Journal of

Materials Chemistry A
View Article Online
PAPER View Journal | View Issue
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

A solar-driven photocatalytic fuel cell with dual

photoelectrode for simultaneous wastewater
Cite this: J. Mater. Chem. A, 2015, 3,
3416 treatment and hydrogen production†
Zhongyi Wu, Guohua Zhao,* Yajun Zhang, Jian Liu, Ya-nan Zhang and Huijie Shi

A solar-driven dual photoelectrode photocatalytic fuel cell (PFC) based on n-type semiconductor
photoanode and p-type semiconductor photocathode is reported for wastewater treatment with
simultaneous hydrogen production. The PFC shows a superior performance for phenol degradation and
hydrogen production with a maximum TOC removal rate of 84.2% and a total hydrogen production rate
of 86.8 mmol cm
2 in 8 hours which is much higher than other similar researches. Compared to the
system comprised of a photoanode (photocathode) and Pt for phenol degradation (hydrogen
production), the results prove that there is a synergistic effect between the two photoelectrodes. This is
because the electrons are preferentially stored on the photocathode for hydrogen production and holes
Received 2nd December 2014
Accepted 9th December 2014
on the photoanode for pollutant degradation. Several semiconductors are chosen as photoelectrodes to
investigate the factors influencing the performance of the PFC system. The VOC value of the PFC
DOI: 10.1039/c4ta06604a
increases with the difference of Fermi level between the two photoelectrodes. This PFC system provides
www.rsc.org/MaterialsA a new approach for efficient energy recovery from wastewater.

hydrogen is a kind of green energy because of no secondary

Introduction pollution. However, use of MFCs is not feasible to treat some
Energy and environment quality are two major issues faced by persistent organics due to their biorefractory nature. The harsh
countries in modern society.1,2 Simultaneous wastewater treat- working conditions, slow electron transfer and complicated
ment and energy recovery by solar energy conversion is an bacteria cultivation also obstruct the large-scale commercial
attractive and promising technology. As a result of the rapid application of MFCs.
development of industrialization, environmental pollution, As we know, light is a kind of cost-free, clean and convenient
especially water pollution, is becoming more and more serious energy. So the idea of using light as a substitute for microor-
due to rapidly increasing effluent discharges. As we know, ganisms has come to the fore. So a novel photoelectrocatalytic
pollutant degradation is the main purpose of traditional (PEC) system for hydrogen production has been proposed.
wastewater treatment methods, but a huge amount of chemical Previously, many researchers focused on hydrogen production
energy is ignored at the same time.3 It is urgent for us to seek by PEC water reduction on a single photocatalyst.9,10 A highly
new and efficient methods for wastewater treatment to elimi- active Cu2O/(ZnO/Al2O3)/TiO2/Pt photocathode has been used
nate pollution and obtain energy. for water reduction.11 CdS-cluster-decorated graphene nano-
A microbial fuel cell (MFC) realizes the concept of generating sheets have been applied for highly efficient visible-light-driven
electricity during the degradation of organic matter by a photocatalytic hydrogen production.12 Also, a two-electrode PEC
microbial community.4–6 The MFC converts the energy of system composed of Cu/nanoCu2O/NiOx photocathode and
organic pollutants present in wastewater into electric current. FTO/nanoWO3 photoanode has been used for PEC hydrogen
However, because of the relatively low current efficiency, production.13 However, the PEC system for hydrogen produc-
researchers have switched to produce hydrogen with MFCs tion needs external electric energy input. The cost and energy
combined with electrocatalysis or photocatalysis.7,8 As is known, consumption are both high. As a result, this type of system is
insufficient from the point of view of energy recovery.
At the same time, the PEC system has been used for organic
Department of Chemistry, Key Laboratory of Yangtze River Water Environment, Tongji waste degradation.14 A solar-driven biomass fuel cell has been
University, Shanghai 200092, People's Republic of China. E-mail: [email protected]. used for electricity generation from wastewater.15 A single
edu.cn; Fax: +86-21-65982287; Tel: +86-21-65981180 photoelectrode photocatalytic fuel cell (PFC) under light illu-
† Electronic supplementary information (ESI) available: The pretreatment of FTO;
mination has been applied for refractory organic pollutant
the detailed preparation procedure of TiO2 NTs/Ti, CdS/FTO, CdSe/FTO and Cu2O
NPs/FTO; the scheme of quartz reactor; the GC spectra of hydrogen. See DOI:
degradation and simultaneous electricity generation.16–18 A
10.1039/c4ta06604a visible light-driven PFC composed of BiOCl/Ti photoanode and

3416 | J. Mater. Chem. A, 2015, 3, 3416–3424 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry A

Pt cathode has been used for organic pollutant degradation and aqueous solution (6 mol L
1, 15 mL) under continuous stirring
electricity generation.19 A dual photoelectrode of WO3/W or for 30 min at room temperature. Then the solution was trans-
TiO2/Ti as photoanode and Cu2O/Cu as photocathode under ferred to a Teon-lined stainless steel autoclave, and a piece of
visible light illumination has been applied for simultaneous FTO sample was immersed in the solution with the conductive
wastewater treatment and electricity generation.20,21 In conclu- surface facing down. The autoclave was sealed and placed in an
sion, these works focus on the degradation of pollutants on the oven at 150  C for 4 h with heating and cooling rates of 5  C
photoanode and the electricity generation performance. min
1. Aer cooling the autoclave to room temperature, the
However, there are many factors affecting the performance of TiO2 NRs/FTO was rinsed several times with deionized water
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

