Switching the selectivity of the photoreduction reaction of carbon dioxide by controlling the band structure of a g-C3N4 photocatalyst

Article abstract of DOI:10.1039/c4cc03060e

The selectivity of the CO₂ photoreduction reaction in the presence of water vapour can be modulated by the band structure of a g-C ₃N₄ photocatalyst. The major products obtained using bulk g-C₃N₄ with a bandgap of 2.77 eV and g-C₃N ₄ nanosheets with a bandgap of 2.97 eV are acetaldehyde (CH ₃CHO) and methane (CH₄), respectively. the Partner Organisations 2014.

Full text of DOI:10.1039/c4cc03060e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Switching the selectivity of the photoreduction
reaction of carbon dioxide by controlling the
Cite this: Chem. Commun., 2014,
0, 10837
band structure of a g-C N photocatalyst†
5
3 4
Received 24th April 2014,
Accepted 21st July 2014
a
a
b
a
ac
Ping Niu, Yongqiang Yang, Jimmy C. Yu, Gang Liu* and Hui-Ming Cheng
DOI: 10.1039/c4cc03060e
www.rsc.org/chemcomm
The selectivity of the CO
water vapour can be modulated by the band structure of a g-C
photocatalyst. The major products obtained using bulk g-C with a oxidation is that the downward shift of the band edges of the
bandgap of 2.77 eV and g-C nanosheets with a bandgap of 2.97 eV surface layer of anatase TiO microspheres by introducing boron
are acetaldehyde (CH CHO) and methane (CH
2
photoreduction reaction in the presence of An example of the effect of changing the surface band structure
3 4
N
in modulating the preference of photocatalytic reduction and
3 4
N
N
3 4
2
3
4
), respectively.
can lead to switching of the preference towards photocatalytic
hydrogen and oxygen producing reactions from splitting water.
5
2
Photocatalytic reduction of CO to fuels using sunlight is an ideal These results strongly suggest that surface atomic structure and
solution to the global warming and energy problems. In this band structure are the two important factors to be considered in
approach, the reduction of CO can take place on the solid–liquid designing photocatalysts for specific reactions because they
2
interface or the solid–gas interface of semiconductor photocata- intrinsically determine the adsorption of the reactant molecules
1
–3
lysts.
transfers, leading to a mixture of products such as carbon
monoxide, organic compounds including formic acid (HCOOH), metal oxides and their derivatives (i.e., doped ones), though some
formaldehyde (HCHO), methanol (CH OH), methane (CH ) as sulfides and phosphides have been explored as photocatalysts
well as higher hydrocarbons. These organic compounds are fuels for CO
with different energy densities, and they are also important ca. 2.7 eV has emerged as an attractive visible light photocatalyst
It is a complex reaction involving multiple-electron and the redox potentials of charge carriers in photocatalysts.
Photoreduction of CO is usually conducted on transition
2
3
4
2,3
2
3 4 3 4
reduction. Graphitic-C N (g-C N ) with a bandgap of
6
6–10
industrial chemicals. Concerning the practical use of these for photocatalytic hydrogen evolution,
decomposition of
11–13
14
15
compounds, it is essential to produce them to be as pure as pollutants,
possible. This requires the development of photocatalysts with because its high conduction band minimum favors the reduction
high selectivity for a specific product. Unfortunately, little is half-reaction. It is also used for CO photoreduction or CO
Moreover, a layered structure makes it easy to
reduction. Revealing these crucial factors is tailor its band structure by controlling its thickness without chan-
therefore important but it remains a great challenge. ging the atomic structure. All these properties make g-C a good
organic synthesis and optoelectronic conversion
2
2
16–21
known about the crucial factors controlling the selectivity of the conversion.
photocatalytic CO
2
3 4
N
For a single-component photocatalyst, there are many factors platform to study the effect of band structure on the selectivity of
that can affect its reaction selectivity. A previous study shows that CO photoreduction.
