Photocatalytic Hydrogen Production by Boron Modified TiO2/Carbon Nitride Heterojunctions

Article abstract of DOI:10.1002/cctc.201901703

Boron-doped anatase TiO₂ particles were effectively coupled with carbon nitride (CN) forming nanocomposites. The materials were fully characterized by DR-UV-Vis, N₂ adsorption-desorption isotherms, XRD, Raman, FTIR, TGA, XPS, TEM, el

Full text of DOI:10.1002/cctc.201901703

Accepted Article
Title: Photocatalytic hydrogen production by boron modified TiO2 /
carbon nitride heterojunctions
Authors: Konstantinos C Christoforidis, Tiziano Montini, Maria Fittipaldi,
Juan Josè Delgado Jaén, and Paolo Fornasiero
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201901703
Link to VoR: http://dx.doi.org/10.1002/cctc.201901703
A Journal of
www.chemcatchem.org

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
Photocatalytic hydrogen production by boron modified TiO2 /
carbon nitride heterojunctions
Konstantinos C. Christoforidis[a],[b],[c]*, Tiziano Montini[b], Maria Fittipaldi[d], Juan Josè Delgado Jaén[e]
and Paolo Fornasiero[b]*
Abstract: Boron-doped anatase TiO2 particles were effectively
Among the different catalysts applied in photocatalytic H2
coupled with carbon nitride (CN) forming nanocomposites. The
production process either from pure water splitting or the
materials were fully characterized by DR-UV-Vis, N2 adsorption-
photoreforming of carbohydrates, TiO2 is the most studied[1a, 1d,
desorption isotherms, XRD, Raman, FTIR, TGA, XPS, TEM, electron
2]. This originates from the TiO2 inherent properties such as low
energy loss spectroscopy, photoluminescence, and electron
cost, chemical inertness and non-toxicity. Unfortunately, pure
TiO2 is active only under UV-light irradiation and presents fast
charge recombination rates. Different approaches have been
paramagnetic spectroscopy (EPR) spectroscopy. The developed
heterojunctions were applied as photocatalysts for hydrogen (H2)
evolution by the photoreforming process of ethanol under solar
irradiation, using minimal amount of Pt nanoparticles (0.1 wt.%) as
co-catalyst. The effects of boron addition and CN content were
evaluated and optimized nanocomposite presented 85% increase in
H2 evolution compared with the pure anatase TiO2 catalyst. The
observed higher H2 evolution rates were ascribed to improvements
in charge formation and separation efficiency due to the B-dopant
and the presence of CN.
applied to deal with these limitations including doping with metal
[2-3]
and non-metal elements
as well as the development of
heterostructures and heterojunctions [1a, 2]. Among the different
dopants used to improve the properties of TiO2, boron is an
attractive candidate since B-doped TiO2 has shown
improvement in photoactivity under either UV[4] or visible light
irradiation[5] as well as improvements in light absorption and
formation of photogenerated charges[4, 6]. Over the last years,
carbon nitride (CN) has been also established as a promising
catalyst in solar fuels production processes[1a, 7]. CN is a metal-
free, visible light active semiconductor with relatively high
Introduction
Clean energy supply with no carbon-footprint is one of the most
critical challenges the scientific community is facing. This need
originates from the severe environmental impact of traditional
fossil fuel usage through the production of greenhouse gases
together with their progressive depletion. Semiconductor
photocatalysis and in particular the field of solar fuels, is
currently considered as one of the most practical candidates to
provide sustainable clean energy[1]. From the variety of solar
fuels presented in the literature including fuels derived from CO2
conversion, photocatalytic H2 production from water or
bioavailable substrates is considered one of the most promising
processes able to deal with both sustainable clean-energy
conduction band edge potential (~-1.12 eV vs. NHE)[7d]
.
However, bulk CN presents moderate photoactivity mainly due
to the fast charge recombination rates [8]. Therefore, modification
of the parent material is required to achieve reasonable
7c, 7e, 8c,
activity[1a,
9]. Further on the development of efficient
photocatalyst for H2 production, the formation of a heterojunction
has been also shown as an efficient route to prevent charge
recombination in semiconductors and to improve light absorption
properties[1a, 10]. This is accomplished via charge separation in
the different phases of the heterojunction through the formed
interface/junction and the coupling of wide with narrow band gap
semiconductors[1a, 2, 3b, 11]. In this direction, heterojunctions made
of TiO2 and CN have been effectively applied in a variety of
supply and reduction of greenhouse gas emissions[1a, 1b]
.
photocatalytic reactions[3b, 12]
.
[a]
[b]
Dr. K.C. Christoforidis. Department of Environmental Engineering,
Democritus University of Thrace, Vasilisis Sofias 12, 67100 Xanthi,
Greece. E-mail: kochristo@env.duth.gr
In the present study we performed a combined modification
through a) the development of nanocomposites made of CN and
anatase TiO2 for the construction of heterojunctions and b)
doping of anatase TiO2 with boron. The prepared materials were
fully characterized and applied as photocatalysts in H2
production from the photoreforming of ethanol under simulated
solar irradiation. Both boron and CN improved significantly H2
evolution compared with pure TiO2. Finally, the mechanism of
activity is discussed on the bases of the catalytic and
spectroscopic data.
Dr. K.C. Christoforidis, Prof. T. Montini, Prof. P. Fornasiero.
Department of Chemical and Pharmaceutical Sciences, ICCOM-
CNR and INSTM University of Trieste, Via L. Giorgieri 1, 34127
Trieste, Italy. E-mail: pfornasiero@units.it
Dr. K.C. Christoforidis. Institut de Chimie et Procédés Pour
l’Energie, l’Environnement et la Santé, (ICPEES) ECPM, University
of Strasbourg, 25 rue Becquerel Cedex 2, Strasbourg, France.
