P- and F-co-doped Carbon Nitride Nanocatalysts for Photocatalytic CO2 Reduction and Thermocatalytic Furanics Synthesis from Sugars

Article abstract of DOI:10.1002/cssc.202001172

A new P- and F-co-doped amorphous carbon nitride (PFCN) has been synthesized via sol-gel-mediated thermal condensation of dicyandiamide. Such synthesized P- and F-co-doped carbon nitride displayed a well-defined mesoporous nanostructure and enhanced visible light absorption region up to infrared with higher BET surface area of 260.93 m² g^?1; the highest recorded value for phosphorus-doped carbon nitride materials. Moreover, the formation mechanism is delineated and the role of templates was found to be essential not only in increasing the surface area but also in facilitating the co-doping of P and F atoms. Co-doping helped to narrow the optical band gap to 1.8 eV, thus enabling an excellent photocatalytic activity for the aqueous reduction of carbon dioxide into methanol under visible-light irradiation, which is fifteen times higher (119.56 μmol g^?1 h^?1) than the bare carbon nitride. P doping introduced Br?nsted acidity into the material, turning it into an acid-base bifunctional catalyst. Consequently, the material was also investigated for the thermal conversion of common carbohydrates into furanics.

Full text of DOI:10.1002/cssc.202001172

Chemistry–Sustainability–Energy–Materials
Accepted Article
Title: Mesoporous P- and F-co-doped Amorphous Carbon Nitride:
Nanocatalysts for Photocatalytic CO2 Reduction and
Thermocatalytic Furanics Synthesis from Sugars
Authors: Subodh Kumar, Manoj B. Gawande, Josef Kopp, Štěpán
Kment, Rajender S. Varma, and Radek Zboril
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: ChemSusChem 10.1002/cssc.202001172
Link to VoR: https://doi.org/10.1002/cssc.202001172
0
1/2020

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
P- and F-co-doped Carbon Nitride Nanocatalysts for
Photocatalytic CO
2
Reduction and Thermocatalytic Furanics
Synthesis from Sugars
[
a]
[a, b]
Josef Kopp, [a] Stepan Kment, [a] Rajender S. Varma, [a] and
Subodh Kumar, Manoj B. Gawande,*
Radek Zbořil*[a]
[
[
a]
b]
Dr. S. Kumar, Dr. M. B. Gawande, Josef Kopp, Prof. R. S. Varma, Prof. R. Zbořil
Regional Centre of Advanced Technologies and Materials,
Palacký University Olomouc Šlechtitelů 27, 783 71 Olomouc, Czech Republic
E-mail: manoj.gawande@upol.cz; radek.zboril@upol.cz
Prof. Manoj B. Gawande
Institute of Chemical Technology, Mumbai-Marathwada Campus Jalna, Maharashtra 431213, India.
Supporting information for this article is given via a link at the end of the document.
Abstract: A novel P- and F-co-doped amorphous carbon nitride
PFCN) has been synthesized via sol-gel mediated thermal
approaches, such as doping,[4] exfoliation,[5] templating,[6] and
homogeneous amorphization,[7] have been implemented,
although homogeneous amorphization appears to be the most
effective. Although these results are important with respect to
designing a material with defects to lower the band gap, they do
not provide high surface area. Moreover, extended visible light
absorption beyond 500 nm does not contribute to the
(
condensation of dicyandiamide. Such synthesized P- and F-doped
carbon nitride displayed a well-defined mesoporous nanostructure
and enhanced visible light absorption region up to infrared with higher
2
–1
BET surface area of 260.93 m g ; the highest recorded value for
phosphorus doped carbon nitride materials. Moreover, the formation
mechanism is delineated and the role of templates was found to be
essential not only in enhancing the surface area but also in facilitating
the co-doping of P and F atoms. Co-doping helped to narrow the
optical band gap to 1.8 eV, thus enabling an excellent photocatalytic
activity for the aqueous reduction of carbon dioxide into methanol
under visible light irradiation, which is fifteen times higher (119.56
[
8]
photocatalytic activity of the material. Hence, in order to exploit
the extended light absorption, it is necessary to create the midgap
energy states by heteroatom doping,[9] which is also important for
the thermal catalysis, as carbon nitride has been exploited as a
basic catalyst[
2a,6,10]
or support owing to nitrogen framework in
[11]
the polymeric carbon nanostructure. Therefore, in order to explore
the catalytic activity (photo and thermal), suitable chemical doping
(metal and nonmetal) is one of the effective strategies to
manipulate electronic and chemical nature of the catalyst and
concomitantly generating high surface area of the material. Until
now, various heteroatom-doped carbon nitrides have been
synthesized in diverse morphological forms via many techniques;
–1
–1
µmol g h ) than the bare carbon nitride. P doping introduced the
Brønsted acidity into the material, turning it into an acid-base bi-
functional catalyst. Consequently, the material was also investigated
for the thermal conversion of common carbohydrates into furanics.
[
2b,4,12]
every dopant bestows some typical characteristic
Introduction
properties on the material. For instance, phosphorus doping not
only provides the reduction in band gap but also imparts the
Graphitic carbon nitride, a layered polymeric semiconductor
material, is a suitable catalyst candidate due to its advantageous
features including abundance, nontoxicity, stability, and chemical
tenability.[1] It is generally prepared by thermal condensation of
nitrogen-rich precursors encompassing C-N bonded building
blocks such as melamine, urea and dicyandiamide under inert or
air atmosphere. However, the degree of condensation, surface
properties and the yield of the final material vary with the
precursor and conditions deployed. Carbon nitrides have shown
promising applications in many fields including catalysis (thermal,
acidity to carbon nitride, turning it into a metal-free acid-base bi-
[13]
functional catalyst.
