Hemin-based conjugated effect synthesis of Fe-N/CNT catalysts for enhanced oxygen reduction

Article abstract of DOI:10.1039/d1nj00020a

Metal-nitrogen codoped cathode catalysts, such as M-N-C (M = Fe, Co, Mn,etc.) are considered most promising non-platinum group ORR catalysts, and have received widespread attention. However, the problem involving the high oxygen reduction performance and stability of the catalyst remains to be solved. The unique olefin oxidation polymerization and π-π stacking effect induced hemin to be evenly coated on polypyrrole nanotubes (PPy). Subsequent carbonization produces the Fe-NCNT catalyst with a monodispersed, uniform diameter and large inner cavity. The PPy not only serves as a template for the formation of the hemin polymer, but also provides C, N source to further improve the catalytic activity. The material exhibits excellent ORR activity attributed to the promotion of the π-π stacking effect between hemin and PPy, and the abundant active site of Fe-N_Xderived from hemin. Results show that the Fe-NCNT-800 catalyst with high performance exhibits a maximum onset potential (E_onset= 0.93 V) and half-wave potential (E_1/2= 0.79 V). The RRDE measures points out the complete four-electron transfer pathway of the Fe-NCNT-800 catalyst. The Fe-NCNT catalysts have higher durability of a negligible negative shift (10 mV) ofE_1/2after a 5000 cycle ADT, and a remarkable methanol tolerance capability that is superior to that of the Pt/C catalyst. The synergy between the PPy-derived N-doped carbon nanotubes and Fe-N_Xfacilitates oxygen reduction and electron conduction.

Full text of DOI:10.1039/d1nj00020a

NJC
View Article Online
View Journal | View Issue
PAPER
Hemin-based conjugated effect synthesis of Fe–
N/CNT catalysts for enhanced oxygen reduction†
Cite this: New J. Chem., 2021,
45, 6940
Yue Lu,a Han Zhang, *a Shaojun Liu,a Chenglong Li,a Lixiang Li,a Baigang An
a
and Chengguo Sun*ab
Metal–nitrogen codoped cathode catalysts, such as M–N–C (M = Fe, Co, Mn, etc.) are considered most
promising non-platinum group ORR catalysts, and have received widespread attention. However, the
problem involving the high oxygen reduction performance and stability of the catalyst remains to be
solved. The unique olefin oxidation polymerization and p–p stacking effect induced hemin to be evenly
coated on polypyrrole nanotubes (PPy). Subsequent carbonization produces the Fe–NCNT catalyst with
a monodispersed, uniform diameter and large inner cavity. The PPy not only serves as a template for the
formation of the hemin polymer, but also provides C, N source to further improve the catalytic activity.
The material exhibits excellent ORR activity attributed to the promotion of the p–p stacking effect
between hemin and PPy, and the abundant active site of Fe–NX derived from hemin. Results show that
the Fe–NCNT-800 catalyst with high performance exhibits a maximum onset potential (Eonset = 0.93 V)
and half-wave potential (E1/2 = 0.79 V). The RRDE measures points out the complete four-electron
transfer pathway of the Fe–NCNT-800 catalyst. The Fe–NCNT catalysts have higher durability of a
negligible negative shift (10 mV) of E1/2 after a 5000 cycle ADT, and a remarkable methanol tolerance
capability that is superior to that of the Pt/C catalyst. The synergy between the PPy-derived N-doped
carbon nanotubes and Fe–NX facilitates oxygen reduction and electron conduction.
Received 3rd January 2021,
Accepted 19th March 2021
DOI: 10.1039/d1nj00020a
rsc.li/njc
1. Introduction
few-layer graphene,15 spherical activated carbon,16 and carbon
The growing energy demand, depletion of fossil fuels, and
worsening environmental pollution have motivated vast tech-
nologies, such as proton exchange membrane fuel cells
(PEMFCs), which are the most popular electrochemical energy
storage and conversion devices.1,2 They can convent chemical
energy into electrical, reacting with hydrogen to form water.
Precious metal-based catalysts, such as Pt-based catalysts, are
highly effective ORR cathodic catalysts. However, the limiting
of their high cost, poor stability, susceptibility to CO poisoning,
and poor methanol resistance, hampers large-scale production
applications.3–7 Therefore, it is important to develop high-
performance, stable and wide-ranging electrocatalysts to
replace precious metal catalysts.8
black,17 which are repeatedly applied as templates to support
precursor materials. In particular, carbon nanotubes and few-
layer graphene are called suitable carrier candidates with excellent
performance due to their light weight properties and high
conductivity.18,19 Generally, heteroatom-doped carbon materials
with their excellent catalytic performance activity, low cost, and
perfect chemical stability have demonstrated enormous potential
for improving the ORR reactivity.20 Since the radius of N atoms is
smaller than that of C atoms, the electron-withdrawing properties
of N atoms will produce a net positive charge on adjacent carbon
atoms, which is conducive to the adsorption of O2 and promotes
the breaking of the OQO bond, thereby increasing the oxygen
reduction capacity of N-doped carbon materials.7 Hetero-atom-
doped carbon-based ORR catalysts can be divided into metal-free
heteroatom (N, B, P and S)-doped carbon-based catalysts7,21–23
and non-noble metal (Fe, Co, Mn and Ni)-doped catalysts.24
Recently, iron–nitrogen–carbon elements (Fe–N–C) catalyst
materials were considered most promising alternatives to
Pt-based catalysts with high cost and scarcity for ORR with
comparable performance to Pt-based catalysts.25,26 However, it is
difficult to determine the true reaction center of most well-
developed Fe–N–C catalysts, and the interpretation of the main
reason for the high activity for ORR is still in debate.27,28 Comparing
the XPS and infrared spectra of N-doped materials before and after
Carbon materials occupy an important position in energy
storage and conversion materials (e.g., fuel cells and lithium
ion batteries) due to their ultra-high stability, excellent electrical
conductivity, modifiable structure, and functionality.9,10 Some
examples include carbon nanotubes,11,12 carbon fibers,13,14
a School of Chemical Engineering, University of Science and Technology Liaoning,
Anshan 114051, China. E-mail: cgsun@njust.edu.cn, hzhang0807@163.com
b School of Chemical Engineering, Nanjing University of Science and Technology,
Xiaolingwei 200, Nanjing, Jiangsu 210094, China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d1nj00020a
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
6940
| New J. Chem., 2021, 45, 6940–6949

View Article Online
Paper
NJC
ORR under alkaline conditions, it is speculated that the ortho was poured into the solution. 10 mL of deionized water containing
carbon atom of pyridinic nitrogen is attached to a hydroxyl group FeCl3Á6H2O (405 mg, 1.5 mmol) was added dropwise to the above
during the ORR reaction, which is an intermediate of the oxidative solution, and kept for 24 h at room temperature under stirring.