PFCs. No further research about specic mechanisms of PFCs is and dried in air.37 The SEM image of TiO2 NRs/FTO is shown in
discussed. So it is meaningful to discover theoretical guidance Fig. S2A.†
for the selection of photoelectrodes in the PFC. Meanwhile, a TiO2 NTs/Ti. Firstly, TiO2 NTs were prepared by a two-step
PFC cannot only be used for simultaneous wastewater treat- electrochemical anodic oxidation method according to the
ment in the anode area and energy recovery, but it is also literature. Aer that, they were annealed in air atmosphere for 3
considered that specic attention should be paid to the cathode h with heating and cooling rates of 5  C min
1. The detailed
reaction. preparation procedure is shown is ESI.†38
Therefore, a solar-driven dual photoelectrode PFC for CdSe NPs/FTO. The applied potential of electrodeposition was
simultaneous organic pollutant degradation and hydrogen
1.1 V vs. SCE and the deposition time was 2 h. Then the CdSe
production has come to the fore. This approach combines the NPs/FTO were rinsed thoroughly and annealed at 300  C in N2
advantages of the two systems mentioned above. As we know, atmosphere for 3 h in a tube furnace (SOM, SGL-1200Z). The
the electrons and holes of a semiconductor will be separated detailed preparation procedure is shown is ESI.†39
under light illumination and there will be a photovoltage. For CdS NPs/FTO. The applied potential of electrodeposition was
n-type semiconductor under light illumination photo-gener-
0.65 V vs. SCE and the deposition time was 1000 s. The ther-
ated holes are the majority for oxidation. In contrast, for p- mostatic bath temperature was 70  C. Then the CdS NPs/FTO
type semiconductor under light illumination photo-generated were rinsed and annealed at 500  C in N2 atmosphere for 6 h.
electrons are the majority for reduction.22,23 So we consider The detailed preparation procedure is shown is ESI.†40
whether the photovoltage difference between two different Photocathode
semiconductors can be used for pollutant oxidation on pho- C/Cu NWAs/Cu mesh. A Cu mesh (Alfa Aesar, 100 mesh, 0.11
toanode and simultaneous hydrogen production on photo- mm as wire diameter) was anodized in an NaOH (Aladdin, AR,
cathode. If possible, the photoanode of the PFC should be an 96.0%) solution (3 mol L
1) for 20 min under 10 mA cm
2 to
n-type semiconductor and have a good photocatalytic prop- form Cu(OH)2 NWAs/Cu mesh at 25  C. The Cu(OH)2 NWAs/
erty for pollutant oxidation, such as TiO2, CdS and so on.24–30 Cu mesh was immersed into 3 mg mL
1 glucose solution,
The photocathode of the PFC should be a p-type semi- which was prepared using deionized water, for glucose
conductor whose Ecb (vs. NHE) should be more negative than coating. Then the Cu(OH)2 NWAs/Cu mesh with glucose
EH+/H2 of 0 V for the purpose of hydrogen production, such as coating was annealed at 550  C in N2 atmosphere for 4 h with
Cu2O.31–34 heating and cooling rates of 5  C min
1.41 The SEM images of
To investigate simultaneous pollutant treatment and Cu NWAs/Cu mesh with or without carbon layer are shown in
hydrogen production, a sealed quartz reactor is used for Fig. S2B and C.†
hydrogen collection which is shown in Fig. S1.† Phenol was Cu2O NPs/FTO. The electrochemical deposition was per-
selected as a representative organic pollutant to study the formed potentiostatically with a potential of
0.3 V (versus Ag/
degradation performance of the PFC.35,36 AgCl) for 30 min and the temperature of the electrolyte was kept
Furthermore, the most important thing for the PFC is the at 60  C using a water bath. The detailed preparation procedure
operation mechanism. So for the study of the operation mech- is shown is ESI.†42
anism of the PFC, several n-type and p-type semiconductors
have been chosen to verify its feasibility. Here, n-TiO2, n-CdS, n-
CdSe and p-Cu2O have been chosen as the photoelectrode. Also, Characterization methods
the same semiconductor with different microstructure has also The morphology of the sample photoelectrodes was identied
been chosen. This work provides an effective and low energy by using eld emission scanning electron microscopy (FE-SEM,
consumption approach for simultaneous environmental Hitachi S-4800, Japan). Detailed microstructural features of the
pollutant disposal and energy recovery by using a photocatalytic sample electrodes were investigated by high resolution trans-
system. mission electron microscopy (HR-TEM, JEM2100, JEOL, Japan),
using a JEM-2000EX electron microscope operated at 200 kV.
Experimental section The crystalline structures of the samples were analyzed by X-ray
diffraction (XRD, D8 Focus X-ray diffractometer, Bruker, Ger-
Electrode preparation many), using Cu Ka radiation (l ¼ 1.540598 Å). UV-visible
Photoanode diffuse reectance absorption spectra (UV-vis DRS) were recor-
TiO2 NRs/FTO. Titanium butoxide (0.35 mL) (Aladdin, AR, ded by AvalightDHS UV-vis absorbance measurements (Avantes,
99.0%) was added dropwise into a HCl (SCRC, AR, 36.0–38.0%) Netherlands).

