Journal of CO2 Utilization 16 (2016) 194–203
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
On the general mechanism of photocatalytic reduction of CO2 Elham Karamian, Shahram Sharifnia* Catalyst Research Center, Chem. Eng. Dept., Razi University, Kermanshah 67149-67246, Iran
A R T I C L E I N F O
Article history: Received 15 December 2015 Received in revised form 2 June 2016 Accepted 10 July 2016 Available online xxx Keywords: CO2 reduction Mechanism Reductant Photocatalysis Greenhouse gases
A B S T R A C T
This paper is going to make a global view of the reaction mechanism of photocatalytic reduction of CO2. During the past decades, extensive studies have been conducted to identify the photocatalytic products of CO2 conversion and their reaction mechanisms. There is a need to propose a general mechanism pattern covering the results of these original researches. Among many factors affecting the redox reaction pathway, this paper focuses on the effect of reductant type as an important issue on the product diversity and selectivity of CO2 photocatalysis. We first tried to show the reaction mechanism of CO2 photocatalysis in the solitary presence of the most common reductants, H2, H2O (both gaseous and aqueous phases), CH4, and CH3OH. Then, a general mechanism has been suggested for CO2 conversion in the mixture of the reductants, based on the fact that in the photoreduction process of CO2, all of the reductants can be derived from one of them. ã 2016 Elsevier Ltd. All rights reserved.
1. Introduction The global warming caused by the accelerative accumulation of atmospheric carbon dioxide (CO2) from both natural and anthropogenic resources is the most crucial issue for human being. CO2 is a noncondensing and stable molecule which is not thermodynamically easy to be transformed into other chemicals at mild reaction conditions. In the past decades, photocatalysis of CO2 has been regarded as a promising sustainable pathway for CO2 mitigation. Indeed, by means of photocatalysis, CO2 can be served as a building block for the synthesis of other useful chemicals and chemical intermediates [1,2]. Some semiconductor materials like TiO2 [3], ZnO [4], WO3 [5], and ZnS [6] in the bare or improved forms of them have been used for this purpose. The reduction of CO2 into hydrogenated compounds needs to reducing substances like H2O, H2, CH4, and alcohols (like methanol and ethanol), which act as both the hydrogen sources and holes scavenges [7]. The photocatalytic process of CO2 follows a relative unknown and complex mechanism leading to different products at the same time. Thus, understanding the mechanism of CO2 conversion could be effective in the control of product selectivity. Photocatalytic-based conversion of CO2 mostly produces the organic compounds such as CO, CH4, HCOOH, HCHO, CH3OH and other light hydrocarbons [2,8,9]. The effect of reductant on the yield and distribution of these
* Corresponding author. E-mail addresses:
[email protected],
[email protected] (S. Sharifnia). http://dx.doi.org/10.1016/j.jcou.2016.07.004 2212-9820/ã 2016 Elsevier Ltd. All rights reserved.
products has been studied in many literatures. Although most studies on the mechanism of photocatalytic reduction of CO2 have focused on TiO2, the results could be applicable for other photocatalysts. This paper first highlights the role of reductants (H2O, H2, CH4, and alcohols) in the diversity and distribution of products of CO2 photocatalysis regardless of the potential levels of the excited charge carriers. In other words, it was assumed that the reactions can proceed as much as possible without any limitation relating to the position of conduction and valence band edges. Then, a general mechanism for CO2 reduction in the presence of all the reductants has been offered. 2. Basic principles of photocatalysis Photocatalysis is a process in which the absorption of light photons having energy equal to/greater than the band gap energy (Ebg) of a semiconductor catalyst (SC) excites electron (e) and hole (h+) to the empty conduction and valence bands (CB and VB), respectively. These energetic charge carriers initiate various redox reactions to produce final products [2,10]. Eqs. (1) and (2) show the semiconductor activating reactions. SC + hn ( Ebg) ! h+ e
(1)
h+ e ! h+VB + eCB
(2)
These photoinduced charge carriers may follow different routes. The photogenerated electron and holes could migrate
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
and trap at the shallow and deep trapping sites. In the case of TiO2, TiIIIOH and TiIVOH (in the hydrated surface) are the shallow trapping sites of electron and hole, respectively. The electrons could also be trapped at bulk trapping sites in the form of TiIII. If the electrons fail to find any trapped sites, or the energy band gap of the semiconductor is very small, they recombine and generate thermal energy at the surface or bulk of photocatalyst. Indeed, charge trapping is needed to avoid charge recombination, and increase photoactivity. Finally, the generated electron-hole could be transferred to the adsorbed species on the photocatalyst surface. The final route is the most desirable event for efficient application of photocatalysis [11–13]. The capability of a photocatalyst to succeed a typical redox reaction depends on the positions of the energy levels of the photocatalyst and adsorbed substrate. For driving a reduction reaction, the potential of conduction band should be more negative than the potential required for the reduction reaction. Similarly, an oxidation reaction may happen if the potential of the valence band be more positive than that of the oxidation reaction [2,14]. 3. Photocatalytic oxidation of reductants
In general, OH radicals could be produced via direct oxidation of OH (Eq. (7)), or through the intervention of H2O2 (Eqs. (8)– (14)). Some of these reactions can only produce the adsorbed OH radicals and others, both the adsorbed and free OH radicals [16,18]. OH + h+VB! OHads
(7)
O2 + eCB! O2
(8)
O2 + 2eCB + 2H+ ! H2O2
(9)
O2 + O2 + 2H+ ! H2O2 + O2
(10)
2OHads ! H2O2
(11)
H2O2 + hn ! OHfree/ads + OH
When CO2 is reduced by photogenerated electrons, an equal number of holes has to be served to decrease the probability of charge recombination and increase the lifetime of electrons. By utilization of electron donor reductants, the holes not only could be scavenged, but they also supply the hydrogen needed for synthesis of hydrogenated products. In fact, the hydrogenated products of CO2 conversion are not formed, unless the hydride reductants simultaneously be oxidized by oxidizing agents. Thus, we first need to know about the oxidizing agents and oxidation mechanism of reductants. In this section, the oxidation mechanism of aliphatic has been discussed to find the reactive radicals, products and byproducts of photocatalytic oxidation of methane and methanol as reductants.