reaction. Pyridine nitrogen plays an important role in the oxygen After the reaction, the filter was washed and filtered with a large
reduction reaction, while the ortho carbon next to the pyridinium amount of deionized water and ethanol for several times. The
nitrogen is the active site of oxygen reduction.29 Hossen et al. proved product was dried in a 40 1C vacuum oven and denoted as PPy.
that the presence of a higher amount of surface oxides and pyridinic
N is the main reason for the better performance towards ORR in 2.3 Activation of PPy and synthesis of hemin–PPy,
alkaline media with different Fe–N–C oxygen reduction reaction hemin–NCNT
catalysts.30
A certain amount of PPy nanotubes were mounted into a muffle
furnace, heated to 300 1C and maintained for 1 h. The furnace was
then allowed to cool naturally to room temperature (RT), and the
How to be promising candidates for the ORR electrocatalyst
often requires two factors to consider: first, the maximized
active site with an exposed active site in the active substance;
activated product was denoted as activation-PPy (A-PPy nanotubes).
second, the pores and surface structure of the template material
A-PPy (30 mg) and hemin (30 mg) were dispersed in 40 mL
N,N-dimethylformamide (DMF, 40 mL) under sonication, and
heated to 80 1C in the oil bath. Then, azobisisobutyronitrile
(AIBN) initiator was added in the mixture solution, and the
reaction was kept at 80 1C for 24 h under stirring. After cooling
to RT, the product was washed several times with ethanol and
deionized water, and dried under vacuum at 60 1C for 6 h. The
as-prepared black solid was denoted as hemin–PPy.
to anchor and expose the active sites.31,32 The porphyrin-based
biomass material due to its highly conjugated structure (M–N4),
and cheap advantage is regarded as the most popular precursor.33
Herein, we have developed a facile olefin oxidation polymeri-
zation approach to synthesize hollow tubular Fe/N-codoped car-
bon nanotubes (Fe–NCNT) for ORR catalysts. The formation of
this structure originates from the p–p stacking effect between the
precursor of polypyrrole nanotubes and hemin, which is different
Under the same conditions, a comparative sample was
from physical adsorption. The remarkable electrocatalytic perfor-
prepared using N-doped carbon nanotubes purchased from
mance and stability was attributed to the uniform decoration of
Xianfeng Nanomaterials Technology Co., Ltd as a template,
Fe–NX active sites on the surface of the carbon nanotubes, and
which was labeled hemin–NCNT.
was beneficial to rapid mass transfer. The ORR performance was
achieved with the half-wave potential (E1/2) of 0.79 V, and exhibited
first-rate performance methanol resistance and higher stability in
0.1 M KOH. A catalytic performance comparable to Pt has been
demonstrated, making it a good candidate for ORR electrocatalysts
2.4 Synthesis of Fe–NCNT, NCNT, C–Fe–NCNT
The hemin–PPy was placed in a horizontal tube furnace, and was
heated for 2 h at 5 1C minÀ1 under Ar atmosphere and maintained
for different temperatures (700, 800, 900 and 1000 1C). The
furnace was then allowed to cool naturally to RT, and the final
in energy storage/conversion and other related applications.
product was denoted as Fe–NCNT-X (X = 700–1000). According to
different mass ratios of A-PPy and hemin, the obtained powders
2. Experimental
2.1 Materials and instrument
were named as Fe–NCNT (5 : 30, 15 : 30, 50 : 30). N-Doped carbon
nanotubes (NCNT) and C–Fe–NCNT were prepared with the form
A-PPy and hemin–NCNT under the same procedure for Fe–NCNT.
All chemicals were directly used without purification. Hemin
was purchased from the Hetianlong Biotechnology Co., Ltd.
The standard catalyst of Pt/C (20% Pt on Vulcan XC-72) was
purchased from Johnson Matthey Company, and 5 wt% Nafion
solution was purchased from DuPont Co., Ltd. N-Doped carbon
nanotubes were purchased from Pioneer Nanomaterials Tech-
nology. Other chemicals and all solvents were received from
Sinopharm Chemical Reagent Company. The horizontal tubu-
lar resistance furnace (model: eXPRESS-LINE) was purchased
from Thermcraft Electric Furnace Co., Ltd of the United States.
The morphologies, composition, and structure of the pro-
ducts were characterized by SEM (Apreos), TEM (Tecnai
G2 F20), XRD (D/Max2500 with Cu Ka radiation), FT-IR (EQUI-
NOX55), BET method N2 adsorption isotherm (ASAP-2020),
Raman (HORIBA Xplora Plus) and XPS (PHI5300X).