2
modifying the surface atomic structure of anatase TiO micro-
In this work, two g-C N photocatalysts with different band
3 4
2
spheres with dominant {001} facets by introducing fluorine can structures, namely, bulk g-C N and g-C N nanosheets with
3
4
3 4
sensitively change the adsorption capability towards azo dyes and bandgaps of 2.77 eV and of 2.97 eV (Fig. 1) were used to study
4
thus realize the selective decomposition of different dyes.
the effect of the band structure of photocatalysts on the selectivity
of the photoreduction of CO . Bulk g-C was synthesized
2
3 4
N
a
through the thermal condensation of dicyandiamide in air accord-
ing to a reported procedure. The synthesis of the nanosheets was
Shenyang National Laboratory for Materials Science, Institute of Metal Research,
6
Chinese Academy of Sciences, 72 Wenhua RD, Shenyang 110016, P. R. China.
E-mail: gangliu@imr.ac.cn; Fax: +86 24 23971682; Tel: +86 24 23971088
conducted using a simple top-down strategy developed in our
previous study, namely the thermal oxidation etching of bulk
b
22
Department of Chemistry and Institute of Environment, Energy and Sustainability,
The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, China
g-C N in air. The resultant nanosheets have a very similar
c
3 4
Chemistry Department, Faculty of Science, King Abdulaziz University,
3 4
geometric structure to their parent bulk g-C N . One distinctive
Jeddah 21589, Saudi Arabia
feature of the nanosheets obtained from the bulk is a 0.2 eV larger
bandgap associated with their small thickness of around 2 nm.
†
Electronic supplementary information (ESI) available: Experimental details,
SEM images, XRD patterns and XPS spectra. See DOI: 10.1039/c4cc03060e
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 10837--10840 | 10837

View Article Online
Communication
ChemComm
photocatalytically active in reducing CO
2
and its selectivity of the
photoreduction reactions is associated with its bandgap. On the
other hand, the competitive hydrogen production during the CO2
photoreduction was estimated as shown in Fig. S1 (ESI†). Under
UV-visible light the nanosheets give a hydrogen production rate
comparable to that of their CH
much higher hydrogen production rates than bulk g-C
4
production. The nanosheets show
under
3 4
N
both UV-visible light and visible light. This is largely because of the
advantageous band structure of the nanosheets and also the large
surface area.
Prior to an analysis of the origin of the reaction selectivity, it is
necessary to investigate the stability of g-C N photocatalysts in the
photoreduction of CO2 in terms of their morphology, crystal
3
4
Fig. 1 Schematic of the generation of CH
4
and CH
3
CHO on bulk g-C
3 4
N
and a g-C N nanosheets in the photoreduction of CO in the presence of
water vapor.
3
4
2
structure and chemical state. Fig. S2 (ESI†) shows scanning electron
microscopy images of two g-C
reactions. Compared to the fluffy state of the nanosheets, bulk
g-C appears as particles of several micrometers width. No
detectable change was observed after the reactions. X-ray diffraction
XRD) patterns (Fig. S3, ESI†) of these photocatalysts show two
3 4
N photocatalysts before and after the
3 4
N
(
characteristic peaks with the centers at around 13.11 and 27.31,
which result from the inter-plane and intra-plane periodic stacking
units in g-C
reactions suggest a good retention of the crystal structure of g-C
The chemical states of carbon and nitrogen of g-C photo-
catalysts were investigated using X-ray photoelectron spectroscopy
XPS). Fig. 3 compares the XPS spectra of C 1s and N 1s in bulk
g-C N before and after the reaction. Two major peaks of C 1s at
3 4
N , respectively. The unchanged XRD patterns after the
3 4
N .
3 4
N
(
3
4
2
84.6 and 288.2 eV originate from adventitious carbon and C–(N)
3
23
(the main state of C in g-C N ), respectively. In the spectra of N 1s,
3 4
two peaks at 398.6 and 400.3 eV are assigned to the nitrogen species
23
in the forms of N–(C)
show oxygen species adsorbed on g-C
2
and N–(C)
3
, respectively. The spectra of O 1s
with a binding energy of
Fig. 2 Irradiation time-dependent generation of CH
bulk g-C and g-C nanosheets in the photoreduction of CO
A, B) UV-visible light and (C, D) visible light (l 4 400 nm).