195
free/ads
H2O2 + eCB ! OHads + OH
H2O2 + O2 ! OH
free/ads + OH
(12)
(13)
+ O2
(14)
Superoxide radical as a weak oxidizing species is less important in starting oxidation reaction. O2 is the product of either oneelectron reduction of molecular oxygen by a conduction band electron (Eq. (8)), or oxidation of H2O2 by valence band hole (Eq. (15)) or by OH (Eqs. (16) and (15) is the major reaction to give superoxide radical anion from H2O2 [16]. H2O2 + h+VB + 2OH ! O2 + 2H2O
(15)
H2O2 + OH + OH ! O2 + 2H2O
(16)
3.1. Oxidizing species In order to understand the reaction mechanism of CO2 photocatalysis, it is essential to clarify the oxidizing species generated at the excited photocatalyst surface or bulk phase. Apart from the holes as primary oxidizing species that drive a direct oxidation process, the other main oxidizing species include hydroxyl radicals (free radical: OHfree and adsorbed radical: OHads) [15], superoxide (O2) [16], and singlet oxygen (1O2) [17]. Additionally, in some photocatalytic oxidation processes, participation of hydrogen peroxide (H2O2) and molecular oxygen has been reported [13,16]. OH radicals are mainly created by oxidation of surface hydroxyl or adsorbed water. The oxidation process initiated by OH is considered as indirect oxidation. In the case of TiO2, OH formation was suggested by the reaction pathway including a nucleophilic attack of water on a surface trapped hole (Eqs. (3)–(6)). The final reaction (Eq. (6)) is based on breaking the O O bond in TiO OH [13]. [Ti-O-Ti] + h+VB + H2O ! [Ti-O HO-Ti] + H+
+
O-Ti] + H [Ti-O HO-Ti] ! [Ti-O
(3)
Singlet oxygen, a strong oxidant, is known as the product of reaction between O2 and a trapped hole. In spite of the high reactivity of 1O2, its short lifetime (2 ms) on TiO2 surface than that of OH (ca. 10 ms) makes it less important compared with hydroxyl radical [13,17]. 3.2. Photocatalytic oxidation of aliphatic (alkanes and alcohols) Since alkanes (mainly methane) and primary alcohols (mainly methanol) are usually utilized as common reductants for photocatalytic reduction of CO2, it needs to be investigated their photocatalytic oxidation mechanisms. For the oxidation of alkanes, the subsequent reactions are thought to be initiated by direct (Eq. (17)) or indirect (Eq. (18)) oxidation processes. Based on the underlying mechanism, alcohols are the first by-product of photocatalytic oxidation of alkanes, so this mechanism also involves the oxidation of primary alcohols. RCH3 + h+VB ! RCH2 + H+
(17)
RCH3 + OH ! RCH2 + H2O
(18)
(4)
OH HO-Ti] [Ti-O O-Ti] + H2O ! [Ti-O
(5)
[Ti-O OH HO-Ti] ! [Ti-O HO-Ti] + OH
(6)
In the next step, alcohols can be yielded from the produced alkyl radicals via two different pathways, Eq. (19) or according to consecutive Eqs. (20)–(23):
196
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
RCH2 + OH ! RCH2OH
RCH2 + O2 ! RCH2OO
(19)
(20)
RCH2OO + RCH3 ! RCH2OOH + RCH2
(21)
RCH2OOH + eCB ! RCH2O+-OH
(22)
RCH2O + RCH3 ! RCH2OH + RCH2
(23)
Then, the preceding alcohols can be further oxidized to their corresponding aldehydes and carboxylic acids (Eqs. (24)–(27)). RCH2OH + OH ! RCH2O + H2O
(24)
RCH2O + OH ! RCHO + H2O
(25)
RCHO + OH ! RCHOOH
(26)
RCHOOH + OH ! RCOOH + H2O
(27)
If the hydroxyl radical attacks to secondary alcohols, ketones can be obtained (Eqs. (28)–(29)). For instance, 2-hexanol and 3hexanol in the existence of OH give corresponding ketones i.e., 2hexanone and 3-hexanone. RCHOHR' + OH ! RCHOR' + H2O
(28)
RCHOR' + OH ! RCOR' + H2O
(29)
It must be emphasized that the alkanes and all other volatile organic compounds can be finally mineralized to CO2 and H2O from the photodecarboxylation reaction of carboxylic acids in the presence of a semiconductor photocatalyst. The decarboxylation reaction takes place via photo-Kolbe process. The mechanism includes OH radical attack, and formation of carboxylate ions, which can discharge through photogenerated holes, as given in Eqs. (30)–(31) [18–21]. RCOOH + OH + h+VB ! RCOO + H2O
RCOO ! R + CO2
(30)
CO2 is a stable molecule (DG = –400 kJ mol1) with linear configuration having two double bonds between carbon and oxygen atoms that its conversion into value added chemicals without any catalyst and input energy is not possible [22]. Apart from the thermodynamics and kinetics limitations, the large energy gap between LUMO (lowest unoccupied molecular orbital) and HOMO (highest unoccupied molecular orbital) (13.7 eV) as well as a large electron affinity of CO2 (-0.6 0.2 eV) make it inactive molecule [23,24]. Photocatalysis has been shown as a facile method for activating and converting CO2 into useful chemicals. Adsorption of CO2 onto a photocatalyst surface is the best way for lowering its energy barrier, because the linear structure of CO2 transforms to a bent form and thus it shows a high reactivity. This is due to the fact that the LUMO level of CO2 decreases as the molecule bends. One-electron transfer to form the surface-bound CO2 on a photocatalyst initiates the sequential chemical reactions. These reactions mostly involve the transfer of an electron or a proton, breaking CO bonds, and creating new C H bonds. Depending on the number and potential of the charge carriers taking part in the chemical reaction, operating conditions, and the type of reductant, the distribution of the final products can be different [25,26]. In order to obtain photocatalytic products like hydrocarbons, a reducing agent for supplying hydrogen is needed. The main reductants used to succeed CO2 reduction are H2O, H2, CH4, and CH3OH [2]. Based on the different literatures, photocatalytic reduction of CO2 depending on the reductant reagent follows different reaction pathways. It may be said that the carbon-free reductants (H2O and H2) lead to the formation of C1 products, but the carbon-containing species (CH4 and CH3OH) may even form C2 and C3 products. The knowledge about what exactly happened in CO2 photocatalysis is still limited, but it is believed that formation of CO2 anion radical by capturing an electron from the CB of a photocatalyst is the first step (Eq. (36)) [7,27]. CO2 is an acidic molecule and shows strong affinity to the basic surface. The formation of anionic radical CO2 on acidic metal oxide like TiO2 is due to the physisorbed interaction between CO2 and photocatalyst surface [28]. The presence of abundant Ti3+ sites on the TiO2 surface creates a low-coordinated oxygen species, which increases the number of basic sites on titania [29]. Some experimental reports have discussed about photoreduction of CO2 using basic metal oxides such as ZrO2 [30], MgO [31], and Ga2O3 [32] which have shown the chemisorbed interaction with CO2. CO2 + eCB ! CO2
(31)
For secondary alcohols, the following oxidation mechanism in water leading to formation of ketones was suggested (Eqs. (32)– (35)) [18]. RR0 CHOH + h+VB ! RR0 C OH + H+
(32)
RR0 C OH ! RC(¼O)R0 + H+ + eCB
(33)
RR0 C OH + O2 ! RR0 COO OH
(34)
RR0 COO OH ! RC(¼O)R0 + O2 + H+
4. Photocatalytic reduction of CO2
(35)
(36)
This anionic radical can be simultaneously protonated to format ion (Eq. (37)), which is a way for reducing the potential required CO2 reduction, or it may undergo disproportionation to form CO and carbonate (Eq. (38)) or dimer to give oxalate ion (Eq. (39)) [22,33].
CO2 + H+ ! HCOO
(37)
CO2 + CO2 ! CO + CO32–
(38)
(39)CO2 + CO2 ! [O2CCO2]2 CO as the typical first product can be also generated from CO2 by dissociating on the oxygen vacancy sites (Eq. (40)) [34]. [SC + VO] + CO2 ! CO + SC
(40)
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
Based on the experimental observations, the products of CO2 photoreduction with the common reductants are often the same (i.e., CO, HCOOH, HCHO, CH3OH, CH4, and C2H5OH) [33,35,36]. The Scheme 1 shows how many reducing electrons, oxidizing holes, and protons (H+) are needed for the formation of the main products. Up now, the photocatalytic reduction of CO2 with a variety of semiconductors and in the presence of different reductants has been investigated. One of the pioneering works in this field was reported by Inoue et al. [37]. They observed the possibility of CO2 reduction in aqueous solution by several semiconductors like WO3, TiO2, ZnO, CdS, GaP, and SiC under UV irradiation. The photocatalytic products were found to be formaldehyde, formic acid, methyl alcohol, and trace amounts of methane. In the following, we will discuss how each of these reductants may affect the reaction mechanism and product diversity.