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. A, 2015, 3, 3416–3424 | 3417
 View Article Online

Journal of Materials Chemistry A Paper

Photoelectrochemical measurement
The photoelectrochemical characterizations of the photo-
electrodes were carried out by using a three-electrode congu-
ration with the samples, platinum foil and Ag/AgCl2 as working
electrode, counter electrode and reference electrode, respec-
tively. The working electrode parameters were controlled by an
electrochemical workstation (CHI660C, USA). The PFC was set
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

up using two-electrode congurations. The photoanode and

photocathode were immersed in the solution in a quartz reactor
and illuminated with one light source. Herein, a 300 W PSL-
SXF300 Xe lamp (Changtuo, China) was used as simulated solar
light source. Without further description, all experiments were
carried out under AM 1.5 (light intensity, 100 mW cm
2) solar
light irradiation by using an AM 1.5 lter supplied by Perfect-
Light (Beijing, China). All the experiments were carried out in a
rectangular quartz reactor (V ¼ 435 mL) with the working area of
the photoanode and photocathode both of 2 cm2, moderate
stirring, without applying external potential (see Fig. S1 in ESI†).
Linear sweep voltammetry was used to obtain the I–V charac-
teristics curve for one typical cycle at a scan rate of 0.02 V s
1.20
The I–V characteristic curves of the PFCs were measured in a
two-electrode system, where the anode served as the working
electrode and the cathode as the counter and reference elec-
trodes. The voltage and power were obtained as a function of
current density, and normalized to the photocathode area. The Fig. 1 (A) The photocurrent of TiO2 NRs/FTO, C/Cu2O NWAs/Cu and
power density (P) was calculated from P ¼ I  V. PFC system, where the substrate is 0.05 mol L
1 phenol and support
Organic compound degradation was carried out in a quartz electrolyte is 0.1 mol L
1 Na2SO4 under AM 1.5 illumination (light density
reactor under AM 1.5 (light intensity of 100 mW cm
2) solar of 100 mW cm
2), and the areas of both photoelectrodes are 2 cm2. (B)
The TOC removal rates of phenol (20 mg L
1) for TiO2 NRs/Pt and TiO2
light irradiation, too. The target organic compound was phenol
NRs, C/Cu2O NWAs/Cu PFC under light illumination and in the dark.
(20 mg L
1) and 0.1 mol L
1 Na2SO4 was used as the electrolyte.
The TOC removal rate of phenol was monitored by a multi N/C
3100 TOC/TNb analyzer (Analytikjena, Germany).
The total hydrogen amount produced was calculated by 0.4 mA cm
2. The short circuit current (JSC) of the PFC
multiplying the hydrogen content in a 1 mL gas sample by the composed of TiO2 NRs/FTO photoanode and C/Cu2O NWAs/Cu
headspace volume of the PFC reactor (200 mL).43 The hydrogen photocathode reaches 0.50 mA cm
2 which is more than two
was measured by gas chromatography (GC 7900, Techcomp, times higher than the reported JSC of TiO2/Ti-Cu2O/Cu system
(0.23 mA cm
2).21 This value is higher than the sum of the
Shanghai, China).
photocurrents of the photoanode and photocathode. This
means that the PFC system has a synergistic facilitating effect
Results and discussion on the photoresponse of the photoanode and photocathode.
Fig. 1B shows the performance of phenol degradation by pho-
PEC performance, TOC removal and hydrogen production of tocatalysis with TiO2 photoelectrode and TiO2 NRs/FTO-C/Cu2O
PFC NWAs/Cu PFC with or without light illumination. As we can see
Fig. 1A presents the photocurrent of TiO2 NRs/FTO, C/Cu2O from the TOC removal rate of phenol, the degradation efficiency
NWAs/Cu, and TiO2 NRs/FTO-C/Cu2O NWAs/Cu PFC. SEM of the PFC system is improved signicantly with a high TOC
images and UV-vis absorption spectra of them are shown in removal rate compared with that photocatalyzed by TiO2 pho-
Fig. S2 and S3.† There is almost no photoresponse for the TiO2 toanode and Pt system. There is no TOC removal of phenol
NRs/FTO photoanode. This may be because TiO2 is a broadband when no light is incident on the TiO2 NRs/FTO photoanode and
semiconductor and it only has a response to ultraviolet (UV) C/Cu2O NWAs/Cu photocathode. Because TiO2 is a UV-light-
light. However, the light source we used is a solar light source in response semiconductor, there is a weak response to the
which the UV light constitutes only about 5%. So under the simulated solar light. So the degradation rate of phenol is very
solar light illumination, there is almost no photoresponse for poor, only 6.4%. However, the PFC has a strong response to
the TiO2 NRs/FTO photoanode. Because Cu2O is a narrowband solar light due to the presence of the C/Cu2O NWAs/Cu photo-
semiconductor which has a response to visible light and there is cathode and the synergistic effect of the two photoelectrodes. So
a large proportion of visible light in solar light, the C/Cu2O its degradation rate of phenol reaches as much as 84.2%. The
NWAs/Cu photocathode has a strong photoresponse to the incident photon-to-electron conversion efficiency (IPCE) of the
simulated solar light whose photocurrent reaches as much as PFC can be calculated by the equation:

3418 | J. Mater. Chem. A, 2015, 3, 3416–3424 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry A

Table 1 The photocurrent density of photoelectrodes and PFCs

PFC

No.a Photoanode Photocathode J1c (mA cm
2) J2d (mA cm
2) JSCe (mA cm
2) 0
JSC ¼ J1 + J2 (mA cm
2)