On the other hand, fluorine doping can
improve the optical absorption (band gap reduction) and
photocatalytic activity by facilitating the migration and separation
of photogenerated charge carriers.[14] We envisioned that co-
doping may be a promising approach and this unique combination
of P- and F-co-doped carbon nitride with high surface area could
generate an exceptionally good thermal as well as a photocatalyst.
Interestingly, despite the reports on co-doped carbon nitrides with
[15]
assorted combinations (B/F, B/S, P/O, S/O, and P/S),
to the
photo-, and electrophotocatalysis).[2] However, pristine g-C
N still
3 4
best of our knowledge, there is no report available on the powder
synthesis of P- and F-co-doped amorphous carbon nitride until
suffers from its restricted visible-light harvesting capacity, ready
recombination of charge carriers, and very low surface area,
which is prerequisite for the efficient catalysis.[3] In order to
develop the carbon nitride material with extended visible light
absorption and higher surface area, various modification
now. Zhang et al. synthesized the P-doped analogue using the
[16]
6
same precursor (BmimPF ) which lacked F doping.
The growing demand for energy and continuous depletion in fossil
fuel reserves insinuate an urgent need to substitute the
1
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
petroleum-based feedstock with renewable resources of energy,
the major renewable energy resources being biomass,
hydropower, geothermal, wind, and solar. Moreover, abundant
dicyandiamide might have also weakened the in-plane hydrogen
bonding between the strands of polymeric melon units during
calcination, imparting amorphous nature to the final material.
These results highlight the significant role of the templates used
to produce a material with distinctive textural properties. The
above-mentioned outcomes indicate that using
carbon dioxide,
implications, is a safe and inexpensive C
a
greenhouse gas with climate change
source that needs to be
1
viewed as an inexhaustible renewable resource. Amongst these,
biomass and carbon dioxide can exclusively generate liquid fuels
and chemicals, thus offering a potential substitute for fossil fuels.
In this context, methanol and furanic compounds (5-
hydroxymethyl-2-furaldehyde (HMF) and furfural) are quite
important and can be produced photochemically and thermally
from the carbon dioxide and carbohydrates (glucose/fructose and
xylose), respectively.[17]
We report here, for the first time, a novel synthetic protocol that
provides P-, F-co-doped mesoporous carbon nitride (PFCN) with
high surface area and lower band gap. Inspired by its extended
visible light absorption and acid-base bi-functional nature, we
explored this new class of catalyst for the aqueous photocatalytic
2
reduction of CO into methanol and conducted thermal catalytic
conversion of carbohydrates (glucose and xylose) into high-value
furanics compounds (HMF and furfural).
Scheme 1. Stepwise synthesis of porous phosphorus- and fluorine-co-doped
carbon nitride.
Results and Discussion
the combination of TEOS and CTAB templates successfully leads
to co-doping of P and F and higher surface area with a significant
change in the matrix network of PFCN.
SEM and TEM images of PFCN revealed a rough, agglomerated
and crumpled structure with 3–20 nm pores on its surface
indicating that the multilayered sheets are mesoporous in nature
High surface area P- and F-co-doped carbon nitride (PFCN) was
synthesized with enhanced optical absorption via a simple
thermal condensation of dicyandiamide and 1-butyl-3-
6
methylimidazolium hexafluorophosphate (BmimPF ) in presence
of hexadecyltrimethylammonium bromide (CTAB) and tetraethyl
orthosilicate (TEOS) at 550 °C. Templates roles were investigated
by the controlled polycondensation of dicyandiamide and
(
Figure 1a; Figure S2a–c). The HR-TEM image displays the
interpenetrating carbon nitride networks with long-range disorder
suggesting amorphous nature of the material (Figure 1b; Figure
S2d), which is further confirmed by the SAED pattern (Figure S2e).
However, local ordering with lateral dimension of 1 to 3 nm and
lattice spacing of 3–5 A° can also be seen along with the curly
ribbons. EDX spectra and corresponding HAADF images of
PFCN confirmed the presence of carbon, nitrogen, phosphorus,
and fluorine elements and their uniform distribution all over the
surface (Figure 1c–f; Figure S2f and S3). Carbon nitride has less
folded nanosheets with comparatively smooth surface, which
clearly indicates that phosphorus doping changes the structural
features of the material while different templates are used (Figure
S4a).
BmimPF
of templates to synthetize PCN-C, PCN-T, PFCN, and PCN,
respectively, under the same reaction conditions.