197
For instance, fabrication of photocatalysts with oxygen vacancy sites may enhance the CO2 adsorption through locating one oxygen atom of CO2 at bridging oxygen vacancy (in the metal oxide photocatalysts) [25]. Studies show that the CO2 conversion by liquid or vapor water usually does not follow the same pathways. Table 1 provides a comparison between the products of CO2 + H2O photocatalysis in the gas/solid and liquid/solid regimes in some literatures. Generally, two basic fundamental mechanisms have been proposed for photocatalysis of CO2 + H2O, once by Subrahmanyam [42] and the other by Sasirekha [43]. The former mechanism allows CO2 to form methane and methanol via a series of reactions;CO2 !HCOOH ! HCOH ! CH3OH ! CH4, and the latter mechanism assumes that methane and methanol are formed in parallel reactions through formation of carbon monoxide;CO2 ! CO ! CH4, CH3OH. These two models have been applied in the following sections.
4.1. Photocatalytic reduction of CO2 by H2O Among all reductants and hole scavengers for photocatalytic reduction of CO2, H2O still remains the most naturally abundant source of hydrogen, as it is inexpensive and readily available. The reduction reaction of CO2 by H2O needs to the simultaneous water splitting to form hydrogen, which is necessary for useful carbonbased chemical products [38,39]. In comparison to the CO2 reduction, reducing H2O to H2 (Eqs. (41)–(43)) is more interesting in term of kinetics and thermodynamics. In thermodynamics point of view, the standard reduction potentials of H2O/H2 is zero (at pH = 0.0) which is more than the standard reduction potential of CO2/CO2 (-1.9 V). In kinetics point of view, reduction of water is a 2-electrons transfer process, but reduction of CO2 to the most hydrogenated compounds needs to 4–8 electrons [14,33]. These mean that the H2 formation is more favorable than formation of other products of CO2 reduction. Therefore, it is better to measure the volume of hydrogen and other products for understanding whether the water splitting is a rival for CO2 reduction [40,41]. H2O + h+VB ! H+ + OH
(41)
H+ + eCB ! H
(42)
4.1.1. Aqueous media When CO2 is photocatalyzed in the condensate water media, the total formation rate of CH3OH may be higher than other products [see Table 1, and Ref. [55]]. Since CH3OH can be directly used as a fuel, it may be more advantageous for applying the liquid systems. The reaction mechanism can be initiated either by direct reduction of CO2 through conduction band electrons or by dissolving it and formation of carbonic acid, bicarbonate or carbonate ions depending on the pH of solution (CO2 + H2O Ð H2CO3 Ð HCO3 + H+ (pKa = 6.4), HCO3 Ð CO32– + H+ (pKa = 10.3)). The solubility and chemical form of CO2 are affected by the pH value of solution as follows: For pH > 10 and pH < 4, aqueous CO2 solutions mainly contain CO32, and H2CO3/CO2, respectively. Near pH = 7, the three carbonic acid, bicarbonate, and carbonate ions may be presented. As each of these forms of CO2 have different adsorption characteristics on photocatalyst surface, the different reduction pathways are expected [1,22,56,57]. Scheme 2 shows the possible mechanism of photocatalytic reduction of CO2 in the aqueous water solution [33,42,43,58–66]. It seems from the mechanism that the significant formation of HCOOH is the reason why methanol is often the major product in the aqueous media. Although, carbon monoxide and methane can be concurrently formed by Eqs. (44) and (45), respectively [7,8]. 2CO2 + 4eCB ! 2CO + O2
(44)
(43) H + H ! H2 In addition, a comparison between polarities of H2O and CO2 reveals that H2O due to its polar nature (1.85 D) has a greater tendency to be adsorbed on photocatalyst surface (such as TiO2) than CO2 with a lower dipole moment (0 D). It follows that H2O is often more successful for adsorption on metal oxide photocatalysts [41]. To overcome this limitation, some ways have been suggested.
2•H + 2h+VB
4•H + 4h+VB •
CO2–
•
6 H + 6h
+
VB
8•H + 8h+VB 12•H +12h+VB
HCOOH
Formic acid
CO + H2O
Carbon monoxide
HCHO + H2O
Formaldehyde
CH3OH + H2O
Methanol
CH4 + 2H2O
Methane
C2H5OH + 3H2O
Ethanol
Scheme 1. The total pathways for production of the main products of photocatalysis of CO2.
CO + 6e + 6H+ ! CH4 + H2O
(45)
2
In spite of the more stable nature of HCO3 and CO3 than CO2, meaning the more difficult reduction of them than CO2 itself, carbonate and bicarbonate ions may act as hole scavengers and improve photocatalytic activity [33,57]. Also, based on Eqs. (46)– (47), CO2 could be produced easily by oxidation of peroxycarbonate, which confirms the important role of bicarbonate ions as the hole consumers [67]. 2CO3 ! C2O62
(46)
C2O62 + 2 h+VB ! 2CO2 + O2
(47)
Glyoxal (HOCCOH), glycol aldehyde (CH2OHCHO) and acetaldehyde (CH3COH) can also be formed by photocatalysis of CO2 in the aqueous water solution. As shown in Scheme 3, the mechanism is thought to be initiated by oxidation of formaldehyde via OH to give formyl radical. Glyoxal is a production of dimerization of two formyl radicals. This by-product due to its high electron affinity is easily reduced to trans-ethan-1, 2-semidione which can be further
198
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
Table 1 Summarized results of some reports on gaseous and aqueous phases photocatalysis of CO2 + H2O by different photocatalysts. Photocatalyst
Reactants
ZnO GaP CdS TiO2 TiO2 (P25) N3 dye-0.5 wt% Cu- 0.5 wt% Fe-TiO2 ZnS/MMT CdSe-Pt/TiO2 PbS-Cu/TiO2 Cu-TiO2 HNb3O8 TiO2 TiO2 (Anat.) Cu-Fe/TiO2-SiO2
CO2
H2O
Saturated Saturated Saturated Saturated Gas Saturated Saturated Gas Gas Saturated Saturated Saturated Saturated Gas (0.72 bar)
Liquid (100 mL) Liquid (100 mL) Liquid (100 mL) 100 mL, 0.1 N NaOH Gas Water vapor 0.2 M NaOH solution Gas Saturated vapor 300 mL, 0.2 N NaOH Water vapor (7 kpa) Saturated vapor Liquid (57.5 mL) Vapor (0.72 bar)
Light source
Major Product
Ref.
500 W Xe/HP Hg 500 W Xe/HP Hg 500 W Xe/HP Hg 500 W W-halogen UV (365 nm) Concentrated natural sunlight (20 mWcm2) 8 W Hg lamp (254 nm) 300 W Xe ( > 420 nm) 300 W Xe (UV cut) Hg (254 nm) 350 W Xe lamp UV light (253.7 nm) 990 W Xe lamp (340 nm) 150 W Hg and solar light
HCHO, CH3OH HCHO, CH3OH HCHO, CH3OH HCOOH CH4 CH4 CH3OH, CH4 CH4, CH3OH CO, CH4 CH3OH CH4 CH4 HCOOH CH4, C2H4
[37] [37] [37] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54]
H+
H2
CH3OH
H2•COH
OH•
H2 O2
O2
h+VB •
O2
+
h
CO32–
VB
•
CO3
h
HCO3H2CO3
e–CB
+
–
HCHO
CO3
VB
e–CB
H+
HCOO
CO2
H2 O
H• 2h+VB
e–CB
•
HCOOH
OH– HCO3– e–CB
e–CB
e–CB
HCOO•
•CO2–
H• e–CB
HCOO–
e–CB
HCOO• h+VB
H+
H2 O
OH• Scheme 2. The proposed reaction mechanism of aqueous phase photocatalysis of CO2 + H2O.
