Facile synthesis of N-doped carbon supported iron species for highly efficient methane conversion with H2O2 at ambient temperature

Article abstract of DOI:10.1016/j.apcata.2021.118052

A series of N-doped carbon supported highly-dispersed Fe catalysts are prepared, and their catalytic performances for the direct conversion of methane to CH₃OH and HCOOH with by-product CO₂ are tested at ambient temperature. The nitrogen doping can greatly improve the metal-support interaction, and anchor the Fe species on the support. The catalytic activity of the catalyst is further enhanced by the modification of hydroxylamine hydrochloride. The optimized 2.5 wt%Fe/NC?HH catalyst shows 475 μmol/g_cat for CH₃OH and 832 μmol/g_cat for HCOOH after 1 h reaction. The effect of the type of nitrogen species on the catalytic performance of the catalyst has been studied in detail. In addition, other transition metals (Ni, Co, Cu and Mn) as active centers have been also studied, and none of them is effective for the conversion of methane. Additionally, it is found that the methane conversion over the prepared catalysts proceed via a radical mechanism.

Full text of DOI:10.1016/j.apcata.2021.118052

Applied Catalysis A, General 615 (2021) 118052
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Facile synthesis of N-doped carbon supported iron species for highly
efficient methane conversion with H2O2 at ambient temperature
Li Zhang, Yan Lin*
School of Environmental Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
A series of N-doped carbon supported highly-dispersed Fe catalysts are prepared, and their catalytic perfor-
mances for the direct conversion of methane to CH3OH and HCOOH with by-product CO2 are tested at ambient
temperature. The nitrogen doping can greatly improve the metal-support interaction, and anchor the Fe species
on the support. The catalytic activity of the catalyst is further enhanced by the modification of hydroxylamine
Fe catalysts
Methane
Methanol
N-doped carbon
Heterogeneous catalysis
hydrochloride. The optimized 2.5 wt%Fe/NCꢀ HH catalyst shows 475 μmol/gcat for CH3OH and 832 μmol/gcat
for HCOOH after 1 h reaction. The effect of the type of nitrogen species on the catalytic performance of the
catalyst has been studied in detail. In addition, other transition metals (Ni, Co, Cu and Mn) as active centers have
been also studied, and none of them is effective for the conversion of methane. Additionally, it is found that the
methane conversion over the prepared catalysts proceed via a radical mechanism.
1. Introduction
complex co-enzymes [19–22]. It has been reported that the soluble ho-
mogeneous catalysts, such as Hg, Pt, Pd, Au, Ru, and Rh complexes, can
Methane is the principal component of natural gas, shale gas, coalbed
gas, etc. as well as a promising and cheap feedstock for valuable
chemicals [1–4]. Therefore, the efficient conversion of methane into
value-added chemicals has become an important research direction [5,
6]. Methane molecule is a regular tetrahedral space configuration, and
its chemical properties are quite stable due to the low polarizability, low
effectively catalyze the oxidation of methane to methanol in liquid
systems [10,23–29]. However, these catalytic systems generally require
relatively high reaction temperatures (180–220 ◦C) and strong acidic
media. Recently, an increasing number of studies have been conducted
on utilizing heterogeneous catalysts to achieve selective oxidation of
methane to methanol due to the ease of recycling and environmental
friendliness. Transition metal exchanged zeolites, such as
iron-containing zeolite (Fe-ZSM-5), copper-containing zeolite
(Cu-ZSM-5) and copper-exchanged mordenite zeolite (Cu-MOR), which
could form stable dinuclear centers similar to methane monooxygenase,
were widely developed for direct methane oxidation [5,14,30–34].
Bokhoven and co-workers reported a direct selective anaerobic oxida-
tion of methane to methanol over Cu-MOR using water as oxidant,
however, this reaction is a stepwise process and the reaction tempera-
ture is as high as 200 ◦C [32]. Flytzani-Stephanopoulos and co-workers
developed a noble metal-based Rh–ZSM-5 catalyst suspended in aqueous
solution, which could catalyze the direct oxidation of methane to
methanol, formic acid and acetic acid at 150 ◦C using O2 and CO as
co-oxidants [35].
–
proton and electron affinity, high ionization energy, and high C H bond
strength (439 kJ⋅molꢀ 1), which leads to the difficult activation of CH4
[7–11]. Conventionally, the industrial production mainly adopts an in-
direct route, that is, partial oxidation of methane into syngas (CO and
H2), and then thermo-catalytic conversion of the syngas into liquid ox-
ygenates and hydrocarbons at high temperature and pressure [12–17].
Since the syngas-based route is an energy-intensive production process,
the direct conversion of methane to high value-added chemicals under
mild reaction conditions become a desirable and more economical
alternative way [2,18]. However, the direct methane conversion is
considered as a “Dream reaction”, therefore, developing catalysts
capable of selectively activating methane under mild conditions has
become a crucial research hot.
In nature, methane monooxygenase can catalyze the one-step
oxidation of methane at room temperature, relying on the active sites
of dimeric Fe and dimeric Cu species, but the delivery of oxygen involves
It is consensus that the low reaction temperature is beneficial to
avoid over-oxidation of produced oxygenates [12]. However, it is still a
great challenge to develop low-cost and earth-abundant catalysts with
* Corresponding author.
E-mail address: linyan2002@sjtu.edu.cn (Y. Lin).
https://doi.org/10.1016/j.apcata.2021.118052
Received 22 July 2020; Received in revised form 10 February 2021; Accepted 12 February 2021
Available online 18 February 2021
0926-860X/© 2021 Elsevier B.V. All rights reserved.