2.5 Electrochemical characterization
Electrochemical measurements were accomplished in a three-
electrode system using a Garmy Reference 3000 potentiostat
(Garmy Instruments, USA). Ag/AgCl was used as a reference
electrode, and a square platinum plate having a side length of
1 cm was used as the counter electrode. The operation experi-
ments were all performed at room temperature. The glassy
carbon electrode was polished on the suede material with a
0.5 mm polishing powder (Al2O3) suspension before the ink was
dispensed. All potentials were converted to the potential versus
RHE according to E(RHE) = E(Ag/AgCl) + 0.0592 Â pH + 0.1989
when measured against an Ag/AgCl electrode in this study.
As to the working electrode preparation, 4 mg of the sample
was dispersed ultrasonically in a mixture of ethanol (400 mL),
2.2 Synthesis of PPy
The polypyrrole nanotubes (PPy) were prepared using the soft isopropanol (100 mL) and Nafion (25 mL, 5 wt%) to form a
template method with modified procedures.34,35 Methyl orange uniform suspension. The catalyst ink (8 mL) was pipetted onto
(49 mg, 0.15 mmol) as a surfactant was dissolved in 20 mL the polished glassy carbon electrode with a diameter of 5 mm
deionized water, and the pyrrole monomer (105 mL, 1.5 mmol) for all samples, including commercial Pt/C.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6940–6949
| 6941

View Article Online
NJC
Paper
The electrochemical test conditions of the catalyst sample template method with methyl orange (MO) as the soft template.
are as follows: the cyclic voltammetry (CV) was cycled 5 times in The morphological structure of the as-prepared products was
N2 and O2-saturated 0.1 M KOH solution at potential between investigated through scanning electron microscope (SEM) and
1.169 V to 0.169 V at a scan rate of 50 mV sÀ1. Linear sweep transmission electron microscopy (TEM). Fig. 1a and b show
voltammetry (LSV) was measured in O2-saturated 0.1 M KOH that PPy has the diameter of about 200 nm and a wall thickness
solution at different rotating rates from 400 to 2500 rpm with a of about 30 nm. The lengths of the nanotubes are up to several
scan rate of 10 mV sÀ1. The methanol tolerance test and the micrometers. It demonstrates that the monodispersed, uniform
durability test of the catalyst were all measured by a constant diameter and smooth surface hollow tube structures have been
voltage current method. The methanol tolerance test was taken successfully prepared (Fig. S2, ESI†). Second, the PPy was
before and after adding 5 M methanol in O2-saturated 0.1 M activated at 300 1C in the air to improve the dispersibility of
KOH solution. The experimental setup parameters and condi- PPy in DMF (noted as A-PPy). As shown in Fig. 1c, the PPy and
tions of the durability test of the catalyst were identical, except A-PPy nanotubes can disperse in DMF well, and form black
that the time was set to 8 h. The stability test was measured at mixtures after sonicating at room temperature. However, after
1600 rpm in O2-saturated 0.1 M KOH solution before and after standing for 1 day at room temperature, PPy slowly settled
5000 cycles of LSV testing at a scan rate of 10 mV sÀ1
.
down and the DMF became clear. In contrast, A-PPy still
The electron transfer number (n) and the kinetic limit current remained as a good dispersion. The long-term dispersion and
density ( Jk) were calculated by the K–L equation, which is the stability of A-PPy can be attributed to the oxygen-containing
relationship between the reciprocal of the experimentally mea- groups (–O–C, CQO) and defect of A-PPy containing the higher
sured current density ( JÀ1) and the number of revolutions (oÀ1/2). solubility onto the surface of PPy. Then, AIBN-initiator induced
The relationship is a straight line, which can be used to calculate hemin to be polymerized on the outer wall of A-PPy in the
the number of transferred electrons by the slope. The reciprocal of manner of olefin polymerization like a coating using the strong
the intercept is the dynamic limit current density Jk.
p–p stacking effect (hemin–PPy, Fig. S1, ESI†), which effectively
In the RRDE test, the disk electrode was scanned cathodi- restrained the hemin to have uneven stacking and enhanced
cally in an O2-saturated 0.1 M KOH solution. The ring potential conductivity.37 SEM and TEM images of A-PPy is shown in
was maintained at 1.25 V (V vs. RHE), and the scan rate waÀs Fig. S3 (ESI†). PPy-300 still maintained a hollow tubular struc-
0.5 mV sÀ1. The electron transfer number (n) and the HO2
ture after 1 h at 300 1C in air. Finally, Fe/N-codoped carbon
nanotubes (Fe–NCNT) were obtained by annealing the inter-
mediates at different temperatures (700–1000 1C) in an Ar
atmosphere. The SEM images of the as-obtained Fe–NCNT were
not only free from agglomeration, but also uniform in dia-
meter. As shown in Fig. 1e and f, the surfaces of the nanotubes
become rough and porous after treatment under flowing Ar.
Meanwhile, the average thickness of the nanotube wall
yield can be calculated according to the equation.36
3. Results and discussion
The synthesis strategy of Fe–NCNT-X (X = 700–1000) is schema-
tically illustrated in Scheme 1. First, polypyrrole (PPy) with a
hollow tubular structure was prepared by using the soft
Scheme 1 Schematic illustration of the Fe/N-codoped carbon nanotubes electrocatalyst.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
6942
| New J. Chem., 2021, 45, 6940–6949

View Article Online
Paper
NJC
Fig. 1 SEM images of (a) PPy and (d) Fe–NCNT-800. TEM images of (b) PPy, and (e and f) Fe–NCNT-800 (the inset is the thickness of the nanotube wall
distribution histogram). (c) Digital photograph of the equivalent PPy and A-PPy nanotubes dispersed in DMF (4 mL) at room temperature for 1 day.
