which of Fe N showed four strong peaks at binding energies (BE) of
2 2 2
Fig. 4 High resolution XPS spectra of Fe 2p and N 1s of (a and b) Fe N-R, (c and d) Fe N-C, (e and f) Fe N-D.
1
8280 | New J. Chem., 2019, 43, 18277--18284 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
Table 3 XPS analysis of the Fe, N, C and O contents and the N/Fe atom
ratio
Element contents (at%)
Samples
Fe
N
C
O
N/Fe atom ratio (%)
Fe
Fe
Fe
2
N-R
2
N-C
2
N-D
18.93
28.16
15.64
5.97
7.47
1.07
33.75
24.24
20.16
41.35
40.13
63.14
31.54
26.53
6.84
284.8 eV, 397.5 eV, 530.5 eV and 710.5 eV corresponding to C 1s,
N 1s, O 1s and Fe 2p, respectively. The C 1s peak at 284.8 eV was
the internal standard. The high resolution Fe 2p spectra of
2 2 2
Fe N-R, Fe N-C and Fe N-D are shown in Fig. 4a, c and e,
respectively. They all showed two strong peaks at BEs of
7
10.5 eV and 724.2 eV corresponding to Fe 2p3/2 and Fe 2p1/2
,
Fig. 6 Fe M o¨ ssbauer spectra of the Fe
2
N-R catalyst at room temperature.
3
8
which were attributed to FeO
x
species. These FeO
x
were
2
formed by the surface oxidation of Fe N and it did not change
the relative nitrogen content of Fe N, which has been reported by and Table 4. The results indicated that z-Fe N, e-Fe N (x r 2.1)
2
2
x
3
2,39
III
others.
In addition, the peak at 707.2 eV should be attributed and O –Fe –N species accounted for 49.14%, 42.3% and 8.6%
x 4
to Fe–N. The high resolution N 1s spectra of Fe N-R, Fe N-C in Fe N-R, respectively. This meant the main species were
3
2
2
2
2
III
and Fe
2
2 x 2 x 4
N-D are shown in Fig. 4b, d and f, respectively. The XPS z-Fe N and e-Fe N in Fe N-R. Also, a small amount of O –Fe –N
peaks of 396.4 and 397.7 eV can be assigned to N–Fe and N–Fe species were formed due to the surface oxidation of Fe N,
2
3
2,40
39,41
coordination compounds, respectively.
Since the peak inten- which was consistent with the HRTEM and XPS results.
sity of N 1s of Fe N-D was too weak, and the error of the peak
fitting was big, so the peak fitting was ignored. The XPS analysis
data of element contents are listed in Table 3, indicating that the
2
3
.2 Catalytic activity testing
3
The NH decomposition reaction test was carried out at a gas
hourly space velocity (GHSV) of 6000 mL gcat
ꢀ
1
ꢀ1
N/Fe atom ratio of the three Fe N catalysts was in the order of
2
h . As shown in
Fe N-R 4 Fe N-C 4 Fe N-D. The relative nitrogen content of
2
2
2
Fig. 7a, the activity of Fe N-R was superior to that of Fe N-C and
2
2
Fe
2
N-R was the highest among the three samples.
N-R, we performed
HRTEM imaging and EDS mapping at the edge of Fe N-R as
shown in Fig. 5a and b. Fig. 5a shows a clear boundary between
Fe N-D under the same condition. Specifically, the NH con-
2
3
To further explore the structure of Fe
2
version of Fe
2
N-R was 30% at 723 K, largely exceeding the
2
conversion delivered by both Fe
Similarly, Fe N-R exhibited 68% conversion at 773 K, whereas
the NH conversion of Fe N-C and Fe N-D was lower than 50%.
Fig. 7b shows the Arrhenius fitting curves over three Fe
catalysts with different morphologies. The activation energies of
2
N-C and Fe N-D at 723 K.