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
high catalytic activity and selectivity towards low-temperature methane
conversion. Compared with thermo-catalytic methane conversion,
photocatalytic methane conversion can decrease the reaction tempera-
ture with the aid of light energy via transferring photonic energy to
energetic carriers. Recently, Xie et al. prepared an atomically dispersed
iron species on titanium dioxide through wet impregnation [13]. The
optimized catalyst performed active conversion of methane to methanol
under room temperature and moderate light irradiation in the presence
of H2O2. However, photocatalysis still requires extra energy-input,
which increases the difficulty of industrialization. Cui et al. developed
an efficient graphene-confined single Fe atoms catalyst (Fe‒N‒C) which
could directly convert methane into C1 oxygenated chemicals at
ambient temperature [4], which encouraged us to explore carbon-based
catalysts that could be used for direct conversion of methane.
2.1.3. Synthesis of catalysts
The catalysts were obtained by a conventional impregnation method.
In a standard procedure, 0.25 g of support (N-doped carbon or the
treated activated carbon) and 0.20 g of hydroxylamine hydrochloride
(NH4OCl) were dispersed in 20 mL of distilled water and stirred for 20
min at room temperature. Then, a required amount of FeCl3⋅6H2O was
added to the above solution and continuously stirred for 10 h. After all
the water was evaporated at 80 ◦C, the precursor was further dried at
110 ◦C for 12 h, followed by washing with distilled water for three times
to obtain Fe/NCꢀ HH (with NH4OCl), Fe/NC (without NH4OCl), Fe/
Cꢀ HH (no doping N), and Fe/C. For better comparison, we also pre-
pared M/NCꢀ HH (M = Ni, Cu, Mn and Co) by the above impregnation
method with corresponding metal chloride hydrate.
It was found that the catalytic activity of the catalyst prepared with
Fe(NO3)3 precursor was slightly lower than that prepared with FeCl3,
therefore, we chose to use FeCl3 precursor.
N-doped carbon materials, such as N-doped graphene and N-doped
carbon nanotube, have been extensively applied in preparation of cat-
alysts used for electrocatalytic reactions and other reactions. They are
proved to play very important roles in optimizing the catalytic perfor-
mance of these catalysts. For example, Zhu and Yang et al. reported a Fe/
2.2. Catalysts characterization
Powder X-ray diffraction (XRD) patterns with 2θ = 10ꢀ 80◦ were
N/C catalyst derived from
a cationic Cd(II)-based metal-organic
framework (MOF), which showed high performance in oxygen reduction
reaction (ORR) [36]. N-Co3O4@NC for oxygen evolution reaction (OER)
was prepared using a zeolitic imidazolate framework-67 as precursor via
a controllable N-doping strategy [37]. The authors revealed that
N-doping led to the formation of abundant defects, which promoted the
OER activity and stability. In addition, Fan et al. prepared N-doped
carbon supported Rh nanoparticles by the pyrolysis of ferric citrate and
NH4Cl, whose activity for ammonia borane hydrolysis was higher than
that of many other Rh-based catalysts [38]. The application of N-doped
carbon materials in the direct conversion of methane is rarely found,
therefore, we intend to conduct in-depth research on its application in
methane conversion reaction in this study.
obtained on a LabX XRD-6100 (SHIMADZU) apparatus with a Cu K
α
radiation (40 kV, 30 mA, λ = 0.15418 nm). Inductively coupled plasma-
atomic emission spectrometry (ICP-AES) measurements were carried
out on a Varian 720 ICP-AES. Scanning electron microscopy (SEM)
measurements were conducted on a CARL ZEISS SIGMA-500 equipped
with energy dispersive X-ray spectroscopy (EDX). Transmission electron
microscope (TEM) and HRTEM images were recorded on a FEI-Tecnai
F20 microscope operating at 200 kV. The STEM-mapping measure-
ment was performed with a high-angle annular-dark-field (HAADF)
detector on the lacey support film to eliminate the effect of carbon on
carbon-coated copper grid. N2 adsorption and desorption isotherms
were obtained at liquid nitrogen temperature on a Micromeritics ASAP
2020 Sorptometer after the samples were degassed at 80 ◦C for 6 h. X-ray
photoelectron spectroscopy was implemented on an ESCALAB 250Xi
Herein, a class of N-doped activated carbon supported Fe species are
reported to be prepared through a facile synthesis strategy, in which the
Fe species are anchored on the N-doped activated carbon support via
simple impregnation method. Characterization results indicate that the
Fe species are highly dispersed on the surface of the catalyst and have a
strong interaction with the support. Direct conversion of methane is
examined on the catalyst with H2O2 as the oxidant in aqueous solution,
and the catalytic performance under different reaction conditions was
systematically investigated. The possible reaction pathways are also
discussed in detail based on the experiment results.
spectrometer using Al K
α
radiation with a pressure of ca. 1 × 10–9 Torr.
The results were calibrated based on the binding energy of C 1s peak at
284.8 eV. Raman spectra were collected on a LabRAM HR Evolution
equipped with a 532 nm laser beam under room temperature. Fourier
transform infrared (FT-IR) spectra were collected on a Bruker Vertex 70
spectrometer using KBr as the dispersion medium. 1H- NMR spectra
were recorded at room temperature on a Bruker Avance III 400 MHz
Nuclear Magnetic Resonance. Typically, 0.1 mL of sample and 0.4 mL of
D2O were placed in an NMR tube to be characterized, and the water-
suppression was done for the H NMR spectra.
2. Experimental
2.1. Catalysts synthesis
2.3. Catalytic activity test
All chemicals were analytical grade and purchased from Sinopharm
Chemical Reagent Co., Ltd. unless otherwise specified. They were used
as received without further purification, except for activated carbon.
The methane oxidation reaction was conducted in a 100 mL
stainless-steel autoclave containing a Teflon liner vessel with magnetic
stirring. A K-type thermocouple covered with a quartz glass cover was
used to measure the temperature of the reaction liquid system. In a
typical CH4 oxidation experiment, 20 mg catalyst was added to 15 mL of
H2O2 aqueous solution. After purged three times with methane (99.999
%), the reactor vessel was pressurized to the required pressure of
methane, and then it was heated to the desired temperature under
constant stirring at 950 rpm. After reaction, the reactor was rapidly
cooled in ice-water to a temperature below 10 ◦C. The reaction gas was
collected in a gas sampling bag and the liquid was collected in a vial
after filtering out the catalyst. The analysis of the liquid products were
performed by gas chromatography (GC-9800 series; Kechuang) equip-
ped with a Porapak Q column and flame ionization detector (FID) and
liquid chromatogram (SHIMADZU, SPD-16 series) with a C18 column
and UV–vis detector. The gas phase products were analyzed by GC-FID
gas chromatography (GC-9800 series; Kechuang) with a Porapak Q
column and a methanizer. A Shimadzu GCMS-QP2010 Plus (a gas
chromatograph combined with a mass spectrometer) was applied to
2.1.1. Pre-treatment of activated carbon
Firstly, 10 g of activated carbon was washed with distilled water for
several times until no suspended dust in the supernatant. Then, the
activated carbon was transferred to 2 mol/L of nitric acid and stirred at
50 ◦C for 12 h. Finally, after filtering and washing with distilled water to
remove adherent nitric acid on activated carbon surface, the activated
carbon was dried at 80 ◦C overnight.
2.1.2. Synthesis of N-doped carbon (NC)
The typical route for synthesis of the N-doped carbon was described
as follow: the treated activated carbon was transferred into an alumina
crucible, and then placed in the center of the quartz tube furnace. After
purging the system with high purity N2 for several times, the sample was
calcined in flowing NH3 (50 mL/min) for 3 h at 600 ◦C with a heating
rate of 5 ◦C/min.
2

