Review

www.advmat.de

Comprehensive Understanding of the Thriving Ambient

Electrochemical Nitrogen Reduction Reaction
Xue Zhao, Guangzhi Hu,* Gao-Feng Chen, Haibo Zhang,* Shusheng Zhang,
and Haihui Wang*

1. Introduction
The electrochemical method of combining N2 and H2O to produce ammonia
(i.e., the electrochemical nitrogen reduction reaction [E-NRR]) continues Our society remains heavily reliant on
the supply of non-renewable fossil fuels
to draw attention as it is both environmentally friendly and well suited for
such as coal, oil, and natural gas, and the
a progressively distributed farm economy. Despite the multitude of recent booming development of science and the
works on the E-NRR, further progress in this field faces a bottleneck. On the advancement of mining technology have
one hand, despite the extensive exploration and trial-and-error evaluation of triggered a rapid decline of fossil fuel
E-NRR catalysts, no study has stood out to become the stage protagonist. On reserves and resulted in a severe energy
the other hand, the current level of ammonia production (microgram-scale) is depletion crisis.[1] Given this harsh back-
ground, countermeasures need to be
an almost insurmountable obstacle for its qualitative and quantitative deter- explored to preserve the valuable achieve-
mination, hindering the discrimination between true activity and contamina- ments of the human civilization. The
tion. Herein i) the popular theory and mechanism of the NRR are introduced; gradual reduction of our dependence on
ii) a comprehensive summary of the recent progress in the field of the E-NRR traditional energy sources and the devel-
and related catalysts is provided; iii) the operational procedures of the E-NRR opment of renewable energy technologies
should be viewed as top-priority tasks.
are addressed, including the acquisition of key metrics, the challenges faced,
A bird’s-eye view of the global resource
and the most suitable solutions; iv) the guiding principles and standardized distribution shows that H2O, CO2, O2,
recommendations for the E-NRR are emphasized and future research direc- and N2 are widely available and recy-
tions and prospects are provided. clable sustainable resources that can be
converted into usable materials, fuels, or
other strategically significant chemicals.
X. Zhao, Prof. G. Hu Among these resources, CO2, O2, and H2O are widely used.
Institute for Ecological Research and Pollution Control of Plateau Lakes For example, recent years have witnessed great advances in
School of Ecology and Environmental Science H2O splitting based on the electrochemical hydrogen evolution
Yunnan University
Kunming 650504, China
reaction (HER) and the oxygen evolution reaction (OER)[2] as
E-mail: [email protected] well as progress in electrochemical O2[3] and CO2[3d,4] reduc-
Dr. G.-F. Chen tion reactions (ORR and CO2RR, respectively). In the near
School of Chemistry and Chemical Engineering future, these reactions are expected to be driven by solar, wind,
South China University of Technology and/or tidal energy to benefit the society. However, the artifi-
Guangzhou 510641, China cial fixation (under mild conditions) of N2, which accounts for
Dr. H. Zhang 78% of the atmosphere, is still in its infancy after more than
College of Chemistry and Molecular Sciences
Wuhan University 100 years of research. The efficient conversion of N2 to
Wuhan 430072, China ammonia (NH3) under ambient conditions continues to draw
E-mail: [email protected] attention, and the effective utilization of N2 resources remains
Prof. S. Zhang an elusive and challenging task.[5]
College of Chemistry and Molecular Engineering NH3 is of key importance to numerous industrial produc-
Zhengzhou University
Zhengzhou 450000, China
tion processes and human life in general, and it is a crucial
Prof. H. Wang
factor determining social progress. The global NH3 production
Beijing Key Laboratory of Membrane Materials and Engineering reached 146 million tons in 2015 and is expected to increase by
Department of Chemical Engineering 40% by 2050.[6] The significance of NH3 is due to the following
Tsinghua University factors. 1) As the main component of nitrogen fertilizers, NH3,
Beijing 100084, P. R. China called the blood of agriculture, helps to sustain the lives of
E-mail: [email protected]
people worldwide.[7] 2) NH3 is the basic raw material for most
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202007650.
industrial chemicals and is widely used for the production of
plastics, explosives, pharmaceuticals, dyes, synthetic fibers,
DOI: 10.1002/adma.202007650 and resins.[8] 3) As a non-carbon hydride source, NH3 is an

Adv. Mater. 2021, 33, 2007650 2007650 (1 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

excellent green energy carrier, featuring a high energy density process, legumes can effectively capture N2 from the sur-
of 5.52 kWh kg−1 and a high hydrogen content of 17.6 wt%.[9] rounding environment under ambient conditions and convert it
In contrast to H2 (direct energy source), NH3 (energy storage into nitrogen-containing substances that can be readily assimi-
chemical) can be easily liquefied for storage and transportation lated. Unsurprisingly, significant work has been done to eluci-
and affords a green mixture of N2 + H2O after being used as a date the active factors driving this N2 fixation and thus pave the
fuel.[10] way to the artificial fixation of N2 under ambient conditions.
The Haber–Bosch process, which exploits the direct combi- The binding of N2 and H2O in leguminous plants is promoted
nation of N2 and H2 to form NH3 (N2 + 3H2 → 2NH3), was by Fe/Mo nitrogenases, which convert N2 into NH3 and related
discovered more than 100 years ago and laid the foundation for substances according to the equation of N2 + 6H+ + nMg-ATP +
the massive industrial production of NH3. This process uses 6e− → 2NH3 + nMg-ADP + nPi.[11b,16] The Fe-containing protein
Fe- or Ru-based metal catalysts, but because of the inherent promotes Mg-ATP hydrolysis to generate energy/electrons, and
inertness of the NN bond in N2 (bond energy = 941 kJ mol−1, the hydrogen bonds and redox-active sites of the surrounding
energy required to break the first NN bond = 410 kJ mol−1),[11] cofactors provide coordination sites for the attachment and
high temperatures (350–550 °C) and pressures (15–25 MPa) are protonation of N2.[17] Although the above N2 fixation method
required.[12] Besides, H2 used in this process is almost entirely is too slow to replace the Haber–Bosch process, the role of
produced by the steam reforming of CH4, which is accom- Fe/Mo as an active center for enzymatic N2 fixation shows that
panied by the emission of large amounts of CO2 (≈2 tons of in the presence of a suitable metal-based catalyst, N2 can com-
CO2 per ton of generated NH3), with annual CO2 emissions bine with H2O to form NH3 under ambient conditions.
reaching 300 million tons.[11b,13] The energy input for NH3 syn- Inspired by nitrogenases, researchers have developed many
thesis in the Haber–Bosch process equals 485 kJ mol−1, with catalysts for the reduction of N2 to NH3, which are used in
the related annual energy consumption accounting for 1–3% light- or electricity-driven N2 reduction systems. Both of these
of the energy released by fossil fuel combustion.[14] Although systems combine N2 and H2O into NH3, but their action mech-
the Haber–Bosch process remains the primary NH3 produc- anisms are very different.[18] Although progress in the reduction
tion route, it does not have high efficiency; the yield of NH3 in of N2 to NH3 under mild conditions has been slow, this topic
a round of the Haber–Bosch process equals 15–20%, and the has attracted considerable attention in recent years (Figure 1).
final yield of 97% can only be achieved by the recycling of unre- Compared with NH3 production by photocatalytic N2 reduc-
acted syngas.[15] Therefore, in the face of the increasing demand tion, that by the electrocatalytic N2 reduction reaction (E-NRR)
for NH3, we cannot continue to rely on the capital- and energy- has drawn significant interest because of the following. 1) Light
intensive Haber–Bosch process, whose long-term operation not sources usually have a wide wavelength range, but only light
only aggravates the depletion of fossil energy but also imposes of a specific wavelength can effectively activate N2. Thus, most
a heavy burden on the environment. This dramatic situation of the light in the full band cannot be used, which results in
has inspired the exploration of environmentally benign and low energy efficiency.[19] 2) The activation of N2 is difficult to
resource-sustainable NH3 production methods, including those complete in an instant, and the photocarrier often suffers
that can be implemented under ambient conditions. from rapid electron–hole recombination. 3) At present, the
Although the inherent inertness of N2 leads to the require- performance of available photocatalytic NH3 fixation methods
ment of extreme heat and pressure in the Haber–Bosch is insufficient for electrochemical N2 fixation. The vigorous

Figure 1. Report frequency of elements involved in E-NRR catalyst under ambient conditions since 2017.

Adv. Mater. 2021, 33, 2007650 2007650 (2 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

development of the E-NRR can be attributed to the facts that N2 can be explained as follows. 1) Unlike that of the CC triple
this method is more economical, greener, and easier to operate bond of acetylene (∆H0 = −171 kJ mol−1), the first protonation
than other ones. 1) The E-NRR employs the readily available of N2 is endothermic (∆H0 = +37.6 kJ mol−1), that is, thermo-
H2O and N2 as reactants and generates NH3 and O2, which dynamically forbidden.[23] 2) The triple bond in the linear N2
do not cause environmental pollution. 2) The E-NRR can be molecule is very strong (941 kJ mol−1) and has high ionization
driven by small amounts of electricity and can utilize additional energy of 15.6 eV.[11b,23,24] 3) N2 exhibits a negative electron
electricity derived from clean (e.g., solar, wind, tidal, and water affinity (−1.9 eV); additional electron input requires external
potential) energy sources.[20] 3) Compared with the capital- and energy, thus it is resistant to most Lewis acids/bases.[11a] 4) The
energy-intensive Haber–Bosch process, the E-NRR is more suit- energy gap between the highest occupied molecular orbital
able for the distributed small-batch production of NH3, avoiding (HOMO) and the lowest unoccupied molecular orbital (LUMO)
secondary energy consumption during transportation and of N2 is 10.82 eV, which greatly hinders the effective transfer
storage. 4) For different electrocatalysts, the value of the applied of electrons, as illustrated by the low efficiency of NH3 produc-
potential can be fine-adjusted to rapidly optimize NH3 produc- tion by photocatalytic N2 reduction.[24,25] It is worth noting that
tion and energy consumption.[21] These advantages continue although the first protonation of N2 is endothermic, subsequent
to inspire the exploration of the E-NRR. Although numerous protonation to form NH3 is relatively easy. Thus, the overall
E-NRR catalysts have been reported to date, the E-NRR field is Gibbs free energy change of the NRR is negative,[9b] as exem-
still in its infancy as no electrocatalyst capable of effectively pro- plified by the conversion of N2 to NH3 in natural nitrogenase
moting the reaction has been established; the best-performing systems, the Haber–Bosch process, photocatalysis, and electro-
E-NRR catalysts are still far from achieving the performance catalysis. Vegge et al. proved the feasibility of the E-NRR under
of the Haber–Bosch process. In addition, most catalysts used ambient conditions through theoretical calculations.[26] They
in aqueous electrolytes also promote the competitive HER, used a volcano diagram model to show the possibility of E-NRR
which results in low current efficiency and poor selectivity for occurrence on various metal surfaces, laying the foundation for
the E-NRR.[22] Finally, further progress is hindered by the dif- the rational design of E-NRR electrocatalysts (Figure 2a,b).
ficulty of unambiguously proving that the detected NH3 is pro- The E-NRR involves multistep N2 protonation and multi-
duced by the electrocatalytic reaction of N2 with H2O. Based on electron transfer, as illustrated below (main reaction = E-NRR;
the above challenges, the optimization of E-NRR systems and side reaction = conversion of N2 to N2H4 and the HER).[11a,b,23,27]
the promotion of further standardization in the E-NRR industry
are considered as essential tasks. N2 + H+ + e − ↔ N2H, E 0 = −3.20 V vs RHE (1)
Herein, we comprehensively introduce the current status of
E-NRR systems to provide guidance for the rational develop-
ment and design of E-NRR catalysts and the correct completion N2 + 2H+ + 2e − ↔ N2H2(g) , E 0 = −1.10V vs RHE (2)
of E-NRR operation. Initially, the intrinsic characteristics of N2,
the E-NRR mechanism, and activity metrics of E-NRR systems N2 + 4H+ + 4e − ↔ N2H4(g) , E 0 = −0.36V vs RHE (3)
are briefly introduced, and pioneering works on electrochem-
ical N2 reduction reactions (including those at higher tempera- N2 + 6H+ + 6e − ↔ 2NH3(g) , E 0 = +0.55V vs NHE (4)
tures) are discussed in detail to show the evolution of E-NRR
development. Next, we discuss the research status of metal- and
N2 + 6H2O + 6e − ↔ 2NH3(g) + 6OH− ,
non-metal-based catalysts promoting the E-NRR under ambient (5)
conditions and provide some suggestions to improve their effi- E 0 = −0.736V vs SHE, pH = 14
ciency. In addition, the actual operation of E-NRR is addressed,
and the challenges faced by different operation processes are N2 + 4H2O + 4e − ↔ N2H4(g) + 4OH− ,
(6)
pointed out. Finally, we summarize the current bottlenecks of E 0 = −1.160V vsSHE, pH = 14
E-NRR development and provide 1) suggestions for the rational
design of E-NRR catalysts based on the reported experimental N2 + e − = N2(aq)
–
, E 0 = −3.37 V vsRHE, pH = 14 (7)
and theoretical results, 2) appropriate catalyst characterizations
and suggestions for the standardization of E-NRR operation
2H+ + 2e − ↔ H2(g) , E 0 = 0V vsSHE, pH = 0 (8)
processes, 3) insights into the optimization of E-NRR electro-
chemical systems, and 4) future prospects for the E-NRR.
2H2O(l) + 2e − ↔ H2(g) + 2OH− ,
(9)
E 0 = −0.828 V vs SHE, pH = 14
2. Understanding the E-NRR
Equations (1)–(3) show that the main obstacle to the elec-
In the N2 molecule, each N atom has an electronic configura- trochemical conversion of N2 to NH3 (i.e., the rate-limiting
tion of 2s22p3, and the four atomic orbitals (1 × s, 3 × p) of each step) is the first one-electron reduction/protonation of N2 as
atom combine to produce four bonding (2 × σ and 2 × π) and the subsequent six- and four-electron reductions are easier.
four anti-bonding (2 × σ* and 2 × π*) molecular orbitals. The Comparison of Equation (7) with Equation (1) shows that in an
inert nature of N2 makes it a good protective gas for most reac- electrochemical system, the activation of N2 via H binding is
tions, with lightning being one of the few natural factors that more likely than that via reduction to N2− as the latter requires
can convert N2 into NOx. The high thermodynamic stability of more severe conditions (pH = 14). The two groups of r­ eactions

Adv. Mater. 2021, 33, 2007650 2007650 (3 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 2. a) The “volcano” model shows the possible situation of E-NRR activity on various metal surfaces, where the solid and dashed lines represent
the absence and presence of H-bonds effect, respectively. Reproduced with permission.[26a] Copyright 2012, PCCP Owner Societies. b) Volcano diagram
of association and dissociation mechanism (free energy ΔGN* as a descriptor), where the dotted line represents hydrogen competition reaction.
Reproduced with permission.[26b] Copyright 2013, PCCP Owner Societies. c) Possible routes for proton-coupled electron transfer and thermodynamic
quantities related to the reaction step, where ET stands for electron transfer and PT stands for proton transfer. d) The selectivity of PCET path and
SPET path depends on pH. Reproduced with permission.[32] Copyright 2017, Royal Society of Chemistry.

described by Equations (4)/(8) and (5)/(9) show that in addi- under ambient conditions. In the associative pathway, the NN
tion to the E-NRR, the competitive HER is unavoidable in bond is not completely cleaved until the first NH3 molecule is
aqueous solutions and it is the main side-reaction of E-NRR released. Depending on the influence of the electrocatalyst on
systems. In addition, if N2H4 cannot be further protonated on N2, the associative pathway can be further divided into distal
the catalyst surface to break the remaining NN bond, the for- hydrogenation, alternating hydrogenation, and enzymatic
mation of this species competes with the generation of NH3 hydrogenation. In distal and alternating hydrogenation path-
(Equations (3) and (6)). The relatively large equilibrium poten- ways, only one N atom of N2 is adsorbed on the catalyst surface
tial, numerous intermediate states, and widespread side (Figure 3b). In the distal hydrogenation pathway, the N atom
­reactions (mainly the HER) of the E-NRR greatly hinder its far from the catalyst first combines with H provided by the sur-
exploration, which puts forward higher requirements for the rounding environment to form *NH3, while the N atom near
reasonable design of E-NRR catalysts and the overall optimiza- the catalyst surface does not bind with H atoms until the first
tion of the electrochemical system. NH3 is released; it is then hydrogenated to release the second
Theoretical studies have established several possible mecha- NH3. In the alternating hydrogenation pathway, the two N
nisms for the conversion of N2 to NH3 on the catalyst surface, atoms of N2 alternately combine with H atoms to release two
mainly including the dissociative pathway, the associative NH3 molecules almost simultaneously.[11b] In the enzymatic
pathway, and the Mars–van Krevelen (MvK) pathway.[11a,26a,28] hydrogenation pathway, the two N atoms of N2 are bound to the
The dissociative pathway is mainly operational in the Haber– catalyst surface simultaneously, and they alternately (or simulta-
Bosch process, featuring the cleavage of the NN bond of N2 neously) combine with H atoms to release two NH3 molecules
on the catalyst surface under the action of high temperature (Figure 3c). Recently, a new mechanism completely different
to form activated atomic nitrogen (N*) that is subsequently from dissociative and associative pathways has been proposed
hydrogenated to generate NH3 (Figure 3a). The conversion of for the conversion of N2 to NH3 (the MvK pathway). In general,
N2 to NH3 via the dissociative pathway is driven by high ther- the formation of NH3 on the surface of transition metal nitrides
modynamic energy and is therefore difficult to implement is believed to follow the MvK pathway (or mixed pathways

Adv. Mater. 2021, 33, 2007650 2007650 (4 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 3. a–c) Schematic diagram of dissociative pathway and associative pathways. Reproduced with permission.[74] Copyright 2016, American Chem-
ical Society. d) Schematic diagram of Mars-van Krevelen pathways.

including the MvK pathway). According to this mechanism, the electron transfer (PCET) (Figure 2c) and sequential proton–elec-
N on the transition metal nitride is reduced to NH3, and the N2 tron transfer (SPET) pathways in the conversion of N2 to NH3
provided by the environment fills the vacancy of the extracted remains unclear. It is often assumed that the PCET pathway
lattice N atoms to complete catalyst recovery and prepare for occurs on the surface of solid metal catalysts, whereas certain
a new reaction (Figure 3d).[29] The MvK mechanism has been works also show the importance of the SPET pathway under
recognized by researchers and compared with the associative these conditions.[30] Kumar et al. used theoretical calculations to
pathway to reveal that in the former case, N2 is electrochemi- show that the ammonization of N2 in the E-NRR preferentially
cally converted into NH3 with a lower overpotential.[29b] In the follows the SPET pathway; specifically, H+ ions attack the acti-
MvK mechanism, the “gas (N2)-liquid (H2O)-solid (catalyst)” vated N2, and electrons are then introduced into the obtained
three-phase reaction is decomposed into “liquid (H2O)-solid positively charged intermediate.[31] The potential barrier of
(catalyst)” and “gas (N2)-solid (solid)” reactions, which greatly proton transfer in the SPET pathway (28.78 kcal mol−1) is lower
promotes mass transfer kinetics. In view of the poor solubility than that in the PCET pathway (31.03 kcal mol−1). Although the
of N2 in water, the MvK mechanism can compensate for the former pathway is preferred in the E-NRR, the selectivity for
lack of N2 supply to a certain extent. This compensation can be these two pathways depends on the pH of the environment
attributed to the large area effect of the catalyst that can capture (Figure 2d).[32] To avoid the undesired occurrence of the HER,
traces of N2 in the electrolyte to fill the N defects. Research on the above studies did not consider pH. Therefore, regarding the
the mechanism of N2 conversion to NH3 in the E-NRR remains selectivity of PCET and SPET pathways, more detailed theoret-
largely theoretical, with only a handful of reports containing ical calculations are needed to optimize the physical and chem-
experimental data. At present, some detection methods can be ical properties of E-NRR catalysts.
introduced into the E-NRR to determine the key steps of the
N2 to NH3 conversion. For instance, in situ electrochemical
infrared and in situ electrochemical Raman spectroscopy can 3. E-NRR Activity Metrics
be used to track subtle real-time changes in the electrolyte and
catalysts on the working electrode. Moreover, experimental data 3.1. NH3 Yield
analysis can be combined with theoretical calculations to eluci-
date the type of NH3 formation mechanism occurring on a spe- NH3 production, the most important and ultimately pursued
cific catalyst, which will have far-reaching influence on E-NRR goal of the E-NRR, is commonly assessed in terms of NH3
development. yield. Therefore, the objective and effective reporting of NH3
Despite the abundance of experimental and theoretical yield is essential for promoting communication within the sci-
studies on the E-NRR, the selectivity for proton-coupled entific community. At present, NH3 yield is mainly reported

Adv. Mater. 2021, 33, 2007650 2007650 (5 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

in the forms of µg h−1 cm−2, µg h−1 mg−1, µmol h−1 cm−2, or slightly fluctuate. In this case, one can state that the cata-
µmol h−1 mg−1, µmol s−1 cm−2, µmol s−1 mg−1, µmol h−1 g−1, lyst meets the E-NRR stability standard. The stability of electro-
µmol s−1 g−1, mol s−1 cm−2, mol s−1 mg−1, and mol s−1 g−1, which catalysts is often affected by their components and structure.
all represent the amount of NH3 produced per hour (or per Therefore, researchers need to have a scientifically rigorous
second) per unit of catalyst. Here, cm−2 represents the apparent attitude toward the research target to report the stability metrics
geometric area of the catalyst on the working electrode, and of specific catalysts and thus promote the further optimization
mg−1 represents the content of the catalyst on the working elec- of E-NRR catalysts.
trode. It should be noted that when using mg−1 as the normal-
ized catalyst quantity, some reports employ the overall quality
of the catalyst as a benchmark, while others refer to the most 3.4. Other Metrics
active part of the catalyst (such as the metal on the carrier) as
a benchmark for normalization. In general, NH3 yield can be At present, the reported literature focuses almost exclusively
expressed as rNH3 = (CNH3 × V)/(t × Δ), where V is the volume of on the three main E-NRR activity metrics of NH3 yield, FE,
the electrolyte in the cathode cell (mL), t is the time (s, min, or and stability. However, to promote a comprehensive evaluation
h), and ∆ is the catalyst amount (cm2, mgcat, or mgmetal). Here, of E-NRR progress, researchers are advised to consider and
we recommend that researchers unify the form of NH3 yield emphasize the following metrics.
in their reports to facilitate rapid communication and com-
parison between peers. As the current NH3 yields achieved in 1) Turnover frequency (TOF) quantifies the amount of reactants
the E-NRR are at µg levels, one is advised to use the units of converted per active site per unit time (quantified by the pro-
µg h−1 cm−2 and/or µg h−1 mg−1, which is also the form pre- duced NH3) and is calculated as TOF = N/Mt, where t is the
ferred by most researchers. reaction time, N is the amount of N2 converted to NH3 within
t, and M is the number of active sites.
2) For NH3 yield and FE as metrics, one is advised to consider the
3.2. Faradaic Efficiency overpotential (the standard potential of the E-NRR is 0.092 V
vs RHE) at the optimal potential and the corresponding Tafel
As the E-NRR is often accompanied by the HER and N2H4 gen- slope (theoretically, the value of the 60 or 120 mV dec−1 node
eration as side reactions, it is crucial to quantify the selectivity can be used as a benchmark[33]), even though the E-NRR is
of NH3 generation in the electrocatalytic system. Ideally, all (or accompanied by serious side reactions.
most) of the external input energy should be used to generate
NH3 via the E-NRR. Therefore, the ratio of the amount of elec-
tricity used to generate NH3 (QNH3) to the total amount of elec- 4. Progress of E-NRR Catalyst Development
tricity passed through the electrode (Qtotal) can be defined as the
Faradaic efficiency (FE) of NH3 production in the E-NRR and 4.1. Metal-Based E-NRR Catalysts
is calculated according to Equation (10), where F is the Faraday
constant (96 500), CNH3 is the concentration (mol L−1) of NH3 4.1.1. Noble Metal Based E-NRR Catalysts
in the electrolyte, V is the volume (mL) of the electrolyte, and Q
is the total amount of electricity passing through the electrode. Ru-Based E-NRR Catalysts: Early theoretical calculations
have revealed that Ru-based catalysts efficiently promote the
FE = 100% × (3F × CNH3 × V )/Q total (10) E-NRR. Some theoretical studies have shown that the E-NRR
activity of the stepped Ru crystal plane significantly exceeds
FE can intuitively describe the effect of side reactions on the that of the flat Ru crystal plane, further demonstrating that
main reaction of NH3 production and, to a certain extent, guide the former can more strongly adsorb N2.[34] Ru is almost at the
the design and screening of highly selective catalysts. Currently, peak of the volcano diagram of the theoretical limit potential
the NH3 selectivity (FE) of the E-NRR remains relatively low. (U) on the metal surface vs. N2 adsorption energy (∆EN*) and
Notably, NH3 yield and FE often exhibit a leverage effect, which therefore favors N2 dissociation and NH3 generation via the
is often manifested as low FE at high NH3 yield or high FE at associative pathway with a low activation barrier.[34a,35] In addi-
low NH3 yield (based on the available reports). Therefore, FE is tion, Ru has an empty d-orbital and can therefore accept the
an important activity metric for guiding E-NRR development. lone pairs of electrons provided by N2 to activate the NN
bond. Initially, researchers used graphite as a carrier to pre-
pare a Ru/C catalyst for the E-NRR,[36] achieving an NH3 yield
3.3. Catalyst Stability of 0.21 µg h−1 cm−2 and an FE of 0.28% at a high voltage of
−1.10 V versus RHE. The nanoscale effect of metal morphology
The stability of an E-NRR electrocatalyst is a key indicator for was exploited to prepare nanodispersed Ru-based materials
evaluating its practical value. In other words, the decomposition (Ru/C) with greatly increased E-NRR activity due to the high
or deactivation of the E-NRR catalyst during operation hinders surface exposure.[37] Geng et al. incorporated Ru3+ into zeolite
its use in industrial production even if the initial E-NRR activity imidazole framework (ZIF) and carbonized the resulting mate-
is excellent. Generally, after the E-NRR catalyst has been reused rial to obtain Ru single atoms distributed on N-doped carbon
more than 10 times or has been continuously used for more (Ru SAs/N-C), that is, atomically dispersed Ru. Compared with
than 10 h, the original NH3 yield and FE should be maintained bulk or dispersed Ru cluster catalysts, Ru SAs/N-C achieved a

Adv. Mater. 2021, 33, 2007650 2007650 (6 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 4. a) Ru SAs/N-C i) preparation route and ii) ammonia yield and iii) FE driven by it. Adapted with permission.[38] Copyright 2019, Wiley-VCH.
b) The stable geometric configuration and charge density difference of N2 on Ru1@C2N, Ru1@T-C3N4, and Ru1@γ-graphene, where blue and yellow
represent positive and negative charges, respectively. Adapted with permission.[40] Copyright 2019, Wiley-VCH. c) DFT results (side view) of boron
monolayer and TM adsorption (i), calculated energy profiles for NRR catalyzed by Ru/Bα (ii: distal) and Ru/Bβ (iii: distal) respectively. Reproduced
with permission.[42] Copyright 2019), Royal Society of Chemistry.

