Download Full-Text PDF

inorganics
Review
Dye-Sensitized Photocatalytic Water Splitting and
Sacrificial Hydrogen Generation: Current Status and
Future Prospects
Pankaj Chowdhury 1,2 , Ghodsieh Malekshoar 1 and Ajay K. Ray 1, *
1
2
*
Department of Chemical and Biochemical Engineering, Western University, London, ON N6A 5B9, Canada;
[email protected] (P.C.); [email protected] (G.M.)
Trojan Technologies, London, ON N5V 4T7, Canada
Correspondence: [email protected]; Tel.: +1-519-661-2111 (ext. 81279)
Academic Editor: Matthias Bauer
Received: 16 March 2017; Accepted: 6 May 2017; Published: 18 May 2017
Abstract: Today, global warming and green energy are important topics of discussion for every
intellectual gathering all over the world. The only sustainable solution to these problems is the use of
solar energy and storing it as hydrogen fuel. Photocatalytic and photo-electrochemical water splitting
and sacrificial hydrogen generation show a promise for future energy generation from renewable
water and sunlight. This article mainly reviews the current research progress on photocatalytic
and photo-electrochemical systems focusing on dye-sensitized overall water splitting and sacrificial
hydrogen generation. An overview of significant parameters including dyes, sacrificial agents,
modified photocatalysts and co-catalysts are provided. Also, the significance of statistical analysis as
an effective tool for a systematic investigation of the effects of different factors and their interactions
are explained. Finally, different photocatalytic reactor configurations that are currently in use for
water splitting application in laboratory and large scale are discussed.
Keywords: dye-sensitization; photocatalytic; photoelectrochemical; hydrogen; water splitting; photoreactor
1. Introduction
Global energy systems transition shows a clear energy shift from coal to oil to hydrogen
(Figure 1) [1]. The element hydrogen is found in abundance in the universe—in water, life forms,
and hydrogen fuel. The three major issues, among different driving forces that contribute to the energy
transition towards hydrogen, are (i) growing requirement for energy; (ii) insufficiency of oil in the
immediate future and (iii) possibilities of climate change. By 2050 the expected energy demand will be
doubled as world population grows to 10 billion. Coal and oil are promising energy sources, but we
need some other energy sources as well that are present in much larger quantities and can meet the
future energy demand [2].
Human activity, approximately, produces 6 billion tons of carbon dioxide annually, with 80%
coming from the burning of the fossil fuels. Coal generates 430 g of CO2 /kWh, while natural gas
produces 190 g of CO2 /kWh [3]. The atmospheric CO2 level of 270 ppm during industrial revolution
rose to 370 ppm during the 20th century, and the current level of CO2 is 406 ppm [4].
Currently, most hydrogen is produced and used by chemical and petroleum industries through
natural gas reforming, coal gasification, thermal water splitting, and electrolysis. About 96% of
worlds’ hydrogen is currently produced from natural gas, oil or coal (48% from natural gas, 30% from
oil, 18% from coal, and the remaining 4% via water electrolysis) [3]. These processes require big
infrastructure, significant energy sources and run enormous environmental cost. Furthermore,
if hydrogen is not generated on-site, a greater quantity of energy is needed to distribute the element.
Inorganics 2017, 5, 34; doi:10.3390/inorganics5020034
www.mdpi.com/journal/inorganics
Inorganics 2017, 5, 34
Inorganics 2017, 5, 34
2 of 35
2 of 35
used to
to produce
produce hydrogen,
hydrogen, but
but the
the basic
basic feedstock
feedstock is
is inadequate
inadequate and
and would
would
Biomass could also be used
also need
need large
large energy
energy input
input [2,3,5].
[2,3,5].
also
Figure 1.1.Global
Global
energy
system
transition,
1850–2150.
Reproduced
with permission
from [1].
Figure
energy
system
transition,
1850–2150.
Reproduced
with permission
from [1]. Copyright
Copyright2002.
Elseriver, 2002.
Elseriver,
The most promising technology for hydrogen generation from renewable water and sunlight
The most promising technology for hydrogen generation from renewable water and sunlight
would be photocatalytic water splitting. Fujishima and Honda demonstrated water splitting in the
would be photocatalytic water splitting. Fujishima and Honda demonstrated water splitting in the early
early 1970s, with a TiO2 photoanode and Pt cathode in the presence of UV light irradiation.
1970s, with a TiO2 photoanode and Pt cathode in the presence of UV light irradiation. Additionally,
Additionally, a power supply or pH difference between a catholyte and an anolyte were used to
a power supply or pH difference between a catholyte and an anolyte were used to apply some external
apply some external bias to the above process [6]. The last few decades saw a notable progress in this
bias to the above process [6]. The last few decades saw a notable progress in this research area using
research area using UV light [7]. The elements of group IV (Ti and Zr), group V (Nb and Tc) and
UV light [7]. The elements of group IV (Ti and Zr), group V (Nb and Tc) and group VI (W) acted
group VI (W) acted as wide band gap photocatalytic materials for water splitting. In spite of such
as wide band gap photocatalytic materials for water splitting. In spite of such development over
development over time, photocatalytic systems for water splitting as well as hydrogen generations
time, photocatalytic systems for water splitting as well as hydrogen generations are still experiencing
are still experiencing many technical challenges. The narrow wavelength range of UV light is its
many technical challenges. The narrow wavelength range of UV light is its fundamental difficulty,
fundamental difficulty, and therefore utilization of the more profuse visible light for photocatalytic
and therefore utilization of the more profuse visible light for photocatalytic reactions appears to be a
reactions appears to be a promising alternative [2].
promising alternative [2].
Dye-sensitization is an innovative technology that could play a major role in developing an
Dye-sensitization is an innovative technology that could play a major role in developing
efficient and cost-effective visible light active semiconductor photocatalyst in the near future. Dyean efficient and cost-effective visible light active semiconductor photocatalyst in the near future.
sensitization technology, so far, has been used in solar cell research to generate electricity. Presently,
Dye-sensitization technology, so far, has been used in solar cell research to generate electricity.
dye-sensitization processes also found applications in visible-light-assisted photocatalytic water
Presently, dye-sensitization processes also found applications in visible-light-assisted photocatalytic
splitting to produce hydrogen. Several authors have used numerous dyes such as porphyrins,
water splitting to produce hydrogen. Several authors have used numerous dyes such as
coumarin, phthalocyanines and carboxylate derivatives of anthracene to sensitize different
porphyrins, coumarin, phthalocyanines and carboxylate derivatives of anthracene to sensitize different
semiconductor photocatalysts. Among the various photosensitizers, transition metal-based
semiconductor photocatalysts. Among the various photosensitizers, transition metal-based sensitizers6
sensitizers have been shown to be the best. Transition metals such as Ru(II), Fe(II) and Os(II) form d
have been shown to be the best. Transition metals such as Ru(II), Fe(II) and Os(II) form d6 complex
complex and undertake strong charge transfer absorption throughout the visible range [8]. Ru(II)
and undertake strong charge transfer absorption throughout the visible range [8]. Ru(II) polypyridine
polypyridine complexes are mostly applied in dye-sensitized solar cells. In the field of dye-sensitized
complexes are mostly applied in dye-sensitized solar cells. In the field of dye-sensitized solar cell,
solar cell, Gratzel stated cis-[Ru(dcbH2)2(NCS)2] (N3), as the best sensitizer, where thiocyanate (NCS−)
Gratzel stated cis-[Ru(dcbH2 )2 (NCS)2 ] (N3), as the best sensitizer, where thiocyanate (NCS− ) was
was applied as an ancillary ligand [9].
applied as an ancillary ligand [9].
In the case of photocatalytic overall water splitting/sacrificial hydrogen generation, researchers
In the case of photocatalytic overall water splitting/sacrificial hydrogen generation, researchers have
have widely used different organic dyes, inorganic sensitizers, and coordination metal complexes.
widely used different organic dyes, inorganic sensitizers, and coordination metal complexes.
The toxicity and high cost of Ruthenium-based dyes make them uneconomical in industrial
The toxicity and high cost of Ruthenium-based dyes make them uneconomical in industrial
applications. On the other hand, organic dyes are less toxic, low-cost, and can be used for large-scale
applications. On the other hand, organic dyes are less toxic, low-cost, and can be used for large-scale
dye sensitization processes. Several organic dyes such as Eosin Y, Rose Bengal, Merocyanine, Cresyl
dye sensitization processes. Several organic dyes such as Eosin Y, Rose Bengal, Merocyanine,
violet and Riboflavin have been employed for spectral sensitization of semiconductors photocatalysts
Cresyl violet and Riboflavin have been employed for spectral sensitization of semiconductors
[10,11].
photocatalysts [10,11].
We begin the review with the explanation of thermodynamics and kinetics of water splitting
processes. Then we focus on the research progresses on visible-light-assisted overall water splitting
Inorganics 2017, 5, 34
3 of 35
We begin the review with the explanation of thermodynamics and kinetics of water splitting
processes. Then we focus on the research progresses on visible-light-assisted overall water
splitting and sacrificial hydrogen generation with photocatalytic and photoelectrochemical systems.
Basic phenomena
dye-sensitization are well described here in the areas mentioned above.3 ofThe
Inorganics 2017,of
5, 34
35 review
Inorganics 2017,
34
3 ofhydrogen
35
also elucidates
the 5,usefulness
of design of experiment as an effective tool for photocatalytic
and sacrificial hydrogen generation with photocatalytic and photoelectrochemical systems. Basic
generation studies. Finally, we discuss different photocatalytic reactor configurations that are either in
phenomena
of dye-sensitization
are well
here inand
the photoelectrochemical
areas mentioned above.
The review
and sacrificial
hydrogen generation
withdescribed
photocatalytic
systems.
Basic
use or will
be applied in future in photocatalytic water splitting and hydrogen generation systems.
also elucidates the usefulness of design of experiment as an effective tool for photocatalytic hydrogen
phenomena of dye-sensitization are well described here in the areas mentioned above. The review
generation
studies.
Finally, weofdiscuss
photocatalytic
reactor
configurations
that are
either
also elucidates
the usefulness
designdifferent
of experiment
as an effective
tool
for photocatalytic
hydrogen
2. Photocatalytic/Photoelectrochemical
Water Splitting:
Thermodynamics
and Kinetics
in
use or willstudies.
be applied
in future
in photocatalytic
water splitting
and hydrogen
generation
systems.
generation
Finally,
we discuss
different photocatalytic
reactor
configurations
that are
either
in
use
or
will
be
applied
in
future
in
photocatalytic
water
splitting
and
hydrogen
generation
systems.
Salvador systematically explained the general photocatalytic water splitting reaction scheme
2. Photocatalytic/Photoelectrochemical Water Splitting: Thermodynamics and Kinetics
in the presence of a semiconductor photocatalyst (Equations (1)–(4)) [12]. Upon illumination of
2. Photocatalytic/Photoelectrochemical
Water
Splitting:
Thermodynamics
and Kinetics
Salvador systematically explained the
general
photocatalytic
water splitting
reaction scheme in
light with photon energy greater than or equal to the bandgap energy of the semiconductor, electrons,
the presence
a semiconductor
photocatalyst
(Equations
(1)–(4))water
[12]. splitting
Upon illumination
of light
Salvadorofsystematically
explained
the general
photocatalytic
reaction scheme
in
and holes are generated in conduction and valence bands of the semiconductor, respectively. The charge
with
photon
energy
greater
than
or
equal
to
the
bandgap
energy
of
the
semiconductor,
electrons,
and
the presence of a semiconductor photocatalyst (Equations (1)–(4)) [12]. Upon illumination of light
carriersholes
finally
and/or
reduce
water
molecules
gaseousrespectively.
hydrogen
andcharge
oxygen
in 2:1
areoxidize
generated
conduction
valence
ofproducing
the
semiconductor,
The
with photon
energyin
greater
than orand
equal
to thebands
bandgap
energy
of the semiconductor,
electrons,
and
stoichiometric
ratio
by
volume
(Figure
2)water
[13].
Operating
principle
of photo-electrolytic
water
carriers
finally
oxidize
and/or
reduce
molecules
gaseous
hydrogen
and The
oxygen
insplitting
holes are
generated
in conduction
and
valence
bands ofproducing
the
semiconductor,
respectively.
charge
2:1
stoichiometric
ratio and/or
by
volume
(Figure
2)molecules
[13].
Operating
principle
ofwater
photo-electrolytic
water
in the photoelectrochemical
cell
isreduce
quite
similar
to the
photocatalytic
splitting
(Figure
carriers
finally oxidize
water
producing
gaseous
hydrogen
and oxygen
in 3) [13].
splitting
in
the
photoelectrochemical
cell
is
quite
similar
to
the
photocatalytic
water
splitting
(Figure
2:1
stoichiometric
ratio
by
volume
(Figure
2)
[13].
Operating
principle
of
photo-electrolytic
water
At the interface between the electrolyte and photo electrode, the photo-generated holes react with
[13]. Atinthe
between the electrolyte
and
photo
the photo-generated
holes
react
splitting
theinterface
photoelectrochemical
cell is quite
similar
the photocatalytic
waterelectrons
splitting
(Figure
water to3)
generate
gaseous
oxygen
and hydrogen
ion
(H+ ).toelectrode,
The
photo-generated
flow
through
with
water
to generate
and hydrogen
ion electrode,
(H+). The photo-generated
electrons
3) [13].
At the
interface gaseous
betweenoxygen
the electrolyte
and photo
the photo-generated
holes flow
react
the external
circuit
to
the
photocathode,
captured
by
the
hydrogen
ion
at
the
photocathode-electrolyte
+). The
through
the to
external
circuit
to theoxygen
photocathode,
captured
hydrogen
ion at the photocathodewith water
generate
gaseous
and hydrogen
ionby
(Hthe
photo-generated
electrons flow
boundary
and
produce
hydrogen
gas.
electrolyte
boundary
and
produce
hydrogen
gas.
through the external circuit to the photocathode, captured by the hydrogen ion at the photocathodeelectrolyte boundary and produce hydrogen gas.
Figure 2. Aux-GSH-sensitized TiO2 film as photoanode of PEC. Reproduced with permission from [13].
Figure 2. Aux -GSH-sensitized TiO2 film as photoanode of PEC. Reproduced with permission from [13].
Copyright
American
ChemicalTiO
Society,
2014.
Figure 2. Au
x-GSH-sensitized
2 film as
photoanode of PEC. Reproduced with permission from [13].
Copyright American Chemical Society, 2014.
Copyright American Chemical Society, 2014.
Figure 3. Aux-GSH-sensitized Pt/TiO2 NPs photocatalytic system for water splitting reaction under
visible light illumination. Reproduced with permission from [13]. Copyright American Chemical
Figure
3. Aux-GSH-sensitized
Pt/TiO2NPs
NPs photocatalytic
photocatalytic system
for for
water
splitting
reaction
under under
Figure 3.
Aux -GSH-sensitized
Pt/TiO
system
water
splitting
reaction
2
Society,
2014. illumination. Reproduced with permission from [13]. Copyright American Chemical
visible light
visible light illumination. Reproduced with permission from [13]. Copyright American Chemical
Society, 2014.
Society, 2014.
Inorganics 2017, 5, 34
4 of 35
Photocatalytic water splitting reaction scheme:
Semiconductor + 2hϑ → 2h+ + 2e−
1
O2 + 2H + EO2 /H2 O
2
+
−
2H + 2e → H2 EH2 O/H2
H2 O + 2h+ →
1
hϑ
H2 O → H2 ( g) + O2 ( g)
2
(1)
(2)
(3)
(4)
2.1. Thermodynamics of Water Splitting
The photon energy is transformed into chemical energy through water splitting giving a positive
value of Gibbs free energy. On the other hand, photocatalytic degradation reactions are downhill
reactions. This downhill-type reaction is known as a photoinduced reaction and has been widely
studied using TiO2 photocatalysts [6]. At standard conditions, electrolysis of water reversibly produces
a stoichiometric mixture of hydrogen and oxygen (2:1) at a potential of 1.23 V which is obtained from
the equation of Gibbs free energy:
∆Go = −nF(∆Eo )
(5)
where, ∆Go is standard Gibbs free energy change, n is the number of electrons, F is Faraday’s constant
and ∆Eo is the standard electrical potential of the reaction. The electromotive force or EMF (∆E) is
defined as the maximum potential difference of an electrochemical reaction [3].
The Gibbs free energy change (∆G) at STP is positive for an endothermic process such as water
splitting reaction (Equation (4)). Such a non-spontaneous reaction will only occur if adequate energy is
provided by the incident photons. The standard Gibbs free energy change (∆Go ) is the negative value of
maximum electrical work corresponding to 237.14 kJ·mol−1 or 2.46 eV for Equation (4). Since this is a
two-electron redox process, photocatalytic water splitting is possible if the semiconductor photocatalyst
possesses a band gap energy (Eg ) greater than 1.23 eV [3,14].
The two energetic requirements for reactions (2) and (3) to take place spontaneously are:
En∗ = EC > E(H2 O/H2 )
(6)
Ep∗ = EV < E(O2 /H2 O)
(7)
where, En *, Ep * are the individual energies of photo-generated electrons and holes respectively. It is
assumed that after thermalization within the semiconductor bands En * and Ep * reach energy levels
near Ev (valence band energy) and Ec (conduction band energy), respectively [12]. According to
Equations (6) and (7) the energy of photoelectrons must be above the energy level of the (H2 O/H2 )
redox couple and the energy of photo holes below the energy level of the (O2 /H2 O) redox couple.
This is shown in the following equations:
E(H2 O/H2 ) − E(O2 /H2 O) = 1.23 eV
(8)
Now in the absence of any over potential, the minimum semiconductor band gap is given by
Equation (9):
Eg = EC − EV = 1.23 eV
(9)
2.2. Kinetic Constraint of Water Splitting
The kinetic requirement of water splitting reaction is very well explained by Salvador [12]. In this
section, we are mainly presenting his ideas in modifying the band gap energy. In Equation (9) he has
not considered the overpotential for oxygen (η O2 ) and hydrogen (η H2 ) evolution, the energy difference
Inorganics 2017, 5, 34
5 of 35
between EC and EF (energy of equilibrium Fermi level at dark) and electrolyte resistance (η R ). Thus,
the revised equation for band gap energy (Eg ) would be as follows:
Eg = 1.23 + ηO2 + ηH2 + ηR + (EC + EF )
(10)
All these over potential values are maintained on the lower side to minimize the band gap
energy. The oxygen over potential control is slightly difficult than the other factors. In electrolytic
oxygen evolution on metals, at first chemisorbed water molecule reacts with a hole (h+ ) and generate
HO• radical which is not applicable for semiconductors. A water molecule can more easily inject
electrons into the semiconductor VB (valence band) under illumination when they are chemisorbed
since electron transfer can then happen directly without the necessity of a tunneling process [12]. If we
consider the energy diagram of semiconductor and electrolyte levels, we will see that the upper level
of water VB (4.65 eV), is below E(H2 O/H2 ) , i.e., for water splitting Eg ≥ 4.65 eV. Again, photogeneration
of hole occurs at the upper level of the semiconductor VB which is then transferred to filled extrinsic
level of solvated HO− ions. Then Eg ≥ E(H2 O/H2 ) − E(HO• /HO− ) = 2.80 eV. In both cases, the band gap
energy (Eg ) shows much higher value than the optimum Eg value [12].
To avoid such limitations, the following reaction scheme was considered:
M − (H2 O)S + h+ → M − (HO• )S + H+
(11)
where, s denotes the surface adsorbed species; M denotes transition metals surface to which water and
intermediate (HO• )S species are organized. Here water molecules are chemisorbed, and holes can be
captured inelastically by band gap surface states associated with water molecules [12]. Reaction (12) is
possible if the following energy criterion is achieved:
[E(HO• /HO− ) − EV ] ≥ Ea
S
(12)
where, Ea is the energy of activation which is basically the reorganization energy (Ψ) of the E(HO• /HO− )s
redox couple. If the HO− is strongly adsorbed onto the semiconductor surface then the interaction
with the surrounding water molecule will be less and Ψ value will be low, which ultimately reduces
the overpotential for the Reaction (12). The shift of E(HO• /HO− )s in the adsorbed state with respect to
the same couple in the aqueous phase is given by:
E(HO• /HO− ) − E(HO• /HO− ) = ∆Gads HO− − ∆Gads (HO• )
s
aq
(13)
Therefore, if the interaction with the semiconductor of both oxidation states of the redox couple
(HO• and HO− species) is the same [i.e., ∆Gads (HO• ) = ∆Gads (HO− )], the redox potential of the system
in the adsorbed state will be the same as in the solution [12].
3. Visible-Light-Driven Overall Water Splitting/Sacrificial Hydrogen Generation
Fujishima and Honda described the photocatalytic water splitting over a TiO2 single crystal
in 1972 [15]. After that, a remarkable progress on water splitting has been observed in the last few
decades under UV light [7,16]. The basic problem with UV photocatalysis was a 4% solar spectrum
compared to visible light (46%). We know that the band gap energy (Eg ) > 1.23 eV for photocatalytic
water splitting from the thermodynamic point of view. Now the semiconductor photocatalyst will
be visible light active if the band gap energy is <3.0 eV. The photocatalytic materials for visible light
water splitting should possess a suitable band gap energy (1.23 eV < Eg < 3.0 eV). Also the band
position of the semiconductor plays a significant role in visible light excitation of the photocatalyst [14].
Photocatalysts such as TiO2 , ZnO, and SnO2 have large band gaps (3–3.8 eV) and utilize only UV
portion of the solar light, and therefore displays low conversion efficiencies. Figure 4 shows band
positions of several semiconductor photocatalysts in contact with aqueous electrolyte at pH 1 [3].
Inorganics 2017, 5, 34
6 of 35
Only a few chalcogenides (CdS, CdSe, etc.) have a band gaps between 1.23 eV and 3.0 eV, which can
be stimulated by visible light. However, these photocatalysts cannot be used because of the severe
Inorganics 2017, 5, 34problem. Currently, the energy conversion efficiency from solar to hydrogen through
6 of 35
photo-corrosion
photocatalytic water splitting is very low. This is primarily for two reasons: (i) rapid recombination
approaches
been and
applied
to modify
photocatalyst
to radiation.
make themDifferent
visible light
active. Some
of
the chargehave
carriers;
(ii) inability
to the
utilize
visible solar
approaches
have
popular
approaches
include
doping
with
cation/anion,
valence
band
controlled
photocatalyst,
been applied to modify the photocatalyst to make them visible light active. Some popular approaches
composite
photocatalyst,
spectral-sensitization
and metal
ion implantation
[17].
include
doping
with cation/anion,
valence band(dye-sensitization)
controlled photocatalyst,
composite
photocatalyst,
Table 1 shows a comparative
study of different
photocatalyst
modification
methods
along
with their
spectral-sensitization
(dye-sensitization)
and metal
ion implantation
[17]. Table
1 shows
a comparative
advantages
and
limitations.
study of different photocatalyst modification methods along with their advantages and limitations.
position of
of different
differentphotocatalyst
photocatalystmaterial
materialininpresence
presenceofofaqueous
aqueouselectrolyte
electrolyte
=
Figure 4. Band position
at at
pHpH
= 1.
Reproduced
with
permission
from
[3].
