ACS H2O2 production

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

pubs.acs.

org/journal/ascecg Research Article

Enhancement in Photocatalytic H2O2 Production over g‑C3N4


Nanostructures: A Collaborative Approach of Nitrogen Deficiency
and Supramolecular Precursors
Hossein Fattahimoghaddam, Tahereh Mahvelati-Shamsabadi, and Byeong-Kyu Lee*
Cite This: ACS Sustainable Chem. Eng. 2021, 9, 4520−4530 Read Online

ACCESS Metrics & More Article Recommendations *


sı Supporting Information
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

ABSTRACT: The poor selectivity and activity of the photocatalysts


Downloaded via UNIV OF CONNECTICUT on May 15, 2021 at 15:36:17 (UTC).

developed has hindered the photocatalytic production of hydrogen


peroxide (H2O2). Therefore, modification techniques need to be
pursued to improve the selectivity and activity of photocatalysts
toward the two-electron oxygen reduction reaction. In this study,
homogeneous self-modification by nitrogen vacancies is adopted as
an efficient technique to narrow the band gap and induce the mid-
gap states in the nanostructures of g-C3N4 synthesized by cyanuric
acid−melamine supramolecular adducts. The presence of the
vacancies in the structure and their respective effects on the
optoelectronic features of the catalysts were thoroughly investigated
by different characterizations. The optimized photocatalyst showed
greatly improved H2O2 production of 200 μM under 1 h visible light
irradiation, as compared with that of the bulk (35 μM) and the
pristine nanostructured sample (85 μM), with acceptable reusability. This study also provides a new perspective on devising
synergistic approaches to modify the performance of supramolecular-based carbon nitride photocatalysts.
KEYWORDS: hydrogen peroxide production, photocatalysis, defect engineering, supramolecular structure, density of states

■ INTRODUCTION
Hydrogen peroxide (H2O2) has been widely used as a clean or
catalytic applications that include the photodegradation of
environmental pollutants,12−14 hydrogen evolution,15−17 CO2
environment-friendly oxidant for organic synthesis and reduction,18,19 and H2O2 production20 under visible light
environmental remediation, leaving only water as a by- irradiation. As seminal studies, Shiraishi et al. reported the
product.1−5 H2O2 can also play a key dual role in electricity ability of g-C3N4 to produce H2O2 through the two-electron
generation in membrane-less fuel cells or as an emerging liquid reduction reaction of molecular oxygen.21,22 They thoroughly
fuel.5 In these systems, H2O2 in the liquid phase with an energy explained the mechanism and revealed the role of the formed
density comparable to that of H2 in the compressed gaseous 1,4-endoperoxide species in suppressing the one-electron
phase can be employed as both an oxidant and energy carrier.6 reduction of O2 and improving the selectivity. They then
Conventional methods for H2O2 production comprise an prepared mesoporous g-C3N4 and investigated the effect of
anthraquinone method and a direct production with H2 and surface defects on its photocatalytic performance over H2O2
O2. However, these approaches suffer from several unavoidable production.23 The authors rationalized the results of this study
challenges, such as abundant energy and organic solvents as primary amines being able to act as active sites for the four-
consumption, as well as the potential explosion hazard of the electron reduction of oxygen and the H2O2 decomposition
mixture of oxygen and hydrogen.1,7 The photocatalytic oxygen reaction leading to decreased selectivity of H2O2 production.
reduction reaction has been recently introduced as an energy- As a result, they proposed that g-C3N4 catalysts having a large
saving, environment-friendly, facile, and safe method in this surface area with a lower content of primary amines would be
regard. Nevertheless, developing photocatalysts with appro-
priate performance is still a matter of considerable importance
in the sustainable conversion of solar light to chemical energy. Received: December 7, 2020
Graphitic carbon nitride (g-C3N4) as a metal-free photocatalyst Revised: February 15, 2021
with unique optical, chemical, and economic characteristics has Published: March 16, 2021
been widely investigated over the past few years for different
applications.8−11 Suitable band gap and band edge potentials of
g-C3N4 make it a promising candidate for different photo-

© 2021 American Chemical Society https://dx.doi.org/10.1021/acssuschemeng.0c08884