Prof. M. Fittipaldi. INSTM and Department of Physics and
Astronomy, via Sansone 1, University of Florence, via della
Lastruccia 3-13, 50019 Sesto Fiorentino (FI), Italy.
[c]
[d]
[e]
Dr. J.J. Delgado Jaén. Departamento de Ciencia de los Materiales e
Ingeniería Metalúrgica y Química Inorgánica, Facultad de Ciencias,
Universidad de Cádiz, Campus Río San Pedro, 11510, Puerto Real,
Cádiz, Spain.
Results and Discussion
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
The structure of TiO2 and CN were assessed by XRD. The XRD
patterns are given in Figure 1. The TiO2 was in the anatase
phase (JCPDS: 21-1272) in all prepared materials: the pure TiO2
sample, the TiCNx and TiCN5By series. It has been shown
A
previously that the TiO2 crystal form is not affected by either the
[3b,
presence of CN or boron as a dopant
12b]. However, the
presence of boron somehow affected the XRD patterns of the
TiCN5By samples. Increasing the nominal boron content in the
TiCN5By series results in a broadening of the diffraction peaks.
This implies a decrease of the TiO2 crystallite size/crystallinity.
The anatase crystallite sizes of all samples extracted using the
Scherrer equation are presented in Table 1.[13] This indicates
that doping with small amounts of boron brings a distortion in the
anatase structure in accordance with the literature[14]. The actual
B/Ti molar ratio has been determined from ICP-AES results,
observing that it is significantly lower than the nominal one: 0.71
for TiCN5B5, 1.22 for TiCN5B10, 2.19 for TiCN5B20 and 1.14
for TiCN0B10. Notably, these results are in agreement with
reports on B-doped TiO2 materials[15] and could be rationalized
considering that H3BO3 is highly volatile and undergoes to
decomposition to B2O3 even at the mild temperature employed
during calcination of the present materials (350°C). Consistently,
the minor differences in the XRD observed upon B doping are
related to the low amount of incorporated B. The CN XRD
pattern exhibit two characteristic reflections at approximately 13
oC and
TiO2
TiCN2
TiCN5
TiCN7
TiCN10
CN
80
20
40
60
2
B
22 24 26 28 30
2
TiCN5B5
TiCN5B10
TiCN5B20
Table 1. BET surface area, crystallite size and CN content of the prepared
materials.
TiCN0B10
80
Catalyst
TiO2
BET surface area (m2/g) Crystallite size (nm)[a] CN % wt.[b]
20
40
60
2
69.8
64.4
70.6
77.9
72.6
79.6
108.5
120.8
69.9
86.9
15.1
15.7
15.9
15.1
15.1
16.2
13.3
12.0
16.4
–
–
TiCN2
0.6
1.3
1.9
4.2
1.3
1.1
1.2
–
Figure 1. XRD patterns of the TiCNx (A) and the TiCN5By series (B).
TiCN5
o
TiCN7
27.4 C corresponding to the (1 0 0) (interplanar spacing of the
tri-s-triazine unit) and (0 0 2) (interlayer stacking) crystal planes
TiCN10
TiCN5B5
TiCN5B10
TiCN5B20
TiCN0B10
CN
[7b]
of CN
. These two features confirmed the successful
formation of CN under the conditions applied [9d]. However, no
diffraction peak corresponding to CN was observed in the
nanocomposites. This probably originates from the low amount
of CN used.[9d]
Raman spectra of the un-doped and boron doped TiO2/CN
composite samples (Figure 2) agree well with XRD results
showing the exclusive presence of the anatase structure with
peaks at ca. 144.8, 196, 399, 518 and 640 cm-1, corresponding
to the Eg(1), Eg(2), B1g(1), B1g(2), A1g(1) and Eg(3) modes,
–
[a] calculated using the Scherrer equation and the (1 0 1) diffraction peak of
anatase. [b] Extracted from TGA.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
Figure 2. Raman spectra of the TiCNx (A) and the TiCN5By series (B).
Figure 3. FTIR spectra of the pure TiO2 and CN catalysts and the TiCN10
composite.
corresponding to CN were detected, probably due to the low CN
concentration. Since no evidence was given for the presence of
CN in the composite materials using either XRD or Raman,
Fourier Transform Infrared
respectively.[3a] CN does not affect the spectral characteristics.
However, boron induces a small shift of the Eg(1) mode towards
lower wavenumbers (i.e. 144 cm-1, Figure S1). This may
originate from the presence of crystalline defects in the TiO2
Figure4. High resolution XPS spectrum of the TiCN5B20 sample in the B 1s
region.
framework due to the presence of boron[16]
. No peaks
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
Spectroscopy (FTIR) was applied to verify the presence of CN in
the composite. CN showed the expected band vibrations (Figure
3). The characteristic stretching vibration modes of the
heptazine ring are observed in the 1200-1680 cm-1 region,
confirming the presence CN.[7b, 17] The CN bands were detected
only in the composite with the highest CN content (TiCN10).
Such
data
Figure 5. SEM images of the pure TiO2 sample (A-B) and TiCN5B10 composite (D). HRTEM of the pure TiO2 (C), as well as the TiCN5B10 catalyst (E-F) are
also included. The Pt particles are indicated with the arrow in panel C.
suggest that the CN structure is maintained during the
hybridization process.
composite materials under the thermal treatment applied for the
calcination step of the TiO2 phase at 350 ◦C. The content of CN
in the composite materials was calculated from the weight
remained and is given in Table 1.
