Photocatalytic degradation of phenol wastewater over Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst under solar light irradiation

Article abstract of DOI:10.1016/j.molliq.2018.10.160

A series of carbon nanotubes (CNT) modified g-C₃N₄/BiVO₄ photocatalysts were synthesized via wet-impregnation method and evaluated via degradation of phenol under solar light irradiation. The physicochemical properties of the as-developed photocatalyst were characterized using FTIR, XRD, FESEM, XPS, SAP and DR-UV Vis. The formation of g-C₃N₄/CNT/BiVO₄ photocatalysts resulted in remarkable enhancement in the performance in which almost six times higher degradation rate in comparison to the pristine g-C₃N₄ and obeyed the pseudo-first-order kinetics and Temkin adsorption model. Congruously, the synergistic interaction between 2 wt% of CNT and 5 vol% of H₂O₂ as an oxidizing agent was capable of removing 80.6% of phenol within 120 min. The profound photodegradation performance monitored was attributed to the better crystallinity structure obtained as shown in XRD and XPS analysis. Furthermore, the intimate contact between the CNT, g-C₃N₄ and BiVO₄ in the heterostructure sample as shown in FESEM micrograph images does help in allowing a smooth electron-hole pair separation and migration, resulting in more available ?OH and ?O₂ ^ˉ radicals for photocatalytic degradation activities. The possible Z-scheme reaction mechanism has been proposed and active species trapping experiments have been carried out to find the role of active radical species responsible for the phenol degradation. Additionally, the g-C₃N₄/CNT/BiVO₄ photocatalysts retained excellent stability even after several cycles. Congruently, a mathematical representation for understanding the interaction between CNT loading and H₂O₂ for photodegradation of phenol using response surface methodology (RSM) was successfully generated.

Full text of DOI:10.1016/j.molliq.2018.10.160

Accepted Manuscript
Photocatalytic degradation of phenol wastewater over Z-scheme
g-C3N4/CNT/BiVO4 heterostructure photocatalyst under solar
light irradiation
Mohamad Fakhrul Ridhwan Samsudin, Nurfatien Bacho, Suriati
Sufian, Yun Hau Ng
PII:
S0167-7322(18)33295-1
DOI:
Reference:
https://doi.org/10.1016/j.molliq.2018.10.160
MOLLIQ 9905
To appear in:
Journal of Molecular Liquids
Received date:
Revised date:
Accepted date:
28 June 2018
29 October 2018
31 October 2018
Please cite this article as: Mohamad Fakhrul Ridhwan Samsudin, Nurfatien Bacho, Suriati
Sufian, Yun Hau Ng , Photocatalytic degradation of phenol wastewater over Z-scheme
g-C3N4/CNT/BiVO4 heterostructure photocatalyst under solar light irradiation. Molliq
(2018), https://doi.org/10.1016/j.molliq.2018.10.160
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
Photocatalytic Degradation of Phenol Wastewater over Z-Scheme g-
C3N4/CNT/BiVO4 Heterostructure Photocatalyst under Solar Light
Irradiation
Mohamad Fakhrul Ridhwan Samsudina, Nurfatien Bachoa, Suriati Sufiana, * Yun Hau Ngb
a Chemical Engineering Department, Universiti Teknologi PETRONAS, 32610, Bandar Seri
Iskandar, Perak, Malaysia.
b School of Energy and Environment, City University of Hong Kong, Kowloon, Hong Kong
SAR, P. R. China.
Corresponding authors: suriati@utp.edu.my (+605-3687587)

ACCEPTED MANUSCRIPT
Abstract
A series of carbon nanotubes (CNT) modified g-C3N4/BiVO4 photocatalysts were synthesized via
wet-impregnation method and evaluated via degradation of phenol under solar light irradiation.
The physicochemical properties of the as-developed photocatalyst were characterized using FTIR,
XRD, FESEM, XPS, SAP and DR-UV Vis. The formation of g-C3N4/CNT/BiVO4 photocatalysts
resulted in remarkable enhancement in the performance in which almost six times higher
degradation rate in comparison to the pristine g-C3N4 and obeyed the pseudo-first-order kinetics
and Temkin adsorption model. Congruously, the synergistic interaction between 2 wt. % of CNT
and 5 vol. % of H2O2 as an oxidizing agent was capable of removing 80.6 % of phenol within 120
minutes. The profound photodegradation performance monitored was attributed to the better
crystallinity structure obtained as shown in XRD and XPS analysis. Furthermore, the intimate
contact between the CNT, g-C3N4 and BiVO4 in the heterostructure sample as shown in FESEM
micrograph images does help in allowing a smooth electron-hole pair separation and migration,
¯
resulting in more available •OH and •O2 radicals for photocatalytic degradation activities. The
possible Z-scheme reaction mechanism has been proposed and active species trapping experiments
have been carried out to find the role of active radical species responsible for the phenol
degradation. Additionally, the g-C3N4/CNT/BiVO4 photocatalysts retained excellent stability even
after several cycles. Congruently, a mathematical representation for understanding the interaction
between CNT loading and H2O2 for photodegradation of phenol using response surface
methodology (RSM) was successfully generated.
Keyword: graphitic carbon nitride, bismuth vanadate, carbon nanotube, photocatalyst, phenol
degradation, wastewater.
2

ACCEPTED MANUSCRIPT
1 Introduction
The exponential growth of the pharmaceuticals industries over the past decade had exerted
substantial pressure on sustainability, mainly on the pillars of the environment. While the gap
towards an improved well-being brings many intended benefits, the environmental sustainability
challenge is also growing on a large scale and complexity, resulting in severe environmental
impacts especially in water pollutions crisis. Moreover, the presence of the phenolic compounds
in the discharged wastewater from pharmaceuticals industries has been identified as one of the
alarming pollutants due to its toxicity, hazardous and carcinogenic properties [1]. Due to its
toxicity and hazardous characteristic, the discharge of phenolic compounds into the aquatic
environment poses a serious threat not only towards the water resources safety but also to human
beings and aquatic life [2]. The ingestion of phenols even at low concentration could cause
impaired vision, eyes, mouth and skin irritation, diarrhea [3, 4].
The recent progress on the current wastewater treatment technologies for the removal of
phenolic compounds includes conventional treatments such as distillation, precipitation, chemical
oxidation and solvent extraction [5]. However, the conventional technique has multiple drawbacks
such as low removal efficiency and high operational costs which are unfavourable for large-scale
application [6]. To alleviate the phenolic wastewater pollution issue, engineering photocatalytic
system for pollutants degradation via semiconductor based photocatalyst had elicited research
fascination as a promising technology to counter the current limitations of the conventional
wastewater treatment. The photocatalytic degradation of pollutants via photocatalyst system is
favourable over the conventional method due to its peculiar features including green, cheap and
simple processes. Moreover, this technique does not require non-renewable energy consumption
as it exploits the sustainable solar energy. Applicable both for gaseous and aqueous treatment,
3