4520 ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

efficient photocatalysts for H2O2 production. In addition, Lei a unique structure at the expense of a decent amount of
et al. in their recently published work introduced inverse opal precursor and finally obtained an inadequate yield of
g-C3N4 with carbon vacancies as photocatalysts for such a approximately 5%. In the present study, we attempt to develop
purpose.24 Nevertheless, the practical application of g-C3N4 is a new strategy to enhance the catalyst yield and improve the
restricted by the low efficiency originating from their poor photocatalytic performance of the catalyst over oxygen
light-harvesting capability, high charge recombination rate, low reduction to H2O2. The idea is founded upon a collaboration
surface area, and so forth.25,26 Nanostructuring has been of employing the homogeneous modifications with vacancies,
considered an efficient approach for the photocatalytic activity as well as our previous results in obtaining high-density g-C3N4
improvement of g-C3N4 by facilitating charge transfer and nanostructures by utilizing CM supramolecular adducts as the
providing more active surface sites.27−30 On the other hand, precursor. In this method, an acceptable yield of g-C3N4 is
the quantum confinement effect dictates a band gap widening achieved by decreasing the calcination time of the CM
to the nanostructures and limits their capability of absorbing complex under air. Then, the obtained g-C3N4 structures are
visible light.27,31 Therefore, adopting new strategies to engineer recalcined under an argon environment to produce nanostruc-
the band gap energy of these semiconductors aimed at tures of graphitic carbon nitride with NVs. By adopting this
capturing the highest portion of sunlight would be of great strategy, we could successfully increase the final yield of the
interest. Several reports have been presented to address this catalysts, and by increasing the light-harvesting and charge
limitation using extrinsic modifications, such as heteroelement transfer efficiency enhance the photocatalytic performance of
doping,32−34 heterojunction design,35,36 and geometrical the catalysts. To the best of our knowledge, this is the first
structuring.37−39 Intrinsically modifying approaches have also study that reports the synergistic effect of using the CM
been recently employed to increase the light-harvesting complex-based carbon nitride with the nitrogen-vacancy
efficiency of these photocatalysts by utilizing homogeneous defects and utilizing such a catalyst for visible light-assisted
amorphization40,41 and homogeneous self-modifications with H2O2 production.
vacancies.42,43 These defect-containing structures exhibited an
extended response to visible light by the incredible lowering of
the band gap of these semiconducting materials, compared to
■ EXPERIMENTAL SECTION
CM Adduct Preparation. The cyanuric acid−melamine com-
that of the pristine g-C3N4. Niu et al. introduced nitrogen- plexes were prepared by the same procedure described in the author’s
vacancy defects by a temperature controlling route to promote previous study.56 Briefly, cyanuric acid and melamine in equimolar
the photocatalytic activity of g-C3N4.44 Inspired by their amounts were thoroughly mixed in distilled water for 6 h. The formed
findings of the outstanding role of oxygen vacancies in the supramolecular adducts were centrifuged and dried at 70 °C
photocatalytic performance of oxides, they inferred that overnight. The obtained CM complexes were used as the precursor
nitrogen vacancies (NVs) may positively affect the photo- for the synthesis of g-C3N4 nanostructures.
catalytic activity of nitrides. Therefore, they introduced NVs g-C3N4 Sample Preparation. The pristine nanostructured g-
C3N4 sample (CN) was prepared by heating of the obtained
into the structure of g-C3N4 by increasing the polycondensa- preorganized CM adducts from room temperature (RT) to 500 °C in
tion temperature, and the resultant sample showed extended 30 min and maintaining this temperature for 2 h in a muffle furnace
visible light utilization due to a narrowed band gap. Later, under static air. Afterward, the as-obtained CN sample was
several studies reported the effect of nitrogen defects on the recalcinated in a tubular furnace under an argon atmosphere to
improvement of the photocatalytic or electrocatalytic perform- prepare a nitrogen-deficient sample (CN-ND). Unless otherwise
ance of polymeric carbon nitrides for a variety of stated, the CN sample was reheated from RT to 600 °C in 20 min and
applications.45−49 For example, Tu et al. investigated the was kept at this temperature for 60 min to fabricate the CN-ND
application of g-C3N4 with tunable NVs for hydrogen sample. To find the optimum level of NVs, two other samples were
evolution and CO2 reduction. In that study, they reported a also prepared by the same procedure as the CN-ND sample, with (30
and 90 min) recalcination time entitled as CN-ND (L) and CN-ND
direct correlation between the optimum level of NVs in the (H), respectively. A control bulk g-C3N4 sample, denoted as MCN,
structure and photocatalytic H2 and CO production.50 What is was also synthesized via the same process as the CN sample, by using
more, g-C3N4 has been conventionally prepared by using low- melamine as the precursor.
cost nitrogen containing organic starting materials, such as Photocatalytic Experiments. Photocatalytic H2O2 production
melamine, urea, thiourea, cyanamide, dicyandiamide, and so under visible light irradiation was conducted to evaluate the
on, via thermal polycondensation.51 However, utilizing the photocatalytic activity of different catalyst samples. A 300 W xenon
preorganized supramolecular structures formed by hydrogen lamp equipped with UV and infrared (IR) cutoff filters was employed
bonds between cyanuric acid and melamine molecules (CM as the illumination system. Photocatalytic experiments for H2O2
production were conducted in 20 mL of aqueous solution containing