2
2
4
0
the Fe
2 x
N nanocrystal and a surface FeO coating. The surface
3
2
2
oxidation phenomena have also been reported for MoP and
MoN. This was also consistent in the presence of the O element
on the surface from the XPS study. However, FeO could not be
2
N
4
1
ꢀ
1
ꢀ1
x
Fe N-R, Fe N-C, and Fe N-D were 128.21 kJ mol , 149.50 kJ mol
2 2 2
ꢀ
1
detected by XRD, proving that FeO
x
were amorphous species, or
and 138.93 kJ mol , respectively. Among these Fe N catalysts,
2
that the quantity of FeO was too small to be picked up by XRD,
x
Fe
with the highest catalytic activity of Fe
Our previous work has reported that Fe
active component for the ammonia decomposition reaction.
2
N-R has the lowest activation energy, which is well correlated
or that oxygen existed in the form of iron oxynitride. Further-
more, Fig. 5b clearly shows the homogeneous distribution of
both Fe and N elements within the particles.
2
N-R.
2
N was the main
3
2
To further explore the composition of Fe N-R, Fe M o¨ ssbauer
2
The XPS results showed that Fe N-R had a higher N/Fe atom
2
spectroscopy measurements were carried out as shown in Fig. 6
ratio than Fe N-C and Fe N-D, indicating that more Fe N
2
2
2
species were formed on Fe N-R than on the other two catalysts.
2
Also, small crystallite size Fe N-R was observed from the XRD
2
pattern. Therefore, the Fe
for ammonia decomposition among the three catalysts due to
the high content and good dispersion of Fe N species. Although
Fe N-C had more Fe N content than Fe N-D, they exhibited a
2
N-R had the optimal catalytic activity
2
2
2
2
similar catalytic activity because of the big crystallite size of
Fe N-C (Table 1). These results indicated that the morphologies
2
of the Fe O precursor had a great influence on the elemental
2
3
composition and crystallite size of the final product, which
in turn affected the final catalytic activity. The M o¨ ssbauer
measurements showed that the main species were z-Fe
2
N and
Fig. 5 (a) HRTEM image and (b) EDS mapping of Fe
2
N-R.
e-Fe N-R. It has been reported in the literature
N (x r 2.1) in Fe
x
2
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18277--18284 | 18281

View Article Online
Paper
NJC
2
Table 4 Fe M o¨ ssbauer parameters of the Fe N-R catalyst at room temperature
ꢀ
1
ꢀ1
Catalysts
Fe N-R
CS (mm s
)
QS (mm s
)
H (kOe)
A (%)
Assignment
2
Singlet
Sextet 1
Sextet 2
0.4169(56)
0.2260(20)
0.2850(13)
—
—
49.1(90)
8.6(14)
42.3(18)
2
z-Fe N
III
0.0160(20)
0.0230(12)
346.40(14)
219.23(89)
O
x
–Fe –N
4
e-Fe
x
N (x r 2.1)
Fig. 7 (a) NH
3
decomposition reaction test over Fe
2
N-R, Fe
2
N-C, and Fe
2
ꢀ1
N-D. (b) Arrhenius plots over Fe
2
N-R, Fe
2
2
N-C, and Fe N-D from 673 K to 773 K.
ꢀ
1
The reaction condition: 100 mg catalyst, 0.1 MPa and 6000 mL gcat
h .
that both species exhibited an excellent catalytic activity for been carried out as shown in Fig. S3b (ESI†). When GHSV was
3
2,42
ꢀ1 ꢀ1
ꢀ1 ꢀ1
ammonia decomposition.
activity for ammonia decomposition.
A comparison of the catalytic activities of different Fe based catalytic activity remained stable after 36 hours for ammonia
This was why Fe
2
N-R has a high increased from 6000 mL gcat
h
to 24000 mL gcat
h , the
NH conversion decreased at the same temperature, but the
3
31,43–46
catalysts for ammonia decomposition are shown in Table 5.
This indicated that the Fe N-R catalyst exhibited excellent catalytic
decomposition.