C1 TiO2 NRsb C/Cu2O/Cu 0.01 0.41 0.50 0.42

C2 TiO2 NRs Cu2O 0.01 0.25 0.33 0.26
C3 TiO2 NTs/Ti C/Cu2O/Cu 0.005 0.41 0.25 0.415
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

C4 TiO2 NTs/Ti Cu2O 0.005 0.25 0.20 0.255

C5 CdS C/Cu2O/Cu 0.30 0.41 0.61 0.71
C6 CdS Cu2O 0.30 0.25 0.31 0.55
C7 CdSe C/Cu2O/Cu 0.10 0.41 0.16 0.51
C8 CdSe Cu2O 0.10 0.25 0.11 0.35
a
The number of PFC is applicable to the following gures and tables. b The substrate of material denoted without FTO, and this is also applicable to
the following tables. c J1 are the photocurrent density of photoanode. d J2 are the photocurrent density of photocathode. e JSC is measured in the
initial 10 minutes while the PFC is working, and this is also applicable to the following tables.

 

1240I
IPCE % ¼  100% (1)
lJlight

where I is the photocurrent density (mA cm
2), Jlight is the

incident irradiance (mW cm
2) and l is the incident light
wavelength. The IPCE curve is shown in Fig. S5.†
At the same time, the photocurrent and degradation rate of
phenol for different PFCs have been studied. Herein, eight
kinds of PFCs mentioned above were chosen to study the
photocurrent. Table 1 shows that JSC of TiO2 NRs/FTO-C/Cu2O
NWAs/Cu PFC reaches as much as 0.50 mA cm
2. JSC (TiO2
NRs/FTO PFC) is higher than JSC (TiO2 NTs/Ti PFC) with the
same photocathode. This is because the photoresponse of
photoanodes determines the relationship of the two JSC values Fig. 2 The TOC removal rate of phenol (20 mg L
1) of 8 PFCs using two-
electrode configuration under AM 1.5 illumination, with 0.1 mol L
1
when the photocathode is xed. This could be attributed to the
Na2SO4 as support electrolyte. Here, C1 to C8 are the numbers of the
single crystal structure of the TiO2 NRs/FTO (see Fig. S2 A3†). PFCs as in Table 1.
The performance of the TiO2 NRs/FTO-C/Cu2O NWAs/Cu PFC is
the best of all. For the PFC with the same photoanode, JSC of
C/Cu2O/Cu is larger than that of Cu2O/FTO. The reason specu-
lated is not only the photoresponse of the photocatalyst to solar degradation. In these 8 PFCs, the CdSe/FTO-Cu2O/FTO PFC has
light but also the conductivity of the substrates. the lowest TOC removal rate of phenol of 42%.19,20
These TOC removal rates for phenol by the 8 PFCs have been In the hydrogen production experiments, the closed reactor
studied. Here the concentration of the phenol is 20 mg L
1, and was purged with N2 to ensure air-proof conditions, and the gas
the support electrolyte is 0.1 mol L
1 Na2SO4. In Fig. 2, the TOC composition was monitored in the whole cycles. The standard
removal rate of phenol on TiO2 NRs/FTO-C/Cu2O NWAs/Cu PFC H2 peak of the GC spectra appears when the retention time is
is the highest, which reaches 84.2%. The result is consistent 1.2 min (Fig. S6†) and the hydrogen-producing rates in 8 h of
with that of JSC comparison. However, there is an exception. the 8 PFCs are shown in Fig. 3. The average hydrogen-producing
With the same photocathode, the TOC removal rate of phenol of rate of the TiO2 NRs/FTO-C/Cu2O NWAs/Cu PFC reaches about
the PFC with CdS/FTO photoanode is lower than that of TiO2 11 mmol h
1 cm
2 (normalized to the cathode area) and for 8 h
NRs/FTO and TiO2 NTs/Ti, but JSC of CdS/FTO photoanode PFC the total rate is 86.8 mmol cm
2 which is the best of all. This
is higher. This is because JSC of the PFC will continuously decay result is higher than the hydrogen production by BPEC.43 For
with light illumination (see Fig. S4†) and JSC in Table 1 is only comparison, the hydrogen-producing rate of Pt-C/Cu2O NWAs/
measured at the initial 10 minutes. JSC of CdS/FTO PFC decays Cu reaches only 1.4 mmol h
1 cm
2 and the total is 10.6 mmol
more rapidly due to the serious photocorrosion of CdS. The cm
2. The result of hydrogen-producing rate of different PFCs is
opposite trend is observed for JSC of CdS/FTO photoanode consistent with that of JSC, too. As with the TOC removal rates,
which is higher than that of TiO2 NTs/Ti. JSC of CdS/FTO for the CdS/FTO photoanode PFC, the hydrogen-producing rate
photoanode decays obviously with increasing time. Aer decreases much in the last 4 h due to the serious photo-
6 hours, JSC is only a tenth of the initial photocurrent density. So corrosion of CdS. So its hydrogen-producing rate is not the
we can see that there is little effect in the last 2 hours on phenol highest. However, when the semiconductor anode is replaced

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. A, 2015, 3, 3416–3424 | 3419
 View Article Online