1–2 orders of magnitude (120 µg h−1 mgcat−1) higher NH3 yield E-NRR performance of Ru have also been reported. Liu et al.
and an FE of 29.6% (Figure 4a).[38] The high catalytic perfor- used DFT calculations to demonstrate that electron-deficient
mance of Ru SAs/N-C is closely related to the high exposure 2D coordinated B can endow a Ru atomic site with an ability
degree of atomic Ru active sites and the synergistic effect of to convert N2 to NH3 via an associative pathway (Figure 4c).[42]
the carrier. Other reports also demonstrated the correlation The introduction of another metal also has a certain auxiliary
between the E-NRR activity and the carrier of Ru-based cata- effect on the E-NRR activity of Ru.[43] For example, an Ru cata-
lysts. For instance, RuCl3 with UiO-66 as the domain-limited lyst coated on a titanium felt plate (Ru/Ti) was shown to have a
carrier was carbonized to obtain monoatomic Ru@ZrO2/NC. high defect density and a strong electron affinity for N2, effec-
Experiments combined with theoretical calculations showed tively promoting the E-NRR.[43a] Although the stepped crystal
that the carrier in Ru@ZrO2/NC helps to improve E-NRR per- plane, defect (vacancy) engineering, and the carrier effect can
formance, with Ru@NC in Ru@ZrO2/NC promoting the NRR be propitious to the E-NRR performance of Ru-based catalysts
pathway and Ru@ZrO2 inhibiting the competitive HER.[39] (Ru > Os > Rh), the experimental and theoretical results sug-
Wang et al. used density functional theory (DFT) to compre- gest that Ru-based catalysts also have prominent HER perfor-
hensively evaluate the promotional effect of the carrier (C2N, mance.[44] Further control of the selectivity of Ru-based catalysts
C3N4, and γ-graphene) of monoatomic Ru on the E-NRR per- for NH3 production via the E-NRR is their greatest challenge,
formance of the active Ru center (Figure 4b).[40] The N or C in which forces researchers to shift their attention to other (non)
these carriers coordinated to Ru atoms to prevent their aggre- metal-based E-NRR catalysts.
gation and leaching as well as to accelerate electron transfer. Pd-Based E-NRR Catalysts: Pd-based catalysts are cheaper
In addition, the related simulation results showed that N2 is than Ru-based ones and can also promote the E-NRR, helping
most likely to be adsorbed on the active site via an associative to realize this reaction at low overpotential. The Pd-catalyzed
pathway. For T-C3N4 as the carrier, the energy of N2 adsorption E-NRR features the conversion of N2 to *N2H as the rate-lim-
energy at the Ru atoms was maximal, while the E-NRR over- iting step, the energy barrier of which can be reduced by Grot-
potential was minimal. Subsequent experiments showed that thuss-like hydride transfer from hydrogenated Pd (α-palladium
monoatomic Ru dispersed on g-C3N4 (SAs/g-C3N4) exhibited hydride, α-Pd-H). According to this mechanism, protons in
better E-NRR performance (NH3 yield = 23.0 µg h−1 mgcat−1; an aqueous medium are transferred from one water molecule
FE = 8.3%) than bulk Ru, further revealing that the presence of to another via oxygen atoms, as in a relay race.[45] As E-NRR
g-C3N4 changed the d-electron state of Ru and uplifted the Ru catalysts, Pd-based materials easily absorb H from the environ-
Fermi level.[41] The effects of other carriers and dopants on the ment and can therefore act as intermediaries for the transfer

Adv. Mater. 2021, 33, 2007650 2007650 (7 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 5. a) DFT calculation of the free energy pathways of HER (insets) and related steps of N2RR on Au, Pt, and α-PdH (211) surface (including
atomic structure insets). Reproduced with permission.[46] Copyright 2018, Springer Nature. b) Preparation route of Nanoporous PdH0.43. c) Possible
pathways for reactions involving lattice hydrogen on palladium hydride in nitrogen reduction reactions. d) In situ Raman spectra during PdH0.43 catalytic
N2 reduction. e) Gibbs free energy of NRR on np-Pd and np-PdH0.43 (based on enzymatic mechanism). Adapted with permission.[47] Copyright 2020,
Wiley-VCH. f) Preparation strategy of PdO-Pd. g) Alternating hydrogenation pathway of NRR at PdO–Pd interface. Reproduced with permission.[48]
Copyright 2019, Royal Society of Chemistry.

of protons from H2O to N2. Wang et al. prepared α-Pd-H After Pd is hydrogenated (α-Pd-H), the surface becomes able
with a hydrogenated surface on carbon black, showing that to support a Grotthuss-like proton hopping mechanism, which
this catalyst exhibits an enhanced E-NRR performance[46] and reduces the reaction energy of N2 by 0.3 eV and thus promotes
(compared to Pt and Au) features surface *H kinetics ther- the E-NRR at low overpotentials (Figure 5a). The role of palla-
modynamically favorable for the conversion of N2 to *N2H. dium hydride catalysts in the E-NRR has been further explored,

Adv. Mater. 2021, 33, 2007650 2007650 (8 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

for example, porous PdH0.43 was shown to achieve excellent the Au catalyst and its support (including the inhibition of the
E-NRR performance (NH3 yield = 20.4 µg h−1 mg−1, FE = 43.6%) competitive HER).
at a low overpotential of −150 mV (Figure 5b).[47] Isotope labe- The easiest thing to understand is that the fluffy (e.g.,
ling experiments revealed that the lattice hydrogen atoms of porous, dendritic, or nanosheet-like) structure of the catalyst
PdH0.43 are the active hydrogen source for the E-NRR, while results in the increased exposure of the Au surface (often con-
in situ Raman spectroscopy and DFT calculations showed that sidered as the active site of the E-NRR). Carrier morphology
palladium hydride reduces the activation energy of the rate- can be controlled to achieve the dispersion of Au active sites.
limiting N2 to N2H conversion (Figure 5c–e). Lv et al. made Wang et al. constructed a porous Au-membrane E-NRR catalyst
full use of the ability of palladium hydride to promote the (porous Au film on porous Ni foam, pAu/NF), showing that
E-NRR and ingeniously constructed a Pd-based E-NRR catalyst it exhibits ideal E-NRR activity due to the abundant exposure
(PdO–Pd) with dual interface functions on carbon nanotubes of Au active sites.[52] ZIFs are another common carrier that
(CNTs) (Figure 5f).[48] In the PdO–Pd-catalyzed E-NRR, PdO can provide abundant nanometal attachment sites. Au-based
and Pd act as active sites and proton transfer intermediaries, porous nanomaterials supported by ZIF-8 (NPG@ZIF-8) pro-
respectively, reducing the overpotential. The Pd on the side of mote the smooth progress of the E-NRR, and their activity
the interface has p-π feedback d-orbital electrons, which can ascribed to the rich Au sites and the dual functions of the ZIF-8
decelerate the electrochemical adsorption of (and hence, acti- shell, namely the prevention of Au loss and HER suppression
vate) N2. PdO on the other side of the interface can capture (Figure 6a).[53] An E-NRR catalyst with Au clusters anchored on
H to form α-Pd-H, thus providing a source of hydrogen for the porous framework of polytetrafluoroethylene was reported
activated N2 (Figure 5g). Doping engineering can regulate the to exhibit high E-NRR performance (FE = 37.8%) due to its
electronic distribution and structure of Pd, which is crucial rich Au active site exposure, the hydrophilicity of the Si-based
for optimizing the E-NRR performance of Pd-based catalysts. cathode, and poor HER activity.[54] In addition, the morpho-
On the one hand, specific dopants can be used to suppress the logical characteristics of Au active centers could be adjusted to
competitive HER, as exemplified by the introduction of Pb into increase the chance of collision with N2. Wang et al. prepared
Pd-based catalysts.[49] Zhao et al. found that the introduction of flower-like Au nanoparticles (Au nanoflowers) by ultrasonica-
Pb into Pd-based materials (i.e., PdPb nanosponges) restricts tion, revealing that the 2D structure of “petals” increased the
the adsorption of *H on Pd to greatly inhibit the HER[49a] and exposure of the Au surface,[55] with the NH3 yield and FE
promotes the E-NRR by reducing the free energy change of the reaching 25.57 µg h−1 mgcat−1 and 6.05%, respectively. Hollow
rate-limiting step. Thus, porous PdPb catalysts can be prepared Au nanocages (AuHNCs) were also shown to exhibit enhanced
using morphology control to maximize active site exposure E-NRR performance, as exemplified by a AuHNC prepared
and thus enhance E-NRR performance.[49b] Moreover, Pd-based using the difference in the reduction potentials of Ag and Au
materials can also be doped with Cu to suppress the HER. In and having an edge size of 39 nm.[56] Compared with solid Au-
this case, the H covering the Pd surface is transferred to the Cu based materials, this AuHNC not only provided about twice the
surface, thereby releasing Pd sites with E-NRR activity and dis- number of Au active sites (on inner and outer surfaces), but
favoring the occurrence of the undesired HER on Pd.[50] On the also promoted the high-frequency collisions of N2 with the
other hand, dopants change the electronic structure of Pd.[51] cavity surface, which extended the retention time of N2 and
Pang et al. prepared a series of alloys with different Pd:Ag thus facilitated its capture. The above features resulted in high
ratios and selected that (Pd1Ag1) with the best E-NRR activity. E-NRR activity (FE = 30.2%). Shi et al. prepared sub-nanoscale
The authors suggested that the introduction of Ag with a large Au supported on TiO2 (Au/TiO2) and evaluated its E-NRR per-
lattice constant (a = 4.09 Å) exerts a tensile strain that changes formance,[57] claiming that the presence of AuOTi moieties
the electronic structure (i.e., increases the surface exposure) of guarantees the stability of Au/TiO2 and the high dispersion of
Pd, further claiming that the low coordination state of surface Au, while the high atomic exposure provides abundant coordi-
atoms and the porous structure with a large specific surface natively unsaturated Au sites to accelerate the conversion of N2
area are of key importance for E-NRR performance enhance- to NH3. Bao et al. prepared tetrahedral Au nanorods (Au THH)
ment. As Pd-based catalysts exhibit prominent HER activity, with a high-index crystal plane (730) distribution (Figure 6b)[10]
much research has been directed at inhibiting the undesired and suggested that the low-coordinate Au atoms provided
HER. At present, the NH3 yields of Pd-based E-NRR catalysts by the stepped (730) plane surrounded by the (210) and (310)
have not yet achieved a major breakthrough, and surface modi- planes facilitate the adsorption of N2 and thus promote the
fication engineering and the mechanism of surface NH3 evolu- E-NRR. Another feasible strategy is the amorphization of the
tion require further exploration. crystalline phase of the metal catalyst. Previous studies sug-
Au-Based E-NRR Catalysts: Au-based materials are consid- gested the existence of an interaction between Au in the amor-
ered as potential E-NRR catalysts and have been extensively phized state and certain small molecules such as CO, H2, and
researched as such. On the one hand, the unoccupied d-orbital O2.[58] Li et al. loaded reduced graphene oxide with amorphous
of Au can accept the lone pairs of electrons from N2. On the Au nanoparticles and compared the obtained composite (a-Au/
other hand, the HER activity of Au-based materials is much CeOx-RGO) with crystalline nanoAu on RGO, revealing that
lower than that of Ru-, Pt-, and Pd-based ones, that is, Au-based a-Au/CeOx-RGO had higher E-NRR activity (Figure 6c).[59]
materials favor the E-NRR over the HER. At present, many This enhanced performance was ascribed to the fact that com-
studies have dealt with the development of Au-based E-NRR pared to crystalline Au, the metastable amorphous Au has a
catalysts, with the design strategy focusing on the control of larger number of dangling bonds and binds N2 more strongly.
micromorphology (or crystal plane) and the synergy between With the advancement of nanomaterial preparation methods,

Adv. Mater. 2021, 33, 2007650 2007650 (9 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 6. a) Schematic diagram of the formation of NPG@ZIF-8 nanocomposite. Adapted with permission.[53] Copyright 2019, Wiley-VCH. b) Geo-
metric model of the (730) crystal plane composed of (210) and (310) sub-facets. Adapted with permission.[10] Copyright 2017, Wiley-VCH. c) Schematic
diagrams of Au-CeOx-RGO and c-Au/RGO as electrocatalysts to drive electrochemical NRR, respectively. Adapted with permission.[59] Copyright 2017,
Wiley-VCH. d) Schematic diagram of Au/M-BOP synthesis and i) electrocatalytic nitrogen reduction reaction process; ii) SEM and iii) TEM images of
Au/M-BOP; iv) ammonia yield caused by Au/M-BOPs catalysts at different potentials; v) model diagram of N2 conversion to NH3 on Au/M-BOPs under
electrochemical conditions. Reproduced with permission.[62a] Copyright 2019, Royal Society of Chemistry. e) Charge density difference of N2 adsorption
on i) Au6/TiO2-VO and ii) TiO2-VO. Reproduced with permission.[64] Copyright 2020, Royal Society of Chemistry.

researchers have obtained nanomaterials with isolated Au atom of the Ti3C2 carrier.[61] The network structure of Ti3C2 effectively
distributions. For example, Au/C3N4 (C3N4-anchored single intercepted the passing N2 to increase reactant flux, and experi-
Au atoms) prepared using the structural confinement strategy ments and theoretical calculations showed that the interface
exhibited an E-NRR activity 22.5-fold higher than that of ordi- between Ti3C2 and high-index Au nanoclusters exhibits a high
nary Au nanoclusters.[60] N2 adsorption energy and can weaken the NN bond to destroy
Doping can also improve the E-NRR performance of Au- the E-NRR intermediate state and thus reduce the reaction
based materials, probably via the mutual regulation of elec- energy barrier. Closo-[B12H12]2− has excellent electron-donating
tronic structure. Functional carriers can affect the E-NRR properties and can simultaneously aid the in situ preparation
performance of Au sites. Liu et al. anchored Au nanoparticles on of nanoAu and stabilize the skeleton of the resulting catalyst.
2D-structured Ti3C2 (Au/Ti3C2), ascribing the enhanced E-NRR Zhao et al. prepared nanoAu anchored in metal-boron organic
performance of this composite to the “spider net” morphology polymers (Au/MBOPs) derived from closo-[B12H12]2−, showing

Adv. Mater. 2021, 33, 2007650 2007650 (10 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

that this catalyst featured excellent E-NRR performance (NH3 cluster.[66] The surface plasmon resonance of Au can generate
yield = 75.89 µg h−1 mgcat−1, FE = 10.35%). This superior per- a local electric field to promote the transfer of hot electrons on
formance was attributed to the highly dispersed Au active sites the Ru surface to achieve N2 activation and further hydrogena-
and the charge transfer environment provided by the carrier tion. The above authors made the most of the local plasmon
(closo-[B12H12]2−), while the high stability was ascribed to the resonance effect of nanoAu, which was not only applied in
stable carrier and the ability of closo-[B12H12]2− to chemisorb the photocatalyzed NRR, but also played a guiding role in the
transition metals (Figure 6d).[62] Recently, these authors inno- E-NRR field. In a prominent report, the researchers constructed
vatively anchored Au on a carrier (CB[7]-K2[B12H12]) with K+ an E-NRR catalyst containing Ni–Au pairs and showing an FE
constricted by closo-[B12H12]2− to simultaneously increase both of 67.8%.[67] Theoretical calculations suggested that the coopera-
NH3 yield and FE. According to the DFT calculations, the coor- tion between Ni and Au to promote the E-NRR is due to the
dination or repulsion of K+, which is coordinated by water to fact that Au can accept electrons from Ni into an electron-rich
form [K(H2O)6]+, effectively prevents H3O+ from approaching state and thus stimulate the smooth progress of the E-NRR
the Au surface, thereby suppressing the HER. At the same on Au (Figure 7a,b). It is worth noting that electron-rich Au-
time, the transferred charge enhances the interaction between based catalysts can well promote the E-NRR. Zheng et al. fab-
*N2 and the Au surface, thus reducing the energy barrier of ricated an unconventional E-NRR catalyst by depositing a layer
the rate-limiting step.[63] 2D nanoAu on OV-rich TiO2 (OVs = of CoOx on nanoAu to adjust the Au electronic structure, that
oxygen vacancies, i.e., the vacancies formed in metal oxides or is, to promote the conversion of Au nanoclusters to the highly
other oxygen-containing compounds when oxygen atoms in priced Au+ oxidation state.[68] Further theoretical calculations
the crystal lattice leave their original positions) exhibited a high suggest that the Au oxidation state can control the adsorption
NH3 yield of 64.6 µg h−1 mg−1 and an FE of 29.5% at different of N2 and reduce the reaction free energy of most intermediate
polarization potentials.[64] Theoretical calculations show that states in the E-NRR (Figure 7c,d). The effect of the Au oxida-
OVs can be used as electron donors for the E-NRR. The energy tion state on the E-NRR has not been adequately explored (both
barrier of the rate-limiting step greatly decreased after the intro- experimentally and theoretically), and the potential of Au-based
duction of nanoAu into OV-rich TiO2, which indicated that Au materials as E-NRR catalysts has not yet been fully tapped.
and OVs jointly promoted the E-NRR (Figure 6e). Liu et al. also Therefore, researchers not only need to rationally design
pointed out the synergistic effects of Au and OVs in CeO2 on the Au-based catalysts to make the most of its HER inertness and
electrochemical synthesis of NH3.[65] In addition, some studies d-electron structure characteristics, but they should also per-
have probed the effect of metal doping on the E-NRR perfor- form more spectral studies (similar to that of Yao et al.[69]) to
mance of Au-based catalysts, revealing the great potential of Ru explain the evolution of the E-NRR on Au.
and the excellent surface plasmon resonance effect of nanoAu. Compared with that of Ru, the use of Pd, Au, Rh, and Ag as
Hu et al. used these two advantages of Ru and Au to construct E-NRR catalysts has been less explored, and existing research
an E-NRR catalyst with Ru nanowires overlaid on the nanoAu has only considered the effects of morphology, as exemplified

Figure 7. a) HAADF-TEM image and EDX mapping results of Au–Ni couples. b) Electron transfer in Au–Ni couples promotes the electrochemical
conversion of N2 to NH3. Reproduced with permission.[67] Copyright 2019, American Chemical Society. c) Preparation process of Au/CoOx sample.
d) The electron density difference of Au/CoOx, where the cyan and pink isosurfaces respectively indicate i) the accumulation and depletion of electrons;
ii) schematic diagram of the free energy profile and optimized geometry of the NRR process (based on the distal mechanism) at the Au0 and Au1+
positions of Au/CoOx samples. Adapted with permission.[68] Copyright 2019, Wiley-VCH.

Adv. Mater. 2021, 33, 2007650 2007650 (11 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

by works on Ag nanosheets[70] and Rh nanosheets.[43a,71] As a coordination of N and Fe in the N-doped Fe-based catalyst (Fe-
state-of-the-art HER catalyst, Pt is hardly considered as a com- NC) ensures the dispersion and stability of Fe sites and allows
ponent of E-NRR catalysts, featuring extremely poor E-NRR for both HER suppression and E-NRR enhancement.[76] In
­selectivity.[72] In short, as noble metal–based catalysts can well addition, hybridization with OV-containing species can pro-
promote the HER, OER, ORR, and the CO2RR, researchers mote the E-NRR performance of zero-valent Fe-based catalysts.
tend to first consider the use of such catalysts in the initial stage Chu et al. doped Fe into CeO2 to form a Fe-CeO2 heterojunc-
of the E-NRR. Numerous studies have shown the problems of tion, showing that the resulting hybrid had a better E-NRR
noble metal–based E-NRR catalysts, that is, their tendency to performance than CeO2 (Figure 8c).[76e] On the one hand, Fe-
favor the HER and the inherently high cost. Undeniably, many doped CeO2 exhibited a nanosheet morphology that increased
precious metal-based catalysts perform well in the E-NRR and the exposure of active sites and electron transfer. On the other
can therefore be optimized in terms of morphology/electronic hand, Fe doping promoted the generation of OVs in CeO2 and
structure to take advantage of their superiorities and minimize thus reduced the E-NRR energy barrier. Furthermore, some
their disadvantages. For example, structural regulation can researchers have constructed N-doped FeMo monoatomic
inhibit the HER activity of Pt-based catalysts and thus release catalysts (such as FeMo@NG and Mo3Fe3C) according to the
their E-NRR potential. Hao et al. showed that N2 can be elec- compositional characteristics of the active factors in nitro-
trochemically reduced to NH3 at isolated Pt sites on a WO3 genase and demonstrated that these catalysts exhibit higher
plate, while the HER activity of these sites can be inhibited.[73] E-NRR activity than N-doped single-metal (Fe or Mo) ones
Specifically, the electron-rich Pt more easily adsorbs H atoms, (Figure 8d).[77] The geometric synergy of the FeMo dimer in
and the number of surface-adsorbed H atoms is positively FeMo@NG is believed to promote the E-NRR selectivity and
correlated to HER activity. Isolated Pt atoms are often coordinated the catalytic kinetics of the latter. Compared with zero-valent Fe,
by atoms such as O and N that anchor Pt to make it more electron- Fe oxides such as Fe2O3 and Fe3O4 are more abundantly studied
deficient and thus less prone to adsorb H atoms and be attacked by N2. as E-NRR catalysts, and researchers continue to increase their
NRR activity and inhibit their HER activity through fine struc-
tural design.[78] Many works on Fe-based E-NRR catalysts show
4.1.2. Non-Noble Metal Based E-NRR Catalysts that both pristine and doped Fe2O3 exhibit E-NRR activity. For
example, Fe2O3 nanorods,[79] α-Fe2O3,[80] and γ-Fe2O3[81] show
As the E-NRR performance of noble metal based catalysts E-NRR activity. The surface of iron oxides is often rich in OVs,
remains far from ideal, many non-noble (mainly transition) which can accommodate the Lewis-basic N2 and provide it with
metal based materials have been explored as alternatives. The electrons.[80a] Therefore, Fe3O4 also exhibits a certain E-NRR
empty d-orbitals of transition metals can accept the lone pairs activity (which can even exceed that of Fe2O3), as exempli-
of N2, while non-noble metal surfaces can adsorb N2 (via MN fied by Fe3O4/Fe and Fe3O4 supported on a Ti mesh, although
bond formation) to weaken the NN bond, with the resulting the origin of the synergy for factors other than OVs remains
p-feedback bonds promoting the formation of π*. The active unclear.[82] Furthermore, the E-NRR performance of Fe2O3 can
factor of natural plant nitrogenase, namely, the Fe/Mo cofactor, be enhanced via doping or anchoring by carriers, as exempli-
has inspired the design of numerous Fe- and Mo-based E-NRR fied by hybrids containing CNTs,[83] graphene oxide (GO),[84] or
catalysts. In addition, E-NRR catalysts based on early transition other transition metals.[77b,85] Specifically, substances such as
metals such as Ti have also been developed. Herein, based on CNTs and GO can not only improve the electron transport per-
the abundant recent research on the active (or main) compo- formance for the E-NRR, but also play a significant role in dis-
nents of non-noble metal–based E-NRR catalysts, we introduce persing and protecting the active center. Zhang et al. reported
the advances in the development of Fe-, Mo-, Ti-, and other an Fe-based E-NRR catalyst loaded with a non-precious-metal
non-noble metal based catalysts. perovskite and pointed out that the perovskite OVs promoted
Fe-Based E-NRR Catalysts: As one of the active components the adsorption and further hydrogenation of N2 (Figure 8e).[85a]
of nitrogenase, Fe is expected to effectively catalyze the E-NRR It should be noted that Fe-based catalysts may be oxidized (or
under ambient conditions. At present, research on Fe-based reduced) during the electrochemical process. To deeply under-
E-NRR catalysts mainly focuses on Fe and its oxides and the stand the origins of E-NRR activity, one needs to use advanced
regulation of their activity through doping. The application of testing methods to perceive changes in the catalyst during the
zero-valent Fe in the E-NRR is relatively rare, usually involving E-NRR. Although Fe acts as an active factor of nitrogenase,
the atomization of Fe active sites and the introduction of OVs the current E-NRR performance of Fe-based catalysts does not
to enhance E-NRR activity. Li et al. developed a graphene-sup- make this metal stand out. Thus, researchers should seek more
ported FeN3 catalyst for the E-NRR,[74] showing that the highly inspiration from the natural N2 fixation process to improve the
polarized Fe-N3 site is the key to promoting the E-NRR and E-NRR activity of such catalysts.
can be further weakened for effective N2 adsorption. Another Mo-Based E-NRR Catalysts: Mo is another active factor of
E-NRR catalyst based on monoatomic Fe was developed by Lv natural nitrogenase, and Mo-based catalysts are believed to
et al., who impregnated a ZIF framework with an Fe precursor have better E-NRR performance than Fe-based ones. Most
and subjected it to carbonization and acid treatment to obtain reported Mo-based E-NRR catalysts contain zero-valent Mo,[86]
an N-doped carbon framework catalyst with single-atom Fe dis- MoxSy,[87] MoxN,[88] and MoxC.[86h,89] Researchers often use
persion (ISAS-Fe/NC) (Figure 8a)[75] and perfect E-NRR activity active site exposure, morphological control, doping (with B,
(NH3 yield = 62.9 µg h−1 mgcat−1, FE = 18.6%). The active site C, N, and S), and electronic structure regulation to improve
was identified as FeN4 (Figure 8b), which indicated that the the E-NRR activity of Mo-based catalysts. In some studies, the

Adv. Mater. 2021, 33, 2007650 2007650 (12 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 8. a) Schematic diagram of the preparation route of ISAS-Fe/NC. b) There are three possible ways for the free energy distribution of NRR on
ISAS-Fe/NC, among which gray, red, sky blue, and white spheres represent C, Fe, N, and H atoms, respectively. Reproduced with permission.[75] Copy-
right 2019, Elsevier B. V. c) Conceptual diagram of electrochemical reduction of N2 to NH3 on Fe–CeO2 heterojunction. Reproduced with permission.[76e]
Copyright 2020, Royal Society of Chemistry. d-i) Schematic illustration of electrocatalytic reactor and MoFe@NG catalyst for N2 conversion to NH3;
Reproduced with permission.[77a] Copyright 2019, Royal Society of Chemistry. d-ii) Relative potential energies (ΔE) result of possible intermediates for
the conversion of N2 to NH3 on Mo2C (top), Mo3Fe3C (middle), and Fe3C (bottom), where the inset shows the structural optimization of N2 adsorp-
tion on the corresponding metal carbide. Reproduced with permission.[77b] Copyright 2018, Elsevier B. V. e) Possible pathway for the electrochemical
conversion of N2 to NH3 on LaFeO3 perovskite catalyst. Reproduced with permission.[85a] Copyright 2019, American Chemical Society.