To verify the presence of boron and to further elucidate the
surface composition and chemical state of the prepared
materials XPS was applied. Peaks attributed to Ti, O, C, and N
are detected in the XPS survey spectra (Figure S2). High
resolution XPS spectra are given in Figure S3. For all materials,
the peaks centered approximately at 458.5 and 464.3 eV are
ascribed to the Ti 2p3/2 and 2p1/2 spin-orbit components of Ti4+
surface species on anatase TiO2[18]. The characteristic C 1s and
N 1s peaks of CN were detected at 288.4 and 398.5 eV,
respectively[18]. The presence of boron is verified by the peak
centered at approximately 191.9 eV that is attributed to the B 1s
(Figure 4)[19]. Boron peaks in this region have been suggested to
occupy interstitial positions[19]. No peaks attributed to H3BO3 and
B2O3 were detected,
The composition and the stability of the materials under thermal
treatment were examined using TGA. The TGA profiles of all
materials are given in Figure S4. Pure CN is decomposed at
temperatures higher than 570 ◦C. On the contrary, the onset for
all hybrid materials is observed at lower temperatures,
approximately at 500 ◦C. This verifies the stability of the
In addition to the above techniques that give information on the
bulk structure and composition of the hybrids, morphological
features and the verification of the coupling of the two materials
were studied using SEM and TEM. TEM images were taken
from the used catalysts after the photocatalytic reaction. Figure
5 shows representative SEM and TEM images of the pure TiO2
sample (Figure 5A, B and C). A nearly perfectly spherical
morphology is observed although no template was used in the
synthesis process. The size of the microspheres ranges
between ca. 0.6 and 1.1 μm and are composed of
interconnected TiO2 nanocrystals. Figure 5B clearly shows the
porosity of the microspheres. The Pt nanoparticles with size in
the range of 2-3 nm are also observed in Figure 5C. TiO2
microspheres are also observed in the TiO2/CN composites.
However, the size of the TiO2 microspheres (0.2 to 0.5 μm) is
significantly smaller compared with the bare TiO2 sample. In the
case of the composite, in addition to the TiO2 microspheres, CN
structures covered with
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
concordance with the assumption that the CN sheets are
covered with TiO2 nanocrystals. This is well illustrated by the
higher magnification SEM image of this area included in figure
6C.
The textural properties of the materials were studied using
nitrogen physisorption isotherms (Figure S5) since surface area
is very important in photocatalytic applications. Pure TiO2
presents
a
mesoporous structure typical for TiO2
nanoparticles[20]. Bare CN presents a hysteresis loop at high
partial pressures (0.9 < P/P0 < 1) which probably originates from
pores formed through the assembly of secondary particles. The
presence of CN induced small changes in the nanocomposites
compared with the pristine TiO2, increasing slightly the BET
surface area. However, as indicated by the N2 adsorption-
desorption isotherms, materials containing CN show higher
absorption capability at high partial pressure. The introduction of
boron in the anatase structure induced an increase in the
surface area. This is manifested in the sample prepared using
the highest boron concentration and is linked with the lower
crystallinity observed with XRD. The same trend in surface area
with varying B-content has been previously observed for B-
doped TiO2 materials prepared under similar conditions in the
absence of CN[20]
.
Figure 6. SEM images of pure TiO2 (A) and TiCN5B10 (B-C) indicating the
areas where the EDS spectra included in (D) were recorded.
Photocatalytic H2 production from aqueous ethanol solution was
performed under simulated solar light irradiation and H2
evolution was monitored for a period of 24 h under flow
conditions. The corresponding data are presented in Figure S6.
The linearity observed over 24 h verify the stability of the
prepared materials under the catalytic conditions applied. To
further establish the stability of the prepared catalyst, the XRD
and FTIR spectra of the post reaction catalyst were also
recorded for the TiCN5B10 case (Figure S7). Figure 7 shows H2
production rates using the TiCNx (Figure 7A) and the TiCN5By
(Figure 7B) catalysts. Pure CN presents approximately 20 times
lower H2 production (ca. 0.053mmol g-1 h-1) than pure TiO2 under
identical conditions. Therefore it is not included in the data
presented in Figure 7. In the TiCNx series a small variation in H2
evolution is observed by varying the CN content. H2 evolution
presents a volcano-type dependence on CN content. Activity is
improved until the CN content reaches 5%. Further increase of
the CN amount results in a decrease in activity. The optimized
TiCN5 composite presents ca. 12% higher activity than the pure
TiO2 catalyst. This most likely originates from the formation of a
heterojunction that allows efficient charge separation as
TiO2 nanocrystals are also observed (Figure 5E). Figure 5F
clearly shows that 5-10 nm nanoparticles of TiO2 are randomly
decorating the CN sheets. The decrease of the microspheres
size in the composite materials compared with the pure TiO2 can
originate from the deposition of TiO2 nanocrystals on CN in the
second case. The digital diffraction pattern of the full image
shows many spots forming two cycles that corresponds 2.3 and
1.9 Å. Unfortunately, these spacing can’t be univocally assigned
to anatase and others TiO2 phases cannot be excluded. This
result can be interpreted considering that CN nanosheets are
covered by TiO2 crystals with different orientations. Interestingly,
one single spot is observed for a spacing of 3.4 Å, which may be
due to the (002) family of planes of CN where the TiO2
nanoparticles are anchoring. Additionally, small CN sheets are
also detected at the edges of the TiO2 microspheres (Figure 5E).
It must be highlighted that the TEM images correspond to the
used catalysts. Thus, the detection of CN structures covered
with TiO2 nanocrystals after the photocatalytic test indicates a
stable and strong interaction between the two phases.
Energy dispersive X-ray spectra (EDS) was used to confirm the
intimal interaction of both phases. Figure 6D includes the EDS
spectra obtained in the areas marked with number in the SEM
images (Figure 6A and B). According to the spectra 1 and 4, we
can observe that TiO2 microspheres are composed by Ti and O
in both the pure titania and the composite samples. It must be
mentioned that both spectra show a small contribution at ~0.45
eV that correspond to the L-level emission lines of Ti and not to
the Ka emission of nitrogen that is observed at 0.39 eV. The
analysis performed on others areas of the composite sample
show the presence of Ti, O, C and N. This is in good
previously suggested in similar TiO2/CN composites [1a, 3b, 18, 21]
.