4
and CH
3
CHO for
3 4
N
3
N
4
3
N
4
2
under
inner core electrons at 532 eV. No binding energy shift or new peak
after the reaction is detected in the C 1s, N 1s and O 1s spectra.
The analysis of the chemical states of C, N and O of the nanosheets
(
The characteristics of the photoreduction reaction of CO were
2
studied in a solid–gas reaction system containing the g-C
photocatalyst, CO , and water vapor. Fig. 2 shows the irradiation
time-dependent generation of products using the two g-C
photocatalysts. In all cases, CH and CH CHO are the two major
carbon–hydrogen containing products. Under UV-visible light, the
nanosheets show a high selectivity toward the generation of CH
Fig. 2A) while the bulk g-C N has a preference toward the
3 4
N
2
3 4
N
4
3
4
(
3
4
production of CH CHO (Fig. 2B). The generation rate of CH with
3
4
the nanosheets is more than an order of magnitude higher than
that of the bulk g-C . In contrast, the generation rate of CH CHO
with the nanosheets is an order of magnitude lower than that of
bulk g-C . Similar trends were also observed under visible light
Fig. 2C and D), although the absolute generation rates of CH and
CH CHO for the nanosheets and bulk g-C are much lower than
3
N
4
3
3 4
N
(
4
3
3 4
N
those under UV-visible light. These results clearly suggest that
g-C N photocatalysts with different band structures have different
3
4
effects on CO
wavelength of light irradiation. Control experiments showed no
activity in generating CH and CH CHO in the absence of g-C or
CO or light irradiation. All these results indicate that g-C
2
photoreduction, which is independent of the
Fig. 3 Comparison of high resolution XPS spectra of C 1s, N 1s and O 1s of
4
3
3
N
4
bulk g-C
3 4 2
N (a) before and (b) after CO photoreduction under UV-visible
is light.
2
N
3 4
10838 | Chem. Commun., 2014, 50, 10837--10840
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
also supports no binding energy shift and the absence of a new peak
(Fig. S4, ESI†). Overall, the chemical states of lattice carbon and
nitrogen in g-C N samples remain unchanged after the reactions.
3
4
These results together with the analysis from SEM and XRD suggest
a good stability of g-C N photocatalysts in CO photoreduction.
3
4
2
According to the literature, three different mechanisms for the
conversion of CO into methane on the semiconductor surface
2
have been proposed. These involve the production of three
2
different intermediates: formaldehyde, carbene and glyoxal.
3
Considering the presence of CH CHO in the final products, the
mechanism of photoreduction of CO on g-C N photocatalysts
2
3 4
Fig. 4 Schematic of the band structures of bulk g-C
3 4
N (the right panel)
could follow the glyoxal pathway containing the dimerization
and g-C nanosheets (the left panel) referred to the redox potentials of
3 4
N
2
process (Fig. S5, ESI†). Since the completion of the glyoxal different redox reactions. VB: valence band; CB: conduction band.
pathway involves both the reduction and oxidation processes,
efficient transfer of photo-generated electrons and holes from
2
À1
large specific surface area of the nanosheets (306 vs. 50 m g of
g-C
tion of CH
the result that the major final product with bulk g-C
3
N
4
to the reactants are essentially important for the produc-
as the product at the end of this pathway. Based on
is
bulk g-C N ) is favorable for photocatalytic reaction activity by
providing abundant active sites for the adsorption of reactant
species. This might partially favor the reduction of CO to CH by
increasing the adsorption capability of intermediate products
and thus promoting the subsequent elementary steps.