Copyright
Springer,
2008.
1. Reproduced with permission from [3]. Copyright Springer, 2008.
Inorganics 2017, 5, 34
7 of 35
Table 1. Modification of photocatalyst materials for visible light activity.
Photocatalyst Modification
Technique
Mechanism of the Process
1.
Noble metal (Au, Ag, Cu, Pt,
Ru, etc.) loading
Fermi levels of the noble metals are less than
the semiconductor, and thus the conduction
band electron transfer occurs more easily to
the metal particles deposited on the surface
of the semiconductor
2.
Doping with transition metal
ions such as Cr, Co, Ni, Mn,
and Fe
No.
Results/Comments
References
(i) suppress the e− /h+ recombination
(ii) help electron transfer in water-splitting
(iii) noble metals are expensive
(iv) much higher metal deposition results
lower efficiency
platinum and gold loading is
beneficial than palladium loading
due to their opposite work function
and electron affinity
[17–21]
Upon incorporation of metal ion into
semiconductor photocatalyst, an impurity
energy level is formed in the band gap
of semiconductor
(i) doping with transition metal ions expand the
photo-response of semiconductor photocatalyst
into the visible region
(ii) deep doping create recombination centers
Metal ions doped near the
semiconductor photocatalyst
surface for efficient charge transfer
[17,22]
3.
Doping with anions such as C,
P, S, N, O, and F
Upon doping band gap narrowing occurs in
the semiconductor photocatalyst
(i) doping with anions in TiO2 shift its
photo-response to the visible region
(ii) anions hardly form any recombination
centers like cations
TiO2 band gap narrow down
occurs due to the overlapping of 2p
orbitals of O with p orbitals of N
[17,23]
4.
Composite semiconductor;
small band gap CdS (2.4 eV)
and large band gap
semiconductors such as TiO2
(3.2 eV), SnO2 (3.5 eV)
A large band gap semiconductor is attached
to a small band gap semiconductor;
conduction band electrons are injected from
the small band gap semiconductor to the
large band gap semiconductor
(i) combination of small and large band gap
semiconductors provide improved
photocatalytic performance due to adequate
charge separation
CdS–SnO2 system produces
hydrogen under visible light
irradiation in the presence of EDTA;
CdS–TiO2 system produces
hydrogen under visible light in the
presence of Na2 S.
[17,24,25]
5.
Valence band
controlled photocatalyst
Metals such as bismuth (6s orbital), tin
(5s orbital) and silver (4d orbital) contribute
to the formation of valence band of these
photocatalysts above the 2p orbital
of oxygen
(i) this process makes the photocatalysts visible
light active
BiVO4 and AgNbO3 are suitable
for O2 evolution, and Pt/SnNb2 O6
is suitable for H2 evolution
[17,26–28]
6.
Metal ion implantation
In this method, high energy ions are injected
into the semiconductor which modifies the
electronic structure
(i) overlapping of the d-orbital of Titanium with
implant metal d-orbital occurs and that leads to
a reduction of the band gap
Metal ion implantation is very
efficient for red shift, and no
electron mediator is required
Dye-sensitization
Upon visible light illumination, the excited
dye molecule introduce electron into the
conduction band of semiconductor to begin
the photocatalytic reaction; the conduction
band electron is transmitted to noble metal
at the semiconductor surface to undertake
water reduction
(i) hydrogen generation rate is improved by dye
absorption onto photocatalyst surface
(ii) careful design of dye and semiconductor
couple is necessary to prevent charge trapping
and recombination which lowers the efficiency
of the dye-sensitized semiconductor
Transition metal-based dyes are the
best dyes regarding intense charge
transfer in the visible spectral range
7.
Advantage/Disadvantage
[17]
[8,17]
Inorganics 2017, 5, 34
8 of 35
The
following
in
Inorganics
2017, 5, 34section will discuss the basics and finer details of dye-sensitization processes
8 of 35
photoelectrochemical and photocatalytic water splitting.
The following section will discuss the basics and finer details of dye-sensitization processes in
photoelectrochemical
and photocatalytic
water
splitting.
4. Dye-Sensitization
Process:
Visible Light
Active
Photocatalyst
The
process of enlarging
photocatalytic
of semiconductor to the visible region is
4. Dye-Sensitization
Process:the
Visible
Light Activeactivity
Photocatalyst
identified as spectral sensitization. In the case of dye-sensitization, the sensitization of a large band
The process of enlarging the photocatalytic activity of semiconductor to the visible region is
gap semiconductor is accomplished with a dye [29]. The history of the semiconductor sensitization has
identified as spectral sensitization. In the case of dye-sensitization, the sensitization of a large band
been gap
explained
explicitly
by Hagfeldt and
[30].
The
mechanism
of photoelectrochemistry
semiconductor
is accomplished
withGratzel
a dye [29].
The
history
of the semiconductor
sensitization and
photography
are very similar
regarding
photo-assisted
charge
separationof
atphotoelectrochemistry
the liquid-solid interface.
has been explained
explicitly
by Hagfeldt
and Gratzel [30].
The mechanism
The process
of spectralare
sensitization
an n-type
semiconductor,
alsoseparation
known asatanodic
sensitization,
and photography
very similarofregarding
photo-assisted
charge
the liquid-solid
interface.
The process
spectral
sensitization
of an n-typemechanism
semiconductor,
also explained
known as anodic
was first
reported
duringofthe
1960s.
Photosensitization
is well
by several
sensitization,
was
first
reported
during
the
1960s.
Photosensitization
mechanism
is
well
explained
researchers in photo-electrochemistry and photocatalysis. In those cases, the photochemically by
excited
several
researchers
photo-electrochemistry
and photocatalysis.
In those
cases, the photochemically
molecule
may
transferinan
electron to the semiconductor
electrode
or catalyst.
The role of the dye
excited
transfer anofelectron
to the semiconductor
electrode
or first
catalyst.
The
of the [31].
molecule
formolecule
spectralmay
sensitization
silver bromide
was explained
for the
time,
byrole
Bourdon
dye molecule for spectral sensitization of silver bromide was explained for the first time, by Bourdon
When silver bromide was alone, intrinsically generated photo holes moved freely in the crystal.
[31]. When silver bromide was alone, intrinsically generated photo holes moved freely in the crystal.
However, adsorbed dye at silver bromide surface significantly altered such behavior under the light.
However, adsorbed dye at silver bromide surface significantly altered such behavior under the light.
OnceOnce
the dye
absorbed
photons,
a latent
image
waswas
formed
which
waswas
similar
to the
the molecule
dye molecule
absorbed
photons,
a latent
image
formed
which
similar
to intrinsic
the
absorption
band
of
silver
bromide
(Figure
5)
[31].
intrinsic absorption band of silver bromide (Figure 5) [31].
Figure 5. Energy level diagram of the dye-silver bromide system. Reproduced with permission from
Figure 5. Energy level diagram of the dye-silver bromide system. Reproduced with permission
[31]. Copyright American Chemical Society, 1965.
from [31]. Copyright American Chemical Society, 1965.
The first case of the dye-sensitized nanocrystalline solar cell was conveyed by O’Regan and
Gratzel
[32],
where
they
used a mesoporous
oxide layer ofsolar
TiO2 cell
nanoparticles.
To construct
the n- and
The
first
case
of the
dye-sensitized
nanocrystalline
was conveyed
by O’Regan
type
electrode,
a
high-surface-area
TiO
2
films
were
deposited
on
a
conducting
glass
surface.
Figure
6
Gratzel [32], where they used a mesoporous oxide layer of TiO2 nanoparticles. To construct the n-type
[32]
shows
the
schematic
diagram
of
the
dye-sensitized
photovoltaic
cell
indicating
the
energy
levels
electrode, a high-surface-area TiO2 films were deposited on a conducting glass surface. Figure 6 [32]
of the
different
phases.
In the dye-sensitized
photo-electrochemical
dye molecule
absorbs
a
shows
schematic
diagram
of the dye-sensitized
photovoltaiccell,
cell the
indicating
the energy
levels
of
photon and injects electron to the conduction band of TiO2. A redox species regenerate the dye
different phases. In the dye-sensitized photo-electrochemical cell, the dye molecule absorbs a photon
molecule in the solution, and the circuit is then completed.
and injects electron to the conduction band of TiO2 . A redox species regenerate the dye molecule in
The electron transfer mechanism in the dye-sensitized photocatalyst is quite similar as in dyethe solution,
thecell.
circuit
is then
completed.
sensitizedand
solar
The dye
molecule
has a dual role; as a sensitizer, it absorbs visible light, and as
The
electron
transfer
mechanism
in the photocatalyst
dye-sensitized
is quite
similar
a molecular
bridge
it attaches
semiconductor
withphotocatalyst
an electron donor
(Figure
7) [33].as in
dye-sensitized
solar
dye molecule
has and
a dual
role;
as a sensitizer,
it absorbs
light,
Youngblood
et al.cell.
[33]The
explained
the forward
back
electron
transfer kinetics
in thevisible
presence
of and
important
components
such
as
photosensitizers,
electron
relays,
and
photocatalysts.
At
first,
the
dye
as a molecular bridge it attaches semiconductor photocatalyst with an electron donor (Figure 7) [33].
moleculeet
adsorbs
and forms
excited state.
excitedtransfer
dye molecule
injects
an electron
Youngblood
al. [33]photon
explained
the forward
and Then
backthe
electron
kinetics
in the
presence of
into the
conduction such
bandas
ofphotosensitizers,
the semiconductorelectron
and thereby
transformed
to oxidizedAt
form.
important
components
relays,
and photocatalysts.
first,The
the dye
oxidized dye is subsequently reduced to the ground state by electron transfer from anther
molecule adsorbs photon and forms excited state. Then the excited dye molecule injects an electron into
photocatalyst for water oxidation [9]. Now the conduction band electrons can reduce water to
the conduction band of the semiconductor and thereby transformed to oxidized form. The oxidized dye
hydrogen on the reduction site over the photocatalyst in the process of water splitting. However, the
is subsequently reduced to the ground state by electron transfer from anther photocatalyst for water
oxidation [9]. Now the conduction band electrons can reduce water to hydrogen on the reduction site
Inorganics 2017, 5, 34
Inorganics 2017, 5, 34
9 of 35
9 of 35
over the photocatalyst in the process of water splitting. However, the electrons in the conduction band,
Inorganicsin
2017,
9 of 35
electrons
the5, 34
conduction
if not used quickly toundergo
produceback
hydrogen/photocurrent,
undergo
if not
used quickly
to produceband,
hydrogen/photocurrent,
electron transfer to regenerate
back electron transfer to regenerate the sensitizer.
the sensitizer.
electrons in the conduction band, if not used quickly to produce hydrogen/photocurrent, undergo
back electron transfer to regenerate the sensitizer.
Figure
Schematicdiagram
diagramofofthe
thedye-sensitized
dye-sensitized photovoltaic
photovoltaic cell
energy
Figure
6. 6.
Schematic
celltotoindicate
indicatethe
theelectron
electron
energy
Figure
6. Schematic
diagramReproduced
of the dye-sensitized
photovoltaic
to Copyright
indicate theNature
electronPublishing
energy
level
in the
different phases.
with permission
fromcell
[32].
level in the different phases. Reproduced with permission from [32]. Copyright Nature Publishing
level in
the different phases. Reproduced with permission from [32]. Copyright Nature Publishing
Group,
1991.
Group,
1991.
Group, 1991.
Figure
transfer and
and charge
chargerecombination
recombinationpathway
pathway
Figure7. 7.Kinetic
Kineticscheme
scheme for
for forward
forward electron
electron transfer
in in
Figure 7. Kinetic scheme for forward electron transfer and charge recombination pathway in sensitized
evolvingcatalyst.
catalyst.Reproduced
Reproduced
with
permission
from
sensitized
semiconductors
and O2 evolving
with
permission
from
sensitized
semiconductorscontaining
containingH
H22 and
semiconductors
containing
H
catalyst. Reproduced with permission from [33].
2 and O
2 evolving
[33].
Copyright
AmericanChemical
Chemical
Society,
2009.
[33].
Copyright
American
Society,
Copyright American Chemical Society, 2009.
earlystage,
stage,dye-sensitized
dye-sensitized photocells
photocells were
into
dyeIn In
thethe
early
were prepared
preparedby
bydipping
dippingthe
theelectrodes
electrodes
into
dyeelectrolyte
solutions.
Experimental
results
also
showed
that
only
adsorbed
dye
molecule
on
thethe
In the early
stage,Experimental
dye-sensitized
photocells
were that
prepared
by dipping
electrodes
into
electrolyte
solutions.
results
also showed
only adsorbed
dyethe
molecule
on
electrodesurface
surface
couldproduce
produce photocurrent
photocurrent
[34].
Thus,
ofofdye
with
dye-electrolyte
solutions.
Experimental
results also
that only adsorbed
dye
molecule
electrode
could
[34]. showed
Thus, chemisorption
chemisorption
dyemolecule
molecule
withon
anchoring
groupswould
would
be
beneficial
as the
the dye
dye can
be
attached
covalent
to with
groups
beneficial
as
can
be effectively
effectively
attachedvia
via
covalent
bonding
to
theanchoring
electrode
surface
couldbe
produce
photocurrent
[34].
Thus, chemisorption
of
dye bonding
molecule
the semiconductor surface. The covalent bond provides a strong electronic coupling between the
the
semiconductor
surface.
The
covalent
bond
provides
a
strong
electronic
coupling
between
the
anchoring groups would be beneficial as the dye can be effectively attached via covalent bonding to the
molecular orbital of adsorbed dye and semiconductor, facilitating fast electron injection rate [35].
molecular orbital
of adsorbed
dye bond
and semiconductor,
facilitating
electron
injection
[35].
semiconductor
surface.
The covalent
provides a strong
electronicfast
coupling
between
therate
molecular
Galoppini [35] provided an extensive review of the synthesis and properties of sensitizers with
Galoppini
[35] provided
an
extensive review
of the synthesis
andinjection
properties
of[35].
sensitizers
with
orbital
of
adsorbed
dye
and
semiconductor,
facilitating
fast
electron
rate
Galoppini
different types of chromophore-linkers arrays. The review focused on the design of linkers that attach [35]
different
of chromophore-linkers
arrays. The
review
focused
the designwith
of linkers
that types
attach of
provided
antypes
extensive
of the
thesemiconductor
synthesis
andto
properties
of on
sensitizers
different
the sensitizer
to the review
surface of
form sensitizer-bridge-anchor
arrays.
Figure
8
the
sensitizer
to
the
surface
of
the
semiconductor
to
form
sensitizer-bridge-anchor
arrays.
Figure
8
chromophore-linkers
arrays.
The review
focused
thethe
design
of linkers
that attachsurface
the sensitizer
[35] shows a schematic
representation
of linker
thaton
binds
sensitizer
to semiconductor
via
[35]
shows
a
schematic
representation
of
linker
that
binds
the
sensitizer
to
semiconductor
surface
via
anchoring
group.
Proper designtoofform
linkers
is important for several reasons
as (i)8 to
fix shows
the
to theansurface
of the
semiconductor
sensitizer-bridge-anchor
arrays.such
Figure
[35]
a
andistance
anchoring
group.
Proper
design
of
linkers
is
important
for
several
reasons
such
as
(i)
to
fix
of the dye from
the semiconductor
surface;
(ii) toto
modify
the properties
of thevia
dye;
tothe
schematic representation
of linker
that binds the
sensitizer
semiconductor
surface
an (iii)
anchoring
distance
of the dye of
from the
semiconductor
surface;
(ii) to modify
the interfacial
properties of the dye;
(iii) to
stop
aggregation
chromophores;
and
to prepare
models
transfer
group.
Proper
design of the
linkers
is important
for(iv)
several
reasons
such for
as (i) to fix theelectron
distance
of the dye
stop
aggregation
of
the
chromophores;
and
(iv)
to
prepare
models
for
interfacial
electron
transfer
This is also a promising method to understand the electron transfer processes at the
from studies.
the semiconductor
surface; (ii) to modify the properties of the dye; (iii) to stop aggregation of
studies.
This is also a interface
promising
to them
understand
the and
electron
transfer
processes
molecule–nanoparticle
andmethod
to control
in a rational
predictable
manner
[35]. at the
the chromophores; and (iv) to prepare models for interfacial electron transfer studies. This is also a
molecule–nanoparticle interface and to control them in a rational and predictable manner [35].
promising method to understand the electron transfer processes at the molecule–nanoparticle interface
and to control them in a rational and predictable manner [35].
Inorganics 2017, 5, 34
10 of 35
Inorganics 2017, 5, 34
Inorganics 2017, 5, 34
10 of 35
10 of 35
8. Schematic
diagram
of linkers
bindsensitizer
sensitizer to
to metal
b: b:
bridge,
A: anchoring
FigureFigure
8. Schematic
diagram
of linkers
toto
bind
metaloxide
oxide(MO).
(MO).
bridge,
A: anchoring
group. Reproduced with permission from [35]. Copyright Elsevier, 2004.
group. Reproduced with permission from [35]. Copyright Elsevier, 2004.
There are few other parameters which are equally important in the dye-sensitization process: (i)
There
are fewSchematic
othercompatibility
parameters
which
are equally
important
in
the
dye-sensitization
process:
(i) the
the energy
between
excited
sensitizer
and
the
band
of the
Figure 8. level
diagram of linkers
to bind
sensitizer
to metal oxide
(MO).
b:conduction
bridge, A: anchoring
+
energysemiconductor;
level
compatibility
between
excited
andElsevier,
the
conduction
oftothe
semiconductor;
(ii) redox
potential
of from
the sensitizer
dye/dye
couple,
and
stability
of band
the dye
undergo
about
group.
Reproduced
with
permission
[35]. Copyright
2004.
108 turnover
cycles
[36].
(ii) redox
potential
of the
dye/dye+ couple, and stability of the dye to undergo about 108 turnover
cycles [36].There are few other parameters which are equally important in the dye-sensitization process: (i)
5.
Review
Dye-Sensitized
Overall
Waterexcited
Splitting
the
energyonlevel
compatibility
between
sensitizer and the conduction band of the
+ couple, and stability of the dye to undergo about
semiconductor;
(ii) redox potential
ofWater
the dye/dye
5. Review
on
Dye-Sensitized
Overall
Splitting
This section will consider the overall water splitting into hydrogen and oxygen (volume ratio
108 turnover cycles [36].
2:1)
using dye-sensitized
photocatalysts.
Dye-sensitization
techniqueand
hasoxygen
a significant
role
in 2:1)
This section
will consider the
overall water
splitting into hydrogen
(volume
ratio
employing a visible fraction of the solar spectrum for water splitting application. Abe et al. [37]
Review on Dye-Sensitized
Overall
Water Splitting technique has a significant role in employing
using 5.
dye-sensitized
photocatalysts.
Dye-sensitization
reported the first example of overall water splitting in the presence of visible light using coumarin
a visible
fraction
of the
spectrum
forwater
water
splitting
application.
Abe et(volume
al.
[37]ratio
reported
section
will solar
consider
the overall
splitting
and oxygen
dyes This
(C343,
and NKX-series
(2677,
2697, 2311,
2587))
and into
metalhydrogen
oxide semiconductors.
The oxidized
the first
example
of
overall
water
splitting
in
the
presence
of
visible
light
using
coumarin
2:1) using
technique
a significant
role long
in dyes
forms
of thedye-sensitized
dye moleculesphotocatalysts.
possess two or Dye-sensitization
more thiophene rings
in theirhas
structures
and enjoy
(C343,lifespans
and NKX-series
(2677,
2587)) and
oxide
semiconductors.
oxidized
employing
a visible reversible
fraction2697,
ofoxidation-reduction
the2311,
solar spectrum
for metal
water
application. activities
Abe etThe
al.and
[37]
to undergo
cycles.
Thesplitting
oxidation/reduction
the
reported
the
first
example
of overall
water
theinpresence
of visible
light
using
coumarin
formsstability
of the of
dye
molecules
possess
two
orsplitting
more in
thiophene
in and
their
structures
and
enjoy
dyes
were
studied
by cyclic
voltammetry
(CV)
both rings
aqueous
organic
solutions
[38].
dyes (C343,
NKX-series
(2677, oxidation-reduction
2697, 2311,
oxide
semiconductors.
The
oxidized
Dye-adsorbed
Pt/H
4Nbreversible
6O17 photocatalyst
was2587))
usedand
for metal
hydrogen
evolution
while IrO
2 and Pt
colong lifespans
to and
undergo
cycles.
The
oxidation/reduction
activities
forms ofWO
the3 photocatalyst
dye moleculeswere
two or
more
thiophene
in their
andofenjoy
loaded
applied
achieve
oxygenrings
evolution
the
presence
I3−and
/I− long
redox
and the
stability
of dyes werepossess
studied
bytocyclic
voltammetry
(CV)ininstructures
both
aqueous
organic
lifespans
to
undergo
reversible
oxidation-reduction
cycles.
The
oxidation/reduction
activities
and
the
couple
as
a
shuttle
redox
mediator
(Figure
9)
[38].
The
photocatalytic
reaction
continued
up
to
2
days
solutions [38]. Dye-adsorbed Pt/H4 Nb6 O17 photocatalyst was used for hydrogen evolution
while
stability any
of dyes
were studied
by cyclic
voltammetry
(CV) in both
organic
[38].
without
deactivation.
However,
the
quantum efficiency
was aqueous
less thanand
0.1%
at 500 solutions
nm. NKX-2677IrO2 and
Pt
co-loaded
WO
photocatalyst
were
applied
to
achieve
oxygen
evolution
in
the
presence
of
36O17 photocatalyst was used for hydrogen evolution while IrO2 and Pt coDye-adsorbed
Pt/H
4Nb
sensitized
Pt/TiO
2 photocatalyst showed efficient H2 evolution with a steady rate with an aqueous
−
−
−/I− redox
I3 /I loaded
redox couple
as a shuttle
redox
mediator
(Figure
9) [38].
The photocatalytic
reaction
continued
3 photocatalyst
were
applied
to achieve
oxygen
evolution
the presence
solutionWO
of triethanolamine
(TEOA).
However,
the reduction
of I3− to I−in
occurred
on theofPtI3co-catalysts
up to 2on
days
without
any
deactivation.
However,
the
quantum
efficiency
was
less
than
0.1%
at 500 nm.
couple
as
a
shuttle
redox
mediator
(Figure
9)
[38].
The
photocatalytic
reaction
continued
up
to
2
days
TiO2 suppressing the H2 evolution consequently. Abe et al. [38] demonstrated that selective
without
any
deactivation.
However,
the
quantum
efficiency
was
less
than
0.1%
at
500
nm.
NKX-2677NKX-2677-sensitized
Pt/TiO2 photocatalyst
showed
H172 suppressed
evolution with
a steady of
rate
with
loading of Pt nanoparticles
in the interlayer
spaces efficient
of H4Nb6O
the reduction
I3−to
I− an
sensitized
Pt/TiO
2 photocatalyst showed efficient H2 evolution with a steady−rate with
an aqueous
aqueous
solution
of triethanolamine
(TEOA). However, the reduction of I3 to I− occurred
on the Pt
on Pt
nanoparticles
(Figure 10) [38].
solution of triethanolamine (TEOA). However, the reduction of I3− to I− occurred on the Pt co-catalysts
co-catalysts
on TiO2 suppressing the H2 evolution consequently. Abe et al. [38] demonstrated that
on TiO2 suppressing the H2 evolution consequently. Abe et al. [38] demonstrated that selective
selective loading of Pt nanoparticles in the interlayer spaces of H4 Nb6 O17 suppressed the reduction
of
loading
of Pt nanoparticles in the interlayer spaces of H4Nb6O17 suppressed the reduction of I3−to I−
I3 − to on
I− Pt
onnanoparticles
Pt nanoparticles
(Figure
10)
[38].