E-NRR performance of zero-valent Mo was improved by atomic addition to zero-valent Mo, other Mo-based materials also
dispersion adjustment. Hui et al. prepared a graphite diacety- exhibit E-NRR activity. Defect-rich flower-like nanoMoS2 was
lene (GDY)-anchored monatomic Mo catalyst (Mo0/GDY) with reported to achieve an NH3 yield and FE of 29.28 µg h−1 mgcat−1
both E-NRR and HER activity (Figure 9a).[86e] In addition, an and 8.34%, respectively, while defect-free MoS2 achieved values
N-doped porous carbon–supported Mo single-atom E-NRR of only 13.41 µg h−1 mgcat−1 and 2.18%, respectively.[87b] Further
catalyst (SA-Mo/NPC) and a Mo single-atom E-NRR catalyst DFT calculations showed that defective MoS2 can reduce the
anchored on gt-C3N4 were reported.[86c,d] The large specific potential energy of the rate-limiting step in the E-NRR. Based
surface area of these carriers not only allows one to effectively on theoretical calculations and experiments, Zhang et al. con-
disperse Mo atoms and increase the exposure of active sites, cluded that the role of MoS2 in the E-NRR is similar to that in
but also ensures the rapid transmission of electrons during the the HER. Compared with the inert basal plane, the MoS2 edge
E-NRR. The crystal phase of Mo also affects its E-NRR activity. has a higher E-NRR activity, and the positive charge distributed
Wang et al. prepared Mo nanoparticles with (110) and (210) around the Mo atom can effectively accept the lone pair of N
crystal plane orientations, showing that the former catalyst fea- to facilitate N2 adsorption (Figure 9c).[87a] The E-NRR activity of
tured higher E-NRR activity than commercial Mo (210).[86a] The MoS2 can be increased by loading on a functional carrier. For
energy of N adsorption on Mo (110) was lower than that of H example, the E-NRR activity of MoS2 supported on reduced
adsorption, which was the main factor promoting the E-NRR. graphene oxide (RGO) exceeds that of unsupported MoS2,
Researchers have also prepared Mo-based materials in the form which can be attributed to the increased exposure of MoS2
of nanorods, obtaining an E-NRR performance at least 100 active sites and electron migration rate due to the 2D struc-
times that of commercial Mo foils, which was ascribed to the ture of RGO.[87d,e] N-doping and Co-doping can also change the
exposure of Mo (110) surfaces. Moreover, the doping of Mo into number of SVs on MoS2. Doping with N increases not only the
SnS2 was shown to adjust the catalyst electronic structure and exposure of MoS2 active sites, but also the conductivity of this
generate abundant S vacancies (SVs). Chu et al. believe that the material, thus favoring its performance in electrocatalysis with
thus formed SVs promote the activation of N2 at the MoSnS electron transfer.[87c] After doping with Co, the SVs in MoS2
sites and reduce the E-NRR energy barrier (Figure 9b).[86g] In were partially modified, which caused the E-NRR overpotential

Adv. Mater. 2021, 33, 2007650 2007650 (13 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 9. a) Synthesis of catalyst Mo0/GDY and evolution of catalytic reaction (includes hydrogen evolution reaction and N2 to NH3 reaction). Repro-
duced with permission.[86e] Copyright 2019, American Chemical Society. b) Electron contour maps of the (001) slice for i) SnS2 and ii) Mo-SnS2-Vs;
iii) free energy diagrams of NRR consecutive/enzymatic reaction pathways over the Mo–Sn–Sn trimer. Reproduced with permission.[86g] Copyright 2020,
Royal Society of Chemistry. c) MoS2 deformed charge density isosurface (top view), where i) red and green represent charge accumulation and loss,
respectively; ii) free-energy profile for NRR at MoS2 edge site; iii) the structure of key intermediates in the rate determination step; iv) the deformed
charge density of the *NNH intermediate (white, blue, cyan, and yellow represent H, N, Mo, and S elements, respectively). Adapted with permission.[87a]
Copyright 2018, Wiley-VCH. d) DFT simulation of the interaction between Vs and N2 on the MoS2−x basal planes (i: undoped Co; ii: doped Co); free
energy diagrams of the conversion of N2 to NH3 on defective MoS2−x and Co doped MoS2−x (with defective) (iii). Reproduced with permission.[87f]
Copyright 2019, American Chemical Society.

to decrease from 1.62 to 0.59 eV (Figure 9d).[87f ] In addition to N (or prepared using N-containing materials) should be evalu-
S, one can also use C, N, and B to change the electronic struc- ated with extreme caution using rigorous pollutant removal
ture and morphology of Mo, and the coordinated MoxN, MoxC, and 15N2 isotope tracking experiments.
and MoBC2 species exhibit certain E-NRR activity. The intro- Ti-Based E-NRR Catalysts: As a transition metal, Ti has
duction of C or B can change the position of the center of the empty d-orbitals that can interact with Lewis bases such as
Mo d-band (transfer to the Fermi level) to promote the adsorp- N2. The reported Ti-based E-NRR catalysts are mainly 2D
tion of N2 at Mo sites and weaken the NN bond.[86b] Unlike MXene (Ti3C2Tx) nanosheets, TiO2, and their derivatives. Ini-
other Mo-based catalysts, MoxN has a unique action mecha- tially, Ti3C2Tx was proven to exhibit E-NRR activity, with N2
nism in the E-NRR and has been proven to follow the MvK (or adsorbed on Ti in the middle of the MXene layer and not on
hybrid) pathway; that is, the N in NH3 comes from MoxN, and the MXene base surface. Thus, Ti is the active site of Ti3C2Tx.[91]
the generated N vacancies are compensated by the provided Zhao et al. showed that 2D Ti3C2Tx (T = F, OH) MXenes can
N2. Some researchers have shown that MoxN may decompose promote the E-NRR, achieving NH3 yields and FEs of up to
during the E-NRR and thus demonstrated that the detected 20.4 µg h−1 mgcat−1 and 9.3% (at −0.4 V), respectively. Moreover,
NH3 comes from the conversion of N in MoxN.[90] Therefore, DFT calculations suggest that the NN bond is elongated upon
we believe that the E-NRR performance of catalysts containing the adsorption of N2 on Ti3C2Tx.[20b] Generally, the F of HF used

Adv. Mater. 2021, 33, 2007650 2007650 (14 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 10. a-i) Synthesis mechanism of Ti3C2Tx nanosheets and ii) ammonia yield and FEof the NRR reaction driven by it. Reproduced with permis-
sion.[92] Copyright 2019, Royal Society of Chemistry. b) The E-NRR activity of MnO2–Ti3C2Tx is higher than that of the condition where i) MnO2 and
Ti3C2Tx act alone; ii) deformation charge density of N2 adsorbed MnO2–Ti3C2Tx heterostructure; iii) Gibbs free energy profile of N2 converted to NH3 on
the surface of MnO2(110)-MXene by distal pathway. Reproduced with permission.[95] Copyright 2019, Royal Society of Chemistry. c) Schematic diagram
of i) the preparation process of TiC/C NFs and ii) the Gibbs free energy profile of N2 converted to NH3 by enzymatic mechanism on it. Reproduced
with permission.[99] Copyright 2019, Royal Society of Chemistry. d) Schematic diagram of the preparation of Zr–TiO2 with oxygen vacancies (Vo) and
bi-Ti3+ sites and the activation of N2 on it. Reproduced with permission.[100] Copyright 2019, Springer Nature.

as the solvent is partially retained in Ti3C2Tx to reduce its con- NFs) prepared by electrospinning were found to exhibit E-NRR
ductivity, which seriously affects the electrochemical process, as activity. Further DFT calculations suggested that during TiC/C
it involves electron transport. Based on this, Li et al. developed NF-catalyzed E-NRR, N2 is converted to NH3 with a low energy
F-free Ti3C2Tx (T = O, OH) nanosheets, achieving an NH3 yield barrier according to the enzymatic pathway (Figure 10c).[99]
of 36.9 µg h−1 mgcat−1 (Figure 10a).[92] This result inspired others Moreover, Ti3+ was reported to play an important role in the
to increase the E-NRR catalytic activity and selectivity of Ti3C2Tx E-NRR. Cao et al. first predicted the formation of a bi-Ti3+ pair
by masking the residual F.[93] In addition, Ti3C2Tx with E-NRR on anatase TiO2 and the activation of N2 for the E-NRR based
activity can also enhance the electron transfer performance of on DFT calculations.[100] These authors introduced large Zr4+
MnO2.[94] The researchers speculated that the E-NRR perfor- ions (with a d-electron configuration and oxide structure) into
mance of MnO2-doped Ti3C2Tx should exceed that of both pris- TiO2, showing that the induced strain promoted the forma-
tine MnO2 and Ti3C2Tx, and this hypothesis was confirmed by tion of a bi-Ti3+ pair in TiO2 and thus favored the formation of
experiments and simulations (Figure 10b).[95] As TiO2 is a tran- OVs (Figure 10d). Further experiments proved that TiO2 with
sition metal oxide with low conductivity, the HER on its surface a bi-Ti3+ pair exhibited excellent E-NRR performance. Subse-
is suppressed. Zhang et al. deposited TiO2 nanosheet arrays on quently, Sun et al. reported that Ti3+-doped TiO2−x nanowires
Ti plates and found that the electrochemical process promoted exhibited good E-NRR activity, revealing that Ti3+ doping is the
OV formation and thus facilitated the adsorption and activation key to reducing the E-NRR energy barrier.[101]
of N2.[96] Qin et al. pyrolyzed MIL-125 (Ti) to obtain C-TixOy/C Other Non-Noble Metal Based E-NRR Catalysts: Along with
nanomaterials and adjusted the concentration of OVs and TiC Fe-, Mo-, and Ti-based catalysts, other transition metal (e.g.,
bonds therein through temperature control. Subsequent experi- Bi)-based catalysts are also used for the E-NRR. Qiao et al.
ments and DFT calculations revealed the promotional effect of constructed 2D Bi nanosheets with p-electron delocalization
OVs in Ti-based materials on their E-NRR performance.[97] Tran- through morphological engineering and electronic regula-
sition metal carbides also have the potential to adsorb N2, as tion, showing that the E-NRR performance of this catalyst was
excessive empty d-orbitals can promote π-orbital feedback and, at least ten times that of Bi nanoparticles.[102] In addition, the
hence, N2 activation.[98] TiC/carbon hybrid nanofibers (TiC/C E-NRR activity of Bi-based catalysts has been increased via

Adv. Mater. 2021, 33, 2007650 2007650 (15 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

defect engineering[103] and morphology control.[104] In another phthalocyanines conducted by Furuya et al.[150] As only the influ-
notable work on the Bi-based E-NRR catalysts by Hao et al.,[105] ence of the coordination metal in the phthalocyanine was con-
the doping of Bi nanocrystals with K+ increased the NH3 yield sidered, the question of whether the phthalocyanine molecule
and FE to 200 mmol g−1 h−1 (or 0.052 mmol h−1 cm−2) and 66%, has a synergistic effect inspired further exploration. Polyimide
respectively. These values were 1–2 orders of magnitude higher is commonly used in the E-NRR because of its good conduc-
than those reported for other E-NRR catalysts. The authors tivity. Initially, experimental observations and theoretical calcu-
believed that the 6p energy band of Bi partially overlapped with lations revealed that Li+ embedded into polyimide can prevent
the 2p orbital of N, and the resulting strong interaction resulted charge transport and HER occurrence (Figure 11a).[151] Inspired
in high E-NRR activity, with K+ stabilizing the intermediate of by this, Wang et al. showed that a polyimide material (PEBCD)
the E-NRR and regulating proton transfer. However, when Choi containing Li+ exhibited ideal E-NRR activity (carbon cloth was
et al. followed the above procedures[105] as closely as possible to used as a carrier).[152] Specifically, Li2SO4 was employed as the
replicate the K+-assisted activation of Bi nanoparticles for cata- electrolyte, and cyclic voltammetry (CV) was used to show that
lytic N2 reduction, the desired E-NRR activity was not observed, the electron-rich CO group of PEBCD bound the electron-
and a detailed account of this work as well as the related dis- deficient Li+ to form an OLi+ active site in the electrocatalytic
cussions have been published in Chemrix.[106] This discrepancy process, thus hindering the smooth progress of the HER on
further reminds researchers that they should carefully report PEBCD (Figure 11b). Further theoretical calculations suggested
their results, including detailed catalyst preparation methods that the OLi+ active sites on PEBCD can adsorb N2 to form
(with operation videos highly welcome), strict control of pollu- [OLi+]N2-Hx and promote the alternate hydrogenation of N2
tion sources, and the standardization of NH3 detection. To com- to NH3 (Figure 11c). The above report suggests that catalyst
prehensively present the real status of E-NRR catalyst research, surface engineering can suppress the undesired HER in the
Table 1 lists the active characteristics of non-noble metal based E-NRR process.
E-NRR catalysts that have not been discussed in detail herein.

4.2.2. Carbon-Based E-NRR Catalysts

4.2. Non-Metal-Based E-NRR Catalysts
For a long time, carbon-based materials have occupied an irre-
Despite the dominance of metal-based E-NRR catalysts, their placeable position in electrocatalysis, regardless of whether
non-metal-based counterparts have also drawn attention, offering directly involved in the catalytic process or used as carriers for
important advantages. 1) Although the nanometal prepara- hybrid materials. This phenomenon is due to the porous (or
tion process has been improved to allow the fabrication of layered) structure of carbon materials, their excellent electrical
single-atom-dispersed nanometal catalysts, the limited surface conductivity, abundant possibilities for surface engineering
exposure and inefficient loading still restrain the full utilization (defect introduction and heteroatom doping), and low cost. The
of metal active sites. 2) Even though the surface of a transition reported carbon-based E-NRR catalysts can be divided into pure
metal with d-orbitals can adsorb N2 to form MN bonds and carbon, single heteroatom doped carbon, and multiple heter-
thus weaken the NN bond, these metals can also form MH oatom doped carbon materials. Surface defects often become
species to promote the HER, which is easier to achieve than the the active sites for the adsorption of N2, and the heteroatom
E-NRR.[7a,26a,149] 3) In the E-NRR, the metals of many metal-based (e.g., N, B, P) doping of carbon materials often leads to changes
catalysts can be easily oxidized or reduced, which severely limits in their band gap, charge density, and spin density, thereby
their recycling. 4) Although metal consumption is not felt at the achieving the synergistic promotion of the E-NRR.[153] Generally,
laboratory scale, the consumption of large amounts of valuable the E-NRR activity of pure carbon-based materials is extremely
metal resources on the industrial scale is uneconomical. Previ- low. Li et al. obtained defect-rich CC-450 by simple heat treat-
ously, non-metal-based catalysts have been widely used in the ment of commercial carbon cloth (CC), and proved that this
OER, HER, ORR, and the CO2RR, and are expected to reach the catalyst had better E-NRR performance than the defect-free
level of metal-based catalysts; that is, non-metal-based catalysts CC.[154] However, many studies suggest that doping with heter-
may also promote the E-NRR. Of course, many non-metal-based oatoms having a strong electron affinity can increase the posi-
E-NRR catalysts such as polymer, carbon-based, and boron-based tive charge density of adjacent C atoms and thus stimulate the
materials have been developed, usually featuring large specific E-NRR activity of carbon-based materials.[155] In the subsequent
surface areas, the presence of dopants, compositional defects sections, we introduce the research status of carbon-based
(vacancies), and special electronic structures and thus exhibiting E-NRR catalysts, namely N-doped carbon-, B-doped carbon-,
satisfactory E-NRR activity. Herein, we introduce the research and oxygen group element doped carbon-based catalysts.
status of non-metal-based E-NRR catalysts, namely organic N-Doped Carbon-Based E-NRR Catalysts: N-doping can not
polymer, carbon-based, boron-based, and phosphorus-based only cause the charge transfer effect, but can also facilitate
ones, and put forward critical comments. the accumulation of positive charges on C atoms near N to
make the adsorption of the Lewis-basic N2 more favorable.
For example, researchers combined poly(acrylic acid) and a
4.2.1. Organic Polymer E-NRR Catalysts poly(ionic liquid), namely poly[1-cyanomethyl-3-vinylimida-
zolium bis(trifluoromethanesulfonyl)imide] (PCMVImTf2N),
The E-NRR activity of certain polymers has been noticed as in a dispersion of multi-wall carbon nanotubes (MWCNTs)
early as. This was inspired by the NRR experiment on metal in N,N-dimethylformamide and transferred the obtained

Adv. Mater. 2021, 33, 2007650 2007650 (16 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 1. E-NRR activity of some non-noble metal based catalysts.

Catalysts Electrolyte Potential (vs RHE) Ammonia yield FE [%] Ref.

−1 −1
B-TiO2 0.1 m Na2SO4 −0.8 V 14.4 µg h mgcat. 3.4 [107]
V-TiO2 0.5 m LiClO4 −0.4 V 17.73 µg h−1 mgcat.−1 15.3 [108]
VN NSs 0.1 m HCl −0.5 V 8.40 × 10−11 mol s−1 cm−2 2.25 [109]
VN NPs 0.05 m H2SO4 −0.1 V 3.30 × 10−10 mol s−1 cm−2 6.0 [110]
VN NW/CC 0.1 m HCl −0.3 V 2.48 × 10−10 mol s−1 cm−2 3.85 [111]
−1 −1
mVOx-rGO 0.1 m Na2SO4 −0.35 V 18.84 µg h mgcat. 16.97 [112]
Cr3C2@CNF 0.1 m HCl −0.3 V 23.9 µg h−1 mgcat.−1 8.6 [113]
Cr2O3 0.1 m Na2SO4 −0.9 V 4.96 × 10−11 mol s−1 cm−2 6.78 [114]
Cr2O3 nanofiber 0.1 m HCl −0.75 V 28.13 µg h−1 mgcat. −1 8.56 [115]
CrO0.66N0.56 1 mm H2SO4 2.0 V (cell voltage) 8.9 × 10−11 mol s−1 cm−2 6.7 [116]
Mn3O4 0.1 m Na2SO4 −0.8 V 11.6 µg h−1 mgcat. −1 3.0 [117]
CoO QD/RGO 0.1 m Na2SO4 −0.6 V 21.5 µg h−1 mgcat. −1 8.3 [118]
C@CoS@TiO2 0.1 m Na2SO4 −0.55 V 8.09 × 10−10 mol s−1 cm−2 28.6 [119]
Co–Nx–C 0.05 m H2SO4 −0.3 V 37.6 µg h−1 cm−2 17.6 [120]
(−0.9 V)
CoP HNC 1 m KOH −0.4 V 5.39 µg h−1 cm−2 7.36 [121]
CoS2/NS-G 0.05 m H2SO4 −0.05 V 25.0 µg h−1 mgcat.−1 25.9 [122]
GDY/Co2N 0.1 m HCl −0.2 V 219.72 µg h−1 mgcat.−1 58.60 [123]
−1 −1
Mo–Co 0.1 m Na2SO4 −0.1 V 89 µg h mgcat. 13.5 [124]
Fe–Ni LDH 0.1 m Na2SO4 −0.5 V 1.72 × 10−10 mol s−1 cm−2 23 [125]
N-NiO/CC 0.5 m LiClO4 −0.5 V 22.7 µg h−1 mgcat.−1 7.3 [126]
BCN/Ni 0.1 m HCl −0.3 V 16.72 µg h−1 cm−2 13.06 [127]
Ni 0.1 m LiCl 1.8 V (cell voltage) 3.58 × 10−11 mol s−1 cm−2 17.2 [128]
Porous Ni Propanol/H2O — 1.54 × 10−11 mol s−1 cm−2 0.89 [129]
Cu/PI-300 0.1 m KOH −0.3 V 12.4 µg h−1 cm−2 6.56 [130]
Cu-TiO2/CP 0.5 m LiClO4 −0.55 V 21.31 µg h−1 mgcat.−1 21.99 [131]
−1 −1
Nb3O7(OH)/CFC 0.1 m Na2SO4 −0.4 V 622 µg h mgcat. 39.9 [132]
Nb2O5 nanofiber 0.1 m HCl −0.55 V 43.6 µg h−1 mgcat.−1 9.26 [133]
NbSe2 nanosheet 0.1 m Na2SO4 −0.4 V and 89.5 µg h−1 mgcat. −1 13.9 [134]
−0.45 V
SnO2 0.1 m Na2SO4 −0.7 V 1.47 × 10−10 mol s−1 cm−2 2.17 [135]
Sn/SnS2 0.1 m PBS −0.8 V 23.8 µg h−1 mgcat. −1 6.5 [136]
SnS2@Ni 0.1 m Na2SO4 −0.5 V 9.17 × 10−10 mol s−1 cm−2 10.8 [137]
LaF 0.5 m LiClO4 −0.45 V 55.9 µg h−1 mgcat. −1 16 [138]
LaFeO3 0.1 m HCl −0.55 V 18.59 µg h−1 mgcat.−1 8.77 [139]
2D W2N3 0.1 m KOH −0.2 V 11.66 µg h−1 mgcat.−1 11.67 [140]
W18O49 0.1 m Na2SO4 −0.2 V 5.3 µg h−1 mgcat. −1 12.1 [141]
PC/Sb/SbPO4 0.1 m HCl −0.25 V 25 µg h−1 mgcat.−1 31 [142]
−0.15 V (−0.25 V)
Sb2Te3 0.1 m KOH 0V 13.8 µg h−1 cm−2 27.7 [143]
Sc/NC 0.1 m HCl −0.1 V 20.40 µg h−1 cm−2 11.2 [144]
Y/NC 0.1 m HCl −0.1 V 23.2 µg h−1 cm−2 12.1 [144]
Ta2O5 0.1 m HCl −0.7 V 15.9 µg h−1 mgcat.−1 8.9 [145]
Re2MnS6 0.1 m Na2SO4 −0.3 V 15 µg h−1 mgcat.−1 14 [146]
Fe-ReS2 0.1 m Na2SO4 −0.2 V 80.4 µg h−1 mgcat. −1 12.3 [147]
DyF3 0.1 m Na2SO4 −0.45 V 10.9 µg h−1 mgcat.−1 8.8 [148]

Adv. Mater. 2021, 33, 2007650 2007650 (17 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 11. a) Schematic illustration of further passivation of HER activity by reducing the number of active sites via Li+ association (brown, green, red,
blue, and purple spheres represent C, H, O, N, and Li atoms, respectively). Adapted with permission.[151] Copyright 2016, Wiley-VCH. b) CV curves of
PEBCD/C electrode and carbon cloth electrode (in 0.5 m Li2SO4 electrolyte at different pH values). c) Prediction of relative energy of electrochemical
reduction of N2 to NH3 at [O–Li+] active site. Reproduced with permission.[152] Copyright 2017, American Chemical Society.

mixture to a glass substrate. Subsequent pyrolysis at 900 °C (N2-TPD) experiments. This behavior inspired researchers to
afforded an N-doped porous carbon membrane (NCM).[156] focus on the forms of N in these three carbon materials. NPC-
Although the pristine NCM had a certain E-NRR activity, the 750 and NPC-850 mainly contained pyridinic and pyrrolic N
corresponding NH3 yield and FE were only 0.08 g h−1 m−2 (pyridinic N was more abundant in NPC-750), while NPC-950
(8 µg h−1 cm−2) and 5.2%, respectively, whereas much higher mainly contained graphitic N. This shows that the presence of
values of 0.36 g h−1 m−2 (36 µg h−1 cm−2) and 22%, respec- pyridinic and pyrrolic N in NPC is the key to its high E-NRR
tively, were observed after the NCM had been doped with Au. performance, whereas the presence of graphitic N does not
In another study, a mixture of cicada sloughs and ZnCl2 was enhance this performance. The origin of the influence of dif-
pyrolyzed to obtain a micro/mesoporous nitrogen-rich carbon ferent N forms was probed by DFT calculations, which showed
material (NCF-900) with notable E-NRR activity, which was that the pyridinic and pyrrolic nitrogen of NPC act as the active
ascribed to the high content of doped N and the porous mor- sites of the E-NRR, while the adsorption of N2 on graphitic N
phology.[157] Lv et al. obtained nitrogen-vacancy (NV)-enriched is difficult (Figure 12c). In addition, the porous structure of
metal-free polymer carbonitrides (PCN-NV4) by pyrolyzing NPC is also conducive to N2 capture and active intermediate
the products of melamine polycondensation[153] and showed stabilization, with higher HER overpotentials leading to an
that PCN-NV4 can effectively catalyze the E-NRR (NH3 yield = increased selectivity for the E-NRR. Some studies have shown
8.09 µg h−1 mgcat−1, FE = 11.59%). DFT calculations revealed that carbon defects can enhance the E-NRR activity of carbon-
that the binuclear end (two C atoms connected to the imida- based materials. Mukherjee et al. pyrolyzed ZIF-8 to obtain an
zole) near a given NV can accelerate through-space electron N-doped porous carbon-based material, achieving an NH3 yield
transfer to facilitate N2 adsorption and significantly elongate of 3.4 × 10−6 mol h−1 cm−2 (57.8 µg h−1 cm−2) and an FE of 10.2%
the NN bond to afford a highly activated state (Figure 12a). (at −0.3 V vs RHE).[159] Further DFT calculations suggested that
As the E-NRR activity of PCN-NV4 significantly exceeded that carbon vacancies at sites comprising three pyridinic N atoms
of NV-free PCN, NV defect engineering was concluded to may be the active sites for the E-NRR, favoring the adsorption of
increase the E-NRR activity of carbon-based materials. The N2 and the dissociation of the NN bond (Figure 12d). In addi-
origin of the enhanced E-NRR activity of N-doped carbon-based tion, the presence of Fe or alkali metals was shown to affect the
materials was further probed by detailed experiments and the- E-NRR activity of N-doped porous carbon. In particular, Fe can
oretical calculations.[158] Organic precursor (ZIF-8) pyrolysis inhibit the E-NRR active sites to promote the undesired HER,
yielded N-doped porous carbon (NPC, Figure 12b) that achieved whereas the electron-donating effect of alkali metals such as K+
an NH3 yield of 1.4 mmol h−1 g−1 (23.8 µg h−1 mg−1), and the is conducive to the smooth progress of the E-NRR. Song et al.
chemical adsorption capacity of NPC-750, NPC-850, and NPC- reported an unusual E-NRR catalyst that employs the cutting-
950 (the number represents the pyrolysis temperature) for N2 edge electric field effect to activate the E-NRR process.[160] Spe-
was determined to increase in the order NPCS-950 < NPCS- cifically, the authors constructed a metal-free N-doped carbon-
850 < NPCS-750 using temperature-programmed N2 desorption based material with nanospikes, achieving an NH3 yield of

Adv. Mater. 2021, 33, 2007650 2007650 (18 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 12. a-i) The geometry of PCN (with nitrogen vacancy) with N2 adsorbed; ii,iii) the charge density difference of PCN (with nitrogen vacancy)
adsorbed with N2, where the yellow and blue isosurfaces represent the accumulation and reduction of charge, respectively; iv) free energy map of NRR
converted to NH3 on PCN with nitrogen vacancies. Adapted with permission.[153] Copyright 2018, Wiley-VCH. b) Schematic diagram of NPC prepara-
tion. c) Free energy diagram for converting N2 to NH3 on NPC with i) pyridine N and ii) pyrrole N. Reproduced with permission.[158] Copyright 2018,
American Chemical Society. d-i) Schematic diagram of the structure of three pyridine N with carbon vacancies, (blue, gray, and white balls represent N,
C, and H atoms, respectively); ii) free energy prediction of N2 conversion to NH3 at 3N position in N-doped carbon. Reproduced with permission.[159]
Copyright 2018, Elsevier B. V.

97.18 ± 7.13 µg h−1 cm−2 and an FE of 11.56 ± 0.85% (at −1.19 V vs and experimentally.[161a] The E-NRR activity of this material was
RHE). Further control experiments showed that the nanospike rationalized as follows. 1) After doping with B, graphene does
tip can form a local electric field that promotes the capture and not lose its sp2 orientation and maintains a 2D planar conju-
hydrogenation of N2 near the electrode to afford NH3. The pro- gated structure with a high surface area, which favors the full
motional effect of the counter cation in aqueous solutions on exposure of catalytically active sites. 2) The electron deficiency
the nanospike E-NRR activity increased in the order of K+ < of B allows it to withdraw electrons from the graphene network
Na+ < Li+, and the limitation of the aqueous environment near and thus promote the attachment of N2 via its lone pair. 3) The
the nanospike tip was shown to inhibit the HER and facilitate electron-deficient B can chemically adsorb the negatively polar-
the entry of N2 into the electric field area. Although N-doping ized O atoms[162] to inhibit the normal progress of the HER
can improve the E-NRR performance of carbon-based mate- and thus increase the selectivity for the E-NRR. Additionally,
rials to a certain extent, researchers need to carefully handle the the lower electronegativity of B (2.04) compared to that of C
NH3 pollution due to the presence of nitrogen in catalyst mate- (2.55) may lead to a significant differentiation of electron den-
rials, for example, using strict pollution pre-control procedures sity and thus change the equilibrium state of the carbon ring
and 15N2 isotope labeling. structure in B-doped graphene (Figure 13a). The LUMO and
Boron-Doped Carbon-Based E-NRR Catalysts: The electron- HOMO of non-doped and B-doped graphene show that after
deficient B atom is Lewis-acidic and thus has the potential to doping (B replaces some C atoms), the electron density is rear-
interact with the Lewis-basic N2. As predicted, researchers ranged, and the uniform distribution of graphene electrons is
have developed B-doped carbon-based materials with E-NRR lost (Figure 13b). Subsequently, ultrasound and heat treatment
activity.[161] Zheng et al. developed a B-doped graphene (BC3) were used to convert commercial bulk B4C to B-doped gra-
and confirmed its excellent E-NRR activity both theoretically phene quantum dots (BGQDs), which achieved an NH3 yield

Adv. Mater. 2021, 33, 2007650 2007650 (19 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 13. a) Schematic diagram of the atomic orbital of BC3 after combining N2. b) LUMO (blue) and HOMO (red) of B doped with G (left) and
undoped G (right). Reproduced with permission.[161a] Copyright 2018, Elsevier B. V. c) Schematic diagram of preparation of B4C–BGQDs. Reproduced
with permission.[161b] Copyright 2019, Royal Society of Chemistry. d) Schematic of the computational models with B–N pairs (gray, pink, red, blue, and
green balls represent C, B, O, N, and H atoms, respectively). e) Free energy evolution diagram of N2 converted to NH3 on BN-1 (the inset is the geo-
metric structure of BN-1). Adapted with permission.[161c] Copyright 2019, Wiley-VCH.