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
following equation:
CH3OH(m ol) + HCOOH(m ol) + CO2(m ol)
Table 1
Textural properties of the prepared materials.
Vp (cm3/
g)
Da
TOF =
Entry
Sample
Fe content by ICP
(wt%)
SBET
Fe(m ol) × tim e(h)
(m2/g)
(nm)
1
2
3
4
5
6
7
C
–
1320
1140
1280
990
0.81
0.71
0.65
0.61
0.64
0.65
0.64
2.4
2.5
2.5
2.5
2.5
2.5
2.5
3. Results and discussion
NC
–
C-HH
–
–
1.9
2.3
2.1
3.1. Physicochemical properties of the catalysts
NC-HH
2.5 wt%Fe/C
2.5 wt%Fe/NC
2.5 wt%Fe/C-
HH
1180
1040
1020
As listed in Table S1, the nitrogen content for NC, Cꢀ HH, and
NCꢀ HH was measured by SEM-EDX, and the values were 4.7 wt%, 0.9
wt% and 5.8 wt%, respectively. The final contents of Fe in the prepared
Fe supported catalysts were determined by ICP-AES, and the results are
summarized in Table 1. The actual Fe contents increased from 0.4–5.1
wt% for the prepared Fe/NCꢀ HH materials, in accordance with the
stoichiometric values from 0.5 to 5.5 wt%. However, the actual amount
of Fe for 2.5 wt%Fe/C was 1.9 wt%, far below the stoichiometric one.
This could be explained by that the weak interaction between Fe species
and activated carbon support without doping N.
8
0.5 wt%Fe/
NC-HH
0.4
1.3
2.3
3.2
4.1
5.1
1040
970
890
890
790
710
0.65
0.60
0.56
0.57
0.50
0.46
2.5
2.5
2.5
2.5
2.5
2.6
9
1.5 wt%Fe/
NC-HH
10
11
12
13
2.5 wt%Fe/
NC-HH
3.5 wt%Fe/
NC-HH
4.5 wt%Fe/
NC-HH
The XRD patterns of the as-prepared various samples were illustrated
in Fig. 1. All the XRD patterns showed two typical broad peaks located at
24◦ and 44◦, corresponding to the (002) and (100) planes of graphite
[39]. However, we also found that a tiny peak appeared after the acti-
vated carbon was roasted with NH3, which belongs to graphite C‒3R
(PDF#26ꢀ 1079). In a graphite crystal structure, the carbon atoms are
arranged in layers in hexagonal rings. There are two different poly-
morphic types due to the different stacking sequence of layers in the
vertical direction, that is, two-layered repeated graphite-2 h and
three-layered repeated graphite-3R. This indicates that roasting in
ammonia atmosphere is beneficial to the formation of graphite-3R.
There were no characteristic diffraction peaks for iron species
observed in the XRD patterns, proving that iron species possessed poor
crystallinity after 110 ◦C treatment or were highly dispersed with very
small crystallites in the N-doped carbon matrices below the detection
limit of XRD. Moreover, the addition of hydroxylamine hydrochloride
did not show obvious effect on the structure of the catalysts according to
the XRD patterns.
5.5 wt%Fe/
NC-HH
further identify all products detected.
The stability of the 2.5 wt%Fe/NCꢀ HH was investigated for the
direct conversion of CH4 under the optimized reaction conditions (15
mL aqueous solution with 20 mg catalyst and 5000 μ
mol H2O2, 25 ◦C,
ρ(CH4) 4 MPa, 900 rpm). After each run, the catalyst was collected from
the reaction solution by filtering and dried at 70 ◦C for 5 h to evaporate
all the adsorbed reactants and products.
Methanol, formic acid and carbon dioxide were the products detec-
ted within the detection limits. All experimental data have been cor-
rected based on the amount of adsorbed species. Therefore, the
selectivity for one product was estimated by dividing the amount of the
total amount of methanol, formic acid and carbon dioxide. The con-
version of H2O2 was measured by spectrophotometry (Shimadzu UV-
2100 UV–vis spectrometer) using titanium potassium oxalate as the
color-developing agent at the best absorption wavelength of 390 nm
[12].Gain factor (GF), defined as total mol of CH3OH and HCOOH/mol
of H2O2 consumed, was applied to present H2O2 based selectivity and
efficient utilization of H2O2.
The morphologies of the as-prepared samples were characterized by
SEM and TEM measurements, respectively. The SEM images (Fig. 2a and
b) of the representative 2.5 wt%Fe/NCꢀ HH showed that the catalyst
was mainly consisted of various irregular monoliths. These monoliths
were not dense, but a lot of holes were uniformly distributed on their
The total TOF of the two main liquid products was calculated by the
Fig. 1. XRD patterns of (a) C, NC, Cꢀ HH, NCꢀ HH, 2.5 wt%Fe/C, 2.5 wt%Fe/NC, 2.5 wt%Fe/Cꢀ HH, 2.5 wt%Fe/NCꢀ HH and (b) Fe/NCꢀ HH catalysts with different
Fe content.
3

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
Fig. 2. (a, b) SEM images and (c–h) EDX-mapping of 2.5 wt%Fe/NCꢀ HH.
surfaces. Fig. 2(c–h) showed that all the elements were homogenously
distributed on the surface of N-doped carbon support, illustrating that
this catalyst preparation method has certain advantages in preparing
higly disepersed actives sites. The EDX presented in Fig. S1 in supporting
information revealed that the samples were consisted of Fe, Cl, O, C and
N elements. As listed in Table S2 in supporting information, the doping
mass ratio of N element was about 6.3 %, which was greater than that of
NCꢀ HH. From the TEM images in Fig. 3a and f, the porous structure of
activated carbon could be clearly observed. Also, TEM and HRTEM
images for the 2.5 wt%Fe/NCꢀ HH confirmed that no metal oxides
clusters or nanoparticles were present on the N-doped carbon support,
and all the elements were uniformly distributed, consistent with the SEM
results. Instead, a huge agglomerate around 200 nm appeared in the
TEM image for the 2.5 wt%Fe/C. EDS-mapping for 2.5 wt%Fe/C verified
that the huge agglomerate was made of Fe species. These results fully
indicate that nitrogen doping effectively strengthens the interaction
between iron species and activated carbon carrier and plays a crucial
role in the high dispersion and stable existence of iron species.
The N2 adsorption-desorption isotherms and pore size distributions
illustrated in Fig. 4 and Fig. S2 were used to investigate the morphol-
ogies and textural properties of the prepared samples, and the quanti-
ficational data were summarized in Table 1. All the N2 adsorption-
desorption isotherms exhibited typical type IV with H3 hysteresis loop
in the relative pressure ranged from 0.4 to 1.0 (p/p0), suggesting the
4

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
Fig. 3. (a) TEM image, (b) STEM image and (c–e) corresponding elemental distributions for 2.5 wt%Fe/NCꢀ HH; (f) TEM image, (g) STEM image and (h–i) cor-
responding elemental distributions for 2.5 wt%Fe/C; (j) HRTEM for 2.5 wt%Fe/NCꢀ HH.
presence of the mesoporous structure according to the classification of
IUPAC [40]. This also confirmed that modification by N doping and Fe
addition would not damage the mesostructure. BJH pore size analysis
showed that the prepared materials had reasonably narrow pore size
distributions, and we could also conclude that there were some micro-
pores existed in this material. Table 1 summarized the basic physical
properties of the prepared materials dried at 110 ◦C. Overall, the
modification of activated carbon and the addition of Fe would result in a
5