(Figure 10) [38].
Figure 9. Visible-light-induced water splitting using dye-adsorbed Pt/H4Nb6O17 photocatalyst in the
presence of I3/I− shuttle redox mediator. Reproduced with permission from [38]. Copyright American
Chemical Society, 2013.
Figure 9. Visible-light-induced water splitting using dye-adsorbed Pt/H4Nb6O17 photocatalyst in the
Figurepresence
9. Visible-light-induced
water splitting using dye-adsorbed Pt/H Nb O17 photocatalyst
in the
of I3/I− shuttle redox mediator. Reproduced with permission from 4[38].6Copyright
American
−
presence
of I3 /ISociety,
shuttle
redox mediator. Reproduced with permission from [38]. Copyright American
Chemical
2013.
Chemical Society, 2013.
Inorganics 2017, 5, 34
11 of 35
Inorganics 2017, 5, 34
11 of 35
Inorganics 2017, 5, 34
11 of 35
Figure 10.
10. Conceptual
layered
Figure
Conceptual reaction
reaction scheme
scheme for
for declining
declining backward
backward reaction
reaction using
using nanostructured
nanostructured layered
semiconductor. Reproduced
Reproduced with
with permission
permission from
from [38].
[38]. Copyright
Copyright American
American Chemical
Chemical Society,
Society, 2013.
2013.
semiconductor.
10.synthesized
Conceptual reaction
schemeTiO
for declining
backward reaction
using nanostructured
Le etFigure
al. [39]
Co-doped
2 nanomaterials
by impregnation
methodlayered
and sensitized
Le etsemiconductor.
al. [39] synthesized
Co-doped
TiO2 nanomaterials
by impregnation
method
and
Reproduced
with permission
fromphotocatalytic
[38]. Copyright
American
Chemical
Society,
2013.sensitized
with Rhodamine
B dye
for visible-light-assisted
water splitting.
The dye-sensitized
with Rhodamine B dye for visible-light-assisted photocatalytic water splitting. The dye-sensitized
photocatalyst was active in the range of 450–600 nm and showed the stoichiometric evolution of
Le et al.
[39]active
synthesized
2 nanomaterials
impregnation
method and sensitized
photocatalyst
was
in theCo-doped
range of TiO
450–600
nm and by
showed
the stoichiometric
evolution of
hydrogen
and oxygenB through
water splitting. In photocatalytic
the absence ofwater
dye, splitting.
Co/TiO2 The
photocatalyst
was not
with
Rhodamine
dye
for
visible-light-assisted
dye-sensitized
hydrogen and oxygen through water splitting. In the absence of dye, Co/TiO2 photocatalyst
was not
visible
light
active,
and
the
rate
of
water
splitting
was
negligible.
Upon
sensitization
with
Rhodamine
photocatalyst was active in the range of 450–600 nm and showed the stoichiometric evolution of
visible light active, and the rate of water splitting was negligible. Upon sensitization with Rhodamine
B, Co/TiO
2 photocatalyst
showedwater
higher
hydrogen
(~9.07
H2/dye/h) in
the
hydrogen
and oxygen through
splitting.
In the evolution/dye/h
absence of dye, Co/TiO
2 photocatalyst
was
notentire
B, Co/TiO2 photocatalyst showed higher hydrogen evolution/dye/h (~9.07 H2 /dye/h) in the entire
rangevisible
of reaction
time.and
According
towater
the authors,
B-sensitized
photocatalyst
was six times
light active,
the rate of
splittingRhodamine
was negligible.
Upon sensitization
with Rhodamine
rangeB,ofCo/TiO
reaction
time. According
to the
authors,
Rhodamine
B-sensitized
wasentire
six times
2 photocatalyst
showed
higher
hydrogen
evolution/dye/h
(~9.07photocatalyst
H2/dye/h) in the
active than unsensitized
photocatalyst
regarding
water
splitting.
activerange
thanofunsensitized
photocatalyst
regarding
water splitting.
reaction
authors, Rhodamine
photocatalyst
was six times
Alibabaei
et al. time.
[40] According
developedtoathe
dye-sensitized
photoB-sensitized
electrosynthesis
cell (DSPEC)
for solar
Alibabaei
et
al. [40] developed
a dye-sensitized
photo
electrosynthesis cell (DSPEC) for solar
active
than
unsensitized
photocatalyst
regarding
water
splitting.
water splitting utilizing a mesoporous SnO2/TiO2 core/shell nanostructured electrode for solar water
water splitting
utilizing
a mesoporous
SnO2 /TiO2 core/shell
nanostructured
electrode
for solar
Alibabaei
et al. [40]
developed a dye-sensitized
photo electrosynthesis
cell (DSPEC)
for solar
splitting. The electrode was prepared with a surface-bound
Ru(II) polypyridal-based chromophorewater
splitting
utilizing
a
mesoporous
SnO
2
/TiO
2
core/shell
nanostructured
electrode
for
solar
water
water splitting. The electrode was prepared with a surface-bound Ru(II) polypyridal-based
photocatalyst assembly having both a light absorber and water oxidation photocatalyst (Figure 11)
splitting. The electrode was
prepared
with a both
surface-bound
Ru(II) polypyridal-based
chromophorechromophore-photocatalyst
assembly
having
a light absorber
and water oxidation
photocatalyst
[40]. photocatalyst
The chromophore-photocatalyst
assembly
was
stabilized
on
TiO
2 surface via atomic layer
assembly
having both a light absorber
and water
(Figure
(Figure 11) [40]. The
chromophore-photocatalyst
assembly
was oxidation
stabilizedphotocatalyst
on TiO2 surface
via11)
atomic
deposition
of Al
2O3 or TiO2 to provide long-term
water
splittingonability
even atvia
pHatomic
7 (in phosphate
[40]. The
chromophore-photocatalyst
assembly was
stabilized
TiO2 surface
layer
layer deposition of Al2 O3 or TiO2 to provide long-term water splitting ability
even at pH 7 (in
II–RubII–OH
buffer).
In the of
DSPEC,
tin oxide water
(FTO)splitting
|SnO2/TiO
2|–[Ru
2]
(Al
2O3 or
deposition
Al2O3 orfluorine-doped
TiO2 to provide long-term
ability
even aat
pH 7 (inIIphosphate
+
II
phosphate buffer). In the DSPEC, fluorine-doped tin oxide (FTO) |SnO2 /TiO
|–[Ru
a –Rub –OH2 ]4
2
−2
II
II
TiO2)buffer).
photoanodes
were illuminated
withtinvisible
= 445
nm,
I = 90
mW·cm
the
In the DSPEC,
fluorine-doped
oxide light
(FTO)(λ|SnO
2/TiO
2|–[Ru
a –Ru
b –OH2)] in(Al
2Opresence
3 or
(Al2 OTiO
TiO2 ) photoanodes
were illuminated
with
visible
lightnm,
(λ I==445
nm, I =−290
cm−2 ) in the
3 or
2) photoanodes
with
visible
light
(λ = 445
90 mW·cm
) inmW
the ·presence
of a platinum
cathode,were
and illuminated
with a small
applied
bias
water
splitting
was
achieved
leading
to a high
presence
of a platinum
cathode,
and
small applied
bias
waterwas
splitting
wasleading
achieved
of a platinum
cathode,
with
a).with
small aapplied
bias water
splitting
achieved
to aleading
high to
−2
photocurrent
density
(1.97and
mA·cm
−2).
a high
photocurrent
density
mA
·cm−2 ).
photocurrent
density
(1.97(1.97
mA·cm
Figure 11. Cartoon is representing an atomic layer deposition (ALD) core/shell electrode surface with
Figure 11. Cartoon is representing an atomic layer deposition (ALD) core/shell electrode surface with
ALD
stabilization
of aan
surface-bound
Reproduced
with permission
from
[40]. with
Figure
11. overlayer
Cartoon is
representing
atomic layerassembly.
deposition
(ALD) core/shell
electrode
surface
ALD Copyright
overlayerNational
stabilization
of aofsurface-bound
assembly. Reproduced with permission from [40].
Academy
Sciences,
2015.
ALD overlayer stabilization of a surface-bound assembly. Reproduced with permission from [40].
Copyright National Academy of Sciences, 2015.
Copyright National Academy of Sciences, 2015.
Ideally, a photo-electrochemical water splitting system would use abundant materials with high
quantum yield and multiple absorbers to harvest the solar spectrum efficiently. The efficacy of the
Ideally, a photo-electrochemical water splitting system would use abundant materials with high
quantum yield and multiple absorbers to harvest the solar spectrum efficiently. The efficacy of the
Inorganics 2017, 5, 34
12 of 35
Inorganics 2017, 5, 34
12 of 35
Ideally, a photo-electrochemical water splitting system would use abundant materials with high
quantum
yield and
multiple absorbers
to harvest
the solarturnover
spectrum
efficiently.
The turnover
efficacy offrequency
the water
water oxidation
photocatalysts
is measured
regarding
number
(TON),
oxidation
regarding
turnover
(TON),
turnover frequency
(TOF) and
(TOF) andphotocatalysts
overpotentialisatmeasured
a suitable
TOF. Swierk
andnumber
Mallouk
[41] reviewed
the development
of
overpotential
at
a
suitable
TOF.
Swierk
and
Mallouk
[41]
reviewed
the
development
of
photoanodes
photoanodes for water splitting with dye-sensitized photoelectrochemical cells (DSSCs). In
for
water splitting
photoelectrochemical
cellscatalyst,
(DSSCs).namely
In photoanodes
there are
photoanodes
therewith
are dye-sensitized
four different kinds
of water oxidation
(i) heterogeneous
four
different
kinds of water
oxidation
catalyst, namelyand
(i) heterogeneous
catalysts;to
(ii)them,
molecular
catalysts;
(ii) molecular
catalysts;
(iii) polyoxometalates;
(iv) cubanes. According
in the
catalysts;
(iii)
polyoxometalates;
and
(iv)
cubanes.
According
to
them,
in
the
case
of
photoanode,
it is
case of photoanode, it is quite difficult to maintain the relative rates of forward and backward electron
quite
difficult
maintain the
relative Normally
rates of forward
and backward
electron
transfer
at the
sensitizer
transfer
at thetosensitizer
molecule.
back electron
transfer
happens
on the
timescale
of
molecule.
Normally
back
electron
transfer
happens
on
the
timescale
of
hundreds
of
microseconds,
hundreds of microseconds, so the regeneration of the reduced sensitizer should be on the
so
the regeneration
of theUse
reduced
sensitizer
should
be on the
microsecond
timescale. Use of(BiP)
electron
microsecond
timescale.
of electron
transfer
mediators
such
as benzimidazole-phenol
and
transfer
mediators
such
as
benzimidazole-phenol
(BiP)
and
ruthenium
polypyridyals
can
be
a
solution
ruthenium polypyridyals can be a solution to the above issue [42]. Swierk and Mallouk [41] suggested
to
the above
issue [42].ofSwierk
and Mallouk
[41] suggested
a possible
a two-absorber
a possible
application
a two-absorber
Z-scheme
to effectively
captureapplication
the visible of
and
near-infrared
Z-scheme
to
effectively
capture
the
visible
and
near-infrared
parts
of
the
solar
spectrum
(Figure
12)other
[41].
parts of the solar spectrum (Figure 12) [41]. Using one photosystem at the photocathode and the
Using
photosystem
the photocathode
and of
theunwanted
other at the
photoanode
will eliminate
the
at the one
photoanode
will at
eliminate
the possibility
reactions
of colloidal
Z-schemes.
possibility
of
unwanted
reactions
of
colloidal
Z-schemes.
Moreover,
it
will
allow
the
use
of
two
Moreover, it will allow the use of two different sensitizers at the photocathode and photoanode
different
sensitizers
at harvesting.
the photocathode and photoanode delivering better light harvesting.
delivering
better light
Figure 12. A two-absorber Z-scheme with an 8-photon, 4-electron water splitting system. Reproduced
Figure 12. A two-absorber Z-scheme with an 8-photon, 4-electron water splitting system. Reproduced
with permission from [41]. Copyright Royal Society of Chemistry, 2012.
with permission from [41]. Copyright Royal Society of Chemistry, 2012.
Overall water splitting processes with different dye-sensitized photocatalytic and
Overall
water splitting
processes
with differentin
dye-sensitized
photocatalytic and photoelectrochemical
photoelectrochemical
systems
are summarized
Table 2.
systems are summarized in Table 2.
Inorganics 2017, 5, 34
13 of 35
Table 2. Dye-sensitized overall water splitting under visible light: photocatalytic and photoelectrochemical processes.
No.
Process Details
1.
Photocatalytic water splitting on organic
dye-sensitized KTa(Zr)O3 photocatalyst
2.
Light Source and Accessories
Other Experimental Details
Results/Comments
References
Xenon lamp (300 or 500 W)
A total of 17 sensitizers (dye) were used to modify
the photocatalytic activity of PtO x /KTa(Zr)O3 for
water splitting
Modification with cyanocobalamin (B12) was the
most promising with energy conversion efficiency
of about 0.013%
[43]
Ruthenium polypyridyl-sensitized solar
cells for water splitting in the presence of
a biometric electron transfer mediator;
Filtered white light illumination (λ = 450 nm
light) at 4.5 mW·cm−2 intensity
Iridium oxide nanoparticles are bonded to
benzimidazole-phenol mediator and
2-carboxyethylphosphonic acid
anchoring molecule
QY (2.3%) is more than doubled with the use of an
electron transfer mediator
[42]
3.
Heteroleptic ruthenium dye-sensitized
photoelectrochemical cell for
water splitting
450 nm light at 7.8 mW·cm−2
Modified ruthenium dye adsorbed on IrO2 ·nH2 O
nanoparticles in aqueous solution (1 M) of
Na2 S2 O8 ; this system undergoes rapid electron
transfer into anatase TiO2
QY of approximately 0.9%
[33]
4.
Water splitting from bipolar
Pt/dye-sensitized TiO2
photoanode arrays
Xenon lamp (2500 W) with IR filter, light
intensity 100 mW·cm−2
Anode: Ru-dye Z-907 coated TiO2 , Electrolyte:
iodide/iodine, Cathode: Pt
QY 3.7%
[44]
5.
Water splitting in a photoelectrochemical
device with a molecular Ru catalyst
assembled on dye-sensitized nano-TiO2
Xenon lamp (500 W) with 400 nm
cut-off filter
Sensitizer: [Ru(bpy)2 (4,40 -(PO3 H2 )2 bpy)]2+ ;
Nafion membrane was used to immobilize the
hydrophobic catalyst; Pt cathode
Strong acidic pH of commercial Nafion membrane
led to a fast decay of the photocurrent
[45]
6.
Organic dye-sensitized Tandem
photoelectrochemical cell for
water splitting
White LED light (λ > 400 nm), light intensity
100 mW·cm−2
Photocathode: organic dye P1 as photo absorber
and a complex molecular Co1 as H2 generation
catalyst on nano-NiO; Photoanode: organic dye L0
as photosensitizer and a molecular complex Ru1
as O2 generation catalyst on mesoporous TiO2
IPCE of 25% at 380 nm under neutral condition
without bias
[46]
7.
Water splitting on Rhodamine-B
dye-sensitized Co-doped TiO2 catalyst
Ozone-free Xenon arc lamp (300 W) with UV
cut-off filter
Photocatalytic reaction was carried out in an outer
irradiated Pyrex cell with Rh–B–Co/TiO2
photocatalyst powder (50 mg) dispersed in pure
water (70 mL)
Dye-sensitized photocatalyst achieved
stoichiometric evolution of H2 and O2 gas; the H2
generation was six times higher than
sensitized photocatalyst
[39]
Inorganics 2017, 5, 34
14 of 35
6. Review on Dye-Sensitized Sacrificial Hydrogen Generation
6. Review on Dye-Sensitized Sacrificial Hydrogen Generation
This section will review dye-sensitized sacrificial hydrogen generation focusing on critical
This section
review dye-sensitized
sacrificial
hydrogen
generationand
focusing
on critical
parameters
such aswill
sensitizers,
sacrificial agents,
semiconductor
photocatalyst
co-catalyst.
parameters such as sensitizers, sacrificial agents, semiconductor photocatalyst and co-catalyst.
6.1. An Overview of Dyes as Photosensitizers
6.1. An Overview of Dyes as Photosensitizers
Different types of dyes have been so far used as sensitizers in photocatalytic dye-sensitized
Different types of dyes have been so far used as sensitizers in photocatalytic dye-sensitized
hydrogen generation. Metal complexes and organic dyes are the most commonly used materials.
hydrogen generation. Metal complexes and organic dyes are the most commonly used materials.
Metal complexes can be divided into two groups such as Ru-complexes and other transition metal
Metal complexes can be divided into two groups such as Ru-complexes and other transition
complexes.
metal complexes.
Ru-complexes have a strong ability to absorb visible light. Therefore, they have been extensively
Ru-complexes have a strong ability to absorb visible light. Therefore, they have been extensively used
used as acceptable sensitizers in dye-sensitized photocatalytic processes. Derivatives of2+Ru(bpy)
as acceptable sensitizers in dye-sensitized photocatalytic processes. Derivatives of Ru(bpy)3 were first
were first reported as sensitizers by Gratzle’s group for dye-sensitized water splitting
on
reported as sensitizers by Gratzle’s group for dye-sensitized water splitting on Pt/TiO2 /RuO2 [47,48].
Pt/TiO2/RuO2 [47,48].
The photoactivity of Ru-complexes based dye-sensitized TiO2 depends on their terminal
The photoactivity of Ru-complexes based dye-sensitized TiO2 depends
on their terminal groups.
groups. Carboxyl [49–52] and phosphonate [53–55] are the two most important terminal groups
Carboxyl [49–52] and phosphonate [53–55] are the two most important terminal groups of Ruof Ru-complexes that have been studied. Choi’s team has studied the photocatalytic activity of
complexes that have been studied. Choi’s team has studied the photocatalytic activity of
semiconductors using Ru-complexes with various terminal groups. It has been reported that the
semiconductors using Ru-complexes with various terminal groups. It has been reported that the
adsorption of those derivatives with phosphonate group is stronger than those with carboxyl groups
adsorption of those derivatives with phosphonate group is stronger than those with carboxyl groups
resulting in higher photoactivity for hydrogen generation. Apparent quantum yield as high as 28%
resulting in higher photoactivity for hydrogen generation. Apparent quantum yield as high as 28%
was reported in this case [56,57].
was reported in this case [56,57].
In addition to Ru-complexes, dyes based on other transition-metal complexes have been used in
In addition to Ru-complexes, dyes based on other transition-metal complexes have been used in
dye-sensitized photocatalytic hydrogen generation. However, research on other transition metals is
dye-sensitized photocatalytic hydrogen generation. However, research on other transition metals is
very limited compared to Ru-complexes. Metal porphyrins (MPs) and metal phthalocyanines (MPcs)
very limited compared to Ru-complexes. Metal porphyrins (MPs) and metal phthalocyanines (MPcs)
are two notable examples of these metal complexes that have been widely used in dye-sensitization
are two notable examples of these metal complexes that have been widely used in dye-sensitization
photocatalysis [58–62]. Typical molecular structures and symbols of the metal complexes are shown in
photocatalysis [58–62]. Typical molecular structures and symbols of the metal complexes are shown
Figure 13 [63].
in Figure 13 [63].
Figure 13. Basic molecular structures of typical porphyrins and phthalocyanines used in dyeFigure 13. Basic molecular structures of typical porphyrins and phthalocyanines used in dye-sensitized
sensitized photocatalytic H2 production. Reproduced with permission from [63]. Copyright Royal
photocatalytic H2 production. Reproduced with permission from [63]. Copyright Royal Society of
Society of Chemistry, 2015.
Chemistry, 2015.
Copper phthalocyanine, ruthenium bipyridyl, and eosin Y were compared as photosensitizers
Copper phthalocyanine,
ruthenium
bipyridyl,
andby
eosin
were
compared
as photosensitizers
in
in photocatalytic
dye-sensitized
hydrogen
generation
the Y
use
of solar
light irradiation
in a slurry
photocatalytic
dye-sensitized
generation
byoutcome
the use ofofsolar
a slurry
of TiO2/RuO2 using
(MV2+) ashydrogen
an electron
relay. The
this light
studyirradiation
indicated in
the
order of
of
2+ ) as an electron relay. The outcome of this study indicated the order of various
TiO
/RuO
using
(MV
2
2 in the enhancement of solar hydrogen generation as follow: copper phthalocyanine >
various
dyes
Inorganics 2017, 5, 34
15 of 35
Inorganics 2017, 5, 34
15 of 35
dyes in the enhancement of solar hydrogen generation as follow: copper phthalocyanine > ruthenium
bipyridyl
eosin Y [58].
TheYdifference
in performance
of the usedofdyes
is ascribed
their different
ruthenium> bipyridyl
> eosin
[58]. The difference
in performance
the used
dyes isto
ascribed
to their
structure
properties
and the difference
in their electron
different and
structure
and properties
and the difference
in theirinjection
electronability.
injection ability.
Despite
investigation
on metal
complexes
as photo-sensitizer,
their high
cost
andcost
toxicity
Despitethe
theextensive
extensive
investigation
on metal
complexes
as photo-sensitizer,
their
high
and
motivate
the
researchers
to
investigate
better
alternatives
for
them.
Therefore,
metal-free
organic
dyes
toxicity motivate the researchers to investigate better alternatives for them. Therefore, metal-free
have
been
studied
an alternative
thealternative
photocatalytic
dye-sensitization
Metal-free process.
organic
organic
dyes
haveasbeen
studied asinan
in the
photocatalyticprocess.
dye-sensitization
dyes
are
inexpensive,
with
high
diversity
and
light
absorption
ability
which
make
them
Metal-free organic dyes are inexpensive, with high diversity and light absorption ability whichbetter
make
alternatives
for metal complexes
[63]. Metal-free
organic dyes
candyes
be classified
into three
them better alternatives
for metal complexes
[63]. Metal-free
organic
can be classified
into main
three
groups:
(i)
xanthene
dyes;
(ii)
cation-organic
dyes;
and
(iii)
D–π–A
organic
dyes
[64–70].
The
structures
main groups: (i) xanthene dyes; (ii) cation-organic dyes; and (iii) D–π–A organic dyes [64–70]. The
of
some of of
these
dyes
are shown
structures
some
of these
dyes in
areFigure
shown14in[63].
Figure 14 [63].
Figure 14.
14. Typical
some
xanthene
dyes
andand
cation-organic
dyesdyes
usedused
in dyeFigure
Typicalmolecular
molecularstructures
structuresofof
some
xanthene
dyes
cation-organic
in
2 production. Reproduced with permission from [63]. Copyright Royal
sensitized
photocatalytic
H
dye-sensitized photocatalytic H2 production. Reproduced with permission from [63]. Copyright Royal
Society of
of Chemistry,
Chemistry,2015.