H2 O •
•
HCHO
COH
COH +
H2 O
h
HOCCOH
HO•CHCOH e–CB
VB
H+
•
OH e-CB
•
H
CH3COH
e-CB
•
CH2COH
e–CB
CH2OHCOH
H2 O Scheme 3. The proposed reaction mechanism of glyoxal, glycolaldehyde, and acetaldehyde via photocatalytic oxidation of formaldehyde during aqueous phase photocatalysis of CO2 + H2O.
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
reduced to glycolaldehyde. Finally, acetaldehyde is produced from vinoxyl radical (CH2CHO), a known precursor of acetaldehyde which is formed by glycolaldehyde reduction [1,25,68,69]. In accordance with Eqs. (48)–(50), CO and CH4 can also be formed through oxidation of acetaldehyde to unstable acetyl radical, which gives carbon monoxide and methyl radical by decarbonylation [1]. CH3COH + h+VB ! CH3C O + H+
H2 O h+VB
(49)
–
●
H+
OH
CH4
(48) CO
e–CB
3•H •
CO–
OH–
•
•
CH3
C
OH–
CH3OH
Scheme 4. The proposed reaction mechanism of gaseous phase photocatalysis of CO2 + H2O.
(50)CH3 + H ! CH4 The low solubility of CO2 in water is one drawback of utilizing this reductant. Thus, some novel reductants such as triethylamine [70], triethanolamin [71], dimethylformamide [72] and isopropyl alcohol [73] have been used as sacrificial electron donors for improving the efficiency of charge transfer process. By utilizing these solvents, a linear correlation between the dielectric constants and the photogenerated electrons is observed. In an aqueous medium, the polarity of solvent determines the fate of CO2 anion radical for formation of CO or HCOO. As given in Eqs. (51)–(52), in low polarity solvents (or low dielectric-constant solvents) such as CCl4 and CH2Cl2, CO2 owing to a poor solvation may be strongly adsorbed at the photocatayst surface, and it forms CO and H2O. But, in the polar solvents (or high dielectric-constant solvents) such as H2O, another mechanism prevails, so that CO2 is stabilized by solvent molecules, which results in a weak interaction with the photocatalyst surface. Then, upon protonation and further reduction-protonation, CO2 is transformed to formic acid (Eqs. (53)–(55)) [6,7,74,75]. CO2 + eCB ! CO2ads
H
e–CB
e–CB
CO2
CH3C O ! CH3 + CO
●
CO2
•
199
pathways toward the major formation of CH3OH [88]. Methane formation mainly results from a reaction between carbon radicals (from CO reduction) and atomic hydrogen [84]. O2 as a product of water oxidation has a main role in the efficiency of methanation process by trapping the conduction band electrons and preventing CO2 from reduction. Additionally, in the photocatalysis of gaseous CO2 + H2O, it is supposed that some products like CH3OH and CO could not be fully desorbed from the catalyst surface and may be re-oxidized back into CO2 through molecular oxygen [6,89]. Therefore, in the conversion of CO2 by water vapor, a main issue to be taken into consideration is the H2O:CO2 molar ratio that must be optimized. Based on the stoichiometric overall chemical equation (Eqs. (56)–(57)), the amount of water required for the best efficiency of CH4 and CH3OH production is twice as much as CO2. Thus, insufficient amount of water vapor leads to remarkably low yield of product generation. Although the CO2 reactivity increases by increasing the H2O:CO2 ratio, but an excessive amount of H2O suppresses the reaction [9,90].
(51) 4.2. Photocatalytic reduction of CO2 by H2
CO2ads + eCB + 2H+ ! CO + H2O
CO2 + eCB ! CO2sol
CO2sol + H+ ! O COH
O
COH + eCB + H+ ! HCOOH
(52)
(53)
(54)
Since, in the photocatalytic reduction of CO2 by H2O, H2 evolution proceeds preferably through water splitting, thus CO2 photoreduction with H2 is more viable for production of fuels with higher yield rates. In addition, photoreduction of CO2 with H2 is more thermodynamically favorable than the reaction with H2O as shown in Eqs. (56)–(59) [9,25]. CO2(g) + 2H2O(g) ! CH3OH + 1.5O2(g)15O2(g) mol1
DGr = +689 kJ (56)
(55)
In addition, experiments show that addition of NaOH or another alkaline solution dissolves more CO2 than pure water [76]. The use of NaOH not only enhances CO2 solubility, but also improves the CO2 reduction due to more concentration of OH ions, which promote the longer decay time of conduction band electrons. In general, increase in alkalinity, increases adsorption of the slightly acidic CO2 on the photocatalyst surface in the liquid-solid systems. Also, increasing CO2 concentration significantly increases the product yields [74,75]. 4.1.2. Gaseous media By a literature survey on gas phase photocatalysis of CO2, it seems that under gaseous conditions, carbon dioxide usually tends to be reduced to CH4 instead of CH3OH [see Table 1, and Refs. [55,77]]. Based on Scheme 4, this may be assigned to the more production of CO rather than formic acid in the gaseous systems [78–87]. However, other parameters like photocatalyst type, reaction pressure and temperature, etc. could affect the reaction
CO2(g) + 2H2O(g) ! CH4 + 2O2 DGr = +818.3 kJ mol1
(57)
CO2(g) + 3H2(g) ! CH3OH(g) + H2O(g) DGr = +2.9 kJ mol1
(58)
CO2(g) + 4H2(g) ! CH4(g) + 2H2O(g) DGr = -113.6 kJ mol1
(59)
H2 can be supplied from either reverse water-gas-shift reaction or photocatalytic water splitting [9]. H is expected to be the main highly reactive radical in the photocatalytic conversion of CO2 + H2. The Scheme 5 proposed the CO2 photocatalysis mechanism by H2 [9,84,91–96]. As investigated by Lo et al. [92] who reported the photocatalytic conversion of CO2 by TiO2/SO42–, the reaction products and yield rates of them varied significantly with the type of reductant (H2 and H2O). The yield rates of CO and CH4 formation in the presence of H2 were much more than those by H2O and H2 + H2O. Both H2
200
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
4•H
•
•
H
HCHO
H2•COH
•
H
H
CH3OH
H2 O
H2 O
•
H
•
3H
•
2H
HCOOH
e–CB
CO2
e–CB
•
H
•
CO2
CO OH
•
CO
–
–
•
H
4H
C
OH
CH3
•
CH3
•
•
–
2•H
CH4
•
C2H6
CH4 • •
CH2
CH2
C2H4
Scheme 5. The proposed reaction mechanism of photocatalysis of CO2 + H2.