Journal of Materials Chemistry A Paper

and hydrogen-producing rate of 86.84 mmol cm
2 which are

much higher than those of other parallel studies.20,21

Operation mechanism of dual photoelectrode PFC

The dual photoelectrode PFC is composed of a conguration of
two photocatalytic semiconductor electrodes including n-type
semiconductor photoanode and p-type photocathode. Under
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

light illumination, both of the photoelectrodes can generate

electron–hole pairs. The driving force of the PFC is the Fermi
level difference between the two photoelectrodes. The Fermi
level of the photoanode should be more negative than that of
Fig. 3 Hydrogen production of the 8 PFCs and Pt-C/Cu2O/Cu.
the photocathode, so that a bias will be produced. Then the
electrons on the photoanode will transfer through the external
circuit to combine with the holes on the photocathode.20 Holes
will be accumulated at the photoanode and electrons at the
by Pt, the hydrogen-producing rate is greatly decreased. These
photocathode. So it is a Z-scheme, and there is a cooperative
results also prove that there is a synergistic facilitating effect
synergistic facilitating effect.44–46 This is why there is a great
between the two photoelectrodes of the PFC which is conrmed
enhancement of TOC removal rate of phenol and hydrogen-
by the results of TOC removal rates of phenol. In Fig. 4, when
producing rate of the PFC compared to the single photoanode
the current density is 0.32 mA cm
2 the power density reaches a
(or photocathode) and Pt system. So the PFC can be used for
maximum (Pmax) of 0.073 mW cm
2. Table 2 lists the current–
simultaneous pollutant degradation and hydrogen production.
voltage characteristics of the 8 PFC systems using phenol (0.05
For example, TiO2 NRs/FTO and C/Cu2O NWAs/Cu mesh are
mol L
1) as substrate and Na2SO4 (0.1 mol L
1) as electrolyte.
chosen as the photoanode and photocathode, respectively. The
The ll factor (FF) which is an important indicator of output
Fermi level of the TiO2 NRs/FTO photoanode is more negative
characteristics of a PFC can be calculated based on these
than that of the C/Cu2O NWAs/Cu mesh photocathode, so in
parameters by the following equation:
this case a self-driven PFC can be operated under solar light
FF ¼ Pmax =JSC VOC (2) illumination. The remaining holes on TiO2 NRs/FTO photo-
where Pmax and JSCVOC represent the real maximum power anode and electrons on C/Cu2O NWAs/Cu mesh photocathode
density and theoretical maximum power density yielded from can be used for simultaneous organic pollutant degradation
the PFC system, respectively. JSC and VOC are short-circuit and hydrogen production.
current density and open-circuit voltage of the PFC system. Fig. 5 shows the working mode of the PFC. In this PFC, two
Table 2 also summarizes the cell parameter such as VOC, JSC, half-reactions are involved: the organics oxidized by solar light
Pmax and FF, TOC removal rates and hydrogen-producing rates illumination at the anode and water reduction to hydrogen at
of all PFCs. The results also prove the hydrogen-producing rate the cathode. The overall reaction process can be described as in
and TOC removal rate of phenol increases with JSC of PFCs. Here eqn (3)–(7):
we can see the TiO2 NRs/FTO-C/Cu2O NWAs/Cu PFC is the best
TiO2 + hn / hvb1+ + ecb1
(3)
one with its JSC of 0.50 mA cm
2, TOC removal rate of 84.2%,
C6H5OH + 5H2O + 16hvb1+ / 6CO2 + 16H+ (4)

Cu2O + hn / hvb2+ + ecb2
(5)

ecb1
+ hvb2+ / hn0 (6)

2H+ + 2ecb2
/ H2 (7)

The photovoltages of TiO2 NRs/FTO photoanode, C/Cu2O

NWAs/Cu mesh photocathode and the PFC combining them
are measured to conrm this operation mechanism. As shown
in Table 3, the photovoltages of TiO2 NRs/FTO photoanode
and C/Cu2O NWAs/Cu mesh photocathode were measured
with and without AM 1.5 illumination in an electrolyte of 0.1
mol L
1 Na2SO4. The TiO2 NRs/FTO photoanode and C/Cu2O
Fig. 4 The I–V characteristic curve (red) and power density curve
(blue) of TiO2 NRs/FTO-C/Cu2O NWAs/Cu PFC. I–V curves were NWAs/Cu mesh photocathode have a quick photoresponse
determined by performing linear sweep voltammetry with a two- and the voltage value changes from
0.08 V to
0.33 V and
electrode system in one typical cycle. The scan rate was 0.02 V s
1. from 0.02 V to 0.18 V instantaneously when the light source is

3420 | J. Mater. Chem. A, 2015, 3, 3416–3424 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry A

Table 2 The photoelectrochemical performance, TOC removal and hydrogen production of PFCs

PFC

No. Photoanode Photocathode VOC (V) JSC (mA cm
2) Pmax (mW cm
2) FF TOCa (%) H2b (mmol cm
2)

C1 TiO2 NRs C/Cu2O/Cu 0.41 0.50 0.073 0.36 84.2 86.84

C2 TiO2 NRs Cu2O 0.53 0.25 0.040 0.30 70.1 57.64
C3 TiO2 NTs/Ti C/Cu2O/Cu 0.36 0.33 0.034 0.29 72.2 61.32
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

C4 TiO2 NTs/Ti Cu2O 0.48 0.20 0.030 0.31 66.4 54.36

C5 CdS C/Cu2O/Cu 0.50 0.61 0.092 0.30 55.0 50.05
C6 CdS Cu2O 0.62 0.31 0.060 0.31 51.8 47.51
C7 CdSe C/Cu2O/Cu 0.27 0.16 0.014 0.32 48.7 37.78
C8 CdSe Cu2O 0.39 0.11 0.013 0.30 45.1 33.67
a
TOC is the TOC removal rate of phenol. b H2 is the hydrogen-producing rate.