(g) HAADF-STEM image of Fe–NCNT-800 and corresponding elemental maps of C, N and Fe.
changed from 30 nm (original PPy) to about 43 nm (inset of characteristic peaks and SEM (Fig. 1a). The spectra of the hemin–
Fig. 1b and f). Moreover, the formation of carbon nanotubes PPy exhibited interesting features at 1000 cmÀ1, which corre-
throughout the PPy-derived catalysts may enhance the electro- sponds to the vibration peak of the porphyrin skeleton. In the
nic conductivity and corrosion resistance of the catalyst.33 The 1700 cmÀ1 region, the spectrum of hemin–PPy is dominated by
high-angle annular dark field-scanning transmission electron the signals of the CQO characteristic peak of –COOH in hemin,
microscope (HAADF-STEM) images and corresponding energy- which proved that hemin succeeded in a functionalized modifica-
dispersive X-ray (EDX) mapping analysis of Fe–NCNT-800 tion on A-PPy. After pyrolysis at high temperature, Fig. S5 (ESI†)
(Fig. 1g and Fig. S4, ESI†) reveal that C, N and Fe are homo- illustrated that the infrared characteristic absorption peaks of the
geneously distributed along the catalyst, which could also be pyrrole and porphyrin rings in Fe–NCNT (X = 700–1000) disap-
observed from the results obtained by the full spectrum of XPS pear, which suggests the transformation of the pyrrole and
(Fig. S7, ESI†). In contrast, the direct calcination of PPy at porphyrin rings into Fe/N-codoped carbon nanotubes.34,35
800 1C under Ar atmosphere without hemin coating resulted in
the formation of the N-doped carbon nanotubes (noted as crystallographic phase of the as-prepared composites. The
NCNT, Fig. S12, ESI†). diffraction peak at around 25.01 and 44.51 corresponds to the
X-Ray diffraction (XRD) was performed to investigate the
The FT-IR spectra of PPy, A-PPy, hemin–PPy and Fe–NCNT are (002) and (101) planes of the graphite structure (Fig. 2b). The
shown in Fig. 2a. The infrared characteristic peak of PPy is XRD peaks of Fe–NCNT-700 and Fe–NCNT-800 indicate the
1539 cmÀ1, which corresponds to the C–C stretching vibration presence of Fe2O3 (JCPDS No. 01-1053) and Fe3O4 (JCPDS No.
on the pyrrole ring. The C–N stretching vibration occurs at 75-0033). The experimental parameters were carefully explored
1475 and 1445 cmÀ1, and the 1162 cmÀ1 peak is the plane via control experiments. With the increasing pyrolysis tempera-
deformation vibration of C–H and C–N. The 900 and 778 cmÀ1 ture, Fe (JCPDS No. 87-0721) and Fe3C (JCPDS No. 85-1317) were
features correspond to the C–H deformation vibration outside the observed, resulting from the reaction of the Fe species with
plane of the ring. The presence of PPy was confirmed by the graphite carbon derived from PPy and hemin.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6940–6949
| 6943

View Article Online
NJC
Paper
Fig. 2 The FT-IR spectrum of (a) PPy, A-PPy, hemin–PPy and Fe–NCNT-800. (b) XRD patterns of the Fe–NCNT-X (X = 700–1000) catalyst. (c and d) N2
adsorption–desorption isotherm and the corresponding pore size distribution of Fe–NCNT-X (X = 700–1000).
The N2 adsorption–desorption isotherms of Fe–NCNT-X (402.4 eV) and oxidized N (404.2 eV) (Fig. 3 and Fig. S8,
(X = 700–1000) are type IV profiles (Fig. 2c and Table S1, ESI†), ESI†),40,41 and the relative content of the N sites is summarized
indicating the presence of mesopores and micropores in these in Table S2 (ESI†). The relative content of graphitic N and
samples. The surface areas of Fe–NCNT-700, 800, 900 and 1000 graphite oxidized N content increased with the increase in
are 52.5, 104.3, 48.7 and 55.2 m2 gÀ1, respectively. The pore size temperature from 800 1C to 1000 1C. Interestingly, the
distribution calculated by the DFT method, as shown in Fig. 2d, Fe–NCNT-800 sample has the relative largest content of
indicates that the samples are mainly controlled by the meso- pyridinic-N and Fe–NX content. According to the previous
porous structure, and the total pore volumes of Fe–NCNT-X reports, it has been demonstrated that Fe–NX is mainly respon-
(X = 700–1000) are 0.19, 0.29, 0.24 and 0.25 cm3 gÀ1 (Fig. 2d). sible for the electrocatalytic activity in ORR.42 The Fe species
The Barrett–Joyner–Halenda (BJH) pore size distribution was anchored in the Fe/N-codoped carbon nanotubes by olefin
showed that the four samples have large pore structures of polymerization and pyrolysis process, and the dominant Fe–N
between 56–57 nm, which is the central pore size of the coordination bonds were postulated to be planar Fe–N4 and Fe–
nanotube structure (Fig. S6, ESI†). The change in the specific N2, respectively (Scheme 1). Moreover, the pyridinic N can
surface area depends on the temperature of the pyrolysis. The generate Lewis base sites, and the carbon atom next to pyr-
pyrolysis temperature was increased from 700 1C to 800 1C, and idinic N is favorable for O2 adsorption during the ORR reaction,
the specific surface area was doubled. However, at a higher having a good electron transfer rate and promoting the cleavage
temperature up to 900 1C, the specific surface area rapidly of O–O under alkaline conditions.30 Therefore, the high Fe–NX
decreases. Excessive temperature causes the collapse of the and pyridinic N content is an important reason for Fe–NCNT-
structural pores. Therefore, the specific surface area of the 800 excellent oxygen reduction performance.