ACCEPTED MANUSCRIPT
photocatalysis technology reportedly can degrade a wide range of pollutants and toxic compounds
without causing any secondary pollutants [7].
Graphitic carbon nitride (g-C3N4) is a polymer semiconductor with a moderate band gap
energy of 2.7 eV has elicited significant interest as the next generation of the photocatalyst in
engineering photocatalytic field for environmental pollutants degradation [5]. This is due to its
excellent physiochemical properties such as high thermal stability, photochemical stability and
appealing electronic structure which favours the photocatalytic degradation of recalcitrant
pollutants [8]. Congruently, the polymeric nature of this conjugated materials allows for facile
modification of the photocatalyst to improve its optical properties besides permit multiple
excitations from absorption of a single photon, both of which are favourable for efficient pollutants
degradation [9]. Nevertheless, the practical application of g-C3N4 photocatalyst is greatly hindered
by some of its individual properties such as high band gap energy, the high recombination rate of
photogenerated electron-hole pairs and slow electron transfer which lead to lower photocatalytic
performance [7, 10 - 12].
Hence, various strategies have been adopted such as energy band engineering, doping with
non-metallic elements and development of heterostructure system in order to overcome the
individual drawbacks of pristine g-C3N4 [5, 13]. Recently, bismuth-based photocatalyst especially
monoclinic bismuth vanadate (BiVO4) shows promises as potential frontier semiconductor
photocatalyst owing to its suitable band gap energy (around 2.4 eV), great chemical stability and
environmental friendliness [14]. Nonetheless, due to the inherent defect of a single component, the
photocatalytic performance of pristine BiVO4 is substandard owing to its sluggish photocharge
carrier separation and migration and slow water oxidation kinetics [15]. Strategies such as the
formation of the Z-scheme heterostructure system and the surface morphology adjustments have
4

ACCEPTED MANUSCRIPT
been demonstrated are great promise in enhancing the overall photocatalytic degradation
performance. With respect to the appropriate band structure location of g-C3N4 and BiVO4
photocatalyst, the photoexcited electrons in the conduction band of the BiVO4 photocatalyst can
easily migrate to the adjacent photogenerated holes in the valence band of g-C3N4, results in the
formation of Z-scheme mechanism which will inhibit the fast recombination rate of electron-hole
pairs.
Moreover, to allow smooth electron-hole pairs separation and migration, the addition of
carbon materials such as graphene, carbon nanotubes (CNTs) and carbon quantum dots have
received much highlights due to their superior optical and electrical property which not only
ameliorates the photocatalyst specific surface area but also facilitate the migration of the electron
transfer between the adjacent of the coupled photocatalysts in the heterostructure system [5, 11].
With regard to aforementioned advantages, the addition of CNTs as an electron bridge mediator
will greatly enhance the overall photocatalytic degradation performance of Z-scheme g-
C3N4/BiVO4 photocatalyst owing to its unique properties such as their large surface areas,
diameter, p–p electrostatic interactions, and the shorter equilibrium time than other materials [15].
Gratifyingly, the one-dimensional structure of the CNTs provides superior electron transport [11].
Congruously, the addition of CNTs can reduce the photocharge carrier recombination rate of
semiconductor photocatalysts, which is favourable for photocatalytic activity.
Herein, a novel Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst is developed
and the performance of the as-developed Z-scheme photocatalyst is evaluated via phenol removal
under simulated solar light irradiation. The addition of CNT in the Z-scheme g-C3N4/CNT/BiVO4
heterostructure photocatalyst will greatly improve the overall surface area and inhibit the fast
recombination rate of electron-hole pairs. In addition, the synergistic interaction between CNTs
5

ACCEPTED MANUSCRIPT
and H2O2 as an oxidizing agent for photodegradation of phenol is analysed via response surface
methodology (RSM). Moreover, the kinetic and adsorption isotherm of the degradation behaviours
of the as-developed the Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst have been
thoroughly investigated along with the possible photocatalytic degradation reaction mechanism.
To the best of our knowledge, this is the first report to discuss in detail the beneficial effect of
CNTs as an electron mediator in the Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst
along with its kinetic, adsorption isotherm and the possible reaction mechanism in treating the
phenol.
2 Materials and Methods
Materials
All reagents and solvents used in the fabrication of photocatalyst and photocatalytic
degradation activity are analytical grade and used as received without any further purification.
Urea (purity of 99%), nitric acid were purchased from Merck KgaA while Phenol, Bi2O3 (purity
of 99.9%), V2O5 (purity of 99.0%) were purchased from Sigma-Aldrich, respectively. CNT
(OD=20 nm) were purchased from carbon Nano-Material. All of the solutions are prepared by
using deionized water.
Experimental Design
The experimental design matrix was developed by using Central Composite Design (CCD)
type methodology, fitted with a quadratic design model as suggested by the software, by using
Design-Expert software Version 8.0.7.1 (Stat-Ease, Inc.). This design tools provide adequate
6