complex) has been recently introduced for g-C3N4 synthesis,
1 mg/mL of catalyst and 5% (v/v) ethanol as the hole scavenger. The
with highly efficient photocatalytic performance.12,52−55 mixture was then subjected to oxygen purging for 30 min to obtain an
In the preliminary studies conducted in our group, the oxygen saturated suspension. After that, the mixture was illuminated
authors could successfully prepare g-C3N4 nanostructures with for 1 h. Next, the catalyst was separated, and the H2O2 concentration
exceptional activity by calcination of the CM adducts prepared in the solution was determined by an iodometric method. In detail, 2
in water as a starting material for 3 h under static air mL of 100 mM potassium iodide and 50 μL of ammonium molybdate
atmosphere.56 The characterizations revealed that using the tetrahydrate (10 mM) were added to the supernatant and mixed for 5
CM adducts could form a highly dense structure of g-C3N4 min, and the resultant mixture was analyzed using UV−vis
with an extremely large surface area and relatively low primary spectrophotometry by applying the absorbency of iodine at 352
nm.24 To create an oxygen-free atmosphere, N2 was purged in the
amines, compared to the bulk structures obtained by melamine reaction mixture for 30 min instead of oxygen, and the photocatalytic
as a precursor. However, the photocatalysts obtained suffered experiment was performed by the same procedure described above.
from a marginal band gap of nearly 2.95 eV as a result of a Ethanol was omitted from the reaction mixture to see the effect of
strong quantum confinement effect limiting their visible light hole scavenger. The H2O2 decomposition kinetics in the presence of
absorption capability. Moreover, the authors could obtain such photocatalysts was studied under the same conditions as the H2O2

4521 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 1. (a) XRD patterns and (b) FT-IR spectra of the CN and CN-ND samples.

Figure 2. SEM images of (a) CM complex hexagonal assembly, (b) 2D layered morphology of hexagonal rosettes nanosheets, (c) CN sample, and
(d) CN-ND sample. HRTEM images of (e) CN sample and (f) CN-ND sample.

production reaction by adding the catalysts to a H2O2 solution with


100 μM initial concentration. The mixture was purged with N2 to
evacuate the dissolved oxygen and irradiated for 60 min.
■ RESULTS AND DISCUSSION
Structural Characteristics. Figure 1a shows the X-ray
Active Species Determination Experiments. To determine the diffraction (XRD) patterns of the CN and CN-ND samples. As
active species involved in photocatalytic H2O2 production in the is evident, both samples exhibit the typical structure of g-C3N4
presence of the CN-ND sample, benzoquinone (BQ, 1 mM) was with two original peaks. In detail, the low-angle diffraction
employed to trap the superoxide anion radical (•O2−). In addition, N2 peak located at about 13° (100) is attributed to the heptazine
was bubbled into the reaction media for 30 min to evacuate the units, while the (002) peak at around 27° is characterized for
dissolved oxygen. The mechanistic role of •O2− radicals was further
the graphitic materials with interlayer stacking. While the
studied by electron paramagnetic resonance (EPR) analysis, using 5,5-
dimethyl-1-pyrroline N-oxide (DMPO) as the spin-trapping reagent. (100) peak exhibits enhanced intensity, the (002) shows a less
The reaction mixture composed of 10 mL of EtOH/DW (5% v/v), 5 pronounced peak in the CN-ND sample, compared to those in
mg of the catalyst sample, and 0.1 mmol DMPO was stirred in dark the CN sample. The increased intensity of the (100) peak
and was then subjected to light irradiation for 20 min. shows the longer atomic range order within each atomic layer
4522 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 3. High-resolution XPS spectra of (a,b) N 1s and (c,d) C 1s of the samples, (e) EPR spectra of the samples, and (f) schematic illustration of
NVs in the structure.

of the sample. This phenomenon may be a result of the heat The identical Fourier transform IR spectroscopy (FT-IR)
treatment, as owing to the thermal energy, the atoms in a spectra suggest a similar chemical structure of the two samples
single atomic layer can fluctuate from their original position. (Figure 1b). The s-triazine units in the structures were
Therefore, heating for 60 min under an argon atmosphere identified by the peak at 809 cm−1. A broad peak at 1200−
could result in the formation of a more ordered atomic layer of 1600 cm−1 indicates the stretching mode of C−N heterocycles.
the respective sample. The reduced intensity of the (002) peak The stretching vibration of N−H and OH groups attributed to
can be the result of two main reasons. First, it shows the the absorbed water and residual N−H groups on the surface
exfoliation of the layers under heat treatment. Second, it
was shown by the peak at around 3000−3500 cm−1. Despite
demonstrates the formation of more stacked atomic layers in
the mentioned similarities, the overall FT-IR peak intensity
the structure. The formation of such a dense structure of the
atomic layers can also be a result of the fluctuation in every shows a decline in the CN-ND sample, compared to that in the
single sheet. A redshift to the higher angles in the (002) peak CN sample. This intensity reduction is attributed to the
can prove the formation of such stacked atomic layers in the reduced nitrogen content in the CN-ND sample. The
structure. The magnified XRD peaks of (100) and (002) for magnified FT-IR spectra in the range of 750−900 cm−1 for
the CN and CN-ND samples, presented in Figure S1 of the the breathing mode of s-triazine (Figure 1b, inset) clearly
Supporting Information, clearly show the aforementioned demonstrate the lower intensity for the triazine units in the
alterations of the structure. nitrogen-deficient sample.
4523 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 4. (a) UV−vis absorption spectra and (b) PL emission spectra of the samples.