Thermogravimetric analysis (TGA) was used to monitor the
progress (Figure S1). In addition, poly-condensation of pure
dicyandiamide and decomposition of CTAB and TEOS were also
recorded. CTAB template seemed to enhance the poly-
6
using CTAB, TEOS, CTAB+TEOS, and in the absence
condensation of dicyandiamide and BmimPF
and lose more weight at the same temperature. It might be
because of the ion exchange[18] between CTAB and BmimPF
6
mixture (PCN-C)
6
,
which leads to the formation of less thermally stable [Bmim]Br‾
species[19] during the micelle formation. These micelles thermally
decompose into hydrocarbons and other volatile intermediates,
thus allowing to generate pores to produce mesoporous carbon
nitride. However, the PCN-C sample showed comparatively low
surface area and no fluorine doping (Table S1 and S2). It might
be because of the higher decomposition temperature of
dicyandiamide than micelles; consequently, the pores generated
were occupied with uncondensed dicyandiamide. However, in
case of the PFCN sample, the TEOS polymerized around the
micelles during the sol-gel process becomes transformed into
silica matrix on thermal treatment. The silica matrix not only
prevents the structural pores to collapse but concurrently traps
6
the PF ‾ species to react with amino groups of uncondensed
carbon nitride precursor to join the C-N framework; the removal of
the silica template by subsequent washing with ammonium
fluoride solution generates the silica-free, P-, F-co-doped
materials with high surface area (Scheme 1). Moreover, the
micelles surrounded by the TEOS throughout the matrix of
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
Figure 1. a) TEM image; b) HR-TEM image; c) HAADF image; and d–f) STEM
elemental mapping images of PFCN.
In addition to XRD, Raman spectra were obtained to recognize
the templates-dependent structural changes in the skeleton of
carbon nitride after doping with P and F; mainly, two bands
appeared at around 1329 cm–1 (the D band) and 1611 cm (the
G band) for the synthesized material (Figure S5c). The shape,
–1
FT-IR spectroscopy ascertained the functional groups present in
the synthesized material (Figure S5a). A peak at 807 cm–1 as well
as a series of vibrations ranging between 1200 and 1650 cm–1
were observed in the spectrum of the material, which can be
attributed to the characteristic bending vibration of tri-s-triazine
units and stretching vibrations of C-N heterocycles of
D G
position, and relative intensity of these D and G bands (I /I ) are
used to estimate the structural changes including surface
disorder. However, weak intensity of the G bands in the Raman
spectrum of PFCN compared to the native carbon nitride alone
suggests that the use of templates significantly altered the
structural arrangement in the form of defects, thus increasing the
[
20]
dicyandiamide, respectively.
Compared to the CN sample,
PFCN shows no obvious change in the peak position but the
merging of peaks presumably indicates the overlapping stretching
vibrations corresponding to C=C and C-F. Moreover, there are
two new peaks of P-related group at 950 and 1055 cm–1 which
can be assigned to P-N stretching mode and P=O, respectively
that appears exclusively for the PFCN sample, pointing to the high
I
D G
ratio from 1.35 to 1.44 for CN and PFCN, respectively. CN
/I
and PFCN show nitrogen-doped carbon-type Raman spectra with
D bands, which suggests a multilayer structure. Moreover,
2
broadening of peaks indicates an increase in the amorphous
nature of the material.
The main objective of XPS analysis is to interpret the atomic
concentration and chemical interactions of C, O, N, P, and F
atoms in the synthesized materials (CN and PFCN). A survey
scan of CN showed mainly three peaks centered at 288.31,
[
21]
P doping. In addition, broadening of the band in the range of
200–3500 cm–1 indicates the presence of absorbed water
3
molecule and uncondensed/free N-H stretching of the amine
groups.
The BET analysis was executed to learn about the textural
3
98.65, and 531.72 eV corresponding to C1s, N1s, and O1s,
respectively. The appearance of two new peaks in PFCN
centered at 133.51 and 686.31 eV, corresponding to P2p and F1s
atom, confirms that phosphorus and fluorine doping was
successful (Figure S5d); notably, fluorine doping along with the P
has been elusive in previous reports even using the same
precursor. This may be because of the lack of reaction between
fluorine species with the melem unit in view of the higher
2
properties of the synthesized material, and the N adsorption-
desorption isotherms with pore size distribution (Figure 2a). The
synthesized material showed a type IV nitrogen adsorption-
desorption isotherm with a H3-shaped hysteresis loop signifying
the well-defined mesoporous nanostructure with slit-like pores,
derived from the overlapping and stacking of nanosheets. N
2
absorption at low pressure is related to the filling of micropores
indicating the presence of a very weak microporosity. However,
the contribution of micropore surface area to the total surface area
is found to be zero, calculated by the T-plot method. The isotherm
of PFCN presents high absorption at a high relative pressure
decomposition temperature of BmimPF
removal of PF ‾ at higher temperature presumably leads to no
doping of F. In our case, BmimPF could have taken part in the
6
ionic liquids. The
6
6
micelles formation with CTAB and generated less thermally stable
¯
[
Bmim]Br species; the formed micelles are further surrounded by
(
0
P/P ), which suggests the existence of large mesopores.
the silica due to the polymerization of TEOS during sol-gel
reaction, thus limiting the escape of fluorine species unreacted. In
order to study the bonding states of P and F atoms, high-
resolution C1s, N1s, O1s, F1s, and P2p XPS spectra were
recorded and de-convoluted (Figure 2c,d; Figure S6a–f).
Furthermore, the above-mentioned observations are also
supported by the respective pore size distribution results, which
may be due to the random pore formation induced by the
decomposition of CTAB and subsequent removal of silica
template; a very broad pore size distribution in the 3–30 nm region
is observed (Figure S5b). These results are in agreement with the
TEM results and confirm the purely mesoporous nature of the
material. PFCN showed higher surface area of 260.93 m2 g ,
which is the highest value, reported so far for phosphorus- and
fluorine-doped carbon nitride materials.