ACCEPTED MANUSCRIPT
information using least but sufficient number of experimental run, making them the most widely
exploited approach in RSM [16]. Although there are various reaction factors that can affect the
photocatalytic degradation of phenol in this study, CNT dosage and total vol. % of H2O2 are
investigated towards the response function of phenol removal from aqueous solution (%). The
developed CCD contains a total of 13 experimental runs; nine (9) experiments with four (4)
replicate runs at the centre point which intended to estimate the experimental error of the studied
system.
Synthesis of pristine g-C3N4 photocatalyst
Graphitic-carbon nitride (g-C3N4) was synthesized from urea by a facile template-free
method (CAS number 57-13-6, manufactured by Merck) [17]. In a typical synthesis, a respective
amount of urea powder was weighed and put in an alumina crucible, placed in a furnace and heated
for 120 minutes at 500 °C with 5 °C/min of heating rate. Once the reaction is complete, the furnace
was allowed to cool down to the room temperature. The resultant light green-yellowish powder
was collected for use without any further treatment.
Synthesis of the pristine BiVO4 photocatalyst
The monoclinic BiVO4 photocatalyst was synthesized via a solid-liquid state reaction [13].
The photocatalyst was prepared in the presence of Vanadium Oxide,V2O5 (≥99 assay, CAS number
1314-62-1, manufactured by Merck) and Bismuth (II) Oxide,Bi2O3 (99.9% trace metal basis,10
µm, CAS number 1304-76-3, manufactured by Sigma-Aldrich) dissolved in aqueous nitric acid,
(HNO3) solution. The suspension was stirred at room temperature. The vivid yellow product
solution is then collected using suction filtration membrane and washed with deionized water
7

ACCEPTED MANUSCRIPT
before left overnight and dried in a drying oven at the temperature of 110 °C. Following the
formation of yellow powder, the produced BiVO4 photocatalyst was grounded into a fine powder
by using a mortar and pestle until no lumps are observed. The final product exists in a solid yellow
powder.
Synthesis of the Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst
The g-C3N4/CNT/BiVO4 heterostructure photocatalyst was synthesized via the wet
impregnation method [18]. 0.5 g of g-C3N4, 20 wt. % of BiVO4 and an appropriate amount of
carbon nanotubes, CNT (OD: 20 nm, length: 5 µm, manufactured by Carbon Nano-material
Technology Co., Ltd.) as formulated by the CCD in Section 2.2 were weighed and prepared. All
reagent were mixed and added to a 30 ml deionized water. The suspension was put under the
magnetic stirring condition and stirred continuously for 60 minutes. Afterward, the solution is
heated until it becomes slurry while simultaneously stirred. The precursor which was in the form
of the thick green paste was left overnight and dried in a drying oven at the temperature of 80 °C.
The obtained product exists in a small lump, grounded into a fine powder until no lumps are
observed. Similar procedures were employed for the synthesis of g-C3N4/BiVO4 and g-C3N4/CNT
photocatalyst.
Physicochemical Characterization
The physicochemical properties of the as-prepared heterostructure photocatalysts were
investigated via several characterization techniques. XRD analysis was carried out to analyse the
crystalline structure of the developed photocatalyst materials by X-ray diffractometer (XRD;
8

ACCEPTED MANUSCRIPT
Model: X’Pert3 powder and Empyrean, PANalytical) excited by Cu Kα radiation from 20 to 80°
at a step size of 0.017°. The morphological features of the as developed photocatalyst were
examined via Zeiss Supra 55VP Field-Emission Scanning Electron Microscopy (FESEM) with a
magnification of 10 kx. The specific functional group identification and types of chemicals bonds
in the developed photocatalyst were examined via Fourier transform infrared spectroscopy (FTIR)
analysis by Shimadzu 8400S Spectrophotometer from 4000 to 400 cm-1 infrared spectra.
The chemical state and elemental composition of the as developed photocatalyst were analysed
by using X-ray Photoelectron Spectroscopy (XPS) (Thermo scientific K-alpha with standard
calibration of copper, silver and gold). All the signals (N 1s, C 1S, Bi 4f, V 2p and O 1s) were
fitted using a Gaussian-Lorentzian curve and Non-linear least square fitting (NLLSF)
photoelectron peak analysis. The energy threshold structure of the as-developed photocatalyst was
examined via the DR-UV-Vis spectrometer (Cary 100s). Afterward, the Kubelka-Munk method
was used to plot the graph of absorbance against wavelength. The band gap energy of all samples
was calculated using the following equation.
(αhn)=A (hn - Eg) 1/2
(1)
Where α, hn, Eg and A denote absorption coefficient, incident photon energy, direct band gap
energy and a constant, respectively. The surface area and porosity (SAP) analysis (Micromeritic
ASAP 2000) were investigated to examine the specific surface area, pore volume and pore size of
the as-developed photocatalyst. The analysis was carried out through nitrogen adsorption-
desorption isotherm at 110°C by employing the Brunauer-Emmett-Teller (BET).
9

ACCEPTED MANUSCRIPT
Photocatalytic Degradation Setup
The photocatalytic degradation experiments were carried out for a reaction mixture of 100
mL with the phenol concentration kept at 10 mg/L. The specific volume of hydrogen peroxide,
H2O2 is added to the solution as formulated by CCD experimental design matrix. The mixture was
first sonicated for 30 minutes to ensure uniform photocatalyst distribution throughout the solution.
Prior to illumination, the mixture was stirred in the dark for 60 minutes to obtain adsorption-
desorption equilibrium between the pollutants and photocatalyst [19]. The solution was stirred and
irradiated under a simulated sunlight with an illumination intensity of 500 W halogen lamp (as
shown in Figure S1). The degradation rate of phenol was monitored by using a UV—1800 UV-
Vis spectrophotometer (Shimadzu) at the pre-determined wavelength at 269.60 nm obtained from
the calibration curve model. The photodegradation efficiency of phenol was calculated as the
difference between the final and initial phenol concentration in the aqueous solution and obtained
as follows [13]:
Rate of Phenol Removal (%) = (C - Cf) x 100
(2)
o
C
o
Where Co and Cf are the initial and final concentration of phenol extracted at the respected
time, respectively. Comparison studies were also performed using pristine g-C3N4, g-C3N4/BiVO4
and g-C3N4/CNT photocatalyst under the same experimental condition. For the kinetics and
adsorption isotherm study, the similar procedures were employed as described previously.
However, instead of measuring only the initial and final phenol concentration, the absorbance and
concentration were measured at 15 minutes time interval for a total of 150 minutes reaction (i.e.
30 minutes dark reaction and 120 minutes light reaction). The obtained experimental data then
analysed using the respective model.
10