The effect of B doping was studied on the optimum CN content
(i.e. 5% CN). As illustrated in Figure 7B, the introduction of
boron results in improved H2 evolution. The TiCN5B10
composite presents the highest activity corresponding to an 85%
and 66% improvement in H2 production compared with the pure
TiO2 and the TiCN5 catalyst, respectively. It is noted that in both
series the same trend in the activity is observed when
considering the surface area of the catalyst. However, the
increase in activity drops from 85% to 20%. A photocatalytic
reaction under pure visible light irradiation (λ > 420 nm) was also
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
performed using the most active TiCN5B10 catalyst. Under
these conditions no H2 production was detected.
Figure 7. Photocatalytic H2 evolution using the TiCNx (A) and the TiCN5By composites (B).
Figure 8. Photoluminescence emission spectra of the pure CN and TiO2 and
the TiCN5 and TiCN5B10 composite materials.
light only in the UV region presenting band gap energy (Eg) of
3.22 eV (Figure S8C). On the other hand, CN absorbs light at
longer wavelengths with Eg approximately at 2.8 eV. As
expected, the CN in the composite materials increases light
absorption in the visible region (Figure S8A). Boron does not
affect the band
To investigate the role of CN and boron on photoactivity
photoluminescence, EPR and DR-UV-Vis spectra were collected.
The DR/UV-Vis spectra are shown in Figure S8. TiO2 absorbs
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
Figure 9. EPR spectra of the pure TiO2, TiCN0B10 and TiCN5B10 catalysts
taken under irradiation.
In the TiCN5B10 sample, an additional set of signals centered at
approximately 3308, 3404 and 3474 Gauss are detected. This
feature is not observed in either the pure TiO2 or TiCN0B10
sample. Based on their position, these EPR signals most likely
do not correspond to h+ in the anatase structure[13, 24b, 25, 27], nitric
gap energy (Eg ~ 3.20 eV) but a small increase in light
absorption in the visible region is detected (Figure S8B)[22]
.
•
oxide (NO) or nitrogen bulk centers (Nb ) [28]. The last two cases
These data clearly indicate that modification of TiO2 with boron
and coupling with CN expands light absorption which could be
beneficial in photocatalysis.
are also excluded based on the FTIR and microscopy analysis
that provide direct evidence for the stability of CN under
Besides light absorption, formation of photogenerated charges
and charge recombination are also important properties since
they are directly related with the photocatalytic performance. PL
emission spectra have be used to give an estimation of the
[3b,
charge separation efficiency in TiO2 and CN based materials
17], although the mechanism cannot be precisely identified based
on the current knowledge[23]. Steady-state PL spectra are
presented in Figure 8. The broad emission band centered at
approximately 435nm can be attributed to electron-hole
recombination. In terms of intensity, CN presents the highest
emission intensity followed by the pure TiO2 catalyst. The
catalysts with the highest catalytic activity in the two series (i.e.
TiCN5 and TiCN5B10 composites) show a significant decrease
in the PL spectral intensity as compared with the reference
catalysts. This suggests that charge recombination is
suppressed in the case of the composites.
conditions applied in the synthesis step. A possible interpretation
Scheme 1. Proposed charge separation mechanism in the TiO2/CN
heterojunction
To study the photogenerated charges and investigate their
impact in H2 evolution, in-situ EPR spectroscopy was employed.
EPR spectroscopy provides high resolution information for
paramagnetic centers and their environment and has been
extensively applied to study photogenerated charges on TiO2 as
trapped species[13, 24]. Figure 9 presents the EPR spectra of the
pristine TiO2, B-doped TiO2 (TiCN0B10) and the B-doped TiO2
coupled with CN (TiCN5B10) under irradiation. Two distinct EPR
signals are detected. The low-field feature, labeled A in Figure 9,
centered at g ~ 2.007 most probably originates from trapped
holes (h+) species in the anatase structure [13, 24b, 25]. The broad
component of the spectrum centered at high magnetic field
(labeled as B) expands from 3380 to 3760 G and has been
previously attributed to photogenerated e- trapped at Ti3+ centers.
The broad character is associated with the disordered
environment of such paramagnetic centers at the crystal surface
of these new EPR features can be a Ti3+ center interacting with
a nitrogen nucleus with I = 1. A theoretical spectrum obtained via
computer simulation using g values 1.987, 1.976 and 1.96 and
hyperfine values 78, 57 and 57 Gauss is given in Figure 9.
Previous studies have also reported orthorhombic Ti3+ centers in
the anatase structure[28a]. No other EPR species are considered
in the simulated spectrum. This interaction can potentially
originate from the coupling of TiO2 with CN forming a composite.
This interpretation is supported by the microscopy study and
further suggests a strong coupling between TiO2 and CN.
Based on the above, it may be concluded that the presence of
boron and CN improves the light absorption properties, improves
charge recombination phenomena via efficient charge
separation and, therefore, enhances charge formation. The
coupling of TiO2 and CN as revealed by TEM and the improved
charge handling properties based on the PL data suggests the
[13, 26]
.