(28.6 µg h−1 mgcat−1) exceeding that observed for boron carbide (NSG) promoted the electrochemical conversion of N2 to NH3,
nanosheets (Figure 13c).[161b] Although the authors believed that featuring higher E-NRR activity than non-doped or single ele-
B4C quantum dots promote fast charge transfer and enhance ment (S or N)-doped graphene.[164] DFT calculations suggested
N2 adsorption, the related action mechanism should be probed that S/N co-doping produces more defects in graphene and
deeper, as in the work of Wang et al. Chen et al. reported a promotes the transfer of electrons, allowing NSG to adsorb
B/N co-doped graphene as an E-NRR catalyst and ascribed its N2 and weaken the NN bond. S/N co-doped carbons can also
enhanced activity to the presence of BN pairs.[161c] The E-NRR be prepared by pyrolyzing CC impregnated with ammonium
inertness of graphene was predicted by theoretical calculations persulfate as an N and S source. The S/N co-doped defective
and confirmed through experiments. However, heteroatom carbon (CC-APS-800) prepared in this way exhibited E-NRR
doping can adjust the electronic structure of graphene and thus activity (NH3 yield = 9.87 × 10−10 mol s−1 cm−2, FE = 8.11%) that
enhance its catalytic activity. Inspired by this strategy, Chen was ascribed to the abundant defects and heteroatoms. Besides,
et al. prepared B- and N-doped graphene and demonstrated O-doped carbon-based materials such as amorphous oxygen-
(both experimentally and theoretically) that the E-NRR activity doped carbon nanosheets,[165] oxygen-doped hollow carbon
of graphene doped with a single heteroatom cannot reach that microtubes,[166] and oxygen-doped graphene[167] are also used as
of B/N co-doped graphene.[161c] Therefore, E-NRR activity was E-NRR catalysts. Wang et al. used DFT calculations to show that
ascribed to the presence of BN pairs, and C atoms near these the CO and OCO groups formed in oxygen-doped carbon-
pairs were identified as the E-NRR active sites (Figure 13d,e). As based materials promote the E-NRR, whereas CO bonds do
discussed earlier, doping with B and N makes the neighboring not significantly contribute.[167]
C atoms positively charged and thus suitable for N2 adsorption.
Oxygen Group Element Doped Carbon-Based E-NRR Catalysts:
In addition to B or N doping, doping with O, S, and P can also 4.2.3. Boron-Based E-NRR Catalysts
be used to adjust the electronic properties of carbon-based mate-
rials and thus change their electrochemical performance.[163] As mentioned earlier, the electron deficiency and low elec-
Above, we mentioned that O can affect the E-NRR performance tronegativity of B determine its ability to influence electron
of electrocatalysts via the formation of an OLi+ active state.[152] transfer in the E-NRR. B-based non-metallic materials are
Some researchers also reported the effect of S doping on the effective E-NRR catalysts with experimentally and theoretically
E-NRR performance of carbon-based materials. Sulfur quantum confirmed activities.[31,168] Sun et al. reported that metal-free
dot-graphene nanocomposites were shown to have good B4C nanosheets (Figure 14a) exhibited an NH3 yield and FE of
E-NRR activity, featuring an NH3 yield of 28.56 µg h−1 mgcat−1 26.57 µg h−1 mgcat−1 and 15.95% (at −0.75 V vs RHE), respectively.
and an FE of 7.07%.[164] In addition, the synergistic effect of The evolution of E-NRR on the B4C (110) surface was probed by
S/N co-doping was revealed to increase the catalytic activity of DFT calculations, and the *NH2−NH2 → *NH2−NH3 pro-
graphene.[163d] Tian et al. reported that S/N co-doped graphene cess was considered to be the rate-limiting step.[168d] According

Adv. Mater. 2021, 33, 2007650 2007650 (20 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 14. a) Transmission electron microscope image of B4C nanosheet. b) Free energy diagram of N2 converted to NH3 on B4C (see ref. [168d] for
details). Adapted with permission.[168d] Copyright 2018, Wiley-VCH. c) The optimal structure of N2 adsorbed on 3B and 3B(H) respectively. d) The dif-
ference in electron density between 3B and 3B(H) after adsorption of N2. e) Free energy diagram of N2 converted to NH3 at 3B position. f) Free energy
diagram of N2 converted to NH3 at 3B(H). g) Changes of 3B(H) DOS before and after adsorption of N2. Reproduced with permission.[168e] Copyright
2019, American Chemical Society.

to the suggested mechanism, N2 is captured by distal adsorp- on 3B (the three adjacent B atoms on B4C were defined as 1B,
tion and then bound to B at both ends of B4C via the enzy- 2B, and 3B) is advantageous and occurs in the form of remote
matic pathway. Calculations show that the initial conversion of association, with the NN bond concomitantly stretching from
*NN to *NH*NH2 and that of *NH*NH2 to *NH2*NH2 1.112 to 1.132 Å (Figure 14c). 2) A large amount of charge accu-
are easy to implement, with the barrier of the rate-limiting mulates at the junction of 3B and N2 molecules, and differen-
step (*NH2*NH2 → *NH2 + *NH3) equaling only 0.34 eV tial charge analysis shows that a charge of 0.38 e− is injected
(Figure 14b). Although the free energy of NH3 desorption into N2 molecules (Figure 14d). 3) DFT predicts the desorption
(*NH3 → NH3) reaches 1.73 eV, the accumulated free energy of NH3 to be difficult and suggests that the increase in NH3
of 2.7 eV is sufficient to complete this process spontaneously. coverage will reduce the NH3 desorption energy to 0.49–1.11 eV,
In another similar work, a 2D B nanosheet material (BNS) was revealing the occurrence of NH3 accumulation on the electrode
prepared and shown to selectively promote the E-NRR.[168e] It is surface and thus showing that NH3 can smoothly leave the
worth noting that this study reported detailed DFT calculation catalyst surface (Figure 14e). 4) The surface hydrogenation of
results that fully revealed the mechanism of the E-NRR on BNS B (104), which becomes the active site of the E-NRR, can effec-
and can be summarized as follows. 1) The adsorption of N2 tively promote the E-NRR and inhibit the HER (Figure 14f).

Adv. Mater. 2021, 33, 2007650 2007650 (21 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 15. a) B atom sp3 hybrid orbitals can form B-N π-back bonding with N. b) Proposed 8 kinds of 2D materials and potential B-sites. c) Volcano
diagrams for the 14 catalysts for N2 to NH3. d) Relationship between ΔGmaxHER and ΔGmaxNRR on 14 catalysts. e) Based on the ΔGmaxHER and
ΔGmaxNRR versus Bader charge of single boron in or on 2D substrate, the calculation results of 14 catalysts combination were obtained. Reproduced
with permission.[168f] Copyright 2019, American Chemical Society.

5) The 2p orbital of B has a certain degree of overlap with the B can promote the E-NRR and inhibit the HER, which provides
2p orbital of *N2, partly because of σ-donation and π feedback a theoretical basis for the selection (design) of B-doped cata-
bonding (Figure 14g). The empty d-orbitals of transition metals lyst carriers. Sun et al. performed theoretical guidance experi-
can accept the lone pair of N2, and the sp3 hybridization of B ments to show that P doping can promote the E-NRR activity
atoms can increase the BN feedback bond and thus increase of B by weakening the NN triple bond and enhancing B-N2
the NN π* anti-bonding orbital (Figure 15a). Based on this, binding.[169] In summary, the experimental exploration of
Liu et al. inserted elemental B into eight kinds of 2D materials B-based E-NRR catalysts with a theoretical research background
to construct 21 types of B-based conceptual catalysts for stud- should be further investigated, as B is a cheap element that can
ying the local charge transfer (in carriers) and bonding modes regulate E-NRR activity.
of B in the E-NRR (Figure 15b).[168f ] Owing to structural relaxa-
tion and the larger distance between B and N, they first ruled
out seven unsuitable models (G-S1, BN-D1, BS-D1, BS-D2, S-A, 4.2.4. Phosphorus-Based E-NRR Catalysts
S-S1, and T-S1) and presented the remaining 14 models in the
volcano diagram of the relationship between ΔGmax and E(N*). Although P-based materials are less frequently used in the
As GA, BN-A, and h-S1 require low energy input, they are the E-NRR, they are potential E-NRR catalysts. On the one hand,
preferred E-NRR catalysts (Figure 15c). In addition, the authors P and N are in the same group of the periodic table and have
compared the energy requirements of the above 14 model mate- similar valence electron configurations (P: 3s23p3; N: 2s22p3),
rials in the HER and the NRR. In the relationship between thus having similar compatible properties.[170] On the other
∆GmaxHER and ∆GmaxNRR (Figure 15d), G-A, h-S1, BP-A, and t-S1 hand, black P has a layered crystal structure with interlayer
lie above the red dotted line, that is, feature less H adsorption van der Waals interactions[171] and can therefore be further
and thus have good E-NRR selectivity. Figure 15e presents the exfoliated into single layers that provide abundant active sites
relationship between ∆GmaxHER, ∆GmaxNRR, and Bader charge, for N2 adsorption. At present, some phosphorus-based mate-
revealing that although small or even negative δ can reduce the rials have been shown to have excellent E-NRR activity.[172]
energy required for the E-NRR, it also causes a large ΔG(H*). Wang et al. used fully exfoliated black P nanoplatelets (FL-BP
In short, positively charged B was concluded to promote the NSs) as E-NRR catalysts to achieve a high NH3 yield of
undesired HER. On the contrary, as B → N2 charge transfer 31.37 µg h−1 mgcat−1.[172c] Further DFT calculations indicate that
enables the adsorption and activation of N2, negatively charged the molecular orbital of the asymmetric electron distribution of

Adv. Mater. 2021, 33, 2007650 2007650 (22 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

FL-BP NSs is the key to E-NRR activity. The HOMO and LUMO with NH3 (e.g., that of air, electrocatalytic equipment, reactors
electron densities of FL-BP NSs are concentrated on the jagged or electrolytes, and catalysts), accuracy of product detection,
edges, and N2 can therefore be converted to NH3 via an asso- and accurate identification of catalyst active centers. To promote
ciation mechanism (alternative hydrogenation) on these edge the healthy development of the E-NRR, we introduce the related
structures. Ding et al. believed that the poorly conductive black operation procedures in detail and point out the existing prob-
P is not conducive to charge transfer in the E-NRR and there- lems and correct treatment methods, thus providing a quick
fore used SnO2−x nanotubes with electrochemical activity as introduction and reference for researchers and evaluators (such
conductive carriers. These authors combined P and Sn in a self- as the reviewers of research reports) to facilitate the establish-
assembled form (Sn-P coordination) to anchor black P quantum ment of standards in the E-NRR field.
dots (BP QDs) in SnO2−x nanotubes (BP@SnO2−x).[172d] In this
way, the BP@SnO2−x composite integrates the functions of BP
and SnO2−x nanotubes and promotes the smooth progress of 5.1. Electrocatalytic Devices
the E-NRR. In addition, the N/O co-doping of a phosphorene
E-NRR catalyst was shown to hinder the HER and allowed The electrocatalytic device in which the E-NRR occurs com-
one to achieve an NH3 yield and FE of 18.79 µg h−1 mg−1 and prises the electrochemical workstation, electrolytic cell, and
21.51%, respectively.[172a] DFT calculations showed that this high electrolyte as the core components. In general, the electrochem-
E-NRR activity was due to the increased hydrophobicity of the ical workstations of regular manufacturers can meet the needs
N/O co-doped phosphorene surface, which hindered the HER of E-NRR experiments, and researchers need to carefully read
in an alkaline environment and promoted the smooth progress the instructions for specific models and accurately calibrate
of the E-NRR. From this viewpoint, P-based materials are also their parameters according to the operation manual.
potential E-NRR catalysts and should be experimentally and
theoretically studied to increase their E-NRR activity.
5.1.1. Electrolytic Cells

5. E-NRR Experimental Procedure and Its At present, the reported electrolytic cell equipment mainly
includes back-to-back cells, polymer electrolyte membrane
Challenges
(PEM)-type cells, H-type cells, and single-chamber cells.
The electrochemical conversion of N2 to NH3 under ambient Figure 16a shows a schematic diagram of a back-to-back cell
conditions is a green and sustainable strategy that can not only structure including two chambers separated by a proton-
alleviate the high energy consumption and environmental pol- exchange membrane. In this structure, each chamber is placed
lution caused by the Haber–Bosch process, but can also meet near the proton-exchange membrane with a catalyst, gas dif-
the convenience of industrial and agricultural production. At fusion layer (GDL), and electrodes. In back-to-back cells, the
present, this seemingly perfect process is facing big challenges E-NRR is mainly performed in the cathode cell, while the anode
such as extremely low NH3 yields, widespread contamination cell generates O2. In a system based on back-to-back cells, a

Figure 16. Simple structure diagram of a) back-to-back electrolytic cell, b) polymer electrolyte membrane (PEM) type electrolytic cell, c) H-type cell,
and d) single chamber cell.

Adv. Mater. 2021, 33, 2007650 2007650 (23 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

solid electrolyte polymer membrane (such as the perfluorosul- electrochemical reduction. In the same way, one can possibly
fonic acid proton-conductive membrane produced by DuPont apply the flow cell to the E-NRR to solve the same problem. In
(Nafion)) separates the cathode from the anode and conducts this case, electrolyte circulation increases the time of contact
protons to achieve the E-NRR in the cathode cell. The widely between N2 and the electrolyte and thus increases the solubility
used Nafion proton-exchange membrane offers the benefits of of N2 in the electrolyte.
a mature production process and high electrical conductivity
under ambient conditions. Currently, back-to-back cells have
been introduced in some reviews (Amar et al., Garegounis 5.1.2. Electrolytes
et al., Gidde et al., and Kyriakou et al.)[5b,c,173] and experimental
reports (Xu et al. and Tao et al.).[72a,d,174] Although these reports E-NRR occurs on the surface of the electrode (catalyst), while
use back-to-back cells, not all of them are used in the water- electrolytes act as the media for N2 enrichment, proton transfer,
based E-NRR; they also include the electrochemical synthesis of and electron transport. Here, we review the electrolytes used
NH3 from N2 and H2. The overall structure of a PEM-type cell for the E-NRR at ambient temperature, including aqueous,
is similar to that of a back-to-back cell except for the fact that solid, organic solvent, ionic liquids, and lithium-mediated
in the former case, the E-NRR occurs in an aqueous environ- electrolytes.
ment. A PEM-type cell is divided into the cathode chamber and Aqueous Electrolytes: Aqueous electrolytes such as LiCl, HCl,
the anode chamber separated by a proton-exchange membrane. H2SO4, Na2SO4, LiClO4, and KOH solutions are the ones most
The E-NRR occurs in the cathode chamber, while the anode widely used for the E-NRR. Liquid electrolytes generally have
chamber produces O2 (Figure 16b). The aqueous environment the advantages of simplicity, economy, easy addition/recycling,
increases the accuracy of the applied potential (because of the and extremely high fluidity (plasticity), thus being appro-
introduction of the reference electrode) and enhances the con- priate for various electrolytic cells. Generally, aqueous electro-
ductivity of the proton-exchange membrane to increase the lytes can enhance proton exposure and transfer efficiency, but
proton transmission efficiency. However, in this case, NH3 pro- they also promote the HER. However, proper modification of
duced in the cathode chamber was reported to pass through aqueous solutions can increase the selectivity for the E-NRR.
the proton-exchange membrane to induce cross-contamination. For example, the HER can be suppressed by the addition of
Therefore, a GDL is often provided on the side of the cathode organics (such as methanol, isopropanol, and ethylenediamine)
chamber to reduce NH3 migration. For detailed reports, into aqueous systems or through the use of electrolytes con-
researchers can refer to the work of Kordali and Chem.[36,83a,175] taining Li+ ions that can act on O-containing catalysts to form
The H-type cell is the most widely used electrolytic cell in OLi+ active sites.[152] In view of the wide range of sources
water-based E-NRR systems, accounting for >90% of the and absence of carbon traces remaining in the environment,
reported works. Figure 16c shows a schematic structure of the aqueous electrolytes have become the ultimate goal of the
H-type cell, which includes a cathode chamber and an anode E-NRR, which is also of interest to most researchers.
chamber filled with the electrolyte and separated by a proton- Solid Electrolytes: Solid electrolytes (SEs), which are mainly
exchange membrane. The working and reference electrodes used in back-to-back cells, generally have high ion transmis-
are assembled in the cathode chamber, the counter electrode sion efficiencies and low H2O contents, which ensures the
is located in the anode chamber, and the E-NRR occurs in the smooth progress of the E-NRR while suppressing the HER.
cathode chamber. This advanced electrocatalytic system allows In the typical work of Xu et al., SmFe0.7Cu0.3−xNixO3 and NiO-
one to precisely control the potential applied to the working Ce0.8Sm0.2O2−δ were used as SEs to realize the electrochemical
electrode and to easily add and separate raw materials and reac- conversion of N2 to NH3 at 25–100 °C, with H2 employed as the
tants. However, as NH3 in H-type cells was reported to migrate H source.[174a] As Nafion was selected as the proton-exchange
from the cathode chamber to the anode chamber, one should membrane in this work, the absence of water at higher tem-
monitor the NH3 content of both chambers or develop more peratures greatly reduced the proton conduction efficiency
advanced proton-exchange membranes. The single-chamber of this membrane. Later, the researchers introduced H+/Li+/
cell is an electrocatalytic device equipped with a three-electrode NH4+ into the Nafion membrane to avoid the attenuation of
system without a proton-exchange membrane (Figure 16d). As proton transmission efficiency.[72d] However, the introduced Li+
the OER and the HER can be successfully completed in one competed with protons to reduce the rate of N2 protonation to
chamber, some researchers have performed the E-NRR in NH3. Moreover, the introduced NH4+ unavoidably caused NH3
single-chamber cells.[43a,128–129] It should be noted that although pollution, making it unclear whether the detected NH3 really
the single-chamber cell avoids the use of proton-exchange came from the electrochemical reduction of N2. In addition,
membranes, the generated NH3 is easily oxidized to NOx− on researchers have developed a new type of polymer gel SE that
the anode. Therefore, the use of a single-chamber cell for the can limit the transmission of water and increase the selectivity
E-NRR is not recommended. Control of the cathode and anode for the E-NRR.[176] Although SEs are rarely used in the E-NRR,
products in two isolated chambers allows one to independently they deserve further research. However, to be practically appli-
control the catalyst and reaction type in the two chambers. cable, SEs should have i) physical and chemical properties that
The solubility of N2 in most common electrolytes is are stable during the E-NRR, ii) a wide electrochemical sta-
extremely low, and the use of a flow cell is therefore believed bility window, and iii) a certain hydrophobicity and high ionic
to enhance N2 enrichment. Generally, the use of a flow cell conductivity.
greatly increases the solubility of CO2 in the electrolyte, Organic Solvent Electrolytes: The solubility of N2 in aqueous
thereby increasing the current density of the corresponding electrolytes is as low as 1.98 × 10−3 g per 100 gH2O (at 20 °C), but

Adv. Mater. 2021, 33, 2007650 2007650 (24 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

it is higher in some organic substances, for example, fatty alco- active protons (e.g., H2O, methanol, and acetic acid) or less
hols, in which case it is at least 30 times that in water. In addi- proton-active sources (for example, 2-methyl-2-propanol) are
tion, the occurrence of the undesired HER in purely aqueous used, the FE of the N2 to NH3 conversion is lower. In contrast,
electrolytes often lead to low E-NRR current efficiency. There- some solvents with moderately active protons, such as 2-pro-
fore, researchers have attempted to use electrolytes based on panol and ethanol, allow for the most efficient Li-mediated
organic solvents or organic solvent/water mixtures to increase NH3 synthesis. This behavior can be ascribed to the fact that
E-NRR efficiency. Tsuneto et al. investigated the E-NRR activity Li easily reacts with solvents having active protons (to produce
of Ti, Mo, Co, Ni, Fe, Ag, and Cu catalysts in tetrahydrofuran- H2), whereas solvents with an overly weak acidity cannot easily
based electrolytes containing ethanol/LiClO4 at current effi- decompose Li3N.
ciencies of 5–8% (under ambient conditions).[177] At a pressure To circumvent the limitations of the Li-mediated E-NRR,
of 5.1 MPa, the current efficiency and NH3 production rate McEnaney et al. and Kim et al. reported some meaningful
reached 58% and 1 × 10−9 mol s−1 cm−2 (Fe was used as a cata- results from the perspective of electrochemical device improve-
lyst), respectively. Kim et al. used a mixed solution of isopro- ment. McEnaney et al. split the Li-mediated E-NRR into several
panol and water as the electrolyte, achieving an NH3 yield and independent parts and recombined them into a cyclic network,
current efficiency on porous Ni of 1.54 × 10−11 mol s−1 cm−2 and successfully solving the problem of HER occurrence and the
1.0%, respectively.[129] As isopropanol is easily reduced at the influence of solvents on NH3 synthesis.[179] Specifically, LiOH
cathode, the NH3 production rate and FE under ambient condi- was electrolyzed in a LiCl/KCl melt to form Li, which com-
tions increased to 3.58 × 10−11 mol s−1 cm−2 and 17.2%, respec- bined with flowing N2 to form Li3N. Finally, Li3N flowed into
tively, when the electrolyte was replaced with LiCl-containing water and was hydrolyzed into NH3 and LiOH, with the latter
ethylenediamine.[128] For the same catalyst, the NH3 yield and entering the next cycle (Figure 17a). To prevent H2O, O2, and
FE in an organic electrolyte exceed those in an aqueous elec- LiOH from contacting the electrolyzed Li, the electrode was
trolyte, which indicates that the use of an organic solvent can covered with a porous alumina diffusion layer (Figure 17b).
effectively enhance the capture of N2 and inhibit the HER. Nev- Kim et al. designed similar devices to isolate the individual
ertheless, organic electrolytes often need to provide the proton reaction units in the Li-mediated E-NRR.[180] In their setup, an
source required by the E-NRR, and some alcohols and a small aqueous phase was introduced at the anode during the elec-
proportion of water become the sacrificial hydrogen sources. trolysis of Li+, while an Li+-conductive glass ceramic membrane
Compared to aqueous electrolytes, organic electrolytes enhance (LISICON) was used to isolate the anode chamber for Li+ elec-
the E-NRR. However, this strategy requires the use of large trolysis from other reaction areas, thus avoiding the influence
amounts of organic solvents and thus deviates from the goal of of water (Figure 17c). Although the LISICON membrane effec-
green NH3 synthesis under ambient conditions. tively isolated the electrolyte without affecting the migration
Li-Mediated Electrolytes: Li is the lightest metal and solid ele- of Li+, its low conductivity caused the entire E-NRR system to
ment, and like all alkali metals, it is highly reactive and flam- require a higher operating voltage.
mable. The properties of Li are similar to those of H, which The introduction of the circulation device to independently
is located directly above Li in the periodic table. For example, operate the different units of the Li-mediated E-NRR solves the
similar to H, Li easily combines with N2 to form Li3N (6Li + problem of the mutual influence of different reactions, but it
N2 = 2Li3N). As Li3N can rapidly release NH3 upon protonation, cannot achieve true continuous operation in time and space,
this concept can be applied to the production of NH3. even if a rotating circulation device is used. To overcome this
Tsuneto et al. significantly contributed to Li-mediated syn- impediment, Lazouski et al. used a gas diffusion electrode as a
thesis of NH3.[178] Initially, Al, Zn, Fe, Ni, Cu, Ag, and Pb were catalyst carrier to accomplish the rapid electrochemical reaction
used as working electrodes in a tetrahydrofuran/ethanol mixed of gases with low solubility. In their setup, the GDL promotes
electrolyte containing LiClO4 to evaluate the conversion of N2 to the contact of low-solubility gas with water (or non-aqueous
NH3. Among these metals, Ti and Ag showed the highest cata- electrolyte), thereby increasing the reaction rate. This strategy
lytic activity, with the respective NH3 outputs reaching 14.1 and shows certain advantages for the E-NRR.[181] As the authors did
14.5 µmol (FE reached 8.2 and 8.4%, respectively). Al and Pb not perform the corresponding Li cycle test, we speculated that
had the lowest activity as they can form alloys with Li to passi- the reduction of NH3 production in the reaction system may be
vate it. When Mo was used as the working electrode, the output related to the depletion of Li; therefore, further exploration is
of NH3 was positively correlated with the applied charge, which needed.
verified that the detected NH3 was produced via the conversion Compared to the direct reaction between N2 and H2O in the
of N2 on Mo. Further research showed that when the electrolyte aqueous phase, the Li-mediated E-NRR generally exhibits a
does not contain ethanol or LiClO4 is replaced with NaClO4 or very high FE but faces some technical problems such as 1) the
Bu4NClO4, almost no NH3 can be detected, which indicates that need for high applied potentials of ≥3 V, 2) the safety hazards
the presence of both ethanol and Li+ is crucial for NH3 produc- of Li usage and high Li recovery costs, and 3) the lack of a
tion. In this system, N2 is electrochemically reduced to NH3 method for the realization of a true continuous-cycle opera-
under the mediation of lithium (Li+ → Li → Li3N → NH3 + tion. However, if the efficient conversion of N2 to NH3 can be
EtOH).[177] Electrons are injected into Li+ under the action of achieved, one should continue to develop and design novel
current, and the Li deposited on the electrode surface reacts electrochemical devices and catalysts to improve the status
with N2 to form Li3N that is further converted into NH3 and quo of Li-mediated E-NRR. Given the strong binding ability
EtOH. Tsuneto et al. also showed that the proton source affects of Li and N2, this is an efficient way to fix the abundant N2 in
the Li-mediated synthesis of NH3. When certain solvents with the atmosphere.

Adv. Mater. 2021, 33, 2007650 2007650 (25 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 17. a) The lithium-mediated E-NRR cycle device designed by McEnaney et al., LiOH is electrolyzed in LiCl/KCl molten salt to form Li, then Li
combines with flowing N2 to form Li3N, and finally Li3N is injected into water to decompose into NH3 and LiOH, and finally LiOH enter the next cycle.
b) The electrode is covered with a porous alumina diffusion layer to prevent H2O, O2, and LiOH from contacting the Li produced by electrolysis. Repro-
duced with permission.[179] Copyright 2017, Royal Society of Chemistry. c) This is a device designed by Kim et al. to isolate the various reaction units in
lithium-mediated E-NRR, in which an Li+ ion conductive glass ceramic membrane (LISICON) is used to isolate the anode chamber for electrolysis of
Li+ from other reaction areas. Adapted with permission.[180] Copyright 2018, Wiley-VCH.