ACCEPTED MANUSCRIPT
3 Results and Discussion
Chemical Stability Analysis
Figure 1 illustrates the FTIR spectra analysis of all as-developed photocatalyst in order to
investigate the presence of the specific functional group. The FTIR spectra for pristine g-C3N4
depicts the sharp peak located at 809 cm-1 is assigned to the characteristic breathing vibration mode
of the s-triazine ring [20] and out of plane bending modes of the heterocyclic C-N [21,22]. The
presence of four distinctive peaks at 1249, 1324, 1418, and 1572 cm-1 are ascribed to the typical
stretching of C-N aromatic skeletal [23]. In addition, the presence of spectra at 1635 cm-1
corresponded to the C=N stretching vibration modes, which presumably provide an improvement
in the organic pollutants adsorption due to its π structure characteristic. The wide absorption
spectra at about 3000-3400 cm-1 could be assigned to the stretching vibration mode of N-H and O-
H [11, 20]. Meanwhile, the presence of small band peak at 1079 cm-1 and another additional peak
at 3436 cm-1 detected for pristine CNT was due to the detection of O-H absorption bands and C-O
stretching modes, respectively. For the binary hybrid, in addition to the g-C3N4 FTIR footprint
spectra, the photocatalyst was also composed of a sharp absorption spectrum within 733-750 cm-1
which correspond to the asymmetric stretching vibration modes of VO43- and Bi-V peaks [13].
Furthermore, the FTIR spectrum of the Z-scheme heterostructure photocatalyst shows all the
characteristic peak of g-C3N4, BiVO4 and CNT, thus signify the successful loading of CNT and
subsequently confirm the formation of Z-scheme g-C3N4/CNT/BiVO4 heterostructure
photocatalyst. Nevertheless, the peaks on the Z-scheme heterostructure curve spectra and binary
photocatalyst were difficult to distinguish due to the low loading of CNT. In addition, the peaks
belonging to the CNT might locate in the same region as C-N [11].
11

ACCEPTED MANUSCRIPT
Crystallographic and Morphological Properties
XRD measurements were performed to observe the effects of nanostructure engineering of
CNT and BiVO4 on the crystalline structure of g-C3N4. Figure 2 illustrates the X-ray diffraction
peaks of binary g-C3N4/BiVO4 and Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst
along with the pristine CNT and g-C3N4, respectively. The pristine CNT exhibit typical diffraction
peaks at 26.79 and 42.57o of 2θ which could be indexed to the (002) and (100) lattice plane for
hexagonal graphitic material which is in agreement with the standard data for MWCNT (JCPDS
No. 01-0646) [24]. Moreover, the XRD patterns of pristine g-C3N4 show the formation of two
characteristic peaks deconvoluted at 13.95 and 27.91º of 2θ. The high-intensity peak at 27.91º
corresponds to the (002) crystal plane and d-spacing of 2.05 Å indicate a graphite-like interlayer
stacking microstructure [23]. In addition, the presence of the peak at 13.95º can be ascribed to the
(100) lattice plane which signifies the presence of the tri-s-triazine unit in-planar ordering (JCPDS
No. 01-0646) [7].
For the binary g-C3N4/BiVO4 photocatalyst, additional diffraction peaks were observed at
19.59, 29.12, 35.10 and 47.31º which can be assigned to (011), (121), (040) lattice plane of the
monoclinic scheelite BiVO4 (JCPDS No. 14-0688) [13, 14]. To highlight further, the coexistence
of both elements in the binary hybrid photocatalyst is demonstrated by the slight change in the
peak shape and position of g-C3N4 in the binary hybrid. The peak shape and position shift in the
spectra also signify that there were no other impurities formed during the photocatalyst fabrication
process [13]. Interestingly, the intensity of (002) planes for the binary and Z-scheme
heterostructure photocatalyst decreases when compared to pristine g-C3N4, indicating the
disruption of the bulk structure of g-C3N4, showing that the composite photocatalyst developed
12

ACCEPTED MANUSCRIPT
differentially from the pristine g-C3N4 along the plane [17]. The increase in the intensity of (011)
crystal plane of BiVO4 indicates that it is reached in electron density which will promote the
formation of radicals leading to an improved photocatalytic performance. The existence of both g-
C3N4 and BiVO4 characteristic peaks indicates the successful formation of binary hybrid
photocatalyst via wet impregnation method. The findings were in agreement with the engineering
nanostructure studies reported by the previous study [23, 25].
The Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst shows only g-C3N4 and
BiVO4 characteristics peaks without typical CNT peaks were detected, indicates a low amount of
CNT dosage was used in the formation of Z-scheme heterostructure photocatalyst [26].
Congruously, the low amount of CNT were highly dispersed during the synthesis process within
Z-scheme heterostructure with respect to its nanostructure characteristic, hence showing no CNT
characteristic peaks in the XRD pattern. Similar notion reported previously claimed that the
unnoticed CNT stacking is due to the high possibility of CNT peak overlapping with other crystal
planes that coexist within the heterostructure [7, 11, 26].
The morphological features and crystal microstructure of the developed photocatalyst were
further investigated through FESEM analysis. Figure 3 illustrates the morphological structure of
the binary and Z-Scheme heterostructure photocatalyst along with the individual sample for
comparison. It can be observed that the pristine g-C3N4 and pristine BiVO4 photocatalyst exhibit
sheet-like structure with a smooth surface and cuboid decahedral shape-like structure, respectively.
[13, 18]. The network of g-C3N4 consists of irregular folding, large particle size and wrinkles.
Meanwhile, the surface morphology of g-C3N4/CNT/BiVO4 is composed of a cubic-decahedral-
like structure of BiVO4 and highly dispersed CNT on the crystal surface of g-C3N4. In addition,
13

ACCEPTED MANUSCRIPT
the crystal surface of g-C3N4/CNT/BiVO4 heterostructure samples was slightly rough in
comparison to the binary g-C3N4/BiVO4 photocatalyst. This observation suggested the attachment
of CNT on the crystal surface of the ternary heterostructure. The added CNT could be dispersed
on the surface structure of g-C3N4 during the synthesis process owing to its nanostructure
characteristic which is in agreement with the studied reported previously [11]. As a result, the
smooth photocharge carrier separation and migration were obtained which further improve the
performance of the photocatalytic degradation due to active surface area enhancement in the Z-
scheme heterostructure photocatalyst. It was noteworthy to highlight that there was no significant
change observed on the overall morphology of g-C3N4 in the ternary hybrid photocatalyst due to
the low loading ratio of BiVO4 and CNT to the g-C3N4 photocatalyst [7, 11, 26].
3.3
Total Surface Area and Porosity Analysis
The specific surface area of the as-prepared photocatalysts was examined by N2 adsorption-
desorption isotherm. Based on the IUPAC-BET classification, all of the as-prepared photocatalysts
can be identified as type II isotherm which represents unrestricted monolayer-multilayer
adsorption of macroporous structure (as shown in Figure S2) [13, 18]. The type II isotherm
indicates a presence of an interconnected network of pore structure with different size and
morphologies, consistent with the FESEM analysis mentioned in the previous section [13]. It was
noteworthy to highlight that type II isotherm is an outstanding feature of photocatalyst used in
absorption studies which are expected to facilitate the photocatalytic performance efficiency.
The specific surface area, (SBET), pore volume and pore size are summarized in Table 1.
The pristine CNT shows a remarkably high specific surface area while BiVO4 photocatalyst shows
14