2015.
Society
Xanthene-based dyes can be better alternatives for Ru-complexes as they can similarly absorb
Xanthene-based dyes can be better alternatives for Ru-complexes as they can similarly absorb
visible light (400–600 nm). Eosin Y, rhodamine B, rhodamine 6G, erythrosine, erythrosin B, erythrosin
visible light (400–600 nm). Eosin Y, rhodamine B, rhodamine 6G, erythrosine, erythrosin B,
yellowish, fluorescein, eosin bluish, rose bengal, uranine, and phloxin, belongs to this category that
erythrosin yellowish, fluorescein, eosin bluish, rose bengal, uranine, and phloxin, belongs to this
has been studied as sensitizers for dye-sensitized photocatalytic hydrogen generation. Among them,
category that has been studied as sensitizers for dye-sensitized photocatalytic hydrogen generation.
eosin Y has been widely used and is proven to be an acceptable sensitizer [11,68,70,71].
Among them, eosin Y has been widely used and is proven to be an acceptable sensitizer [11,68,70,71].
Some cationic organic dyes such as thiazole orange, pyronin Y, methylene blue, thionine acetate,
Some cationic organic dyes such as thiazole orange, pyronin Y, methylene blue, thionine acetate,
safranine O, and acridine orange hydrochloride, were examined as photosensitizers for hydrogen
safranine O, and acridine orange hydrochloride, were examined as photosensitizers for hydrogen
generation [72,73], in addition to xanthene dyes.
generation [72,73], in addition to xanthene dyes.
Lu’s group has studied co-sensitization strategy as a suitable technique to improve the efficiency
Lu’s group has studied co-sensitization strategy as a suitable technique to improve the efficiency
of hydrogen generation on Pt deposited graphene using two-beam monochromic light (520 and 550
of hydrogen generation on Pt deposited graphene using two-beam monochromic light (520 and 550 nm)
nm) irradiation. Eosin Y and Rose Bengal have been used as photosensitizers. A quantum yield of
irradiation. Eosin Y and Rose Bengal have been used as photosensitizers. A quantum yield of 37.3%
37.3% has been achieved using EY/RB co-sensitized graphene/Pt. The individual quantum yields
has been achieved using EY/RB co-sensitized graphene/Pt. The individual quantum yields were
were 13.9% and 21.7%, using EY and RB, respectively. Co-sensitization provides an extended light
13.9% and 21.7%, using EY and RB, respectively. Co-sensitization provides an extended light response,
response, resulting in hydrogen generation enhancement [74].
resulting in hydrogen generation enhancement [74].
Peng’s group also investigated the effect of co-sensitization on g-C3N4 to broaden spectral
Peng’s group also investigated the effect of co-sensitization on g-C3 N4 to broaden spectral
response. Two different types of photosensitizers were applied: indole-based
D–π–A organic dye (LIresponse. Two different types of photosensitizers were applied: indole-based D–π–A organic dye (LI-4)
4) and an asymmetric zinc phthalocyanine derivative (Zn-tri-PcNc). Their co-sensitized system
and an asymmetric zinc phthalocyanine derivative (Zn-tri-PcNc). Their co-sensitized system showed a
showed a wide visible/NIR light absorption region (400–800 nm) with apparent quantum yields of
16.3%, 7.7%, and 1.7% at 420, 500, and 700 nm monochromatic light irradiation, respectively [75]. The
proposed mechanism for this process is illustrated in Figure 15 [75].
Inorganics 2017, 5, 34
16 of 35
wide visible/NIR light absorption region (400–800 nm) with apparent quantum yields of 16.3%, 7.7%,
and 1.7% at 420, 500, and 700 nm monochromatic light irradiation, respectively [75]. The proposed
mechanism
this process is illustrated in Figure 15 [75].
Inorganics 2017,for
5, 34
16 of 35
Figure 15.
15. Visible/near-IR-light-induced
Visible/near-IR-light-induced HH
2 production
over
g-C
3N34N
co-sensitized
by organic
dyesdyes
and
Figure
over
g-C
by organic
2 production
4 co-sensitized
and
phthalocyanine
derivative.
Reproduced
withpermission
permissionfrom
from [75].
[75]. Copyright American
the the
zinczinc
phthalocyanine
derivative.
Reproduced
with
American
Chemical
Chemical Society,
Society,2015.
2015.
6.2. An
An Overview
Overview of
of Sacrificial
Sacrificial Agents
Agents
6.2.
Sacrificial agent
agent or
or electron
electron donor
donor isis aa major
major component
component of
of photocatalytic
photocatalytic dye-sensitized
dye-sensitized
Sacrificial
hydrogen
generation.
Electron
donors
are
required
for
dye
regeneration
to
pursue
continuous
hydrogen generation. Electron donors
dye regeneration to pursue aa continuous
hydrogen generation
compounds
such
as acetonitrile
[76], methanol
[77–79],
Na2S–
hydrogen
generation scheme.
scheme.Different
Different
compounds
such
as acetonitrile
[76], methanol
[77–79],
Na22S–Na
SO3 [80,81],
EDTA
[49],
and
TEOA
are
reported
as
electron
donors.
Na
SO
[80,81],
EDTA
[49],
and
TEOA
are
reported
as
electron
donors.
2
3
Effect of
of different
different electron
electrondonors
donors including
including ascorbic
ascorbic acid,
acid, acetone,
acetone, EDTA,
EDTA,FeCl
FeCl22, HQ, methanol,
Effect
2
+
phenol, and TEOA on
2/Pt
by
phenol,
on hydrogen
hydrogen generation
generationwas
wasinvestigated
investigatedwith
withRu(bpy)
Ru(bpy)3 sensitized
sensitizedTiO
TiO
2 /Pt
Kaneko’s
group.
Their
results
showed
that
only
EDTA
and
TEOA
were
effective
in
hydrogen
by Kaneko’s group. Their results showed that only EDTA and TEOA were
hydrogen
production. In
In aa dye-sensitized
dye-sensitized photocatalysis,
photocatalysis, holes
holes are
are not
not produced
produced as
as opposed
opposed to
to conventional
conventional
production.
photocatalytic processes.
processes. Therefore,
hydrogen generation
generation and
and the
the role
role of
of the
the
photocatalytic
Therefore, the
the mechanism
mechanism for
for hydrogen
sacrificial
agent
are
different.
This
could
be
the
reason
that
some
compounds
like
methanol
are
not
sacrificial agent are different. This could be the reason that some compounds like methanol are not
effective in
in dye-sensitization
dye-sensitization while
while they
they perform
perform well
well as
as aa sacrificial
sacrificial agent
agent in
in photocatalysis
photocatalysis without
without
effective
sensitizer[82].
[82].Moreover,
Moreover,the
thedonating
donatingability
ability
sacrificial
agents
hydrogen
generation
depends
sensitizer
ofof
sacrificial
agents
onon
hydrogen
generation
depends
on
on
pH
of
the
solution.
In
Kaneko’s
report,
the
optimum
pH
for
hydrogen
generation
was
7
in
the
pH of the solution. In Kaneko’s report, the optimum pH for hydrogen generation was 7 in the presence
presence
of EDTA,
whilepH
alkaline
pH
was
best for
TEOA
as andonor
electron
of
EDTA, while
alkaline
was the
best
forthe
TEOA
as an
electron
[82].donor [82].
Photocatalytic
hydrogen
generation
on
TiO
2
/Pt
sensitized
by
Ru(bpy)
was studied
by
Photocatalytic hydrogen generation on TiO2 /Pt sensitized by Ru(bpy)23+ was studied
by Furlong
Furlong
et
al.,
using
EDTA
and
TEA.
It
was
shown
that
both
EDTA
(at
pH
=
4)
and
TEA
(at
pH
=
10)
et al., using EDTA and TEA. It was shown that both EDTA (at pH = 4) and TEA (at pH = 10)
werestrongly
stronglyadsorbed
adsorbedon
onTiO
TiO22surface
surfaceleading
leadingto
toinferior
inferiordye
dye(Ru
(Ru(dcbpy)
adsorption and
and poor
poor
were
(dcbpy)23+ )) adsorption
dye-sensitization
[49].
dye-sensitization [49].
Recently, Peng’s
Peng’s group
group studied
studied and
and comprehensively
comprehensively discussed
discussed the
the effect
effect of
of different
different electron
electron
Recently,
donors on
on hydrogen
hydrogen production
production by
by dye-sensitized
dye-sensitized photocatalysis
photocatalysis [62,75,83].
[62,75,83]. In
In one
one study,
study, they
they used
used
donors
three
widely
used
electron
donors,
TEOA,
EDTA,
and
ascorbic
acid
(AA)
to
study
the
hydrogen
three widely used electron donors, TEOA, EDTA, and ascorbic acid (AA) to study the hydrogen
generationrate
rateover
overZn-tri-P
Zn-tri-Pc N
cNc/g-C
in Figure
Figure 16
16 [62].
[62]. It
It can
can
generation
N44. .The
Theresults
resultsof
of this
this study
study are
are shown
shown in
c /g-C33N
be
observed
that
the
best
photoactivity
of
Zn-tri-P
c
N
c
/g-C
3
N
4
was
obtained
in
the
presence
of
AA
be observed that the best photoactivity of Zn-tri-Pc Nc /g-C3 N4 was obtained in the presence of AA
compared to
to that
that of
of with
with TEOA
TEOA and
and EDTA.
EDTA. However,
However,considering
considering the
the redox
redox potentials
potentials of
of sacrificial
sacrificial
compared
agentsand
andthe
theHOMO
HOMOlevel
levelofofZn-tri-P
Zn-tri-P
cN,c,the
oxidizeddye
dyewas
was
better
regenerated
presence
agents
better
regenerated
in in
thethe
presence
of
cN
c theoxidized
of
TEOA
and
EDTA
than
with
AA.
As
a
result,
the
recombination
of
oxidized
dye
and
the
excited
TEOA and EDTA than with AA. As a result, the recombination of oxidized dye and the excited electron
electron was increased in AA leading to less hydrogen production. This is in contradiction to the
obtained result which showed AA to be the best sacrificial agent among the three. They proposed
that different pH values might be the reason for such difference. It was suggested that molecular
interaction among the various compounds is important in addition to the energy levels, and
adsorption behavior of dyes is strongly related to pH of the solution. Therefore, the association of the
Inorganics 2017, 5, 34
17 of 35
was increased in AA leading to less hydrogen production. This is in contradiction to the obtained
result
which
AA to be the best sacrificial agent among the three. They proposed that different
Inorganics
2017, showed
5, 34
17 of 35
pH values might be the reason for such difference. It was suggested that molecular interaction among
the
various compounds
is important
in addition
to thehand,
energy
levels,
and adsorption
behavior
of dyes
regeneration
and hydrogen
generation.
On the other
EDTA
is known
to strongly
compete
with
is
strongly
related
to
pH
of
the
solution.
Therefore,
the
association
of
the
oxidized
dye
might
be
better
carboxylic acid-based dyes in adsorption onto TiO2 surface, resulting in lower hydrogen generation
with
acidicinAA
than16
with
the basic TEOA leading to better dye regeneration and hydrogen generation.
as shown
Figure
[62].
On the
other
hand,
EDTA
is known
tostudies
stronglyindicate
compete
with
carboxylic
dyes
in adsorption
In general, the findings
of these
the
importance
ofacid-based
the electron
donors
and their
onto
TiO2 surface,
resulting
in lower
hydrogen
as shown
Figure
16 [62].
combinational
effects
on these
compounds
withgeneration
other factors
such asinpH
and dye
concentration.
Figure 16.
16. (a)
(a) Photocatalytic
Photocatalytic H
H2 production
production rates
ratesover
overZn-tri-PcNc/g-C
Zn-tri-PcNc/g-C3N
4 in the presence of various
Figure
2
3 N4 in the presence of various
electron
donors
under
different
light
irradiation
conditions.
Conditions:
1.0 wt
wt %
% Pt-loaded
Pt-loaded
electron donors under different light irradiation conditions. Conditions: 10
10 mg
mg 1.0
−1
−
1
catalyst
with
5.0
μmol·g
Zn-tri-PcNc,
10
mL
of
water
containing
50
mM
AA,
10
vol
%
TEOA
or or
10
catalyst with 5.0 µmol·g Zn-tri-PcNc, 10 mL of water containing 50 mM AA, 10 vol % TEOA
mM
EDTA;
(b)(b)
the
and the
the CB/VB
CB/VB (for
(for
10
mM
EDTA;
therelative
relativepositions
positionsofofthe
thesacrificial
sacrificialreagents’
reagents’ redox
redox potentials
potentials and
g-C33N
as well
wellas
asthe
theHOMO/LUMO
HOMO/LUMO(for
(forZn-tri-PcNc)
Zn-tri-PcNc)
levels.
Reproduced
permission
g-C
N44),), as
levels.
Reproduced
withwith
permission
fromfrom
[62].
[62]. Copyright
American
Chemical
Society,
Copyright
American
Chemical
Society,
2014. 2014.
6.3. An
Overviewthe
of Photocatalyst
Modification
In general,
findings of these
studies indicate the importance of the electron donors and their
combinational
on these compounds
with other factors
as pH
and dye concentration.
Like othereffects
photocatalytic
processes, semiconductors
are such
essential
compounds
in dye-sensitized
photocatalysis. To facilitate electron transfer and achieve dye-sensitized hydrogen generation, the CB
6.3. An Overview of Photocatalyst Modification
level of the semiconductors should fall below the dye’s LUMO level rise above the potential of water
reduction.
The properties
of the
semiconductor
should also
the
adsorptioninofdye-sensitized
the dye on its
Like other
photocatalytic
processes,
semiconductors
aresupport
essential
compounds
surface.
photocatalysis.
To facilitate electron transfer and achieve dye-sensitized hydrogen generation, the
The of
materials
that have been
so farfall
studied
asthe
a semiconductor
in dye-sensitized
photocatalysis
CB level
the semiconductors
should
below
dye’s LUMO level
rise above the
potential of
can
be
classified
into
two
main
groups:
metal-based
semiconductors
and
metal-free
semiconductors.
water reduction. The properties of the semiconductor should also support the adsorption of the dye
Metal-based
on
its surface.semiconductors that have been widely studied are TiO2, ZnO, SnO2 [25,59], etc. On the
otherThe
hand
the most
metal-free
materials
used in dye-sensitization
are MWCNTs
[84],
materials
that famous
have been
so far studied
as a semiconductor
in dye-sensitized
photocatalysis
GO/RGO
[52,66,70,74,77,85–87],
and g-Cmetal-based
3N4 [62,75,81].
can
be classified
into two main groups:
semiconductors and metal-free semiconductors.
Lu’s group
studied photocatalytic
hydrogen
sensitized
Metal-based
semiconductors
that have been
widelygeneration
studied areusing
TiO2 ,eosin
ZnO,YSnO
etc. On the
2 [25,59],multiwalled
carbonhand
nanotube
(MWCNT/Pt)
catalyst with
TEOA
as the
sacrificial agent are
under
visible light
other
the most
famous metal-free
materials
used
in dye-sensitization
MWCNTs
[84],
illumination.
The apparent quantum
efficiency
of 12.14% was obtained. MWCNT was treated by
GO/RGO
[52,66,70,74,77,85–87],
and g-C
3 N4 [62,75,81].
HNOLu’s
3 which
ledstudied
to the formation
of –COOH
and generation
–OH on MWCNT,
providing
anchoring
sites for
group
photocatalytic
hydrogen
using eosin
Y sensitized
multiwalled
eosin
Y.
Acid
treatment
of
CNT
introduces
oxygen-containing
groups,
such
as
hydroxyl,
carbonyl,
carbon nanotube (MWCNT/Pt) catalyst with TEOA as the sacrificial agent under visible light
and carboxylicThe
functionalities
providing
electrostatic
stabilization
when CNT
is dispersed.
The main
illumination.
apparent quantum
efficiency
of 12.14%
was obtained.
MWCNT
was treated
by
role of3 MWCNT
capture
electrons
and to decrease
recombination
[84].sites
The for
PL
HNO
which ledistotothe
formation
of –COOH
and –OHthe
on electron-hole
MWCNT, providing
anchoring
spectra
MWCNT
is shown
in introduces
Figure 17 [84].
Significant quenching
the as
photoluminescence
of
eosin
Y. of
Acid
treatment
of CNT
oxygen-containing
groups,of
such
hydroxyl, carbonyl,
eosincarboxylic
Y can befunctionalities
observed. This
confirmselectrostatic
the transferstabilization
of photogenerated
electrons
from eosin
Y to
and
providing
when CNT
is dispersed.
The main
MWCNT
which results
in better
separation
of decrease
e−/h+ pairs.
role
of MWCNT
is to capture
electrons
and to
the electron-hole recombination [84]. The PL
Theofsame
group
also studied
the17function
of reduced
graphene
oxide
for dye-sensitized
spectra
MWCNT
is shown
in Figure
[84]. Significant
quenching
of the
photoluminescence
of
hydrogen generation [85.] In this work eosin Y is used as a photosensitizer, TEOA as a sacrificial
agent, Pt loaded on the surface of RGO sheets as a catalyst and RGO as an electron mediator for
efficient transfer of the electrons. In this study, the apparent quantum efficiency of 9.3% for hydrogen
generation at 520 nm as obtained. The excited electrons are trapped by RGO owing to their excellent
electron mobility. Subsequently, the electrons are transferred to Pt nanoparticles and eventually are
Inorganics 2017, 5, 34
18 of 35
consumed in water reduction reaction [85]. The higher apparent quantum efficiency of 18.5% was
obtained over RB sensitized graphene sheets with dispersed Pt nanoparticles under visible light
irradiation of 550 nm [86]. Graphene is known as a promising material in designing hybrids
for
Inorganics 2017, 5, 34
18 of 35
photocatalysis [77,88]. Besides improving electron transfer and charge separation, dispersible
graphene sheets provide great interfaces for highly dispersing nanoparticles, enhancing hydrogen
eosin
Y cancompared
be observed.
This
the
transfer
of photogenerated
electrons
eosin Y to
generation
to GO
andconfirms
MWCNT.
The
suggested
mechanism of this
systemfrom
is illustrated
in
Inorganics
2017,
5, 34 results in better separation of e− /h+ pairs.
18 of 35
MWCNT
which
Figure 18
[86].
consumed in water reduction reaction [85]. The higher apparent quantum efficiency of 18.5% was
obtained over RB sensitized graphene sheets with dispersed Pt nanoparticles under visible light
irradiation of 550 nm [86]. Graphene is known as a promising material in designing hybrids for
photocatalysis [77,88]. Besides improving electron transfer and charge separation, dispersible
graphene sheets provide great interfaces for highly dispersing nanoparticles, enhancing hydrogen
generation compared to GO and MWCNT. The suggested mechanism of this system is illustrated in
Figure 18 [86].
Figure 17.
17. PL
Y/CMWCNT
(b),(b),
Eosin
Y/UMWCNT
(c), and
Figure
PL spectra
spectra of
of Eosin
EosinY/DMWCNT
Y/DMWCNT(a),
(a),Eosin
Eosin
Y/CMWCNT
Eosin
Y/UMWCNT
(c),
Eosin
Y
(d)
in
TEOA
solution
at
pH
7.
The
excitation
wavelength
is
450
nm.
Reproduced
with
and Eosin Y (d) in TEOA solution at pH 7. The excitation wavelength is 450 nm. Reproduced with
permissionfrom
from[84].
[84].Copyright
CopyrightAmerican
AmericanChemical
ChemicalSociety,
Society,2007.
2007.
permission
Recently, high apparent quantum efficiency (30.3%) for dye-sensitized photocatalytic hydrogen
The same group also studied the function of reduced graphene oxide for dye-sensitized hydrogen
generation under visible light using decorated graphene in the presence of Ni nanoparticles has been
generation [85]. In this work eosin Y is used as a photosensitizer, TEOA as a sacrificial agent, Pt loaded
reported. Eosin Y and trimethylamine were used as dye and electron donor. The prepared composite
on the surface of RGO sheets as a catalyst and RGO as an electron mediator for efficient transfer
(GN) performs better in dye-sensitized hydrogen generation than in single Ni and reduced graphene
of the electrons. In this study, the apparent quantum efficiency of 9.3% for hydrogen generation
Figure
PL spectra ofcan
Eosin
Y/CMWCNT
(b), Eosin
Y/UMWCNT
(c),Ni
and
oxide.
This17.
enhancement
be Y/DMWCNT
explained by(a),
theEosin
significant
synergetic
effect
of RGO and
[70].
at 520 nm as obtained. The excited electrons are trapped by RGO owing to their excellent electron
Eosin Y (d) in TEOA solution at pH 7. The excitation wavelength is 450 nm. Reproduced with
mobility.
Subsequently, the electrons are transferred to Pt nanoparticles and eventually are consumed
permission from [84]. Copyright American Chemical Society, 2007.
in water reduction reaction [85]. The higher apparent quantum efficiency of 18.5% was obtained
over Recently,
RB sensitized
with
dispersed
Pt nanoparticles
under visible
light irradiation
of
highgraphene
apparent sheets
quantum
efficiency
(30.3%)
for dye-sensitized
photocatalytic
hydrogen
550 nm [86].under
Graphene
is known
as a decorated
promising graphene
material inindesigning
hybrids
fornanoparticles
photocatalysis
[77,88].
generation
visible
light using
the presence
of Ni
has
been
Besides
improving
electron
transfer
and
charge
separation,
dispersible
graphene
sheets
provide
great
reported. Eosin Y and trimethylamine were used as dye and electron donor. The prepared composite
interfaces
for highly
nanoparticles,
enhancing
hydrogen
compared
GO and
(GN)
performs
betterdispersing
in dye-sensitized
hydrogen
generation
than in generation
single Ni and
reducedtographene
MWCNT.
The
suggested
mechanism
of
this
system
is
illustrated
in
Figure
18
[86].
oxide. This enhancement can be explained by the significant synergetic effect of RGO and Ni [70].
Figure 18. The proposed photocatalytic mechanism for efficient hydrogen evolution over a dyesensitized graphene/metal photocatalyst under visible light irradiation. Reproduced with permission
from [86]. Copyright Elsevier, 2013.
6.3.1. Role of Co-Catalyst
Co-catalysts have an essential role to enhance the activity and stability of photocatalysts. In a
dye-sensitized system, co-catalysts can suppress the charge recombination which occurs between the
electron coming from the excited state dye and the oxidation state dye. This recombination process
Figure 18. The proposed photocatalytic mechanism for efficient hydrogen evolution over a dyeFigure 18. The proposed photocatalytic mechanism for efficient hydrogen evolution over a
sensitized graphene/metal photocatalyst under visible light irradiation. Reproduced with permission
dye-sensitized graphene/metal photocatalyst under visible light irradiation. Reproduced with
from [86]. Copyright Elsevier, 2013.
permission from [86]. Copyright Elsevier, 2013.