Fig. 5 The working mode diagram of the PFC system.

turned on. So the photovoltages of TiO2 NRs/FTO photoanode valence band. So the position of the Fermi level can be deter-
and C/Cu2O NWAs/Cu mesh photocathode were 0.25 V and mined through the position of the conduction band (Ecb) and
0.16 V, respectively. Then the theoretical photovoltage of the valence band (Evb). Here, DV which is the difference between Ecb
PFC could be calculated to be 0.41 V which is the sum of the of n-type semiconductor and Evb of p-type semiconductor
photovoltage values of the TiO2 NRs/FTO photoanode and stands for the difference of Fermi level between two photo-
C/Cu2O NWAs/Cu mesh photocathode. The experimental electrodes. Table 4 shows the energy band parameters of several
photovoltage (VOC) of the PFC system is about 0.41 V which semiconductors. Theoretically, the greater the difference
matches that obtained by theoretical calculation. This result between Ecb of photoanode and Evb of photocathode, the greater
indicates that the photovoltage of the PFC authentically orig- the VOC of the PFC will be. Table 4 lists the difference between
inated from the difference between the Fermi levels of the two Ecb of photoanode and Evb of photocathode of the PFCs.
photoelectrodes. Such a PFC system can work under solar light Therefore VOC of CdS NPS/FTO-C/Cu2O NWAs/Cu PFC, TiO2
illumination without other forms of additional energy as a NRs/FTO-C/Cu2O NWAs/Cu PFC and CdSe NPS/FTO-C/Cu2O
result of the mismatch of the Fermi levels of the NWAs/Cu PFC would be in the descending order: VOC (CdS NPS/
photoelectrodes. FTO-C/Cu2O NWAs/Cu) > VOC (TiO2 NRs/FTO-C/Cu2O NWAs/Cu)
In order to further study the inuence of Fermi level of the > VOC (CdSe NPS/FTO-C/Cu2O NWAs/Cu).
semiconductor photoelectrodes in the PFC, several other Here we can see that when the photoanode of the PFC is
semiconductors, n-CdS NPS/FTO, n-CdSe NPs/FTO and n-TiO2 CdSe/FTO, VOC of the PFC is lower than the sum of photo-
NTs/Ti, have been chosen as photoanode. As we know, the voltages of photoanode and photocathode while for other PFCs
Fermi level of an n-type semiconductor is close to the conduc- it is equal to the sum. The main reason speculated is that the
tion band and that of a p-type semiconductor is close to the difference of Fermi level between CdSe and Cu2O is the smallest

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. A, 2015, 3, 3416–3424 | 3421
 View Article Online

Journal of Materials Chemistry A Paper

Table 3 The photovoltages of photoelectrodes and PFCs

PFC

No. Photoanode Photocathode V1a (V) V2b (V) VOC (V) 0

VOC ¼ V1 + V2 (V)

C1 TiO2 NRs C/Cu2O/Cu 0.25 0.16 0.41 0.41

C2 TiO2 NRs Cu2O 0.25 0.28 0.53 0.53
C3 TiO2 NTs/Ti C/Cu2O/Cu 0.20 0.16 0.36 0.36
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

C4 TiO2 NTs/Ti Cu2O 0.20 0.28 0.48 0.48

C5 CdS C/Cu2O/Cu 0.34 0.16 0.50 0.50
C6 CdS Cu2O 0.34 0.28 0.62 0.62
C7 CdSe C/Cu2O/Cu 0.11 0.16 0.19 0.27
C8 CdSe Cu2O 0.11 0.28 0.30 0.39
a
V1 is the photovoltage of photoanode. b V2 is the photovoltage of photocathode.

Table 4 Ecb of photoanode and Evb of photocathode of PFCs and for the TiO2 photoanodes with different structure, when the
difference between them22 PFCs are composed of the same photocathode, VOC of the PFC
with TiO2 NRs/FTO is higher than that with TiO2 NTs/Ti as
PFC
p-Evb n-Ecb DV ¼ (p-Evb
n-Ecb) photoanode. In Fig. 6B, the dominant crystal facet of TiO2 NTs/
Photocathode Photoanode (V) (V) (V) Ti is (101) while that of TiO2 NRs/FTO is high index crystal facets
(002) and (112). Therefore, VOC of TiO2 NRs/FTO photoanode
p-Cu2O n-CdS 1.10
0.52 1.39
PFC is higher than that of TiO2 NTs/Ti photoanode PFC. Here
p-Cu2O n-TiO2 1.10
0.29 1.62
p-Cu2O n-CdSe 1.10
0.10 1.20