ACCEPTED MANUSCRIPT
a comparatively low specific surface area. Surprisingly, the pristine g-C3N4 shows larger specific
surface area compared to the binary hybrid and Z-scheme heterostructure photocatalyst.
Congruently, the specific surface area of the Z-scheme heterostructure photocatalyst is slightly
higher than the binary composite which can be attributed to the presence of CNT which provide
more available active sites on the g-C3N4 network [12]. Similarly, the pore volume and pore size
analysis illustrates the same patterns as the BET specific area for all of the as-developed
photocatalyst. Hypothetically, photocatalyst with a higher surface area is expected to attain higher
adsorption capacity due to the presence of more active sites within the as-developed
heterostructure. In addition, the bigger the size of the photocatalyst pore structure will greatly
enhanced the adsorption rate of the pollutants onto the surface of the photocatalyst. The synergistic
effect between the components forming the heterostructure successively increase the presence of
the active sites and broaden the light absorption spectra which is expected to enhance the
photocatalytic degradation performance of the ternary nanostructure [27].
Table 1: The BET specific surface area, pore size and pore volume of the as-developed
photocatalyst.
Photocatalyst
Pure CNT
BET Surface
Area (m2/g)
232.90
Pore Volume
(cm3/g)
Pore Size (Å)
134.82
0.8000
Pure BiVO4
2.51
40.27
33.95
34.99
0.0067
0.1338
0.0871
0.1054
104.67
Pristine g-C3N4
g-C3N4/BiVO4
g-C3N4/CNT/BiVO4
130.79
100.37
120.04
15

ACCEPTED MANUSCRIPT
Optical Properties
The performance of the photodegradation by photocatalysis are greatly influenced by the
optical properties of the as-prepared photocatalyst. Hypothetically, the photocatalyst with smaller
band gap energy is highly favourable since it possesses wide light absorption capabilities in
absorbing visible-light energy, towards improving the photodegradation performance efficiency.
Figure 4 (a, b) illustrated the light absorption properties of all as-developed heterostructure
photocatalyst along with the individual sample for comparison. The band gap absorption edge of
all photocatalyst samples display resilient absorption capacity in the visible-light regime which
strongly justify that all of the as-prepared photocatalysts were visible-light-driven photocatalyst.
In addition, the resilient light absorption properties monitored was due to a relationship between
the as-prepared crystalline structure and scattering phenomena.
Figure 4 (b) revealed the estimation of the band gap energy for all as-prepared photocatalyst
via extrapolation from the plot of (αhn)2 versus (hn). The pristine g-C3N4 exhibit band gap
absorption edge at a threshold wavelength of 440 nm, which is attributable to the standard 2.85 eV
medium band gap energy [9, 18]. Subsequently, the band gap energy for pristine BiVO4, binary g-
C3N4/BiVO4 and Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst were found out to
be 2.40, 2.74 and 2.70 eV respectively. The sensitization of g-C3N4 with BiVO4 results in a band
gap energy reduction to 2.74 eV, due to the surface plasma resonance characteristic of BiVO4,
which can increase the heterostructure light-harvesting capabilities [7]. Gratifyingly, the addition
of CNT into the g-C3N4 and BiVO4 network further increase the absorption of visible light, thus
leading to higher photocatalytic activity under visible light irradiation. Remarkably, there is a
significant improvement in the band gap energy of the Z-scheme heterostructure photocatalyst
from 2.85 to 2.70 eV which indicates that the incorporation of CNT had certainly improved the
16

ACCEPTED MANUSCRIPT
visible absorption region of the photocatalyst. It is noteworthy to mention that carbon material
can be doped as a photosensitizer for band gap narrowing of the Z-scheme heterostructure, leading
to faster migration of the photogenerated holes and photoexcited electron [27]. In addition, these
results also suggested a possible bond formation between the components forming the
heterostructure, which introduce localized occupied state within BiVO4 band gap and subsequently
extend the absorption wavelength within the spectrum [11].
Surface Chemical Properties
The chemical composition of the as-prepared photocatalysts was analysed by XPS spectra
Figure 5 (a, b, c, d and e) locates the XPS spectra of C 1s, N 1s, Bi 4f, O 1s and V 2p species that
presence in the as-prepared photocatalyst sample. The C 1s core level peaks of pristine g-C3N4
photocatalyst exhibited at 288.5 eV represent π- π* vibrations, which give the conjugated
properties of the semiconductor photocatalyst. This fascinating feature allows the photocatalyst to
act as an electron sink to suppress the recombination rate of the photogenerated electron-hole pairs
[5]. Meanwhile, the N 1s spectra show a deconvolution peak at a binding energy of 399 eV, the
strongest peaks for N 1s according to the Gaussian fitting. This binding energy is associated with
sp2-hybridized nitrogen (C=N-C) and tertiary nitride (N-C3) group [28]. It is noteworthy to
mention that this asymmetrical features also signify the polymerization of urea in a large amount
[7].
The successful formation of binary g-C3N4/BiVO4 photocatalyst was verified by the
presence of all the characteristic peak of g-C3N4 and BiVO4 which are C 1s, N 1s, Bi 4f, O 1s and
V 2p species. The presence of bismuth ion in the binary photocatalyst induced by the Bi3+ species
17