6.3.1. Role of Co-Catalyst
Co-catalysts have an essential role to enhance the activity and stability of photocatalysts. In a
dye-sensitized system, co-catalysts can suppress the charge recombination which occurs between the
electron coming from the excited state dye and the oxidation state dye. This recombination process
Inorganics 2017, 5, 34
19 of 35
Recently, high apparent quantum efficiency (30.3%) for dye-sensitized photocatalytic hydrogen
generation under visible light using decorated graphene in the presence of Ni nanoparticles has been
reported. Eosin Y and trimethylamine were used as dye and electron donor. The prepared composite
(GN) performs better in dye-sensitized hydrogen generation than in single Ni and reduced graphene
oxide. This enhancement can be explained by the significant synergetic effect of RGO and Ni [70].
6.3.1. Role of Co-Catalyst
Co-catalysts have an essential role to enhance the activity and stability of photocatalysts. In a
19 ofthe
35
co-catalysts can suppress the charge recombination which occurs between
electron coming from the excited state dye and the oxidation state dye. This recombination process
competes
competes with
with the
the hydrogen
hydrogen generation
generation process.
process. Also,
Also, co-catalysts
co-catalysts offer
offer more
more active
active sites
sites for
for the
the
hydrogen
generation
reaction
to
take
place
[63].
In
general,
co-catalysts
can
be
classified
into
two
hydrogen generation reaction to take place [63]. In general, co-catalysts can be classified into two main
main
categories:
(i) noble
co-catalysts
and
(ii) non-noble
co-catalysts.
categories:
(i) noble
metalmetal
basedbased
co-catalysts
and (ii)
non-noble
metalmetal
basedbased
co-catalysts.
Inorganics
2017, 5, 34
dye-sensitized
system,
Noble-Metal Based Co-Catalysts
The
as as
co-catalyst
is Schottky
junction
that that
is created
due to
theto
close
The key
keyconcept
conceptininusing
usingmetals
metals
co-catalyst
is Schottky
junction
is created
due
the
association
of a metal
n-type
semiconductor.
Particular
energy level
alignment
in a Schottky
close association
of a and
metal
and n-type
semiconductor.
Particular
energy
level alignment
in a
junction
a potential
barrier in
preventing
electronelectron
and holes
through
them
Schottky provides
junction provides
a potential
barrier
in preventing
andfrom
holespassing
from passing
through
[21,89].
Schottky
junction
facilitates
the transfer
of electrons
to the
contacting
metal,
as shown
in
them [21,89].
Schottky
junction
facilitates
the transfer
of electrons
to the
contacting
metal,
as shown
Figure
19 19
[19,21,90].
In this
process,
metal
captures
an an
electron
andand
accelerates
thethe
separation
of
in Figure
[19,21,90].
In this
process,
metal
captures
electron
accelerates
separation
photo-generated
charges.
TheThe
energy
band
diagram
of of
a metal-semiconductor
Schottky
junction
is
of photo-generated
charges.
energy
band
diagram
a metal-semiconductor
Schottky
junction
depicted
in Figure
19 [90].
In this
Фb represents
the elevation
of the potential
barrier,barrier,
which
is depicted
in Figure
19 [90].
In figure,
this figure,
Φb represents
the elevation
of the potential
depends
on the on
band
of the of
semiconductor
and the
Fermi
level of
metal
[89]. It[89].
has Itbeen
which depends
thebending
band bending
the semiconductor
and
the Fermi
level
of metal
has
recognized
that the
bandband
of the
should
be higher
than
thethe
Fermi
level
of
been recognized
thatconduction
the conduction
ofsemiconductor
the semiconductor
should
be higher
than
Fermi
level
the
incorporated
metal.
of the
incorporated
metal.
(a)
(b)
Figure
Schottky
junction
with
energy
levellevel
alignment
and (b)
electron
transfer
across the
metal–
Figure 19.
19.(a)(a)
Schottky
junction
with
energy
alignment
and
(b) electron
transfer
across
the
semiconductor
junction.
Reproduced
with
permission
from
[90].
Copyright
Royal
Society
metal–semiconductor junction. Reproduced with permission from [90]. Copyright Royal Society of
of
Chemistry,
2014.
Chemistry, 2014.
Till now, incorporation of noble metals such as Au, Ag, Pt, Pd, Ru, and Rh on semiconductors
Till now, incorporation of noble metals such as Au, Ag, Pt, Pd, Ru, and Rh on semiconductors has
has been widely investigated to achieve high hydrogen generation. Among them, Pt is the most
been widely investigated to achieve high hydrogen generation. Among them, Pt is the most popular
popular one owing to its low Fermi level, and it is considered to be successful in hindering photoone owing to its low Fermi level, and it is considered to be successful in hindering photo-generated
generated charges recombination [91]. Abe et al. [92,93] reported quantum efficiency equal to 2.5%
charges recombination [91]. Abe et al. [92,93] reported quantum efficiency equal to 2.5% for hydrogen
for hydrogen production over merocyanine or coumarin dye-sensitized Pt/TiO2 photocatalysts with
production over merocyanine or coumarin dye-sensitized Pt/TiO2 photocatalysts with the use of
the use of visible light irradiation in a water–acetonitrile mixed solution including
I− as an electron
visible light irradiation in a water–acetonitrile mixed solution including I− as an electron donor.
donor. The influence of various noble metal (Pt, Rh, and Ru) loading on TiO2 for dye-sensitized
The influence of various noble metal (Pt, Rh, and Ru) loading on TiO2 for dye-sensitized hydrogen
hydrogen production using different electron donors such as triethanolamine, acetonitrile, and
production using different electron donors such as triethanolamine, acetonitrile, and triethylamine
triethylamine under visible light irradiation was investigated [94]. It was observed that the
under visible light irradiation was investigated [94]. It was observed that the adsorption of dye with
adsorption of dye with the surge of the dispersed metal. In this study, relatively high quantum
efficiency (10.3%) was obtained using 10 wt % eosin-Rh/TiO2 photocatalyst.
Noble-Metal Free Co-Catalysts
Although it is common to deposit noble metals as co-catalyst on semiconductors to improve
Inorganics 2017, 5, 34
20 of 35
the surge of the dispersed metal. In this study, relatively high quantum efficiency (10.3%) was obtained
using 10 wt % eosin-Rh/TiO2 photocatalyst.
Noble-Metal Free Co-Catalysts
Although it is common to deposit noble metals as co-catalyst on semiconductors to improve their
Inorganics 2017, 5, performance,
34
20 of to
35
photocatalytic
the scarcity and high cost of these metals is an issue. Researchers prefer
find inexpensive and earth-abundant alternatives for metals to be applicable in large scale processes.
Nickel
sulfide
is on
onenon-noble
of these metal
materials
that have
studied metal
for photocatalytic
hydrogen
Recently,
some
studies
co-catalysts
suchbeen
as transition
oxides, transition
metal
production
[64].
NiS
x-decorated graphene (NiSx/G) exhibited high photoactivity for hydrogen
hydroxides, and transition metal sulfides have been conducted. Ni [64,70,95], Cu [96,97], and Co [98,99]
generation
underand
the molybdenum
visible light when
it is sensitized
EYexamples
using TEOA.
In situ research
chemical to
deposition
based
materials
disulfides
[100,101]byare
of current
develop
method
was
applied
as
a
straightforward
and
scalable
process.
Very
high
quantum
yield
of 32.5%
noble-metal-free co-catalysts for hydrogen generation.
was obtained
at 430 is
nm.
The
of this study
NiSstudied
x/G can be a promising substitute for
Nickel sulfide
one
of results
these materials
thatshow
havethat
been
for photocatalytic hydrogen
noble
metals.
production [64]. NiSx -decorated graphene (NiSx /G) exhibited high photoactivity for hydrogen
Du’s group
found
nickel
(Ni(OH)2by
) as
active
co-catalyst
for EY-sensitized
generation
under the
visible
light hydroxide
when it is sensitized
EYan
using
TEOA.
In situ chemical
deposition
hydrogen
generation.
Different
transition
metal
based
hydroxides
and
oxides
such
as
cobalt
oxide
method was applied as a straightforward and scalable process. Very high quantum yield of 32.5%
(CoO
x), cobalt hydroxide (Co(OH)2), nickel oxide (NiOx), ferric hydroxide (Fe(OH)3) and copper
was obtained at 430 nm. The results of this study show that NiSx /G can be a promising substitute for
hydroxide
(Cu(OH)2) as co-catalyst on the surface of TiO2 semiconductor were compared with nickel
noble
metals.
hydroxide
co-catalyst.
As shown
in (Ni(OH)
Figure 20
[95], Ni(OH)2 could outperform others for
Du’s group
found nickel
hydroxide
2 ) as an active co-catalyst for EY-sensitized hydrogen
photocatalytic
hydrogen
production
[95].
generation. Different transition metal based hydroxides and oxides such as cobalt oxide (CoOx ),
Inhydroxide
another study,
Co and
CoSx nanoparticles were deposited on the graphene surface by onecobalt
(Co(OH)
2 ), nickel oxide (NiOx ), ferric hydroxide (Fe(OH)3 ) and copper hydroxide
step
photo-reduction
and
in
situ
chemical
methods. were
The high
apparent
quantum
efficiency
(Cu(OH)2 ) as co-catalyst on the surface
of deposition
TiO2 semiconductor
compared
with
nickel hydroxide
of 8.7% withAs
the
use ofinEY-Co/G
obtained
520 nm illumination [98]. These results propose
co-catalyst.
shown
Figure 20was
[95],
Ni(OH)under
2 could outperform others for photocatalytic hydrogen
a
potential
alternative
for
noble
metals
as
co-catalysts.
production [95].
20. Hydrogen production
production from
from CoO
CoOxx,, Co(OH)
Co(OH)22, NiO
NiOxx, Ni(OH)
Ni(OH)22, Fe(OH)33 and Cu(OH)
Cu(OH)22,
Figure 20.
respectively in
inEY
EYsolution
solutionat at
9 5for
h. Using
W Xenon
arcaslamp
as the
light
source
pHpH
9 for
h. 5Using
300 W300
Xenon
arc lamp
the light
source
and
a cut
and
filter to the
eliminate
thebelow
photons
420 nm. Reproduced
with permission
from [95].
filteratocut
eliminate
photons
420 below
nm. Reproduced
with permission
from [95]. Copyright
Copyright
Elsevier, 2014.
Elsevier, 2014.
In another
addition,
recent
that molybdenum disulfide (MoS2) could be a suitable coIn
study,
Coresearch
and CoSproves
x nanoparticles were deposited on the graphene surface by one-step
catalyst in photocatalytic
hydrogendeposition
generation
processes
[100,101].
Lu’s
group efficiency
reported ofa
photo-reduction
and in situ chemical
methods.
The high
apparent
quantum
MoS
2
/graphene
nanohybrid
as
an
active
and
low-cost
material
for
solar
hydrogen
production
[101].
8.7% with the use of EY-Co/G was obtained under 520 nm illumination [98]. These results propose
a
A
one-pot
hydrothermal
reaction
of
Na
2
MoO
4
·2H
2
O
and
NH
2
CSNH
2
in
exfoliated
graphite
oxide
potential alternative for noble metals as co-catalysts.
(GO)In
aqueous
suspensions
synthesized
MoS
2/RGO
nanohybrid.
The apparent
of
addition,
recent research
proves
that
molybdenum
disulfide
(MoS2 quantum
) could beefficiency
a suitable
24%
at
460
nm
over
an
eosin
Y-sensitized
MoS
2
/RGO
photocatalyst
has
been
reported.
This
enhanced
co-catalyst in photocatalytic hydrogen generation processes [100,101]. Lu’s group reported a
photoactivity
of MoS2/RGO is ascribed to the electrical coupling and synergistic effect between MoS2
MoS
2 /graphene nanohybrid as an active and low-cost material for solar hydrogen production [101].
and
RGO sheets.
The two-dimensional
conductive
network is provided by RGO resulting in the
A
one-pot
hydrothermal
reaction of Na2 MoO
4 ·2H2 O and NH2 CSNH2 in exfoliated graphite oxide
efficient
transfer
of
photo-excited
dye
to
the
active
sites of MoS
2. In
this study,
different
halogen
(GO) aqueous suspensions synthesized MoS2 /RGO nanohybrid.
The
apparent
quantum
efficiency
of
substitutions to the xanthene ring, along with other xanthene dyes have also been examined. The
order of hydrogen generation using different dye as photo-sensitizer is: eosin Y > rose bengal >
dichlorofluorescein > fluorescein sodium > rhodamine B.
7. Application of Design of Experiment in Photocatalytic Hydrogen Generation
Inorganics 2017, 5, 34
21 of 35
24% at 460 nm over an eosin Y-sensitized MoS2 /RGO photocatalyst has been reported. This enhanced
photoactivity of MoS2 /RGO is ascribed to the electrical coupling and synergistic effect between MoS2
and RGO sheets. The two-dimensional conductive network is provided by RGO resulting in the
efficient transfer of photo-excited dye to the active sites of MoS2 . In this study, different halogen
substitutions to the xanthene ring, along with other xanthene dyes have also been examined.
The order of hydrogen generation using different dye as photo-sensitizer is: eosin Y > rose bengal >
dichlorofluorescein > fluorescein sodium > rhodamine B.
7. Application of Design of Experiment in Photocatalytic Hydrogen Generation
Although photocatalytic water splitting/hydrogen generation is an emerging approach used
for renewable hydrogen generation, only a few researchers have developed methodologies on the
basis of the statistical design of experiments. Most of the studies use the single-variable-at-a-time
(SVAT) approach to examine the effect of different parameters such as solution pH, photocatalyst
concentration, electron donor concentration, sensitizer (dye) concentration, and light intensity on
hydrogen generation rate. However, such SVAT approach is time-consuming and expensive and
unable to provide optimization towards response(s). Moreover, this approach does not consider the
interaction effect of two or more factors that can limit the process efficiency. In contrast, the effect
of selected operating factors could be screened by using a factorial design methodology. This helps
evaluate the interaction effects of two or more factors and determine the significant factors that affect
the response with a minimum number of experiments [100,102,103]. Factorial designs can be either
a full factorial design or a fractional factorial design at 2 or 3-levels. In a 2-level full factorial design,
the number of experiments is 2k , where k is the number of variables. This 2k full factorial design is
useful if the number of variables is less than or equal to four. If the number of variables affecting the
process is greater than four, it is recommended to perform a fractional factorial design [104]. Besides,
the optimal conditions can be obtained by a robust statistical method of optimization method titled
central composite design (CCD). This approach is rapid and more accurate compared to the SVAT
approach, which also suffers from overlooking the interaction effect of the various factors [100,102,103].
Dye-sensitized photocatalytic water splitting/hydrogen generation systems have been widely
studied on account of its superior visible light harvesting capabilities. There is, however, lack of
information on investigations of the principal process parameters which influence sacrificial hydrogen
generation/water splitting based on robust experimental design methodology and statistical analysis.
In Ray’s group, Chowdhury et al. [5] filled such gap, developing a method for eosin Y-sensitized
photocatalysis for sacrificial hydrogen generation based on factorial design analysis. They also
defined the optimum conditions for the significant factors that influence the hydrogen generation.
They performed a factorial design at two levels and four factors to investigate the potential of eosin
Y-sensitized TiO2 /Pt photocatalyst under the visible solar light in the presence of triethanolamine
(TEOA) as an electron donor for hydrogen generation. Solution pH interacted with Eosin Y and
triethanolamine also influenced the lifespan of dye-sensitized TiO2 /Pt. The visible light irradiation
time had the maximum effect on hydrogen generation followed by the effect of platinum content (wt %)
in TiO2 . Figure 21 below provides the two-factors interaction effect at all possible combinations of the
parameters. Parallel lines indicate an insignificant interaction between the two factors (Figure 21) [5].
“Pareto analysis” as well as conventional regression analysis techniques were used to analyze
experimental data. Authors used Minitab 15 software to generate regression function with the
confidence level of 95%. The regression function obtained is as follows:
H2 generation
(% increment)
= −26.7 + 7.9 × ( TEOA) + 16.5 × ( Eosin Y ) + 25.8 × ( Pt) + 38.8
×(I − time) − 3.6 × ( TEOA) × ( Pt) + 5.7 × ( TEOA) × (I − time)
+6.3 × ( Eosin Y ) × (I − time) − 15.2 × ( Pt) × (I − time)
(14)
Inorganics 2017, 5, 34
22 of 35
where, TEOA (triethanolamine) concentration in M, Eosin Y concentration in M, Pt (platinum) content
in wt % and visible light irradiation time (I-time) in min.
Inorganics 2017, 5, 34
22 of 35
Inorganics 2017, 5, 34
22 of 35
Figure 21. Interaction effect plot for electron donor (TEOA), sensitizer (Eosin Y dye), platinum, and
Figure 21. Interaction effect plot for electron donor (TEOA), sensitizer (Eosin Y dye), platinum, and
visible light irradiation time (I-time) for the percent increment in H2 generation. Reproduced with
visible
light irradiation
time (I-time)
for the
percent increment in H2 generation. Reproduced with
permission
from [5]. Copyright
Elsevier,
2011.
permission
from
[5]. Copyright
Elsevier,
2011.donor (TEOA), sensitizer (Eosin Y dye), platinum, and
Figure 21.
Interaction
effect plot
for electron
visible
light study,
irradiation
time (I-time)
the[100]
percent
increment
in H2 generation.photocatalytic
Reproduced with
In another
Malekshoar
andfor
Ray
proposed
a noble-metal-free
system
permission
from [5].
Copyrightgeneration
Elsevier, 2011.
for
eosin
Y-sensitized
hydrogen
under
visible
solar
light
with
TEOA
electron
donor.
They
In another study, Malekshoar and Ray [100] proposed a noble-metal-free photocatalytic
system
used a statistical design of experimental analysis to find the impacts of different parameters such as
for eosinInY-sensitized
hydrogen
generation
under
visible
solar
light
with
TEOA
electron
study,
Malekshoar and
Ray [100]
proposed
a noble-metal-free
photocatalyticoptimize
systemdonor.
solutionanother
pH, initial
concentrations
of TEOA,
eosin
Y and (NH
4)2MoS4, and subsequently,
Theyfor
used a Y-sensitized
statistical design
of experimental
analysis solar
to find
the
impacts
of different
parameters
hydrogen
generationAt
under
light
with
TEOA
donor.
They
theeosin
photocatalytic hydrogen
generation.
first,visible
a full factorial
design
was electron
conducted
to screen
suchused
as
solution
pH,
initial
concentrations
of
TEOA,
eosin
Y
and
(NH
)
MoS
,
and
subsequently,
a statistical
of experimental
analysis
to find ascent
the impacts
different
parameters
suchfor
as
4 2composite
4
different
factors,design
followed
by the path
of steepest
and of
central
design
optimize
thepH,
photocatalytic
generation.
At
first,
a full
factorial
design
was conducted
solution
initial
concentrations
TEOA, eosin
Y and
(NH
4)2MoS
4, and subsequently,
optimize
optimization.
Analysis
ofhydrogen
Varianceof (ANOVA)
confirmed
the
competence
of the model
for the to
the
photocatalytic
hydrogen
generation.
At first,
aand
full(NH
factorial
design
was conducted
to screen
screen
different
followed
byforthe
path
of
ascent
central
composite
design
response
andfactors,
the optimum
values
solution
pHsteepest
4)2MoS4 and
concentration
were found
to be for
different
followed
by the
path ofconfirmed
steepest
ascent
and central
composite
design
for
optimization.
Analysis
of
Variance
(ANOVA)
the production.
competence
the
model
for
the response
at 10.7 andfactors,
1.19 3 M,
respectively,
to maximize
the
hydrogen
Aof
response
surface
plot
and
optimization.
Analysis
of
Variance
(ANOVA)
confirmed
the
competence
of
the
model
for
the
the
corresponding
contour
plot
for
the
factors
(Figure
22)
[100]
showed
that
both
the
solution
pH
and
and the optimum values for solution pH and (NH4 )2 MoS4 concentration were found to be at 10.7
response
and
the optimum had
values
for solution
pH
(NH4)2MoS
were
found
to be
(NH4)
MoS
4 concentration
a positive
on and
the production.
response
up4 concentration
toAcertain
point.
After
reaching
and 1.19
3 2M,
respectively,
to maximize
theeffect
hydrogen
response
surface
plot and the
at
10.7
and 1.19point,
3 M, respectively,
to maximize
the
hydrogen
production.
A response
surface
plot
and
the
maximum
a
further
increase
in
those
factors
had
an
adverse
effect
on
the
response.
It
corresponding contour plot for the factors (Figure 22) [100] showed that both the solutionwas
pH and
the
corresponding
contour
plot
for
the
factors
(Figure
22)
[100]
showed
that
both
the
solution
pH
and
−1
predicted that 4790 μmol·g·cat would be produced at the optimal conditions. The average hydrogen
(NH4)
had a positive effect on the response up to certain point. After
reaching
2 MoS
4 concentration
(NH4)
2MoS
4 concentration had a positive effect on the response up to certain point. After
reaching
production
after 3 h of the solar irradiation was found to be 4890 ± 151 μmol·g·cat−1, which
is in
the maximum
point,
a
further
increase
in
those
factors
had
an
adverse
effect
on
the
response.
the
maximum
point, a with
further
in value.
those factors had an adverse effect on the response. It wasIt was
reasonable
agreement
theincrease
predicted
−
1
−1
predicted
that that
47904790
µmol
·g·cat would
producedatatthe
theoptimal
optimal
conditions.
average
hydrogen
predicted
μmol·g·cat
would be produced
conditions.
TheThe
average
hydrogen
−1 , which
production
of the
solar
irradiationwas
wasfound
found to be
be 4890
, which
is in is in
production
afterafter
3 h 3ofh the
solar
irradiation
4890 ±±151
151μmol·g·cat
µmol·g·−1cat
reasonable
agreement
with
predictedvalue.
value.
reasonable
agreement
with
thethe
predicted
Figure 22. Response surface and contour plot for solution pH and (NH4)2MoS4 concentration.
Reproduced with permission from [100]. Copyright Elsevier, 2016.
Figure 22. Response surface and contour plot for solution pH and (NH4)2MoS4 concentration.
Figure 22. Response surface and contour plot for solution pH and (NH4 )2 MoS4 concentration.
Reproduced with permission from [100]. Copyright Elsevier, 2016.
Reproduced with permission from [100]. Copyright Elsevier, 2016.
Inorganics 2017, 5, 34
23 of 35
8. Photoreactor Configurations for the Improvement of Water Splitting/Hydrogen Generation
Photoreactor development is one of the crucial areas towards the industrial application of
photocatalysis. There are mainly three design parameters that need to be optimized for successful
scaling up of photo-reactors, such as (i) photoreactor geometry (deals with arrangement of light
source and the state of photocatalyst); (ii) type of photocatalyst (immobilized or slurry form) and
(iii) utilization of incident light (area of illuminated photocatalyst) [105]. In a conventional reactor,
we normally deal with factors such as reaction kinetics, proper contact between reactants and catalyst,
effective mass transfer, etc. In the case of photoreactor, in addition to all the above factors, we also
must consider the photocatalyst illumination factor because of its utmost importance in photocatalyst
activation [105]. An optimum light illumination would increase the absorbed photons per unit
time and per unit volume and thereby improve the overall process efficiency. However, it is not an
easy job to provide uniform light distribution inside the photoreactor to a large surface area of the
photocatalyst. There are several phenomena that govern the light distribution viz. light scattering,
opacity, light penetration depth and local volumetric light absorption [105]. Ray [105] came up with
a factor called illuminated surface area of photocatalyst (κ, m2 ·m−3 ) representing the amount of
photocatalyst inside the photoreactor which is adequately illuminated to promote the reaction. Again,
the photocatalyst can be used as immobilized or slurry form inside the photoreactor. Immobilized
photoreactors suffer from lower efficiency due to mass transfer limitation and high light scattering.
