Integrated nano-architectured photocatalysts for photochemical CO2 reduction

Subhash Chandra Shit a, Indrajit Shown *bc, Ratul Paul a, Kuei-Hsien Chen bd, John Mondal *a and Li-Chyong Chen *de
aCatalysis & Fine Chemicals Division, CSIR-Indian Institute of Chemical Technology, Uppal Road, Hyderabad 500007, India. E-mail:;
bInstitute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan
cAmrita Centre for Nanosciences and Molecular Medicine, Amrita Vishwa Vidyapeetham, Kochi-682041, India. E-mail:
dCenter for Condensed Matter Sciences, National Taiwan University, Taipei 10617, Taiwan
eCenter of Atomic Initiative for New Materials, National Taiwan University, Taipei 10617, Taiwan. E-mail:

Received 10th August 2020 , Accepted 13th October 2020

First published on 15th October 2020

Recent advances in nanotechnology, especially the development of integrated nanostructured materials, have offered unprecedented opportunities for photocatalytic CO2 reduction. Compared to bulk semiconductor photocatalysts, most of these nanostructured photocatalysts offer at least one advantage in areas such as photogenerated carrier kinetics, light absorption, and active surface area, supporting improved photochemical reaction efficiencies. In this review, we briefly cover the cutting-edge research activities in the area of integrated nanostructured catalysts for photochemical CO2 reduction, including aqueous and gas-phase reactions. Primarily explored are the basic principles of tailor-made nanostructured composite photocatalysts and how nanostructuring influences photochemical performance. Specifically, we summarize the recent developments related to integrated nanostructured materials for photocatalytic CO2 reduction, mainly in the following five categories: carbon-based nano-architectures, metal–organic frameworks, covalent-organic frameworks, conjugated porous polymers, and layered double hydroxide-based inorganic hybrids. Besides the technical aspects of nanostructure-enhanced catalytic performance in photochemical CO2 reduction, some future research trends and promising strategies are addressed.

1. Introduction

Anthropogenic carbon dioxide (CO2) is one of the major sources accounting for more than 60% of the greenhouse gases related to global warming. Burning fossil fuels and the huge rate of deforestation have remarkably increased the atmospheric CO2 level.1–3 In addition to control of CO2 production, materials scientists are developing more aggressive strategies such as CO2 capture and storage (CCS) and utilization of CO2.4–6 In this context, it is practically impossible to prevent CO2 emissions owing to recent human lifestyles and enormous combustion of fossil fuels. In fact, the CCS strategy is limited due to the danger of leakage along with requirement of energy in fuel compression and transportation. In order to overcome the pitfalls of previous strategies, new strategies are utilization and conversion of CO2 to useful chemicals or fuels, for which the methods include biological,7 thermochemical,8–10 electrochemical,11,12 photochemical,13,14 photoelectrochemical15 and photothermal16,17 reduction processes. The major advantages and disadvantages of these processes are summarized in Table 1. Among them, solar light-driven photochemical CO2 reduction has received great attention as a clean sustainable process with valuable chemicals like CH4, CH3OH, HCOOH etc.18,19 The solar light-driven photochemical CO2 reduction process, or artificial photosynthesis, which mimics natural photosynthesis, is essentially an ideal clean renewable process to solve energy and environmental problems.
Table 1 Advantages and disadvantages of the processes used for CO2 reduction
Name of process Advantages Disadvantages
Electrochemical • Operates under moderate electrolyzer conditions • High overpotential and low energy efficiency
• Opportunities to use industrial or municipal wastewater as an electrolyte • Hydrogen evolution reaction (HER) affects the product selectivity
• Promising faradaic efficiency broadens the scope for practical applications • CO2 solubility issues in electrolyte
• High product selectivity • High cost and low stability of metal-based electrocatalysts
Photochemical • Solar-energy-driven environmentally friendly process • Slurry-based process suffers due to the HER reaction
• Operates under ambient conditions • Low photochemical quantum efficiency
• Maintains the carbon cycle • Complicated reaction mechanism
• Low-cost photocatalyst materials • Faster charge carrier recombination
• HER reaction is suppressed in the gas-phase process
Biological • Enzyme- and microorganism-based eco-friendly and green process • Higher cost of enzymes cripples the potential for industrial applications
• Provides highly stereo-specific and regio-/chemo-selective products • Reactivity depends on cofactor reactions
• Low energy consumption process • Slow process and enzyme-specific product selectivity
• Easy to scale up • Large reactor size is required
Photoelectrochemical (PEC) • External electrical bias improves photogenerated charge carrier separation • High overpotential
• Isolated redox reactions and less partial oxidation • The HER occurs at the expense of CO2 reduction
• High overall quantum efficiency can be achieved of around 8–12% • Often requires a co-catalyst, which is favorable for the HER but suppresses CO2 reduction
• Applied external bias affects product selectivity • Photocorrosion of the electrode materials
• Kinetic limitations
Thermochemical • Mature process and high efficiency • High-temperature process
• Most important method used in industry • Competitive and adverse thermodynamic and kinetic aspects suppress the conversion and selectivity
• Catalysts prepared via cost-effective synthetic methods • Product formation depends on the nature of the catalyst
• Photothermic effect causes heat generation, which enhances the overall conversion • Complicated reaction mechanism
Photothermal • More ecofriendly as compared with thermal techniques • Understanding this reaction mechanism is very challenging
• Comparatively less energy input is required • Product selectivity is poor

The enormous interest in the photocatalysis research field has grown significantly after the pioneering discovery of Fujishima and Honda in 1972.20 Afterwards, various materials were developed for CO2 photoreduction, including metal oxides and non-oxides,21–23 metal sulfides,24,25 metal-free semiconductors,26 molecular catalysts,27 and other diverse catalysts.28–30 According to the Web of Science database in the area of photocatalytic CO2 reduction, China, Japan, Taiwan and other Asian countries together contributed more than 50% of significant footprint as compared with the rest of the world as shown in Fig. 1. Moreover, due to the geographical location of Asian countries, India and Taiwan have abundant solar light; therefore, the development of solar light-based photochemical CO2 conversion to solar fuels is a promising process to reduce their future carbon emissions.

image file: d0nr05884j-f1.tif
Fig. 1 Contribution by country to CO2 photocatalytic reduction research (Web of Science, statistics to 2019).