Ionic Liquid Electrolytes: N2 is well soluble in many ionic Ltd., including Nafion 115, Nafion 117, and Nafion 211) under
liquids, for example, the solubility of N2 in [P66614][C8F17CO2] E-NRR conditions,[185] observing the following phenomena.
equals 6.5 mmol L−1,[182] with high values also observed for 1) During electrolysis, the content of NH3 pre-introduced into
[HMIM][NTf2], [C4mpyr][eFAP], and [P6,6,6,14][eFAP].[183] In the the cathode chamber decreased, while the NH3 content in the
typical work of Ozin et al., [P6,6,6,14][eFAP] containing trace H2O anode chamber increased, which caused a measurement error
was used as the electrolyte to achieve an FE of 60% and an NH3 of ≈30 wt% upon balancing (Figure 18b). 2) NH3 generated
yield of 14 mg h−1 m−2 (Figure 18a).[183c] As ionic liquids have an in the cathode chamber migrated to the anode chamber and
extremely low vapor pressure, high thermal stability, adjustable was oxidized to N2 or NOx at a high potential of −0.3 V versus
dissolving power, and a wide electrochemical stability window, RHE (Figure 18c).[185,186] 3) In view of the inherent polarity
they are considered to be green solvents capable of replacing and basicity of Nafion, its Brønsted-acidic sites adsorbed
water. Some polyfluorinated (F has a very high electronega- NH3 (Figure 18d), which was subsequently slowly released.
tivity) ionic liquids favoring the capture of N2 can be designed 4) NH3 promoted the degradation of Nafion, possibly by inter-
for the E-NRR. Thus, compared with volatile organic electro- acting with its sulfonic acid groups. These results indicate
lytes, ionic liquids have some inherent advantages as E-NRR that Nafion is not suitable for long-term E-NRR operation.
electrolyte solvents. The high conductivity and ion transmission efficiency of the
Nafion membrane in the E-NRR makes it difficult to find suit-
able alternative membranes, and researchers looking for such
5.1.3. Proton-Exchange Membranes alternatives are advised to consider the following proposals.

H-type cells are the ones most widely used in E-NRR 1) One should simultaneously detect the NH3 content in the
research, with Nafion proton-exchange membranes frequently cathode and anode chambers, and the spent Nafion mem-
employed to separate the cathode and anode chambers. brane should be soaked in neutral 0.1 m Na2SO4 for at least 10
However, some reports indicate that NH3 from the cathode h to collect the released NH3.
chamber can migrate to the anode chamber during the E-NRR 2) The catalyst should be reasonably designed to obtain an op-
to bias NH3 quantitation.[80a,110,184] To verify the existence of timal applied potential of less than −0.3 V versus RHE and
the problems pointed out in these reports, Ren et al. com- thus prevent the degradation of Nafion as much as possible
prehensively tested the changes in Nafion proton-exchange (when the NH3 yield is significant, this point can be ignored,
membranes (procured from DuPont China Holding Co., but strict control experiments are required).

Adv. Mater. 2021, 33, 2007650 2007650 (26 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 18. a) Schematic diagram of the conversion of N2 to NH3 on the Fe-SS electrode (above) and the interaction of ionic liquid ions with N2 (below).
Reproduced with permission.[183c] Copyright 2017, Royal Society of Chemistry. b) Relationship between the change of ammonia concentration in the
cathode and anode compartments (separated by Nafion membrane) and the equilibrium time. c) Changes of ammonia concentration with electrolysis
time under different potentials and open circuit potentials. d) NH3-TPD test results of commercial Nafion 117 membrane. e) Schematic diagram of the
new ENRR equipment equipped with a salt bridge. Adapted with permission.[185] Copyright 2019, Wiley-VCH.

3) After continuous long-term E-NRR operation, it is best to de- in the cathode chamber, while the counter electrode is placed in
tect changes in the Nafion membrane by scanning electron the anode chamber. The counter electrode most often comprises
microscopy. high-purity graphite or Pt. For different electrolytes, the refer-
4) If the E-NRR is performed in an acidic electrolyte, a method ence electrode should be selected according to the type of elec-
can be established to detect the content of NH4+ in the Nafion trode filling solution and electrode material, and it is best to use
membrane. According to this method, the membrane is soaked a reference electrode with a salt bridge if necessary. Generally, the
in electrolytes with different NH4Cl contents for 2 h (the de- catalyst is often dispersed on a specific support (such as carbon
crease in the NH4+ content of the electrolyte is due to the NH4+ paper, CC, and Ni foam) and fixed on the working electrode with
adsorbed on the Nafion membrane). Then, the residual water an electrode clip. In some reports, the catalyst is directly coated
on the Nafion membrane is removed under reduced pressure on the working (e.g., glassy carbon) electrode surface. Although
at low temperature. Next, the membrane is probed by Fourier this strategy is also suitable for the E-NRR, it should prevent
transform infrared (FT-IR) spectroscopy, and the FT-IR signal inaccuracies in the normalization of activity caused by the low
intensity is correlated with the NH4+ content of the membrane catalyst loading and ultra-low current density.
to construct a standard curve. The amount of NH4+ adsorbed on
the Nafion membrane after each potentiostatic polarization is
then obtained from the standard curve by interpolation. 5.2. Pre-Exclusion of System Pollution
5) In addition to Nafion proton-exchange membranes, other
types of membranes can also be used for the E-NRR, among At present, the average NH3 yield of the E-NRR is at a level of
which Celgard 3501 offers the benefits of low cost and low tens of micrograms, which complicates the accurate quantita-
NH3 adsorption.[187] tion of NH3. Therefore, the strict control of NH3 pollution has
6) In addition to the diaphragm, traditional salt bridges (e.g., become a major requirement. As most reported catalyst prep-
those saturated with KCl) may also be used to connect the aration processes introduce N sources, the detected NH3 will
cathode and anode (Figure 18e). indicate false activity if no NH3 (or NOx) pre-exclusion treat-
ment is carried out. In general, one can alternately soak the
catalyst in acid and lye until no NH3 is detected. On the other
5.1.4. Electrode Material hand, some reports indicate that commercial high-purity N2
often contains traces of NH3 or NOx, which may cause uncer-
Usually, a three-electrode system (working, reference, and tainty in the origin of NH3 produced in the E-NRR. There-
counter electrodes) is used for E-NRR experiments. In the H-type fore, researchers are advised to use acid hydrazine or follow
electrolytic cell, the working and reference electrodes are placed the proposal of Andersen[184b] to remove NH3 or nitrogen-like

Adv. Mater. 2021, 33, 2007650 2007650 (27 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 19. Schematic diagram of the gas circuit circulation in E-NRR process (a: simple diagram. b: physical image). Reproduced with permission.[184b]
Copyright 2019, Springer Nature.

compounds from the feed gas in advance (Figure 19) and then certain amount of acid (such as HCl) should be added to the
import the pure feed gas into the E-NRR reactor. In addition, concentrated electrolyte, which greatly enhances the features of
before E-NRR operation, the entire electrolysis device (e.g., 14NH + and 15NH + peaks near δ 7.0.
4 4
electrodes, electrolytic cells, and proton-exchange membranes)
needs to be sufficiently washed to remove any remaining envi-
ronmental NH3. 5.4. Screening of Catalyst Active Center

In the E-NRR catalyst, active centers play a key role. The cor-
5.3. Isotope Labeling Experiments rect identification of such active centers can promote the rapid
development of E-NRR catalysts and is also conducive to cor-
The currently low NH3 production in the E-NRR and the ubiq- rect theoretical modeling. Generally, if ABC is used as a cata-
uitous NH3 pollution require researchers to realize the impor- lyst, the E-NRR activities of A, B, C, AB, AC, BC, and ABC
tance of 15N2 isotopic labeling experiments and describe the should be studied separately to determine the key components
operation process in their reports. Unless the reported NH3 or synergistic effects of the material. In addition to composi-
yield significantly exceeds the total NH3 content of the envi- tion identification, some crystal phase and morphology factors
ronment or the material, the authors are required to provide determining the exposure of active sites and electron transfer
the results of isotopic labeling experiments when nitrogen- information need to be considered.
containing materials are introduced during catalyst preparation
or when the reported NH3 yield is low and background pollu-
tion cannot be ignored. It is worth mentioning that despite the 5.5. Test Items of E-NRR Procedure
high price of 15N2, researchers are still advised to perform iso-
tope labeling experiments, as the cost of this input is justified In the field of E-NRR, where standardization is urgently
by the achieved scientific progress. The saturation solubility of needed, many reports have proposed a complete set of oper-
N2 in aqueous solutions is extremely low, and in a long-term ating procedures.[188] These manipulation procedures are all
electrochemical experiment, most of the delivered 15N2 will focused on obtaining real ammonia activity, which is essential
be released into the environment to result in a great waste of to guide researchers to achieve correct E-NRR operation. Here,
resources. Considering that some research groups have insuffi- we only introduce the conventional test items:
cient research funds, we recommend that researchers use a gas
circulation pump with a four-way valve (Figure 19) described by 1) N2-TPD. The ability of the catalyst to adsorb N2, which is a
Andersen et al.[184b] to recycle 15N2. necessary prerequisite for E-NRR realization, can be deter-
Although several studies have used proton nuclear magnetic mined through N2-TPD experiments.
resonance (1H NMR) spectroscopy to qualitatively or quanti- 2) CV experiments can be performed to probe the electrochemi-
tatively detect 14NH4+ and 15NH4+, none of these studies pro- cal stability of the catalyst.
vide detailed parameters for NMR spectrometers. In general, 3) Linear sweep voltammetry (LSV) experiments can be used to
the coupling constant JH is the product of the chemical shift compare current density in N2- and Ar-saturated electrolytes.
interval between two peaks and the magnetic field frequency If a higher current density is observed in the former case,
of the NMR spectrometer (e.g., 400, 600, and 800 MHz), so the catalyst has an obvious effect on N2, and its E-NRR per-
the chemical displacement interval between the two peaks formance can be further evaluated. In general, researchers
decreases with increasing magnetic field frequency. At the are advised to use the potential that results in the largest cur-
same time, as the sensitivity of NMR spectrometers with low rent density interval as the applied potential point for the first
magnetic field frequencies is low, different instrumental set- study.
tings may result in the inability to collect NH3 signals. Under 4) Potentiostatic polarization experiment (j–t curve). Potentio-
the normal configuration of the 1H NMR spectrometer, a static polarization can be performed for a certain period of

Adv. Mater. 2021, 33, 2007650 2007650 (28 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

time (e.g., 2 h) at different potentials, and the FE and normal- dihydrate (C5H4FeN6Na2O3), salicylic acid, sodium citrate, and
ized NH3 yield can be obtained at these potentials. sodium hypochlorite, with its action principle described in
5) Stability test. The catalyst recycling performance can be Equation (12). In this case, the reaction needs to be carried out
tested through potentiostatic polarization experiments, and in 1 m NaOH, with sodium nitroprusside acting as a catalyst
catalyst stability can also be evaluated by N2-to-Ar-to-N2∙∙∙ and sodium citrate acting as a buffer to stabilize the pH. The
cycling. The stability metric of E-NRR catalysts usually mani- obtained product is usually blue but changes color from light
fests as a low fluctuation of the j–t curve and a low change in yellow to blue-green with increasing NH3 concentration, which
NH3 yield and FE. is linearly related to absorbance at 655 nm. NH3-sensitive
6) Tolerance test. Generally, industrial NH3 production is run composite electrodes use a pH glass electrode as the indicator
for a long time of up to several months to save handling costs. electrode and an Ag–AgCl electrode as the reference electrode
In the E-NRR field, researchers are also required to conduct (Figure 20a). The thin semi-permeable membrane between the
continuous constant-potential polarization experiments for pH glass electrodes can isolate the external solution. If the water
at least 10 h, and the performance is viewed as stable when sample is alkaline (pH >11), ammonium salts are converted to
the j–t curve does not show significant single-direction fluc- NH3, which diffuses through the semi-permeable membrane
tuation. that does not let through water and other ions. Consequently,
the equilibrium of NH3 + H2O ↔ NH4+ + OH− in the NH4Cl
electrolyte film layer shifts to the right to change the concentra-
5.6. NH3 Detection Methods tion of OH− and thus affect the pH, which is recorded by the
glass electrode. As the electric potential measured by the NH3-
The detection of NH3 production in E-NRR systems is a major sensitive electrode is positively correlated with the logarithm of
challenge. At present, the quantitative detection of NH3 is NH3 concentration in the aqueous sample, it can be used for
mainly achieved using spectrophotometry,[189] ion chromatog- the determination of NH3 produced in the E-NRR. Ion chro-
raphy,[190] NH3-sensitive electrode methods,[191] titrimetry,[192] matography uses the principle of ion exchange to determine
mass spectrometry (MS),[193] and 1H NMR spectroscopy.[194] specific ionic strength. By selecting a suitable ion-exchange
Among these techniques, one should single out the Nessler’s column (such as Origin 9512HPBNWP NH3 and Radiometer
reagent, indophenol blue, fluorescence, and enzymatic methods Analytical ISE25NH4-9 NH4+), one can reliably quantify NH4+
that rely on spectrophotometry and offer the benefits of detec- ions. 1H NMR can quantitatively and qualitatively analyze NH3
tion simplicity and rapid results. Nessler’s reagent and indo- in the electrolyte and has excellent resolution for 14NH4+ and
phenol blue methods are the two most widely used methods 15NH +.[184b] In mass spectrometry, 14NH + and 15NH + can be
4 4 4
in the E-NRR field. Equation (11) describes the action prin- distinguished based on the difference in their mass/charge
ciple of the former method. The colorimetric reagent of this ratio (m/z), and quantitative measurements can be performed
method mainly comprises K2HgI4, KOH solution, and Rochelle using the relationship between the mass-spectrometric signal
salt (KNaC4H4O6∙4H2O, masking agent for other ions used and analyte amount. In 1H NMR, 14NH4+ affords three peaks
to reduce their interference with NH3 determination). Under of equal intensity, while the signal of 15NH4+ is split into two
alkaline conditions, [HgI4]2− reacts with NH3 to produce a red- peaks of equal intensity, which allows one to distinguish these
brown complex (Hg2ONH2I) with strong absorption at 420 nm ions. Moreover, NH3 can be quantified using integrated peak
that is proportional to NH3 content. The colorimetric agent of areas. However, when these methods are used to detect NH3 in
the indophenol blue method comprises sodium nitroprusside the E-NRR, one should consider their limitations. For example,

Figure 20. a) Simple schematic diagram of ammonia-sensitive electrode structure. b) The color of the ammonia solution in Nessler’s reagent (top left)
and indophenol blue (top right) at different pH values, where the bottom left and bottom right are the corresponding detected ammonia concentra-
tions. Adapted with permission.[195] Copyright 2019, Wiley-VCH.

Adv. Mater. 2021, 33, 2007650 2007650 (29 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

15NH + and H3O+ have the same nominal m/z ratio. The E-NRR of NH3, and researchers can choose either of them based on
4
is mainly carried out in aqueous solutions, and H2O therefore the properties of the electrolyte or invoke the assistance of 1H
produces a high background signal in the related mass spectra NMR. Finally, we recommend that the authors provide the
to hinder the accurate identification of the 15NH4+ signal.[193] results of two detection methods at the same time and use
Zhao et al.[195] compared the effects of pH, other ions, 14NH + and 15NH + 1H NMR detection in cases of low NH yield
4 4 3
and organic molecules on the results of NH3 quantitation by and/or the use of catalysts that may contain N sources.
Nessler’s reagent and indophenol blue methods and reached the
following conclusions. 1) The Nessler’s reagent method, indo-
2 [ HgI4 ]
2−
phenol blue method, and ion chromatography can accurately + NH3 + 3OH− → Hg 2ONH2I + 7I− + 2H2O (11)
quantify NH3 at low concentrations (<500 µg L−1), although the
indophenol blue method has a large error at high NH3 levels
and thus needs to be used together with other methods. 2) The
Nessler’s reagent method has good sensitivity for NH3 in the (12)
entire pH range, whereas the indophenol blue method only has
good NH3 detection sensitivity under neutral and alkaline con-
ditions (Figure 20b). 3) Low concentrations of Ru3+, Fe2+, In3+,
and Ni2+ can affect the NH3 detection accuracy of the Nessler’s NADH + NH+4 + 2 − oxoglutarate →
(13)
reagent method, which is therefore not applicable at high con- L − glutamate + NAD+ + H2O
centrations of these metal ions. 4) When the catalyst contains
carbon (or carbon carrier), a control test is required. 5) When
the catalyst contains nitrogen or a nitrogen source is introduced 6. Summary and Outlook
during the preparation process, the catalyst needs to be repeat-
edly washed with ultrapure water until NH3 is not detected The electrochemical reduction of N2 under ambient condi-
before it can be used for the E-NRR experiment. Furthermore, tions instead of the resource- and energy-intensive Haber–
the colorimetric agents of Nessler’s reagent and indophenol Bosch process is the best strategy to address the needs of our
blue methods need to be stored in the dark at low temperature present-day society and alleviate the deterioration of the global
and have a limited shelf life of less than three weeks. Sodium environment. In this review, we elaborate on the recent devel-
hypochlorite used in the indophenol blue method needs to be opments of the E-NRR, including the intrinsic characteristics
prepared and used on the spot to ensure the same dilution con- of the E-NRR, advanced E-NRR catalysts, E-NRR activity met-
centration, and should be guaranteed to come from the same rics, E-NRR operating procedures, and related challenges.
batch of sodium hypochlorite mother liquor, which is the big- Over the past few decades, improvement in the E-NRR field
gest factor affecting absorbance. For ion chromatography, one has been sullen, whereas the past three years have witnessed
should ensure that the electrolyte does not contain Na+ or K+, as rapid development. Even so, the blueprint for the E-NRR has
these ions seriously compromise NH3 detection accuracy.[183a,185] only been introduced and it faces significant challenges. To aid
Therefore, when using electrolytes or catalysts containing Na+/ in the rapid and effective development of the E-NRR, we have
K+, researchers should avoid using ion chromatography or set summarized the main points of this review with the hope of
up rigorously controlled experiments. Fluorescence spectros- accelerating E-NRR standardization and ensuring rapid com-
copy is used less frequently, relying on the reaction of phtha- munication among the research community. In addition to the
laldehyde (OPA) and sulfuric acid with ammonium salts. The detailed text, we put forward our recommended brief process
accuracy of this method is often affected by organic amines, and for E-NRR in Figure 21. In addition, Table 2 summarizes the
the fluorescence method with the participation of OPA cannot parameters of some typical E-NRR catalysts discussed in this
detect traces (micro- and millimolar levels) of NH3.[196] Titri- review.
metric and enzymatic methods are not commonly used in the
E-NRR because of their restrictions. The former methods are 1) Theoretical predictions can well guide experiments, including
suitable for the detection of high NH3 concentrations (5 mg L−1) catalyst design and electrocatalytic environment optimization.
but require cumbersome and strictly controlled operating con- Most E-NRR studies perform DFT calculations on the conver-
ditions. Specifically, the NH3 solution needs to be adjusted to sion of N2 into NH3 on a specific catalyst surface, but only few
pH 9.5 and distilled (the distillate is collected in a mixed solu- reports summarize and analyze the detailed E-NRR mechanism
tion containing boric acid and an indicator), with sulfuric acid and theoretical E-NRR paths on various catalysts. Although it is
subsequently used to titrate the thus enriched solution until necessary to report the possible E-NRR pathways on a specific
the indicator turns light purple. Distillation under alkaline catalyst, it is even more necessary for researchers to analyze the
­conditions may cause the loss of the volatile NH3. According to E-NRR mechanism that integrates a large number of E-NRR
the principle described by Equation (13), the enzymatic method catalysts to obtain predictions with a certain trend. In addition
can quantify NH3 by detecting the decrease of NADH content to the use of big data to guide catalyst design, detailed predic-
through the change of absorbance.[197] Although the enzymatic tions of the E-NRR process on metal or non-metal active cent-
method is sufficiently sensitive for NH3 detection, it needs to ers can greatly reduce the period of experimental exploration, as
be operated under the optimal operating conditions of pH 8.0 in the work of Liu et al.[164] and Li et al.[199]
and 20–25 °C.[198] In contrast, Nessler’s reagent and indophenol 2) Design of advanced catalysts. At present, numerous metal-
blue methods are the easiest ones to achieve the rapid detection and non-metal-based E-NRR catalysts are known. The biggest

Adv. Mater. 2021, 33, 2007650 2007650 (30 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 21. Suggested solution for the benchmarking of electrochemical nitrogen reduction.

problems faced by metal-based catalysts are the severe HER operating procedures, numerous studies have provided clear
and lack of active site exposure. Most transition metals have hints. Specifically, CV, LSV, and constant potential polari-
empty d-orbitals that benefit the adsorption of N2 while si- zation experiments (j–t curves) can not only determine the
multaneously promoting the HER. In other words, further activity/non-activity of a specific catalyst, but can also obtain
modification engineering (such as surface hydrophobization the NH3 yield, FE, and stability characteristics.
and heteroatom [e.g., O] doping) is necessary to suppress 4) Ensuring the accuracy of E-NRR activity is one of the chal-
the HER on the metal surface and thus improve E-NRR lenges faced by the E-NRR field. At present, the NH3 yield in
selectivity. Another strategy to improve the performance of the E-NRR is at the microgram level, and the ubiquitous NH3
metal-based E-NRR catalysts is to provide a sufficiently large pollution results in strict electrochemical test procedures.[200]
surface with active sites, for example, by loading nanomet- Although there is no mandatory processing standard, re-
als on a high-specific-surface-area support (such as porous searchers must address the following. i) Electrocatalytic de-
carbon and graphene) or by preparing single-atom-dispersed vices must strictly exclude possible NH3 pollution such as
metal-based materials. Certainly, some surface modification that from the electrolytic cells, electrolytes, electrodes, and
techniques such as crystal-plane control and the construc- proton-exchange membranes. ii) N2 and Ar gases used as
tion of surface defects (such as N, S, and O vacancies) are feeds should be subjected to impurity (mainly NH3 or NOx)
more effective than others. The development of non-metal- removal pretreatment. iii) The catalyst should be pretreated
based E-NRR catalysts is necessary as these catalysts offer the to eliminate any NH3 sources (including organic amines). iv)
benefits of low cost and ease of processing. Although B, C, Researchers are required to provide the results of 15N2 labe-
N, S, and P have potential E-NRR activity, related research ling experiments, especially for low NH3 yields or catalysts
reports are few, and extensive experimental exploration and containing N sources. v) The authors are advised to report
theoretical calculations are still required. In this regard, re- NH3 yields obtained using two or more detection methods
searchers must be able to adopt appropriate characterization and conduct strict control experiments.
methods to present information on the catalyst structure to 5) Advanced in situ methods such as in situ FT-IR spectroscopy,
correctly construct a theoretical calculation model. The MvK in situ UV–vis spectroscopy, in situ mass spectrometry, and
mechanism can promote the contact/exchange between the in situ Raman spectroscopy should be developed.[201] Here,
reactants and can reduce the negative effect of the poor solu- “in situ detection” means that the intermediates formed in
bility of N2 in water to a certain extent. Therefore, it is mean- the electrolyte and catalyst during the E-NRR are detected in
ingful to construct an N-doped E-NRR catalyst and optimize real time to promote the development of the E-NRR theory
its NRR activity, for example, by increasing the density of N and guide catalyst modification.
sites. At the same time, it is necessary to eliminate possible 6) To explore suitable E-NRR catalysts as soon as possible, re-
NH3 pollution owing to the presence and decomposition of searchers should pay attention to the following points. i) It
NH3 (or NOx−) containing catalysts. is necessary to develop water-soluble organic solvents with
3) Reasonable and comprehensive electrochemical test items. high N2 solubility (such as specific ionic liquids) and dope
Although there is no consensus on the fixation of E-NRR them into water-based electrolytes to improve the kinetics of

Adv. Mater. 2021, 33, 2007650 2007650 (31 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. The activity characteristics of some typical E-NRR catalysts mentioned in this review.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
Au1Co1@GO 0.5 m K2SO4 −0.2 36.82 µg h−1 mgcat.−1 22.03 Dissociative When Co is introduced into Au to form [7a]
an alloy, the free energy of nitrogen adsorption and
the activation bond of nitrogen can be reduced.
THH Au NRs 0.1 m KOH −0.2 1.648 µg h−1 cm−2 4.02 Associative Tetrahexahedral gold nanorods with alternating [10]
step-like crystal faces surrounded by high-index
crystal faces
Vr-ReSe2@CBC 0.1 m Na2SO4 −0.25 28.3 µg h−1 cm−2 42.5 Associative The introduced selenium vacancy could [14b]
modulate the electronic structures of ReSe2
nanosheet by positively shifting its d-band position;
Cover both sides of Vr-ReSe2@CBC nanofibers with
hydrophobic CBC layers to inhibit HER, thereby
improving Faraday efficiency.
PdRu TPs 0.1 m KOH −0.2 37.23 µg h−1 mgcat.−1 1.85 – Tripod structure provides a large number of exposed [20a]
active sites;
The bimetallic composition can adjust the
coordination environment and electronic state of Pd
and Ru, which helps to enhance the adsorption and
activation of N2 species.
Ti3C2Tx 0.1 m HCl −0.4 20.4 µg h−1 mgcat.−1 9.3 Associative N2 chemically adsorbed on Ti3C2Tx undergoes the [20b]
elongation/weakening of the NN triple bond,
which promotes its catalytic conversion to NH3.
Ru SAs/N-C 0.05 m H2SO4 −0.2 120.9 µg h−1 mgcat.−1 29.6 Associative Ru is dispersed on nitrogen-doped carbon in the [38]
form of isolated atoms, thus exposing more active
sites.
Ru@ZrO2/NC 0.1 m HCl −0.11 3.6 mgNH3 h−1 21 Associative Ru sites with oxygen vacancies can stabilize *NNH, [39]
mgRu−1 *H instability and promote N2 adsorption;
ZrO2 as a carrier can promote the ammonia yield
rate and Faraday efficiency of E-NRR;
Ru@NC in Ru@Zr/NC improved the NRR pathway
while Ru@ZrO2 inhibited HER.
Ru SAs/g-C3N4 0.05 m NaOH 0.05 23.0 µg h−1 mgcat.−1 8.3 Associative Ru SAs/g-C3N4 has easy reaction thermodynamics, [41]
and the E-NRR performance Ru SAs/g-C3N4
from adjusting the d electron energy from
macromolecules to a single atom, thus causing
the d-band center to move upward in the direction
of the Fermi energy level.
PdRu NRAs 0.1 m HCl −0.2 34.2 µg h−1 mgcat.−1 2.4 – The nanorod structure in PdRu NRA can [43b]
expose a large number of active sites and
promote N2 diffusion and electrolyte
penetration during E-NRR;
Alloying Pd and Ru can induce lattice strain
and adjust the electronic configuration of Pd
atoms, which is conducive to the high activity of
the catalyst.
RuPt/C 1 m KOH 0.123 5.1 × 10−9 gNH3 s−1 13.2 – Compared with Ru and Pt catalysts, RuPt has a [43c]
cm−2 synergistic effect on the electrochemical formation
of ammonia.
Pd/C 0.05 m H2SO4 0.1 4.5 µg h−1 mg−1Pd 8.2 Associative The surface kinetics *H of the surface hydrogenated [46]
α-Pd-H facilitates the conversion of N2 to *N2H;
After surface hydrogenation, α-Pd has a Grotthuss-
like proton hopping mechanism, which can reduce
the reaction energy of N2 on it.
PdH0.43 0.1 m PBS −0.15 20.4 µg h−1 mg−1 43.6 Associative The lattice hydrogen atoms in PdH0.43 are used as [47]
the active hydrogen source for the E-NRR process,
and the palladium hydride reduces the activation
energy barrier of the N2 to *N2H rate limiting step.