Double integration of spectrum B that corresponds to surface
trapped e- reveals a difference in the actual amount of such
species in the catalysts studied. The signal intensity of spectrum
B in the TiCN0B10 and TiCN5B10 catalysts is 1.8 and 2.4 times
higher than the pure TiO2, respectively. This can originate from
the formation of more photogenerated e- in the TiCN0B10 and
TiCN5B10 catalysts. The trend observed agrees with the
photocatalytic activity as presented in Figure 7. It should be
noted that the intensity of such species depends greatly on the
illumination conditions applied and the amount of the catalyst
used. Herein, the same amount of sample was used in each
case. Although no detailed quantitative analysis was performed,
the observed trend in Ti3+ signal intensity is in line with the PL
data presented in Figure 8 and photocurrent measurements
efficient formation of a heterojunction. Taken into account the
[1a, 12b, 29]
band energy levels of TiO2 and CN
a heterojunction of
Type-II may be formed. Heterojunctions of Type-II are
characterized by the coupling of semiconductors with correctly
aligned band levels that allow spontaneous interfacial charge
separation. The Conduction Band (CB) and Valence Band (VB)
of the one semiconductor is positioned at more positive (or less
negative) potentials than those of the second. This difference
force e- and h+ to move at opposite directions improving charge
separation. Under solar light irradiation, photogenerated e-/h+
pairs can be formed in both the CN and TiO2 part of the
composite based on their UV-Vis absorption spectrum. Based
on the difference in band structure and in particular the actual
energy level of the CB and VB in the two semiconductors, h+
formed on the VB of TiO2 are spontaneously transferred to the
[4,
performed on other studies using un-doped and B-doped TiO2
6]
.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
VB of CN, while e- in the CB of CN follow the opposite direction
and are transferred to that of TiO2 [3b, 12b]. A schematic illustration
of the suggested mechanism for charge formation and
separation is given in Scheme 1. Due to charge transfer through
the interface of the junction formed, charge separation is favored
enhancing the stability and availability of photoexcited e-. This is
essential in H2 production and in the present work, is controlled
by the modification of the TiO2 phase and the formation of a
heterojunction through the coupling of TiO2 with CN.
of 0.2 mol/L was added in the aqueous phase. The amount of boron was
calculated on the basis of the molar ratio to Ti. Materials with three
different nominal B/Ti molar percentages were synthesized, 5, 10 and
20%. The catalysts in this series are labelled TiCN5By where
y
corresponds to the B/Ti molar percent.
Materials characterization
The amount of B present in the samples was determined by means of
Inductively Coupled Plasma-Atomic Emission Spectrometry (ICP-AES).
The samples were prepared by acid dissolution following the procedure
reported by Wang et al.[30]. 100 mg of the sample were dissolved in a
quartz crucible using 1g (NH4)2SO4 and 2 mL of H2SO4 and heated until a
clear solution is formed. After cooling, few drops of H2O2 30% and 10 mL
of doubly distilled water were added. The pale turbidity observed was
due to undissolved CN, which was removed by filtration on a Millipore
PDVF 0.45 μm filter. The obtained solution was diluted to 100 mL with
doubly distilled water for the analysis with an Optima 8000 instrument
(Perkin Elmer; Waltham, MA, USA) equipped with an integrated
autosampler (S10, Perkin Elmer; Waltham, MA, USA). The amounts of Ti
and B were determined at the operative wavelengths (Ti 339.90 nm; B
249.677 nm) using 5-points calibration curves. The precision of the
measurements as repeatability (as RSD%) for the analysis was always
less than 5%.
Conclusions
The present study presents the development and
characterization of boron-modified TiO2 coupled with CN for the
formation of heterojunctions and their application in H2
photoproduction from aqueous ethanol solution. TiO2 was
synthesized in-situ in the presence of preformed CN.
Modification of TiO2 resulted in improved photoactivity. Both
boron and CN content were optimized in the final composite
against H2 evolution. Materials bearing 5 wt.% CN and 10%
boron presented the highest activity. The study demonstrated
that the enhanced H2 evolution originates from the improved
charge handling properties as revealed by PL.
A TGA Q500 instrument was used for the thermogravimetric analysis.
Samples were heated from 100 to 800 oC with a heating rate of 10 oC
min-1 under air flow. The weight difference between 450 and 600 oC was
taken to calculate the composition of the nanocomposites.
A
Micrometrics ASAP 2020 system was used for the N2 adsorption–
desorption isotherms and analysis was performed at liquid nitrogen
temperature. Samples were initially degassed at 170 oC overnight and
specific surface area was determined using the Brunauer–Emmett–Teller
(BET) model. A Perkin Elmer 2000 instrument was used to acquire the
Fourier transform infrared (FT-IR) spectra using KBr pellets. A Philips
X'Pert diffractometer with a Cu Ka X-ray source was used to collect the
X-ray diffraction patterns. Diffuse reflectance UV-visible spectra were
obtained with a PerkinElmer (Lamda 35) spectrophotometer. BaSO4 was
used as reference. The Kubelka-Munk function was used to estimate the
band gap energies from the samples' optical absorption edges using:
Experimental Section
Materials synthesis
Carbon nitride (CN) was synthesized via thermal polycondensation of
urea (Sigma-Aldrich). 10 g of urea were dehydrated in an alumina
crucible for 2 h at 150 oC in a muffle furnace. The alumina crucible was
covered and the sample was further heated in air at 550 oC for 4h with a
heating rate of 5 oC min-1. The obtained powder was dispersed in 0.1M
HNO3 (Sigma-Aldrich), collected via centrifugation and washed with
water and ethanol three times. Finally, the light-yellow material was dried
in vacuum over night at 90 oC.
2
(
1-R
)
(
)
F R =
2R
For the synthesis of the TiO2/CN composites a given amount of CN (0, 4,
10, 14 or 20 mg) was dispersed in ethanol and sonicated for 2 h at room
temperature. To the above solution, Ti(n-OBu)4 (Sigma-Aldrich) in 3.5 ml
THF was added and further sonicated for 30 min at room temperature.