–1
Deviations in the crystal structure and phase of the synthesized
materials (CN and PFCN) were revealed by the XRD technique
(
3
Figure 2b). CN exhibited mainly two reflections at 15.23° and
1.9°, while the PFCN sample showed only one broad reflection
at around 31.4°. The reflections at 2θ value of 15.23° and 31.9°
correspond to the [100] and [002] graphitic planes arising from the
in-plane structural packing motif of the heptazine units and
interlayer stacking of conjugated aromatic systems, respectively.
The disappearance and broadening of the peaks at 15.23° and
31.4°, respectively, in PFCN indicate a change in the graphitic
structure or absence of a long-range order in the atomic
[
7a]
arrangements of the materials after doping. In comparison to
CN, using the combination of CTAB and TEOS templates to
synthetize the PFCN sample leads to the shift in [002] reflection
towards lower 2θ value, indicating an increase in the interlayer
distance after P doping, which is attributed to the larger radius of
the doped P atom (100 pm) than that of the replaced C atom (70
Figure 2. a) N
spectrum; c) P2p; d) F1s XPS spectra of PFCN.
2
Abs-Desorption isotherm with pore size distribution; b) XRD
[
9]
pm).
3
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
The high-resolution C1s spectrum of PFCN sample mainly
revealed two wide peaks attributed to different types of chemical
interaction of carbon. These peaks can be further deconvoluted
into six associate peaks at 284.76, 285.89, 286.44, 287.34,
P, respectively. It is important to note that no peak for C-P bond
is observed in neither P2p nor C1s high-resolution, indicating that
P replaced the carbon in carbon-nitride skeleton forming P-N
bond. However, the marginal shifting in the peak positions in
PFCN spectra is due to the doping. Based on the compositional
information of C, O, N, P, and F, we observed P and F doping
level at 4.3 and 1.4 atom % for PFCN (Table S2).
288.73, and 289.40 eV, corresponding to the C-C/C=C, C-CF, C-
O/C-N, N-C=N, C-F, and O=C-O functional groups, respectively.
An actual indication of fluorine doping of carbon nitride was
deduced by the presence of one peak with shoulder in F1s core
level spectra, which on deconvolution splits into three peaks
centered at 686.16, 687.13, and 688.41 eV. The first peak
represents the semi-ionic C-F, while the other two peaks
Furthermore, solid-state magic angle spinning nuclear magnetic
resonance (MASNMR) analyses were performed to confirm the
chemical environment of the elements in the synthesized material
network. 13C, 15N, 31P, 19F, and H MAS and CPMAS NMR spectra
1
represent the covalently bound fluorine to carbon (C-F
X
). It is
of PFCN material are shown in Figure 3 and Figure S7. 13
CPMASNMR displays four peaks that suggest the presence of
four types of carbon (Figure S7a). The peaks at 160.0 (C ) and
168.1 (C ) ppm are the characteristic peaks for carbon atoms in
carbon nitrides due to the C-NH
very low intensity at 98.0 (C ) ppm occurs due to the fluorine
doping, confirming the F-C covalent bonding and a little shoulder
at 148.9 (C ) ppm corresponds to the graphitic carbon.
C
worth mentioning here that the intensity of the peak attributable to
C-C/C=C in high-resolution C1s XPS spectrum of CN is very low,
while in the PFCN samples is higher, demonstrating a significant
increase in the conjugation forming C=C bonds. This may
3
4
[
23]
2
or C-NH- bond. A peak with
[
22]
possibly be due to the carbonizing of templates. Moreover,
appearance of a new peak at 399.69 eV in N1s XPS spectrum
1
2
Additionally, 19F spectrum shows a wide peak at –147.6 ppm,
which corresponds to the C-F units, excluding the probability of
absorbed fluorine. Moreover, 15N CPMAS spectrum exhibits five
peaks between 0–300 ppm, which correspond to five different
types of nitrogen environments. The peaks at 97.25, 135.14,
2 3
153.08, and 204.98 ppm are assigned to the –NH , -NH-, N-C ,
and –N=C groups, respectively, and are consistent with the
[
23]
existing reports. However, a new peak arises at 48.54 ppm,
which could be assigned to N-P bond. 31P CPMASNMR spectrum
further confirms the successful phosphorus doping displaying two
intense peaks at –3.63 and 6.40 ppm, indicating that there are
possibly two types of P doping (P-OH and P=O) in the final
[
13]
material. We presume that all the doped P atoms replace the
corner and bay C atoms during the polymerization of carbon
1
nitride skeleton. Significantly, H MASNMR evidence the actual
association of protons with the respective nuclei; 1H MASNMR
spectrum clearly shows four peaks at 4.1, 5.9, 8.8, and 11.3 ppm
represented as H
peaks at 4.1 and 5.9 ppm correspond to the proton associated
without hydrogen-bonded -NH and –NH- groups, respectively,
while an intense broad peak at 8.83 ppm indicates proton
associated with the hydrogen-bonded –NH groups. A new peak
at 11.38 ppm could be assigned to the acidic proton (P-OH). In
1 3 4
, H2, H , and H , respectively (Figure 3c). The
2
2
1
31
order to support our assignment, we also performed H- P
heteronuclear correlation (HETCOR) 2D MASNMR (Figure 3d).
There is a clear cross peak between P
of H nuclei, suggesting the presence of phosphorus in the form
of P-OH. However, no cross peak was observed for the peak P ,
1
2
peak of 31P and H
4
peak
Figure 3. Solid-state NMR spectra of PFCN sample. a) 1P CPMAS; b) 19F; c)
3
1
1
H nuclei and d) H-31P heteronuclear correlation (HETCOR) 2D NMR.