and H2O are oxidized to H, but formation of O2 from H2O would actually retard the photoreduction of CO2, which leads to change in the distribution of products. Additionally, the presence of molecular O2 may accelerate the re-oxidation of adsorbed products like CO and CH3OH back into CO2, as mentioned in Section 4.1.2 Photoreduction of CO2 by H2 usually forms the same products as H2O, but the yield rates and pathways are different. The reaction of CO2 + H2 to CH4 have an exothermic nature (Eq. (59)). This reaction can be carried out in the dark at moderate temperatures and catalyzed by nickel (the Sabatier reaction) [97]. But, this reaction may be photocatalytically initiated at room temperature. As the reaction progresses, a sudden increase of temperature is happened, and depending on temperature, a combination of catalytic and photocatalytic process could take place. Indeed, CO2 reduction by H2, the so-called Sabatier reaction would be carried out thermally at high temperature [98,99]. Therefore, in the photocatalysis of CO2 + H2, temperature is a key parameter that must be respected. 4.3. Photocatalytic reduction of CO2 by CH4 Both CO2 and CH4 are stable greenhouse molecules that their direct conversion into oxygenated compounds is not thermodynamically favorable at mild reaction conditions. Methane has a high C H bond energy of about 439 kJ mol1 [100]. Photocatalytic reduction of CO2 with CH4 at the same time is an ideal redox reaction for recycling them back to fuels. As both of CO2 and CH4 are carbon containing compounds, the formation of heavier products with more than one carbon atom would be expected. Scheme 6 represents a possible reaction mechanism for photocatalysis of CO2 + CH4 [84,101–107]. It must be noted that C2H4 can also be generated by oxidation of CH3CH3 via valence band holes to CH2CH3 which gives CH2CH2 and H through an elimination reaction activated by photons. Moreover, C2H2 production is possible through CO ! CO ! C ! CH2 + H ! C2H2 route.
OH•
CO
–
CO2
e–CB
H
•
C
CO2
2•H
•
CH3COCH3
CH3
h+VB
CH4 •
Due to the stronger reducibility and easier solubility of CO2 in the methanol, it has advantageous to be a suitable alternative as reductant [108]. If CH3OH is utilized for reducing CO2, HCOOH and HCHO are expected to be the main products, because they can be formed from either oxidation of alcohol or reduction of CO2. Although, by aging time and reactions proceeding, a part of these carbonyl compounds could be mineralized to CO2 and H2O, and CO2 may undergo methanation by the excited electrons. It may be interesting that methane yielding in the alcohol photocatalytic systems usually is low, because of the fact that the good hole scavengers like methanol, ethanol and 2-propanol, which are responsible for methane formation, have very low potentials amongst other reactive organic radicals. In other words, the major source of CH4 is CO2 [89]. In the CO2 photoreduction systems containing alcohols, the ambient conditions affect the reaction progress especially mineralization, based on photo-Kolbe process. For example, the presence of molecular oxygen suppresses the reaction leading to the formation of hydrocarbons. It is another reason showing that methanation is occurred only by reduction of CO2 [109]. Scheme 7 offers a possible reaction mechanism for photocatalysis of CO2 + CH3OH [17,108–110]. 2-prpanol (the simplest secondary alcohol) was also served as a reductant and hole scavenger for CO2 photoreduction. It has been shown that acetone, carbon monoxide, carbonate, format, acid
•
3•H H
4.4. Photocatalytic reduction of CO2 by alcohols
OHe–CB
CO
•
The CH4:CO2 ratio is an important parameter that must be respected. The highest conversion is often achieved at the higher ratios of methane to carbon dioxide. It may be assigned to the higher CO2 stability relative to CH4 [102]. As reported by Torabi Merajin et al. [104], the highest photocatalytic activity of TiO2 after 8 h of UV irradiation was achieved at 45%CO2: 45%CH4: 10%He initial feed composition.
•
CH2 C2H6
•CH3
OH•
-
CH3COO–
H+
CH3OH
C3H8 •
H
CH3COOH
•
H HCOO–
H+
HCOOH
Scheme 6. The proposed reaction mechanism of photocatalysis of CO2 + CH4.
Scheme 7. The proposed reaction mechanism of photocatalysis of CO2 + CH3OH.
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
CO2 by any one of the reductants (H2O, H2, CH4, or CH3OH) is able to give various chemicals like saturated and unsaturated hydrocarbons, oxalic acid (HOOCCOOH), methyl acetate (CH3COOCH3), and acetone (CH3COCH3). Also, methyl formate (HCOOCH3) and acetic acid (CH3COOH) are two isomeric structures of glycolaldehyde (HOCH2CHO), which are possible to be formed during photocatalysis of CO2. The presence of these organic compounds have been exactly detected or confirmed in some literatures by identification analyses like FTIR and GC/MS [33,35,36,84, 103–105,107,109]. This general mechanism tells us two main points; (1) if a product such as methyl format is detected, its production pathway could be described by such a mechanism, (2) if the spectroscopic analyses like FTIR show characteristic peaks of any functional group, this general mechanism gives an idea about compounds having that functional group which is possible to be formed. This mechanism does not take into account any limitation related to the potential levels of conduction and valence bands. Indeed, in the practical applications, some of these products may be impossible to be generated, because the potential of conduction band edge determines the possibility of producing any particular product. For example, for a photocatalyst with potential level of CB equal to 0.0 V which is higher than Eo(CO2/CH4) and lower than Eo(CO2/CH3OH), CH3OH may be the dominant product. Based on the general mechanism, in the photocatalytic reaction pathways to form a final product, the reactions may be stopped somewhere that the excited charge carriers of the used photocatayst don’t have enough power to redox the intermediates. In addition, the product accumulation on the photocatalyst surface and deactivation of the photocatalyst can inhibit the formation of some products. Although this general mechanism is mainly based on free oxidizing radicals at the bulk phase. Researches show that the different photocatalysts may behave differently under similar conditions. For instance, ZrO2 and TiO2 which were compared by Lo et al. [94] for photocatalysis of CO2 + H2, had different performances at the
Scheme 8. The proposed reaction mechanism of photocatalysis of 2-propanol.
formic, and methane may be formed. Acetone is produced by oxidation of 2-propanol via excited holes, as given in the Scheme 8 (see also the oxidation mechanism of secondary alcohols, Section 3.2). Format an acid formic are the products of successive attacks of H to CO2. Other products and by-products follow CO2
201
CO2 ! CO ! C ! CH2 ! CH4 and CO2 ! CO + CO32–routes [73,110–112]. Cyclohexanol (a cyclic secondary alcohol) is a good option as solvent and reductant of CO2, because, at room temperature and normal pressure, the saturated mole fraction of CO2 in cyclohexanol is 4.43 103, which is 7.5 times of that in water. Cyclohexanone ((CH2)5CO) may be a product formed by photocatlytic oxidation of cyclohexanol through valence band holes. The other product is cyclohexyl format (C7H12O2), which is made by the esterification of cyclohexanol and formic acid (as product of CO2 reduction) in the bulk phase [113]. 4.5. A general mechanism Photocatalysis of CO2 involves chain redox reactions leading to the generation of a wide range of organic products. According to experimental observations, the reaction products are distributed between C1 C3 hydrocarbons. Based on the fact that during the photocatalysis of CO2 by each of the mentioned reductants, formation of the others is inevitable, a general mechanism can be proposed. Scheme 9 shows that the photocatalytic reduction of
2•H
e–CB
2
HCOO• H++OH–
e–CB
•
CO2– •
●
•
•
COOH
COOH
HOOCCOOH HCOOCH3
2•CH3
OH
HCHO •
CH3COOCH3 H
CH2OH H2•COH •
H
h+VB
CO
h+VB
H+2h+VB
h+VB
•
HCOOH
•
4•H
CO2
H+
HCOO–
H
CH3COO•
CH3COOH
H+
CH3COO–
CH3OH
2•CH3 CH3COCH3 CO–
•
•
H
• C 3H
•
•
•
CH3
H
CH4
•
CH3
C2H6 •
2H •
H
h+VB
•
h+VB
•
CH3
C2H5
C3H8
•
OH
C2H5 C2H4
h+VB
C2H5OH
C3H6
•
C2H3
Scheme 9. The proposed general mechanism of photocatalysis of CO2 in the mixture of reductants.