Fe N-R after NH decomposition reaction at different tem-
2
2
3
activity. The ammonia decomposition reaction test at a high GHSV peratures was denoted as used Fe N-R. The TEM image of the
2
ꢀ
1
ꢀ1
of 24 000 mL gcat
h
has been carried out as shown in Fig. S3a used Fe
2
N-R is displayed in Fig. 9a. It can be seen that,
to compared with the fresh Fe N-R, the morphology of the used
conversion decreased at the same Fe N-R did not change significantly. The XRD pattern of the
N-R is displayed in Fig. 9b, indicating that the main
tion benefited the ammonia decomposition reaction, which was phase was still Fe N and its crystallite size (57.2 nm, in the
N-R inset) had little change compared with fresh Fe N-R (50.8 nm).
ꢀ1
ꢀ1
(
ESI†). With the increase of GHSV from 6000 mL gcat
h
2
ꢀ1
ꢀ1
24000 mL gcat
h
, the NH
3
2
temperature. This indicated that the relatively low GHSV condi- used Fe
2
2
7
consistent with the report in the literature. However, the Fe
2
2
also maintained 90% NH conversion at 873 K with a high GHSV Fig. 9c shows the XRD pattern of Fe N-R after the stability
3
2
ꢀ1
ꢀ1
of 24000 mL gcat
The NH
obtained by a 40 h reaction test at 6000 mL g
h
as listed in Table 5.
test. It could be observed that no new peaks appeared and its
3
2
decomposition stability of the Fe N-R catalyst was
catꢀ
1
ꢀ1
h
as shown
in Fig. 8. It could be clearly seen that the catalytic activity was
almost unchanged in the 40 h test. The stability test of ammonia
ꢀ1
ꢀ1
decomposition at a high GHSV of 24000 mL gcat
h
has also
Table 5 Comparison of the catalytic activity of different catalysts for
ammonia decomposition based on the reported data
Temperature GHSV
3
NH conversion
) (%)
catꢀ
1
ꢀ1
Catalysts
(K)
(mL g
h
Ref.
Fe
Fe
Fe
Fe
Fe/CNTs
Fe/CNFs
Fe
Fe
Fe
2
2
2
3
N-R
N-R
N-R
823
848
873
823
973
873
6000
6000
24 000
6000
5000
6500
7500
24 000
24 000
90
93
90
78
75
51
92
80
89
Present
Present
Present
31
43
44
45
46
46
3
Mo N
2
O
3
O
3
O
3
4
4
/CMK-5 873
@CeO
@TiO
Fig. 8 NH
6000 mL g
3
catꢀ
decomposition stability test of Fe
h ).
2
N-R (GHSV
=
2
873
873
1
ꢀ1
2
1
8282 | New J. Chem., 2019, 43, 18277--18284 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
2 2
Fig. 9 (a) TEM image and (b) XRD pattern of the used Fe N-R, and (c) XRD pattern of Fe N-R after stability test.
crystallite size was 50.0 nm (in the inset), which was almost the
same as the fresh Fe N-R. These results confirmed the high
9 D. A. Hansgen, D. G. Vlachos and J. G. Chen, Nat. Chem.,
2010, 2, 484–489.
2
stability of the phase structure of Fe N-R. This was responsible 10 J. Li, W. Wang, W. Chen, Q. Gong, J. Luo, R. Lin, H. Xin,
2
for the high catalytic stability of Fe N-R.
H. Zhang, D. Wang, Q. Peng, W. Zhu, C. Chen and Y. Li,
Nano Res., 2018, 11, 4774–4785.
2
1
1
1 S. J. Wang, S. F. Yin, L. Li, B. Q. Xu, C. F. Ng and C. T. Au,
Appl. Catal., B, 2004, 52, 287–299.
2 A. M. Karim, V. Prasad, G. Mpourmpakis, W. W. Lonergan,
A. I. Frenkel, J. G. Chen and D. G. Vlachos, J. Am. Chem. Soc.,
4
. Conclusion
It was found that the shapes of Fe
important influence on the crystallite size, active component
content and catalytic performance of the corresponding Fe
catalysts for NH decomposition. The prepared Fe N-R exhib-
ited a small crystallite size as well as a high relative nitrogen
content, resulting in the best catalytic activity compared with
both Fe N-C and Fe N-D for NH decomposition. These results
demonstrated that our strategy would have a promising appli-
cation in the design of more active catalysts for various indus-
trial applications.