of all, so that the small difference of Fermi level would not

maximize VOC of the PFC.
As shown in Table 3, VOC of CdS NPS/FTO-C/Cu2O NWAs/Cu
has a maximum value of 0.50 V among the three PFCs. And the
VOC values of TiO2 NRs/FTO-C/Cu2O NWAs/Cu and CdS NPS/
FTO-C/Cu2O NWAs/Cu are 0.41 V and 0.20 V, respectively. They
also range in the same descending sequence as theoretical
speculation. Therefore, it is reasonable that VOC of the PFC is
determined by the difference of Fermi level between the two
photoelectrodes. What is more, the VOC value of PFC increases
with the difference of Fermi level between the two
photoelectrodes.
As we know, the VOC values of the PFC are also inuenced by
many other factors. These factors can be divided into two
categories: (1) environmental factors such as temperature, pH
value and electrolyte and (2) material factors such as crystal
facet and microstructure. While the environmental factors have
been studied commonly by many researchers, the study of
material factors is fully discussed here. Therefore, Cu2O NPs/
FTO and TiO2 NTs/Ti were chosen as the reference photo-
electrodes. Table 3 also shows that the VOC value of the PFCs
with Cu2O NPs/FTO photocathode is higher than that of PFCs
with C/Cu2O NWAs/Cu photocathode when their photoanodes
are the same. Herein, the key factor that inuences VOC of PFCs
is the crystal facet structure of Cu2O.
In Fig. 6A, the XRD pattern of Cu2O/FTO shows the dominant
crystal facet of Cu2O is (220) which is the high index crystal facet
while the dominant crystal facet of C/Cu2O NWAs/Cu is (111).
The (111) facet has the best photocatalytic ability. This is
because the high index crystal facet (220) has a higher surface Fig. 6 (A) The XRD patterns of Cu2O NPs/FTO and C/Cu2O NWAs/Cu.
energy than the low index crystal facet (111). In a similar way, (B) The XRD patterns of TiO2 NRs/FTO and TiO2 NTs/Ti.

3422 | J. Mater. Chem. A, 2015, 3, 3416–3424 This journal is © The Royal Society of Chemistry 2015
 View Article Online

Paper Journal of Materials Chemistry A

we can see that in addition to the difference of Fermi level, the 8 F. F. Ajayi, K. Y. Kim, K. J. Chae, M. J. Choi, S. Y. Kim,
microstructure such as crystal facet structure of the photo- I. S. Chang and I. S. Kim, Int. J. Hydrogen Energy, 2009, 34,
electrodes also inuences VOC of PFCs. VOC of PFCs increases 9297–9304.
with the ratio of high index crystal facet of the photoelectrodes. 9 Y. Gao, X. Ding, J. Liu, L. Wang, Z. Lu, L. Li and L. Sun, J. Am.
Chem. Soc., 2013, 135, 4219–4222.
10 B. Zhang, W. F. Yao, C. P. Huang, Q. J. Xu and Q. Wu, Int. J.
Conclusions Hydrogen Energy, 2013, 38, 7224–7231.
11 A. Paracchino, V. Laporte, K. Sivula, M. Grätzel and
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