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
Fig. 4. (a) Nitrogen adsorption-desorption isotherms and (b) BJH pore size distributions of the representive 2.5 wt%Fe/NCꢀ HH sample.
710 m2/g and 0.46 cm3/g, respectively. This is mainly due to the sin-
tering of the support and the blockage of the pore system of the catalyst.
Raman spectroscopy of the various samples (Fig. 5a) was provided
for the further study of the structure and properties of carbonaceous
species. All the Raman spectra contained two peaks at 1330 and 1580
cmꢀ 1, which could be readily assigned to disordered carbon (D band)
and graphitic carbon (G band), respectively. The disorders and defects
caused by structural imperfections induced the D band, whereas, the G
2
–
band produced through C C stretching vibration of all pairs of sp
C
atoms in carboatomic ring or long-chain [41]. Generally, the intensity
ratio of D band to G band (ID/IG) was used to measure the graphitic
degree. It was obvious that the ID/IG ratio of 0.75 for N-doped carbon
was larger than that of 0.66 determined for activated carbon, indicating
that the graphitic content of the carbon support decreased after the
roasting with NH3. Additionally, after the introduction of hydroxyl-
amine hydrochloride, the ID/IG ratio increased from 0.75 for NC to 0.84
–
for NC HH, implying that the introdution of hydroxylamine hydro-
chloride gave rise to the distortion of carbon layers and created more
defects [42], which was conducive to the improvement of catalytic
performance. However, the Fe content had no effect on the the graphitic
degree of the N-doped carbon support. FT-IR spectra (Fig. 5b) were
applied to study the chemical bonding and structure of the N-doped
carbon support. The peaks near 1573 cmꢀ 1 were the characteric peaks of
amorphous carbon structure, suggesting the presence of incompletely
carbonized materials. The peaks from 1000 to 1600 cmꢀ 1 reflected the
–
–
typical stretching vibration modes of CN and C N groups, confirming
–
the existence of chemical bonding between C and N in the N-doped
carbon materials, and hydroxyl groups was also formed on the surface of
the catalysts, which was supported by the typical bands near 3420 cmꢀ 1
[43,44].
The chemical states of the iron and nitrogen species on the catalyst
surface were investigated by XPS and the results were presented in
Fig. 6. Fig. 6a shows the XPS images of the Fe 2p region, and all the
samples exhibited two main peaks located at around 711 eV and 724 eV,
corresponding to Fe 2p3/2 and Fe 2p1/2, respectively. The XPS Fe 2p3/
2 spectrum could be deconvoluted into two peaks centered at about 711
eV and 713 eV, which were associated with an Fe(III) oxidation state,
and the peaks at 713 eV could also reflect the existence of hydroxide
groups [13,45,46]. Furthermore, there were obvious satellite peaks
around 718 eV, which was characteristic for the fingerprint of an Fe(III)
oxidation state [13,45]. In addition, the characteristic peaks corre-
sponding to Fe2+ at around 708.4 eV and 721.6 eV were not observed
[46]. It was found that the binding energy (BE) of Fe 2p3/2 was ar-
Fig. 5. (a) Raman spectra and (b) FT-IR spectra for the various samples.
–
ranged in the order of 2.5wt%Fe/NC HH > 2.5wt%Fe/NC > 2.5wt%
gradual decrease in specific surface area and pore volume. For instance,
the SBET and Vp values for pure activated carbon were 1320 m2/g and
0.81 cm3/g, while for the 5.5 wt%Fe/NCꢀ HH sample, they declined to
–
Fe/C HH > 2.5wt%Fe/C. This result demonstrated that the in-
teractions between the Fe species and support were strengthened with
doping nitrogen and introducing hydroxylamine hydrochloride, which
6

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
Fig. 6. (a) Fe 2p, (b) N 1s XPS images and (c) the survey scan of the various catalysts.
was the basis to ensure the stability of Fe species. Additionally, the Fe
2p3/2 primary peaks shifted to lower values with increasing Fe content
to 5.5 wt%, indicating that high Fe content would weaken the interac-
tion between the Fe species and the N-doped carbon support. The N 1s
XPS images of the prepared catalysts in Fig. 6b could be fitted into three
peaks positioned at the BEs of 398.6 eV, 399.9 eV, and 401.1 eV, which
were attributed to pyridine-type N, Feꢀ Nx and/or pyrrolic-type N, and
3.2. Performance of the catalysts for methane oxidation
Direct methane oxidation reaction over the synthesized catalysts was
carried out in the stainless-steel autoclave as described in the experi-
mental section. The liquid products detected were mainly CH3OH and
HCOOH, and by-product CO2 was also detected under certain reaction
conditions. First of all, blank experiments were carried out to determine
the possible thermal reactions which is shown in Fig. S3. There were no
any products produced without catalyst and/or H2O2, confirming that
the the conversion of CH4 catalyzed by the catalysts was real hetero-
geneous reaction. The amounts of the produced CO2 and the conversions
of H2O2 were listed in Table S3, Table S4 and Table S5 in supporting
information.
–
graphite-type N, respectively [47]. However, 2.5 wt%Fe/C HH has
only one primary peak at 400.2 eV, implying that the formed nitrogen
species by introducing hydroxylamine hydrochloride was mainly Feꢀ Nx
and/or pyrrolic-type N. The N 1s peaks in 2.5wt%Fe/NCꢀ HH, 2.5wt%
Fe/Cꢀ HH and 5.5wt%Fe/NCꢀ HH showed larger binding energies
compared to those in 2.5wt%Fe/NC. This suggested that adding hy-
droxylamine hydrochloride in the preparation process could increase
the stability of Fe‒N‒C structure.
Pure activated carbon support, the modified carbon support without
Fe species and those loaded with Fe species were then evaluated for the
direct conversion of methane reaction with H2O2 as the oxidant. As
listed in Table S1 (entries 1–4) and Table S3 (entries 2–5) in supporting
information, the four supports, C, NC, Cꢀ HH, and NC-H H, have not
showed any catalytic activities, illustrating that Fe oxide species were
the actual active centers. When N2 was used to replace CH4 as feedstock
gas for the reaction, CH3OH, HCOOH and CO2 were not generated
(Table S1, entry 5 and Table S3, entry 6), indicating that Fe active
centers could not catalyze the oxidation of carbon support with H2O2.
Clearly, the introduction of different dopants has a major impact on
the catalytic performance, as shown in Fig. 7a. It is well known that the
preparation of highly dispersed active centers is the key to obtain
excellent catalytic performance. As expected, 2.5 wt%Fe/C showed the
lowest catalytic activity (0.2 h–1) without producing HCOOH, but 1363
There were no characteristic diffraction peaks for Fe species
observed in the XRD patterns, which made it difficult to determine the
phase structure of Fe species. From TEM and FT-IR, it was known that
the Fe species was highly dispersed, but more detailed information
about the exact structure of Fe species was not available. Based on the
XPS results, it could be determined that Fe species existed in a trivalent
state. Limited by experimental conditions, there were currently no more
characterization methods to study the exact structure of Fe species. In
the current study, only Fe catalysts supported on molecular sieves with
similar structures to methane monooxygenase have found dimeric iron
sites. In the previous reported catalysts such as Fe-N-C or Fe/TiO2, they
all had isolated sites. Therefore, we inferred that the Fe species existed in
the form of isolated sites in the prepared catalyst.
μmol/gcat of CO2 was produced (Table S3, entry 7). After nitrogen
7