For comparison, Figure S2 of the Supporting Information presence of pores with a cylindrical shape in the structure.58 As
also presents the XRD pattern and FT-IR spectrum of the is evident, after recalcination, the pore volume and surface area
MCN sample. Comparing the results obtained from structural of the CN-ND sample increased to 0.162 cm3 g−1 and 45.3 m2
characterizations for all of these samples suggests that by using g−1, compared to those in the CN sample of 0.121 cm3 g−1 and
the CM complex as the precursor, the π-conjugated structure 35.9 m2 g−1, respectively. This enhancement in the pore
of g-C3N4 has been completely constructed, and after the volume and BET surface area can originate from the further
recalcination of those structures, the main structure of g-C3N4 thermal exfoliation of the nanostructures because of the post-
has been preserved. heating treatment.
To investigate the formation of the CM adducts and the Elemental analysis data, shown in Table S1 of the
transformation of those adducts to the corresponding g-C3N4 Supporting Information, reveal that the C/N molar ratio
nanostructures, field emission scanning electron microscopy increased from 0.671 to 0.685 in the CN to CN-ND sample.
(FESEM) was employed. In the case of molecular self- This can be considered as evidence of nitrogen atoms
assemblies between melamine and cyanuric acid, forming the deficiency in the structure of the CN-ND sample. However,
CM adducts in three different assemblies, namely, cyclic this nitrogen deficiency did not significantly alter the structural
rosette, crinkled tape, and linear tape, has already been features of the sample, and the backbone of g-C3N4 has been
reported, and the cyclic rosette structure has been identified as preserved in the CN-ND sample, as depicted in the XRD and
the most stable one.57 Figure 2a,b presents SEM images of the FT-IR characterizations.
as-prepared CM complex. These figures clearly show the N 1s and C 1s X-ray photoelectron spectroscopy (XPS)
construction of the 2D hexagonal (cyclic rosette) sheet analyses were performed to investigate the composition and
structures, stacked into three-dimensional nanopillars, for the chemical states of carbon and nitrogen, as well as the formation
self-assemblies of melamine and cyanuric acid in water (CM of surface defects in the samples (Figure 3a−d). For both the
adducts). The SEM images of Figure 2c,d represent the studied samples, a typical signal peak at ca. 398.1 eV
formation of a helical structure for the corresponding carbon originating from two-coordinated nitrogen atoms (N2c) and a
nitride samples. A helix-like structure of several connected shoulder peak at ca. 400.7 eV related to three-coordinated
hexagonal curly nanosheets containing almost hexagonal pores nitrogen atoms (N3c) in the heptazine units were observed in
at their centers has been constructed by calcination of the CM the N 1s spectra. In addition, C 1s spectra showed a dominant
adducts under an air atmosphere in the CN sample (see Figure peak at ca. 288 eV and a weaker peak at 284.5 eV stemming
2c). Meanwhile, after recalcination of this sample under an from three-coordinated carbon (C3c) and adventitious carbon
argon atmosphere, the subcomponent of the helical structure in the form of graphitic sheets, respectively. Furthermore,
has changed from the sheet in the CN sample to the cluster in broad peaks at higher energies have been detected in all XPS
the CN-ND, as shown in Figure 2d. Figure S3 of the spectra that are assigned to the π-electron delocalization in the
Supporting Information shows the magnified FESEM images conjugated systems.59 The delocalization of π-electrons can be
of the two samples, which demonstrate the alterations in the of significant importance in improving the charge mobility of
surface morphology. High-resolution transmission electron the photocatalytic systems. To gain an in-depth understanding
microscopy (HRTEM) was also utilized for the morphological of the structural defects, the peaks for N 1s were deconvoluted
characterizations of the samples and the results were consistent to three components assigned to C−NC, N−C3, and C−
with those of FESEM (Figure 2e,f). NH2 at 398, 399, and 401 eV, respectively. The obtained data
Figure S4 of the Supporting Information shows the nitrogen extracted from the respective peak area clearly illustrate some
adsorption−desorption analysis results, which provided the alterations in the XPS spectra of the CN-ND sample as a result
data for the pore volume and surface area of the samples. The of introducing the NVs in the surface structure, compared to
formation of a mesoporous texture with H1-type hysteresis those of the CN sample. First, the ratio of N2c to N3c showed a
loop was confirmed for the two samples. This indicates the declining trend from 2.88 to 2.77 in CN to CN-ND, indicating
4524 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 5. XPS VB spectra of the (a) CN sample and (b) CN-ND sample.