The volume of Ti(n-OBu)4 was chosen so that the theoretical final amount
of CN to be 0, 2, 5, 7 and 10 wt.%. For example, 835 μl of Ti(n-OBu)4
was the volume added to obtain the composite with 5 wt.% of CN. A 10%
v/v% solution of H2O in THF (final Ti(n-OBu)4/H2O molar ratio equal to
1/100) was added dropwise followed by stirring at room temperature for
24 h. For example, 5.36 ml of H2O were used in the synthesis of the
composite containing 5 wt.% of CN. The final THF/EtOH volume ratio in
all samples was kept constant and equal to 1.4. The material was
collected via filtration using a 0.45 μm PTFE filter, washed with ethanol
three times and dried in vacuum at 90 oC overnight. Finally, the material
was calcined at 350 oC for 5 h with a heating rate of 1 oC min-1. The
samples prepared were labelled as TiCNx were x denotes the nominal
wt.% of CN.
where
R is the reflectance. The Eg values were estimated by
extrapolating the linear part of the plot F(R)hv1/2 versus energy. TiO2 was
treated as indirect while CN as direct semiconductor.
Raman spectra were acquired after excitation with a 532 nm laser using
a Renishaw instrument. Photoluminescence (PL) spectra were recorded
with a Fluorolog FM-32 spectrofluorometer (Horiba Jobin Yvon) after
excitation at 360 nm. The X-ray photoelectron spectroscopy (XPS)
measurements were carried out in an ultrahigh vacuum (UHV)
spectrometer. The Al Kα line (1486.6 eV) of a dual anode X-ray source
was used as incident radiation. All spectra were recorded in constant
pass energy mode (44 and 22 eV for survey and high resolution spectra,
respectively). A Bruker Elexys E500 spectrometer was used to acquire
spectra at X-band frequencies. EPR measurements were contacted at 5
K using a continuous flow 4He cryostat (ESR 900, Oxford Instruments).
microwave power of 0.264 mW and field modulation of 100 kHz and 5 G.
The catalysts (20mg) were irradiated at 405nm directly in the cavity with
2 mW power using a diode laser (Thorlabs code DL5146-251) connected
to an optical fiber with one end inserted in the EPR sample tube.
Theoretical EPR spectra were obtained using the EasySpin software
package[31]. Catalysts were also analyzed by Transmission Electron
For the B-doped TiO2 / CN composite materials a similar process was
adopted. The CN amount was kept constant at 5 wt.% (based on the
optimized catalytic activity) and boric acid from an aqueous stock solution
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
T. Montini, C. Petit, E. Stathatose, R. Godin, J. R. Durrant, M.
Prato, P. Fornasiero, Nano Energy 2018, 50, 468-478; c) E. M.
Dias, K. C. Christoforidis, L. Francas, C. Petit, ACS Appl.
Energy Mater. 2018, 1, 6524-6534; d) X. Wang, K. Maeda, A.
Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen, M.
Antonietti, Nat. Mater. 2009, 8, 76-80; e) W. J. Ong, L. L. Tan,
Y. H. Ng, S. T. Yong, S. P. Chai, Chem. Rev. 2016, 116,
7159-7329.
a) J. Zhang, J. Sun, K. Maeda, K. Domen, P. Liu, M. Antonietti,
X. Fu, X. Wang, Energy Environ. Sci. 2011, 4, 675-678; b) Y.
Zheng, L. Lin, B. Wang, X. Wang, Angew. Chem., Int. Ed.
Engl. 2015, 54, 12868-12884; c) S. Yang, Y. Gong, J. Zhang,
L. Zhan, L. Ma, Z. Fang, R. Vajtai, X. Wang, P. M. Ajayan, Adv.
Mater. 2013, 25, 2452-2456.
a) Z. Zhao, Y. Sun, F. Dong, Nanoscale 2015, 7, 15-37; b) P.
Niu, L. Zhang, G. Liu, H.-M. Cheng, Adv. Funct. Mater. 2012,
22, 4763-4770; c) X. Zhang, X. Xie, H. Wang, J. Zhang, B.
Pan, Y. Xie, J. Am. Chem. Soc. 2013, 135, 18-21; d) K. C.
Christoforidis, T. Montini, E. Bontempi, S. Zafeiratos, J. J. D.
Jaén, P. Fornasiero, Appl. Catal., B 2016, 187, 171-180.
a) X.-H. Li, M. Antonietti, Chemical Society Reviews 2013, 42,
6593-6604; b) Y.-Y. Cai, X.-H. Li, Y.-N. Zhang, X. Wei, K.-X.
Wang, J.-S. Chen, Angew. Chem., Int. Ed. 2013, 52, 11822-
11825; c) S. A. Ansari, M. H. Cho, Sustainable Energy Fuels
2017, 1, 510-519; d) L. Wu, Q. Li, C. Yang, X. Ma, Z. Zhang, X.
Cui, J. Mater. Chem. A 2018, 6, 20947-20955 ; e) Z. Xu, C.
Zhuang, Z. Zou, J. Wang, X. Xu, T. Peng, Nano Res. 2017, 10,
2193-2209.
Microscopy (TEM) using a Field Emission Gun JEOL2010 instrument
equipped operated at 200kV. This instrument has a spatial resolution at
Scherzer defocus conditions of 0.19 nm. Scanning Electron Microscopy
(SEM) images were recorded on a Field Emission Gun FEI Nova 450
electron microscope with an accelerating voltage of 5 kV. Energy
dispersive X-ray spectra (EDS) were obtained in the same microscope
using 20 kV.