1
which indicates deprotonated phosphorus probably in the form of
P=O. Moreover, the existence of characteristic peaks in the 13C
and 15N solid-state NMR confirms that the melem unit structure
remained intact after the phosphorus doping.
suggests the formation of N-P bonds in the PFCN sample after
doping. We identified only one strong broad peak at 133.42 eV
position in P2p spectrum, which represents N=P-N bonds
formation and can be deconvoluted into two signals, assigned to
two different chemical environments of phosphorus atom in N=P-
N. The peaks positioned at 133.29 and 134.42 eV represent the
phosphorus atom forming phosphonic acid moiety (P-OH) and
phosphates (P=O), respectively. Additional evidence for this
comes from the O1s XPS spectrum; it displays a broad peak with
a shoulder, which can be deconvoluted into five different peaks.
Peaks at 530.87, 532.09, and 534.1 eV are due to the absorbed
oxygen, -N-C-O, and absorbed water, respectively. The additional
two peaks at 530.31 and 533.08 eV represent the O=P and HO-
Optical properties of PFCN were studied by the UV-Vis diffuse
reflectance (DRS) spectra as shown in Figure 4a. Based on the
spectra, the band gaps of the samples were acquired from the
2
transformed Kubelka–Munk equation (Figure S8a): [F(R)·hv] =
2
A(hv – Eg), where F(R) = α = (1 – R) /2R, R is the percentage of
reflected light, hv is the incident photon energy, Eg is the band
gap energy and A is constant (value depends on the transition
probability).[24] In contrast to CN, PFCN shows a red shift in the
absorption edge, which suggests the narrowing band gap. In fact,
the co-doping narrows the band gap of PFCN to 1.83 eV
4
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
compared with CN (2.83 eV), enhancing the light-harvesting
capability of PFCN in the visible-light region (700 nm > λ > 420
nm). However, PFCN displays a very strong tail absorption in the
visible-light region; the tail, also known as Urbach tail, appeared
due to the photoexcitation from the valence band (VB) to the
midgap electronic states. In other words, the photogenerated
electron can easily jump from the VB to the midgap state or from
the midgap states to the conduction band (CB). These midgap
states are generated as a result of P doping by the hybridization
of C 2s2p, N 2s2p and P 3s3p and appear below the conduction
band (CB) of PFCN (Figure 4d). In order to recognize the
narrowing of bandgap for PFCN material, XPS valence band
spectra of CN and PFCN were studied and the position of valence
bands were determined by extrapolating the edges of spectra
source for 24 h. The highest methanol yield obtained was 119.56
μmol h g , which was higher in comparison with previously
−
1 −1
reported heteroatom doped and undoped carbon nitride catalyzed
photoreduction of CO
2
to methanol.[ There are two possible
26]
reasons for PFCN’s higher photocatalytic activity: lower band gap
and more existing defects in the structure.[27] Lower band gap
corresponds to broadened light response, so that electrons could
3 4
be generated upon irradiation. Incorporation of F in the C N
2
network can modify its electronic structure by converting C sp to
3
[14]
C sp hybridization thus narrowing the band gap. Moreover, F
atom has lone-pair electrons and C has a large π-conjugated
3
N
4
system. The incorporation of F into the C/N scaffold could
promote the formation of p-π conjugation, which might increases
the charge density on the HOMO thus reducing the photocarrier
transfer barrier[28] and enabling the enhanced photocatalytic
(
Figure 4b, c). CN and PFCN showed band tail with its end at 1.72
and 1.47 eV respectively, which are higher than their respective
dominant edges value. This could be due to the amorphous
nature and presence of dangling bonds.[7a] These values imply
that CB edge is shifted towards VB by 0.75 eV and VB shifts
towards the CB by 0.25 eV, thus narrowing the band gap by 1.0
eV. In order to reveal the position of midgap states in the band
gap, the transition energy (Et) from VB to the midgap states for
2
activity of PFCN for the aqueous reduction of CO to methanol.
On the other hand, defects (structural and chemical) facilitate the
charge separation. For this, photoluminescence (PL) emission
analyses of PFCN and CN samples were performed under an
excitation wavelength of 390 nm to study the recombination of
photogenerated charge carriers. Evidently, there is no peak for
PFCN unlike the CN sample at 440 nm (Figure S8b) in the PL
spectrum, which indicates inhibition of charge carrier
recombination due to their trapping in the defects associated with
long-range disorder.[7a] Moreover, midgap states generated on P
doping also act as an electron reservoir to trap the unstable
electrons thus inhibiting the direct recombination of electron-holes
and providing a shorter path length for electron excitation to the
CB; the electrons residing at the midgap states are known to have
PFCN was calculated to be 1.52 eV by the Kubelk–Munk method
25]
(
Figure S8a).[ The exact position of midgap states for PFCN
was determined to be at −0.05 V vs. NHE, which is higher than
the [2H+/H (0.0 V), CO /CH OH (0.03V) and H O/O (1.23 V)]
and thus still satisfying the thermodynamic requirements (Figure
d).