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
that Fe species had a weak interaction with the unmodified carbon
support, and doping N can greatly enhance the interaction and prevent
the aggregation of Fe species. As shown in Fig. 7b, the liquids collected
after reaction, by 2.5 wt%Fe/C and 2.5 wt%Fe/Cꢀ HH were pale yellow,
but those from 2.5 wt%Fe/NC and 2.5 wt%Fe/NCꢀ HH were as clear as
distilled water. The above results clearly demonstrated that N doping
mainly guaranteed the stable existence of iron species on the catalyst
surface and inhibits the generation of CO2, and hydroxylamine hydro-
chloride modification is mainly responsible for improving the activity of
the catalyst. Additionally, the gain factors of 2.5 wt%Fe/Cꢀ HH, 2.5 wt%
Fe/NC and 2.5 wt%Fe/C were 1.5, 1.0 and 0.1 (Table 2), respectively,
which indicated that N-doping and modification of hydroxylamine hy-
drochloride could effectively improve the utilization efficiency of H2O2.
As is well known, different metals have different impacts on the
catalytic performance. Hence, N-doped carbon supported other transi-
tion metal catalysts whose structures were almost the same as 2.5 wt%
Fe/NCꢀ HH were prepared, including 2.5 wt% Ni/NCꢀ HH, 2.5 wt% Co/
NCꢀ HH, 2.5 wt% Cu/NCꢀ HH and 2.5 wt% Mn/NCꢀ HH. Surprisingly,
none of them had any catalytic effects on the methane oxidation reaction
under the identical reaction conditions as 2.5 wt% Fe/NCꢀ HH
(Table S1, entries 6–9 and Table S3, entries16‒19). As listed in Table S3,
for 2.5 wt% Ni/NCꢀ HH, 2.5 wt% Co/NCꢀ HH, 2.5 wt% Cu/NCꢀ HH and
2.5 wt% Mn/NCꢀ HH, the H2O2 had almost no consumption. They could
not effectively catalyze the productive decomposition of H2O2 might be
the main reason for their inactivity to CH4 conversion reaction.
Both the yields and selectivities toward liquid products were affected
by the amount of Fe, therefore, the effect of Fe loading amount on the
catalytic performance was further studied. As shown in Fig. 8a, the yield
Fig. 7. (a) The product yields for various samples. (b) The collect reaction
solutions,1: 2.5 wt%Fe/C; 2: 2.5 wt%Fe/Cꢀ HH; 3: 2.5 wt%Fe/NC; 4: 2.5 wt%
Fe/NCꢀ HH; 5: H2O. Reaction conditions: 15 mL, 25 ◦C, 1 h, ρ(CH4) 4 MPa, H2O2
of HCOOH increased linearly from 53 μmol/gcat to 1371 μmol/gcat as the
Fe amount increased from 0.5 to 5.5 wt%, whereas the CH3OH yield
gradually decreased when Fe content exceeded 2.5 wt%. Additionally, it
was interesting to find that the total yields of CH3OH and HCOOH did
not change too much when the Fe content varied between 2.5 wt% and
5.5 wt%, which demonstrated that the increased HCOOH likely derived
from the oxidation of CH3OH. The promotion of the catalytic activity
was due to the more active sites with the increase of Fe content from 0.5
to 2.5 wt%. However, when the Fe content increased, the amount of CO2
was also significantly increased with the Fe content (Table S3, entries
10–15), implying an over-oxidation of CH4 to CO2, which might be the
reason for the suppression in the formation of CH3OH and HCOOH.
When it comes to the utilization efficiency of H2O2, the gain factor
increased from 0.5 to 1.2 with the Fe content increased from 0.5 to 2.5,
while when the Fe content continued to increase, the gain factor did not
change much (Table 2).
5000 μmol, catalyst 20 mg, 900 rpm.
Table 2
The gain factors for H2O2 to produce CH3OH and HCOOH.
Entry
Sample
H2O2 conv. (%)
Gain factor (×10–2
)
1
2
3
4
5
6
7
8
9
2.5 wt%Fe/C
23.6
18.8
23.3
19.2
38.5
43.1
54.1
60.7
63.8
0.1
1.0
1.5
0.5
0.9
1.2
1.1
1.0
1.0
2.5 wt%Fe/NC
2.5 wt%Fe/C-HH
0.5 wt%Fe/NC-HH
1.5 wt%Fe/NC-HH
2.5 wt%Fe/NC-HH
3.5 wt%Fe/NC-HH
4.5 wt%Fe/NC-HH
5.5 wt%Fe/NC-HH
–
Fig. 8b showed the catalytic activities of the 2.5 wt%Fe/NC HH for
doping (2.5 wt%Fe/NC) (1.1 h–1) and hydroxylamine hydrochloride
modification (2.5 wt%Fe/Cꢀ HH) (2.4 h–1), the catalytic performance of
the catalyst was significantly enhanced. Among them, the promoting
effect of hydroxylamine hydrochloride modification on HCOOH pro-
duction was more obvious, but the yield of CO2 over 2.5 wt%Fe/Cꢀ HH
the direct methane oxidation reaction at different temperatures from 20
to 70 ◦C. Even at low temperature of 20 ◦C, the catalyst still possessed
relatively high catalytic activity with 279 μmol/gcat of CH3OH and 117
μ
mol/gcat of HCOOH. Under the reaction conditions, H2O2 molecules
could adsorbed on the surface of the catalysts and reduced by the Fe
•
•
species to form radicals ( OH and O2ꢀ ) as the active oxidative species.
At the same time, CH4 activation over Fe species proceeded to form CH3
intermediate CH3 intermediate and reacted with radicals, producing a
series of products. This would be analyzed in detail in the following
sections. The yields of CH3OH and HCOOH increased along with the
increase of reaction temperature to 50 ◦C. As the reaction temperature
further raised to 70 ◦C, the yield of HCOOH continued to increase, but
the yield of CH3OH decreased slightly. Howerver, the TOF was always
increased from 1.5 h–1 to 55 h–1with the reaction temperature. High
still remained high, at 1535 μmol/gcat (Table S3, entry 9). When the
catalyst was simultaneously modified by doping N and adding hydrox-
ylamine hydrochloride, the improvement of the catalytic activity (3.2
h
–1) presented superposition effect. Combinding with the XPS result that
the formed nitrogen species on the surface of 2.5 wt%Fe/Cꢀ HH was
Fe–Nx and/or pyrrolic-type N, it could be concluded that Feꢀ Nx and/or
pyrrolic-type N could only improve the catalytic activity, especially the
production of HCOOH. For 2.5 wt%Fe/NC and 2.5 wt%Fe/NCꢀ HH,
nitrogen doping not only increased the catalyst activity, but also
significantly inhibited the production of CO2 (Table S3, entries 8 and
10). This result implied that the introduction of pyridine-type N and
graphite-type N could increase the selectivities of CH3OH and HCOOH.
In addition, no HCOOH was produced over 2.5 wt%Fe/C, revealing that
nitrogen species were essential for the formation of HCOOH in this
catalyst system. From the TEM and XPS results in Fig. 3, it was known
–
reaction temperature is conducive to the activation of the C H bond of
methane, so the activity of the catalyst can be increased when the re-
action temperature is elevated. However, as the reaction temperature
increased, the formation of CO2 will be also enhanced (Table S4, entries
1–6). The yield of CO2 was up to 16,996
μmol/gcat at the reaction
temperature of 70 ◦C. Therefore, the over-oxidation of CH4 to CO2
should be account for the reduction of CH3OH at 70 ◦C. Temperature
8