2 3
O precursors had an
2009, 131, 12230–12239.
2
N
1
1
1
1
1
3 Y. Li, L. Yao, Y. Song, S. Liu, J. Zhao, W. Ji and C. Au, Chem.
Commun., 2010, 46, 5298.
4 D. Huang, C. Jiang, F. Liu, Y. Cheng, Y. Chen and K. Hsueh,
Int. J. Hydrogen Energy, 2013, 38, 3233–3240.
5 H. Zhang, Y. A. Alhamed, W. Chu, Z. Ye, A. AlZahrani and
L. Petrov, Appl. Catal., A, 2013, 464, 156–164.
6 Y. Li, J. Wen, A. M. Ali, M. Duan, W. Zhu, H. Zhang, C. Chen
and Y. Li, Chem. Commun., 2018, 54, 6364–6367.
7 Y. Fujiwara, J. M. Lee, M. Tsujimoto, K. Kongpatpanich,
T. Pila, K. Iimura, N. Tobori, S. Kitagawa and S. Horike,
Chem. Mater., 2018, 30, 1830–1834.
3
2
2
2
3
Conflicts of interest
There are no conflicts to declare.
1
1
8 J. S. J. Hargreaves, Coord. Chem. Rev., 2013, 257, 2015–2031.
9 Z. Schnepp, M. Thomas, S. Glatzel, K. Schlichte, R. Palkovits
and C. Giordano, J. Mater. Chem., 2011, 21, 17760.
0 V. Tagliazucca, K. Schlichte, F. Sch u¨ th and C. Weidenthaler,
J. Catal., 2013, 305, 277–289.
Acknowledgements
2
2
2
This work was supported by the National Natural Science
Foundation of China (21406184).
1 V. Tagliazucca, M. Leoni and C. Weidenthaler, Phys. Chem.
Chem. Phys., 2014, 16, 6182–6188.
2 R. Pelka, K. Kiełbasa and W. Arabczyk, J. Phys. Chem. C,
References
2014, 118, 6178–6185.
1
2
J. A. Turner, Science, 2004, 5686, 972–974.
G. W. Crabtree, M. S. Dresselhaus and M. V. Buchanan,
Phys. Today, 2004, 34, 7.
23 P. Yu, J. Guo, L. Liu, P. Wang, G. Wu, F. Chang and P. Chen,
ChemSusChem, 2016, 9, 364–369.
24 X. Duan, G. Qian, X. Zhou, D. Chen and W. Yuan, Chem.
Eng. J., 2012, 207–208, 103–108.
3
4
5
6
7
8
H. Zhang, M. L. Huang, J. Wen, Y. P. Li, A. R. Li, L. H. Zhang,
A. M. Ali and Y. D. Li, Chem. Commun., 2019, 55, 4699–4702. 25 J. Ji, X. Duan, G. Qian, X. Zhou, G. Tong and W. Yuan, Int.
S. Yin, Q. Zhang, B. Xu, W. Zhu, C. Ng and C. Au, J. Catal.,
004, 224, 384–396.
J. Hydrogen Energy, 2014, 39, 12490–12498.
2
26 D. V. Leybo, A. N. Baiguzhina, D. S. Muratov, D. I. Arkhipov,
E. A. Kolesnikov, V. V. Levina, N. I. Kosova and D. V.
Kuznetsov, Int. J. Hydrogen Energy, 2016, 41, 3854–3860.
Y. F. Chang, J. G. McCarty, E. D. Wachsman and V. L. Wong,
Appl. Catal., B, 1994, 4, 283–299.