2
2
3
2
2
4
a
prolonged lifetime as compared to free excitons.[28b,29]
Consequently, PFCN can provide more charge carriers for the
photocatalytic reduction of CO
2
in comparison with CN. PFCN
-
absorbed the visible light efficiently and generated electron (e )
+
and holes pair (h ). Holes oxidize the water molecules to give
+
.
protons (H ) and hydroxide radicals ( OH). The hydroxide radicals
react further with water molecules to produce oxygen (O ) and
protons. These protons and CO adsorbed on the catalyst surface
2
2
was then reduced to methanol by photo-generated electrons.
Shifting the conduction band towards redox potential of
CO /CH OH limits the formation of liquid side products such as
2 3
HCOOH and HCHO. We believe that these studies will help to
design other possible catalysts for the photocatalytic applications
in the future.
Since PFCN catalyst contains Brønsted acid sites as revealed by
1
the XPS and H SSNMR analysis, we further explored this new
class of catalyst for the thermal catalytic conversion of sugars
(
glucose and xylose) into high-value furanic compounds (HMF
and furfural). In a typical experiment, glucose or xylose (0.2 mmol)
and PFCN catalyst (50 mg) were put into water/THF mixture (3.0
mL) and the resulting mixture was heated at 130 °C in a sealed
pressure reactor under nitrogen atmosphere for 5 h. The highest
yields of the HMF and furfural under above conditions were found
to be 68 and 91%, respectively (Scheme 2). As suggested in the
literature,[30] the presence of both acidic and basic sites facilitates
the conversion of carbohydrates into the furanics. Lewis bases
are particularly important for the isomerization step, whereas
Brønsted acidity is required for the dehydration step, which could
be the reason for enhancement of the reaction rate. In analogy
with the existing reports, we assume that the reaction probably
involves the isomerization of glucose and xylose to fructose and
xylose, followed by their dehydration to HMF and furfural,
Figure 4. a) UV-vis spectra; b) and c) valence band XPS spectra; d) Band
alignments of CN and PFCN.
This result encouraged us to test its photocatalytic prowess for
the reduction of carbon dioxide under visible light irradiation at
room temperature using blue LED light source (Kessil 150N, 34W)
2
with the light density of 22 mW/cm . Initially, glass tube was
charged with water (20 mL), PFCN (100 mg) and then carbon
dioxide was purged for 30 min in the mixture with continuous
stirring. The tube was then sealed, connected with a carbon
dioxide balloon through the needle and exposed to the light
5
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
respectively. Furthermore, to ascertain the acidity generation on
phosphorus doping, we determined the total acid content of the
PFCN material by titration,[31] which is found to be 931 µmol g–1
improvement of sustainable protocols for fine chemicals from bio-
based resources, particularly for large-scale reactions.
(
see associated text in Supporting Information).
Acknowledgements
The authors gratefully acknowledge the support by the
Operational Program Research, Development and Education -
European Regional
Development
Fund,
project
no.
CZ.02.1.01/0.0/0.0/15_003/0000416 of the Ministry of Education,
Youth and Sports of the Czech Republic. The authors thank to Dr.
V. Ranc for Raman analysis, Ms. J. Stráská and Ms. P. Bazgerová
for SEM/TEM analysis, Dr. Juri Ugolotti and Mr. I. Popa for solid-
state NMR analysis, and Dr. S. Kalytchuk for PL analysis. The
authors also thank Mr. O. Tomanec and Mr. M. Petr for HRTEM-
elemental mapping data and HR-XPS measurements,
respectively.
Scheme 2. PFCN-catalyzed synthesis of furanics.
Recycling of the catalyst is a critical step toward developing a
greener and sustainable catalytic system for chemical
transformations. After completion of the respective substrate
reaction, the catalyst was separated by filtration, washed with
water, ethanol and dried at 80 °C. The recovered catalyst was
then regenerated by stirring in 1M HCl solution followed by
heating at 400 °C for one hour before next run. The recovered
catalyst was reused for five subsequent runs. In all experiments,
the yield of the products remained almost the same, indicating
that the developed catalyst can be recycled efficiently without
significant loss in catalytic activity. Moreover, the spent catalyst
was also analyzed by TEM and FT-IR to see any possible
morphological changes and functionality of the catalysts (Figure
S9); barely any noticeable change was discerned in peak features,
indicating that the catalyst is quite stable and truly heterogeneous
in nature during the reaction. The comparison of textural and
electronic properties, catalytic activity of dual template-assisted
synthesized PFCN material with the earlier documented methods
is shown in Tables S3 and S4, Supporting Information.
Keywords: Amorphous carbon nitride • Bi-functional catalysts •
Biomass upgradation • P- and F-co-doping • Photocatalytic
reduction of carbon dioxide
[
1]
a) W. Iqbal, B. Yang, X. Zhao, M. Rauf, M. Waqas, Y. Gong, J. Zhang,
Y. Mao, Catal. Sci. Technol. 2018, 8, 4576-4599; b) A. Thomas, A.
Fischer, F. Goettmann, M. Antonietti, J.-O. Müller, R. Schlögl, J. M.
Carlsson, J. Mater. Chem. 2008, 18, 4893-4908.
[
2]
a) Y. Gong, M. Li, H. Li, Y. Wang, Green Chem. 2015, 17, 715-736; b)
W.-J. Ong, L.-L. Tan, Y. H. Ng, S.-T. Yong, S.-P. Chai, Chem. Rev. 2016,
116, 7159-7329; c) J. Liu, H. Wang, M. Antonietti, Chem. Soc. Rev. 2016,
45, 2308-2326; d) L. Zhou, H. Zhang, H. Sun, S. Liu, M. O. Tade, S.