Fe–NCNT-800 reaches the maximum and the most pores, it
facilitates the adsorption of O2 and electron transport, and contributed by hemin (Fig. 3c and Fig. S16, ESI†). The high-
promotes the progress of the ORR reaction.38,39
resolution spectra of the Fe 2p electrons of the Fe–NCNT-800
The Fe element in the Fe–NCNT-800 composite material was
X-Ray photoelectron spectroscopy (XPS) was used to deter- composite material were deconvoluted into two pairs of peaks:
mine the chemical composition, chemical state, and bond type Fe2+ (710.6 eV and 722.3 eV) and Fe3+ (712.9 eV and 725.0 eV),
between the elements of the Fe–NCNT-X (700–1000) catalyst. As and the satellite peak is 717.8 eV 43,45 (Fig. 3c, Fig. S9 and
expected, the presence of C, N and Fe elements in the four Table S3, ESI†). Only the Fe–NCNT-1000 sample contained the
catalyst samples is shown in the survey spectra of the XPS Fe(0) species, implying that the iron species was reduced to
measurements (Fig. S7, ESI†). In addition, the high-resolution form metallic elements or clusters of metallic elements at
N 1s spectrum of the samples under different pyrolysis tem- higher temperature conditions, corresponding to the XRD test
peratures can be deconvoluted into five peaks: pyridinic N results (Fig. 2b). The Fe–NCNT-800 sample was further analyzed
(398.4 eV), Fe–NX (399.2 eV), pyrrole N (400.8 eV), graphite N by depth XPS with different etching times of Ar plasma treatment.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
6944
| New J. Chem., 2021, 45, 6940–6949

View Article Online
Paper
NJC
Fig. 3 (a) High-resolution XPS spectroscopy of N 1s of Fe–NCNT-800, (b) nitrogen content in Fe–NCNT-X (X = 700–1000) of four different N
configurations, (c) high-resolution XPS spectroscopy of Fe 2p of Fe–NCNT-800, (d) depth XPS spectra of N 1s for Fe–NCNT-800 at an Ar etching time of
0 and 60 s.
The high-resolution depth spectra of N 1s shows that, com- (different pyrolysis temperature and mass ratio of PPy/hemin)
pared with the original Fe–NCNT-800, the etching time is 60 s, were also tested (Fig. 5a and Fig. S10, ESI†). Notably, the
and the photoelectrons go through the hemin-derived carbon reduction peak position for Fe–NCNT-800 occurs at 0.77 V,
shell. The relative content of graphitic N increased proportion- and is higher than those of Fe–NCNT-X (X = 700, 900 and 1000),
ally, while the Fe–NX and pyridinic–N contents decreased with implying the superior catalytic activity of Fe–NCNT-800 among
Ar plasma treatment, proving that hemin succeeded in functio- these catalysts.44,45
nalized modification on the PPy nanotubes (Fig. 3d and
In order to gain deeper insight into the electrocatalytic
Table S5, ESI†). The high-resolution XPS of the Fe 2p electrons activity and kinetics, the linear scanning voltammetry (LSV)
and the corresponding element atomic content of the Fe–NCNT- curves of Fe–NCNT-800, C–Fe–NCNT-800, NCNT-800 and com-
800 samples at different etching times are shown in Fig. S17 and mercial Pt/C were recorded on RDE with a rotation rate of
Table S7 (ESI†). It reveals that the iron content of the Fe–NCNT- 1600 rpm at a scan rate of 10 mV sÀ1 (Fig. 4b and Table S6,
800 composite material changes slightly with the increase of the ESI†). When compared with Fe–NCNT and commercially Pt/C
etching time, which confirmed that the Fe element is uniformly catalysts, the best ORR activity was achieved by Fe–NCNT-800,
distributed throughout the Fe–NCNT-800 composite nanotube. as judged by the maximum onset potential (Eonset) of 0.93 V and
This is consistent with the Fe element mapping in the HAADF- halfwave potential (E1/2) of 0.79 V, which is close to the Pt/C
STEM image of the tested Fe–NCNT-800.
with Eonset of 1.06 V and E1/2 of 0.82 V. It was worth mentioning
The unique structures and compositions mentioned above that Fe–NCNT-800 had the largest limit current density
were used to explore the catalytic activity of the samples toward (À5.02 mA cmÀ2), exceeding Pt/C (À4.64 mA cmÀ2). This result
ORR, which were performed by rotating disk electrode (RDE) at can infer that the Fe–NX active sites have a strong interaction
room temperature. The cycle voltammetry (CV) was carried out with peroxide, which is beneficial to promote the four-electron
in 0.1 M KOH solution with O2-saturated or N2-saturated reduction process of ORR. In contrast, the NCNT-800 catalyst
conditions (Fig. 4a). A well-defined oxygen reduction peak shows poor catalytic activity, indicating that the N-doped
was observed at 0.77 V (V vs. RHE), which indicates that PPy-based catalyst barely contributed much to improving the
Fe–NCNT-800 can undergo oxygen reduction reaction under overall activity of the catalyst. The ORR activity is significantly
O2 saturation. However, no reaction peak emerged in the N2- enhanced as the hemin was coated on PPy, implying that the
saturation solution in the same voltage window. To estimate activity of the Fe–NCNT-800 catalyst can be largely attributed to
the electrocatalytic activities, the CV curves of Fe–NCNT the Fe–NX sites instead of the C–NX sites. In addition,
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6940–6949
| 6945

View Article Online
NJC
Paper
Fig. 4 Electrocatalytic performance evaluation of Fe–NCNT-X (X = 700–1000), NCNT and commercial Pt/C in O2-saturated 0.1 M KOH solution. (a) CV
curves of Fe–NCNT-800 in N2-saturated and O2-saturated 0.1 M KOH at a scan rate of 50 mV sÀ1. (b) LSV curves of Fe–NCNT-800, C–Fe–NCNT-800,
NCNT and the Pt/C catalyst at 1600 rpm. (c) LSV polarization curves of Fe–NCNT-800 at different rotation rates from 400 to 2500 rpm and inset: the
fitted K–L plots of Fe–NCNT-800 at different potentials. (d) The inset figure is the K–L plots of Fe–NCNT-800 (oÀ1/2 vs. jÀ1/2; o and j are the speed of the
angular rotation and current density, respectively). (e) Tafel slopes of Fe–NCNT-800 and Pt/C. (f) LSV before and after a 5000 cycle ADT of ORR on Fe–
NCNT-800 in O2-saturated 0.1 M KOH. (g) RRDE measurements for Fe–NCNT-800, C–Fe–NCNT-800, NCNT and the Pt/C catalyst. (h) H2O2 yield and
electron transfer number (n) of Fe–NCNT-800, C–Fe–NCNT-800, NCNT-800 and Pt/C. (i) Chronoamperometric responses of Pt/C and Fe–NCNT-800
catalysts with sequential injection of O2-saturated 5 M MeOH at 600 s.