Photochemical CO2 reduction to solar fuels production is one of the promising technologies that could potentially provide clean energy as well as CO2 reduction. In this process, CO2 is directly reduced through multi-electron photochemical reaction by using homogeneous and heterogeneous catalysis. In homogeneous catalysis, the metal complexes are usually acting as both light harvesting antennas and catalysts. Various research groups have efficiently worked on homogeneous photocatalytic CO2 reduction by using combination of organometallic photosensitizers and semiconductor catalysts. Those reported systems based on noble- and/or rare-metal complexes provided highly efficient and durable photocatalysts. However, large-volume CO2 reduction with rare-earth metal or noble metal catalyst systems is not feasible for practical demands. In the past few years, molecular photocatalytic systems of mainly earth-abundant metal macrocycles (Co(II), Ni(II), Cu(I), Fe(II), Mn(II), Zn(II) macrocycles, etc.)31–33 have progressively been used for homogeneous photocatalytic CO2 reduction to selective products. These catalyst systems provided an efficiency around 4 to 30% with a reasonable turnover number (TON) of 10 to 150.34 Recently Ishitani and coworkers have been developing Cu(I) photosensitizer and Mn(I) catalyst systems for photocatalytic CO2 reduction with high efficiency (around 57%) and high TON. Although these molecular catalyst systems provide very promising and high photocatalytic activity, this kind of homogeneous catalyst has several disadvantages as compared with large-scale heterogeneous catalysis. In homogeneous catalysis, the catalysts, reactants and products are of the same phase; therefore all catalytic sites are available for reaction and provide high selectivity. However, the products are difficult to separate. Large amounts of organic solvent used and waste materials produced are big drawbacks for large-scale industrial processes. In contrast, for heterogeneous catalysis, the catalysts, reactants and products are in different phases and are easy to separate and re-use, which is more amenable for industrial scale-up implementation. To take major advantage of heterogeneous catalysis, many works focused on inorganic semiconductor systems taking advantage of solid–liquid or solid–gas interface photocatalytic CO2 reduction.35

Photocatalytic CO2 reduction is a complex process and produces high-energy carbon compounds. In this process, photocatalysts are mainly semiconductor materials, which absorb photons from light based on their band gap, separate the photogenerated charges (e and h+) and transport them to the surface with active sites for the catalytic reaction. These reaction sites are mainly located either directly on the surface of the semiconductor or indirectly at the interface between the semiconductor and co-catalyst. Successful donation of the photoinduced carriers to the accepter molecules at the surface, which leads to reduction and oxidation reactions, competes with various charge carrier recombination processes. In the surface recombination, carriers can recombine with their counterparts of opposite charge on the surface. In another way, recombination of two oppositely charge carriers can occur in the bulk of the photocatalyst, namely volume recombination.36 Key catalyst development strategies facilitate the separation of photoinduced carriers in the photocatalytic process, and afford maximum light harvesting from the solar spectrum. Various photocatalysts, including metal oxides, metal sulfides, nonmetals, composite photocatalysts, graphene-based photocatalysts etc., have been developed.

A significant number of review articles on evolving topics have focused on (1) design of promising photocatalysts;37–40 (2) importance of co-catalysts;41 and (3) analysis methods, reaction conditions, reactor design42etc. It has been shown that other than chemical doping with foreign atoms or semiconductor defect engineering, morphology control or nanostructuring are also effective methods to tailor electronic properties, to manipulate the surface chemistry, and to modify the charge recombination of semiconductor photocatalysts. Therefore, morphology control or nanostructuring in semiconductor photocatalysts is explored in photocatalytic CO2 reduction. Moreover, over the past few years there were many reviews selectively focused on special kinds of nanostructured materials such as metal–organic framework (MOF),43–45 covalent organic framework (COF),46 metal free47 and g-C3N4 based48 materials for photocatalytic CO2 reduction.

Although nanostructured photocatalyst materials have been selectively explored, the specific importance and characteristics of aforementioned materials stimulated us to explore all types of nano-architectured catalysts comprehensively for photochemical CO2 reduction in both liquid- and gas-phase reactions. Here in this review, we will focus on the following five categories of materials and their development strategies. First, we point out improvement of CO2 reduction employing metal oxide-based photocatalysts, which reveals strategies like surface defects, along with Z-scheme heterostructuring and controlling nanostructure within the atomic-scale range. Second, enhancement of photocatalytic activity and stability of g-C3N4 and GO based materials is discussed through composite formation with metals, metal halides, metal oxides and metal complexes. Third, we explored MOF-based photocatalysts containing different metals (Ti, Zr, Fe, Cu, Ru, Co and Mg) for CO2 reduction and their specific mechanism. Fourth, up-to-date COF-based materials for reduction activity are also mentioned. Fifth, recent developments of conjugated porous polymer (CPP)- and layered double hydroxide (LDH)-based inorganic-hybrid photocatalyst systems towards visible light-driven reactions have been specified. Lastly, we mention the advantages and disadvantages of all these types of aforementioned nanostructured photocatalysts. Moreover, we provide highlights of the most recent rational catalyst design and structure engineering with the aim of gaining a fundamental understanding of the photochemical CO2 reduction process.