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
Fig. 8. (a) The effect of Fe content on the catalytic activities of xFe/NCꢀ HH. Reaction conditions: 15 mL, 1 h, 25 ◦C, ρ(CH4) 4 MPa, H2O2 5000
μ
mol, catalyst 20 mg,
mol, (c) 25 ◦C,
900 rpm. (b-d) Catalytic performance of 2.5 wt%Fe/NCꢀ HH under different reaction conditions. Reaction conditions: (b) ρ(CH4) 4 MPa, H2O2 5000
μ
ρ(CH4) 4 MPa and (d) 25 ◦C, H2O2 5000
μmol, 15 mL, 1 h, catalyst 20 mg, 900 rpm.
Fig. 9. (a) Time-on-line yields of products for 2.5 wt%Fe/NCꢀ HH, and (b) Stability of the 2.5 wt%Fe/NCꢀ HH for the direct oxidation of methane under the
optimized conditions with each test for 1 h. Reaction conditions: 15 mL, 25 ◦C, ρ(CH4) 4 MPa, H2O2 5000
mol, catalyst 20 mg, 900 rpm.
μ
had a great influence on the gain factor. In a certain temperature range
(<70 ◦C), high temperature was conducive to the productive decom-
position of H2O2 (Table S4).
found that the H2O2 dosage had no apparent effect on the gain factor
(Table S4).
The influence of CH4 pressure on the catalytic performance was
explored at 25 ◦C. As illustrated in Fig. 8d, a progressive increase of the
yields of CH3OH and HCOOH and TOF occurred when the CH4 pressure
increased. This might be because the higher the pressure of methane, the
greater the amount of methane dissolved in the water. It was also found
that the selectivity of CO2 was sharply decreased from 73.9% to 14.2%
and the H2O2 conversion decreased from 71.5% to 35.4%, respectively,
with the CH4 pressure increased from 1 MPa to 5 MPa (Fig. S6d and
Table S4, entries 2 and 12–15). This revealed that higher CH4 pressure
could inhibit the formation of CO2. Meanwhile, as can be seen from
Table S4, gain factor has a larger value under high methane pressure.
In order to verify if the conversion of CH4 over the prepared catalysts
The effect of H2O2 amount on the catalytic performance for the CH4
oxidation reaction was also investigated as shown in Fig. 8c. When no
H2O2 was present, no CH4 was converted (Fig. S3), and increasing H2O2
from 1000
μ
mol to 5000
μ
mol increased the yields of CH3OH and
HCOOH. However, when the amount of H2O2 exceeded 5000
μmol,
there have no any positive effects on the production of liquid products.
Furthermore, as the oxidant, H2O2 amount is important for controlling
the selectivity of CO2 (Fig. S6c). When the amount of H2O2 exceeded
5000 μmol, CO2 began to be produced, and the selectivity for CO2 also
increased with the amount of H2O2 (Table S4, entries 2 and 7–11), which
led to the decrease of the yields of CH3OH and HCOOH. However, it was
9

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
Fig. 10. (a) TEM image, (b) SEM image, (c) N2 sorption isotherms and pore size distribution, (d) XRD pattern, (e) Fe 2p and (f) N 1s XPS images of the 2.5wt%Fe/
NCꢀ HH after stability test.
is a closed catalytic cycle, the influence of reaction time on the catalytic
activity of 2.5 wt%Fe/NCꢀ HH is given in Fig. 9a. The yields of CH3OH
and HCOOH were found to increase rapidly during the initial 60-min
reaction stage, and steadily increased in the following time, demon-
strating that 2.5 wt%Fe/NCꢀ HH was an effective catalyst for trans-
formation of CH4 with a closed catalytic cycle when H2O2 was used as an
hydroxylamine hydrochloride could improve the catalytic activity of the
catalyst. Therefore, combined with different characterization results, we
analyzed the reason why the modification of hydroxylamine hydro-
chloride could improve the activity of catalyst. The modification of
hydroxylamine hydrochloride could create more defects, increase sta-
bility of Fe species and the number of Feꢀ Nx and/or pyrrolic-type N.
Meanwhile, the formed nitrogen species by introducing hydroxylamine
hydrochloride was only Feꢀ Nx and/or pyrrolic-type N, which could be
used to study the role of different nitrogen species.
oxidant. The yield of CH3OH at reaction time of 10 min was 146
μmol/
g
cat, while the yield of HCOOH was 28 μmol/gcat and no CO2 was pro-
duced (Table S4, entry 16). This implied that there was no induction
stage in CH3OH and HCOOH production, but CO2 formation had an
induction stage.
The stability of the 2.5 wt%Fe/NCꢀ HH was further investigated
under the given reaction conditions. As shown in Fig. 9b, the 2.5 wt%Fe/
NCꢀ HH could be recycled for four times in the stability test without
significant decreases in catalytic performance. ICP-AES experiment
showed that no Fe traces were detected in the reaction solution, which
was in good agreement with the former reasult that the Fe species had
Overall, the role of hydroxylamine hydrochloride was basically
clarified. As we can see in Fig. 7(a), the catalytic activities of 2.5 wt%Fe/
Cꢀ HH and 2.5 wt%Fe/NCꢀ HH were higher than those of 2.5 wt%Fe/C
and 2.5 wt%Fe/NC, which indicated that the modification of
10