that N2c are the preferred sites for the formation of NVs as a where Eu is the Urbach energy and α is the absorption
result of heating at 600 °C in an argon atmosphere. This can coefficient. Therefore, the Urbach energy can be calculated by
be explained in terms of the unsaturated coordination and the the inverse of the slope of the line obtained from plotting ln α
higher population of the N2c compared to those of the N3c versus the energy of the photon. The inset of Figure S5 of the
atoms.44 Second, the ratio of the delocalized π electrons in the Supporting Information shows the respective Urbach plots for
CN-ND is higher than that in the CN sample for both N 1s these samples. The calculated Eu values show a higher value of
and C 1s spectra. This means that the recalcination treatment 0.21 eV for the nitrogen-deficient sample, compared to that of
could enhance the delocalization of the π-electrons in the post- 0.19 eV for the CN sample. All these red-shifted energy states,
heated sample, compared to that in the untreated sample, which are consistent with the strong yellowish color of the
indicating an improvement in charge mobility. sample with NVs compared with the pale yellow-colored CN
Further investigation of the presence of NVs was performed sample (Figure S5 of the Supporting Information, inset), could
by studying the EPR spectra of the samples. Figure 3e shows positively affect the light-harvesting efficiency of this sample.
that an apparently enhanced signal at g = 2.00152 was To understand the role of CM-complex utilization on
observed in the CN-ND sample, compared to that in the CN modifying the optoelectronic properties of the catalysts, Figure
sample. This can be attributed to the presence of unpaired S6 of the Supporting Information shows the Kubelka−Munk
electrons on the carbon atoms of the conjugated structure of function versus the incident light energy, as well as the
nitrogen-deficient carbon nitride. Based on all the structural respective Urbach plot for the MCN sample. These plots
characterizations, Figure 3f presents a simplified chemical clearly show the absence of any mid-gap state energy and a
structure model for the CN-ND sample. lower Urbach energy (0.14 eV) for this sample, compared with
Optical Properties and Band Structures. UV−vis that for the CM complex-based samples. This means that
diffuse reflectance spectroscopy was employed to examine employing the preorganized CM adducts (or complex) as the
the light-harvesting and band tail structure of the samples. The precursor could positively enhance the optical and electronic
UV−vis absorption spectra in Figure 4a present an extended features of the respective catalysts. Figure 4b shows the steady-
light absorption for the CN-ND sample, compared to those for state photoluminescence (PL) spectra of the samples. The
quenched PL intensity of the CN-ND sample can be an
the CN. A closer look at these spectra illustrates the presence
indication of the suppressed recombination rate of the charge
of two distinct electron transfer states in both samples. One of
carriers in the sample with NVs, with trapping of the charge
them corresponds to π−π* electron transfer in the π-
carriers.
conjugated structure (band gap), while the second one,
The valance band (VB) energies of the samples were
which requires a lower energy than the band gap, is attributed
determined by XPS VB spectroscopy. In the XPS VB spectra,
to the mid-gap states. To calculate the band gap and the mid- the main shoulder structure of the first peak at the lowest
gap state energy of the samples, the transformed Kubelka− binding energy sides corresponds to the VB position of the
Munk function versus the incident light energy was plotted. samples; however, the presence of some minor shoulders at
Figure S5 of the Supporting Information shows that the band lower binding energies than the VB can be a sign of mid-gap
gap and mid-gap state energies decreased from 2.92 and 2.47 states near the VB. Figure 5a,b shows the VB positions of both
eV in the CN to 2.87 and 2.38 eV in the CN-ND sample, samples are located at 1.68 eV. However, the density of the
respectively. Moreover, the UV−vis absorption spectra of the mid-gap states is higher in the nitrogen-deficient sample (CN-
samples depicted an extended shoulder tail in the visible range ND), compared to that in the CN sample, which confirms the
(Urbach tail) for the CN-ND sample assigned to the electronic facilitated charge mobility of this sample.
states formed within the band gap (defect-associated Table 1 shows a summary of the data obtained from UV−vis
states).15,44 Urbach plots can be used to estimate the energy spectroscopy and XPS pertaining to the electronic band
of the band tail of these mid-gap states based on eq 1 structures of the samples. The higher density of the energy
introduced by Urbach in 195360 states along with their red-shifted reconstruction indicates
simultaneous enhancement in charge mobility and visible light
α = α0 exp(hν /Eu) (1) harvesting in the CN-ND sample, compared to those in the
4525 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Table 1. Electronic Band Structure of the Samples deficient sample (CN-ND) showed the highest production rate
of 200 μM H2O2 under the same photocatalytic conditions.
potential vs NHE (eV)
This exceptional increase in the photocatalytic performance of
energy band CN CN-ND the samples is attributed to the presence of the nitrogen-
conduction (CB) −1.24 −1.19 vacancy defects in the structure. These structural defects
Urbach energy −1.05 −0.98 increase the light-harvesting capability and suppress the charge
mid-gap state (near CB) −0.79 −0.7 recombination rate, owing to the favorable reconstruction of
mid-gap state (near VB) 1.0 1.05 the band alignments. To show the optimum level of NVs, two
mid-gap state (near VB) 0.85 other samples [CN-ND (L) and CN-ND (H)] were prepared
valance (VB) 1.68 1.68 by recalcination of the CN sample for 30 and 90 min. The
obtained results show that all nitrogen-deficient samples
exhibited better photocatalytic performance, compared with
CN sample. These enhancements are expected to improve the
the pristine CN sample (Figure S7a of the Supporting
photocatalytic performance of the obtained photocatalysts.
Photocatalytic Hydrogen Peroxide Production Per- Information). However, an optimum level of NVs is needed
formance. All the experiments in this section (except the for best performance. This can be related to the dual
stability tests) were run in triplicate at least, and the averages functionality of the defects that can simultaneously act as
are reported in the graphs. Figure 6a shows the photocatalytic trapping sites and recombination centers for the charge
production of H2O2 in the presence of the photocatalysts. This carriers. Therefore, precise optimization is required to obtain
graph shows that while the photocatalytic production of H2O2 the desired amount of the defects in the structure.50 As a result,
in the presence of the bulk MCN sample could yield just 30 in the CN-ND (L) sample with 30 min recalcination time, an
μM H2O2 in 1 h reaction time, utilizing the CN sample showed inadequate number of defects as trapping sites may cause
a significant improvement in the production rate, of performance reduction, compared to the optimized sample. On
approximately 55 μM of H2O2. This improved performance the other hand, over-introduction of the vacancies in the CN-
can be attributed to the enhanced photocatalytic activity of the ND (H) sample with 90 min recalcination time can lead to a
CN sample, due to the formation of porous nanostructures higher recombination rate and lower photocatalytic perform-
with a large surface area, as well as the improved optoelectronic ance than the optimized CN-ND sample. The electrochemical
properties of the nano sample, compared to the bulk MCN impedance spectroscopy (EIS) was employed as an efficient
sample. These nanostructures can provide more active surface tool to illustrate the charge transfer efficiency of the samples.
sites for the reaction and facilitate charge transfer by reducing Figure S7b displays the EIS Nyquist plots of all the samples.
the charge diffusion length and formation of mid-gap states The arc diameter of the Nyquist plot demonstrates the sum of
that can act as trapping sites. Moreover, recalcination of the the electrolyte solution resistance and the interfacial charge-
CN sample in an argon atmosphere at 600 °C could effectively transfer resistance between the electrode and the electrolyte.
enhance the photocatalytic activity. Employing the nitrogen- Since the electrolyte resistance is the same for all the samples,

Figure 6. (a) Photocatalytic H2O2 production using different samples under visible light of 420 < λ < 800 nm, (b) wavelength dependence of
photocatalytic H2O2 production over CN and CN-ND samples, (c) cycling experiments for H2O2 production over CN-ND sample, and (d) Kf and
Kd for H2O2 production experiments.