2. Photocatalytic CO2 reduction

Current photochemical CO2 reduction, which is mainly based on the development of various homogeneous and heterogeneous photocatalysts over the past 30 years, is encountering different kinds of challenges. Although photocatalytic CO2 reduction is one of the emerging technologies to solve energy and environmental issues, the low quantum efficiency and product selectivity are the major barriers for large-scale industrial application. These problems can be outlined more fully as follows. (1) Light absorption: The majority of oxide-based semiconductor photocatalysts are of wide band gap. Therefore, they only absorb UV light and cannot utilize the maximum energy from the visible solar spectrum. (2) Gas adsorption: The linear CO2 molecule needs to be adsorbed on the heterogeneous catalyst surface and start the charge transfer, and initiates the surface reaction, leading to further conversion to bent CO2. (3) Charge recombination: The photogenerated electrons and holes are recombined due to the inherent bulk and surface recombination process of semiconductors. The fast charge recombination process is a major demerit for semiconductor photocatalysts. (4) The multielectron redox reaction provides several products and leads to poor product selectivity during photocatalytic CO2 reduction. Meanwhile, it is unfortunate that no single photocatalyst satisfies all the aforesaid requirements. In order to meet the requirements, scientists are now looking to developing suitable photocatalysts for CO2 reduction. Generally, the following characteristics for photocatalysts are essential requirements for most favorable CO2 photoreduction: (i) the valence band should be more positive for the holes acting as an electron acceptor; (ii) the electrons reside in the conduction band (CB) that should be sufficiently negative with respect to onset reduction potential of CO2; (iii) the adsorption capability of reactants like CO2 or CO32− onto the photocatalyst should be high; (iv) no toxic byproduct should be evolved during the photochemical reaction; (v) the required catalyst should possess adequate stability.49,50 In photochemical processes, mainly CO2 conversion into liquid or gaseous hydrocarbons by homogeneous or heterogeneous photocatalysis, reduction of CO2 has been found to occur through multielectron pathways in gaseous and liquid phases. The reduction products include carbon monoxide, formic acid or formate, oxalic acid or oxalate through 2-electron reactions, formaldehyde through 4-electron reactions, methanol through 6-electron reactions, methane through 8-electron reactions, and ethylene and ethanol through 12-electron reactions, as shown in Table 2.42 Moreover, CO2 reduction driven by solar energy can be performed in several ways. In search of the “Holy Grail” of sustainability, while the photochemical approach, namely so-called artificial photosynthesis, is showing some promise, considering the various constraints in thermodynamics for multielectron CO2 reduction reactions, the photoelectrochemical (PEC) approach is a definitive objective for a manageable, inexhaustible and green technology for CO2 reduction.51 In the PEC process, CO, CH4 or other hydrocarbons are produced from solar light-driven CO2 reduction using a semiconductor photocathode, which directly reduces CO2 molecules into hydrocarbons. While at the same time a separate semiconductor photoanode oxidizes water molecules to oxygen via photo-oxidation reaction. Although PEC and PC processes depend on the semiconductor materials, dissimilar from PC, which does not need any external electrical energy input, a PEC system mainly consists of the electrolyte, a working electrode and a counter electrode, typically configured with a pair of p-type photocathode/anode or cathode/n-type photoanode or p-type photocathode/n-type photoanode as working and counter electrode, respectively. Usually, to construct the photocathode and photoanode for a PEC cell, different types of semiconductor (n- and p-type) materials will be selected based on the corresponding redox reactions.52–54 In heterogeneous catalysis, both aqueous and nonaqueous electrolytes are used in PEC. However, a nonaqueous system possesses better solubility of CO2 than an aqueous system. Though positive thermodynamic and kinetic factors influence PEC for CO2 reduction reaction, nevertheless, a few limiting factors like inevitably competing with the hydrogen evolution reaction and intermediate products restrict it for wider practical application.
Table 2 Multielectron reactions of carbon dioxide and the thermodynamic redox potentials
Reaction E° (V vs. NHE, pH = 7)
CO2 + 2H+ + 2e → CO + H2O −0.51
CO2 + 2H+ + 2e → HCOOH −0.58
CO2 + 4H+ + 4e → HCHO + H2O −0.55
CO2 + 6H+ + 6e → CH3OH + H2O −0.39
CO2 + 8H+ + 8e → CH4 + 2 H2O −0.24
2CO2 + 8H+ + 8e → CH3COOH + H2O −0.31
2CO2 + 10H+ + 10e → CH3CHO + 3H2O −0.36
2CO2 + 12H+ + 12e → C2H5OH + 3H2O −0.33
2CO2 + 14H+ + 14e → C2H6 + 4H2O −0.27
3CO2 + 16H+ + 16e → CH3CH2CHO + 5H2O −0.32
3CO2 + 16H+ + 16e → CH3COCH3 + 5H2O −0.31

In the development of heterogeneous photocatalytic systems, metal or metal oxide nanoparticles together with semiconductors are utilized as co-catalysts. In the photocatalytic CO2 reduction process, the photocatalyst should fulfill several important criteria: the semiconductor system must contain a minimum band gap that is sufficient to absorb solar light, the valence band edge position must cover the oxygen redox potential, whereas the position of the conduction band edge should be sufficiently negative to cover the onset reduction potential of CO2 to various hydrocarbons. Additionally, the photogenerated charge carrier recombination rate must be low. Thermodynamically, the CO2 molecule is of linear geometry and very stable with a closed-shell electronic configuration. In the presence of single electrons on the semiconductor surface, its symmetry is lost, accompanied by conversion to a bent structure with C2V symmetry due to the increase in repulsive force between the electrons on the central carbon atom and free electron pairs on the covalently attached oxygen atoms. The single electron reduction of CO2 into CO2˙ radical requires a strong negative potential around −1.9 V (vs. NHE), which is unfavorable. Whereas proton-assisted multielectron CO2 reduction to hydrocarbons is more favorable due to less negative potential required for molecular species production through a radical reaction mechanism in the presence of protons. The prospect of a photocatalyst system (semiconductors) is highly influenced by the band structure configuration of the semiconductor catalyst and redox potentials of the adsorbate on the photocatalyst surface.