[8]
Photocatalytic activity evaluation
The photocatalytic activity in H2 production of all catalysts was evaluated
using a solar simulator (LOT-Oriel) equipped with a 150 W Xe lamp and
an Atmospheric Edge Filter. A power density of approximately 25 and
180 mW cm−2 is achieved in the UV (300-400 nm) and in the visible
region (400-1000 nm), respectively. All reactions were contacted at flow
conditions (15 ml min-1 argon flow) over a period of 24 h[32]. The catalyst
(60 mg) was suspended into 80 ml of an aqueous ethanol solution (50
v.%) and Pt (0.1 wt.%) was photodeposited as co-catalyst by adding
Pt(NO3)2. Temperature was maintained at 25 oC by means of a cryostat
during the reaction. The reaction mixture was purged with argon (15 ml
min-1) for 40 min under steering to remove air prior to irradiation. An
Agilent 7890 gas chromatographer equipped with a Carboxen 1010
PLOT (Supelco, 30 m × 0.53 mm ID, 30 μm film) column using Ar as
carrier followed by a Thermal Conductivity Detector (TCD) and DB-
225ms column (J&W, 60 m × 0.32 mm ID, 20 μm film) using He as
carrier followed by a mass spectrometer (MS) HP 5975C was used to
detect the evolved gaseous products on-line.
[9]
[10]
[11]
[12]
S. J. A. Moniz, S. A. Shevlin, D. J. Martin, Z.-X. Guo, J. Tang,
Energy Environ. Sci. 2015, 8, 731-759.
a) Z. Tong, D. Yang, T. Xiao, Y. Tian, Z. Jiang, Chemical
Engineering Journal 2015, 260, 117-125; b) A. Crake, K. C.
Christoforidis, R. Godin, B. Moss, A. Kafizas, S. Zafeiratos, J.
R.Durrant, C. Petit, Appl. Catal., B 2019, 242, 369-378; c) B.
Chai, T. Peng, J. Mao, K. Li, L. Zan, Phys. Chem. Chem. Phys.
2012, 14, 16745-16752; d) X. Fan, T. Wang, B. Gao, H. Gong,
H. Xue, H. Guo, L. Song, W. Xia, X. Huang, J. He, Langmuir
2016, 32, 13322-13332.
K. C. Christoforidis, M. Fernandez-Garcia, Catal. Sci. Technol.
2016, 6, 1094-1105.
X. Lan, L. Wang, B. Zhang, B. Tian, J. Zhang, Catal. Today
2014, 224, 163-170.
M. Fittipaldi, V. Gombac, A. Gasparotto, C. Deiana, G. Adami,
D. Barreca, T. Montini, G. Martra, D. Gatteschi, P. Fornasiero,
ChemPhysChem 2011, 12, 2221-2224.
M. Szkoda, K. Siuzdak, A. Lisowska-Oleksiak, J. Karczewski,
J. Ryl, Electrochem. Comm. 2015, 60, 212-215.
K. C. Christoforidis, M. Melchionna, T. Montini, D. Papoulis, E.
Stathatos, S. Zafeiratos, E. Kordouli, P. Fornasiero, RSC Adv.
2016, 6, 86617-86626.
A. Crake, K. C. Christoforidis, R. Godin, B. Moss, A. Kafizas,
S. Zafeiratos, J. R. Durrant, C. Petit, Appl. Catal., B 2019, 242,
369-378.
M. Quesada-González, N. D. Boscher, C. J. Carmalt, I. P.
Parkin, ACS Appl. Mater. Interfaces 2016, 8, 25024-25029.
M. Bettinelli, V. Dallacasa, D. Falcomer, P. Fornasiero, V.
Gombac, T. Montini, L. Romano, A. Speghini, J. Hazard.
Mater. 2007, 146, 529-534.
a) X. Zhou, B. Jin, L. Li, F. Peng, H. Wang, H. Yu, Y. Fang, J.
Mater. Chem. 2012, 22, 17900-17905; b) X. Zhou, F. Peng, H.
Wang, H. Yu, Y. Fang, Chem. Commun. 2011, 47, 10323-
10325.
M. L. Guo, X. D. Zhang, C. T. Liang, Physica B 2011, 406,
3354-3358.
L. Mascaretti, V. Russo, G. Zoppellaro, A. Lucotti, C. S. Casari,
S. Kment, A. Naldoni, A. L. Bassi, J. Phys. Chem. C 2019, 123,
11292-11303.
Acknowledgements
Financial support from the French National Research Agency
(ANR) under the Program “Make Our Planet Great Again” (ANR-
18-MOPGA-0014) is fully acknowledged. T.M. acknowledges
financial support from MIUR PRIN2017 project 2017PBXPN4.
[13]
[14]
[15]
Keywords: boron, carbon nitride, hydrogen, photocatalysis,
[16]
[17]
photoreforming, TiO2.
[1]
a) K. C. Christoforidis, P. Fornasiero, ChemCatChem 2017, 9,
1523-1544; b) M. G. Walter, E. L. Warren, J. R. McKone, S. W.
Boettcher, Q. Mi, E. A. Santori, N. S. Lewis, Chem. Rev. 2010,
110, 6446-6473; c) K. C. Christoforidis, P. Fornasiero,
ChemCatChem 2019, 11, 368-382; d) X. Chen, S. Shen, L.
Guo, S. S. Mao, Chem. Rev. 2010, 110, 6503-6570.
Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han, C. Li, Chem. Rev.
2014, 114, 9987-10043.
a) K. C. Christoforidis, A. Iglesias-Juez, S. J. A. Figuero, M. D.
Michiel, M. A. Newton, M. Fernandez-Garcia, Catal. Sci.
Technol. 2013, 3, 626-634; b) L. Chen, X. Zhou, B. Jin, J. Luo,
X. Xu, L.Zhang, Y. Hong, Int. J. Hydrogen Energy 2016, 41,
7292-7300.
[18]
[19]
[20]
[2]
[3]
[21]
[4]
[5]
P. Carmichael, D. Hazafy, D. S. Bhachu, A. Mills, J. A. Darr, I.
P. Parkin, Phys. Chem. Chem. Phys. 2013, 15, 16788-16794.
a) V. Gombac, L. D. Rogatis, A. Gasparotto, G. Vicario, T.