comparing the performance of the C–Fe–NCNT-800 samples suggesting the superior ORR kinetics of Fe–NCNT-800 and the
prepared from commercial N-doped carbon tubes (Fig. 4b and advantage of hollow nanotube structures for the electrocatalytic
Fig. S13, ESI†), it further highlights the excellent performance activity (Fig. 4e). In contrast, the effect of the pyrolysis temperature
of the Fe–NCNT-800 sample, which is attributed to the fact that on the ORR performance of the hemin–PPy samples is shown in
the activated A-PPy is more conducive to the p–p uniform Fig. 5c and Fig. S15 (ESI†). Fe–NCNT-800 gives the maximum
modification of hemin, making the active sites of the derived electron transfer number, a number close to that for the
Fe–NCNT more uniform. It is noteworthy that our Fe–NCNT is 4-electron reaction process of Pt/C. In contrast, the other catalysts
superior to most carbon Fe–N–C electrocatalysts previously obtained under different temperatures at 0.51 V, 700 1C and
reported (Table S8, ESI†).
900 1C exhibited lower electron transfer numbers close to that
Furthermore, the electron transferred number (n) and of the 3-electron reaction process, indicating the poorer selectivity
kinetic current density (Jk) of Fe–NCNT-X (X = 700–1000) were in the ORR electrocatalytic system. It should be noted that the
determined according to the LSV curves calculated from the electron transfer number of Fe–NCNT-1000 has not yet reached
Koutecky–Levich (K–L) equation. The good parallel linearity of the 3-electron reaction process, suggesting that the atomically
the K–L plots of Fe–NCNT-800 from 0.3 to 0.6 V indicated dispersed Fe sites aggregated into metallic Fe particles or clusters
first-order reaction kinetics for ORR (Fig. 4d). Notably, the in the excessive temperature can reduce the electrocatalytic
Fe–NCNT-800 exhibited highly kinetic current density Jk of behavior during the reduction process.
28.23 mA cmÀ2 and a 4-electron ORR reaction mechanism close
To further evaluate the ORR pathway of the catalyst in this
to Pt/C, as shown in Fig. S15 (ESI†).3,46–48 According to the LSV study and determine the amount of H2O2 produced, a rotating
curves, the calculated Tafel slope of the Fe–NCNT-800 catalyst ring disk electrode (RRDE) measurement was performed. In
(90 mV decÀ1) was smaller than that for Pt/C (88 mV decÀ1), fact, the production of H2O2 will not only reduce the electron
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
6946
| New J. Chem., 2021, 45, 6940–6949

View Article Online
Paper
NJC
Fig. 5 (a) Cyclic voltammetry (CV) curves of Fe–NCNT-X (X = 700–1000) in O2-saturated 0.1 M KOH at a scan rate of 50 mV sÀ1. (b) Chronoampero-
metric testing of Fe–NCNT-800 and Pt/C in O2-saturated 0.1 M KOH. (c) LSV curves of Fe–NCNT-X (X = 700–1000) at 1600 rpm and inset: the fitted K–L
plots of Fe–NCNT-X (X = 700–1000) at 0.51 V. (d) LSV curves of hemin: PPy are synthesized at different ratios (mass ratio) and pyrolyzed at 800 1C in O2-
saturated 0.1 M KOH. (e) Fe/N-doped carbon nanotubes show the potential applications for ORR.
transfer efficiency, but also generate HO2À radicals by cracking, Fig. 5d and Table S5 (ESI†). Obviously, the order of the catalytic
which is toxic to electrode materials, such as proton exchange activity increase in the alkaline electrolyte is as follows: Fe–
membranes.30 As shown in Fig. 4g, the Fe–NCNT-800 catalyst NCNT-800 (5 : 30) o Fe–NCNT-800 (15 : 30) o Fe–NCNT-800
has the largest disk current of À1.242 mA and a relatively (50 : 30) o Fe–NCNT-800 (30 : 30). With the increase of the
smaller ring current of 0.0041 mA. The H2O2 yield percentage precursor hemin content from 5 mg to 30 mg, the catalytic
is 0.99% and the electron transfer number (n) is also about 4.0 activity of the Fe–NCNT-800 material also gradually increased.
by calculation (Fig. 4h), which indicates that lower H2O2 This could be attributed to the increase in the number of active
production demonstrates the effective ORR catalytic perfor- centers (Fe–NX in the catalyst to further enhance the catalytic
mance of the Fe–NCNT-800 catalyst and is accompanied by a activity of the Fe–NCNT-800 material. When the amount of
complete four-electron transfer pathway.