Wang, W. Jin, Catal. Sci. Technol. 2016, 6, 7002-7023; e) N. Mansor, T.
S. Miller, I. Dedigama, A. B. Jorge, J. Jia, V. Brázdová, C. Mattevi, C.
Gibbs, D. Hodgson, P. R. Shearing, Electrochim. Acta 2016, 222, 44-57;
f) A. Wang, C. Wang, L. Fu, W. Wong-Ng, Y. Lan, Nano-micro lett. 2017,
9, 47; g) F. K. Kessler, Y. Zheng, D. Schwarz, C. Merschjann, W. Schnick,
X. Wang, M. J. Bojdys, Nature Rev. Mater. 2017, 2, 17030; h) S. Cao, J.
Low, J. Yu, M. Jaroniec, Adv. Mater. 2015, 27, 2150–2176.
[
3]
a) J. Zhu, P. Xiao, H. Li, S. A. Carabineiro, ACS Appl. Mater. Inter. 2014,
6
, 16449-16465; bS. Cao, J. Low, J. Yu, M. Jaroniec, Adv. Mater. 2015,
7, 2150-2176.
2
Conclusion
[
[
4]
5]
L. Jiang, X. Yuan, Y. Pan, J. Liang, G. Zeng, Z. Wu, H. Wang, Appl. Cata.
B. Environ. 2017, 217, 388-406.
We have synthesized P- and F-co-doped porous carbon nitride
a) H. Xu, J. Yan, X. She, L. Xu, J. Xia, Y. Xu, Y. Song, L. Huang, H. Li,
Nanoscale 2014, 6, 1406-1415; b) X. She, H. Xu, Y. Xu, J. Yan, J. Xia,
L. Xu, Y. Song, Y. Jiang, Q. Zhang, H. Li, J. Mater. Chem. A 2014, 2,
2563-2570.
(
PFCN) with enhanced visible light absorption and surface area
–
using dicyandiamide, BmimPF
6
ionic liquid taking CTAB and
TEOS as templates via sol-gel mediated thermal condensation
process. The presence of both soft and hard templates not only
resulted in the high surface area of 260.93 m2 g–1 but also
revealed unprecedented results for the P, F co-doping. High
surface area and co-doping results in higher activity of the
material towards visible light photoreduction of carbon dioxide into
methanol. The activity of the nanocatalyst was found to be
impeccable compared to most of the documented protocols with
the added advantage that it could also be used as a thermal
catalyst for the conversion of glucose and xylose into HMF and
furfural, respectively. The high activity of the catalyst is due to its
high surface area as well as the presence of Lewis base (nitrogen
in carbon nitride) and Brønsted acid (doped phosphorus). Notably,
this bi-functional catalyst described herein provides clear
advantages in terms of environmental impact due to no use of
metal and other additives, which bodes well for its adoption in
industrial processes. The obtained results are important for
[6]
K. S. Lakhi, D.-H. Park, K. Al-Bahily, W. Cha, B. Viswanathan, J.-H. Choy,
A. Vinu, Chem. Soc. Rev. 2017, 46, 72-101.
[
7]
a) Y. Kang, Y. Yang, L. C. Yin, X. Kang, G. Liu, H. M. Cheng, Adv. Mater.
2015, 27, 4572-4577; b) M. Z. Rahman, P. C. Tapping, T. W. Kee, R.
Smernik, N. Spooner, J. Moffatt, Y. Tang, K. Davey, S. Z. Qiao, Adv.
Funct. Mater. 2017, 27, 1702384.
[8]
A. B. Jorge, D. J. Martin, M. T. Dhanoa, A. S. Rahman, N. Makwana, J.
Tang, A. Sella, F. Corꢀ, S. Firth, J. A. Darr, J. Phys. Chem. C. 2013, 117,
7178-7185.
[9]
J. Ran, T. Y. Ma, G. Gao, X.-W. Du, S. Z. Qiao, Energ Environ. Sci. 2015,
8, 3708-3717.
[
[
10] X. Jin, V. V. Balasubramanian, S. T. Selvan, D. P. Sawant, M. A. Chari,
G. e. Q. Lu, A. Vinu, Angew. Chem. Int. Ed. 2009, 48, 7884-7887.
11] a) Z. Chen, Q. Zhang, W. Chen, J. Dong, H. Yao, X. Zhang, X. Tong, D.
Wang, Q. Peng, C. Chen, Adv. Mater. 2018, 30, 1704720; b) Z. Chen, E.
Vorobyeva, S. Mitchell, E. Fako, M. A. Ortuño, N. López, S. M. Collins,
P. A. Midgley, S. Richard, G. Vilé, Nature Nanotechnol. 2018, 13, 702.
[12] Y. Zheng, J. Liu, J. Liang, M. Jaroniec, S. Z. Qiao, Energy Environ. Sci.
012, 5, 6717-6731.
2
6
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
[
13] D.-H. Lan, H.-T. Wang, L. Chen, C.-T. Au, S.-F. Yin, Carbon 2016, 100,
1-89.
14] a) Y. Wang, Y. Di, M. Antonietti, H. Li, X. Chen, X. Wang, Chem. Mater.
010, 22, 5119-5121; b) M. Xu, B. Chai, J. Yan, H. Wang, Z. Ren, K.-W.
8
[
2
Paik, Nano 2016, 11, 1650137.