Montini, D. Barreca, G. Balducci, P. Fornasiero, E. Tondello,
M. Graziania, Chem. Phys. 2007, 339, 111-123; b) S. In, A.
Orlov, R. Berg, F. García, S. Pedrosa-Jimenez, M. S. Tikhov,
D. S. Wright, R. M. Lambert, J. Am. Chem. Soc. 2007, 129,
13790-13791; c) M. Fittipaldi, V. Gombac, T. Montini, P.
Fornasiero, M. Graziani, Inorg. Chim. Acta 2008, 361, 3980-
3987.
[22]
[23]
[24]
a) T. Berger, M. Sterrer, O. Diwald, E. Knozinger,
ChemPhysChem 2005, 6, 2104-2112; b) T. Berger, M. Sterrer,
O. Diwald, E. Knozinger, D. Panayotov, T. L. Thompson, J. J.
T. Yates, J. Phys. Chem. B 2005, 109, 6061-6068; c) T.
Berger, M. Sterrer, S. Stankic, J. Bernardi, O. Diwald, E.
Knözinger, Mat. Sci. Eng. C Mater 2005, 25, 664-668; d) M.
Fittipaldi, D. Gatteschi, P. Fornasiero, Catal. Today 2013, 206,
2-11.
[6]
[7]
N. Lu, X. Quan, J. Li, S. Chen, H. Yu, G. Chen, J. Phys. Chem.
C 2011, 111, 11836-11842.
a) J. Zhang, Y. Chen, X. Wang, Energy Environ. Sci. 2015, 8,
3092-3108; b) K. C. Christoforidis, Z. Syrgiannis, V. L. Parola,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
[25]
R. Scotti, I. R. Bellobono, C. Canevali, C. Cannas, M. Catti, M.
D’Arienzo, A. Musinu, S. Polizzi, M. Sommariva, A. Testino, F.
Morazzoni, Chem. Mater. 2008, 20, 4051-4061.
J. Biedrzycki, S. Livraghi, E. Giamello, S. Agnoli, G. Granozzi,
J. Phys. Chem. C 2014, 118, 8462.
[29]
a) D. O. Scanlon, C. W. Dunnill, J. Buckeridge, S. A. Shevlin,
A. J. Logsdail, S. M. Woodley, C. R. A. Catlow, M. J. Powell, R.
G. Palgrave, I. P. Parkin, G. W. Watson, T. W. Keal, P.
Sherwood, A. Walsh, A. A. Sokol, Nat. Mater 2013, 12, 798-
801; b) H.-Z. Wu, L.-M. Liu, S.-J. Zhao, Phys. Chem. Chem.
Phys. 2014, 16, 3299-3304; cY. Yang, Z. Zeng, G. Zeng, D.
Huang, R. Xiao, C. Zhang, C. Zhou, W. Xiong, W. Wang, M.
Cheng, W. Xue, H. Guo, X. Tang, D. He, Appl. Catal., B 2019,
258, 117956.
Z. Wang, Z. Ni, D. Qiu, T. Chen, G. Tao, P. Yang, J. Anal. At.
Spectrom. 2004, 19, 273-276.
S. Stoll, A. Schweiger, J. Magn. Reson. 2006, 178, 42-55.
G. Carraro, C. Maccato, A. Gasparotto, T. Montini, S. Turner,
O. I. Lebedev, V. Gombac, G. Adami, G. V. Tendeloo, D.
Barreca, P. Fornasiero, Adv. Funct. Mater. 2014, 24, 372-378.
[26]
[27]
a) C. P. Kumar, N. O. Gopal, T. C. Wang, M.-S. Wong, S. C.
Ke, J. Phys. Chem. B 2006, 110, 5223-5229; b) S. O.
Baumann, M. J. Elser, M. Auer, J. Bernardi, N. Husing, O.
Diwald, Langmuir 2011, 27, 1946-1953; c) M. D’Arienzo, J.
Carbajo, A. Bahamonde, M. Crippa, S. Polizzi, R. Scotti, L.
Wahba, F. Morazzoni, J. Am. Chem. Soc. 2011, 133, 17652-
17661; d) H.-H. Lo, N. O. Gopal, S.-C. Sheu, S.-C. Ke, J. Phys.
Chem. C 2014, 118, 2877-2884.
a) A. E. Giannakas, E. Seristatidou, Y. Deligiannakis, I.
Konstantinou, Appl. Catal., B 2013, 132-133, 460-468; b) S.
Livraghi, A. Votta, M. C. Paganini, E. Giamello, Chem.
Commun. 2005, 498-500; c) D. Dvoranová, M. Mazúr, I.
Papailias, T. Giannakopoulou, C. Trapalis, V. Brezová,
Catalysts 2018, 8, 47.
[30]
[31]
[32]
[28]
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901703
ChemCatChem
FULL PAPER
FULL PAPER
Photocatalytic hydrogen production is
significantly improved via doping TiO2
with boron and coupling with carbon
nitride. A strong interaction between
TiO2 and carbon nitride is formed
resulting to the successful formation of
a heterojunction. Both doping and the
formation of a heterojunction improve
the electronic properties of the
catalysts. The enhanced photoactivity
is ascribed to the improved
Konstantinos C. Christoforidis*, Tiziano
Montini, Maria Fittipaldi, Juan Josè
Delgado Jaén and Paolo Fornasiero*
TiO2 + B-doping + CN Junction
B-TiO2/CN
B-TiO2
2.0
Page No. – Page No.
1.5
1.0
0.5
0.0
TiO2/CN
Title
TiO2
TiO2
TiO2/CN B-TiO2 B-TiO2/CN
Catalysts
abundance of photogenerated
electron.
This article is protected by copyright. All rights reserved.