Adv. Mater. 2021, 33, 2007650 2007650 (32 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
PdO/Pd/CNTs 0.1 m NaOH 0.1 18.2 µg h−1 mg−1 11.5 – PdO and Pd play a synergistic effect between the [48]
active site and the proton H transition, reducing the
overpotential of the E-NRR process;
Pd on the interface side has p-π donated d-orbital
electrons, so it can reduce the electrochemical
reaction kinetics of chemical adsorption of N2,
thereby activating N2;
PdO on the other side of the interface can trap
H to form α-Pd-H, thereby providing a source of
hydrogen for activated N2.
PdPb nanosponges 0.1 m HCl 0.05 25.68 µg h−1 mg−1cat. 5.79 Associative The existence of Pb limits the adsorption [49a]
of *H on Pd, so the introduction of Pb into
palladium-based materials can inhibit the
operation of HER on Pd;
The existence of Pb can also reduce the free energy
of speed-decision in E-NRR, thereby promoting the
smooth operation of E-NRR.
Pd3Pb nanowire 0.1 m Na2SO4 −0.2 18.2 µg h−1 mgcat−1 21.46 – Construct porous PdPb nanowires to expose more [49b]
active sites and thus enhance the E-NRR activity of
this hybrid material.
Pd0.2Cu0.8/rGO 0.1 m KOH −0.2 2.80 µg h−1 mgcat.−1 ≈0.6 – When Pd is doped with Cu, Cu can compete [50]
the H covering the Pd surface to the
Cu surface, thus releasing the Pd site with
E-NRR activity and weakening the side reaction
of HER on Pd.
Pd1Ag1 alloy 1 m KOH −0.2 24.1 µg h−1 mgcat.−1 ≈1.5 – The introduction of Ag with a large lattice [51a]
constant with tensile strain effect can change
the electronic structure of Pd;
The factors that improve the catalytic activity of
Pd1Ag1 alloy E-NRR are: 1) The porous structure
of Pd1Ag1 alloy makes Pd1Ag1 alloy have a high
specific surface, so more active sites are exposed;
2) The presence of metal atoms with incomplete
coordination on the surface of Pd1Ag1 alloy is one of
the key factors to realize the enhancement of
E-NRR activity.
PdCu/NC 0.05 m H2SO4 −0.45 69.2 µg h−1 mgcat.−1 24.8 Associative An E-NRR catalyst with diatomic Pd–Cu sites was [203]
prepared on N-doped carbon by modulating single-
atom Pd sites with Cu;
The introduction of Cu not only moves the partial
density of states of Pd to the Fermi level, but
also promotes the d-2π* coupling between Pd
and adsorbed N2, which leads to the chemical
adsorption of N2 and inhibits the release of
hydrogen.
pAu/NF 0.1 m Na2SO4 −0.2 9.42 µg h−1 cm−2 13.3 – The interconnected porous structure exposing a [52]
large number of Au active sites is the reason for
enhancing E-NRR.
NPG@ZIF-8 0.1 m Na2SO4 −0.8 28.7 µg h−1 cm−2 44 – ZIF-8 as a carrier can expose more Au active sites [53]
and use its cavity to prevent Au nanoparticles
from leaching;
Hydrophobic ZIF-8 can inhibit a large amount
of water from entering the cavity of ZIF-8 to
inhibit HER
Au flowers 0.1 m HCl −0.2 25.57 µg h−1 mgcat.−1 6.05 – The 2D petal-like structure exposes Au to the [55]
interface more and thus increases the catalytic
activity for E-NRR.

Adv. Mater. 2021, 33, 2007650 2007650 (33 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
AuHNCs 0.5 m LiClO4 −0.4 3.9 µg h−1 cm−2 30.2 – Compared with solid nano-Au, AuHNC not only [56]
(at −0.5 V) provides about twice the active site exposure rate,
but also the cage structure can increase the number
of collisions between N2 and the inner surface to
achieve the purpose of capturing N2.
Au/TiO2 0.1 m HCl −0.2 21.4 µg h−1 mg−1cat. 8.11 – The existence of Au-O-Ti ensures the stability of Au/ [57]
TiO2 and the dispersion of Au, and the high atom
exposure ratio provides abundant unsaturated
coordination Au sites, thus promoting the
conversion of N2 to NH3.
a-Au/CeOx-RGO 0.1 m HCl −0.2 8.3 µg h−1 mg−1cat. 10.1 – The amorphous Au existing in RGO has a large [59]
number of “dangling bonds,” and this metastable
state has a stronger binding ability with N2 than the
crystalline state Au, which is the reason why a-Au/
CeOx-RGO has a higher E-NRR activity.
Au/C3N4 0.005 m H2SO4 −0.1 1305 µg h−1 mgAu−1 11.1 Associative In the 2D C3N4, the structure-constrained Au atom [60]
promotes the maximization of active sites, and its
E-NRR performance is 22.5 times that of ordinary
gold nano clusters.
Au/Ti3C2 0.1 m HCl −0.2 30.06 µg h−1 mg−1 18.34 Associative The mesh structure of Ti3C2 can effectively intercept [61]
the passing N2, increasing the flux of reactants;
The interface with high adsorption energy for N2
between Ti3C2 and high-priced Au nanoclusters
weakens NN, and can break the E-NRR
intermediate state to lower the reaction energy
barrier.
Au/M-BOP 0.1 m HCl −0.2 75.89 µg h−1 mgcat.−1 10.35 – With MBOPs as the carrier, and the presence of [62a]
closo-[B12H12]2− makes the E-NRR reaction in an
electric field atmosphere, which is beneficial to the
electron transfer reaction.
CB[7]K2[B12H12]@Au 0.1 m HCl −0.4 41.69 µg h−1 mgcat.−1 29.53 Associative Using bifunctional CB[7]-K2[B12H12] as a carrier, a [63]
highly dispersed supported Au was prepared, which
greatly improved the active site of E-NRR;
Potassium cations effectively prevent H3O+ from
approaching the gold surface, which is due to the
repulsion of the positively charged coordination
spheres of the K+ water complex, thereby inhibiting
the HER.
Au/TiO2 0.01 m HCl −0.4 64.6 µg h−1 mg cat.−1 29.5 Associative The oxygen vacancy (VO) in TiO2 has a strong [64]
(−0.3 V) electron donating effect, which is very important for
the activation of N2;
The introduction of Au adjusts the electronic
structure of the active site and thus promotes the
electrochemical and thermodynamic rate limiting
steps.
Au@CeO2 0.01 m H2SO4 −0.4 28.2 µg h−1 cm−2 9.5 Associative There are oxygen-rich vacancies in the CeO2 [65]
nanoparticle shell in Au@CeO2, which is conducive to
N2 adsorption and activation of E-NRR;
The combination of a large number of oxygen
vacancies in the shell of CeO2 nanoparticles and
the Au nanoparticles of Au@CeO2 synergistically
enhance the conversion of N2 to NH3.
Au6/Ni 0.05 m H2SO4 −0.14 7.4 µg h–1 mgcat.–1 67.8 Associative The reason why Ni and Au synergistically promote [67]
the high FE of E-NRR is that Au can accept electrons
from Ni and thus enter an electron-rich state,
thereby promoting the smooth progress of E-NRR
on Au.

Adv. Mater. 2021, 33, 2007650 2007650 (34 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
Au/CoOx 0.05 m H2SO4 −0.5 15.1 µg h−1 cm−2 19 Associative CoOx can adjust the electronic structure of Au [68]
(promote the transfer of Au nanoclusters to the
high-priced Au1+ oxidation state), and the oxidation
state of Au can control the adsorption of N2
and reduce the free energy of reaction of most
intermediate states in E-NRR.
Au/o-CFP 0.1 m Na2SO4 −0.1 40.6 µg h−1 mgcat.−1 31.3 – A three-phase electrocatalyst (Au/o-CFP) (three- [204]
phase: gas, liquid, and solid) made by preparing Au
nanoparticles on hydrophobic carbon fiber paper
in situ;
The hydrophobic CFP surface promotes effective
three-phase contact (TPCP) of N2 (gas), electrolyte
(liquid) and Au NP (solid);
High concentration of N2 molecules can directly
contact the electrocatalyst surface, thereby inhibiting
HER and enhancing E-NRR.
Pt SAs/WO3 0.1 m K2SO4 −0.2 342.4 µg h−1 mgPt−1 31.1 Associative N2 can be electrochemically reduced to NH3 at the [73]
isolated Pt site on the WO3 plate, where the Pt-3O
site is the activation center of the NN triple bond.
ISAS-Fe/NC 0.1 m PBS −0.4 62.9 µg h−1 mgcat.−1 18.6 Associative Fe in the form of a single atom can increase the [75]
active site exposure of Fe-based catalysts;
The atomically dispersed FeN4 sites can activate
N2 molecules.
Fe SANC 0.1 m KOH 0 7.48 µg h−1 mg−1 56.55 Associative The Fe SANC structure can effectively capture [76a]
N2, and its energy barrier is 2.38 kJ mol−1. The
obtained local high concentration of N2 will facilitate
the subsequent adsorption, which has a particularly
low Gibbs free energy of -0.28 eV.
Fe-N/C 0.1 m KOH −0.2 34.83 µg h−1 mg−1 9.28 Associative FeN3 species is the main catalytic activity center [76b]
of E-NRR.
FePc/C 0.1 m Na2SO4 −0.3 137.95 µg h−1 10.5 Associative Theoretical calculation of five possible sites in FePc [76c]
mg−1FePc (Pc: phthalocyanine) (iron, pyrrole-N1, pyrrole-N2,
pyridine-N3, and carbon atoms near pyridine-N)
found that the E-NRR activity is mainly determined
by the Fe center.
Fe–CeO2 −0.5 26.2 µg h−1 mg−1 14.7 Associative On the one hand, Fe-doped CeO2 exhibits nanosheet [76e]
state that can increase the exposure of active sites
and promote electron transport;
On the other hand, the doping of Fe promotes the
generation of OV in CeO2, so Fe and CeO2 (with OV)
together reduce the energy barrier of E-NRR.
FeMo@NG 0.25 m LiClO4 −0.4 14.95 µg h−1 mg−1 41.7 Associative The composition characteristics of active factors in [77a]
(−0.2 V) nitrogenase were simulated to construct N-doped
FeMo single-atom catalyst (FeMo@NG);
The geometric synergistic effect of FeMo dimers in
FeMo@NG promotes the selectivity of E-NRR and
the catalytic kinetics of FeMo@NG.
Mo3Fe3C 0.1 m Li2SO4 −0.05 72.5 µmol h−1 gcat.−1 27 Associative DFT calculations prove that Mo3Fe3C is more [77b]
conducive to the polarization and activation of N2
molecules than Mo2C and Fe3C;
When using Mo3Fe3C as a catalyst, the energy
required for the formation of *N2H is lower, which
is more beneficial to E-NRR;
The interaction between Mo atoms and Fe atoms
is more conducive to the activation and
hydrogenation of N2.

Adv. Mater. 2021, 33, 2007650 2007650 (35 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
LCFN82 (LaFeO3) 2.4 V 13.46 µg h–1 mgcat.–1 1.9 – The oxygen vacancies (OV) in the perovskite [85a]
(cell (1.8 V) (LaFeO3) can promote N2 adsorption and
voltage) subsequent hydrogenation.
SA-Mo/NPC 0.1 m KOH −0.3 34.0 µg h–1 mgcat.–1 14.6 – The SA-Mo/NPC with N-doped porous carbon [86d]
supporting Mo single atom exhibits a high active
site exposure rate, and a carrier with a high specific
surface area is conducive to the rapid transmission
of electrons in the E-NRR process.
Mo0/GDY 0.1 m HCl −0.1 V 2.0 µg h–1 mgcat.–1 15.6 – The strong electron-enriched GDY (graphdiyne) [86e]
(vs SCE) environment retains Mo0 through strong p-d
coupling;
In E-NRR process, the fast and reversible Mo-C1
charge transfer not only inhibits the HER side
reaction, but also ensures the extremely high N2
hydrogenation rate related to the thermodynamic
trend.
Mo-SnS2/CC 0.5 m LiClO4 −0.5 41.3 µg h–1 mgcat.–1 20.8 Associative Mo doping can adjust the electronic structure [86g]
(−0.4 V) of SnS2, and make SnS2 spontaneously generate
enriched S vacancies (Vs);
DFT calculations show that the coexistence
of Mo dopants and Vs can create active sites
for trimers (Mo–Sn–Sn), which can strongly activate
N2 and reduce the energy barrier of E-NRR.
MoS2/CC 0.1 m Na2SO4 −0.5 8.1 × 10−11 mol s−1 1.17 Associative Compared with the inert base surface, the edge [87a]
cm−1 E-NRR of MoS2 is more active, and the positive
charge distributed around the Mo atoms can
effectively accept the lone electrons from N, so it
can adsorb N2.
DR MoS2/CPE 0.1 m Na2SO4 −0.4 29.28 µg h–1 mgcat.–1 8.34 Associative Inert N2 can be activated by bare Mo atoms on the [87b]
edge of the defect;
DFT shows that the depletion of electrons
occurs in the region between the two N atoms,
while the accumulation of electrons occurs around
the NMo and NH bonds, so the charge is
transferred from the N2 molecule to the edge of the
diatomic Mo.
Co-doped MoS2−x 0.01 m H2SO4 −0.3 0.6 mmol h−1 gcat.−1 ≈10 Associative After Co doping, the sulfur vacancies (Sv) in MoS2 [87f ]
were partially modified, which directly caused the
overpotential of E-NRR to decrease from 1.62 to
0.59 eV.
MoN NA/CC 0.1 m HCl −0.3 3.0 × 10−10 mol s−1 1.15 Mvk MoxN follows the Mvk mechanism [88b]
cm−2 mechanism (or hybrid mechanism) in E-NRR, that is,
N in NH3 comes from MoxN, and the resulting
N vacancies are compensated by N2 in the
environment.
Ti3C2Tx (TO, OH) 0.1 m HCl −0.3 36.9 µg h−1 mgcat.−1 9.1 – The inertness of F will reduce the electrical [90]
conductivity of the material and thus will seriously
affect the electrochemical process dependent on
electron transport. This research has developed
Ti3C2Tx (TO, OH) nanosheets without F, which
increases the ammonia in the E-NRR.
MXene/TiFeOx-700 0.05 m H2SO4 −0.2 21.9 μg h−1 mgcat.−1 25.44 – In order to avoid the influence of F on [93]
electron transport, by masking the remaining
F in Ti3C2Tx, the high catalytic activity and
selectivity of E-21.9 µg h−NRR was
realized.

Adv. Mater. 2021, 33, 2007650 2007650 (36 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
TiO2/Ti3C2Tx 0.1 m HCl −0.55 32.17 µg h−1 mgcat.−1 16.07 Associative Oxygen-rich vacant TiO2 nanoparticles (NPs) grown [94b]
(−0.45 in situ on Ti3C2Tx nanosheets (TiO2/Ti3C2Tx) to
V) prepare Ti3C2Tx MXene;
The untreated high-conductivity Ti3C2Tx nanosheets
can not only promote electron transmission, but
also avoid self-aggregation of TiO2 NPs;
The oxygen vacancy serves as the main active site
for NH3 synthesis;
Compared with Ti3C2Tx or TiO2(101) alone,
TiO2(101)/Ti3C2Tx has the lowest ENRR energy
barrier (0.40 eV).
TiO2/Ti 0.1 m Na2SO4 −0.7 9.16 × 10−11 mol s−1 2.5 – TiO2 nanosheet arrays are deposited on the [96]
cm−2 highly conductive Ti plate, and the subsequent
electrochemical process will induce OV to be
enriched on TiO2, thus promoting the adsorption
and activation of N2.
Zr-TiO2 0.1 m KOH −0.45 8.90 µg h−1 cm−2 17.3 – Zr4+ with d-electron configuration and oxide structure [205]
is introduced into TiO2, and the large-size Zr4+
induced strain effect promotes the formation of
bi-Ti3+ pairs in TiO2, which in turn induces a large
amount of OV enrichment. TiO2 with bi-Ti3+ pair has
excellent E-NRR performance.
Bi NS 0.1 m Na2SO4 −0.8 13.23 µg h–1 mgcat.–1 10.46 – The high E-NRR electrocatalytic activity of [102]
Bi NS can be attributed to the sufficient
exposure of the edge sites in the mosaic
bismuth nanosheets and the effective p-orbital
electron delocalization
BiNCs with K+ 0.5 m K2SO4 −0.6 0.052 mmol h–1 66 Associative The 6p energy band of Bi can partially [105]
cm–2 cover the 2p orbital of N, and the strong
interaction caused by this promotes the high
activity of E-NRR, in which K+ plays a dual function
of stabilizing the E-NRR reaction intermediate and
regulating proton transfer.
PEBCD with Li+ 0.5 m Li2SO4 −0.7 2.01 µg h–1 cm–2 2.85 Associative Cyclic voltammetry (CV) proved that the electron- [152]
(−0.5 V) rich CO group on PEBCD combined with the
electron-deficient Li+ to form the OLi+ active
site during the electrocatalytic process, thereby
inhibiting the smooth progress of HER on PEBCD;
Theoretical calculations believe that the OLi+ active
sites on PEBCD can adsorb N2 to form [OLi+]
N2Hx and promote the alternate hydrogenation of
N2 to NH3.
PCN-NVx – −0.2 8.09 µg h–1 mgcat.–1 11.59 Associative A metal-free polymerized carbon [153]
and nitrogen material (PCN-NV4) enriched with a
large amount of nitrogen vacancy (NV) was
obtained by pyrolyzing the product of
polycondensation of melamine;
DFT calculations show that the binuclear
end (two C connected to the imidazole) near
NV can promote space electron transfer to
effectively adsorb N2, and the NN triple bond in
adsorbed N2 is significantly elongated (strongly
activated state);
Since the E-NRR catalytic activity of PCN-
NV4 is significantly higher than that of PCN
without NV, NV defect engineering is
believed to improve the E-NRR activity of
carbon-based materials.

Adv. Mater. 2021, 33, 2007650 2007650 (37 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
NPC 0.05 m H2SO4 −0.9 1.40 mmol h−1 gcat.−1 1.42 Associative Pyridine nitrogen and pyrrole nitrogen in NPC [158]
are the key to high-performance E-NRR, while the
presence of graphite nitrogen has no gain in E-NRR
performance;
The porous structure of NPC is conducive to
capturing N2 and stabilizing active intermediates,
and the higher HER overpotential can improve the
selectivity of E-NRR.
C-ZIF 0.1 m KOH −0.3 3.4 × 10−6 10.2 Associative DFT calculations believe that the carbon vacancies [159]
mol cm−2 h−1 in the three pyridine N sites are possible active
sites for E-NRR, where N2 can be adsorbed and
the dissociation of the NN triple bond can be
achieved;
Fe can inhibit the active site of E-NRR and
promote the side reaction of HER. On the
contrary, the electron donating effect of alkali
metal K+ is conducive to the smooth
progress of E-NRR.
CNS 0.25 m LiClO4 −1.19 97.18 µg h−1 cm−1 11.56 – The tip of the nanospikes can form a local [160]
electric field, and N2 near the electrode can
be captured and converted into NH3 under this
electric field;
The counter cations in the aqueous solution
can enhance the E-NRR activity of the
nanospikes in the order of K+ < Na+ < Li+, while
restricting the water environment near the tip of
the nanospikes can inhibit HER and promote N2 to
enter electric field.
BG 0.05 m H2SO4 −0.5 9.8 µg h−1 cm−1 10.8 Associative After doping with B, the sp2 orientation [161a]
of graphene is not destroyed, and the
2D planar conjugated structure with high
surface area continues to be maintained,
which is conducive to the full exposure of catalytic
active sites;
Due to the electron deficiency of B, it can
induce the electron loss of graphene, which
can promote the adhesion of N2 containing lone
electron pairs;
The electron-deficient B has proved to be
able to chemically adsorb the negatively
polarized O atoms, which can inhibit the
normal progress of HER, thereby improving the
selection of E-NRR.
BCN 0.1 m HCl −0.3 7.75 µg h−1 mgcat.−1 13.79 Associative Both experiments and theoretical calculations show [161c]
that the E-NRR activity of graphene doped with a
single heteroatom cannot reach the activity of B-N
pair doping;
The activity of E-NRR comes from the active
triggering of the B–N pair doped in graphene, and
the C atom close to the B–N pair is the active site
of E-NRR.
B-BCN 0.05 m Na2SO4 −0.6 8.39 µg h−1 cm−2 9.87 Associative The B-rich B-BCN has more Lewis acid sites, [206]
which is beneficial to the adsorption of N2 Lewis
bases;
Compared with N-BCN, the stronger bond
interaction between *NN and the closest B atom
on B-BCN determines that N2→*NN has no
energy input.

Adv. Mater. 2021, 33, 2007650 2007650 (38 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Table 2. Continued.

Catalysts Electrolyte Potential Ammonia yield rate FE [%] E-NRR Factors that enhance E-NRR or nhibit HER Ref.
(V vs RHE) pathway
NSG 0.1 m HCl −0.6 7.7 µg h−1 mgcat.−1 5.8 – S, N co-doped graphene materials (NSG) can realize [164]
the electrochemical conversion of N2 to NH3, and
compared to undoped or single-doped (S or N)
graphene, NSG has better E- NRR activity;
DFT calculations show that S and N co-doping can
produce more defects on graphene and promote
electron transfer, so NSG can adsorb N2 and weaken
NN.
BNS 0.1 m Na2SO4 −0.8 13.22 µg h−1 mgcat.−1 4.04 Associative The adsorption of N2 on 3B is favorable (they [168e]
defined the three adjacent Bs on B4C as 1B, 2B, and
3B), and it is adsorbed on B atom in the form of
remote association (the bond length of NN triple
bond stretched from 1.112 to 1.132 Å);
There is a large amount of charge
accumulation at the position where 3B and N2
molecules bind, and differential charge analysis
shows that the N2 molecules are injected with
charges (0.38e−);
The hydrogenation on the surface of B(104) can
effectively promote E-NRR and inhibit the side
reaction of HER, and this site becomes the active
site of E-NRR;
There is a certain degree of coincidence between
the 2p orbital of B and the 2p orbital of *N2, which
is attributed in part to σ feedback and π reverse
bonding.
FL-BP NSs 0.01 m HCl −0.7 31.37 µg h−1 mgcat.−1 5.07 Associative The fully stripped black phosphorous [172c]
(−0.6 V) nanosheets (FL-BP NSs) were used as E-NRR
catalysts to achieve a high ammonia yield of
31.37 µg h−1 mgcat.−1;
The electron density of HUMO and LUMO in FL-BP
NSs is concentrated on the jagged edge, and N2
can be converted into NH3 by way of association
mechanism (alternative hydrogenation) on this edge
structure.

“–” means that the information cannot be obtained.

N2-catalyst contact. ii) In addition to the commonly used H- to guide healthy E-NRR development, including E-NRR activity
type electrolytic cells, we advise researchers to employ flow metrics, product (NH3) detection standardization, catalyst sta-
cells for the E-NRR as these cells promote the efficient cap- bility metrics, and correct theoretical calculations. Although
ture of N2 by the electrolyte, which is an important factor in- the E-NRR field currently faces significant challenges, the
fluencing the rate of NH3 production. iii) At the initial stage advancement of catalyst preparation technology, in situ testing
of the E-NRR, we recommend that researchers increase the methods, and product detection methods is expected to deepen
catalyst feed to increase the electrocatalytic active area and our understanding of the E-NRR. It is believed that in the near
NH3 production; this will provide a clear distinction between future, distributed electrochemical NH3 production equipment
NH3 produced by the E-NRR and that introduced from the driven by clean energy will spread worldwide, and the Haber–
environment. These two methods can solve the current di- Bosch process will gradually lose its dominant position.
lemma and effectively discriminate E-NRR catalysts with
high catalytic activity.
Acknowledgements
In short, although it is still far from the industrialization
This work was financially supported by the National Key R&D Program
level (NH3 yield of up to 6120 µg h−1 cm−2, FE of up to 50%,
of China (2020YFB1505602, 2016YFB0100304, 2019YFC1804400),
and current density of up to 300 mA cm−2), the E-NRR field has National Natural Science Foundation of China (21677171, 21505154, and
taken an important step forward (Figure 22).[5c,173a,184a,202] Cur- 51621001), and the Double Tops Joint Fund of the Yunnan Science and
rently, there is an urgent need for a recognized field standard Technology Bureau and Yunnan University (2019FY003025).

Adv. Mater. 2021, 33, 2007650 2007650 (39 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Figure 22. E-NRR is in the initial exploration stage, and it is still a long way from industrialization. Related metric refs. [5c,173a,184a,202].

Conflict of Interest X. Liu, A. K. Engstfeld, S. Horch, B. Seger, I. E. L. Stephens,

K. Chan, C. Hahn, J. K. Nørskov, T. F. Jaramillo, I. Chorkendorff,
The authors declare no conflict of interest. Chem. Rev. 2019, 119, 7610; d) C. Chen, J. F. Khosrowabadi Kotyk,
S. W. Sheehan, Chem 2018, 4, 2571.
[5] a) R. Schlögl, Handbook of Heterogeneous Catalysis, Wiley-VCH,
Weinheim, Germany 2008; b) I. A. Amar, R. Lan, C. T. G. Petit,
Keywords S. Tao, J. Solid State Electrochem. 2011, 15, 1845; c) S. Giddey,
electrochemical nitrogen reduction reaction, E-NRR, nitrogen fixation, S. P. S. Badwal, A. Kulkarni, Int. J. Hydrogen Energy 2013, 38, 14576.
synthetic ammonia [6] a) Mineral Commodity Summaries, U.S. Geological Survey, Reston,
VA 2009; b) H. J. M. van Grinsven, L. Bouwman, K. G. Cassman,
Received: November 10, 2020 H. M. van Es, M. L. McCrackin, A. H. W. Beusen, J. Environ. Qual.
Revised: February 1, 2021 2015, 44, 356.
Published online: July 1, 2021 [7] a) Q. Wang, G. Zheng, S. Hao, X. Liu, J. Zheng, Y. Wang, Z. Su,
N. Xu, Y. He, L. Lei, X. Zhang, ACS Sustainable Chem. Eng. 2020,
8, 44; b) K. A. Brown, D. F. Harris, M. B. Wilker, A. Rasmussen,
N. Khadka, H. Hamby, S. Keable, G. Dukovic, J. W. Peters,
[1] a) G. W. Crabtree, M. S. Dresselhaus, M. V. Buchanan, Phys. Today L. C. Seefeldt, P. W. King, Science 2016, 352, 448; c) R. F. Service,
2004, 57, 39; b) D. G. Nocera, Acc. Chem. Res. 2012, 45, 767. Science 2014, 345, 610.
[2] a) J. E. B. Halliday, K. Hampson, N. Hanley, T. Lembo, J. P. Sharp, [8] a) R. Schlögl, Angew. Chem., Int. Ed. 2003, 42, 2004; b) V. Smil,
D. T. Haydon, S. Cleaveland, Science 2017, 357, 146; b) H. Sun, Nature 1999, 400, 415; c) V. Rosca, M. Duca, M. T. de Groot,
Z. Yan, F. Liu, W. Xu, F. Cheng, J. Chen, Adv. Mater. 2020, 32, M. T. M. Koper, Chem. Rev. 2009, 109, 2209.
1806326; c) Z.-P. Wu, X. F. Lu, S.-Q. Zang, X. W. Lou, Adv. Mater. [9] a) T. He, P. Pachfule, H. Wu, Q. Xu, P. Chen, Nat. Rev. Mater. 2016,
2020, 30, 1910274; d) L. Zhang, J. Xiao, H. Wang, M. Shao, ACS 1, 16059; b) J. G. Chen, R. M. Crooks, L. C. Seefeldt, K. L. Bren,
Catal. 2017, 7, 7855. R. M. Bullock, M. Y. Darensbourg, P. L. Holland, B. Hoffman,
[3] a) X. Tian, X. F. Lu, B. Y. Xia, X. W. Lou, Joule 2020, 4, 45; M. J. Janik, A. K. Jones, M. G. Kanatzidis, P. King, K. M. Lancaster,
b) A. Zitolo, N. Ranjbar-Sahraie, T. Mineva, J. Li, Q. Jia, S. Stamatin, S. V. Lymar, P. Pfromm, W. F. Schneider, R. R. Schrock, Science
G. F. Harrington, S. M. Lyth, P. Krtil, S. Mukerjee, E. Fonda, 2018, 360, eaar6611; c) M. Aziz, T. Oda, A. Morihara, T. Kashiwagi,
F. Jaouen, Nat. Commun. 2017, 8, 957; c) U. Martinez, S. Komini Energy Procedia 2017, 143, 674; d) G.-F. Chen, Y. F. Yuan, H. F. Jiang,
Babu, E. F. Holby, H. T. Chung, X. Yin, P. Zelenay, Adv. Mater. 2019, S.-Y. Ren, L.-X. Ding, L. Ma, T. P. Wu, J. Lu, H. H. Wang, Nat. Energy
31, 1806545; d) X. F. Lu, B. Y. Xia, S.-Q. Zang, X. W. Lou, Angew. 2020, 5, 605.
Chem., Int. Ed. 2020, 59, 4634. [10] D. Bao, Q. Zhang, F.-L. Meng, H.-X. Zhong, M.-M. Shi, Y. Zhang,
[4] a) Y. Y. Birdja, E. Pérez-Gallent, M. C. Figueiredo, A. J. Göttle, J.-M. Yan, Q. Jiang, X.-B. Zhang, Adv. Mater. 2017, 29, 1604799.
F. Calle-Vallejo, M. T. M. Koper, Nat. Energy 2019, 4, 732; b) X. Bai, [11] a) X. Cui, C. Tang, Q. Zhang, Adv. Energy Mater. 2018, 8, 1800369;
W. Chen, B. Wang, G. Feng, W. Wei, Z. Jiao, Y. Sun, Acta Phys. b) C. J. M. van der Ham, M. T. M. Koper, D. G. H. Hetterscheid,
-Chim. Sin. 2017, 33, 2388; c) S. Nitopi, E. Bertheussen, S. B. Scott, Chem. Soc. Rev. 2014, 43, 5183; c) S. Wang, X. Hai, X. Ding,