4526 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Figure 7. (a) Photocatalytic H2O2 production in control experiments over the CN-ND sample, (b) EPR spectra of DMPO−•O2− for the CN-ND
sample, and (c) proposed mechanism of photocatalytic H2O2 production over the CN-ND sample.

the differences in the arc diameter of the Nyquist plots indicate successive reaction runs under the same conditions. The
the different interfacial charge transfer efficiency of the catalyst was collected by centrifugation, washed with distilled
samples. The CN-ND sample shows the smallest diameter water, and utilized for the next run. Figure 6c shows the results,
suggesting its lowest resistance for interfacial charge transfer which confirm the acceptable reusability of the sample, and a
from electrode to electrolyte molecules among all the samples. slight decrease in the photocatalytic activity can be the result of
Therefore, it is deemed that this sample not only possesses the some catalyst lost in the separation and washing process. The
strongest capability in separating the photo-generated elec- structural characterizations were performed for the sample
tron−hole pairs but also in transferring them to the reactant after the cycling experiments. Figure S9 of the Supporting
molecules adsorbed on the surface active sites. This implies its Information presents the FT-IR spectra and XRD patterns of
suppressed e−/h+ recombination rate and outstanding the catalyst before and after reuse for the fourth run, and these
competence as a high-performance photocatalyst. exhibit no obvious structural disruption of the catalyst.
Further evaluation of the samples was performed for the Simultaneous decomposition of the generated H2O2 has also
photocatalytic performance of the samples in the visible region been reported as a competitive reaction in related studies.61,62
of light by employing different cutoff filters. The wavelength In this regard, photogenerated electrons and holes are
dependence of the H2O2 production rate for the samples along supposed to be responsible for the decomposition process. A
with their respective UV−vis absorption spectra is presented in kinetic model based on a zero-order kinetic for the formation
Figure 6b. The data illustrate a perfect match between the and a first-order kinetic for the decomposition reactions has
photocatalytic activity of the samples and their optical been proposed for quantifying the rate constants by fitting the
absorption spectra. Although the pristine CN sample exhibited experimental data into the following equations
poor activity at higher wavelengths and only a trace amount of [H 2O2 ] = K f /Kd(1 − exp( −Kdt )) (2)
3 μM H2O2 was produced at λ > 495 nm, the CN-ND sample
could produce an acceptable amount of 12 μM H2O2 in the Kd = −ln(Ct /C0)/t (3)
same conditions. These results confirm the extended light-
harvesting ability of the sample with a proper level of the where Kf and Kd, are the formation and decomposition rate
nitrogen vacancy. constants, and Ct and C0 are H2O2 concentration at time t and
To examine the stability of the CN-ND sample, the initial concentration (100 μmol L−1) for decomposition
photocatalytic H2O2 production was performed for four reactions, respectively. Figure S8a,b of the Supporting
4527 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

Information shows the photocatalytic decomposition of H2O2 another electron to H2O2. Figure 7c summarizes this proposed
over different photocatalysts as well as the corresponding mechanism.
pseudo-first-order kinetic fitting curves. In addition, Figure 6d
shows the calculated rate constants. According to the obtained
results, the formation rate constant of 3.68 μmol L−1 min−1 for
■ CONCLUSIONS
In summary, g-C3N4 nanostructures were easily prepared by
the CN-ND sample is 2.6 times higher than that of 1.44 μmol the thermal condensation of cyanuric acid−melamine supra-
L−1 min−1 for the CN sample. At the same time, the molecular structures. Afterward, NVs in the optimum range
decomposition rate constant (Kd) also increased, but less were successfully incorporated into the surface of the g-C3N4
significantly from 0.0029 to 0.0037 min−1 in the CN to CN- nanostructures by a facile recalcination treatment under an
ND sample, respectively. This indicates that the facilitated inert atmosphere. These nanostructures were employed for
charge mobility and increased light-harvesting capability in the photocatalytic oxygen reduction to H2O2 under visible light
CN-ND sample are favorable for both photocatalytic formation irradiation. A significant enhancement in the H2O2 production
and the decomposition of H2O2. Nevertheless, the enhance- rate was observed by utilizing nitrogen-deficient nanostructures
ment of the formation rate outweighs the increase of the compared to the bulk sample, which was prepared by
decomposition rate, leading to the overall outstanding H2O2 melamine and the pristine nanostructures with no NVs. This
production rate of this sample. improvement can be attributed to two substantial structural
An exact comparison among the results of different studies modifications. First, utilizing the CM complex as the precursor
cannot easily be performed, since different studies work under enabled curly paper-like nanostructures to be obtained with a
different experimental conditions. Nevertheless, the authors relatively high surface area and porous structure, as well as
have tried to prepare a summary of the results of some studies induced energy states within the band gap. Second, as a result
performed under nearly identical conditions to those of the of incorporating the NVs into the nanostructures, the band
experiments in the current study.21−24,63−66 Comparison of the structures were beneficially reconstructed in favor of extended
results shown in Table S2 of the Supporting Information visible light harvesting and charge mobility enhancement.
demonstrates the superior outcome of the current study Therefore, employing the CM complex as the precursor and
compared to the already reported results for photocatalytic inducing the NVs in the structure could synergistically enhance
H2O2 production over a series of g-C3N4-based photocatalysts. the performance of the as-obtained samples for the photo-
Mechanism of Improved Performance. To propose the catalytic reduction of molecular oxygen to H2O2 under visible
possible mechanism of photocatalytic H2O2 generation, control light irradiation. Improvements with respect to the robustness
experiments for H2O2 production were also conducted by and photocatalytic performance of CM complex-based carbon
using the CN-ND catalyst, and Figure 7a presents the results. nitride materials entail further scrutiny, and this study would
H2O2 was not detected in the experiments performed under pave the way for more investigations into this extensive subject.
the dark or in the absence of the photocatalysts, indicating that
the oxygen reduction to H2O2 has been driven by the
photocatalytic mechanism. In addition, H2O2 could hardly be

*
ASSOCIATED CONTENT
sı Supporting Information
produced in the absence of ethanol as an electron donor/hole The Supporting Information is available free of charge at
scavenger, and only 10 μM H2O2 was obtained in the reaction https://pubs.acs.org/doi/10.1021/acssuschemeng.0c08884.
media containing water as an electron donor. This can be
explained by considering the thermodynamic constraints for Chemicals; characterization equipment; additional XRD,
water oxidation. Holes in the VB of the CN-ND sample cannot FT-IR, and SEM characterizations; BET analysis data;
oxidize water, as the EVB of the catalyst (+1.68 eV) is much elemental analysis; band gap determination graphs;
smaller than the required potential for water oxidation (2.68 additional experimental results for the effect of different
eV). Consequently, ethanol is needed as the electron donor in recalcination times; H2O2 decomposition kinetics
the system to provide the required protons. The photocatalytic results; and comparison table (PDF)
H2O2 production was also dramatically inhibited in the
experiments performed by purging nitrogen or adding p-BQ
(1 mM) as a superoxide anion scavenger into the reaction
■ AUTHOR INFORMATION
Corresponding Author
suspension before the illumination. These results indicate the Byeong-Kyu Lee − Department of Civil and Environmental
mechanistic role of •O2− in photocatalytic H2O2 production Engineering, University of Ulsan, Ulsan 44610, Republic of
through an oxygen reduction reaction. Further evidence for the Korea; orcid.org/0000-0002-5075-6231; Phone: 82-52-
presence of •O2− as an intermediate in the photocatalytic 259-2864; Email: [email protected]
reaction was obtained by EPR analysis. Figure 7b exhibits the
characteristic peaks of DMPO−•O2− adducts after light Authors
illumination. According to the EPR, trapping, and control Hossein Fattahimoghaddam − Department of Civil and
experiment results, the following explanation can be elaborated Environmental Engineering, University of Ulsan, Ulsan
for this mechanism. Under visible light irradiation, electrons 44610, Republic of Korea
and holes are generated in the conduction and VBs of the Tahereh Mahvelati-Shamsabadi − Department of Civil and
catalyst. In the presence of ethanol as the scavenger, holes in Environmental Engineering, University of Ulsan, Ulsan
the VB are trapped, and protons (H+) are produced. On the 44610, Republic of Korea
other side, conduction band electrons or those trapped in the Complete contact information is available at:
mid-gap states, having more negative potential than that of https://pubs.acs.org/10.1021/acssuschemeng.0c08884
O2/•O2− (−0.33 eV vs NHE), can reduce the molecular
oxygen to superoxide anion radical (•O2−) by a one-electron Notes
reduction reaction. Finally, •O2− can be further reduced by The authors declare no competing financial interest.
4528 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering


pubs.acs.org/journal/ascecg Research Article

ACKNOWLEDGMENTS (18) He, Y.; Zhang, L.; Teng, B.; Fan, M. New application of Z-
scheme Ag3PO4/g-C3N4 composite in converting CO2 to fuel.
This work was supported by the National Research Environ. Sci. Technol. 2015, 49, 649−656.
Foundation of Korea (NRF) grant funded by the Korea (19) Cao, S.; Low, J.; Yu, J.; Jaroniec, M. Polymeric photocatalysts
government (MIST: Ministry of Science and ICT) (no. based on graphitic carbon nitride. Adv. Mater. 2015, 27, 2150−2176.
2019R1A2C2085250). (20) Masih, D.; Ma, Y.; Rohani, S. Graphitic C3N4 based noble-


metal-free photocatalyst systems: a review. Appl. Catal., B 2017, 206,
556−588.
REFERENCES (21) Shiraishi, Y.; Kanazawa, S.; Sugano, Y.; Tsukamoto, D.;
(1) Campos-Martin, J. M.; Blanco-Brieva, G.; Fierro, J. L. Hydrogen Sakamoto, H.; Ichikawa, S.; Hirai, T. Highly selective production of
peroxide synthesis: an outlook beyond the anthraquinone process. hydrogen peroxide on graphitic carbon nitride (g-C3N4) photo-
Angew. Chem., Int. Ed. 2006, 45, 6962−6984. catalyst activated by visible light. ACS Catal. 2014, 4, 774−780.
(2) Ai, Z.; Lu, L.; Li, J.; Zhang, L.; Qiu, J.; Wu, M. Fe@Fe2O3 (22) Shiraishi, Y.; Kofuji, Y.; Sakamoto, H.; Tanaka, S.; Ichikawa, S.;
Core−Shell Nanowires as Iron Reagent. 1. Efficient Degradation of Hirai, T. Effects of surface defects on photocatalytic H2O2
Rhodamine B by a Novel Sono-Fenton Process. J. Phys. Chem. C production by mesoporous graphitic carbon nitride under visible
2007, 111, 4087−4093. light irradiation. ACS Catal. 2015, 5, 3058−3066.
(3) Qin, Y.; Song, F.; Ai, Z.; Zhang, P.; Zhang, L. Protocatechuic (23) Kofuji, Y.; Ohkita, S.; Shiraishi, Y.; Sakamoto, H.; Tanaka, S.;
Acid Promoted Alachlor Degradation in Fe(III)/H2O2 Fenton Ichikawa, S.; Hirai, T. Graphitic carbon nitride doped with biphenyl
System. Environ. Sci. Technol. 2015, 49, 7948−7956. diimide: efficient photocatalyst for hydrogen peroxide production
(4) Hage, R.; Lienke, A. Applications of Transition-Metal Catalysts from water and molecular oxygen by sunlight. ACS Catal. 2016, 6,
to Textile and Wood-Pulp Bleaching. Angew. Chem., Int. Ed. 2006, 45, 7021−7029.
206−222. (24) Lei, J.; Chen, B.; Lv, W.; Zhou, L.; Wang, L.; Liu, Y.; Zhang, J.
(5) Hou, H.; Zeng, X.; Zhang, X., Production of Hydrogen Peroxide Robust Photocatalytic H2O2 Production over Inverse Opal g-C3N4
by Photocatalytic Processes. Angew. Chem., Int. Ed. 2020, 59, DOI: with Carbon Vacancy under Visible Light. ACS Sustainable Chem. Eng.
DOI: 10.1002/anie.201911609 2019, 7, 16467−16473.
(6) An, L.; Zhao, T.; Yan, X.; Zhou, X.; Tan, P. The dual role of (25) Ong, W.-J.; Tan, L.-L.; Ng, Y. H.; Yong, S.-T.; Chai, S.-P.
hydrogen peroxide in fuel cells. Sci. Bull. 2015, 60, 55−64. Graphitic carbon nitride (g-C3N4)-based photocatalysts for artificial
(7) Choudhary, V. R.; Gaikwad, A. G.; Sansare, S. D. Nonhazardous photosynthesis and environmental remediation: are we a step closer
direct oxidation of hydrogen to hydrogen peroxide using a novel to achieving sustainability? Chem. Rev. 2016, 116, 7159−7329.
membrane catalyst. Angew. Chem. 2001, 113, 1826−1829. (26) Qi, Y.; Liu, L.; Liang, Y.; Hu, J.; Cui, W. Graphitic carbon
(8) Mamba, G.; Mishra, A. K. Graphitic carbon nitride (g-C 3 N 4 ) nitride compound photocatalyst. Prog. Chem. 2015, 27, 38−46.
nanocomposites: A new and exciting generation of visible light driven (27) Niu, P.; Zhang, L.; Liu, G.; Cheng, H.-M. Graphene-Like
photocatalysts for environmental pollution remediation. Appl. Catal., Carbon Nitride Nanosheets for Improved Photocatalytic Activities.
B 2016, 198, 347−377. Adv. Funct. Mater. 2012, 22, 4763−4770.
(9) Ye, S.; Wang, R.; Wu, M.-Z.; Yuan, Y.-P. A review on g-C3N4 for (28) Zhang, J.; Guo, F.; Wang, X. An optimized and general
photocatalytic water splitting and CO2 reduction. Appl. Surf. Sci. synthetic strategy for fabrication of polymeric carbon nitride
2015, 358, 15−27. nanoarchitectures. Adv. Funct. Mater. 2013, 23, 3008−3014.
(10) Zheng, Y.; Lin, L.; Wang, B.; Wang, X. Graphitic carbon nitride (29) Gao, Q.; Hu, S.; Du, Y.; Hu, Z. The origin of the enhanced
polymers toward sustainable photoredox catalysis. Angew. Chem., Int. photocatalytic activity of carbon nitride nanotubes: a first-principles
Ed. 2015, 54, 12868−12884. study. J. Mater. Chem. A 2017, 5, 4827−4834.
(11) Safaei-Ghomi, J.; Akbarzadeh, Z.; Teymuri, R. ZnS nano- (30) Li, X.-H.; Zhang, J.; Chen, X.; Fischer, A.; Thomas, A.;
particles immobilized on graphitic carbon nitride as a recyclable and Antonietti, M.; Wang, X. Condensed graphitic carbon nitride
environmentally friendly catalyst for synthesis of 3-cinnamoyl nanorods by nanoconfinement: promotion of crystallinity on
coumarins. Res. Chem. Intermed. 2019, 45, 3425−3439. photocatalytic conversion. Chem. Mater. 2011, 23, 4344−4348.
(12) Zhou, C.; Lai, C.; Huang, D.; Zeng, G.; Zhang, C.; Cheng, M.; (31) Qin, Z.; Huang, Z.; Wang, M.; Liu, D.; Chen, Y.; Guo, L.
Hu, L.; Wan, J.; Xiong, W.; Wen, M.; Wen, X.; Qin, L. Highly porous Synergistic effect of quantum confinement and site-selective doping in
carbon nitride by supramolecular preassembly of monomers for polymeric carbon nitride towards overall water splitting. Appl. Catal.,
photocatalytic removal of sulfamethazine under visible light driven. B 2020, 261, 118211.
Appl. Catal., B 2018, 220, 202−210. (32) Mahvelati-Shamsabadi, T.; Lee, B.-K. Photocatalytic H2
(13) Zhou, L.; Liu, Z.; Guan, Z.; Tian, B.; Wang, L.; Zhou, Y.; Zhou, evolution and CO2 reduction over phosphorus-doped g-C3N4
Y.; Lei, J.; Zhang, J.; Liu, Y. 0D/2D plasmonic Cu2-xS/g-C3N4 nanostructures: Electronic, Optical, and Surface properties. Renewable
nanosheets harnessing UV-vis-NIR broad spectrum for photocatalytic Sustainable Energy Rev. 2020, 130, 109957.
degradation of antibiotic pollutant. Appl. Catal., B 2020, 263, 118326. (33) Bui, T. S.; Bansal, P.; Lee, B.-K.; Mahvelati-Shamsabadi, T.;
(14) Dong, H.; Guo, X.; Yin, Y. A facile synthesis of goethite- Soltani, T. Facile fabrication of novel Ba-doped g-C3N4 photocatalyst
modified g-C3N4 composite for photocatalytic degradation of tylosin with remarkably enhanced photocatalytic activity towards tetracycline
in an aqueous solution. Res. Chem. Intermed. 2018, 44, 3151−3167. elimination under visible-light irradiation. Appl. Surf. Sci. 2020, 506,
(15) Ran, J.; Ma, T. Y.; Gao, G.; Du, X.-W.; Qiao, S. Z. Porous P- 144184.
doped graphitic carbon nitride nanosheets for synergistically (34) Moon, G.-h.; Fujitsuka, M.; Kim, S.; Majima, T.; Wang, X.;
enhanced visible-light photocatalytic H2 production. Energy Environ. Choi, W. Eco-friendly photochemical production of H2O2 through
Sci. 2015, 8, 3708−3717. O2 reduction over carbon nitride frameworks incorporated with
(16) Wang, X.; Maeda, K.; Thomas, A.; Takanabe, K.; Xin, G.; multiple heteroelements. ACS Catal. 2017, 7, 2886−2895.
Carlsson, J. M.; Domen, K.; Antonietti, M. A metal-free polymeric (35) Jourshabani, M.; Lee, B.-K.; Shariatinia, Z. From Traditional
photocatalyst for hydrogen production from water under visible light. Strategies to Z-scheme Configuration in Graphitic Carbon Nitride
Nat. Mater. 2009, 8, 76−80. Photocatalysts: Recent Progress and Future Challenges. Appl. Catal.,
(17) Mahvelati-Shamsabadi, T.; Lee, B.-K. Design of Ag/g-C3N4 on B 2020, 276, 119157.
TiO2 nanotree arrays via ultrasonic-assisted spin coating as an (36) Mahvelati-Shamsabadi, T.; Goharshadi, E. K.; Karimi-
efficient photoanode for solar water oxidation: Morphology Nazarabad, M. Z-scheme design of Ag@g-C3N4/ZnS photoanode
modification and junction improvement. Catal. Today 2020, 358, device for efficient solar water oxidation: An organic-inorganic
412−421. electronic interface. Int. J. Hydrogen Energy 2019, 44, 13085−13097.