ACCEPTED MANUSCRIPT
was evidenced by the existence of double peaks at 159.8 and 164.8 eV which are assigned to Bi
4f7/2 and Bi 4f5/2, respectively [25]. Meanwhile, the detection of V 2p species was deconvoluted
around 518 eV binding energy, corresponds to V5+ of BiVO4. On the other hand, the regional
spectra of O 1s centered at 531 eV are associated with the Bi-O bonds and the crystal lattice of
oxygen (O2-).
The full scan range for g-C3N4/CNT/BiVO4 photocatalyst reveals the presence of C 1s, N
1s, Bi 4f, O 1s and V 2p species with the exception of C 1s species for CNT albeit successful CNT
loading has been validated by FTIR, FESEM and DR-UV-Vis analysis as discussed previously.
The inherent heterostructure and surface morphology characteristic of the CNT itself suggest a
possibility of the components to be well-dispersed on the surface of g-C3N4, which suggest why
the peak intensities are hard to observe [27]. Compared to the pristine g-C3N4, the C 1s, N 1s, Bi
4f7/2, and V 2p species present in the g-C3N4/CNT/BiVO4 photocatalyst shift positively to higher
binding energy. This occurrence presumably due to the interaction between g-C3N4, CNT and
BiVO4 which induced the C 1s and N 1s orbitals of the g-C3N4 more towards the inner shift during
the development of the Z-scheme heterostructure photocatalyst [29]. Congruously, the O 1s and
Bi 4f5/2 species shift to a lower binding energy suggested an interaction between the components
forming the heterostructure during the synthesizing process of the photocatalyst [13].
Photocatalytic Degradation Performance of Phenol
The photocatalytic degradation performance of the synthesized photocatalysts was evaluated
by the measuring the photodegradation of phenol in aqueous solution at room temperature under
simulated sunlight irradiation. For comparison, the photocatalytic degradation of phenol was also
18

ACCEPTED MANUSCRIPT
measured using pristine g-C3N4, g-C3N4/CNT and binary g-C3N4/BiVO4 photocatalyst under the
same experimental condition as shown in Figure 6. It was found that the phenol degradation
efficiency increase in the following order: pristine g-C3N4 < g-C3N4/CNT < binary g-C3N4/BiVO4
photocatalyst < Z-scheme g-C3N4/CNT/BiVO4 photocatalyst. The pristine g-C3N4 photocatalyst
exhibits the lowest photocatalytic degradation performance with only 15.4 % phenol removal after
120 minutes. The limited photocatalytic degradation performance observed could be attributed to
its high band gap energy of 2.85 eV, consequently limits its light absorption abilities in harnessing
the astronomical part of solar energy, evidently from DR-UV-Vis analysis. On the other hand, the
incorporation of BiVO4 and CNT resulted in significant enhancement in the photocatalytic
performance of g-C3N4, with the binary g-C3N4/BiVO4 photocatalyst showing almost four times
higher degradation performance and Z-scheme g-C3N4/CNT/BiVO4 photocatalyst almost six times
higher in comparison to the pristine g-C3N4. Compared to the single component, the binary hybrid
of g-C3N4/ BiVO4 exhibit better performance due to the presence of heterojunction between the
two photocatalyst. The addition of BiVO4 helps to promote better optical absorption ability over
the well-contacted structure in heterojunction with a narrow band gap energy of 2.78 eV, consistent
with the DR-UV-Vis analysis.
Furthermore, the Z-scheme g-C3N4/CNT/ BiVO4 heterostructure photocatalyst exhibit the
highest photocatalytic activity, degrading 80.6% of the phenol within 120 minutes. The
incorporation of CNT had resulted in higher photocatalytic performance owing to its ability as an
electron bridge mediator which is responsible to facilitate photogenerated charge carrier separation
[6]. The Z-scheme photocatalyst sample exploits the heterojunction advantages and the significant
individual photocatalytic properties of the component, resulting in enhancement of the
photocatalytic degradation performance in treating the phenol.
19

ACCEPTED MANUSCRIPT
Likewise, the BET surface area, pore size and pore volume of the Z-scheme photocatalyst
is slightly higher in comparison to the binary hybrid photocatalyst and pristine g-C3N4, results in
more targeted pollutants can be adsorbed in the Z-scheme heterostructure photocatalyst and
ameliorate the diffusion rate between the surface of the photocatalyst and the targeted pollutants.
Nevertheless, the photocatalytic performance degradation of pristine g-C3N4 demonstrates the
lowest photocatalytic activity at only 15.8 % despite having a large BET surface area in
comparison to the Z-scheme heterostructure sample. It was noteworthy to highlight that this
occurrence can be partly attributed to the individual drawbacks of the semiconductor materials-
low quantum efficiency of visible light utilization which lead to a lower photocatalytic activity
[12]. To highlight further, some recent studies revealed that the BET specific surface area show
limited to no effects on the photocatalytic activity of the photocatalyst, which is aligned with the
present study [14, 30].
Synergistic Interaction of CNT loading and H2O2 on Enhancing Photocatalytic
Degradation of Phenol
To further illustrate the effect of the presence of mediator on the photodegradation
performance, the synergistic interaction between CNT loading and H2O2 on the removal of phenol
are evaluated by following the design of experiment (DOE) couple with central composite design
(CCD) as shown in Table 2.
The reliability of the Response Surface Model was examined from the generated analysis
of variance (ANOVA) in order to determine whether the CCD design methodology provide a
satisfactory estimate to the studied system as shown in Table 3. The overall significance of the
developed model is described via Prob > F values relationship. Based on the results, generated
20

ACCEPTED MANUSCRIPT
value of < 0.0001 which is less than 0.05 threshold indicates that the model term is significant and
adequate. The coefficient of determination, R2 was recorded at 0.9772, indicating that 97.72% of
the response data variability around its means. In addition, the Predicted R2 of 0.7808 is in
reasonable agreement with the Adjusted R2 of 0.9582; i.e. the difference is less than 0.2. This
explained that the experimental data fit the developed model. From the obtained response, the
coefficient of variation (C.V) is determined at 0.7982 %, far away from the threshold value of 10%
which indicate a good predictability. Hence, the developed model was dexterous enough in
explaining the correlation and interaction of the study parameters.
Figure S3 (a, b) depicts the 3D response surface plot and contour plot of the interactive
effects of CNT loading and presence of H2O2 as an oxidizing agent over the Z-scheme g-
C3N4/CNT/ BiVO4 heterostructure photocatalyst on phenol removal at a constant reaction time of
120 minutes. This plot displayed that the CNT loading as the most important parameter had a
completely quadratic impact on the rate of removal its second order term of A2 is significant.
Moreover, it was observed that the percentage removal of phenol increased with an increase in the
CNT dosage when the electron scavenger loading was kept at a constant value. This could be
attributed to the presence of a greater surface area of the composite with higher CNT dosage which
provides more active sites for phenol adsorption, consistent with the BET results discussed earlier
[11, 15]. In addition, the incorporation of CNT improved the photocatalyst response towards the
visible light source which in turn facilitate the electron-hole pair separation and transportation [9].
The tremendous electron properties of CNT also help in reducing the charge carrier recombination
rate of the electron-hole pairs of the photocatalyst, which is favourable for photocatalytic activity.
21