3 4
4
3 4
N
CH CHO while the major final product with the nanosheets is
2
4
3
CH , it is reasonably inferred that the oxidation of CH CHO and
4
3
subsequent reduction process on bulk g-C N is limited. This
3
4
Finally, we should discuss the influence of the layer struc-
inference is further supported by the fact that bulk g-C
produces a much lower amount of the CO byproduct in the
photoreduction of CO than g-C nanosheets.
Band structure is a crucial factor affecting CO
3 4
N
ture of g-C
3
N
4
on the selectivity of CO
2
photoreduction. The
N consist of hydrogen-bonded strands of poly-
meric melon units as shown in Fig. 1. Independent of the
thickness along the c-axis of the bulk g-C N and nanosheets
layers in g-C
3 4
2
3 4
N
2
7
2
photoreduc-
tion. The pioneering work on the photoelectrocatalytic reduction
3 4
studied, the surface atomic structure of the two photocatalysts
can be considered to be the same. The adsorption of the
reactant species is dependent on the surface atomic structure
of the solid so that the adsorption state of the reactant species
on g-C N is also the same. In this situation, the photoreduc-
of CO in aqueous solution by Inoue et al. showed a positive
2
dependence of the reactivity of CO reduction on the energy level
2
difference between the conduction band of the semiconductor
and the redox agents in the solution by comparing a group of
1
semiconductors with different conduction band minima. This
3 4
2 3 4
tion of CO on g-C N could follow a single reaction pathway
was explained by saying that a large energy level difference
increases charge carrier transfer rate between the photogenerated
and then stop at different elementary steps largely depending
on the photoexcited carrier transfer rate as analyzed above. This
may explain why only one major product was produced using
24–26
carriers and the solution species. Recent progress
has shown
that wide-bandgap semiconductors with a high conduction band
g-C
In summary, the photoreduction reactions of CO
g-C photocatalysts with different bandgaps were investigated.
The dependence of the nature of the major product on the band
structure of g-C was shown. Bulk g-C with a bandgap of
.77 eV gives the major product of CH CHO and the nanosheets
3 4
N with a layered structure.
level (more negative) give a high reactivity of CO photoreduction,
2
on two
which supports this hypothesis. In our case, compared to bulk
g-C N , the nanosheets have a larger bandgap by 0.2 eV as a result
3 4
of a lower valence band edge by 80 meV (Fig. S6, ESI†) and high
conduction band edge by 120 meV, as illustrated in Fig. 4. This
2
3 4
N
N
3 4
3 4
N
2
3 4
means that, compared to bulk g-C N , the nanosheets can
3
with a bandgap of 2.97 eV give the major product of CH . These
provide a stronger driving force for the transfer of holes or
electrons as a result of the larger energy level difference between
the electronic band edges and the redox potentials of the
reactants (Fig. 4). In this situation, fast photoexcited electron
transfer to the intermediate species in all elementary steps of
4
results could provide insight into the design of efficient photo-
catalysts with high selectivity for CO photoreduction conversion.
2
The authors thank the Major Basic Research Program, Ministry
of Science and Technology of China (2014CB239401) and the
National Science Fund of China (No. 21090343, 51221264) for the
financial support.
producing CH
slow electron transfer rate of bulk g-C
probability of the latter elementary steps occurring. In addition,
4
is favorable and feasible. In contrast, the relatively
3 4
N
greatly lowers the
in contrast to the bulk, the nanosheets can generate a larger Notes and references
percentage of long-lived charge carriers under the light irradia-
1
2
T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature, 1979, 277, 637.
S. N. Habisreutinger, L. S. Mende and J. K. Stolarczyk, Angew. Chem.,
Int. Ed., 2013, 52, 7372.
V. P. Indrakanti, J. D. Kubicki and H. H. Schobert, Energy Environ.
Sci., 2009, 2, 745.
S. W. Liu, J. G. Yu and M. Jaroniec, J. Am. Chem. Soc., 2010,
132, 11914.
22
tion. This increases the probability of photo-generated electrons
and holes in involving the reduction of CO to CH production.