3. Development of photocatalysts

A vast range of semiconductor photocatalysts with diverse characteristics, such as metal oxides, metal sulfides, carbonaceous hybrids together with various dopants or nanostructure support materials, have been explored for photochemical CO2 reduction. Novel strategic initiatives in the development of photocatalysts to utilize the maximum amount of visible light and to reduce charge recombination rate are desirable.55 Pure metal oxides such as TiO2 are widely studied as cheap and stable photocatalysts to a large extent.56–58 However, the first-generation crystalline photocatalyst systems based on pure metal oxides suffered due to their wide band gaps and limited light harvesting characteristics. In the second generation, metal oxide semiconductors are modified by sensitization and doping methods to help to control the charge recombination and to facilitate the response towards visible light.59–62 Metal and nonmetal doping mainly creates inter-band state and modulates the band structures, which helps charge transfer with adsorbed CO2 molecules. Meanwhile, several metal sulfide-based narrow band gap semiconductors have been explored due to more negative shift of the S 3p orbitals as compared with O 2p of oxides.63 Although these metal sulfides, nitrides and oxysulfides are promising, their overall photocatalytic efficiency still needs to be enhanced in order to meet industrial requirements. On the other hand, due to the emerging effort of nanotechnology, nanomaterials have significantly contributed in the construction of light harvesting semiconductor nanostructures. Therefore, in the third-generation semiconductor photocatalysts for CO2 reduction, several nanostructured semiconductors, heterostructures with size-dependent properties and quantized charging phenomena, open up new opportunities. Nanoarchitectured semiconductor photocatalyst systems possessing various shapes and dimensions, ranging from particles, rods, prisms, wires, hollow tubes/spheres, 2D sheets to 3D objects, have been developed as new strategies for homogeneous and heterogeneous photocatalytic CO2 reduction. Since the scientific discovery of graphene and graphene oxide (GO), development of these novel materials and their hybrids can offer heterogeneous photocatalyst systems with a number of favorable characteristics, such as large surface area, ample active adsorption sites, and efficient charge separation.64 Additionally, more recent semiconductor structural engineering strategies include a range of nano-porous structures, shallow oxygen or sulfur defects and controlled crystal defects related to uncovered facets. These recent efforts on heterostructures together with macro-/meso-/microporous/nanostructural designs open up promising hybrid systems for fourth-generation photocatalysts for CO2 reduction. The most popularly studied homogeneous and heterogeneous photocatalyst systems are categorized as shown in Fig. 2. Due to the abovementioned merits, nanostructure semiconductor photocatalyst systems have been comprehensively explored in various photocatalysis processes. The timeline of development along this direction is schematically shown in Fig. 3.
image file: d0nr05884j-f2.tif
Fig. 2 Categorization of various materials for photocatalytic CO2 reduction.

image file: d0nr05884j-f3.tif
Fig. 3 A schematic representation of the timeline of the development of nanostructure photocatalyst systems for CO2 reduction to solar fuels.

4. Nanostructured photocatalysts

Among various metal oxides, titania (TiO2) is widely used as a semiconductor photocatalyst for CO2 reduction as a result of its excellent inherent attributes as well as easy preparation, eco-friendliness, controllable nanoscale structures, exceptionally high chemical and heat resistance, and superior optoelectronic and photophysical characteristics. However, there are challenges: besides the thermodynamically stable molecular configuration of CO2, many protons and electrons are involved in the multi-step photochemical CO2 reduction to hydrocarbons. The low active surface areas and poor photon energy absorption at visible wavelengths of TiO2-containing hybrid photocatalysts result in low photocatalytic CO2 conversion efficiency. Several attempts have been reported to improve these demerits via rational material design and surface modification. Additionally, mesopore- or/and macropore-containing bulk TiO2 has been developed in order to increase active surface areas. In the progressive development of visible-light semiconductor photocatalysts, diverse concepts have been reported for TiO2-based photocatalysts. Photocatalytic activity of anatase, rutile, and brookite titania nanocrystals for CO2 reduction with surface defect engineering has been studied.65,66 Liu et al. demonstrated that in comparison to the defect-free TiO2 surfaces, defective anatase and brookite TiO2 are highly effective for photochemical CO2 reduction to selective carbon monoxide and methane formation, respectively (Fig. 4a).65 The product selectivity from CO2 reduction is basically influenced by the various surface energies of the TiO2 crystallite faces and their adsorption and dissociation interaction with different intermediates. This work revealed new insights towards surface defects on TiO2 nanocrystal semiconductors for their future development and more insightful understanding of the photochemical CO2 reduction process.
image file: d0nr05884j-f4.tif
Fig. 4 Photocatalytic CO2 reduction with (a) H2O on TiO2 nanocrystals and (b) TiO2 single crystals and mesocrystals with {101} facets. Reprinted with permission from ref. 65 and 66. Copyright 2012 American Chemical Society.

Nanostructuring effects have been introduced to improve the photocatalytic CO2 reduction process, which have a different impact in semiconductor photocatalysis apart from conventional strategies like doping, heterostructuring and sensitization. Nanostructuring of semiconductors can provide the following advantages: size-dependent photon energy absorption, higher extinction coefficients, creating single photon induced multiple excitons, extended carrier lifetime, and improved photostability. Most of the reported TiO2-based photocatalyst systems mainly covered higher thermodynamically stable {101} facets rather than {001} facets. Jaroniec and coworkers reported that co-exposed {101} and {001} facets of anatase TiO2 can form a surface heterojunction within a single TiO2 particle, which is advantageous for the relocation of photon-induced charge carriers to {101} and {001} facets (Fig. 5).67 Moreover, a controlled optimum ratio of the exposed {101} and {001} facets enhanced the photocatalytic activity of anatase toward reduction of CO2 to selective CH4 formation (Fig. 5). This “surface heterojunction” concept provided a novel material design and fabrication tool for advanced development of semiconductor photocatalysts.

image file: d0nr05884j-f5.tif
Fig. 5 The photocatalytic CO2 reduction activity of anatase TiO2 with co-exposed {001} and {101} facets. Reprinted with permission from ref. 67. Copyright (2014) American Chemical Society.