In this work, a dual PFC has been fabricated, which can work E. Thimsen, Nat. Mater., 2011, 10, 456–461.
without an external energy input except for light illumination. 12 Q. Li, B. Guo, J. Yu, J. Ran, B. Zhang, H. Yan and J. R. Gong, J.
Several n-type and p-type semiconductors have been chosen as Am. Chem. Soc., 2011, 133, 10878–10884.
the photoanode and photocathode, respectively. The PFC has a 13 C. Y. Lin, Y. H. Lai, D. Mersch and E. Reisner, Chem. Sci.,
superior VOC of 0.41 V and a JSC of 0.50 mA cm
2 with TiO2 NRs/ 2012, 3, 3482.
FTO as photoanode and C/Cu2O NWAs/Cu as photocathode. 14 M. Antoniadou, D. I. Kondarides, D. Labou, S. Neophytides
The best TOC removal rate of phenol and hydrogen production and P. Lianos, Sol. Energy Mater. Sol. Cells, 2010, 94, 592–597.
rate in 8 hours reaches 84.2% and 86.8 mmol cm
2, which is 15 R. L. Chamousis and F. E. Osterloh, ChemSusChem, 2012, 5,
much higher than reported by other similar studies. For the 1482–1487.
investigation of mechanism, the VOC value of PFC is mainly 16 Y. Liu, J. Li, B. Zhou, H. Chen, Z. Wang and W. Cai, Chem.
inuenced by the difference of Fermi level between the two Commun., 2011, 47, 10314–10316.
photoelectrodes. For the same semiconductors with different 17 Y. Liu, J. Li, B. Zhou, S. Lv, X. Li, H. Chen, Q. Chen and
microstructure, the crystal facet properties also inuence VOC of W. Cai, Appl. Catal., B, 2012, 111–112, 485–491.
the PFC. And the VOC value of the PFC increases with the ratio of 18 B. Wang, H. Zhang, X. Lu, J. Xin and M. K. H. Leung, Chem.
high index crystal facet of photoelectrodes. However, JSC of the Eng. J., 2014, 253, 174–182.
PFC is not only inuenced by the difference of Fermi level 19 K. Li, Y. Xu, Y. He, C. Yang, Y. Wang and J. Jia, Environ. Sci.
between the two photoelectrodes but also by the photoresponse Technol., 2013, 47, 3490–3497.
of the photocatalyst and electrical conductivity of the substrate 20 Q. P. Chen, J. H. Li, X. J. Li, K. Huang, B. X. Zhou, W. M. Cai
of the photocathode. The higher the photocurrent and the and W. F. Shangguan, Environ. Sci. Technol., 2012, 46, 11451–
better the electrical conductivity, the higher the JSC achieved for 11458.
the PFC. The trends of degradation for phenol and hydrogen 21 J. Y. Li, J. H. Li, Q. P. Chen, J. Bai and B. X. Zhou, J. Hazard.
production are the same as those of JSC. So this work provides a Mater., 2013, 262, 304–310.
novel and clear theoretical guidance for photoelectrode selec- 22 M. Grätzel, Nature, 2001, 414, 338–344.
tion and exploits a new application of PFCs for hydrogen 23 B. Kumar, M. Llorente, J. Froehlich, T. Dang, A. Sathrum and
production. C. P. Kubiak, Annu. Rev. Phys. Chem., 2012, 63, 541–569.
24 A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
25 Q. Zheng, B. Zhou, J. Bai, L. Li, Z. Jin, J. Zhang, J. Li, Y. Liu,
Acknowledgements W. Cai and X. Zhu, Adv. Mater., 2008, 20, 1044–1049.
This work was supported by the National Natural Science 26 Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han and C. Li, Chem. Rev.,
Foundations of China (NSFC, 21477085 and 21277099). 2014, 114(19), 9987–10043.
27 A. L. Linsebigler, G. Lu and J. T. Yates, Chem. Rev., 1995,
95(3), 735–758.
Notes and references 28 L. Sang, Y. Zhao and C. Burda, Chem. Rev., 2014, 114(19),
9283–9318.
1 S. Chu and A. Majumdar, Nature, 2012, 488, 294–303. 29 H. J. Zhang, G. H. Chen and D. Bahnemann, J. Mater. Chem.,
2 S. B. Grant, J. D. Saphores, D. L. Feldman, A. J. Hamilton, 2009, 19, 5089–5121.
T. D. Fletcher, P. L. M. Cook, M. Stewardson, B. F. Sanders, 30 Y. Hou, X. Y. Li, Q. D. Zhao, G. H. Chen and C. L. Raston,
L. A. Levin, R. F. Ambrose, A. Deletic, R. Brown, S. C. Jiang, Environ. Sci. Technol., 2012, 46, 4042–4050.
D. Rosso, W. J. Cooper and I. Marusic, Science, 2012, 337, 31 M. Wang, L. Sun, Z. Lin, J. Cai, K. Xie and C. Lin, Energy
681–686. Environ. Sci., 2013, 6, 1211–1220.
3 M. R. Hoffmann, S. T. Martin, W. Choi and 32 W. C. Huang, L. M. Lyu, Y. C. Yang and M. H. Huang, J. Am.
D. W. Bahnemann, Chem. Rev., 1995, 95, 69–96. Chem. Soc., 2012, 134(2), 1261–1267.
4 B. E. Logan and K. Rabaey, Science, 2012, 337, 686–690. 33 E. M. Zahran, N. M. Bedford, M. A. Nguyen, Y. J. Chang,
5 B. E. Logan, B. Hamelers, R. Rozendal, U. Schröder, J. Keller, B. S. Guiton, R. R. Naik, L. G. Bachas and M. R. Knecht, J.
S. Freguia, P. Aelterman, W. Verstraete and K. Rabaey, Am. Chem. Soc., 2014, 136(1), 32–35.
Environ. Sci. Technol., 2006, 40(17), 5181–5192. 34 P. D. Tran, L. H. Wong, J. Barber and J. S. C. Loo, Energy
6 A. Rinaldi, B. Mecheri, V. Garavaglia, S. Licoccia, P. D. Nardo Environ. Sci., 2012, 5, 5902–5918.
and E. Traversa, Energy Environ. Sci., 2008, 1, 417–429. 35 S. Song, Z. Liu, Z. He, A. Zhang, J. Chen, Y. Yang and X. Xu,
7 F. Qian, G. Wang and Y. Li, Nano Lett., 2010, 10, 4686–4691. Environ. Sci. Technol., 2010, 44(10), 3913–3918.

This journal is © The Royal Society of Chemistry 2015 J. Mater. Chem. A, 2015, 3, 3416–3424 | 3423
 View Article Online

Journal of Materials Chemistry A Paper

36 R. Vinu and G. Madras, Environ. Sci. Technol., 2008, 42(3), 41 Z. Zhang, R. Dua, L. Zhang, H. Zhu, H. Zhang and P. Wang,
913–919. ACS Nano, 2013, 7, 1709–1717.
37 J. H. Bang and P. V. Kamat, Adv. Funct. Mater., 2010, 20, 42 Q. T. Du, J. S. Tan, Q. T. Wang, C. Y. Li, X. H. Liu, R. S. Cai,
1970–1976. Y. H. Ding and Y. Q. Wang, J. Anal. Methods Chem., 2012,
38 S. N. Chai, G. H. Zhao, Y. N. Zhang, Y. J. Wang, F. Q. Nong, 406162, 1–8.
M. F. Li and D. M. Li, Environ. Sci. Technol., 2012, 46(18), 43 G. L. Zang, G. P. Sheng, C. Shi, Y. K. Wang, W. W. Li and
10182–10190. H. Q. Yu, Energy Environ. Sci., 2014, 7, 3033–3039.
39 L. F. Fan, G. H. Zhao, H. J. Shi, M. C. Liu, Y. A. Wang and 44 H. Tada, T. Mitsui, T. Kiyonaga, T. Akita and K. Tanaka, Nat.
Published on 11 December 2014. Downloaded by Federal University of Ceará on 10/18/2023 5:38:55 PM.

H. Y. Ke, Environ. Sci. Technol., 2014, 48(10), 5754–5761. Mater., 2006, 5, 782–786.
40 S. Banerjee, S. K. Mohapatra, P. P. Das and M. Misra, Chem. 45 K. Maeda, ACS Catal., 2013, 3, 1486–1503.
Mater., 2008, 20, 6784–6791. 46 P. Zhou, J. Yu and M. Jaroniec, Adv. Mater., 2014, 26, 4920–
4935.

3424 | J. Mater. Chem. A, 2015, 3, 3416–3424 This journal is © The Royal Society of Chemistry 2015