4529 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530
ACS Sustainable Chemistry & Engineering pubs.acs.org/journal/ascecg Research Article

(37) Lei, J.; Chen, B.; Zhou, L.; Ding, N.; Cai, Z.; Wang, L.; In, S.-I.; gates to Carbon Nitride Hollow Spheres. Adv. Funct. Mater. 2013, 23,
Cui, C.; Zhou, Y.; Liu, Y. Efficient Degradation of Antibiotics in 3661−3667.
Different Water Matrices through the Photocatalysis of Inverse Opal (55) Shalom, M.; Inal, S.; Fettkenhauer, C.; Neher, D.; Antonietti,
K-g-C3N4: Insights into Mechanism and Assessment of Antibacterial M. Improving carbon nitride photocatalysis by supramolecular
Activity. Chem. Eng. J. 2020, 400, 125902. preorganization of monomers. J. Am. Chem. Soc. 2013, 135, 7118−
(38) Sun, L.; Yang, M.; Huang, J.; Yu, D.; Hong, W.; Chen, X. 7121.
Freestanding graphitic carbon nitride photonic crystals for enhanced (56) Fattahimoghaddam, H.; Mahvelati-Shamsabadi, T.; Lee, B.-K.
photocatalysis. Adv. Funct. Mater. 2016, 26, 4943−4950. Efficient Photodegradation of Rhodamine B and Tetracycline over
(39) Tian, Y.; Zhou, L.; Zhu, Q.; Lei, J.; Wang, L.; Zhang, J.; Liu, Y. Robust and Green g-C3N4 Nanostructures: Supramolecular Design. J.
Hierarchical macro-mesoporous g-C3N4 with an inverse opal Hazard. Mater. 2021, 403, 123703.
structure and vacancies for high-efficiency solar energy conversion (57) Ji, H.-F.; Xu, X. Hexagonal Organic Nanopillar Array from the
and environmental remediation. Nanoscale 2019, 11, 20638−20647. Melamine−Cyanuric Acid Complex. Langmuir 2010, 26, 4620−4622.
(40) Kang, Y.; Yang, Y.; Yin, L.-C.; Kang, X.; Liu, G.; Cheng, H.-M. (58) ALOthman, Z. A review: fundamental aspects of silicate
An Amorphous Carbon Nitride Photocatalyst with Greatly Extended mesoporous materials. Materials 2012, 5, 2874−2902.
Visible-Light-Responsive Range for Photocatalytic Hydrogen Gen- (59) Mane, G. P.; Talapaneni, S. N.; Lakhi, K. S.; Ilbeygi, H.; Ravon,
eration. Adv. Mater. 2015, 27, 4572−4577. U.; Al-Bahily, K.; Mori, T.; Park, D.-H.; Vinu, A. Highly Ordered
(41) Kang, Y.; Yang, Y.; Yin, L.-C.; Kang, X.; Wang, L.; Liu, G.; Nitrogen-Rich Mesoporous Carbon Nitrides and Their Superior
Cheng, H.-M. Selective breaking of hydrogen bonds of layered carbon Performance for Sensing and Photocatalytic Hydrogen Generation.
nitride for visible light photocatalysis. Adv. Mater. 2016, 28, 6471− Angew. Chem., Int. Ed. 2017, 56, 8481−8485.
6477. (60) Urbach, F. The long-wavelength edge of photographic
(42) Li, S.; Dong, G.; Hailili, R.; Yang, L.; Li, Y.; Wang, F.; Zeng, Y.; sensitivity and of the electronic absorption of solids. Phys. Rev.
Wang, C. Effective photocatalytic H2O2 production under visible 1953, 92, 1324.
light irradiation at g-C3N4 modulated by carbon vacancies. Appl. (61) Zhou, L.; Feng, J.; Qiu, B.; Zhou, Y.; Lei, J.; Xing, M.; Wang, L.;
Catal., B 2016, 190, 26−35. Zhou, Y.; Liu, Y.; Zhang, J. Ultrathin g-C3N4 nanosheet with
(43) Liu, G.; Zhao, G.; Zhou, W.; Liu, Y.; Pang, H.; Zhang, H.; Hao, hierarchical pores and desirable energy band for highly efficient H2O2
D.; Meng, X.; Li, P.; Kako, T.; Ye, J. In Situ Bond Modulation of production. Appl. Catal., B 2020, 267, 118396.
Graphitic Carbon Nitride to Construct p-n Homojunctions for (62) Lin, S.; Zhang, N.; Wang, F.; Lei, J.; Zhou, L.; Liu, Y.; Zhang, J.
Enhanced Photocatalytic Hydrogen Production. Adv. Funct. Mater. Carbon Vacancy Mediated Incorporation of Ti3C2 Quantum Dots in
2016, 26, 6822−6829. a 3D Inverse Opal g-C3N4 Schottky Junction Catalyst for
(44) Niu, P.; Liu, G.; Cheng, H.-M. Nitrogen vacancy-promoted Photocatalytic H2O2 Production. ACS Sustainable Chem. Eng. 2020,
photocatalytic activity of graphitic carbon nitride. J. Phys. Chem. C 9, 481−488.
2012, 116, 11013−11018. (63) Kofuji, Y.; Isobe, Y.; Shiraishi, Y.; Sakamoto, H.; Tanaka, S.;
(45) Chen, J.; Mao, Z.; Zhang, L.; Wang, D.; Xu, R.; Bie, L.; Ichikawa, S.; Hirai, T. Carbon Nitride-Aromatic Diimide-Graphene
Fahlman, B. D. Nitrogen-deficient graphitic carbon nitride with Nanohybrids: Metal-Free Photocatalysts for Solar-to-Hydrogen
enhanced performance for lithium ion battery anodes. ACS Nano Peroxide Energy Conversion with 0.2% Efficiency. J. Am. Chem. Soc.
2017, 11, 12650−12657. 2016, 138, 10019−10025.
(46) Lv, C.; Qian, Y.; Yan, C.; Ding, Y.; Liu, Y.; Chen, G.; Yu, G. (64) Yang, Y.; Zeng, Z.; Zeng, G.; Huang, D.; Xiao, R.; Zhang, C.;
Defect Engineering Metal-Free Polymeric Carbon Nitride Electro- Zhou, C.; Xiong, W.; Wang, W.; Cheng, M.; Xue, W.; Guo, H.; Tang,
catalyst for Effective Nitrogen Fixation under Ambient Conditions. X.; He, D. Ti3C2 Mxene/porous g-C3N4 interfacial Schottky
Angew. Chem. 2018, 130, 10403−10407. junction for boosting spatial charge separation in photocatalytic
(47) Mo, R.; Li, J.; Tang, Y.; Li, H.; Zhong, J. Introduction of H2O2 production. Appl. Catal., B 2019, 258, 117956.
nitrogen defects into a graphitic carbon nitride framework by (65) Yang, Y.; Zeng, G.; Huang, D.; Zhang, C.; He, D.; Zhou, C.;
selenium vapor treatment for enhanced photocatalytic hydrogen Wang, W.; Xiong, W.; Li, X.; Li, B.; Dong, W.; Zhou, Y. Molecular
production. Appl. Surf. Sci. 2019, 476, 552−559. engineering of polymeric carbon nitride for highly efficient photo-
(48) Xue, J.; Fujitsuka, M.; Majima, T. The role of nitrogen defects catalytic oxytetracycline degradation and H2O2 production. Appl.
in graphitic carbon nitride for visible-light-driven hydrogen evolution. Catal., B 2020, 272, 118970.
Phys. Chem. Chem. Phys. 2019, 21, 2318−2324. (66) Shi, L.; Yang, L.; Zhou, W.; Liu, Y.; Yin, L.; Hai, X.; Song, H.;
(49) Zhang, S.; Hu, C.; Ji, H.; Zhang, L.; Li, F. Facile synthesis of Ye, J. Photoassisted Construction of Holey Defective g-C3 N4
nitrogen-deficient mesoporous graphitic carbon nitride for highly Photocatalysts for Efficient Visible-Light-Driven H2 O2 Production.
efficient photocatalytic performance. Appl. Surf. Sci. 2019, 478, 304− Small 2018, 14, 1703142.
312.
(50) Tu, W.; Xu, Y.; Wang, J.; Zhang, B.; Zhou, T.; Yin, S.; Wu, S.;
Li, C.; Huang, Y.; Zhou, Y.; Zou, Z.; Robertson, J.; Kraft, M.; Xu, R.
Investigating the role of tunable nitrogen vacancies in graphitic carbon
nitride nanosheets for efficient visible-light-driven H2 evolution and
CO2 reduction. ACS Sustainable Chem. Eng. 2017, 5, 7260−7268.
(51) Zhang, Y.; Wen, R.; Guo, D.; Guo, H.; Chen, J.; Zheng, Z. One-
step facile fabrication and photocatalytic activities of ZnS@g-
C3N4nanocomposites from sulfatotris(thiourea)zinc(II) complex.
Appl. Organomet. Chem. 2016, 30, 160−166.
(52) Barrio, J.; Shalom, M. Rational design of carbon nitride
materials by supramolecular preorganization of monomers. Chem-
CatChem 2018, 10, 5573−5586.
(53) Feng, L.-L.; Zou, Y.; Li, C.; Gao, S.; Zhou, L.-J.; Sun, Q.; Fan,
M.; Wang, H.; Wang, D.; Li, G.-D.; Zou, X. Nanoporous sulfur-doped
graphitic carbon nitride microrods: A durable catalyst for visible-light-
driven H 2 evolution. Int. J. Hydrogen Energy 2014, 39, 15373−15379.
(54) Jun, Y.-S.; Lee, E. Z.; Wang, X.; Hong, W. H.; Stucky, G. D.;
Thomas, A. From Melamine-Cyanuric Acid Supramolecular Aggre-

4530 https://dx.doi.org/10.1021/acssuschemeng.0c08884
ACS Sustainable Chem. Eng. 2021, 9, 4520−4530

You might also like