On the other hand, there would be no mass transfer limitations in a slurry photoreactor with nano-sized
photocatalyst with negligible diffusion length. According to Chen and Ray [106], a smaller slurry
photoreactor can provide a high ratio of illuminated photocatalyst surface area to the effective
photoreactor volume. However, for larger slurry photoreactor the κ value would be high and thus
slurry photoreactor scale up is not possible. There are few other photoreactor configurations in use
such as (i) external type with light source outside the photoreactor [107]; (ii) immersion type with light
source submerged within photoreactor [108] and (iii) distributive type with the light dispersed from
the source to the photoreactor by reflectors and light conductors [109]. All these photoreactors are
compared regarding κ value to evaluate their design efficiency and scale up prospect.
In the case of photocatalytic overall water splitting, gas separation is the additional design criteria
that need to be addressed along with other parameters. Numerous photoreactors are available in
the area of water and wastewater treatment, but very little research is performed on photocatalytic
water splitting/hydrogen generation photoreactors. In the case of photocatalytic water splitting,
the photoreactors need to include few additional accessories such as (i) evacuation system to eliminate
air/oxygen from photoreactor; (ii) gas separation system to separate hydrogen and oxygen; and (iii)
gas detection system for hydrogen and/or oxygen [110]. Moreover, the photoreactors need to have
proper sealing to protect reactant from the air and to avoid the loss of generated gasses (H2 and O2 )
during photocatalytic water splitting. Photocatalytic membrane reactors (PMR) provide simultaneous
separation of product gas and recovery of photocatalyst and allow continuous operation [111].
8.1. Photocatalytic Membrane Reactor for Separating Hydrogen and Oxygen
Yu et al. [112] mentioned the use of a twin-reactor for the Z-scheme water splitting where they
used Pt/SrTiO3 :Rh for hydrogen generation, and BiVO4 for oxygen generation. The reactor comprises
of two compartments separated by a modified Nafion membrane which stopped the backward reaction
during water splitting (Figure 23). Pt/SrTiO3 :Rh showed visible-light activity around 450–650 nm
range due to Rh doping. BiVO4 also showed visible-light activity because of its characteristic band
gap of 2.38 eV. Photocatalyst concentrations of 2 g·L−1 were used at each compartment along with
2 mM redox mediator for the regeneration of photocatalysts. A 300 W xenon lamp was used as a light
source with an irradiance of 1.76 mW·cm−2 . They reported the hydrogen generation at Pt/SrTiO3 :Rh
photocatalyst as the rate determining step of the water splitting reaction. The use of twin reactor
minimized the deactivation of Pt/SrTiO3 :Rh photocatalyst and the modified Nafion membrane in the
twin reactor did not obstruct the water splitting reaction.
Inorganics 2017, 5, 34
Inorganics2017,
2017,5,
5,34
34
Inorganics
24 of 35
24of
of35
35
24
Figure 23.
A
connected
twin photoreactor
photoreactor with
modified Nafion
Nafion membrane.
membrane. Reproduced
Reproduced
with
Figure
23. A
A connected
connected twin
with aa modified
modified
Reproduced with
permission from
from [112].
Copyright Elsevier,
Elsevier, 2011.
[112]. Copyright
2011.
permission
A
similar
twin
reactor
was
designed
by
Lo
et
al.
[113]
for
the
photocatalytic
water
splitting
under
A
Asimilar
similartwin
twinreactor
reactorwas
wasdesigned
designed by
byLo
Loet
etal.
al.[113]
[113]for
forthe
thephotocatalytic
photocatalyticwater
watersplitting
splittingunder
under
visible
light.
Pt/SrTiO
:Rh
and
WO
were
used
for
hydrogen
generation
and
oxygen
generation,
visible
light.
Pt/SrTiO
33:Rh
and
WO
33 were
used
for
hydrogen
generation
and
oxygen
generation,
visible light. Pt/SrTiO3 :Rh and WO3 were used for hydrogen generation and oxygen generation,
respectively.
In the
the
twin
reactor,
two compartments
compartments
were
separated
by
modified
Nafion
membrane
respectively.
respectively. In
In
thetwin
twinreactor,
reactor, two
two
compartments were
were separated
separated by
byaaamodified
modifiedNafion
Nafionmembrane
membrane
to
generate
hydrogen
and
oxygen
separately.
The
photocatalytic
solution
also
contained
Fe(II)
and
to
generate
hydrogen
and
oxygen
separately.
The
photocatalytic
solution
also
contained
to generate hydrogen and oxygen separately. The photocatalytic solution also contained Fe(II)
Fe(II) and
and
Fe(III)
as
electron-transfer
mediator.
In
this
case,
the
backward
reaction
was
significantly
reduced,
Fe(III)
as
electron-transfer
mediator.
In
this
case,
the
backward
reaction
was
significantly
reduced,
Fe(III) as electron-transfer mediator. In this case, the backward reaction was significantly reduced,
and
they
achieved
much
higher
hydrogen generation
generation rate
rate than
than the
the conventional
conventional Z-scheme
Z-scheme system.
system.
and
andthey
they achieved
achieved much
much higher
higher hydrogen
hydrogen
generation
rate
than
the
conventional
Z-scheme
system.
Kitano
et
al.
[114]
demonstrated
a
thin
film
device
consisting
of
titania
film
and
Ti-foil
(50
μm)
Kitano
et
al.
[114]
demonstrated
a
thin
film
device
consisting
of
titania
film
and
Ti-foil
Kitano et al. [114] demonstrated a thin film device consisting of titania film and Ti-foil(50
(50μm)
µm)
one
side
for
oxygen
evolution,
and
with
platinum
for
hydrogen
generation
on
the
other
side
of
an
Hone
side
for
oxygen
evolution,
and
with
platinum
for
hydrogen
generation
on
the
other
side
of
an
one side for oxygen evolution, and with platinum for hydrogen generation on the other side ofHan
type
photoreactor
(Figure
24).
They
used
an
aqueous
solution
of
sulfuric
acid
(0.5
M)
and
sodium
type
photoreactor
(Figure
24).
They
used
an
aqueous
solution
of
sulfuric
acid
(0.5
M)
and
sodium
H-type photoreactor (Figure 24). They used an aqueous solution of sulfuric acid (0.5 M) and sodium
hydroxide
(1
M)
in
the
platinum and
and titania
titania side,
side, respectively.
respectively.
hydroxide
hydroxide (1
(1 M)
M) in
in the
the platinum
platinum
and
titania
side,
respectively.
24. Diagram
Diagram of
the H-type
H-type photoreactor
photoreactor for
for the
Figure 24.
Diagram
of the
the separation
separation of
of H
H22 and
and O
O222.. Reproduced
Reproduced with
with
Figure
[114].
2006.
permission
from
Copyright
Elsevier,
permission from [114]. Copyright Elsevier, 2006.
In
recent
study,
Sasaki
et
al.
[115]
used
membrane
photoreactor
for
visible-light-driven
water
In
Inaaarecent
recentstudy,
study, Sasaki
Sasaki et
etal.
al.[115]
[115]used
usedaaamembrane
membranephotoreactor
photoreactorfor
forvisible-light-driven
visible-light-drivenwater
water
splitting
in
the
presence
of
an
electron
mediator.
Ru/SrTiO
3
:Rh
and
BiVO
4
photocatalyst
were
used
splitting
in
the
presence
of
an
electron
mediator.
Ru/SrTiO
3
:Rh
and
BiVO
4
photocatalyst
were
splitting in the presence of an electron mediator. Ru/SrTiO3 :Rh and BiVO4 photocatalyst wereused
used
for
hydrogen
and
oxygen generation,
generation, respectively.
respectively.
A
membrane
was
used
to
separate
the
gasses
for
for hydrogen
hydrogen and
and oxygen
oxygen
generation,
respectively. A
A membrane
membranewas
was used
used to
to separate
separate the
the gasses
gasses
−1
−1
−1 and
preventing
thebackward
backwardreaction.
reaction.After
After
h irradiation,
irradiation,
hydrogen
production
rate
ofmL
110
mL·h
preventing
the
backward
reaction.
After
aa hydrogen
production
of
110
preventing the
3 3h3 h
irradiation,
a hydrogen
production
rate rate
of
110
·hmL·h
−1 were obtained at the corresponding compartment (Figure
−1
−1 were
and
oxygen
production
rate
of
43·hmL·h
mL·h
and
oxygen
production
43
were
obtained
at corresponding
the corresponding
compartment
(Figure
oxygen
production
rate rate
of
43of
mL
obtained
at the
compartment
(Figure
25).
25).
25).
Inorganics 2017, 5, 34
25 of 35
Inorganics 2017, 5, 34
Inorganics 2017, 5, 34
25 of 35
25 of 35
Figure 25. Membrane-based
photoreactor
for
the separation
separation of H
H2 and O
O2 generated from
from water
Figure
Membrane-based photoreactor
photoreactor for
for the
the
Figure 25.
25. Membrane-based
separation of
of H22 and
and O22 generated
generated from water
water
splitting
using
solar
light.
Reproduced
with
permission
from
[115].
Copyright
American
Chemical
splitting using solar light. Reproduced with permission from [115]. Copyright American Chemical
splitting using solar light. Reproduced with permission from [115]. Copyright American Chemical
Society, 2013.
2013.
Society,
Society, 2013.
8.2. Dual
Dual Bed
Bed Photoreactor for
for Water Splitting
Splitting
8.2.
8.2. Dual
Bed Photoreactor
Photoreactor for Water
Water Splitting
Hole scavengers play
play a crucial role
role in sacrificial
sacrificial hydrogen generation
generation by reducing
reducing the
Hole
Hole scavengers
scavengers play aa crucial
crucial role in
in sacrificial hydrogen
hydrogen generation by
by reducing the
the
recombination
of
charge
carriers.
However,
the
hydrogen
generation
process
will
only
be
continuous
recombination
However, the
recombination of
of charge
charge carriers.
carriers. However,
the hydrogen
hydrogen generation
generation process
process will
will only
only be
be continuous
continuous
if the hole scavenger is constantly regenerated in the photoreactor. Yan et al. [116] designed a novel
if
in the
the photoreactor.
photoreactor. Yan
Yan et
if the
the hole
hole scavenger
scavenger is
is constantly
constantly regenerated
regenerated in
et al.
al. [116]
[116] designed
designed aa novel
novel
dual bed reactor (Figure 26) where they combined a photocatalytic reaction bed and a hole scavenger
dual
bed
reactor
(Figure
26)
where
they
combined
a
photocatalytic
reaction
bed
and
a
hole
scavenger
dual bed reactor (Figure 26) where they combined a photocatalytic reaction bed and a hole scavenger
regeneration bed for continuous hydrogen generation for a longer duration compared to a single bed
regeneration
regeneration bed
bed for
for continuous
continuous hydrogen
hydrogen generation
generation for
for aa longer
longer duration
duration compared
compared to
to aa single
single bed
bed
photoreactor without such regeneration bed.
photoreactor
without
such
regeneration
bed.
photoreactor without such regeneration bed.
Figure 26. Schematic diagram of a dual bed photoreactor for water splitting. Reproduced with
Figure 26. Schematic diagram of a dual bed photoreactor for water splitting. Reproduced with
permission
from [116].
Copyright
Elsevier,
2011.
Figure
26. Schematic
diagram
of a dual
bed photoreactor
for water splitting. Reproduced with permission
permission from [116]. Copyright Elsevier, 2011.
from [116]. Copyright Elsevier, 2011.
In this photoreactor, Pt–TiO2 photocatalyst was used at a concentration of 2 g·L−1 with 0.4 M KI
In this photoreactor, Pt–TiO2 photocatalyst was used at a concentration of 2 g·L−1 with 0.4 M KI
solution.
High-pressure
mercury
lamp
(300 W) was
used
as
the
light source. The
photoreactor
was
In this
photoreactor,mercury
Pt–TiO2lamp
photocatalyst
wasused
usedas
atthe
a concentration
·L−1 with 0.4was
M
solution.
High-pressure
(300 W) was
light source. of
The2 g
photoreactor
purged
with
nitrogen
for
20
min
before
light
illumination
to
make
the
reactor
oxygen
free.
In
the
KI
solution.
lamp
(300illumination
W) was used
the light
source.oxygen
The photoreactor
purged
with High-pressure
nitrogen for 20mercury
min before
light
to as
make
the reactor
free. In the
regeneration
bed,nitrogen
they used
a min
glassbefore
chromatographic
column
filledthe
with
Cu2Ooxygen
and KI
aqueous
was
purged
with
for
20
light
illumination
to
make
reactor
free.
In the
regeneration bed, they used a glass chromatographic column filled with Cu2O and KI
aqueous
solution.
In
the
photoreactor
bed,
iodide
ion
acted
as
hole
scavenger
and
was
oxidized
to
iodine.
regeneration
bed,photoreactor
they used a glass
with Cu
aqueoustosolution.
2 O and
solution. In the
bed,chromatographic
iodide ion actedcolumn
as holefilled
scavenger
and
was KI
oxidized
iodine.
After
that,
the iodinebed,
reacted
with
Cuacted
2O inside the regeneration bed to produce iodine ion which then
In
the
photoreactor
iodide
ion
as
hole
scavenger
and
was
oxidized
to
iodine.
After then
that,
After that, the iodine reacted with Cu2O inside the regeneration bed to produce iodine ion which
served
as reacted
the holewith
scavenger
again.the
The
dual-bed system
showed
aiodine
steady
high
ratethen
of hydrogen
the
iodine
Cu
O
inside
regeneration
bed
to
produce
ion
which
served
as
2
served as the hole scavenger
again. The dual-bed system showed a steady high rate of hydrogen
production
for
more
than
50
h.
the
hole scavenger
again.
production
for more
thanThe
50 h.dual-bed system showed a steady high rate of hydrogen production for
Bae
et50
al.h.[117] demonstrated the overall photocatalytic water splitting in a simulated dual-bed
moreBae
thanet
al. [117] demonstrated the overall photocatalytic water splitting in a simulated dual-bed
system
under
visible
light irradiation.
They
used
WO3 and Rh-doped
SrTiO3 as
oxygen
and hydrogen
Bae
et
al.
[117] demonstrated
the They
overall
photocatalytic
water splitting
in oxygen
a simulated
dual-bed
system under visible
light irradiation.
used
WO3 and Rh-doped
SrTiO3 as
and hydrogen
2+/Fe3+ redox couple was used as a redox mediator in the
producing
photocatalyst,
respectively.
Fe
system
under
visible light irradiation.
They
Rh-doped
SrTiO
oxygen
and hydrogen
3 and
producing
photocatalyst,
respectively.
Fe2+used
/Fe3+ WO
redox
couple
was used
as3 as
a redox
mediator
in the
water splitting reaction. In the simulated dual-bed at first the hydrogen generation reaction was
water splitting reaction. In the simulated dual-bed at first the hydrogen generation reaction was
performed, then the reaction mixture was filtered, and finally, the filtrate solution was used for
performed, then the reaction mixture was filtered, and finally, the filtrate solution was used for
Inorganics 2017, 5, 34
26 of 35
producing photocatalyst, respectively. Fe2+ /Fe3+ redox couple was used as a redox mediator in the
26 was
of 35
In the simulated dual-bed at first the hydrogen generation reaction
26 of 35
performed, then the reaction mixture was filtered, and finally, the filtrate solution was used for oxygen
oxygen production (Figure 27). They obtained stoichiometric hydrogen and oxygen generation in the
production
(Figure 27).
They27).
obtained
stoichiometric
hydrogenhydrogen
and oxygen
in the dual-bed
oxygen
production
(Figure
They obtained
stoichiometric
andgeneration
oxygen generation
in the
dual-bed photoreactor under visible light.
photoreactor
under
visible
light.
dual-bed photoreactor under visible light.
Inorganics
2017, 5, 34reaction.
water splitting
Inorganics
2017, 5, 34
Figure 27. H2 and O2 generation in a simulated dual-bed photoreactor under visible light. Reproduced
Figure
27.
O
in aasimulated
simulateddual-bed
dual-bed photoreactor under visible light. Reproduced
Figure
27. H
H22and
andfrom
O22generation
generation
in
with
permission
[117]. Copyright
Elsevier, 2009. photoreactor under visible light. Reproduced with
with
permission
from
[117].
Copyright
Elsevier,
permission from [117]. Copyright Elsevier, 2009. 2009.
8.3. Self-Mixing Photoreactor for Large-Scale Application
8.3.
8.3. Self-Mixing
Self-Mixing Photoreactor
Photoreactor for
for Large-Scale
Large-Scale Application
Application
In laboratory scale photo-reactor, one can get uniform mixing with either magnetic stirrer or
In
laboratory
scale
photo-reactor,
one
uniform
with either
magnetic
stirrer or
In laboratory
photo-reactor,
one can
can get
getphotoreactor,
uniform mixing
mixing
either
magnetic
or
mechanical
mixer.scale
However,
in an industrial
suchwith
active
stirring
willstirrer
not be
mechanical
mixer.
However,
in
an
industrial
photoreactor,
such
active
stirring
will
not
be
mechanical mixer.
However,
in an industrial
photoreactor,
active
stirring will
not be photoreactor
economically
economically
feasible.
To overcome
this issue,
Huang etsuch
al. [118]
developed
a slurry
economically
feasible. To
overcome
this issue,
Huang
et al. [118] adeveloped
a slurry photoreactor
feasible.
To
overcome
this
issue,
Huang
et
al.
[118]
developed
slurry
photoreactor
with facile
with facile self-mixing of the flow field inside the photoreactor. The photoreactor was designed
in
with
facile
self-mixing
of
the
flow
field
inside
the
photoreactor.
The
photoreactor
was
designed
in
self-mixing
of the
flow field
inside the
photoreactor.
photoreactor
was designed
in such a particles
way that
such
a way that
enough
turbulence
generated
insideThe
photoreactor
to keep
the photocatalyst
such
a
way
that
enough
turbulence
generated
inside
photoreactor
to
keep
the
photocatalyst
particles
enough
turbulence
inside photoreactor
to keep the photocatalyst
particles
in suspended
in
suspended
form.generated
The photoreactor
was a three-neck-flask
(Figure 28) with
shallow
depth that
in
suspended
form. Thewas
photoreactor
was a three-neck-flask
(Figure
28)
withthat
shallow
depth
that
form.
The
photoreactor
a
three-neck-flask
(Figure
28)
with
shallow
depth
provided
proper
provided proper mixing and circulation of the solution preventing photocatalyst settling.
The
provided
proper
mixing
andsolution
circulation
of the photocatalyst
solution preventing
photocatalyst
settling.
The
mixing
and
circulation
of
the
preventing
settling.
The
photoreactor
floor
was
photoreactor floor was tapered (tapered angle 10°–15°) providing swirl and uniform distribution of
photoreactor
floor
was
tapered
(tapered
angle
10°–15°)
providing
swirl
and
uniform
distribution
◦
◦
tapered (tapered
angle 10 –15 ) providing swirl and uniform distribution of photocatalyst slurry. of
photocatalyst
slurry.
photocatalyst slurry.
Figure 28. Flow diagram of a three-neck-flask photoreactor with passive mixing. Reproduced with
Figure 28. Flow diagram of a three-neck-flask photoreactor with passive mixing. Reproduced with
permission from [118].
Copyright Elsevier, 2011.
permission from [118]. Copyright Elsevier, 2011.
8.4. Solar Photocatalytic Reactor for Water Splitting
8.4. Solar
Solar Photocatalytic Reactor for Water Splitting
In the area of solar water splitting/hydrogen generation, one of the key issues is the efficient
In the area of solar
solar water
water splitting/hydrogen
splitting/hydrogen generation,
generation, one
one of
of the
the key
key issues
issues is
is the efficient
utilization of solar light. As solar light is illuminated over the earth surface in an irregular manner
utilization
utilization of
of solar light. As
As solar
solar light
light is
is illuminated
illuminated over
over the
the earth
earth surface
surface in an irregular manner
throughout the year, solar energy needs to be captured and converted to a storable form of energy
throughout the year, solar energy needs to be captured and converted to a storable form of energy
(e.g. hydrogen energy). However, the efficient collection of solar light is still in a developing stage,
(e.g. hydrogen energy). However, the efficient collection of solar light is still in a developing stage,
and it requires the use of solar concentration to overcome the problem. Till date, researchers have
and it requires the use of solar concentration to overcome the problem. Till date, researchers have
developed different solar photoreactors which are mainly non-concentrating solar reactors, namely
developed different solar photoreactors which are mainly non-concentrating solar reactors, namely
double skin sheet reactor (DSSR), the thin-film-fixed-bed reactor (TFFBR). Compound Parabolic
double skin sheet reactor (DSSR), the thin-film-fixed-bed reactor (TFFBR). Compound Parabolic
Inorganics 2017, 5, 34
27 of 35
throughout the year, solar energy needs to be captured and converted to a storable form of energy (e.g.,
hydrogen energy). However, the efficient collection of solar light is still in a developing stage, and it
requires the use of solar concentration to overcome the problem. Till date, researchers have developed
Inorganics
5,
27
different
solar
photoreactors which are mainly non-concentrating solar reactors, namely double
Inorganics 2017,
2017,
5, 34
34
27 of
of 35
35
skin sheet reactor (DSSR), the thin-film-fixed-bed reactor (TFFBR). Compound Parabolic Collectors
Collectors
(CPC)
collector
where
they
retain
of
(CPC)
is a simple
concentrating
collector where
they retain
properties
of properties
flat
plate collectors
too.
Collectors
(CPC) is
is aa simple
simple concentrating
concentrating
collector
wherethe
they
retain the
the
properties
of flat
flat plate
plate
collectors
too.
Jing
et
al.
[119]
used
a
compound
parabolic
concentrator
(CPC)
based
photocatalytic
Jing
et al. [119]
usedeta al.
compound
parabolic
concentrator
(CPC)
based photocatalytic
reactor for
collectors
too. Jing
[119] used
a compound
parabolic
concentrator
(CPC) basedsolar
photocatalytic
solar
for
generation.
Figure
29
photoreactor
configuration
based
on
hydrogen
generation.
Figure
29 describes
the photoreactor
configuration
based
on CPCs. The
solar
solar reactor
reactor
for hydrogen
hydrogen
generation.
Figure
29 describes
describes the
the
photoreactor
configuration
based
on
◦
CPCs.
The
solar
collector
had
four
CPC
modules
in
series
which
were
located
on
a
support
inclined
collector
had
four
CPC modules
series
which in
were
located
a support
at 35 inclined
on the
CPCs. The
solar
collector
had fourinCPC
modules
series
whichon
were
locatedinclined
on a support
at
horizontal
plane.
The
hydrogen
production
reactor
(SPHR)
set
horizontal
plane.
The solar
photocatalytic
hydrogen production
(SPHR)
set up
consists
a
at 35°
35° on
on the
the
horizontal
plane.