The reactive facets on a semiconductor photocatalyst mainly enhance photogenerated charge carrier separation and reduce bulk recombination. Introducing nanotechnology to reduce the particle size of semiconductor photocatalysts is one of the conventional strategies to shorten the bulk diffusion length of photogenerated charge carriers. However, this nanoscale synthesis technique often causes serious agglomeration of synthesized nanoparticles, hence drastically compromising high surface area and increasing grain boundary recombination. Jiao et al. developed hollow anatase TiO2 single crystals and mesocrystals with dominant {101} facets using F/PO43− as a morphology controlling co-mediator. The combined photophysical properties based on UV–visible absorption spectra and valence band spectra of the hollow mesocrystals have revealed higher conduction and valence band edges than the hollow single crystals as shown in Fig. 4b.66 In various photocatalytic reactions like hydrogen and oxygen evolutions from water splitting, and photoreduction of CO2, the hollow crystals showed superior photocatalytic activities. The hollow mesocrystals with high surface area revealed the synergistic effects of shortened bulk diffusion length of photogenerated carriers for reduced bulk recombination.

Moreover in the last several years, black TiO2 with some surface defects has been reported as a promising photocatalyst for various photochemical processes. Among the efforts in this direction, Billo et al. reported a combined experimental and computational study on Ni-nanocluster-loaded black TiO2 (Ni/TiO2[VO]) with built-in dual active sites for selective photocatalytic CO2 conversion.68 That report reveals that the synergistic effects of intentionally induced Ni nanoclusters and oxygen vacancies in black TiO2 have several positive impacts on the photocatalyst for CO2 reduction. The synthesized Ni/TiO2[VO] significantly enhanced the light harvesting characteristics by lowering the band gap, helped the fast electron transfer process through a less resistive pathway for charge transfer between TiO2 and Ni nanoclusters, and provided highly reactive sites and energetically stable CO2 binding sites with low activation energy (0.08 eV). The Ni/TiO2[VO] photocatalyst has demonstrated highly selective and enhanced photocatalytic activity as compared with commercial TiO2 (P-25). This insightful study also well explored the details of interfacial charge transfer and product formation mechanisms.

Recently Sun et al. reported ultrathin WO3·0.33H2O nanotubes with a large amount of exposed Vo sites on the surface as an outstanding and stable photocatalyst for photochemical reduction of CO2 to CH3COOH in pure water under solar light.69 Interestingly, the overall catalytic behavior of the oxygen-deficient WO3·0.33H2O nanotubes showed promising stability under repeated photocatalytic reactions. Therefore, controlling nanostructure within the atomic scale range is an effective strategy to design future improved photocatalyst systems for CO2 activation. It should also be noted that most semiconductor nanomaterials are mainly composed of single crystalline phases without any grain boundaries and nearly defect free, which effectively decreases the volume recombination.

Apart from the nanostructuring, to further enhance the photocatalytic activity of WO3, Jin et al. developed a hierarchical Z-scheme CdS–WO3 heterostructure for enhanced photocatalytic CO2 activity.70 This Z-scheme heterostructuring could account for an efficient charge separation and exhibits two separate redox catalytic reactions (reduction and oxidation) in two different semiconductor surfaces to minimize the undesirable photo-oxidation reaction. In another approach to improve the photogenerated carrier recombination, Low et al. synthesized direct Z-scheme composite film of titanium dioxide and cadmium sulfide (TiO2/CdS) for photocatalytic CO2 reduction.71 The direct Z-scheme charge-carrier migration pathway in TiO2/CdS composite film was also confirmed by in situ irradiated X-ray photoelectron spectroscopy. In the last several years this direct Z-scheme approach has been frequently utilized to improve photocatalytic performances. In 2018, Xu et al. demonstrated TiO2 nanofiber-coated few-layer CuInS2 nanoplates for photochemical CO2 reduction. This heterostructure hybrid possesses improved surface area and interfacial charge separation by direct Z-scheme.72 The direct Z-scheme photoexcited carrier manipulation processes mainly improve the photocatalytic CO2 reduction ability of the reductive semiconductor surface by coupling with another oxidative semiconductor. Such a multicomponent photocatalyst offers an alternative approach together with nanostructuring to develop highly efficient photocatalytic CO2 reduction.

Tailored nanostructures with proper design are particularly helpful for improving the separation of electron–hole pairs by a direct electron transport pathway. In the past few years several metal oxide (binary and ternary) semiconductor nanostructure photocatalyst systems were reported to show promising photocatalytic CO2 reduction.73 Among them, zinc orthogermanate (Zn2GeO4) exhibited high potential for various applications, such as high-wavelength selectivity in UV photodetectors and photocatalytic water splitting. Zou et al. have reported single-crystalline Zn2GeO4 nanobelts, which were synthesized by a novel solvothermal process, to be a potential photocatalyst for photochemical CO2 reduction to selective methane formation.74 In another work they also prepared hexagonal prism Zn2GeO4 nanorods with different aspect ratios via a wet chemical process. It was also demonstrated that the few crystal defects, beneficial high surface area and rational nanostructure with microporosity of the 1D photocatalyst are the main revealing factors for improved photoreduction of CO2.75Fig. 6 shows ultrathin Zn2GeO4 nanoribbons and nanosheets and their comparative photochemical reactivity for CO2 reduction. Moreover, in the last several years, hierarchical nanostructures of metal oxide semiconductors and their hybrids have been investigated as heterogeneous photocatalysts and showed great potential for photochemical CO2 reduction. In general, nanoarchitectures improve comprehensive light absorption and photogenerated charge carrier separation in photocatalyst systems for excellent catalytic performance.

image file: d0nr05884j-f6.tif
Fig. 6 Ultrathin Zn2GeO4 nanoribbon (a, b) and nanosheet (c) photocatalyst systems for CO2 reduction. (d) Comparative methane formation from photocatalytic CO2 reduction for various Zn2GeO4. Adapted from ref. 74 and 75. Copyright (2010) American Chemical Society.