ACCEPTED MANUSCRIPT
Table 2: Experimental and predicted values of phenol rate of removal from the total 13 runs.
Run
CNT dosage
(Wt. %)
H2O2
(Vol. %)
Phenol Removal
(%)
Experimental
Predicted
73.36
73.36
73.58
73.36
74.83
73.36
73.80
73.35
78.78
80.84
73.36
73.27
74.17
1.25
1.25
0.5
12.5
12.5
20
73.41
73.29
74.14
73.41
75.64
73.29
71.42
73.12
78.20
80.60
73.41
72.47
74.16
1
2
3
1.25
0.5
12.5
5
4
5
1.25
2.31
1.25
1.25
2
12.5
12.5
23.1
1.89
5
6
7
8
9
10
11
12
13
1.25
0.19
2
12.5
12.5
20
22

ACCEPTED MANUSCRIPT
Table 3: ANOVA of the quadratic model for the rate of phenol removal.
Sum of
Squares
Mean
Square
Source
aDF
F-Value
51.43
Prob.>F
<0.0001
Model
88.59
2.07
5
6
17.72
Residual
0.3445
Lack of
Fit
2.07
2
1.03
<0.0001
b Std. Dev =0.5870, c Adeq Precision =20.7632, d C.V. % =0.7928, R2 =0.9772
On the contrary, the addition of H2O2 as an oxidizing agent shows opposite outcomes, in
which the higher the amount of H2O2 had an antagonistic effect towards the photodegradation
performance of phenol under visible light irradiation. Hypothetically, the higher the H2O2
concentration is favourable, since more •OH radicals will be produced within the photocatalytic
system, which is expected to promote the degradation of phenols in the aqueous solution [31].
However, the photocatalytic degradation of phenol in this system started to deteriorate with further
increment in H2O2. Beyond 5 vol.% of the optimum dosage, the excess H2O2 presents in the system
would react with the hydroxyl radicals formed earlier lead to the generation of weaker oxidants
•OH2. Congruously, •OH radicals can also dimerize to H2O2 at high concentration. Hence, the
excess H2O2 scavenger will acts as an inhibiting agent by consuming the •OH radicals which are
responsible to accelerate the degradation of pollutants molecule [32].
H2O2 + •OH → •OH2 + H2O
•OH + •OH → H2O2
(3)
(4)
23

ACCEPTED MANUSCRIPT
Kinetic
Behaviour
of
as-developed
g-C3N4/CNT/BiVO4
Z-Scheme
heterostructure photocatalyst
The kinetic behavior of the best as-developed g-C3N4/CNT/BiVO4 Z-Scheme
heterostructure photocatalyst performance (2 wt.% of CNT and 5 vol.% of H2O2) was investigated
to examine the mechanism of phenol removal. The experimental data were tested with two kinetics
model namely pseudo-first-order and pseudo-second-order. The pseudo-first order model
proposed by Lagrergen suggested an establishment of reversible reaction between the solid and
liquid phase of the system [1 – 3]. The model of the pseudo-first-order kinetic equation can be
expressed as follows:
ln (qe - qt) = ln (qe) - k1t
(3)
Where qe is the amount of phenol adsorbed per unit weight of adsorbent at equilibrium
(mg/g), qt is the amount of phenol adsorbed per unit weight of adsorbent at time t (mg/g) and k1 is
the adsorption rate constant (min-1).
On the other hand, the pseudo-second-order kinetic model considers the involvement of
chemisorption mechanism via electron exchange between the adsorbate and adsorbent over the
entire range of the system [2 – 4]. The expression for the pseudo-second-order kinetic model can
be expressed as follow:
t
1
1
=
+ t
q
e
(4)
2
q
k q
2
t
e
Where k2 is the pseudo-second-order kinetic rate constant of adsorption (g mg-1min-1), other
terms have already defined in pseudo-first-order kinetic. The adsorption kinetic parameters derived
from the application of the pseudo-first-order and pseudo-second-order kinetic equation were
24

ACCEPTED MANUSCRIPT
calculated and listed in Table 4. The R2 correlation coefficient for both pseudo-first-order and
pseudo-second-order model was calculated to be more than the threshold of 80%, with the R2 value
for pseudo-first-order model at 0.9270, higher than the pseudo-second-order model at 0.8318. The
high R2 correlation coefficient of Lagregen pseudo-first-order kinetic model suggested that the
experimental kinetic data obeyed the pseudo-first-order model (as shown in Figure S4). Hence,
this model is a recommendable model for the description of the adsorption of phenol on Z-scheme
g-C3N4/CNT/BiVO4 heterostructure photocatalyst which could be further used in determining the
equilibrium sorption capacity, rate constants and photocatalytic degradation performance of
phenol removal. Furthermore, the previous study also found that the pseudo-first-order is the most
suitable kinetic model for low initial phenol concentration [33].
Table 4: Summary of the parameters of pseudo-first order and pseudo-second order kinetic model
for the phenol removal over Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst.
Absorption at
Equilibrium
Kinetic model
Kinetic Constant
R2
k1
qe
Pseudo-First Order
0.0102
0.0785
0.927
(min-1)
(mg g-1)
K2 (g
mg
Pseudo-Second
Order
qe
0.3474
1.1061
0.8318
(mg g-1)
min-1)
Adsorption Isotherm of as-developed Z-Scheme g-C3N4/CNT/BiVO4
heterostructure photocatalyst
The adsorption isotherms model describes the interaction between solutes and adsorbent in
the system expressed through a mathematical equation at a constant temperature and pH. For this
25