Adv. Mater. 2021, 33, 2007650 2007650 (40 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

K. Chang, Y. Xiang, X. Meng, Z. Yang, H. Chen, J. Ye, Adv. Mater. Chem. 2012, 4, 177; d) J. Bonin, C. Costentin, M. Robert, M. Routier,
2017, 29, 1701774. J.-M. Savéant, J. Am. Chem. Soc. 2013, 135, 14359.
[12] a) M. Kitano, Y. Inoue, Y. Yamazaki, F. Hayashi, S. Kanbara, [31] C. V. S. Kumar, V. Subramanian, Phys. Chem. Chem. Phys. 2017, 19,
S. Matsuishi, T. Yokoyama, S.-W. Kim, M. Hara, H. Hosono, Nat. 15377.
Chem. 2012, 4, 934; b) M. Ali, F. Zhou, K. Chen, C. Kotzur, C. Xiao, [32] A. J. Göttle, M. T. M. Koper, Chem. Sci. 2017, 8, 458.
L. Bourgeois, X. Zhang, D. R. MacFarlane, Nat. Commun. 2016, 7, [33] J. Deng, J. A. Iñiguez, C. Liu, Joule 2018, 2, 846.
11335; c) C. Guo, J. Ran, A. Vasileff, S.-Z. Qiao, Energy Environ. Sci. [34] a) S. Dahl, E. Törnqvist, I. Chorkendorff, J. Catal. 2000,
2018, 11, 45. 192, 381; b) Á. Logadóttir, J. K. Nørskov, J. Catal. 2003, 220,
[13] Y. Tanabe, Y. Nishibayashi, Coord. Chem. Rev. 2013, 257, 2551. 273; c) J. H. Montoya, C. Tsai, A. Vojvodic, J. K. Nørskov,
[14] a) M. Appl, Ullmann’s Encyclopedia of Industrial Chemistry, Wiley- ChemSusChem 2015, 8, 2180; d) A. Hellman, E. J. Baerends,
VCH, Weinheim 2006; b) F. Lai, W. Zong, G. He, Y. Xu, H. Huang, M. Biczysko, T. Bligaard, C. H. Christensen, D. C. Clary, S. Dahl,
B. Weng, D. Rao, J. A. Martens, J. Hofkens, T. Liu, I. P. Parkin, R. van Harrevelt, K. Honkala, H. Jonsson, G. J. Kroes, M. Luppi,
Angew. Chem., Int. Ed. 2020, 59, 13320. U. Manthe, J. K. Nørskov, R. A. Olsen, J. Rossmeisl, E. Skúlason,
[15] S. L. Foster, S. I. P. Bakovic, R. D. Duda, S. Maheshwari, C. S. Tautermann, A. J. C. Varandas, J. K. Vincent, J. Phys. Chem. B
R. D. Milton, S. D. Minteer, M. J. Janik, J. N. Renner, L. F. Greenlee, 2006, 110, 17719.
Nat. Catal. 2018, 1, 490. [35] a) A. L. Garden, E. Skúlason, J. Phys. Chem. C 2015, 119, 26554;
[16] a) L. C. Seefeldt, B. M. Hoffman, D. R. Dean, Curr. Opin. Chem. b) M. Antonietti, M. Oschatz, Adv. Mater. 2018, 30, 1706836;
Biol. 2012, 16, 19; b) K. M. Lancaster, M. Roemelt, P. Ettenhuber, c) A. R. Singh, B. A. Rohr, M. J. Statt, J. A. Schwalbe, M. Cargnello,
Y. Hu, M. W. Ribbe, F. Neese, U. Bergmann, S. DeBeer, Science 2011, J. K. Nørskov, ACS Catal. 2019, 9, 8316.
334, 974; c) T. Spatzal, M. Aksoyoglu, L. Zhang, S. L. A. Andrade, [36] V. Kordali, G. Kyriacou, C. Lambrou, Chem. Commun. 2000, 17, 1673.
E. Schleicher, S. Weber, D. C. Rees, O. Einsle, Science 2011, 334, 940. [37] D. Wang, L. M. Azofra, M. Harb, L. Cavallo, X. Zhang,
[17] a) R. Cai, S. D. Minteer, ACS Energy Lett. 2018, 3, 2736; B. H. R. Suryanto, D. R. MacFarlane, ChemSusChem 2018, 11, 3416.
b) R. Bjornsson, F. A. Lima, T. Spatzal, T. Weyhermüller, P. Glatzel, [38] Z. Geng, Y. Liu, X. Kong, P. Li, K. Li, Z. Liu, J. Du, M. Shu, R. Si,
E. Bill, O. Einsle, F. Neese, S. DeBeer, Chem. Sci. 2014, 5, 3096; J. Zeng, Adv. Mater. 2018, 30, 1803498.
c) S. M. Keable, O. A. Zadvornyy, L. E. Johnson, B. Ginovska, [39] H. Tao, C. Choi, L.-X. Ding, Z. Jiang, Z. Han, M. Jia, Q. Fan, Y. Gao,
A. J. Rasmussen, K. Danyal, B. J. Eilers, G. A. Prussia, A. X. LeVan, H. Wang, A. W. Robertson, S. Hong, Y. Jung, S. Liu, Z. Sun, Chem
S. Raugei, L. C. Seefeldt, J. W. Peters, J. Biol. Chem. 2018, 293, 9629; 2019, 5, 204.
d) D. Sippel, O. Einsle, Nat. Chem. Biol. 2017, 13, 956. [40] Y. Cao, Y. Gao, H. Zhou, X. Chen, H. Hu, S. Deng, X. Zhong,
[18] a) T. Oshikiri, K. Ueno, H. Misawa, Angew. Chem., Int. Ed. 2014, 53, G. Zhuang, J. Wang, Adv. Theory Simul. 2018, 1, 1800018.
9802; b) H. Li, J. Shang, Z. Ai, L. Zhang, J. Am. Chem. Soc. 2015, [41] B. Yu, H. Li, J. White, S. Donne, J. Yi, S. Xi, Y. Fu, G. Henkelman,
137, 6393. H. Yu, Z. Chen, T. Ma, Adv. Funct. Mater. 2020, 30, 1905665.
[19] R. A. J. Janssen, J. Nelson, Adv. Mater. 2013, 25, 13. [42] C. Liu, Q. Li, J. Zhang, Y. Jin, D. R. MacFarlane, C. Sun, J. Mater.
[20] a) H. Wang, Y. Li, C. Li, K. Deng, Z. Wang, Y. Xu, X. Li, H. Xue, Chem. A 2019, 7, 4771.
L. Wang, J. Mater. Chem. A 2019, 7, 801; b) J. Zhao, L. Zhang, [43] a) K. Kugler, M. Luhn, J. A. Schramm, K. Rahimi, M. Wessling,
X.-Y. Xie, X. Li, Y. Ma, Q. Liu, W.-H. Fang, X. Shi, G. Cui, X. Sun, Phys. Chem. Chem. Phys. 2015, 17, 3768; b) H. Wang, Y. Li, D. Yang,
J. Mater. Chem. A 2018, 6, 24031. X. Qian, Z. Wang, Y. Xu, X. Li, H. Xue, L. Wang, Nanoscale 2019, 11,
[21] C. S. Diercks, Y. Liu, K. E. Cordova, O. M. Yaghi, Nat. Mater. 2018, 5499; c) R. Manjunatha, A. Schechter, Electrochem. Commun. 2018,
17, 301. 90, 96.
[22] A. R. Singh, B. A. Rohr, J. A. Schwalbe, M. Cargnello, K. Chan, [44] a) S. Back, Y. Jung, Phys. Chem. Chem. Phys. 2016, 18, 9161;
T. F. Jaramillo, I. Chorkendorff, J. K. Nørskov, ACS Catal. 2017, 7, 706. b) A. Ishikawa, T. Doi, H. Nakai, J. Catal. 2018, 357, 213.
[23] A. E. Shilov, Russ. Chem. Bull. 2003, 52, 2555. [45] N. Agmon, Chem. Phys. Lett. 1995, 244, 456.
[24] H.-P. Jia, E. A. Quadrelli, Chem. Soc. Rev. 2014, 43, 547. [46] J. Wang, L. Yu, L. Hu, G. Chen, H. Xin, X. Feng, Nat. Commun.
[25] a) S. Wang, F. Ichihara, H. Pang, H. Chen, J. Ye, Adv. Funct. Mater. 2018, 9, 1795.
2018, 28, 1803309; b) X.-L. Ma, J.-C. Liu, H. Xiao, J. Li, J. Am. Chem. [47] W. Xu, G. Fan, J. Chen, J. Li, L. Zhang, S. Zhu, X. Su, F. Cheng,
Soc. 2018, 140, 46. J. Chen, Angew. Chem., Int. Ed. 2020, 59, 3511.
[26] a) E. Skúlason, T. Bligaard, S. Gudmundsdóttir, F. Studt, [48] J. Lv, S. Wu, Z. Tian, Y. Ye, J. Liu, C. Liang, J. Mater. Chem. A 2019,
J. Rossmeisl, F. Abild-Pedersen, T. Vegge, H. Jónsson, J. K. Nørskov, 7, 12627.
Phys. Chem. Chem. Phys. 2012, 14, 1235; b) J. G. Howalt, T. Bligaard, [49] a) H. Zhao, D. Zhang, Z. Wang, Y. Han, X. Sun, H. Li, X. Wu, Y. Pan,
J. Rossmeisl, T. Vegge, Phys. Chem. Chem. Phys. 2013, 15, 7785. Y. Qin, S. Lin, Z. Xu, J. Lai, L. Wang, Appl. Catal., B 2020, 265,
[27] a) T. A. Bazhenova, A. E. Shilov, Coord. Chem. Rev. 1995, 144, 69; 118481; b) J. Guo, H. Wang, F. Xue, D. Yu, L. Zhang, S. Jiao, Y. Liu,
b) S. G. Bratsch, J. Phys. Chem. Ref. Data 1989, 18, 1. Y. Lu, M. Liu, S. Ruan, Y.-J. Zeng, C. Ma, H. Huang, J. Mater. Chem.
[28] a) D. Yan, H. Li, C. Chen, Y. Zou, S. Wang, Small Methods 2019, 3, A 2019, 7, 20247.
1800331; b) X. Yan, D. Liu, H. Cao, F. Hou, J. Liang, S. X. Dou, Small [50] M.-M. Shi, D. Bao, S.-J. Li, B.-R. Wulan, J.-M. Yan, Q. Jiang, Adv.
Methods 2019, 3, 1800501. Energy Mater. 2018, 8, 1800124.
[29] a) J. Rittle, J. C. Peters, J. Am. Chem. Soc. 2016, 138, 4243; [51] a) F. Pang, F. Wang, L. Yang, Z. Wang, W. Zhang, Chem. Commun.
b) Y. Abghoui, E. Skúlason, Catal. Today 2017, 286, 69; c) T. J. Del 2019, 55, 10108; b) G. Deng, T. Wang, A. A. Alshehri, K. A. Alzahrani,
Castillo, N. B. Thompson, J. C. Peters, J. Am. Chem. Soc. 2016, Y. Wang, H. Ye, Y. Luo, X. Sun, J. Mater. Chem. A 2019, 7, 21674.
138, 5341; d) J. S. Anderson, G. E. Cutsail, J. Rittle, B. A. Connor, [52] H. Wang, H. Yu, Z. Wang, Y. Li, Y. Xu, X. Li, H. Xue, L. Wang, Small
W. A. Gunderson, L. Zhang, B. M. Hoffman, J. C. Peters, 2019, 15, 1804769.
J. Am. Chem. Soc. 2015, 137, 7803; e) Y. Abghoui, A. L. Garden, [53] Y. Yang, S.-Q. Wang, H. Wen, T. Ye, J. Chen, C.-P. Li, M. Du, Angew.
J. G. Howalt, T. Vegge, E. Skúlason, ACS Catal. 2016, 6, 635. Chem., Int. Ed. 2019, 58, 15362.
[30] a) S. C. S. Lai, S. E. F. Kleijn, F. T. Z. Öztürk, V. C. van Rees Vellinga, [54] J. Zheng, Y. Lyu, M. Qiao, R. Wang, Y. Zhou, H. Li, C. Chen, Y. Li,
J. Koning, P. Rodriguez, M. T. M. Koper, Catal. Today 2010, 154, 92; H. Zhou, S. P. Jiang, S. Wang, Chem 2019, 5, 617.
b) Y. Kwon, S. C. S. Lai, P. Rodriguez, M. T. M. Koper, J. Am. Chem. [55] Z. Wang, Y. Li, H. Yu, Y. Xu, H. Xue, X. Li, H. Wang, L. Wang,
Soc. 2011, 133, 6914; c) P. Rodriguez, Y. Kwon, M. T. M. Koper, Nat. ChemSusChem 2018, 11, 3480.

Adv. Mater. 2021, 33, 2007650 2007650 (41 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

[56] M. Nazemi, S. R. Panikkanvalappil, M. A. El-Sayed, Nano Energy [80] a) X. Cui, C. Tang, X.-M. Liu, C. Wang, W. Ma, Q. Zhang, Chem. -
2018, 49, 316. Eur. J. 2018, 24, 18494; b) M. Knies, M. Kaiser, A. Isaeva, U. Müller,
[57] M.-M. Shi, D. Bao, B.-R. Wulan, Y.-H. Li, Y.-F. Zhang, J.-M. Yan, T. Doert, M. Ruck, Chem. - Eur. J. 2018, 24, 1.
Q. Jiang, Adv. Mater. 2017, 29, 1606550. [81] J. Kong, A. Lim, C. Yoon, J. H. Jang, H. C. Ham, J. Han, S. Nam,
[58] Y. Wang, X. G. Gong, J. Chem. Phys. 2006, 125, 124703. D. Kim, Y.-E. Sung, J. Choi, H. S. Park, ACS Sustainable Chem. Eng.
[59] S.-J. Li, D. Bao, M.-M. Shi, B.-R. Wulan, J.-M. Yan, Q. Jiang, Adv. 2017, 5, 10986.
Mater. 2017, 29, 1700001. [82] a) L. Hu, A. Khaniya, J. Wang, G. Chen, W. E. Kaden, X. Feng, ACS
[60] X. Wang, W. Wang, M. Qiao, G. Wu, W. Chen, T. Yuan, Q. Xu, Catal. 2018, 8, 9312; b) Q. Liu, X. Zhang, B. Zhang, Y. Luo, G. Cui,
M. Chen, Y. Zhang, X. Wang, J. Wang, J. Ge, X. Hong, Y. Li, Y. Wu, F. Xie, X. Sun, Nanoscale 2018, 10, 14386; c) B. H. R. Suryanto,
Y. Li, Sci. Bull. 2018, 63, 1246. C. S. M. Kang, D. Wang, C. Xiao, F. Zhou, L. M. Azofra, L. Cavallo,
[61] D. Liu, G. Zhang, Q. Ji, Y. Zhang, J. Li, ACS Appl. Mater. Interfaces X. Zhang, D. R. MacFarlane, ACS Energy Lett. 2018, 3, 1219.
2019, 11, 25758. [83] a) S. Chen, S. Perathoner, C. Ampelli, C. Mebrahtu, D. Su, G. Centi,
[62] a) X. Zhao, C. Yao, H. Chen, Y. Fu, C. Xiang, S. He, X. Zhou, Angew. Chem., Int. Ed. 2017, 56, 2699; b) S. Chen, S. Perathoner,
H. Zhang, J. Mater. Chem. A 2019, 7, 20945; b) X. Zhao, Z. Yang, C. Ampelli, C. Mebrahtu, D. Su, G. Centi, ACS Sustainable Chem.
W. Wang, Y. Li, X. Zhou, H. Zhang, J. Mater. Chem. A 2020, 8, 7171; Eng. 2017, 5, 7393.
c) Z. Wang, Y. Liu, H. Zhang, X. Zhou, J. Colloid Interface Sci. 2020, [84] J. Li, X. Zhu, T. Wang, Y. Luo, X. Sun, Inorg. Chem. Front. 2019, 6,
566, 135. 2682.
[63] X. Zhao, Z. Yang, A. Kuklin, G. Baryshnikov, H. Ågren, W. Wang, [85] a) S. Zhang, G. Duan, L. Qiao, Y. Tang, Y. Chen, Y. Sun, P. Wan,
X. Zhou, H. Zhang, J. Mater. Chem. A 2020, 8, 13086. S. Zhang, Ind. Eng. Chem. Res. 2019, 58, 8935; b) L. Zeng, X. Li,
[64] S. Zhao, H.-X. Liu, Y. Qiu, S.-Q. Liu, J.-X. Diao, C.-R. Chang, R. Si, S. Chen, J. Wen, W. Huang, A. Chen, J. Mater. Chem. A 2020, 8,
X.-H. Guo, J. Mater. Chem. A 2020, 8, 6586. 7339; c) L. M. Azofra, C. Sun, L. Cavallo, D. R. MacFarlane, Chem. -
[65] G. Liu, Z. Cui, M. Han, S. Zhang, C. Zhao, C. Chen, G. Wang, Eur. J. 2017, 23, 8275.
H. Zhang, Chem. - Eur. J. 2019, 25, 5904. [86] a) D. Yang, T. Chen, Z. Wang, J. Mater. Chem. A 2017, 5, 18967;
[66] C. Hu, X. Chen, J. Jin, Y. Han, S. Chen, H. Ju, J. Cai, Y. Qiu, C. Gao, b) Q. Li, S. Qiu, C. Liu, M. Liu, L. He, X. Zhang, C. Sun, J. Phys.
C. Wang, Z. Qi, R. Long, L. Song, Z. Liu, Y. Xiong, J. Am. Chem. Soc. Chem. C 2019, 123, 2347; c) C. Ling, X. Bai, Y. Ouyang, A. Du,
2019, 141, 7807. J. Wang, J. Phys. Chem. C 2018, 122, 16842; d) L. Han, X. Liu, J. Chen,
[67] Z.-H. Xue, S.-N. Zhang, Y.-X. Lin, H. Su, G.-Y. Zhai, J.-T. Han, R. Lin, H. Liu, F. Lü, S. Bak, Z. Liang, S. Zhao, E. Stavitski, J. Luo,
Q.-Y. Yu, X.-H. Li, M. Antonietti, J.-S. Chen, J. Am. Chem. Soc. 2019, R. R. Adzic, H. L. Xin, Angew. Chem., Int. Ed. 2019, 58, 2321; e)
141, 14976. L. Hui, Y. Xue, H. Yu, Y. Liu, Y. Fang, C. Xing, B. Huang, Y. Li, J. Am.
[68] J. Zheng, Y. Lyu, M. Qiao, J. P. Veder, R. D. Marco, J. Bradley, Chem. Soc. 2019, 141, 10677; f) L. Han, X. Liu, J. Chen, R. Lin, H. Liu,
R. Wang, Y. Li, A. Huang, S. P. Jiang, S. Wang, Angew. Chem., Int. F. Lü, S. Bak, Z. Liang, S. Zhao, E. Stavitski, J. Luo, R. R. Adzic,
Ed. 2019, 58, 18604. H. L. Xin, Angew. Chem., Int. Ed. 2019, 58, 2525; g) K. Chu, J. Wang,
[69] Y. Yao, S. Zhu, H. Wang, H. Li, M. Shao, J. Am. Chem. Soc. 2018, Y.-p. Liu, Q.-q. Li, Y.-l. Guo, J. Mater. Chem. A 2020, 8, 7117; h) Y. Liu,
140, 1496. X. Zhu, Q. Zhang, T. Tang, Y. Zhang, L. Gu, Y. Li, J. Bao, Z. Dai,
[70] H. Huang, L. Xia, X. Shi, A. M. Asiri, X. Sun, Chem. Commun. 2018, J.-S. Hu, J. Mater. Chem. A 2020, 8, 8920.
54, 11427. [87] a) L. Zhang, X. Ji, X. Ren, Y. Ma, X. Shi, Z. Tian, A. M. Asiri, L. Chen,
[71] H.-M. Liu, S.-H. Han, Y. Zhao, Y.-Y. Zhu, X.-L. Tian, J.-H. Zeng, B. Tang, X. Sun, Adv. Mater. 2018, 30, 1800191; b) X. Li, T. Li, Y. Ma,
J.-X. Jiang, B. Y. Xia, Y. Chen, J. Mater. Chem. A 2018, 6, 3211. Q. Wei, W. Qiu, H. Guo, X. Shi, P. Zhang, A. M. Asiri, L. Chen,
[72] a) R. Lan, J. T. S. Irvine, S. Tao, Sci. Rep. 2013, 3, 1145; b) C. Huang, B. Tang, X. Sun, Adv. Energy Mater. 2018, 8, 1801357; c) L. Zeng,
L. Shang, P. Han, Z. Gu, A. M. Al-Enizi, T. M. Almutairi, N. Cao, S. Chen, J. van der Zalm, X. Li, A. Chen, Chem. Commun. 2019, 55,
G. Zheng, J. Colloid Interface Sci. 2019, 552, 312; c) Y.-J. Mao, 7386; d) X. Li, X. Ren, X. Liu, J. Zhao, X. Sun, Y. Zhang, X. Kuang,
L. Wei, X.-S. Zhao, Y.-S. Wei, J.-W. Li, T. Sheng, F.-C. Zhu, N. Tian, T. Yan, Q. Wei, D. Wu, J. Mater. Chem. A 2019, 7, 2524; e) Y. Liu,
Z.-Y. Zhou, S.-G. Sun, Chem. Commun. 2019, 55, 9335; d) R. Lan, W. Wang, S. Zhang, W. Li, G. Wang, Y. Zhang, M. Han, H. Zhang,
S. Tao, RSC Adv. 2013, 3, 18016. ACS Sustainable Chem. Eng. 2020, 8, 2320; f) J. Zhang, X. Tian,
[73] R. Hao, W. Sun, Q. Liu, X. Liu, J. Chen, X. Lv, W. Li, Y.-p. Liu, M. Liu, H. Guo, J. Zhou, Q. Fang, Z. Liu, Q. Wu, J. Lou, J. Am.
Z. Shen, Small 2020, 16, 2000015. Chem. Soc. 2019, 141, 19269.
[74] X.-F. Li, Q.-K. Li, J. Cheng, L. Liu, Q. Yan, Y. Wu, X.-H. Zhang, [88] a) X. Ren, G. Cui, L. Chen, F. Xie, Q. Wei, Z. Tian, X. Sun, Chem.
Z.-Y. Wang, Q. Qiu, Y. Luo, J. Am. Chem. Soc. 2016, 138, 8706. Commun. 2018, 54, 8474; b) L. Zhang, X. Ji, X. Ren, Y. Luo, X. Shi,
[75] F. Lü, S. Zhao, R. Guo, J. He, X. Peng, H. Bao, J. Fu, L. Han, G. Qi, A. M. Asiri, B. Zheng, X. Sun, ACS Sustainable Chem. Eng. 2018, 6,
J. Luo, X. Tang, X. Liu, Nano Energy 2019, 61, 420. 9550.
[76] a) M. Wang, S. Liu, T. Qian, J. Liu, J. Zhou, H. Ji, J. Xiong, [89] a) H. Cheng, L.-X. Ding, G.-F. Chen, L. Zhang, J. Xue, H. Wang, Adv.
J. Zhong, C. Yan, Nat. Commun. 2019, 10, 341; b) Y. Wang, X. Cui, Mater. 2018, 30, 1803694; b) K. Ba, G. Wang, T. Ye, X. Wang, Y. Sun,
J. Zhao, G. Jia, L. Gu, Q. Zhang, L. Meng, Z. Shi, L. Zheng, H. Liu, A. Hu, Z. Li, Z. Sun, ACS Catal. 2020, 10, 7864.
C. Wang, Z. Zhang, W. Zheng, ACS Catal. 2019, 9, 336; c) C. He, [90] B. Hu, M. Hu, L. Seefeldt, T. L. Liu, ACS Energy Lett. 2019, 4, 1053.
Z.-Y. Wu, L. Zhao, M. Ming, Y. Zhang, Y. Yi, J.-S. Hu, ACS Catal. [91] Y. Luo, G.-F. Chen, L. Ding, X. Chen, L.-X. Ding, H. Wang, Joule
2019, 9, 7311; d) F. Xu, L. Zhang, X. Ding, M. Cong, Y. Jin, 2019, 3, 279.
L. Chen, Y. Gao, Chem. Commun. 2019, 55, 14111; e) K. Chu, [92] T. Li, X. Yan, L. Huang, J. Li, L. Yao, Q. Zhu, W. Wang, W. Abbas,
Y.-h. Cheng, Q.-q. Li, Y.-p. Liu, Y. Tian, J. Mater. Chem. A 2020, R. Naz, J. Gu, Q. Liu, W. Zhang, D. Zhang, J. Mater. Chem. A 2019,
8, 5865. 7, 14462.
[77] a) Y. Li, Q. Zhang, C. Li, H.-N. Fan, W.-B. Luo, H.-K. Liu, S.-X. Dou, [93] Y. Guo, T. Wang, Q. Yang, X. Li, H. Li, Y. Wang, T. Jiao, Z. Huang,
J. Mater. Chem. A 2019, 7, 22242; b) B. Qin, Y. Li, Q. Zhang, G. Yang, B. Dong, W. Zhang, J. Fan, C. Zhi, ACS Nano 2020, 14, 9089.
H. Liang, F. Peng, Nano Energy 2020, 68, 104374. [94] a) R. B. Rakhi, B. Ahmed, D. Anjum, H. N. Alshareef, ACS Appl.
[78] N. Cao, G. Zheng, Nano Res. 2018, 11, 2992. Mater. Interfaces 2016, 8, 18806; b) Y. Fang, Z. Liu, J. Han, Z. Jin,
[79] X. Xiang, Z. Wang, X. Shi, M. Fan, X. Sun, ChemCatChem 2018, 10, Y. Han, F. Wang, Y. Niu, Y. Wu, Y. Xu, Adv. Energy Mater. 2019, 9,
4530. 1803406.