In recent years, several narrow band gap metal sulfides have been intensively promoted in photocatalysis as compared to oxides for the following reasons.76 Firstly, the conduction bands of sulfide semiconductors are composed of d and sp orbitals in general, while their valence bands consist of S 3p orbitals, which are much more negative than O 2p orbitals, thus resulting in the conduction band positions of sulfides being negative enough to cover the onset reduction potential of CO2. Secondly, the reduced band gaps of these sulfides are favorable for harvesting the solar spectrum. Moreover, the distinctive chemical and physical properties together with tunable electronic band structure of metal sulfides are advantageous for photocatalytic CO2 reduction. In this direction of research, Feng et al. developed metal-incorporated crystalline chalcogenides such as Au@GeZnS and Pd@GeZnS for photocatalytic CO2 reduction. These crystalline chalcogenides exhibit tunable band structure and enhanced photochemical CO2 reduction to selective CH4 formation.77 In 2017, Xie et al. developed one-unit-cell ZnIn2S4 layers with rich zinc vacancies for CO2 reduction to selective CO under visible light irradiation (Fig. 7). They observed that with increasing the number of Zn vacancies in the ZnIn2S4 layer the formation rate of CO increases around 3.6 times compared to ZnIn2S4 without Zn vacancies.78 Later on, they synthesized atomically thin layers of sulfur-deficient CuIn5S8 having charge-enriched Cu–In dual sites. In that study, they found that highly stable Cu–C–O–In intermediate formed at the Cu–In dual sites, which plays a significant role for the selective CO2 reduction to CH4 over CO.79 Similarly, Lou et al. developed a hierarchical In2S3–CdIn2S4 heterostructured nanotube via a self-templated strategy for photocatalytic CO2 reduction to CO. They demonstrated that the unique structural and compositional features like nanosized interfacial contacts in the hierarchical tubular nanostructures reduce the diffusion length for photogenerated charge-carrier separation and migration. Additionally the large surface area of the tubular In2S3–CdIn2S4 heterostructure (Fig. 7) exhibits improved CO2 adsorption, and consequently helps in the CO2 reduction to CO with high formation rate of around 825 μmol h−1 g−1.80 Similarly, branch-like ZnS-DETA/CdS hierarchical heterostructures developed by Ding et al. show deoxygenative CO2 reduction to CO with high generation rate. They found uniformly distributed heterojunctions in nanodomains and high surface area of ZnS-DETA/CdS hierarchical hybrid specifically enhance visible-light absorption and offer an ample amount of active sites for CO2 adsorption.81 More recently, Xiong et al. have explored spinel-type ternary mixed metal sulfide NiCo2S4 for deoxygenative reduction of CO2 to CO with visible light. They demonstrated that the spinel catalyst could improve separation and transport properties of photogenerated charge carriers, thus significantly improving the CO2 reduction reaction.82 Most recently, Wang et al. synthesized hierarchical Co3O4@CdIn2S4 p–n heterojunction photocatalysts with ultrathin CdIn2S4 nanosheets supported on Co3O4 nanocubes. This hierarchical Co3O4@CdIn2S4 heterostructure exhibited selectivity of around 93.3% for CO generation from photochemical CO2 reduction.83 Owning to the advantages of nanostructured metal sulfides like visible-light absorption, pre-designable crystal structure, tunable band structure, carrier separation and excellent stability, in the last several years these sulfide materials have received increasing attention and research interest for their tremendous potential applications in heterogeneous photochemical CO2 reduction. A wide range of hybrid nanostructures can be directly designed and facilely tuned by introducing selective counter semiconductor oxides or other functional building blocks.

image file: d0nr05884j-f7.tif
Fig. 7 Zn-Rich ZnIn2S4 unit-cell layers for photocatalytic CO2 reduction into CO. Adapted from ref. 80. Copyright (2017) American Chemical Society.