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
•
•
obtained HCOOH and CO2. Oꢀ2 was less oxidizing than O H, hence, it
would combine with CH3* species to form stable Oꢀ CH3* groups, which
in turn hydrolyze to form CH3OH. Undoubtedly, some CH3OH and
HCOOH will be further over-oxidized into CO2 during the reaction
process, which has been proven by the results in Table S6.
Table 3
Effect of scavenger on the formation of oxygenatesa.
Entry
Sample
Scavenger
CH3OH
mol/
gcat
HCOOH
mol/
gcat
CO2
H2O2
conv.
(%)
(
μ
(
μ
(
μ
mol/
)
)
gcat)
336
1
2.5 wt%
Fe/NC-
HH
No
475
832
43.1
scavenger
4. Conclusion
2
3
No
Na2SO3
0
0
0
0
In summary, a facile synthesis method has been developed to prepare
N-doped carbon supported highly dispersed Fe species catalysts due to
the strong interaction between active sites and support. The prepared
catalysts with Fe featured III value states and hydroxide can efficiently
and directly convert methane to C1 oxygenated products at room tem-
perature. Particularly, Fe ꢀ Nx and/or pyrrolic-type N can only help
increasing the activity of the catalyst, while pyridine-type N and
graphite-type N can effectively prevent the formation of CO2. Recycling
tests and characterizations of the reused catalyst demonstrated the 2.5
wt%Fe/NCꢀ HH catalyst was highly stable and possessed good recy-
catalyst
2.5 wt%
Fe/NC-
HH
and/or BQ
Na2SO3
384
393
147
24.9
4
5
2.5 wt%
Fe/NC-
HH
BQ
116
139
706
379
399
206
33.7
20.6
2.5 wt%
Fe/NC-
HH
Na2SO3
and BQ
a
Reaction conditions: 15 mL, 1 h, 25 ◦C, ρ(CH4) 4 MPa, H2O2 5000
μmol,
•
•
clability. Free radicals ( OH and O ꢀ2 ) are the main oxidizing active
species, hence, radical pathways are the reasonable reaction mecha-
nism. This work provides an alternative to understanding and designing
non-precious catalysts for direct methane conversion in mild conditions.
catalyst 20 mg, 900 rpm.
strong interaction with the N-doped support.
After the stability test, the used 2.5 wt%Fe/NCꢀ HH catalyst was
characterized by TEM, SEM, BET, XRD and XPS, and the results were
illustrated in Fig. 10. From SEM image and XRD pattern, it could be seen
that the used catalyst was still consisted of irregular monoliths with
graphite C-3R crystal structure. TEM and BET measurements verified the
maintance of the mesostructure of the used catalyst, and the Fe 2p3/2 BE
values for the Fe species on the catalyst surface revealed that the Fe
species still existed with III valence states after reaction. No obvious
changes in the chemical properties and basic structures of the catalyst
during the redox reaction illustrated its good stability and excellent
applicability.
CRediT authorship contribution statement
Li Zhang: Conceptualization, Methodology, Validation, Investiga-
tion, Writing - original draft, Visualization, Formal analysis. Yan Lin:
Conceptualization, Resources, Writing - review & editing, Supervision,
Project administration, Funding acquisition.
Declaration of Competing Interest
The authors declare no conflict of interest.
Acknowledgement
3.3. Reaction mechanism
As is well known, Fe species are the main active species in Fenton-
type reactions. Generally, hydrogen peroxide could be catalyzed to
Y.L. appreciated the support from the National Key Research and
Development Program of China (No. 2017YFE0127100).
•
•
generate OH and O ꢀ2 , which both have strong oxidizing ability. To find
out the actual oxidative species and the reaction mechanism for selective
methane oxidation in water by hydrogen peroxide over 2.5 wt%Fe/
Appendix A. Supplementary data
–
NC HH catalyst, Na2SO3 and benzoquinone (BQ) as radical scavengers
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2021.118052.
•
•
were added to the reaction system to quench OH and O ꢀ2 , respectively
[14]. As listed in Table 3, when only the scavenger was added, there
could not yield CH3OH, HCOOH and CO2, which could rule out the
possibility that the presence of scavenger would affect the reaction.
After introducing Na2SO3, the yields of CH3OH and HCOOH and H2O2
conversion were reduced by 19.2 %, 52.8 % and 42.3 %, respectively,
and the amount of CO2 was also drastically decreased, suggesting that
References
[1] B.Z. Han, Y. Yang, Y.Y. Xu, U.J. Etim, K. Qiao, B.J. Xu, Z.F. Yan, Chin. J. Catal. 37
(2016) 1206–1215.
[2] X.G. Meng, X.J. Cui, N.P. Rajan, L. Yu, D.H. Deng, X.H. Bao, Chem 5 (2019)
2296–2325.
•
O H played a key role in the formation of HCOOH and CO2. When BQ
[3] M. Ravi, M. Ranocchiari, J.A. van Bokhoven, Angew. Chem. Int. Ed 56 (2017)
16464–16483.
was added, the yields of CH3OH and HCOOH were reduced by 75.6 %
and 15.2 %, respectively, however, the amount of CO2 rose slightly,
[4] X.J. Cui, H.B. Li, Y. Wang, Y.L. Hu, L. Hua, H.Y. Li, X.W. Han, Q.F. Liu, F. Yang, L.
M. He, X.Q. Chen, Q.Y. Li, J.P. Xiao, D.H. Deng, X.H. Bao, Chem 4 (2018)
1902–1910.
•
indicating that Oꢀ2 was the main active species for production of
[5] P. Tomkins, M. Ranocchiari, J.A. van Bokhoven, Acc. Chem. Res. 50 (2017)
418–425.
CH3OH. When adding both of the two scavengers to the reaction system,
the yields of CH3OH, HCOOH and CO2 were all decreased. In order to
better understand the reaction pathways, we designed two experiments
using CH3OH or HCOOH and 4 MPa of N2 as the reactants. As listed in
Table S6, little CH3OH was converted to HCOOH and CO2, similarly, the
HCOOH had aslo decreased small amount. This result showed that the
main formation pathways of CH3OH, HCOOH and CO2 should be rela-
tively independent. The possible reaction pathways were proposed as
following: the H2O2 adsorbed on the surface of the catalysts was cata-
[6] C. Okolie, Y.F. Belhseine, Y. Lyu, M.M. Yung, M.H. Engelhard, L. Kovarik,
E. Stavitski, C. Sievers, Angew. Chem. Int. Ed 56 (2017) 13876–13881.
[7] Y. Kwon, T.Y. Kim, G. Kwon, J. Yi, H. Lee, J. Am. Chem. Soc. 139 (2017)
17694–17699.
[8] R. Palkovits, M. Antonietti, P. Kuhn, A. Thomas, F. Schüth, Angew. Chem. Int. Ed
48 (2009) 6909–6912.
[9] R.A. Periana, D.J. Taube, S. Gamble, H. Taube, T. Satoh, H. Fujii, Science 280
(1998) 560–564.
¨
[10] R.A. Periana, D.J. Taube, E.R. Evitt, D.G. Loffler, P.R. Wentrcek, G. Voss,
T. Masuda, Science 259 (1993) 340–343.
•
•
lyzed decomposition into OH and O ꢀ2 . Meanwhile, Fe species would
´
[11] K. Villa, S. Murcia-Lopez, J.R. Morante, T. Andreu, Appl. Catal. B–Environ 187
(2016) 30–36.
adsorb and activate the CH4 to form CH3 intermediate (CH3*). Because
[12] N. Agarwal, S.J. Freakley, R.U. McVicker, S.M. Althahban, N. Dimitratos, Q. He, D.
J. Morgan, R.L. Jenkins, D.J. Willock, S.H. Taylor, C.J. Kiely, G.J. Hutchings,
Science 358 (2017) 223–227.
•
O H is strong oxidizing agent with an oxidation potential as high as 2.73
V, it would continuously attract the other H from CH3*, and finally
11