Adv. Mater. 2021, 33, 2007650 2007650 (42 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

[95] W. Kong, F. Gong, Q. Zhou, G. Yu, L. Ji, X. Sun, A. M. Asiri, [121] W. Guo, Z. Liang, J. Zhao, B. Zhu, K. Cai, R. Zou, Q. Xu, Small
T. Wang, Y. Luo, Y. Xu, J. Mater. Chem. A 2019, 7, 18823. Methods 2018, 2, 1800204.
[96] R. Zhang, X. Ren, X. Shi, F. Xie, B. Zheng, X. Guo, X. Sun, ACS [122] P. Chen, N. Zhang, S. Wang, T. Zhou, Y. Tong, C. Ao, W. Yan,
Appl. Mater. Interfaces 2018, 10, 28251. L. Zhang, W. Chu, C. Wu, Y. Xie, Proc. Natl. Acad. Sci. U. S. A. 2019,
[97] Q. Qin, Y. Zhao, M. Schmallegger, T. Heil, J. Schmidt, R. Walczak, 116, 6635.
G. Gescheidt-Demner, H. Jiao, M. Oschatz, Angew. Chem., Int. Ed. [123] Y. Fang, Y. Xue, Y. Li, H. Yu, L. Hui, Y. Liu, C. Xing, C. Zhang,
2019, 58, 13101. D. Zhang, Z. Wang, X. Chen, Y. Gao, B. Huang, Y. Li, Angew.
[98] a) R. Michalsky, Y.-J. Zhang, A. J. Medford, A. A. Peterson, J. Phys. Chem., Int. Ed. 2020, 59, 13021.
Chem. C 2014, 118, 13026; b) R. Zhao, H. Xie, L. Chang, X. Zhang, [124] Y. Zhang, J. Hu, C. Zhang, Y. Liu, M. Xu, Y. Xue, L. Liu,
X. Zhu, X. Tong, T. Wang, Y. Luo, P. Wei, Z. Wang, X. Sun, M. K. H. Leung, J. Mater. Chem. A 2020, 8, 9091.
EnergyChem 2019, 1, 100011. [125] Y.-T. Liu, L. Tang, J. Dai, J. Yu, B. Ding, Angew. Chem., Int. Ed. 2020,
[99] G. Yu, H. Guo, W. Kong, T. Wang, Y. Luo, X. Shi, A. M. Asiri, T. Li, 59, 13623.
X. Sun, J. Mater. Chem. A 2019, 7, 19657. [126] X.-h. Wang, J. Wang, Y.-b. Li, K. Chu, ChemCatChem 2019, 11, 4529.
[100] M. Kunitski, N. Eicke, P. Huber, J. Köhler, S. Zeller, J. Voigtsberger, [127] X. Zhao, Z. Yang, A. V. Kuklin, G. V. Baryshnikov, H. Agren, W. Liu,
N. Schlott, K. Henrichs, H. Sann, F. Trinter, L. P. H. Schmidt, H. Zhang, X. Zhou, ACS Appl. Mater. Interfaces 2020, 12, 31419.
A. Kalinin, M. S. Schöffler, T. Jahnke, M. Lein, R. Dörner, Nat. [128] K. Kim, C.-Y. Yoo, J.-N. Kim, H. C. Yoon, J.-I. Han, J. Electrochem.
Commun. 2019, 10, 2877. Soc. 2016, 163, F1523.
[101] B. Li, X. Zhu, J. Wang, R. Xing, Q. Liu, X. Shi, Y. Luo, S. Liu, X. Niu, [129] K. Kim, N. Lee, C.-Y. Yoo, J.-N. Kim, H. C. Yoon, J.-I. Han,
X. Sun, Chem. Commun. 2020, 56, 1074. J. Electrochem. Soc. 2016, 163, F610.
[102] L. Li, C. Tang, B. Xia, H. Jin, Y. Zheng, S.-Z. Qiao, ACS Catal. 2019, [130] Y.-X. Lin, S.-N. Zhang, Z.-H. Xue, J.-J. Zhang, H. Su, T.-J. Zhao,
9, 2902. G.-Y. Zhai, X.-H. Li, M. Antonietti, J.-S. Chen, Nat. Commun. 2019,
[103] a) Y. Wang, M.-M. Shi, D. Bao, F.-I. Meng, Q. Zhang, Y.-T. Zhou, 10, 4380.
K.-H. Liu, Y. Zhang, J.-Z. Wang, Z.-W. Chen, D.-p. Liu, Z. Jiang, [131] T. Wu, H. Zhao, X. Zhu, Z. Xing, Q. Liu, T. Liu, S. Gao, S. Lu, G. Chen,
M. Luo, L. Gu, Q.-H. Zhang, X.-Z. Cao, Y. Yao, M.-H. Shao, A. M. Asiri, Y. Zhang, X. Sun, Adv. Mater. 2020, 32, 2000299.
Y. Zhang, X.-B. Zhang, J. G. Chen, J.-M. Yan, Q. Jiang, Angew. [132] T. Wu, M. Han, X. Zhu, G. Wang, Y. Zhang, H. Zhang, H. Zhao,
Chem., Int. Ed. 2019, 58, 9464; b) C. Lv, C. Yan, G. Chen, Y. Ding, J. Mater. Chem. A 2019, 7, 16969.
J. Sun, Y. Zhou, G. Yu, Angew. Chem., Int. Ed. 2018, 57, 6073. [133] J. Han, Z. Liu, Y. Ma, G. Cui, F. Xie, F. Wang, Y. Wu, S. Gao, Y. Xu,
[104] F. Wang, X. Lv, X. Zhu, J. Du, S. Lu, A. A. Alshehri, K. A. Alzahrani, X. Sun, Nano Energy 2018, 52, 264.
B. Zheng, X. Sun, Chem. Commun. 2020, 56, 2107. [134] Y. Wang, A. Chen, S. Lai, X. Peng, S. Zhao, G. Hu, Y. Qiu, J. Ren,
[105] Y.-C. Hao, Y. Guo, L.-W. Chen, M. Shu, X.-Y. Wang, T.-A. Bu, X. Liu, J. Luo, J. Catal. 2020, 381, 78.
W.-Y. Gao, N. Zhang, X. Su, X. Feng, J.-W. Zhou, B. Wang, C.-W. Hu, [135] L. Zhang, X. Ren, Y. Luo, X. Shi, A. M. Asiri, T. Li, X. Sun, Chem.
A.-X. Yin, R. Si, Y.-W. Zhang, C.-H. Yan, Nat. Catal. 2019, 2, 448. Commun. 2018, 54, 12966.
[106] C. Jaecheol, D. Hoang-Long, C. Manjunath, S. Bryan, [136] P. Li, W. Fu, P. Zhuang, Y. Cao, C. Tang, A. B. Watson, P. Dong,
H. R. S. Alexandr, M. Douglas, ChemRxiv, https://doi.org/10.26434/ J. Shen, M. Ye, Small 2019, 15, 1902535.
chemrxiv.11768814.v1. [137] X. Chen, Y.-T. Liu, C. Ma, J. Yu, B. Ding, J. Mater. Chem. A 2019, 7,
[107] Y. Wang, K. Jia, Q. Pan, Y. Xu, Q. Liu, G. Cui, X. Guo, X. Sun, ACS 22235.
Sustainable Chem. Eng. 2019, 7, 117. [138] P. Li, Z. Liu, T. Wu, Y. Zhang, L. Wang, L. Wang, L. Ji, Y. Zhang,
[108] T. Wu, W. Kong, Y. Zhang, Z. Xing, J. Zhao, T. Wang, X. Shi, Y. Luo, Y. Luo, T. Wang, S. Liu, Y. Wu, M. Liu, X. Sun, J. Mater. Chem. A
X. Sun, Small Methods 2019, 3, 1900356. 2019, 7, 17761.
[109] R. Zhang, Y. Zhang, X. Ren, G. Cui, A. M. Asiri, B. Zheng, X. Sun, [139] C. Li, D. Ma, S. Mou, Y. Luo, B. Ma, S. Lu, G. Cui, Q. Li, Q. Liu,
ACS Sustainable Chem. Eng. 2018, 6, 9545. X. Sun, J. Energy Chem. 2020, 50, 402.
[110] X. Yang, J. Nash, J. Anibal, M. Dunwell, S. Kattel, E. Stavitski, [140] H. Jin, L. Li, X. Liu, C. Tang, W. Xu, S. Chen, L. Song, Y. Zheng,
K. Attenkofer, J. G. Chen, Y. Yan, B. Xu, J. Am. Chem. Soc. 2018, S.-Z. Qiao, Adv. Mater. 2019, 31, 1902709.
140, 13387. [141] M. Yang, R. Huo, H. Shen, Q. Xia, J. Qiu, A. W. Robertson, X. Li,
[111] X. Zhang, R.-M. Kong, H. Du, L. Xia, F. Qu, Chem. Commun. 2018, Z. Sun, ACS Sustainable Chem. Eng. 2020, 8, 2957.
54, 5323. [142] X. Liu, H. Jang, P. Li, J. Wang, Q. Qin, M. G. Kim, G. Li, J. Cho,
[112] W. Fang, J. Zhao, T. Wu, Y. Huang, L. Yang, C. Liu, Q. Zhang, Angew. Chem., Int. Ed. 2019, 58, 13329.
K. Huang, Q. Yan, J. Mater. Chem. A 2020, 8, 5913. [143] Y. Han, W. Cai, X. Wu, W. Qi, B. Li, H. Li, D. Zhang, Y. Pan,
[113] G. Yu, H. Guo, S. Liu, L. Chen, A. A. Alshehri, K. A. Alzahrani, Z. Wang, J. Lai, L. Wang, Cell Rep. Phys. Sci. 2020, 1, 100232.
F. Hao, T. Li, ACS Sustainable Chem. Eng. 2019, 11, 35764. [144] J. Liu, X. Kong, L. Zheng, X. Guo, X. Liu, J. Shui, ACS Nano 2020,
[114] Y. Zhang, W. Qiu, Y. Ma, Y. Luo, Z. Tian, G. Cui, F. Xie, L. Chen, 14, 1093.
T. Li, X. Sun, ACS Catal. 2018, 8, 8540. [145] W. Fu, P. Zhuang, M. OliverLam Chee, P. Dong, M. Ye, J. Shen,
[115] H. Du, X. Guo, R.-M. Kong, F. Qu, Chem. Commun. 2018, 54, ACS Sustainable Chem. Eng. 2019, 7, 9622.
12848. [146] Y. Fu, T. Li, G. Zhou, J. Guo, Y. Ao, Y. Hu, J. Shen, L. Liu, X. Wu,
[116] Y. Yao, Q. Feng, S. Zhu, J. Li, Y. Yao, Y. Wang, Q. Wang, M. Gu, Nano Lett. 2020, 20, 4960.
H. Wang, H. Li, X.-Z. Yuan, M. Shao, Small Methods 2019, 3, [147] F. Lai, N. Chen, X. Ye, G. He, W. Zong, K. B. Holt, B. Pan,
1800324. I. P. Parkin, T. Liu, R. Chen, Adv. Funct. Mater. 2020, 30, 1907376.
[117] X. Wu, L. Xia, Y. Wang, W. Lu, Q. Liu, X. Shi, X. Sun, Small 2018, [148] Y. Li, T. Li, X. Zhu, A. A. Alshehri, K. A. Alzahrani, S. Lu, X. Sun,
14, 1803111. Chem. – Asian J. 2020, 15, 487.
[118] K. Chu, Y.-p. Liu, Y.-b. Li, H. Zhang, Y. Tian, J. Mater. Chem. A 2019, [149] Y. Zheng, Y. Jiao, M. Jaroniec, S. Z. Qiao, Angew. Chem., Int. Ed.
7, 4389. 2015, 54, 52.
[119] Y.-T. Liu, X. Chen, J. Yu, B. Ding, Angew. Chem., Int. Ed. 2019, 58, [150] N. Furuya, H. Yoshiba, J. Electroanal. Chem. Interfacial Electrochem.
18903. 1989, 272, 263.
[120] M. Qin, X. Li, G. Gan, L. Wang, S. Fan, Z. Yin, G. Chen, ACS [151] Y. Wang, X. Cui, Y. Zhang, L. Zhang, X. Gong, G. Zheng, Adv.
Sustainable Chem. Eng. 2020, 8, 13430. Mater. 2016, 28, 7626.

Adv. Mater. 2021, 33, 2007650 2007650 (43 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

[152] G.-F. Chen, X. Cao, S. Wu, X. Zeng, L.-X. Ding, M. Zhu, H. Wang, X.-F. Yu, Angew. Chem., Int. Ed. 2020, 59, 14383; c) L. Zhang,
J. Am. Chem. Soc. 2017, 139, 9771. L.-X. Ding, G.-F. Chen, X. Yang, H. Wang, Angew. Chem., Int. Ed.
[153] C. Lv, Y. Qian, C. Yan, Y. Ding, Y. Liu, G. Chen, G. Yu, Angew. 2019, 58, 2612; d) Y.-T. Liu, D. Li, J. Yu, B. Ding, Angew. Chem., Int.
Chem., Int. Ed. 2018, 57, 10246. Ed. 2019, 58, 16439.
[154] W. Li, T. Wu, S. Zhang, Y. Liu, C. Zhao, G. Liu, G. Wang, H. Zhang, [173] a) V. Kyriakou, I. Garagounis, E. Vasileiou, A. Vourros,
H. Zhao, Chem. Commun. 2018, 54, 11188. M. Stoukides, Catal. Today 2017, 286, 21. b) I. Garagounis,
[155] C. Hu, L. Dai, Adv. Mater. 2017, 29, 1604942. V. Kyriakou, A. Skodra, E. Vasileiou, M. Stoukides, Front. Energy
[156] H. Wang, L. Wang, Q. Wang, S. Ye, W. Sun, Y. Shao, Z. Jiang, Res. 2014, 2, 1.
Q. Qiao, Y. Zhu, P. Song, D. Li, L. He, X. Zhang, J. Yuan, T. Wu, [174] a) G. Xu, R. Liu, J. Wang, Sci. China, Ser. B: Chem. 2009, 52, 1171;
G. A. Ozin, Angew. Chem., Int. Ed. 2018, 57, 12360. b) R. Liu, G. Xu, Chin. J. Chem. 2010, 28, 139.
[157] X. Yang, K. Li, D. Cheng, W.-L. Pang, J. Lv, X. Chen, H.-Y. Zang, [175] G. Qing, R. Ghazfar, S. T. Jackowski, F. Habibzadeh,
X.-L. Wu, H.-Q. Tan, Y.-H. Wang, Y.-G. Li, J. Mater. Chem. A 2018, M. M. Ashtiani, C.-P. Chen, M. R. Smith, T. W. Hamann, Chem.
6, 7762. Rev. 2020, 120, 5437.
[158] Y. Liu, Y. Su, X. Quan, X. Fan, S. Chen, H. Yu, H. Zhao, Y. Zhang, [176] B. L. Sheets, G. G. Botte, Chem. Commun. 2018, 54, 4250.
J. Zhao, ACS Catal. 2018, 8, 1186. [177] A. Tsuneto, A. Kudo, T. Sakata, J. Electroanal. Chem. 1994,
[159] S. Mukherjee, D. A. Cullen, S. Karakalos, K. Liu, H. Zhang, S. Zhao, 367, 183.
H. Xu, K. L. More, G. Wang, G. Wu, Nano Energy 2018, 48, 217. [178] A. Tsuneto, A. Kudo, T. Sakata, Chem. Lett. 1993, 22, 851.
[160] Y. Song, D. Johnson, R. Peng, D. K. Hensley, P. V. Bonnesen, L. Liang, [179] J. M. McEnaney, A. R. Singh, J. A. Schwalbe, J. Kibsgaard, J. C. Lin,
J. Huang, F. Yang, F. Zhang, R. Qiao, A. P. Baddorf, T. J. Tschaplinski, M. Cargnello, T. F. Jaramillo, J. K. Nørskov, Energy Environ. Sci.
N. L. Engle, M. C. Hatzell, Z. Wu, D. A. Cullen, H. M. Meyer, 2017, 10, 1621.
B. G. Sumpter, A. J. Rondinone, Sci. Adv. 2018, 4, e1700336. [180] K. Kim, S. J. Lee, D.-Y. Kim, C.-Y. Yoo, J. W. Choi, J.-N. Kim, Y. Woo,
[161] a) X. Yu, P. Han, Z. Wei, L. Huang, Z. Gu, S. Peng, J. Ma, G. Zheng, H. C. Yoon, J.-I. Han, ChemSusChem 2018, 11, 120.
Joule 2018, 2, 1610; b) W.-B. Qiu, Y.-X. Luo, R.-P. Liang, J.-D. Qiu, [181] N. Lazouski, M. Chung, K. Williams, M. L. Gala, K. Manthiram,
X.-H. Xia, Chem. Commun. 2019, 55, 7406; c) C. Chen, D. Yan, Nat. Catal. 2020, 3, 463.
Y. Wang, Y. Zhou, Y. Zou, Y. Li, S. Wang, Small 2019, 15, 1805029. [182] C. S. M. Kang, X. Zhang, D. R. MacFarlane, J. Phys. Chem. C 2018,
[162] Y. Jiao, Y. Zheng, M. Jaroniec, S. Z. Qiao, J. Am. Chem. Soc. 2014, 122, 24550.
136, 4394. [183] a) J. L. Anderson, J. K. Dixon, J. F. Brennecke, Acc. Chem. Res.
[163] a) C. Hu, L. Dai, Angew. Chem., Int. Ed. 2016, 55, 11736; b) Y. Jiao, 2007, 40, 1208; b) S. Stevanovic, M. F. Costa Gomes, J. Chem.
Y. Zheng, M. Jaroniec, S. Z. Qiao, Chem. Soc. Rev. 2015, 44, 2060; Thermodyn. 2013, 59, 65; c) F. Zhou, L. M. Azofra, M. Ali,
c) M.-H. Yeh, Y.-A. Leu, W.-H. Chiang, Y.-S. Li, G.-L. Chen, T.-J. Li, M. Kar, A. N. Simonov, C. McDonnell-Worth, C. Sun, X. Zhang,
L.-Y. Chang, L.-Y. Lin, J.-J. Lin, K.-C. Ho, J. Power Sources 2018, 375, D. R. MacFarlane, Energy Environ. Sci. 2017, 10, 2516.
29; d) X. Wang, G. Sun, P. Routh, D.-H. Kim, W. Huang, P. Chen, [184] a) G.-F. Chen, S. Ren, L. Zhang, H. Cheng, Y. Luo, K. Zhu, L.-X. Ding,
Chem. Soc. Rev. 2014, 43, 7067. H. Wang, Small Methods 2019, 3, 1800337; b) S. Z. Andersen,
[164] Y. Tian, D. Xu, K. Chu, Z. Wei, W. Liu, J. Mater. Sci. 2019, 54, V. Čolić, S. Yang, J. A. Schwalbe, A. C. Nielander, J. M. McEnaney,
9088. K. Enemark-Rasmussen, J. G. Baker, A. R. Singh, B. A. Rohr,
[165] H. Huang, L. Xia, R. Cao, Z. Niu, H. Chen, Q. Liu, T. Li, X. Shi, M. J. Statt, S. J. Blair, S. Mezzavilla, J. Kibsgaard, P. C. K. Vesborg,
A. M. Asiri, X. Sun, Chem. - Eur. J. 2019, 25, 1914. M. Cargnello, S. F. Bent, T. F. Jaramillo, I. E. L. Stephens,
[166] T. Wu, P. Li, H. Wang, R. Zhao, Q. Zhou, W. Kong, M. Liu, J. K. Nørskov, I. Chorkendorff, Nature 2019, 570, 504.
Y. Zhang, X. Sun, F. Gong, Chem. Commun. 2019, 55, 2684. [185] Y. Ren, C. Yu, X. Tan, X. Han, H. Huang, H. Huang, J. Qiu, Small
[167] T. Wang, L. Xia, J.-J. Yang, H. Wang, W.-H. Fang, H. Chen, D. Tang, Methods 2019, 3, 1900474.
A. M. Asiri, Y. Luo, G. Cui, X. Sun, Chem. Commun. 2019, 55, 7502. [186] W. Peng, L. Xiao, B. Huang, L. Zhuang, J. Lu, J. Phys. Chem. C
[168] a) M.-A. Légaré, M.-A. Courtemanche, É. Rochette, F.-G. Fontaine, 2011, 115, 23050.
Science 2015, 349, 513; b) M.-A. Légaré, G. Bélanger-Chabot, [187] H. Liu, Y. Zhang, J. Luo, J. Energy Chem. 2020, 49, 51.
R. D. Dewhurst, E. Welz, I. Krummenacher, B. Engels, [188] a) C. Tang, S.-Z. Qiao, Chem. Soc. Rev. 2019, 48, 3166;
H. Braunschweig, Science 2018, 359, 896; c) C. Liu, Q. Li, J. Zhang, b) B. H. R. Suryanto, H.-L. Du, D. Wang, J. Chen, A. N. Simonov,
Y. Jin, D. R. MacFarlane, C. Sun, J. Phys. Chem. C 2018, 122, 25268; D. R. MacFarlane, Nat. Catal. 2019, 2, 290.
d) W. Qiu, X.-Y. Xie, J. Qiu, W.-H. Fang, R. Liang, X. Ren, X. Ji, [189] a) L. Zhou, C. E. Boyd, Aquaculture 2016, 450, 187; b) Y. Zhu,
G. Cui, A. M. Asiri, G. Cui, B. Tang, X. Sun, Nat. Commun. 2018, D. Yuan, H. Lin, T. Zhou, Anal. Lett. 2016, 49, 665.
9, 3485; e) X. Zhang, T. Wu, H. Wang, R. Zhao, H. Chen, T. Wang, [190] D. H. Thomas, M. Rey, P. E. Jackson, J. Chromatogr. A 2002, 956, 181.
P. Wei, Y. Luo, Y. Zhang, X. Sun, ACS Catal. 2019, 9, 4609; f) C. Liu, [191] A. LeDuy, R. Samson, Biotechnol. Lett. 1982, 4, 303.
Q. Li, C. Wu, J. Zhang, Y. Jin, D. R. MacFarlane, C. Sun, J. Am. [192] E. W. Rice, R. B. Baird, A. D. Eaton, L. S. lesceri, Standard Methods
Chem. Soc. 2019, 141, 2884. for the Examination of Water and Wastewater, 22nd ed., American
[169] a) Z. Chen, J. Zhao, L. Yin, Z. Chen, J. Mater. Chem. A 2019, 7, Water Works Association, Denver, CO 2012.
13284; b) X. Zhu, T. Wu, L. Ji, C. Li, T. Wang, S. Wen, S. Gao, [193] W. Yu, N. S. Lewis, H. B. Gray, N. F. Dalleska, ACS Energy Lett.
X. Shi, Y. Luo, Q. Peng, X. Sun, J. Mater. Chem. A 2019, 7, 2020, 5, 1532.
16117. [194] J. Liu, M. S. Kelley, W. Wu, A. Banerjee, A. P. Douvalis, J. Wu,
[170] D. W. Boukhvalov, Phys. Chem. Chem. Phys. 2015, 17, 27210. Y. Zhang, G. C. Schatz, M. G. Kanatzidis, Proc. Natl. Acad. Sci. U.
[171] a) L. Li, Y. Yu, G. J. Ye, Q. Ge, X. Ou, H. Wu, D. Feng, X. H. Chen, S. A. 2016, 113, 5530.
Y. Zhang, Nat. Nanotechnol. 2014, 9, 372; b) Z. Guo, H. Zhang, [195] Y. Zhao, R. Shi, X. Bian, C. Zhou, Y. Zhao, S. Zhang, F. Wu,
S. Lu, Z. Wang, S. Tang, J. Shao, Z. Sun, H. Xie, H. Wang, X.-F. Yu, G. I. N. Waterhouse, L.-Z. Wu, C.-H. Tung, T. Zhang, Adv. Sci. 2019,
P. K. Chu, Adv. Funct. Mater. 2015, 25, 6996. 6, 1802109.
[172] a) G. Xu, H. Li, A. S. R. Bati, M. Bat-Erdene, M. J. Nine, D. Losic, [196] N. Amornthammarong, J.-Z. Zhang, Anal. Chem. 2008, 80, 1019.
Y. Chen, J. G. Shapter, M. Batmunkh, T. Ma, J. Mater. Chem. A [197] a) E. Kun, E. B. Kearney, Methods of Enzymatic Analysis, 2nd
2020, 8, 15875; b) Q. Liu, X. Zhang, J. Wang, Y. Zhang, S. Bian, ed. (Ed: H. U. Bergmeyer), Academic Press, Boston, MA 1974;
Z. Cheng, N. Kang, H. Huang, S. Gu, Y. Wang, D. Liu, P. K. Chu, b) G. G. Guilbault, Anal. Chem. 1966, 38, 527.

Adv. Mater. 2021, 33, 2007650 2007650 (44 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

[198] L. F. Greenlee, J. N. Renner, S. L. Foster, ACS Catal. 2018, [203] L. Han, Z. Ren, P. Ou, H. Cheng, N. Rui, L. Lin, X. Liu, L. Zhuo,
8, 7820. J. Song, J. Sun, J. Luo, H. L. Xin, Angew. Chem., Int. Ed. 2020,
[199] L. Li, J. M. P. Martirez, E. A. Carter, ACS Catal. 2020, 10, 133, 349.
12841. [204] J. Zhang, B. Zhao, W. Liang, G. Zhou, Z. Liang, Y. Wang, J. Qu,
[200] L. Shi, Y. Yin, S. Wang, X. Xu, H. Wu, J. Zhang, S. Wang, H. Sun, Y. Sun, L. Jiang, Adv. Sci. 2020, 7, 2002630.
Appl. Catal., B 2020, 278, 119325. [205] N. Cao, Z. Chen, K. Zang, J. Xu, J. Zhong, J. Luo, X. Xu, G. Zheng,
[201] L. Shi, Y. Yin, S. Wang, H. Sun, ACS Catal. 2020, 10, 6870. Nat. Commun. 2019, 10, 2877.
[202] L. Wang, M. Xia, H. Wang, K. Huang, C. Qian, C. T. Maravelias, [206] B. Chang, L. Li, D. Shi, H. Jiang, Z. Ai, S. Wang, Y. Shao, J. Shen,
G. A. Ozin, Joule 2018, 2, 1055. Y. Wu, Y. Li, X. Hao, Appl. Catal., B 2021, 283, 119622.

Xue Zhao received his Master’s degree from Wuhan University in 2019. He is now pursuing
for Ph.D. in the College of Chemistry and Molecular Sciences of Wuhan University. His
current research interests mainly focus on the application of supramolecular nano-metals in
electrochemical and photoelectrochemistry.

Guangzhi Hu received his Ph.D. from Lanzhou Institute of Chemical Physics, Chinese Academy
of Sciences in 2010. After four years of post-doctoral research in the Nano Carbon Materials
Research Laboratory of the Department of Physics, Umeå University, Sweden, he joined the
Xinjiang Institute of Physics and Chemistry of the Chinese Academy of Sciences as a principle
investigator. Currently, Dr. Hu Guangzhi is a full time professor at the Institute for Ecological
Research and Pollution Control of Plateau Lakes of Yunnan University. His research interests
focus on the application of functional nanomaterials in energy and environment.

Gao-Feng Chen is currently pursuing his Ph.D. at South China University of Technology, China.
His research interests focus on electrocatalysis and energy materials, particularly in terms of
the nitrogen reduction reaction, oxygen reduction/evolution reactions, electrolytic hydrogen
production, and supercapacitors.

Adv. Mater. 2021, 33, 2007650 2007650 (45 of 46) © 2021 Wiley-VCH GmbH
 15214095, 2021, 33, Downloaded from https://advanced.onlinelibrary.wiley.com/doi/10.1002/adma.202007650 by Gwangju Institute Of Science And Technology (Gist), Wiley Online Library on [07/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.advmat.de

Haibo Zhang received his Ph.D. from Wuhan University in 2007. After two years studies as a
postdoctoral fellow in Wuhan University, He joined Wuhan University as an assistant professor in
2009. He is currently an associate professor at Wuhan University, China. His interests focus on
boron cluster supramolecular and boron cluster functional materials.

Shusheng Zhang received his Ph.D. from the University of Science and Technology of China
in 1998. Currently, he is a professor at the School of Chemistry and Molecular Engineering of
Zhengzhou University and a member of the Organic and Biological Analysis Committee of the
Chinese Chemical Society, and the director of the Chromatography Committee of the Chinese
Chemical Society. His main research interest is the detection and analysis of environmental
pollutants.

Haihui Wang received his Ph.D. from the Chinese Academy of Sciences in 2003. After one and
a half years’ study as an Alexander von Humboldt research fellow at the Institute of Physical
Chemistry and Electrochemistry, Leibniz University of Hannover, Germany, he continued his
studies there as a postdoctoral fellow in 2005. He is currently a professor at Tsinghua University,
China. His interests focus on inorganic membranes, catalysis, and energy materials.

Adv. Mater. 2021, 33, 2007650 2007650 (46 of 46) © 2021 Wiley-VCH GmbH