S. Ren, F. Huang, J. Zheng, S. Chen and H. Zhang, Int. 27 S. F. Zaman, L. A. Jolaoso, S. Podila, A. A. Al-Zahrani,
J. Hydrogen Energy, 2017, 42, 5105–5113.
X. Ju, L. Liu, P. Yu, J. Guo, X. Zhang, T. He, G. Wu and
P. Chen, Appl. Catal., B, 2017, 211, 167–175.
T. Furusawa, M. Shirasu, K. Sugiyama, T. Sato, N. Itoh and
N. Suzuki, Ind. Eng. Chem. Res., 2016, 55, 12742–12749.
Y. A. Alhamed, H. Driss, M. M. Daous and L. Petrov, Int.
J. Hydrogen Energy, 2018, 43, 17252–17258.
28 L. A. Jolaoso, S. F. Zaman, S. Podila, H. Driss, A. A.
Al-Zahrani, M. A. Daous and L. Petrov, Int. J. Hydrogen
Energy, 2018, 43, 4839–4844.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18277--18284 | 18283

View Article Online
Paper
NJC
2
3
9 S. Podila, S. F. Zaman, H. Driss, A. A. Al-Zahrani, M. A. Daous 38 M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau,
and L. A. Petrov, Int. J. Hydrogen Energy, 2017, 42, 8006–8020.
A. R. Gerson and R. S. C. Smart, Appl. Surf. Sci., 2011, 257,
0 S. Podila, S. F. Zaman, H. Driss, Y. A. Alhamed, A. A. Al-
2717–2730.
Zahrani and L. A. Petrov, Catal. Sci. Technol., 2016, 6, 39 D. M. Borsa and D. O. Boerma, Hyperfine Interactions, 2003,
496–1506. 151, 31–48.
1 S. Zaman, L. Jolaloso, A. Al-Zahrani, Y. Alhamed, S. Podila, 40 F. Yu, H. Zhou, Z. Zhu, J. Sun, R. He, J. Bao, S. Chen and
1
3
H. Driss, M. Daous and L. Petrov, Bulg. Chem. Commun.,
018, 50, 181–188.
2 H. Zhang, Q. Gong, S. Ren, M. A. Arshid, W. Chu and
Z. Ren, ACS Catal., 2017, 7, 2052–2057.
41 J. Yang, F. Zhang, X. Wang, D. He, G. Wu, Q. Yang, X. Hong,
Y. Wu and Y. Li, Angew. Chem., Int. Ed., 2016, 55,
12854–12858.
2
3
3
3
3
3
3
C. Chen, Catal. Sci. Technol., 2018, 8, 907–915.
3 X. Fan, H. Zhang, J. Li, Z. Zhao, C. Xu, J. Liu, C. Xu and 42 S. F. Kurto ˘g lu and A. Uzun, Sci. Rep., 2016, 6, 32279.
J. Liu, Chin. J. Catal., 2014, 35, 286–293.
43 J. Zhang, M. Comotti, F. Schueth, R. Schloegl and D. S. Su,
Chem. Commun., 2007, 1916–1918.
44 X. Duan, G. Qian, X. Zhou, Z. Sui, D. Chen and W. Yuan,
Appl. Catal., B, 2011, 101, 189–196.
4 S. Chouzier, P. Afanasiev, M. Vrinat, T. Cseri and M. Roy-
Auberger, J. Solid State Chem., 2006, 179, 3314–3323.
5 X. Liu, J. Liu, Z. Chang, X. Sun and Y. Li, Catal. Commun.,
2
011, 12, 530–534.
45 A. Lu, J. Nitz, M. Comotti, C. Weidenthaler, K. Schlichte,
C. W. Lehmann, O. Terasaki and F. Sch u¨ th, J. Am. Chem.
Soc., 2010, 132, 14152–14162.
6 J. Liu, S. Yang, W. Wu, Q. Tian, S. Cui, Z. Dai, F. Ren, X. Xiao
and C. Jiang, ACS Sustainable Chem. Eng., 2015, 3, 2975–2984.
7 J. L. Rico and J. S. J. Hargreaves, J. Mater. Sci., 2017, 52, 46 H. Z. Cui, Y. Q. Gu, X. X. He, S. Wei, Z. Jin, C. J. Jia and
188–5193. Q. S. Song, Sci. Bull., 2016, 61, 1–7.
5
1
8284 | New J. Chem., 2019, 43, 18277--18284 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019