The solar
solar photocatalytic
photocatalytic
hydrogenreactor
production
reactor
(SPHR)
setofup
up
consists
of
a
continuously
stirred
tank,
a
recirculation
pump,
and
solar
collector.
They
combined
aa
continuously
stirred tank, astirred
recirculation
and solar
collector.
They
combined
a tubular
glass
consists of a continuously
tank, a pump,
recirculation
pump,
and solar
collector.
They
combined
tubular
glass
reactor
with
CPC
module
to
construct
the
solar
photoreactor
for
hydrogen
generation
reactor
module
construct
thetosolar
photoreactor
hydrogen generation
and generation
achieved a
tubularwith
glassCPC
reactor
withto
CPC
module
construct
the solarfor
photoreactor
for hydrogen
−1.
−
1
and
achieved
a
hydrogen
production
rate
of
1.88
L·h
−1
hydrogen
production
rate
of
1.88
L
·
h
.
and achieved a hydrogen production rate of 1.88 L·h .
Figure
29.
Solar
photocatalytic
hydrogen
production
reactor
set
up.
(1)
(2)
Figure 29.
29.Solar
Solarphotocatalytic
photocatalytic
hydrogen
production
reactor
setDescription:
up. Description:
Description:
(1) support;
support;
(2)
Figure
hydrogen
production
reactor
set up.
(1) support;
(2) storage
storage
tank;
(3)
gas
chromatography
(GC);
(4)
gas
flow
meter;
(5)
gas
collection
apparatus;
(6)
feeding
storage
tank;
(3)
gas
chromatography
(GC);
(4)
gas
flow
meter;
(5)
gas
collection
apparatus;
(6)
feeding
tank; (3) gas chromatography (GC); (4) gas flow meter; (5) gas collection apparatus; (6) feeding pump;
pump;
(7)
pump;
(8)
(9)
sensor;
(10)
pressure
sensor; and
pump;
(7) circulating
circulating
pump;
(8) nitrogen;
nitrogen;
(9) temperature
temperature
sensor;
(10)sensor;
pressure
and (11)
(11)
(7)
circulating
pump; (8)
nitrogen;
(9) temperature
sensor; (10)
pressure
andsensor;
(11) connecting
connecting
tube.
Reproduced
with
permission
from
[119].
Copyright
Elsevier,
2009.
connecting
tube. Reproduced
with permission
from [119].Elsevier,
Copyright
Elsevier, 2009.
tube.
Reproduced
with permission
from [119]. Copyright
2009.
In
In another
another study,
study, Jing
Jing et
et al.
al. [120]
[120] again
again developed
developed three
three photoreactors
photoreactors at
at different
different capacities,
capacities, aa
Inclose
another
study, Jing
et al. [120]a again
developed
three photoreactors
at different
capacities,
small
circulation
photoreactor,
medium
scale
photoreactor
with
simulated
solar
small close circulation photoreactor, a medium scale photoreactor with simulated solar light
light and
and
a
small
close
circulation
photoreactor,
a
medium
scale
photoreactor
with
simulated
solar
light
and
finally
scaled
up
to
a.
CPC
based
direct
solar
reactor
for
photocatalytic
hydrogen
generation
(Figure
finally scaled up to a. CPC based direct solar reactor for photocatalytic hydrogen generation (Figure
finally scaled up to a. CPC based direct solar reactor for photocatalytic hydrogen generation (Figure 30).
30).
30).
Figure
based direct
solar photoreactor
for hydrogen
poduction. Reproduced
with permission
Figure 30.
30. CPC
CPC
Figure
30.
CPC based
based direct
direct solar
solar photoreactor
photoreactor for
for hydrogen
hydrogen poduction.
poduction. Reproduced
Reproduced with
with permission
permission
from
[120].
Copyright
Elsevier,
2010.
from [120].
[120]. Copyright
Copyright Elsevier,
Elsevier, 2010.
2010.
from
Priya
Priya and
and Kanmani
Kanmani [121]
[121] studied
studied photocatalytic
photocatalytic hydrogen
hydrogen generation
generation from
from sulfide
sulfide wastewater
wastewater
Priya
andsunlight
Kanmani
[121]
studied
photocatalytic
hydrogen
generation
from sulfide
wastewater
using
natural
on
sunny
days
between
10:30
a.m.
and
2:00
p.m.
(during
using natural sunlight on sunny days between 10:30 a.m. and 2:00 p.m. (during April
April and
and May)
May) at
at
using
natural
sunlight
on
sunny
days
between
10:30
a.m.
and
2:00
p.m.
(during
April
and
May)
Centre
for
Environmental
Studies,
Anna
University
(India).
The
intensity
of
sunlight
showed
the
Centre for Environmental Studies, Anna University (India). The intensity of sunlight showed the
highest
highest value
value during
during April
April and
and May
May months,
months, and
and the
the average
average global
global solar
solar radiant
radiant was
was about
about 5.37
5.37
−2. They used a method described by Ray [109], to determine the photoreactor volume for the
kWh·m
kWh·m−2. They used a method described by Ray [109], to determine the photoreactor volume for the
3
treatment
treatment of
of 11 m
m3 of
of sulfide
sulfide wastewater.
wastewater. The
The photoreactor
photoreactor volume
volume was
was expressed
expressed as
as follows:
follows:
=
=
(15)
(15)
Inorganics 2017, 5, 34
28 of 35
at Centre for Environmental Studies, Anna University (India). The intensity of sunlight showed
the highest value during April and May months, and the average global solar radiant was about
5.37 kWh·m−2 . They used a method described by Ray [109], to determine the photoreactor volume for
the treatment of 1 m3 of sulfide wastewater. The photoreactor volume was expressed as follows:
Inorganics 2017, 5, 34
VR =
QCin X
κR
(15)
28 of 35
3
3 −1 C is the inlet
where, VRR is
is the photoreactor volume (m
(m3), Q is
is the
the volumetric
volumetric flow
flowrate
rate(m
(m·3s·s−1),), C
in
in is the
−3 ), X is the desired fractional conversion, R is the average
−3),
concentration of
ofthe
thepollutant
pollutant(mol·m
(mol·m
concentration
X is the desired fractional conversion, R is the average mass
−2 ·s−1 ) and κ is the illuminated photocatalyst surface area per
−2·s−1)·m
massdistribution
flow distribution
rate (mol
flow
rate (mol·m
and κ
is the illuminated photocatalyst surface area per unit volume
2 ·m−3 ). Priya and Kanmani [121] calculated the κ value from Equation (1) for their
2
−3
unit
volume
(m
(m ·m ). Priya and Kanmani [121] calculated the κ value from Equation (1) for their photoreactor as
2 ·m−3 . The also created the same using physical characteristics of photocatalyst
photoreactor
as 1000
1000
m2·m−3. The
alsomcreated
the same using physical characteristics of photocatalyst and found κ =
6 m2 ·m−3 . Since the κ value calculated from Equation (1) is much lower than
6
2
−3
and
found
κ
=
8.96
×
10
8.96 × 10 m ·m . Since the κ value calculated from Equation (1) is much lower than the κ value
the κ valuefrom
calculated
from
physical properties,
the photoreactor
scale up
bewith
possible
large
calculated
physical
properties,
the photoreactor
scale up would
be would
possible
largewith
volume.
3 of sulfide wastewater
volume.
alsothe
found
the optimum
photoreactor
55treat
m to1treat
1m
They
alsoThey
found
optimum
photoreactor
length length
of 55 mofto
m3 of
sulfide
wastewater and
and simultaneous
hydrogen
generation.
simultaneous
hydrogen
generation.
photocatalytic
reactor
thatthat
utilized
bothboth
sunlight
and
Baniasadi et
et al.
al.[122]
[122]dealt
dealtwith
witha ahybridized
hybridized
photocatalytic
reactor
utilized
sunlight
UV-visible
lampslamps
along along
with electrodes
to consume
photo-generated
hole andhole
to produce
and
UV-visible
with electrodes
to consume
photo-generated
and tohydrogen
produce
from waterfrom
(Figure
31).(Figure 31).
hydrogen
water
Figure
scale
hybrid
photoreactor
for for
continuous
operation,
artificial
illumination,
and
Figure 31.
31. Pilot
Pilotplant
plant
scale
hybrid
photoreactor
continuous
operation,
artificial
illumination,
natural
sunlight.
Reproduced
with
permission
from
[122].
Copyright
Elsevier,
2012
and natural sunlight. Reproduced with permission from [122]. Copyright Elsevier, 2012
They used solar panels to provide electrons through the electrode-solution interface and to
They used solar panels to provide electrons through the electrode-solution interface and to
generate electricity. The intermittent behavior of sunlight was controlled by the use of UV-visible
generate electricity. The intermittent behavior of sunlight was controlled by the use of UV-visible
lamps inside the concentrator to perform photocatalytic reaction during cloudy days and night time.
lamps inside the concentrator to perform photocatalytic reaction during cloudy days and night time.
9.
9. Conclusions
Conclusions and
and Future
Future Research
Research Directions
Directions
From
energy change,
change, photocatalytic
photocatalytic water
From the
the viewpoint
viewpoint of
of the
the Gibbs
Gibbs free
free energy
water splitting
splitting is
is an
an uphill
uphill
reaction,
similar
to
photosynthesis
in
green
plants.
Therefore,
to
undergo
photocatalytic
water
reaction, similar to photosynthesis in green plants. Therefore, to undergo photocatalytic water splitting
splitting
reaction,
the
energy
of
photoelectrons
must
be
above
the
energy
level
of
the
(H
2O/H2) redox
reaction, the energy of photoelectrons must be above the energy level of the (H2 O/H2 ) redox system
system
the of
energy
photo
holes be
should
bethe
below
the level
energy
the O)
(O2redox
/H2O)system.
redox system.
and theand
energy
photoofholes
should
below
energy
of level
the (Oof2 /H
Again,
2
Again,
for
visible-light-driven
water
splitting
reaction,
the
photocatalytic
materials
should
for visible-light-driven water splitting reaction, the photocatalytic materials should havehave
bandband
gap
gap
energy
(E
g) between 1.23 and 3 eV (1.23 eV < Eg < 3.0 eV). Among various visible-light-driven
energy (Eg ) between 1.23 and 3 eV (1.23 eV < Eg < 3.0 eV). Among various visible-light-driven methods,
methods,
dye-sensitization
known
to be atechnique
promisingtotechnique
develop
a highly
active
dye-sensitization
is known toisbe
a promising
develop a to
highly
visible
active visible
photocatalyst
photocatalyst
for
the
efficient
photocatalytic
overall
water
splitting
and
sacrificial
hydrogen
for the efficient photocatalytic overall water splitting and sacrificial hydrogen generation. The dye
generation.
Theadye
plays
a dual role in process
the dye-sensitization
processlight,
by absorbing
visible
molecule plays
dualmolecule
role in the
dye-sensitization
by absorbing visible
and by bridging
light, and by bridging the electron donor with semiconductor photocatalyst. Again, energy level
compatibility is necessary between the semiconductor conduction band and the excited dye molecule,
for efficient electron transfer from dye to the semiconductor. Dye-sensitized photocatalyst and photoelectrochemical cell show enormous promise for water splitting under visible light. The
photocatalytic system typically uses two different photocatalysts for hydrogen and oxygen
Inorganics 2017, 5, 34
29 of 35
the electron donor with semiconductor photocatalyst. Again, energy level compatibility is necessary
between the semiconductor conduction band and the excited dye molecule, for efficient electron
transfer from dye to the semiconductor. Dye-sensitized photocatalyst and photo-electrochemical cell
show enormous promise for water splitting under visible light. The photocatalytic system typically
uses two different photocatalysts for hydrogen and oxygen generation. Researchers also recommended
a possible application of a two-absorber Z-scheme to capture the visible and near-infrared parts of the
solar spectrum effectively.
In the case of sacrificial hydrogen generation, several parameters are involved such as dyes,
sacrificial agents, photocatalysts, and co-catalysts. In this case, the band gap of a semiconductor is
less important, but the conduction of semiconductor should be lower than the lowest unoccupied
molecular orbital of the dye and higher than the energy level of the (H2 O/H2 ) redox system.
Ruthenium-based dye-sensitized photocatalyst can provide a very high quantum yield (approximately
28%). Xanthene-based dyes such as eosin Y, rhodamine B, erythrosine, fluorescein, rose bengal,
and uranine can be good alternatives for Ru-complexes as they can similarly absorb visible light.
The main purpose of using metal co-catalyst is utilizing them as an electron trap center to accelerate
the separation of photo-generated charges. Among the different metal co-catalysts, platinum
has a comparatively low Fermi level, and it has been effective in inhibiting photo-generated
charge recombination.
Sacrificial electron donors also play a significant role in dye-sensitized hydrogen generation
reaction by influencing the solution pH, dye adsorption, dye regeneration and the charge transfer
from dye to the semiconductor. However, further studies are required to explain the effect of a
sacrificial electron donor in dye-sensitized hydrogen generation system. In sacrificial hydrogen
generation, only a few researchers have developed methodologies based on the statistical design of
experiments. They achieved optimal conditions by a powerful statistical method of optimization
titled central composite design (CCD). Pareto analysis, regression model, and response surface plot
could be used as effective tools in future studies on dye-sensitized overall water splitting process
as well. Also, as mentioned before, the interaction effect between different factors is important and
need to be investigated to have a comprehensive understanding of the photocatalytic systems. As of
our knowledge, there is only one research group that reported statistical analysis for dye-sensitized
hydrogen generation. Therefore it is important to consider this technique in future studies.
In the case of photocatalytic overall water splitting, gas separation is the additional design criteria
that are addressed along with other design parameters. Photocatalytic membrane reactor utilizes a
Nafion membrane to suppress the backward reaction during water splitting. Continuous hydrogen
generation can be achieved with a dual bed photoreactor with a hole scavenger regeneration system.
In large scale application, CPC based direct solar photoreactor and hybrid (UV + Solar) photoreactor
draw significant attention.
Despite the extensive studies on dye-sensitized photocatalytic hydrogen generation, there are
still challenges and gaps in the literature that are required to be addressed in future research.
Developing new photocatalysts and co-catalysts with high photocatalytic activity in visible light
irradiation as well as high stability is still required. Moreover, applying design of experiment and
developing large scale photoreactor for dye-sensitized water splitting/hydrogen generation would
enrich this research area.
Acknowledgments: Funding for the research was provided by Natural Sciences and Engineering Research
Council of Canada (NSERC).
Author Contributions: Pankaj Chowdhury and Ghodsieh Malekshoar wrote this review article while Ajay K. Ray
provided valuable inputs, feedbacks, and comments.
Conflicts of Interest: The authors declare no conflict of interest.
Inorganics 2017, 5, 34
30 of 35
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Dunn, S. Hydrogen futures: Toward a sustainable energy system. Int. J. Hydrog. Energy 2002, 27, 235–264.
[CrossRef]
Chowdhury, P. Solar and Visible Light Driven Photocatalysis for Sacrificial Hydrogen Generation and Water
Detoxification with Chemically Modified TiO2 ; Western University: London, ON, Canada, 2012.
Grimes, C.A.; Varghese, O.K.; Ranjan, S. Hydrogen Generation by Water Splitting; Springer: New York, NY,
USA, 2008.
The Kelling Curve. Available online: https://scripps.ucsd.edu/programs/keelingcurve/ (accessed on 12
March 2017).
Chowdhury, P.; Gomaa, H.; Ray, A.K. Factorial design analysis for dye-sensitized hydrogen generation from
water. Int. J. Hydrog. Energy 2011, 36, 13442–13451. [CrossRef]
Kudo, A.; Miseki, Y. Heterogeneous photocatalyst materials for water splitting. Chem. Soc. Rev. 2009, 38,
253–278. [CrossRef] [PubMed]
Domen, K.; Kudo, A.; Onishi, T. Mechanism of photocatalytic decomposition of water into H2 and O2 over
NiO–SrTiO3 . J. Catal. 1986, 102, 92–98. [CrossRef]
Polo, A.S.; Itokazu, M.K.; Iha, N.Y.M. Metal complex sensitizers in dye-sensitized solar cells. Coord. Chem.
Rev. 2004, 248, 1343–1361. [CrossRef]
Chowdhury, P.; Gomaa, H.; Ray, A.K. Dye-sensitized photocatalyst: A breakthrough in green energy and
environmental detoxification. In Sustainable Nanotechnology and the Environment: Advances and Achievements;
Shamim, N., Sharma, V.K., Eds.; American Chemical Society: Washington, DC, USA, 2013; Volume 1124,
pp. 231–266.
Pei, D.; Luan, J. Development of visible light-responsive sensitized photocatalysts. Int. J. Photoenergy 2012,
2012, 262831. [CrossRef]
Chowdhury, P.; Gomaa, H.; Ray, A.K. Sacrificial hydrogen generation from aqueous triethanolamine with
Eosin Y-sensitized Pt/TiO2 photocatalyst in UV, visible and solar light irradiation. Chemosphere 2015, 121,
54–61. [CrossRef] [PubMed]
Salvador, P. Thermodynamic and kinetic considerations about water splitting and competitive reactions in a
photoelectrochemical cell. New J. Chem. 1988, 12, 35–43.
Chen, Y.-S.; Kamat, P.V. Glutathione-capped gold nanoclusters as photosensitizers. Visible light-induced
hydrogen generation in neutral water. J. Am. Chem. Soc. 2014, 136, 6075–6082.
Lee, J.S. Photocatalytic water splitting under visible light with particulate semiconductor catalysts. Catal. Surv.
Asia 2005, 9, 217–227. [CrossRef]
Fujishima, A.; Honda, K. TiO2 photoelectrochemistry and photocatalysis. Nature 1972, 238, 37–38. [CrossRef]
[PubMed]
Kudo, A.; Kato, H. Photocatalytic decomposition of water into H2 and O2 over novel photocatalyst
K3 Ta3 Si2 O13 with pillared structure consisting of three TaO6 chains. Chem. Lett. 1997, 26, 867–868. [CrossRef]
Ni, M.; Leung, M.K.H.; Leung, D.Y.C.; Sumathy, K. A review and recent developments in photocatalytic
water-splitting using TiO2 for hydrogen production. Renew. Sustain. Energy Rev. 2007, 11, 401–425. [CrossRef]
Sakthivel, S.; Shankar, M.V.; Palanichamy, M.; Arabindoo, B.; Bahnemann, D.W.; Murugesan, V.
Enhancement of photocatalytic activity by metal deposition: Characterization and photonic efficiency
of Pt, Au and Pd deposited on TiO2 catalyst. Water Res. 2004, 38, 3001–3008. [CrossRef] [PubMed]
Subramanian, V.; Wolf, E.E.; Kamat, P.V. Catalysis with TiO2 /gold nanocomposites. Effect of metal particle
size on the Fermi level equilibration. J. Am. Chem. Soc. 2004, 126, 4943–4950. [CrossRef] [PubMed]
Subramanian, V.; Wolf, E.E.; Kamat, P.V. Green emission to probe photoinduced charging events in ZnO–Au
nanoparticles. Charge distribution and Fermi-level equilibration. J. Phys. Chem. B 2003, 107, 7479–7485.
[CrossRef]
Jakob, M.; Levanon, H.; Kamat, P.V. Charge distribution between UV-irradiated TiO2 and gold nanoparticles:
Determination of shift in the Fermi level. Nano Lett. 2003, 3, 353–358. [CrossRef]
Choi, W.; Termin, A.; Hoffmann, M.R. The role of metal ion dopants in quantum-sized TiO2 : Correlation
between photoreactivity and charge carrier recombination dynamics. J. Phys. Chem. 1994, 98, 13669–13679.
[CrossRef]
Inorganics 2017, 5, 34
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
31 of 35
Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visible-light photocatalysis in nitrogen-doped titanium
oxides. Science 2001, 293, 269–271. [CrossRef] [PubMed]
Therefore, W.-W.; Kim, K.-J.; Moon, S.-J. Photo-production of hydrogen over the CdS–TiO2 nano-composite
particulate films treated with TiCl4 . Int. J. Hydrog. Energy 2004, 29, 229–234.
Gurunathan, K.; Maruthamuthu, P.; Sastri, M.V.C. Photocatalytic hydrogen production by dye-sensitized
Pt/SnO2 and Pt/SnO2 /RuO2 in aqueous methyl viologen solution. Int. J. Hydrog. Energy 1997, 22, 57–62.
[CrossRef]
Kudo, A.; Ueda, K.; Kato, H.; Mikami, I. Photocatalytic O2 evolution under visible light irradiation on BiVO4
in aqueous AgNO3 solution. Catal. Lett. 1998, 53, 229–230. [CrossRef]
Kato, H.; Kobayashi, H.; Kudo, A. Role of Ag+ in the band structures and photocatalytic properties of
AgMO3 (M: Ta and Nb) with the perovskite structure. J. Phys. Chem. B 2002, 106, 12441–12447. [CrossRef]
Hosogi, Y.; Tanabe, K.; Kato, H.; Kobayashi, H.; Kudo, A. Energy structure and photocatalytic activity of
niobates and tantalates containing Sn(II) with a 5s2 electron configuration. Chem. Lett. 2003, 33, 28–29.
[CrossRef]
Nazeeruddin, M.K.; Gratzel, M. Encyclopedia of Electrochemistry. In Semiconductor Electrodes and
Photoelectronchemistry; Licht, S., Ed.; Wiley-VCH: Weinheim, Germany, 2002; Volume 6, pp. 407–431.
Hagfeldt, A.; Gratzel, M. Molecular Photovoltaics. Acc. Chem. Res. 2000, 33, 269–277. [CrossRef] [PubMed]
Bourdon, J. Spectral Sensitization of Chemical Effects in Solids. J. Phys. Chem. 1965, 69, 705–713. [CrossRef]
Oregan, B.; Gratzel, M. A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2 films.
Nature 1991, 353, 737–740. [CrossRef]
Youngblood, W.J.; Lee, S.-H.A.; Maeda, K.; Mallouk, T.E. Visible light water splitting using dye-sensitized
oxide semiconductors. Acc. Chem. Res. 2009, 42, 1966–1973. [CrossRef] [PubMed]
Tsubomura, H.; Matsumura, M.; Nomura, Y.; Amamiya, T. Dye sensitised zinc oxide: Aqueous electrolyte:
Platinum photocell. Nature 1976, 261, 402–403. [CrossRef]
Galoppini, E. Linkers for anchoring sensitizers to semiconductor nanoparticles. Coord. Chem. Rev. 2004, 248,
1283–1297. [CrossRef]
Gratzel, M. Dye-sensitized solar cells. J. Photochem. Photobiol. C Photochem. Rev. 2003, 4, 145–153. [CrossRef]
Abe, R.; Shinmei, K.; Hara, K.; Ohtani, B. Robust dye-sensitized overall water splitting system with two-step
photoexcitation of coumarin dyes and metal oxide semiconductors. Chem. Commun. 2009, 24, 3577–3579.