ACCEPTED MANUSCRIPT
photocatalytic system, the interaction is expressed as the ratio of the adsorbed amount of phenols
on the Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst to amount of remaining
phenols left in the aqueous solutions. The adsorption properties and equilibrium data of the
photocatalytic system were calculated and examined through three adsorption isotherms which are
Langmuir, Freundlich, and Temkin as summarized in Table 5.
The Langmuir isotherm model can be applied to a homogeneous monolayer adsorption
process which assumes that the adsorption of the involved molecule can occur only at the definite
localized sites. The mathematical equation of the Langmuir model can be represented as follow
[3, 4]:
C
qe= b1q+bC
(5)
m
e
e
Where qe (mg/g) and Ce (mg/L) are the amount of solute adsorbed per unit weight of
adsorbent at equilibrium and phenol concentration at equilibrium, respectively. In addition, qm
(mg/g) and b (L/mg) is the maximum adsorption capacity and adsorption equilibrium constant
linked to the free energy of adsorption. On the other hand, the Freundlich isotherm model
suggested that the model is applicable for multilayer adsorption with the distinctive non-uniform
heat distribution on the heterogeneous surface. The linearized mathematical equation form for this
model can be expressed as below [3, 4]:
1
ln qe = ln Kf + ln Ce
(6)
n
Where Kf is an empirical constant correlated to the adsorbent sorption capacity
(L/mg)(L/g)1/n and n is a constant which is responsible for indicating the adsorption intensity.
26

ACCEPTED MANUSCRIPT
Meanwhile, the Temkin isotherm model presumes that the adsorption process belongs to the
reaction with the uniform distribution of binding energy. Hypothetically, the adsorption heat of all
molecules within the system will be decreased linearly owing to the existence of the adsorbent-
adsorbate interactions. The mathematical expression for Temkin model can be written as below
[3, 4]:
qe= K1 ln K2 + K1 ln Ce
(7)
Where K1 is associated with the heat adsorption (L/g) and K2 is the Temkin isotherm
constant (dimensionless). The both Temkin isotherm parameters can be calculated from the linear
plots of qe and ln Ce.
Table 5: Summary of parameters of Freundlich, Langmuir and Temkin adsorption isotherm model
for the removal of phenol over Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst.
Model
Freudlich
Langmuir
Temkin
Isotherm Constant
Absorption at
Equilibrium
R2
Kf
3.6604
12.582
36.196
n
b
15.641
0.8028
0.5062
0.9885
(mg g-1)
qe
0.4669
14.62
(mg g-1)
K1
K2
(L g-1)
Based on the adsorption isotherm analysis from the three abovementioned models, it was
found that the adsorption isotherm of the Z-scheme g-C3N4/CNT/BiVO4 heterostructure
27

ACCEPTED MANUSCRIPT
photocatalyst was confirmed by both Freundlich and Temkin isotherm model, based on the high
R2 correlation coefficient. The R2 correlation coefficient for Temkin isotherm model (0.9885) is
higher in comparison to the Freundlich isotherm model (0.8028), suggested that the Temkin
isotherm model is a recommendable model for the description of adsorption isotherm of phenol
removal over Z-scheme heterostructure photocatalyst. The apparent adsorbent-adsorbate
interaction on the adsorption isotherms observed based on the Temkin isotherms was also
consistent with the pseudo-first-order kinetic model discussed previously. Congruously, the high
value of K1 indicates the chemical nature of the adsorption process and since the positive slope
was obtained, the adsorption reaction was considered as an exothermic reaction.
Reusability and Stability of the Z-Scheme g-C3N4/CNT/BiVO4
In the photocatalytic degradation system, recycling the photocatalyst and improving the
photocatalyst stability can be a suitable option for reducing the production cost of the synthesis as
the photocorrosion issues will greatly restrict its practical applications. Hence, the stability of the
as-developed Z-scheme g-C3N4/CNT/BiVO4 heterostructure photocatalyst was evaluated under
the same reaction condition by recycling the photocatalyst three times to assess its reusability. The
photocatalyst was filtered, washed using deionized water and dried in an oven at 80 °C after each
run. Figure 7 (a) depicts the photocatalytic degradation performance of the Z-scheme g-
C3N4/CNT/BiVO4 heterostructure photocatalyst in three cyclic activity. The Z-scheme
heterostructure photocatalyst remains a profound performance with 78.5% rate of phenol removal
after three cyclic activities. The slight reduction in the photocatalytic performance was observed
which can be related to the changes of the Z-scheme heterostructure surface and deactivation of
the photocatalyst by intermediates formed on the surfaces [34]. Nevertheless, the XRD and FTIR
28

ACCEPTED MANUSCRIPT
results for the used g-C3N4/CNT/BiVO4 heterostructure photocatalyst showed that no obvious
changes occurred in the crystal structure after three cyclic activity as illustrated in Figure 7 (b, c).
The obtained results suggest that the as-developed Z-scheme g-C3N4/CNT/BiVO4 heterostructure
photocatalyst exhibits excellent stability in the photocatalytic degradation of phenol and tends not
to suffer from photocorrosion during the degradation process.
Mechanism Route for Photodegradation of Phenol over Z-Scheme g-
C3N4/CNT/BiVO4
It is known that the photocatalytic performance of the as-developed photocatalyst is associated
with its band gap structure. Thus, it is imperative to find out the conduction band (ECB) and valence
band (EVB) of the photocatalyst. The ECB and EVB of the photocatalyst can be calculated using the
following empirical formula [35]:
ECB = X – Ec – 0.5 (Eg)
(8)
(9)
EVB = Eg + ECB
Where X is the absolute electronegativity of the photocatalyst, EC is the energy of free
electrons on the hydrogen scale (ca. 4.5 eV), and Eg is the band gap energy. Based on the DR-UV-
Vis analysis, the band gap energies for g-C3N4 and BiVO4 were 2.85 and 2.40 eV, respectively.
The absolute electronegativity (X) of g-C3N4 and BiVO4 photocatalysts were 4.64 and 6.04 eV,
respectively [35, 36]. The calculated ECB and EVB values of g-C3N4 photocatalyst were -1.285 and
1.565 eV, respectively while the calculated ECB and EVB values of BiVO4 photocatalyst were 0.34
and 2.74 eV, respectively.
29