L. Zhang and Y. Lin
Applied Catalysis A, General 615 (2021) 118052
[13] J.J. Xie, R.X. Jin, A. Li, Y.P. Bi, Q.S. Ruan, Y.C. Deng, Y.J. Zhang, S.Y. Yao,
G. Sankar, D. Ma, J.W. Tang, Nat. Catal. 1 (2018) 889–896.
[30] M.H. Groothaert, P.J. Smeets, B.F. Sels, P.A. Jacobs, R.A. Schoonheydt, J. Am.
Chem. Soc. 127 (2005) 1394–1395.
[14] C. Hammond, M.M. Forde, M.H. Ab Rahim, A. Thetford, Q. He, R.L. Jenkins,
N. Dimitratos, J.A. Lopez-Sanchez, N.F. Dummer, D.M. Murphy, A.F. Carley, S.
H. Taylor, D.J. Willock, E.E. Stangland, J. Kang, H. Hagen, C.J. Kiely, G.
J. Hutchings, Angew. Chem. Int. Ed. 51 (2012) 5129–5133.
[31] V.L. Sushkevich, D. Palagin, M. Ranocchiari, J.A. van Bokhoven, Science 356
(2017) 523–527.
[32] J.S. Woertink, P.J. Smeets, M.H. Groothaert, M.A. Vance, B.F. Sels, R.
A. Schoonheydt, E.I. Solomon, Proc. Natl. Acad. Sci. U. S. A. 106 (2009)
18908–18913.
[15] W.X. Huang, S.R. Zhang, Y. Tang, Y.T. Li, L. Nguyen, Y.Y. Li, J.J. Shan, D.Q. Xiao,
R. Gagne, A.I. Frenkel, F. (Feng) Tao, Angew. Chem. Int. Ed. 55 (2016)
13441–13445.
[33] K.A. Dubkov, V.I. Sobolev, G.I. Panov, Kinet. Catal 39 (1998) 72–79.
[34] K.A. Dubkov, N.S. Ovanesyan, A.A. Shteinman, E.V. Starokon, G.I. Panov, J. Catal.
207 (2002) 341–352.
[16] Y. Tang, Y.T. Li, V. Fung, D. Jiang, W.X. Huang, S.R. Zhang, Y. Iwasawa, T. Sakata,
L. Nguyen, X.Y. Zhang, A.I. Frenkel, F. (Feng) Tao, Nat. Commun. 9 (2018) 1231.
[17] C. Williams, J.H. Carter, N.F. Dummer, Y.K. Chow, D.J. Morgan, S. Yacob, P. Serna,
D.J. Willock, R.J. Meyer, S.H. Taylor, G.J. Hutchings, ACS Catal. 8 (2018)
2567–2576.
[35] J.J. Shan, M.W. Li, L.F. Allard, S. Lee, M. Flytzani-Stephanopoulos, Nature 551
(2017) 605–608.
[36] J.W. Huang, Y.B. Chen, X. Liu, Y.C. Huang, Y.L. Ding, Y. Xu, H.C. Yao, H.B. Zhu,
H. Yang, Chem. Commun. 55 (2019) 6930–6933.
[18] J.J. Shan, M.W. Li, L.F. Allard, S. Lee, M. Flytzani-Stephanopoulos, Nature 551
(2017) 605–608.
[37] Z.C. Wang, W.J. Xu, X.K. Chen, Y.H. Peng, Y.Y. Song, C.X. Lv, H.L. Liu, J.W. Sun,
D. Yuan, X.Y. Li, X.X. Guo, D.J. Yang, L.X. Zhang, Adv. Funct. Mater. 29 (2019),
1902875.
[19] M. Merkx, D.A. Kopp, M.H. Sazinsky, J.L. Blazyk, J. Müller, S.J. Lippard, Angew.
Chem. Int. Ed. 40 (2001) 2782–2807.
[38] F.Y. Zhong, Q. Wang, C.L. Xu, Y. Wang, B. Xu, Y. Zhang, G.Y. Fan, Int. J. Hydrogen
Energy 43 (2018) 22273–22280.
[20] J. Colby, D.I. Stirling, H. Dalton, Biochem. J. 165 (1977) 395–402.
[21] S. Sirajuddin, A.C. Rosenzweig, Biochemistry 54 (2015) 2283–2294.
[39] H.Z. Yang, L. Shang, Q.H. Zhang, R. Shi, G.I.N. Waterhouse, L. Gu, T. Zhang, Nat.
Commun. 10 (2019) 4585.
́
́
[22] D.Y. Osadchii, A.I. Olivos-Suarez, A.. Szecsenyi, G. Li, M.A. Nasalevich, I.
́
A. Dugulan, P.S. Crespo, E.J.M. Hensen, S.L. Veber, M.V. Fedin, G. Sankar, E.
A. Pidko, J. Gascon, ACS Catal. 8 (2018) 5542–5548.
[23] N.J. Gunsalus, A. Koppaka, S.H. Park, S.M. Bischof, B.G. Hashigu-chi, R.A. Periana,
Chem. Rev. 117 (2017) 8521–8573.
[40] H.J. Wu, C.M. Li, H.N. Che, H. Hu, W. Hu, C.B. Liu, J.Z. Ai, H.J. Dong, Appl. Surf.
Sci. 440 (2018) 308–319.
[41] X.-Y. Quek, D. Liu, W. Cheo, H. Wang, Y. Chen, Y. Yang, Appl. Catal. B–Environ 95
(2010) 374–382.
[24] R. Palkovits, C. von Malotki, M. Baumgarten, K. Müllen, C. Baltes, M. Antonietti,
P. Kuhn, J. Weber, A. Thomas, F. Schüth, ChemSusChem 3 (2010) 277–282.
[25] M. Lin, A. Sen, Nature 368 (1994) 613–615.
[42] Y.W. Wu, X.T. Chen, Y. Han, D.T. Yue, X.D. Cao, Y.X. Zhao, X.F. Qian, Environ. Sci.
Technol. 53 (2019) 9081–9090.
¨
[43] A. Samikannu, L.J. Konwar, P. Maki-Arvelab, J.-P. Mikkola, Appl. Catal. B–Environ
[26] T. Zimmermann, M. Soorholtz, M. Bilke, F. Schüth, J. Am. Chem. Soc. 138 (2016)
12395–12400.
241 (2019) 41–51.
¨
[44] L.J. Konwar, P. Maki-Arvela, E. Salminen, N. Kumar, A.J. Thakur, J.-P. Mikkola,
[27] Z.J. An, X.L. Pan, X.M. Liu, X.W. Han, X.H. Bao, J. Am. Chem. Soc. 128 (2006)
16028–16029.
D. Deka, Appl. Catal. B–Environ 176–177 (2015) 20–35.
[45] J.A. Grossnickle, G.C. Vlases, A.L. Hoffman, P.A. Melnik, R.D. Milroy, A. Tankut, K.
M. Velas, Phys. Plasmas 17 (2010), 032506.
[28] M. Muehlhofer, T. Strassner, W.A. Herrmann, Angew. Chem. Int. Ed. 41 (2002)
1745–1747.
[46] J.L. Liu, D. Qian, H.B. Feng, J.H. Li, J.B. Jiang, S.J. Penga, Y.C. Liu, J. Mater. Chem.
A 2 (2014) 11372–11381.
[29] R.A. Periana, D.J. Taube, S. Gamble, H. Taube, T. Satoh, H. Fujii, Science 280
(1998) 560–564.
[47] Z.L. Li, Z.B. Xiao, S.Q. Wang, Z.B. Cheng, P.Y. Li, R.H. Wang, Adv. Funct. Mater. 29
(2019), 1902322.
12