202
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203
same conditions. TiO2 was able to form CH4, CO, and C2H6, whereas ZrO2 could only produce CO. The main products derived from a photocatalytic conversion of CO2 are strongly dependent on the type of photocatalyst used. The control of product selectivity and diversity requires further understanding on the rate determining step leading to the formation of each product. In general, the CO2 photocatalytic reduction is controlled by several parameters like operating conditions, photocatalyst type, nature of reducing agents, and reactor configuration. An overview on the large number of literatures dealing with photocatalytic reduction of CO2 indicates that all types of photocatalysts produce the same products with different rates, efficiencies, and distributions. One of the most important factors on controlling product distribution is the acidic/ basic characteristic of the metal oxide photocatalyst [42,114]. The experimental observations have shown that hydrocarbons (C1 C3) are mainly formed in the photocatalytic process with basic photocatalysts like MgO [114,115], Ga2O3 [32,107,116]. On the contrary, acidic metal oxide photocatalysts such as SiO2 [54,117], and Nb2O5 [118] cause the formation of short chain hydrocarbons (CH4 and CO). As the key step in the photocatalysis of CO2 is its adsorption onto the surface of photocatalyst, thus the photocatalyst materials with basic sites are more favorable. Improvement of photocatalytic properties by different methods (transition and noble metal deposition, non-metal doping, photocatalyst coupling, photosensitization, etc.) can also affect the product selectivity and diversity. For example, in a report about photocatalytic reduction of CO2 using Fe/Cu-TiO2, the reaction products of single doped and co-doped photocatalysts (Cu/TiO2, Fe/TiO2, and Fe/Cu-TiO2) were different. Being Fe as co-dopand of Cu/TiO2 formed more amount of ethylene and a trace amount of methane, while methane was favorably produced by Cu/TiO2 and Fe/TiO2 [119]. It must be noted that in the photocatalysis of CO2, some backward reactions are also possible, making the CO2 conversion process more complex. The strong oxidation power of holes and hydroxyl radicals, and the presence of O2 could oxidize the intermediates and products back into CO2. Therefore, the ambient condition is important, because the photo-Kolbe process is affected by reaction medium (either it is inert, aerated, or under high vacuum) [89]. To clarify the role of photo-Kolbe process in the photocatalytic efficiency of CO2 conversion, let’s take an example. CH3COOH upon oxidation by hVB+ and photo-Kolbe process may recycle back to CO2 as the following equation (Eq. (60)) [120]. This means that the efficiencies and yield rates of the products may be very different from what is expected to be. CH3COOH + h+ ! CH3 + CO2 + H+
(60)
Scheme 10 specifically refers to the reaction mechanism of synthesis of alkanes, alkenes, and syngas by photocatalysis of CO2.
•
CO
CO CH4
•
H
C2H4 •
CO2–
C2H6 ●
•
e–CB
C
•
CH
CH2
•
CH3
C3H6 C3H8
CO2 Reductant
h+VB
H+
e–CB
•
H
H2
Scheme 10. The possible reaction mechanism for production of syngas and various hydrocarbons by photocatalysis of CO2.
Given that the formation rates of these useful products rarely exceed tens of mmol g1h1, the innovation of methods to enhance the photocatalytic efficiency is an urgent need. 5. Conclusion As discussed in this paper, photocatalysis can be effectively applied as a new green route for chemical production. Photocatalytic reduction of CO2 is a suitable choice for CO2 mitigation, by which CO2 as a feedstock can be converted to fuels or value-added chemicals. This paper attempts to explain the effects of reducing agents on the chemical pathways of CO2 reduction. This knowledge can be applicable for control of product selectivity. Although, photocatalytic conversion of CO2 is still far from practical applications due to the low yields of products. The main points of this paper are summarized as follows: CO is the first product of photocatalysis of CO2, which has a multitude of industrial uses like Fischer-Tropsh synthesis or methanol synthesis. In the aqueous phase photocatalysis of CO2 + H2O, CH3OH is expected to be formed at a higher rate than CH4. This can be assigned to the formation of carbonic acid, carbonate, and bicarbonate species that lead to HCOOH production. In contrast, CH4 may be predominant in the gaseous phase photocatalysis of CO2 + H2O. Although, other parameters may actually control and affect the reactions to the formation of other products. Under oxygen-free conditions, the formation rates of photocatalytic products of CO2 + H2 often are more than the products of CO2 + H2O. Molecular oxygen may participate in the reoxidizing the adsorbed products back into CO2. When CH4 is the reductant of photocatalysis of CO2, it is possible for heavier products like CH3COOH and CH3COCH3 to be generated. The molar ratio of CO2:CH4 is a main parameter to be optimized. Since HCOOH can be produced from either reduction of CO2 or oxidation of CH3OH, it may be the main intermediate of photocatalysis of CO2 + CH3OH. The ambient conditions play a major role in the mineralization of photocatalytic products. Based on the proposed general mechanism which involves all the reductants, photocatalysis of CO2 follows a complicated mechanism and gives various products. Although, it is likely that some of these products are not actually be formed or detected due to their small amounts.
References [1] S.N. Habisreutinger, L. Schmidt-Mende, J.K. Stolarczyk, Angew. Chem. Int. Ed. 52 (2013) 7372–7408. [2] M. Tahir, N.A. Saidina Amin, Energy Conve. Manage. 76 (2013) 194–214. [3] P. Pathak, M.J. Meziani, Y. Li, L.T. Cureton, Y.P. Sun, Chem. Commun. 10 (2004) 1234–1235. [4] M.F. Ehsan, T. He, Appl. Catal. B 166-167 (2015) 345–352. [5] P.Q. Wang, Y. Bai, P.Y. Luo, J.Y. Liu, Catal. Commun. 38 (2013) 82–85. [6] T. Baran, S. Wojtyła, A. Dibenedetto, M. Arestac, W. Macyk, Appl. Catal. B 178 (2015) 170–176. [7] G. Liu, N. Hoivik, K. Wang, H. Jakobsen, Sol. Energy Mater. Sol. Cells 105 (2012) 53–68. [8] S. Das, W.M.A. Wan Daud, Ren. Sustain. Energy Rev. 39 (2014) 765–805. [9] Y. Izumi, Coord. Chem. Rev. 257 (2013) 171–186. [10] G. Palmisano, V. Augugliaro, M. Pagliarob, L. Palmisano, Chem. Commun (2007) 3425–3437. [11] S.H. Szczepankiewicz, A.J. Colussi, M.R. Hoffmann, J. Phys. Chem. B 104 (2000) 9842–9850. [12] L. Zhang, H.H. Mohamed, R. Dillerta, D. Bahnemann, J. Photochem. Photobiol. C 13 (2012) 263–276. [13] A. Fujishima, X. Zhang, D.A. Tryk, Surf. Sci. Rep. 63 (2008) 515–582. [14] K. Li, X. An, K.H. Park, M. Khraisheh, J. Tang, Catal. Today 224 (2014) 3–12. [15] Q. Xiang, J. Yu, P.K. Wong, J. Colloid Interface Sci. 357 (2011) 163–167. [16] T. Hirakawa, K. Yawata, Y. Nosaka, Appl. Catal. A 325 (2007) 105–111. [17] Y. Nosaka, T. Daimon, A.Y. Nosaka, Y. Murakami, Phys. Chem. Chem. Phys. 6 (2004) 2917–2918.
