Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Electron configuration wikipedia , lookup
Photovoltaics wikipedia , lookup
Mössbauer spectroscopy wikipedia , lookup
Marcus theory wikipedia , lookup
Surface properties of transition metal oxides wikipedia , lookup
Ultraviolet–visible spectroscopy wikipedia , lookup
Photoelectric effect wikipedia , lookup
Electrochemistry wikipedia , lookup
Hydrogen-bond catalysis wikipedia , lookup
Photoredox catalysis wikipedia , lookup
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/).