2
4
3
4
The high specific surface area of the nanosheets may have an
impact on the photoreduction of CO . There is no doubt that if all
2
the other parameters of the photocatalysts are comparable, the
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 10837--10840 | 10839

View Article Online
Communication
ChemComm
5
6
7
8
9
G. Liu, J. Pan, L. C. Yin, J. T. S. Irvine, F. Li, J. Tan, P. Wormald and 17 J. Mao, T. Y. Peng, X. H. Zhang, K. Li, L. Q. Ye and L. Zan, Catal. Sci.
H. M. Cheng, Adv. Funct. Mater., 2012, 22, 3233. Technol., 2013, 3, 1253.
X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. 18 Y. P. Yuan, S. W. Cao, Y. S. Liao, L. S. Yin and C. Xue, Appl. Catal., B,
Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76. 2013, 140–141, 164.
J. S. Zhang, J. H. Sun, K. Maeda, K. Domen, P. Liu, M. Antonietti, 19 Z. J. Huang, F. B. Li, B. F. Chen, T. Lu, Y. Yuan and G. Q. Yuan,
X. Z. Fu and X. C. Wang, Energy Environ. Sci., 2011, 4, 675. Appl. Catal., B, 2013, 136–137, 269.
G. Liu, P. Niu, C. H. Sun, S. C. Smith, Z. G. Chen, G. Q. Lu and 20 S. W. Cao, X. F. Liu, Y. P. Yuan, Z. Y. Zhang, Y. S. Liao, J. Fang,
H. M. Cheng, J. Am. Chem. Soc., 2010, 132, 11642. S. C. J. Loo, T. C. Sum and C. Xue, Appl. Catal., B, 2014, 147, 940.
Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Phys. Chem. C, 2011, 21 K. Maeda, K. Sekizawa and O. Ishitani, Chem. Commun., 2013, 49, 10127.
1
15, 7355.
22 P. Niu, L. L. Zhang, G. Liu and H. M. Cheng, Adv. Funct. Mater., 2012,
22, 4763.
1
1
0 L. Ge and C. C. Han, Appl. Catal., B, 2012, 117, 268.
1 X. C. Wang, X. F. Chen, A. Thomas, X. Z. Fu and M. Antonietti, 23 A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J. O. Muller,
Adv. Mater., 2009, 21, 1609.
R. Schl ¨o gl and J. M. Carlsson, J. Mater. Chem., 2008, 18, 4893.
2 J. G. Yu, S. H. Wang, J. X. Low and W. Xiao, Phys. Chem. Chem. Phys., 24 S. C. Yan, S. X. Ouyang, J. Gao, M. Yang, J. Y. Feng, X. X. Fan,
1
2
013, 15, 16883.
L. J. Wan, Z. S. Li, J. H. Ye, Y. Zhou and Z. G. Zou, Angew. Chem.,
Int. Ed., 2010, 49, 6400.
1
1
3 L. Ge, C. C. Han and J. Liu, J. Mater. Chem., 2012, 22, 11843.
4 X. F. Chen, J. S. Zhang, X. Z. Fu, M. Antonietti and X. C. Wang, 25 Q. Liu, Y. Zhou, J. H. Kou, X. Y. Chen, Z. P. Tian, J. Gao, S. C. Yan
J. Am. Chem. Soc., 2009, 131, 11658. and Z. G. Zou, J. Am. Chem. Soc., 2010, 132, 14385.
5 Y. J. Zhang, T. Mori, L. Niu and J. H. Ye, Energy Environ. Sci., 2011, 26 G. C. Xi, S. X. Ouyang and J. H. Ye, Chem. – Eur. J., 2011, 17, 9057.
1
1
4, 4517.
27 B. V. Lotsch, M. Doblinger, J. Sehnert, L. Seyfarth, J. Senker,
6 G. H. Dong and L. Z. Zhang, J. Mater. Chem., 2012, 22, 1160.
O. Oeckler and W. Schnick, Chem. – Eur. J., 2007, 13, 4969.
1
0840 | Chem. Commun., 2014, 50, 10837--10840
This journal is ©The Royal Society of Chemistry 2014