E. Karamian, S. Sharifnia / Journal of CO2 Utilization 16 (2016) 194–203 [18] V. Augugliaro, M. Bellardita, V. Loddo, G. Palmisano, L. Palmisano, S. Yurdakal, J. Photochem. Photobiol. C 13 (2012) 224–245. [19] J. Shang, Y. Du, Z. Xu, Chemosphere 46 (2002) 93–99. [20] K.P. Yu, G.W.M. Lee, W.M. Huang, C. Wu, S. Yang, Atmos. Environ. 40 (2006) 375–385. [21] D.S. Muggli, S.A. Keyser, J.L. Falconer, Catal. Lett. 55 (1998) 129–135. [22] I. Ganesh, Ren. Sustain Energy Rev. 31 (2014) 221–257. [23] Y. Jean, F. Volatron, An Introduction to Molecular Orbitals, first ed., Oxford University Press, New York, NY, USA, 1993. [24] R.N. Compton, P.W. Reinhardt, C.D. Cooper, J. Chem. Phys. 63 (1975) 3821–3827. [25] L. Yuan, Y.J. Xu, Appl. Surf. Sci. 342 (2015) 154–167. [26] H. He, P. Zapol, L.A. Curtiss, Energy Environ. Sci. 5 (2012) 6196–6205. [27] K. Kocí, K. Mateju, L. Obalová, S. Krejcíková, Z. Lacny, D. Plachá, L. Capek, A. Hospodková, O. Solcová, Appl. Catal. B 96 (2010) 239–244. [28] Z. Yong, V. Mata, A.E. Rodrigues, Sep. Pur. Technology 26 (2002) 195–205. [29] W.G. Tu, Y. Zhou, Z.G. Zou, Adv. Func. Mater. 23 (2013) 4996–5008. [30] S. Yoshida, Y. Kohno, Catal. Surv. Jpn. 4 (2000) 107–114. [31] Y. Kohno, H. Ishikawa, T. Tanaka, T. Funabiki, S. Yoshida, Phys. Chem. Chem. Phys. 3 (2001) 1108–1113. [32] H. Tsuneoka, K. Teramura, T. Shishido, T. Tanaka, J. Phys. Chem. C 114 (2010) 8892–8898. [33] A. Corma, H. Garcia, J. Catal. 308 (2013) 168–175. [34] L. Liu, Y. Li, Aerosol Air Qual. Res. 14 (2014) 453–469. [35] C.C. Yang, J. Vernimmen, V. Meynen, P. Cool, G. Mul, J. Catal. 284 (2011) 1–8. [36] J. Mao, T. Peng, X. Zhang, K. Li, L. Zan, Catal. Commun. 28 (2012) 38–41. [37] T. Inoue, A. Fujishima, S. Konishi, K. Honda, Nature 277 (1979) 637–638. [38] K. Maeda, J. Photochem. Photobiol. C 12 (2011) 237–268. [39] R. Abe, J. Photochem. Photobiol. C 11 (2010) 179–209. [40] V.P. Indrakanti, J.D. Kubickib, H.H. Schobert, Energy Environ. Sci. 2 (2009) 745–758. [41] P. Akhter, M. Hussain, G. Saracco, N. Russo, Fuel 149 (2015) 55–65. [42] M. Subrahmanyama, S. Kaneco, N. Alonso-Vante, Appl. Catal. B 23 (1999) 169–174. [43] N. Sasirekha, S.J. Sardhar Basha, K. Shanthi, Appl. Catal. B 62 (2006) 169–180. [44] Z. Zhao, J. Fan, M. Xie, Z. Wang, J. Clean. Prod. 17 (2009) 1025–1029. [45] X.H. Xia, Z.J. Jia, Y. Yu, Y. Liang, Z. Wang, L.L. Ma, Carbon 45 (2007) 717–721. [46] J. Wu, C.H. Chiou, Catal. Commun. 9 (2008) 2073–2076. [47] K. Ko9 cí, M. Reli, O. Kozák, Z. Lacný, D. Plachá, P. Praus, L. Obalová, Catal. Today 176 (2011) 212–214. [48] C. Wang, R.L. Thompson, J. Baltrus, C. Matranga, J. Phys. Chem. Lett. 1 (2010) 48–53. [49] C. Wang, R.L. Thompson, P. Ohodnicki, J. Baltrus, C. Matranga, J. Mater. Chem. 21 (2011) 13452–13457. [50] I.H. Tseng, J.C.S. Wu, H.Y. Chou, J. Catal. 221 (2004) 432–440. [51] X. Li, H. Pan, W. Li, Z. Zhuang, Appl. Catal. A 413 (2012) 103–108. [52] S.S. Tan, L. Zou, E. Hu, Catal.Today 115 (2006) 269–273. [53] S. Kaneco, H. Kurimoto, K. Ohta, T. Mizuno, A. Saji, J. Photochem. Photobiol. A 109 (1997) 59–63. [54] T.V. Nguyen, J.C.S. Wu, Sol. Energy Mater. Sol. Cells 92 (2008) 864–872. [55] K. Ko9cí, L. Obalová, O. Šolcová, Chem. Process Eng. 31 (2010) 395–407. [56] I.H. Tseng, W.C. Chang, J.C.S. Wu, Appl. Catal. B 37 (2002) 37–48. [57] N.M. Dimitrijevic, B.K. Vijayan, O.G. Poluektov, T. Rajh, K.A. Gray, H. He, P. Zapol, J. Am. Chem. Soc. 133 (2011) 3964–3971. [58] Q.D. Truong, J.Y. Liu, C.C. Chung, Y.C. Ling, Catal. Commun. 19 (2012) 85–89. [59] K. Chandrasekaran, J.K. Thomas, Chem. Phys. Lett. 99 (1983) 7–10. [60] S. Sato, T. Arai, T. Morikawa, K. Uemura, T.M. Suzuki, H. Tanaka, T. Kajino, J. Am. Chem. Soc. 133 (2011) 15240–15243. [61] M. Abou Asi, C. Hea, M. Su, D. Xia, L. Lin, H. Denga, Y. Xiong, R. Qiua, X.Z. Li, Catal. Today 175 (2011) 256–263. [62] B. Michalkiewicz, J. Majewska, G.K. dziołka, K. Bubacz, S. Mozia, A.W. Morawski, J. CO2 Util. 5 (2014) 47–52. [63] G.R. Dey, A.D. Belapurkar, K. Kishore, J. Photochem. Photobiol. A 163 (2004) 503–508. [64] Z. Xiong, Y. Zhao, J. Zhang, C. Zheng, Fuel Process. Technol. 135 (2015) 6–13. [65] G. Guan, T. Kida, T. Harada, M. Isayama, A. Yoshida, Appl. Catal. A 249 (2003) 11–18. [66] S. Xia, Y. Meng, X. Zhou, J. Xue, G. Pan, Z. Ni, Appl. Catal. B 187 (2016) 122–133. [67] M. Ni, M.K.H. Leung, D.Y.C. Leung, K. Sumathy, Ren. Sustain Energy Rev. 11 (2007) 401–425. [68] I.A. Shkrob, T.W. Marin, H. He, P. Zapol, J. Phys. Chem. C 116 (2012) 9450–9460.