5. Carbon-based nanoarchitectures

5.1. Graphitic carbon nitride (g-C3N4)-based materials

Recently, graphitic carbon nitride (g-C3N4) has been explored as a stable and viable polymeric organic semiconductor photocatalyst for diverse photocatalytic applications.84 Interestingly, it shows an energy band gap in the visible-light range around 2.77 eV with CB and VB positions at −1.1 and 1.6 eV, respectively.85 The strong visible-light absorption of g-C3N4 is mainly associated with its smaller band gap (2.7 eV). Further, g-C3N4 has high resistance to attack from heat, strong acid, and strong alkaline solution.86 Synthesis of g-C3N4 and related photocatalysts can be performed easily by thermal polycondensation of various cheap N-rich precursors, such as dicyanamide, cyanamide, melamine, and urea,87 which makes it economically feasible as compared to metal-containing photocatalysts consisting of expensive precious metals. In recent years, g-C3N4 has stood out as an emerging catalytic material for water splitting, CO2 photoreduction, organic contaminant purification, catalytic organic synthesis, and fuel cells due to various outstanding properties.88 Despite its stronger absorption behavior in the visible-light region as well as outstanding photocatalytic properties, the overall photocatalytic efficiency of single-phase g-C3N4 is restricted by the high recombination rate of photogenerated electrons and holes. Therefore, it is needed to develop a highly efficient photocatalyst based on g-C3N4 by some new strategies to improve the photon-induced charge carrier transfer and separation process to boost the performance of g-C3N4-based photocatalysts.89–92 Following conventional photocatalyst development, homo-junction or hetero-junction between two semiconductors forming a hybrid photocatalyst system could be effective for significantly enhanced photocatalytic performance. Hence, composites based on C3N4 with metal halide, metal oxide, metal sulfide, and other complexes have been attempted, as exemplified below.
5.1.1. Composites with metal halides: surface plasmon resonance effects. Ong et al. developed graphitic carbon nitride/AgX hybrid photocatalysts by a sonochemical deposition–precipitation method at ambient condition where the light-sensitive silver halides AgX (X = Cl and Br) were decorated on the surface of protonated graphitic carbon nitride (pCN). AgX/pCN showed water vapor-assisted photocatalytic CO2 reduction to methane under visible light in ambient atmosphere. Among the AgX/pCN composite photocatalysts, 30AgBr/pCN revealed a selective CH4 formation of 10.92 μmol gcat.−1, which is around 34 and 4 times higher than those of pristine AgBr and pCN, respectively. In addition, AgBr/pCN showed the highest performance among the halide ion catalysts, and the photocatalytic activity was better (by a factor of 1.3) than that of the optimal AgCl/pCN. The photochemical reactivity was improved for two reasons: (1) the surface plasmon resonance effect from Ag and (2) the heterojunction formation in AgBr/pCN hybrid photocatalyst by a charge depletion region between pCN and AgBr, which mainly suppresses the photogenerated carrier recombination process by effective charge transfer and separation. The photogenerated charge separation in the reported composite system is more efficient as a result of alignment of the energy bands between pCN and AgBr to form a conventional staggered gap (Type II) heterojunction, as compared to a straddling gap (Type I) heterojunction in AgCl/pCN.93 The surface plasmon resonance generates “hot electrons” under visible-light absorption on the surface of the photocatalyst, which synergistically influences the light-driven multielectron CO2 reduction process.
5.1.2. Composites with metal oxides: heterojunctions and the Z-scheme. Construction of heterojunction composites by coupling the narrow band gap g-C3N4 with another wide band gap metal oxide semiconductor has been proven to be an efficient approach. He et al. developed a ZnO nanoparticle-functionalized g-C3N4 sheet (ZnO/g-C3N4) composite by the impregnation method for photochemical CO2 reduction. The ZnO/g-C3N4 composite shows improved photocatalytic CO2 reduction performance due to effective photogenerated charge separation at the charge depletion region between two semiconductor interfaces. Optimized ZnO/g-C3N4 composite revealed photocatalytic CO2 reduction to hydrocarbon formation rate around 45.6 μmol h−1 gcat−1, which is 4.9 and 6.4 times higher than that of g-C3N4 and P25, respectively.94 Li et al. used a hard-template route and prepared mesoporous CeO2/graphite carbon nitride (m-CeO2/g-C3N4) nanohybrid, yielding CO and CH4 from photocatalytic CO2 reduction. The heterogeneous nanocomposites displayed improved photocatalytic performance due to synergistic optoelectronic combination between CeO2 and g-C3N4.95 In another approach Zhou et al. developed a nitrogen-doped titanium dioxide (g-C3N4/N-TiO2) composite by in situ urea based simple pyrolysis process. g-C3N4/N-TiO2 photocatalysts showed enhanced catalytic activity for water vapor-assisted gas-phase CO2 photoreduction under visible light irradiation in ambient atmosphere in comparison with g-C3N4 and commercial titania (P25), when used alone. The g-C3N4/N-TiO2 composites showed selective formation of CO from photocatalytic CO2 reduction. The observed maximum CO formation rate is around 14.73 μmol under 12 h photoirradiation, which is four times higher than that from commercial P25.96 In the last several years, advanced design and consecutive synthesis of Z-scheme photocatalysts has become one of the promising strategies in CO2 reduction due to its remarkable efficiency to address environmental depletion. In a different approach to enhance photochemical performance of g-C3N4 photocatalyst, Wang et al. utilized the Z-scheme approach and developed BiOI/g-C3N4. In this hybrid photocatalyst system, the band positions of the two semiconductors effectively allow the electrons from g-C3N4 to recombine with the holes of the nearby conduction band of BiOI without conventional type-II charge transfer process. The BiOI/g-C3N4 composite exhibited an effective photocatalysis reaction under visible-light irradiation (λ > 400 nm) and reduced CO2 to CO, H2 and CH4. It was revealed that the composite provided better catalytic performance than pristine g-C3N4 and BiOI, when used separately, under light energy irradiation. Interestingly the band structure of the BiOI/g-C3N4 composite photocatalyst system does not follow the conventional type-II semiconductor heterostructure. It possesses first generation indirect Z-scheme hetero configuration where VB holes of BiOI and CB electrons of g-C3N4 react with intermediate I3/I ions as an acceptor–donor thus helping BiOI electrons and g-C3N4 holes for respective reduction and oxidation processes. Additionally, the intermediate I3 ion, ˙OH and H2O2 detection technique confirmed photogenerated charge carriers are well separated across the charge depletion region of hybrid material obtained by indirect Z-scheme as shown in Fig. 8.97
image file: d0nr05884j-f8.tif
Fig. 8 Mechanistic pathways for photocatalytic CO2 reduction using BiOI/g-C3N4: (a) a type II charge transfer pathway and (b) a Z-scheme pathway. Reprinted with permission from ref. 97. Copyright 2016 American Chemical Society.