hemin that was added increased to 50 mg, the performance of
We also compared the durability and methanol sustainabil- Fe–NCNT-800 (50 : 30) was reduced. It is explained that the
ity of Fe–NCNT-800 with a standard Pt/C catalyst by chronoam- introduction of excessive Fe content will cause the agglomera-
perometry measurement. The current–time (i–t) curve for Fe– tion of metallic Fe during the pyrolysis process and reduce the
NCNT-800 delivers almost no current loss after an 8 h length electrochemical performance of the catalyst.49
durability test, demonstrating higher stability than the com-
mercial Pt/C (current loss with only 75% current retention just
after 2.5 h) (Fig. 5b). The accelerated durability test (ADT) in O2-
saturated electrolytes was carried out to test the stability of Fe–
4. Conclusions
NCNT-800 and Pt/C for 5000 cycles. The Fe–NCNT-800 exhib- In summary, we demonstrated an effective approach for the
ited outstanding durability (Fig. 4f), with a negligible negative preparation of Fe/N-codoped carbon nanotubes derived from
shift (10 mV) of E1/2 after a 5000 cycle ADT, which is superior to PPy nanotubes as a new efficient ORR catalyst. The olefin
Pt/C with a 20 mV shift of E1/2 (Fig. S14, ESI†). The unexpected oxidation polymerization and p–p stacking effect induced
stability can be attributed to the covalent binding of Fe–NX on hemin assembly on the template, fabricating the structure of
the outer wall of PPy with good mechanical strength, and the carbon nanotubes decorated with Fe–NX. The PPy not only
p–p stacking effect between hemin and PPy. As shown in Fig. 4i, serves as a template for the formation of the hemin polymer,
a very slight fluctuation in the current density could be but also provides a C, N source to further improve the catalytic
observed in Fe–NCNT-800 after the addition of 5 M methanol. activity. The as-prepared Fe–NCNT-800 shows excellent electro-
However, the Pt/C catalyst shows an enormous current loss due catalytic performance in the reactions of the oxygen reduction.
to the methanol oxidation of Pt, indicating that Fe–NCNT-800 This is attributed to the unique Fe/N-codoped compositions in
had excellent tolerance of methanol crossover effects for ORR. the carbon skeleton by the p–p stacking effect between hemin
The influence of the synthesized mass ratio of the precursor and PPy, as well as the superiority of the tubular structure,
hemin to PPy nanotubes for ORR activity was investigated in which affords a high surface area, enhances the utilization
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6940–6949
| 6947

View Article Online
NJC
Paper
17 C. Ouyang, B. Ni, Z. Sun, J. Zhuang, H. Xiao and X. Wang,
Chem. Sci., 2019, 10, 2118–2123.
efficiency of the catalytic active sites, shortens the ion diffusion
length, and facilitates the charge transfer. This proposed
strategy can enlighten the construction of Fe/N-codoped tubu-
lar structure, and broaden the application scope of hemin in
the fields of energy conversion reactions.
18 L. Z. Ouyang, Z. J. Cao, L. L. Li, H. Wang, J. W. Liu, D. Min,
Y. W. Chen, F. M. Xiao, R. H. Tang and M. J. I. J. O. H. E. Zhu,
Int. J. Hydrogen Energy, 2014, 39, 12765–12772.
19 C. Lin, L. Ouyang, C. Zhou, R. Hu and M. J. J. O. P. S. Zhu,
J. Power Sources, 2019, 443, 227276.
20 J. Gu, S. Magagula, J. Zhao and Z. Chen, Small Methods,
2019, 3, 1800550.
21 X. Liu and L. Dai, Nat. Rev. Mater., 2016, 1, 16064–16076.
22 F. Razmjooei, K. P. Singh, D.-S. Yang, W. Cui, Y. H. Jang and
J.-S. Yu, ACS Catal., 2017, 7, 2381–2391.
23 C. H. Choi, M. W. Chung, H. C. Kwon, S. H. Park and
S. I. Woo, J. Mater. Chem. A, 2013, 1, 3694–3699.
24 H. Peng, F. Liu, X. Liu, S. Liao, C. You, X. Tian, H. Nan,
F. Luo, H. Song, Z. Fu and P. Huang, ACS Catal., 2014, 4,
3797–3805.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
The authors gratefully acknowledge the financial support pro-
vided by the National Natural Science Foundation of China
(21701077, 11972178, 51872131 and 51972156), the Talent
Project of Revitalizing Liaoning (XLYC1807114, 2019LNZD01,
2020LNQN17, 2019QN07) and Technology Liaoning Talent
Project Grants (601010326).
25 H.-Y. Wang, C.-C. Weng and Z.-Y. Yuan, J. Energy Chem.,
2021, 56, 470–485.
26 B. Li, S. P. Sasikala, D. H. Kim, J. Bak, I.-D. Kim, E. Cho and
S. O. Kim, Nano Energy, 2019, 56, 524–530.
27 B.-C. Hu, Z.-Y. Wu, S.-Q. Chu, H.-W. Zhu, H.-W. Liang,
J. Zhang and S.-H. Yu, Energy Environ. Sci., 2018, 11,
2208–2215.
Notes and references
1 P. Zuo, Y. Li, A. Wang, R. Tan, Y. Liu, X. Liang, F. Sheng,
G. Tang, L. Ge, L. Wu, Q. Song, N. B. McKeown, Z. Yang and
T. Xu, Angew. Chem., Int. Ed., 2020, 6, 9651–9660.
2 L. Gao, M. Xiao, Z. Jin, C. Liu, J. Zhu, J. Ge and W. Xing,
J. Energy Chem., 2018, 27, 1668–1673.
28 H. Ren, Y. Wang, Y. Yang, X. Tang, Y. Peng, H. Peng, L. Xiao,
˜
J. Lu, H. D. Abruna and L. Zhuang, ACS Catal., 2017, 7,
6485–6492.
29 X. Yang, C. Chen, Z. Zhou and S. Sun, Acta Phys.-Chim. Sin.,
2019, 35, 472–485.
30 M. M. Hossen, K. Artyushkova, P. Atanassov and A. Serov,
J. Power Sources, 2018, 375, 214–221.
31 Y. Han, Y. G. Wang and W. Chen, J. Am. Chem. Soc., 2017,
48, 17269–17272.
32 Z. Ma, L. Xu, L. Liu, L. Wang, X. Zhang and A. Kong, Dalton
Trans., 2019, 48, 2338–2344.
3 Y.-J. Wang, W. Long, L. Wang, R. Yuan, A. Ignaszak, B. Fang
and D. P. Wilkinson, Energy Environ. Sci., 2018, 11, 258–275.