203
[69] I.A. Shkrob, N.M. Dimitrijevic, T.W. Marin, H. He, P. Zapol, J. Phys. Chem. C 116 (2012) 9461–9471. [70] S. Matsuoka, T. Kohzuki, C. Pac, A. Ishida, S. Takamuku, M. Kusaba, N. Nakashima, S. Yanagida, J. Phys. Chem. 96 (1992) 4437–4442. [71] P. Johne, H. Kisch, J. Phototchem. Photobiol. A 111 (1997) 223–228. [72] H. Fujiwara, H. Hosokawa, K. Murakoshi, Y. Wada, S. Yanagida, J. Phys. Chem. B 101 (1997) 8270–8278. [73] S. Kaneco, Y. Shimizu, K. Ohta, T. Mizuno, J. Phototchem. Photobiol. A 115 (1998) 223–226. [74] R.K. de-Richter, T. Ming, S. Caillol, Ren. Sustain. Energy Rev. 19 (2013) 82–106. [75] O. Ola, M.M. Maroto-Valer, J. Photochem. Photobiol. C 24 (2015) 16–42. [76] D. Luo, Y. Bi, W. Kan, N. Zhang, S. Hong, J. Mol. Struct. 994 (2011) 325–331. [77] F. Saladin, L. Forss, I.J. Kamber, Chem. Soc.Chem. Commun. 5 (1995) 533–534. [78] S.K. Parayil, A. Razzaq, S.M. Park, H.R. Kim, C.A. Grimes, S.I. In, Appl. Catal. A 498 (2015) 205–213. [79] H. Yamashita, H. Nishiguchi, N. Kamada, M. Anpo, Res. Chem. Intermed. 20 (1994) 815–823. [80] D. Uner, M.M. Oymak, Catal. Today 181 (2012) 82–88. [81] G. Marcì, E.I. García-López, L. Palmisano, Catal. Commun. 53 (2014) 38–41. [82] M. Hussain, P. Akhter, N. Russo, G. Saracco, Chem. Eng. J. 278 (2015) 279–292. [83] P. Usubharatana, D. McMartin, A. Veawab, P. Tontiwachwuthikul, Ind. Eng. Chem. Res. 45 (2006) 2558–2568. [84] G. Mahmodi, S. Sharifnia, M. Madani, V. Vatanpour, Sol. Energy 97 (2013) 186–194. [85] O.K.Varghese,M.Paulose,T.J.LaTempa,C.A.Grimes,NanoLett.9(2009)731–737. [86] A. Ahmad Beigi, S. Fatemi, Z. Salehi, J. CO2 Util. 7 (2014) 23–29. [87] S.S. Tan, L. Zou, E. Hu, Catal. Today 131 (2008) 125–129. [88] J.C.S. Wu, H.M. Lin, Int. J. Photoenergy 7 (2005) 115–119. [89] G.R. Dey, K.K. Pushpa, Res. Chem. Intermed. 32 (2006) 725–736. [90] S.C. Roy, O.K. Varghese, M. Paulose, C.A. Grimes, ACS Nano. 4 (2010) 1259–1278. [91] M. Tahir, N.A. Saidina Amin, Appl. Catal. A 493 (2015) 90–102. [92] C.C. Lo, C.H. Hung, C.S. Yuan, Y.L. Hung, Chin. J. Catal. 28 (2007) 528–534. [93] K. Teramura, T. Tanaka, H. Ishikawa, Y. Kohno, T. Funabiki, J. Phys. Chem. B 108 (2004) 346–354. [94] C.C. Lo, C.H. Hung, C.S. Yuan, J.F. Wu, Sol. Energy Mater. Sol. Cells 91 (2007) 1765–1774. [95] K. Teramura, S.I. Okuoka, H. Tsuneoka, T. Shishido, T. Tanaka, Appl. Catal. B 96 (2010) 565–568. [96] F. Sastre, A.V. Puga, L. Liu, A. Corma, H. García, J. Am. Chem. Soc. 136 (2014) 6798–6801. [97] P.J. Lunde, Ind. Eng. Chem. Process Des. Dev. 13 (1974) 226–233. [98] J. Albero, H. Garcia, A. Corma, Top. Catal. 59 (2016) 787–791. [99] L.B. Hoch, T.E. Wood, P.G. O’Brien, K. Liao, L.M. Reyes, C.A. Mims, G.A. Ozin, Adv. Sci. 1 (2014) 1400010–1400013. [100] J. Berkowitz, G.B. Ellison, D. Gutman, J. Phys. Chem. B 98 (1994) 2744–2765. [101] M. Tahir, B. Tahir, N.A. Saidina Amin, Mater. Res. Bull. 63 (2015) 13–23. [102] S. Delavari, N.A. Saidina Amin, M. Ghaed, J. Clean. Prod. 111 (2016) 143–154. [103] N. Yazdanpour, S. Sharifnia, Sol. Energy Mater. Sol. Cells 118 (2013) 1–8. [104] M. Torabi Merajina, S. Sharifnia, S.N. Hosseini, N. Yazdanpour, J. Taiwan Inst. Chem. Eng. 44 (2013) 239–246. [105] A. Yarahmadi, S. Sharifnia, Dyes Pigm. 107 (2014) 140–145. [106] D. Shi, Y. Feng, S. Zhong, Catal. Today 98 (2004) 505–509. [107] L. Yuliati, H. Itoh, H. Yoshida, Chem. Phys. Lett. 452 (2008) 178–182. [108] S. Qin, F. Xin, Y. Liu, X. Yin, W. Ma, J. Colloid Interface Sci. 356 (2011) 257–261. [109] J. Chen, F. Xin, S. Qin, X. Yin, Chem. Eng. J. 230 (2013) 506–512. [110] G.R. Dey, K.K. Pushpa, Res. Chem. Intermed. 33 (2007) 631–644. [111] G.R. Dey, J. Natural Gas Chem. 16 (2007) 217–228. [112] B.J. Liu, T. Torimoto, H. Matsumoto, H. Yoneyama, J. Photochem. Photobiol. A 108 (1997) 187–192. [113] G. Song, F. Xin, J. Chen, X. Yin, Appl. Catal. A 473 (2014) 90–95. [114] S. Xie, Y. Wang, Q. Zhang, W. Fan, W. Deng, Y. Wang, Chem. Commun. 49 (2013) 2451–2453. [115] L. Liu, C. Zhao, D. Pitts, H. Zhao, Y. Li, Catal. Sci. Technol. 4 (2014) 1539–1546. [116] H. Park, J.H. Choi, K.M. Choi, D.K. Lee, J.K. Kang, J. Mater. Chem. 22 (2012) 5304–5307. [117] X. Li, W. Li, Z. Zhuang, Y. Zhong, Q. Li, L. Wang, J. Phys. Chem. C 116 (2012) 16047–16053. [118] C. Deleitenburg, A. Trovarelli, J. Catal. 156 (1995) 171–174. [119] T.V. Nguyen, J.C.S. Wu, Appl. Catal. A 335 (2008) 112–120. [120] A. Danon, P.C. Stair, E. Weitz, Catal. Lett. 141 (2011) 11057–11066.