[
15] a) Y. Wang, J. Zhang, X. Wang, M. Antonietti, H. Li, Angew. Chem. Int.
Ed. 2010, 49, 3356-3359; b) H. Ma, Y. Li, S. Li, N. Liu, Appl. Surf. Sci.
2
015, 357, 131-138; c) L. Jiang, X. Yuan, G. Zeng, X. Chen, Z. Wu, J.
Liang, J. Zhang, H. Wang, H. Wang, ACS Sustainable Chem. Eng. 2017,
, 5831-5841; d) P. Babu, S. Mohanty, B. Naik, K. Parida, ACS Appl.
5
Energy Mater. 2018, 1, 5936-5947; e) Y.-C. Lu, J. Chen, A.-J. Wang, N.
Bao, J.-J. Feng, W. Wang, L. Shao, J. Mater. Chem. C 2015, 3, 73-78.
16] Y. Zhang, T. Mori, J. Ye, M. Antonietti, J. Am. Chem. Soc. 2010, 132,
[
[
6294-6295.
17] a) D. S. Simakov, Renewable Synthetic Fuels and Chemicals from
Carbon Dioxide: Fundamentals, Catalysis, Design Considerations and
Technological Challenges, Springer, 2017; b) K. Li, B. Peng, T. Peng,
ACS Catal. 2016, 6, 7485-7527; c) H. Li, Q. Zhang, P. S Bhadury, S.
Yang, Curr. Org. Chem. 2014, 18, 547-597.
[
18] Y. Zhou, W. Lin, H. Wang, Q. Li, J. Huang, M. Du, L. Lin, Y. Gao, L. Lin,
N. He, Langmuir 2010, 27, 166-169.
[
[
19] Y. Cao, T. Mu, Ind. Eng. Chem. Res. 2014, 53, 8651-8664.
20] M. Gao, J. Feng, Z. Zhang, M. Gu, J. Wang, W. Zeng, Y. Lv, Y. Ren, T.
Wei, Z. Fan, ACS Appl. Nano Mater. 2018, 1, 6733-6741.
[
21] S. Hu, L. Ma, J. You, F. Li, Z. Fan, F. Wang, D. Liu, J. Gui, RSC Adv.
2014, 4, 21657-21663.
[
[
22] S. Zhang, K. Dokko, M. Watanabe, Mater. Horizons 2015, 2, 168-197.
23] X. Li, I. V. Sergeyev, F. Aussenac, A. F. Masters, T. Maschmeyer, J. M.
Hook, Angew. Chem.Int. Ed. 2018, 130, 6964-6968.
[
[
[
24] T. Zhang, I. P. A. F. Souza, J. Xu, V. C. Almeida, T. Asefa, Nanomaterials
2018, 8, 636.
25] H.-B. Fang, X.-H. Zhang, J. Wu, N. Li, Y.-Z. Zheng, X. Tao, Appl. Cata.
B: Environ. 2018, 225, 397–405
26] a) F. R. Pomilla, A. Brunetti, G. Marcꢁ, E. I. Garcı
Fontananova, L. Palmisano, G. Barbieri, ACS Sustainable Chem. Eng.
018, 6, 8743-8753; b) J. Mao, T. Peng, X. Zhang, K. Li, L. Ye, L. Zan,
́
a-Lꢂpez, E.
2
Catal. Sci. Technol. 2013, 3, 1253-1260.
[
27] Q. Tay, P. Kanhere, C. F. Ng, S. Chen, S. Chakraborty, A. C. H. Huan,
T. C. Sum, R. Ahuja, Z. Chen, Chem. Mater. 2015, 27, 4930-4933.
28] a) W. Che, W. Cheng, T. Yao, F. Tang, W. Liu, H. Su, Y. Huang, Q. Liu,
J. Liu, F. Hu, Z. Pan, Z. Sun, S. Wei, J. Am. Chem. Soc. 2017, 139,
[
3
021−3026; b) H. Yaghoubi, Z. Li, Y. Chen, H. T. Ngo, V. R.
Bhethanabotla, B. Joseph, S. Ma, R. Schlaf, A. Takshi, ACS Catal. 2015,
, 327−335.
5
[
[
29] J.-Y. Tang, X. Y. Kong, B.-J. Ng, Y.-H. Chew, A. R. Mohamed, S.-P. Chai,
Catal. Sci. Technol., 2019, 9, 2335.
30] a) F. Parveen, S. Upadhyayula, Fuel Process. Technol. 2017, 162, 30-
36; b) X. Cao, S. P. Teong, D. Wu, G. Yi, H. Su, Y. Zhang, Green Chem.
2015, 17, 2348-2352; c) P. Zhao, Y. Zhang, Y. Wang, H. Cui, F. Song,
X. Sun, L. Zhang, Green Chem. 2018, 20, 1551-1559.
[31] S.-H. Chai, H.-P. Wang, Y. Liang, B.-Q. Xu, Green Chem., 2007, 9, 1130-
1136
7
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001172
RESEARCH ARTICLE
Entry for the Table of Contents
A novel synthetic protocol that provides P-, F-co-doped mesoporous carbon nitride (PFCN) with high surface area and lower band
gap. Inspired by its extended visible light absorption and acid-base bi-functional nature, we explored this new class of catalyst for the
photocatalytic reduction of CO
compounds.
2
into methanol and conducted thermal catalytic conversion of carbohydrates into high-value furanics
8
This article is protected by copyright. All rights reserved.