[CrossRef] [PubMed]
Abe, R.; Shinmei, K.; Koumura, N.; Hara, K.; Ohtani, B. Visible-light-induced water splitting based on
two-step photoexcitation between dye-sensitized layered niobate and tungsten oxide photocatalysts in the
presence of a triiodide/iodide shuttle redox mediator. J. Am. Chem. Soc. 2013, 135, 16872–16884. [CrossRef]
[PubMed]
Le, T.T.; Akhtar, M.S.; Park, D.M.; Lee, J.C.; Yang, O.B. Water splitting on Rhodamine-B dye sensitized
Co-doped TiO2 catalyst under visible light. Appl. Catal. B 2012, 111–112, 397–401. [CrossRef]
Alibabaei, L.; Sherman, B.D.; Norris, M.R.; Brennaman, M.K.; Meyer, T.J. Visible photoelectrochemical water
splitting into H2 and O2 in a dye-sensitized photoelectrosynthesis cell. Proc. Natl. Acad. Sci. USA 2015, 112,
5899–5902. [CrossRef] [PubMed]
Swierk, J.R.; Mallouk, T.E. Design and development of photoanodes for water-splitting dye-sensitized
photoelectrochemical cells. Chem. Soc. Rev. 2013, 42, 2357–2387. [CrossRef] [PubMed]
Zhao, Y.; Swierk, J.R.; Megiatto, J.D.; Sherman, B.; Youngblood, W.J.; Qin, D.; Lentz, D.M.; Moore, A.L.;
Moore, T.A.; Gust, D. Improving the efficiency of water splitting in dye-sensitized solar cells by using a
biomimetic electron transfer mediator. Proc. Natl. Acad. Sci. USA 2012, 109, 15612–15616. [CrossRef]
[PubMed]
Hagiwara, H.; Nagatomo, M.; Ida, S.; Ishihara, T. Photocatalytic Splitting of Water into Hydrogen and
Oxygen on Organic Dye Modified KTa(Zr)O3 Catalyst. Energy Procedia 2012, 22, 53–60. [CrossRef]
Park, J.H.; Bard, A.J. Unassisted Water Splitting from Bipolar Pt/Dye-Sensitized TiO2 Photoelectrode Arrays.
Electrochem. Solid State Lett. 2005, 8, G371–G375. [CrossRef]
Li, L.; Duan, L.; Xu, Y.; Gorlov, M.; Hagfeldt, A.; Sun, L. A photoelectrochemical device for visible light
driven water splitting by a molecular ruthenium catalyst assembled on dye-sensitized nanostructured TiO2 .
Chem. Commun. 2010, 46, 7307–7309. [CrossRef] [PubMed]
Inorganics 2017, 5, 34
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
32 of 35
Li, F.; Fan, K.; Xu, B.; Gabrielsson, E.; Daniel, Q.; Li, L.; Sun, L. Organic dye-sensitized tandem
photoelectrochemical cell for light driven total water splitting. J. Am. Chem. Soc. 2015, 137, 9153–9159.
[CrossRef] [PubMed]
Graetzel, M. Artificial photosynthesis: Water cleavage into hydrogen and oxygen by visible light. Acc. Chem.
Res. 1981, 14, 376–384. [CrossRef]
Houlding, V.H.; Gratzel, M. Photochemical H2 generation by visible light: Sensitization of TiO2 particles by
surface complexation with 8-hydroxyquinoline. J. Am. Chem. Soc. 1983, 105, 5695–5696. [CrossRef]
Furlong, D.N.; Wells, D.; Sasse, W.H.F. Colloidal semiconductors in systems for the sacrificial photolysis
of water: Sensitization of TiO2 by adsorption of ruthenium complexes. J. Phys. Chem. 1986, 90, 1107–1115.
[CrossRef]
Kim, Y.I.; Keller, S.W.; Krueger, J.S.; Yonemoto, E.H.; Saupe, G.B.; Mallouk, T.E. Photochemical charge transfer
and hydrogen evolution mediated by oxide semiconductor particles in zeolite-based molecular assemblies.
J. Phys. Chem. B 1997, 101, 2491–2500. [CrossRef]
Lakshminarasimhan, N.; Bae, E.; Choi, W. Enhanced photocatalytic production of H2 on mesoporous
TiO2 prepared by template-free method: Role of interparticle charge transfer. J. Phys. Chem. C 2007, 111,
15244–15250. [CrossRef]
Chen, Y.; Mou, Z.; Yin, S.; Huang, H.; Yang, P.; Wang, X.; Du, Y. Graphene enhanced photocatalytic hydrogen
evolution performance of dye-sensitized TiO2 under visible light irradiation. Mater. Lett. 2013, 107, 31–34.
[CrossRef]
Vinodgopal, K.; Hua, X.; Dahlgren, R.L.; Lappin, A.G.; Patterson, L.K.; Kamat, P.V. Photochemistry of
Ru(bpy)2 (dcbpy)2+ on Al2 O3 and TiO2 surfaces. An insight into the mechanism of photosensitization.
J. Phys. Chem. 1995, 99, 10883–10889. [CrossRef]
Lakadamyali, F.; Reisner, E. Photocatalytic H2 evolution from neutral water with a molecular cobalt catalyst
on a dye-sensitised TiO2 nanoparticle. Chem. Commun. 2011, 47, 1695–1697. [CrossRef] [PubMed]
Lakadamyali, F.; Reynal, A.; Kato, M.; Durrant, J.R.; Reisner, E. Electron transfer in dye-sensitized
semiconductors modified with molecular cobalt catalysts: Photoreduction of aqueous protons. Chem. Eur. J.
2012, 18, 15464–15475. [CrossRef] [PubMed]
Bae, E.; Choi, W.; Park, J.; Shin, H.S.; Kim, S.B.; Lee, J.S. Effects of surface anchoring groups (carboxylate
vs phosphonate) in ruthenium-complex-sensitized TiO2 on visible light reactivity in aqueous suspensions.
J. Phys. Chem. B 2004, 108, 14093–14101. [CrossRef]
Bae, E.; Choi, W. Effect of the anchoring group (carboxylate vs phosphonate) in Ru-complex-sensitized TiO2
on hydrogen production under visible light. J. Phys. Chem. B 2006, 110, 14792. [CrossRef] [PubMed]
Nada, A.A.; Hamed, H.A.; Barakat, M.H.; Mohamed, N.R.; Veziroglu, T.N. Enhancement of photocatalytic
hydrogen production rate using photosensitized TiO2 /RuO2 –MV2+ . Int. J. Hydrog. Energy 2008, 33,
3264–3269. [CrossRef]
Maitani, M.M.; Zhan, C.; Mochizuki, D.; Suzuki, E.; Wada, Y. Influence of co-existing species on charge
transfer in dye-sensitized nanocrystalline oxide semiconductors in aqueous suspension for H2 evolution
under visible light. Appl. Catal. B 2014, 147, 770–778. [CrossRef]
Hagiwara, H.; Ono, N.; Inoue, T.; Matsumoto, H.; Ishihara, T. Dye-Sensitizer Effects on a Pt/KTa(Zr)O3
Catalyst for the Photocatalytic Splitting of Water. Angew. Chem. Int. Ed. 2006, 45, 1420–1422. [CrossRef]
[PubMed]
Astuti, Y.; Palomares, E.; Haque, S.A.; Durrant, J.R. Triplet state photosensitization of nanocrystalline metal
oxide electrodes by zinc-substituted cytochrome c: Application to hydrogen evolution. J. Am. Chem. Soc
2005, 127, 15120–15126. [CrossRef] [PubMed]
Zhang, X.; Yu, L.; Zhuang, C.; Peng, T.; Li, R.; Li, X. Highly asymmetric phthalocyanine as a sensitizer
of graphitic carbon nitride for extremely efficient photocatalytic H2 production under near-infrared light.
ACS Catal. 2014, 4, 162–170. [CrossRef]
Zhang, X.; Peng, T.; Song, S. Recent advances in dye-sensitized semiconductor systems for photocatalytic
hydrogen production. J. Mater. Chem. A 2016, 4, 2365–2402. [CrossRef]
Kong, C.; Min, S.; Lu, G. Dye-sensitized NiSx catalyst decorated on graphene for highly efficient reduction of
water to hydrogen under visible light irradiation. ACS Catal. 2014, 4, 2763–2769. [CrossRef]
Inorganics 2017, 5, 34
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
33 of 35
Watanabe, M.; Hagiwara, H.; Iribe, A.; Ogata, Y.; Shiomi, K.; Staykov, A.; Ida, S.; Tanaka, K.; Ishihara, T.
Spacer effects in metal-free organic dyes for visible-light-driven dye-sensitized photocatalytic hydrogen
production. J. Mater. Chem. A 2014, 2, 12952–12961. [CrossRef]
Latorre-Sanchez, M.; Lavorato, C.; Puche, M.; Fornes, V.; Molinari, R.; Garcia, H. Visible-light photocatalytic
hydrogen generation by using dye-sensitized graphene oxide as a photocatalyst. Chem. Eur. J. 2012, 18,
16774–16783. [CrossRef] [PubMed]
Mou, Z.; Dong, Y.; Li, S.; Du, Y.; Wang, X.; Yang, P.; Wang, S. Eosin Y functionalized graphene for
photocatalytic hydrogen production from water. Int. J. Hydrog. Energy 2011, 36, 8885–8893. [CrossRef]
Li, Y.; Guo, M.; Peng, S.; Lu, G.; Li, S. Formation of multilayer-Eosin Y-sensitized TiO2 via Fe3+ coupling
for efficient visible-light photocatalytic hydrogen evolution. Int. J. Hydrog. Energy 2009, 34, 5629–5636.
[CrossRef]
Chen, S.; Li, Y.; Wang, C. Visible-light-driven photocatalytic H2 evolution from aqueous suspensions of
perylenediimide dye-sensitized Pt/TiO2 catalysts. RSC Adv. 2015, 5, 15880–15885. [CrossRef]
Zhang, W.; Li, Y.; Zeng, X.; Peng, S. Synergetic effect of metal nickel and graphene as a cocatalyst for
enhanced photocatalytic hydrogen evolution via dye sensitization. Sci. Rep. 2015, 5, 10589. [CrossRef]
[PubMed]
Moser, J.; Gratzel, M. Photosensitized electron injection in colloidal semiconductors. J. Am. Chem. Soc. 1984,
106, 6557–6564. [CrossRef]
Kamegawa, T.; Matsuura, S.; Seto, H.; Yamashita, H. A Visible-light-harvesting assembly with a
sulfocalixarene linker between Dyes and a Pt–TiO2 Photocatalyst. Angew. Chem. Int. Ed. 2013, 52, 916–919.
[CrossRef] [PubMed]
Maia, D.L.S.; Pepe, I.; da Silva, A.F.; Silva, L.A. Visible-light-driven photocatalytic hydrogen production over
dye-sensitized β-BiTaO4 . J. Photochem. Photobiol. A Chem. 2012, 243, 61–64. [CrossRef]
Min, S.; Lu, G. Dye-cosensitized graphene/Pt photocatalyst for high efficient visible light hydrogen evolution.
Int. J. Hydrog. Energy 2012, 37, 10564–10574. [CrossRef]
Zhang, X.; Peng, T.; Yu, L.; Li, R.; Li, Q.; Li, Z. Visible/near-infrared-light-induced H2 production over
g-C3 N4 co-sensitized by organic dye and zinc phthalocyanine derivative. ACS Catal. 2015, 5, 504–510.
[CrossRef]
Abe, R.; Sayama, K.; Sugihara, H. Effect of water/acetonitrile ratio on dye-sensitized photocatalytic H2
evolution under visible light irradiation. J. Sol. Energy. Eng. 2005, 127, 413–416. [CrossRef]
Xiao, B.; Wang, X.; Huang, H.; Zhu, M.; Yang, P.; Wang, Y.; Du, Y. Improved superiority by covalently binding
dye to graphene for hydrogen evolution from water under visible-light irradiation. J. Phys. Chem. C 2013,
117, 21303–21311. [CrossRef]
Peng, T.; Ke, D.; Cai, P.; Dai, K.; Ma, L.; Zan, L. Influence of different ruthenium(II) bipyridyl complex on the
photocatalytic H2 evolution over TiO2 nanoparticles with mesostructures. J. Power Sources 2008, 180, 498–505.
[CrossRef]
George, L.; Sappati, S.; Ghosh, P.; Devi, R.N. Surface Site Modulations by Conjugated Organic Molecules To
Enhance Visible Light Activity of ZnO Nanostructures in Photocatalytic Water Splitting. J. Phys. Chem. C
2015, 119, 3060–3067. [CrossRef]
He, K.; Li, M.; Guo, L. Preparation and photocatalytic activity of PANI–CdS composites for hydrogen
evolution. Int. J. Hydrog. Energy 2012, 37, 755–759. [CrossRef]
Yan, H.; Huang, Y. Polymer composites of carbon nitride and poly(3-hexylthiophene) to achieve enhanced
hydrogen production from water under visible light. Chem. Commun. 2011, 47, 4168–4170. [CrossRef]
[PubMed]
Hirano, K.; Suzuki, E.; Ishikawa, A.; Moroi, T.; Shiroishi, H.; Kaneko, M. Sensitization of TiO2 particles by
dyes to achieve H2 evolution by visible light. J. Photochem. Photobiol. A Chem. 2000, 136, 157–161. [CrossRef]
Zhang, X.; Peng, B.; Zhang, S.; Peng, T. Robust wide visible-light-responsive photoactivity for H2 production
over a polymer/polymer heterojunction photocatalyst: The significance of sacrificial reagent. ACS Sustain.
Chem. Eng. 2015, 3, 1501–1509. [CrossRef]
Li, Q.; Chen, L.; Lu, G. Visible-light-induced photocatalytic hydrogen generation on dye-sensitized
multiwalled carbon nanotube/Pt catalyst. J. Phys. Chem. C 2007, 111, 11494–11499. [CrossRef]
Min, S.; Lu, G. Dye-sensitized reduced graphene oxide photocatalysts for highly efficient visible-light-driven
water reduction. J. Phys. Chem. C 2011, 115, 13938–13945. [CrossRef]
Inorganics 2017, 5, 34
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
34 of 35
Min, S.; Lu, G. Promoted photoinduced charge separation and directional electron transfer over dispersible
xanthene dyes sensitized graphene sheets for efficient solar H2 evolution. Int. J. Hydrog. Energy 2013, 38,
2106–2116. [CrossRef]
Wang, D.; Huang, J.; Li, X.; Yang, P.; Du, Y.; Goh, C.M.; Lu, C. Photocatalytic H2 production under visible-light
irradiation based on covalent attachment of manganese phthalocyanine to graphene. J. Mater. Chem. A 2015,
3, 4195–4202. [CrossRef]
Malekshoar, G.; Pal, K.; He, Q.; Yu, A.; Ray, A.K. Enhanced solar photocatalytic degradation of phenol with
coupled graphene-based titanium dioxide and zinc oxide. Ind. Eng. Chem. Res. 2014, 53, 18824–18832.
[CrossRef]
Potje-Kamloth, K. Semiconductor junction gas sensors. Chem. Rev. 2008, 108, 367–399. [CrossRef] [PubMed]
Yuan, Y.-P.; Ruan, L.-W.; Barber, J.; Loo, S.C.J.; Xue, C. Hetero-nanostructured suspended photocatalysts for
solar-to-fuel conversion. Energy Environ. Sci. 2014, 7, 3934–3951. [CrossRef]
Yang, J.; Wang, D.; Han, H.; Li, C. Roles of cocatalysts in photocatalysis and photoelectrocatalysis. Acc. Chem.
Res. 2013, 46, 1900–1909. [CrossRef] [PubMed]
Abe, R.; Sayama, K.; Arakawa, H. Dye-sensitized photocatalysts for efficient hydrogen production from
aqueous I− solution under visible light irradiation. J. Photochem. Photobiol. A Chem. 2004, 166, 115–122.
[CrossRef]
Abe, R.; Sayama, K.; Arakawa, H. Efficient hydrogen evolution from aqueous mixture of I− and acetonitrile
using a merocyanine dye-sensitized Pt/TiO2 photocatalyst under visible light irradiation. Chem. Phys. Lett.
2002, 362, 441–444. [CrossRef]
Jin, Z.; Zhang, X.; Lu, G.; Li, S. Improved quantum yield for photocatalytic hydrogen generation under
visible light irradiation over eosin sensitized TiO2 —Investigation of different noble metal loading. J. Mol.
Catal. A Chem. 2006, 259, 275–280. [CrossRef]
Yan, Z.; Yu, X.; Zhang, Y.; Jia, H.; Sun, Z.; Du, P. Enhanced visible light-driven hydrogen production from
water by a noble-metal-free system containing organic dye-sensitized titanium dioxide loaded with nickel
hydroxide as the cocatalyst. Appl. Catal. B 2014, 160, 173–178. [CrossRef]
Jin, Z.; Zhang, X.; Li, Y.; Li, S.; Lu, G. 5.1% Apparent quantum efficiency for stable hydrogen generation over
eosin-sensitized CuO/TiO2 photocatalyst under visible light irradiation. Catal. Commun. 2007, 8, 1267–1273.
[CrossRef]
Li, X.; Zhang, J.; Kang, S.; Li, G.; Mu, J. Visible light photocatalytic activity of CuO/Cr2 O3 co-loaded
multiwalled carbon nanotubes sensitized with eosin Y for hydrogen evolution from water. Ceram. Int. 2014,
40, 10171–10176. [CrossRef]
Kong, C.; Min, S.; Lu, G. Dye-sensitized cobalt catalysts for high efficient visible light hydrogen evolution.
Int. J. Hydrog. Energy 2014, 39, 4836–4844. [CrossRef]
Yin, M.; Ma, S.; Wu, C.; Fan, Y. A noble-metal-free photocatalytic hydrogen production system based on
cobalt(III) complex and eosin Y-sensitized TiO2 . RSC Adv. 2015, 5, 1852–1858. [CrossRef]
Malekshoar, G.; Ray, A.K. In situ grown molybdenum sulfide on TiO2 for dye-sensitized solar photocatalytic
hydrogen generation. Chem. Eng. Sci. 2016, 152, 35–44. [CrossRef]
Min, S.; Lu, G. Sites for high efficient photocatalytic hydrogen evolution on a limited-layered MoS2 cocatalyst
confined on graphene sheets—The role of graphene. J. Phys. Chem. C 2012, 116, 25415–25424. [CrossRef]
Feilizadeh, M.; Mul, G.; Vossoughi, M. E. coli inactivation by visible light irradiation using a Fe–Cd/TiO2
photocatalyst: Statistical analysis and optimization of operating parameters. Appl. Catal. B 2015, 168, 441–447.
[CrossRef]
Chen, J.; Luo, H.; Shi, H.; Li, G.; An, T. Anatase TiO2 nanoparticles–carbon nanotubes composite:
Optimization synthesis and the relationship of photocatalytic degradation activity of acyclovir in water.
Appl. Catal. A 2014, 485, 188–195. [CrossRef]
Antony, J. Design of Experiments for Engineers and Scientists; Elsevier: New York, NY, USA, 2014.
Ray, A.K. Photocatalytic reactor configurations for water purification: Experimentation and modeling.
Adv. Chem. Eng. 2009, 36, 145–184.
Chen, D.; Ray, A.K. Photodegradation kinetics of 4-nitrophenol in TiO2 suspension. Water Res. 1998, 32,
3223–3234. [CrossRef]
Assink, J.W.; Koster, T.P.M.; Slaager, J.M. Fotokatalytische Oxydatie Voor Afvalwaterbehandeling; Internal Report
Ref. No. 93-137; Instituut voor Milieu en Energie-technologie (TNO): Apeldoorn, The Netherlands, June 2014.
Inorganics 2017, 5, 34
35 of 35
108. Ray, A.K.; Beenackers, A.A.C.M. Development of a new photocatalytic reactor for water purification.
Catal. Today 1998, 40, 73–83. [CrossRef]
109. Ray, A.K. Design, modelling and experimentation of a new large-scale photocatalytic reactor for water
treatment. Chem. Eng. Sci. 1999, 54, 3113–3125. [CrossRef]
110. Xing, Z.; Zong, X.; Pan, J.; Wang, L. On the engineering part of solar hydrogen production from water
splitting: Photoreactor design. Chem. Eng. Sci. 2013, 104, 125–146. [CrossRef]
111. Molinari, R.; Marino, T.; Argurio, P. Photocatalytic membrane reactors for hydrogen production from water.
Int. J. Hydrog. Energy 2014, 39, 7247–7261. [CrossRef]
112. Yu, S.-C.; Huang, C.-W.; Liao, C.-H.; Wu, J.C.S.; Chang, S.-T.; Chen, K.-H. A novel membrane reactor
for separating hydrogen and oxygen in photocatalytic water splitting. J. Membr. Sci. 2011, 382, 291–299.
[CrossRef]
113. Lo, C.-C.; Huang, C.-W.; Liao, C.-H.; Wu, J.C.S. Novel twin reactor for separate evolution of hydrogen and
oxygen in photocatalytic water splitting. Int. J. Hydrog. Energy 2010, 35, 1523–1529. [CrossRef]
114. Kitano, M.; Tsujimaru, K.; Anpo, M. Decomposition of water in the separate evolution of hydrogen and
oxygen using visible light-responsive TiO2 thin film photocatalysts: Effect of the work function of the
substrates on the yield of the reaction. Appl. Catal. A 2006, 314, 179–183. [CrossRef]
115. Sasaki, Y.; Kato, H.; Kudo, A. [Co(bpy)3 ]3+/2+ and [Co(phen)3 ]3+/2+ electron mediators for overall water
splitting under sunlight irradiation using Z-scheme photocatalyst system. J. Am. Chem. Soc. 2013, 135,
5441–5449. [CrossRef] [PubMed]
116. Yan, W.; Zheng, C.L.; Liu, Y.L.; Guo, L.J. A novel dual-bed photocatalytic water splitting system for hydrogen
production. Int. J. Hydrog. Energy 2011, 36, 7405–7409. [CrossRef]
117. Bae, S.W.; Ji, S.M.; Hong, S.J.; Jang, J.W.; Lee, J.S. Photocatalytic overall water splitting with dual-bed system
under visible light irradiation. Int. J. Hydrog. Energy 2009, 34, 3243–3249. [CrossRef]
118. Huang, C.; Yao, W.; Ali, T.; Muradov, N. Development of efficient photoreactors for solar hydrogen
production. Solar Energy 2011, 85, 19–27. [CrossRef]
119. Jing, D.; Liu, H.; Zhang, X.; Zhao, L.; Guo, L. Photocatalytic hydrogen production under direct solar light in a
CPC based solar reactor: Reactor design and preliminary results. Energy Convers. Manag. 2009, 50, 2919–2926.
[CrossRef]
120. Jing, D.; Guo, L.; Zhao, L.; Zhang, X.; Liu, H.; Li, M.; Shen, S.; Liu, G.; Hu, X.; Zhang, X. Efficient solar
hydrogen production by photocatalytic water splitting: From fundamental study to pilot demonstration.
Int. J. Hydrog. Energy 2010, 35, 7087–7097. [CrossRef]
121. Priya, R.; Kanmani, S. Design of pilot-scale solar photocatalytic reactor for the generation of hydrogen from
alkaline sulfide wastewater of sewage treatment plant. Environ. Technol. 2013, 34, 2817–2823. [CrossRef]
[PubMed]
122. Baniasadi, E.; Dincer, I.; Naterer, G.F. Exergy and environmental impact assessment of solar photoreactors
for catalytic hydrogen production. Chem. Eng. J. 2012, 213, 330–337. [CrossRef]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).