4 A. Serov, K. Artyushkova, E. Niangar, C. Wang, N. Dale,
F. Jaouen, M.-T. Sougrati, Q. Jia, S. Mukerjee and
P. Atanassov, Nano Energy, 2015, 16, 293–300.
5 S. Alayoglu, A. U. Nilekar, M. Mavrikakis and B. Eichhorn,
Nat. Mater., 2008, 7, 333–338.
6 G. Yang, C. Erbay, S.-I. Yi, P. de Figueiredo, R. Sadr, A. Han
and C. Yu, Nano Energy, 2016, 22, 607–614.
33 Z. S. Wu, L. Chen, J. Liu, K. Parvez, H. Liang, J. Shu,
H. Sachdev, R. Graf, X. Feng and K. Mullen, Adv. Mater.,
2014, 26, 1450–1455.
¨
7 C. V. Pham, B. Britton, T. Bohm, S. Holdcroft and S. Thiele,
Adv. Mater. Interfaces, 2018, 5, 1800184.
´
ˇ
34 N. V. Blinova, J. Stejskal, M. Trchova, J. Prokes and
8 X. X. Wang, M. T. Swihart and G. Wu, Nat. Catal., 2019, 2,
578–589.
´
M. Omastova, Eur. Polym. J., 2007, 43, 2331–2341.
´
´
´
ˇ
´
´
35 G. Ciric-Marjanovic, S. Mentus, I. Pasti, N. Gavrilov, J. Krstic,
9 W. Wei, D.-w. Wang and Q.-h. Yang, Carbon, 2015, 81, 850.
10 S. Cao and J. Yu, J. Photochem. Photobiol., C, 2016, 27, 72–99.
11 K. Waki, R. A. Wong, H. S. Oktaviano, T. Fujio, T. Nagai,
K. Kimoto and K. Yamada, Energy Environ. Sci., 2014, 7,
1950–1958.
´
J. Travas-Sejdic, L. T. Strover, J. Kopecka, Z. Moravkova,
M. Trchova and J. Stejskal, J. Phys. Chem. C, 2014, 118,
14770–14784.
´
36 C. C. Yang, S. F. Zai, Y. T. Zhou, L. Du and Q. Jiang, Adv.
Funct. Mater., 2019, 29, 1901949.
12 R. Du, N. Zhang, J. Zhu, Y. Wang, C. Xu, Y. Hu, N. Mao,
H. Xu, W. Duan, L. Zhuang, L. Qu, Y. Hou and J. Zhang, 37 S. Zeng, F. Lyu, L. Sun, Y. Zhan, F.-X. Ma, J. Lu and Y. Y. Li,
Small, 2015, 11, 3903–3908.
Chem. Mater., 2019, 31, 1646–1654.
13 M. Wu, Y. Wang, Z. Wei, L. Wang, M. Zhuo, J. Zhang, X. Han 38 Z.-Y. Sui, X. Li, Z.-Y. Sun, H.-C. Tao, P.-Y. Zhang, L. Zhao and
and J. Ma, J. Mater. Chem. A, 2018, 6, 10918–10925.
B.-H. Han, Carbon, 2018, 126, 111–118.
14 C. Liu, J. Wang, J. Li, J. Liu, C. Wang, X. Sun, J. Shen, W. Han 39 C. Si, C. Zhang, J. Sunarso and Z. Zhang, J. Mater. Chem. A,
and L. Wang, J. Mater. Chem. A, 2017, 5, 1211–1220. 2018, 6, 19979–19988.
15 C. Lin, L. Yang, L. Ouyang, J. Liu, H. Wang and M. Zhu, 40 Y. Chen, H. Wang, S. Ji and R. Wang, Catal. Commun., 2018,
J. Alloys Compd., 2017, 728, 578–584. 107, 29–32.
16 A. J. Romero-Anaya, M. Ouzzine, M. A. Lillo-Rodenas and 41 X. Xu, X. Zhang, Z. Xia, R. Sun, H. Li, J. Wang, S. Yu, S. Wang
A. Linares-Solano, Carbon, 2014, 68, 296–307. and G. Sun, J. Energy Chem., 2021, 54, 579–586.
´
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
6948
| New J. Chem., 2021, 45, 6940–6949

View Article Online
Paper
NJC
42 J. Woo, S. Y. Yang, Y. J. Sa, W.-Y. Choi, M.-H. Lee, H.-W. Lee, 46 O. T. Holton and J. W. Stevenson, Platinum Met. Rev., 2013,
T. J. Shin, T.-Y. Kim and S. H. Joo, Chem. Mater., 2018, 30,
6684–6701.
43 D. H. Lee, W. J. Lee, W. J. Lee, S. O. Kim and Y. H. Kim, Phys.
Rev. Lett., 2011, 106, 175502.
57, 259–271.
47 X. Peng, T. J. Omasta, E. Magliocca, L. Wang, J. R. Varcoe
and W. E. Mustain, Angew. Chem., Int. Ed., 2019, 131,
1058–1063.
44 M. Gong, Z. Deng, D. Xiao, L. Han, T. Zhao, Y. Lu, T. Shen, 48 Q. Liu, X. Liu, L. Zheng and J. Shui, Angew. Chem., Int. Ed.,
X. Liu, R. Lin, T. Huang, G. Zhou, H. Xin and D. Wang, ACS
Catal., 2019, 9, 4488–4494.
45 W. Niu, L. Li, X. Liu, N. Wang, J. Liu, W. Zhou, Z. Tang and
S. Chen, J. Am. Chem. Soc., 2015, 137, 5555–5562.
2018, 130, 1218–1222.
49 Y. Chen, S. Ji, Y. Wang, J. Dong, W. Chen, Z. Li, R. Shen,
L. Zheng, Z. Zhuang, D. Wang and Y. Li, Angew. Chem., Int.
Ed., 2017, 56, 6937–6941.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 6940–6949
| 6949