Planetary ball milling was used to produce a hybrid of g-C3N4 and tungsten(VI) oxide (WO3) which was also utilized for photocatalytic CO2 reduction and showed promising catalytic activity. Additionally, photochemical synthesis of silver (Ag) or gold (Au) nanoparticles (NPs) on g-C3N4 hybrid photocatalyst provided an enhancement in methanol yield due to embedded metallic NPs exhibiting surface-plasmon and Schottky-junction effect on the photocatalyst surface, as well as improved photon energy intake and electron–hole recombination for CO2 reduction via a multi-electron process. Catalytic CO2 reduction performance of the AuNP-decorated hybrid Au@g-C3N4/WO3 photocatalyst is reported to be around 1.7 times higher than that of pristine g-C3N4/WO3.98 In this solid-phase indirect Z-scheme, AuNPs were used as an alternative of redox ion pairs, and accelerated the photogenerated charge separation as a forbidden Type II heterostructure. The indirect Z-scheme showed a direction in heterojunction photocatalysts to handle the most challenging criteria of effective separation of photogenerated charge carriers. However, due to several limitations of the first-generation Z-scheme, a direct Z-scheme mechanism is proposed for heterostructure photocatalysts to improve the catalytic performance. As a progressive development of the effective charge-separation processes, the direct Z-scheme concept was introduced in photocatalytic CO2 reduction. Shi et al. developed a visible-light-responsive g-C3N4/NaNbO3 nanowire composite photocatalyst system that showed remarkably high photocatalytic CO2 reduction activity in comparison with pristine g-C3N4 and NaNbO3. The improved photogenerated electron–hole pair separation and transfer at the intimate interface of g-C3N4/NaNbO3 heterojunctions reorganized the band structures of C3N4 and NaNbO3 for an improved CO2 reduction process.99 In 2015, He et al. prepared SnO2−x/g-C3N4 composite photocatalysts by high-temperature calcination of g-C3N4 and Sn6O4(OH)4 mixture. SnO2−x/g-C3N4 composite possesses promising photocatalytic CO2 reduction activity and conventional dye degradation activity. The insertion of SnO2−x on g-C3N4 enhanced the surface area, along with enhanced light absorption and carrier separation to further improve the photocatalytic performance. The heterojunction between the wide band gap SnO2−x and narrow band gap g-C3N4 mainly suppresses the electron–hole pair recombination process via direct Z-scheme mechanism and improves the catalytic activity.100 In all these approaches, due to difference in work function between two semiconductors, an internal electric field is the only guiding force to follow the reverse charge flow as compared with the convention Type-II heterostructure. In the metal oxides family, Fe2O3, a visible light narrow band gap semiconductor, has been proven to be a potential photocatalyst because of its low cost, high stability and environmental friendliness. However, the conduction band position of Fe2O3 is more positive than the redox potential of CO2, and therefore Fe2O3 is not suitable for CO2 reduction. Recently, Guo et al. developed a direct Z-scheme photocatalyst, α-Fe2O3/g-C3N4 heterojunction composite, where the interface mismatch between α-Fe2O3 and g-C3N4 creates an interfacial electric field which has the forbidden type II heterostructure. The Z-scheme α-Fe2O3/g-C3N4 heterojunction promotes photogenerated photoexcited carrier separation and suppresses recombination and shows selective methanol formation under visible-light photocatalytic CO2 reduction. The nanostructure interface between α-Fe2O3 and g-C3N4 and comparison of photocatalytic activity together with a schematic of band structure Z-scheme engineering are shown in Fig. 9.101 Composite-based semiconductor photocatalyst development is a significant pathway towards photochemical CO2 reduction to hydrocarbons by utilizing tunable band alignment of various band structure semiconductors.

image file: d0nr05884j-f9.tif
Fig. 9 A g-C3N4/α-Fe2O3 Z-scheme photocatalyst for CO2 reduction: (a) TEM and HRTEM images, (b) CH3OH formation rates, and (c) the photocatalytic mechanism diagram of g-C3N4/α-Fe2O3 composites. Reprinted with permission from ref. 101. Copyright 2019 Elsevier.

In heterostructure photocatalysis, the Z-scheme mechanism has matured over the year from the indirect to direct concept through a scientific and material evolution process. Though the direct Z-scheme heterostructure mechanism is more mature, the 1st generation indirect Z-scheme concept always creates a fundamental conflict with it. Therefore, in 2019, Yu et al. introduced the S-scheme (step-scheme) heterojunction as a new candidate for direct Z-scheme concept. Recently, S-scheme photocatalysis has been explored as an effective approach to control the charge separation for various photocatalytic reactions.102–104 The S-scheme heterojunction is based on the direct Z-scheme heterojunction comprising reduction photocatalyst (RP) and oxidation photocatalyst (OP) with staggered band structure as a similar charge transfer pathway to solve the carrier recombination process. The proposed S-scheme in comparison with the traditional type-II heterojunction is shown in Fig. 10.104 In recent times, several researchers have been using this S-scheme heterojunction concept in photocatalytic CO2 reduction. Among them, recently Wang et al. developed a CdS/TiO2 hollow nanocomposite hybrid as S-scheme heterojunction for photocatalytic CO2 reduction. The CdS/TiO2 hollow nanocomposite exhibits enhanced photocatalytic CO2 reduction to CH4 formation compared to pristine CdS and TiO2 due to the improved charge separation.105 Similarly, He et al. synthesized a 2D/2D/0D TiO2/C3N4/Ti3C2 composite heterojunction photocatalyst, which showed outstanding activity for photocatalytic CO2 reduction. The outstanding activity could be attributed to the formation of dual heterojunction (the S-scheme at the TiO2/C3N4 interface and the Schottky heterojunction at the C3N4/TCQD interface). The experimental and theoretical study showed that dual heterojunctions lead to better charge carrier separation for improved redox activity.106 We believe more detailed insightful study together with a novel nano-architecture concept will broaden the application of S-scheme heterojunctions for improved CO2 reduction activity.

image file: d0nr05884j-f10.tif
Fig. 10 The comparative band structure configuration and charge-transfer pathway of a type-II heterojunction and S-scheme heterojunction. Reprinted with permission from ref. 104. Copyright 2020 Cell Press.
5.1.3. Composites with metal sulfides: interfacial internal fields. Numerous metal sulfide semiconductor materials and their composites have been proven as promising catalysts for photochemical CO2 reduction in the past several decades. In this respect, Xu and co-workers synthesized g-C3N4/SnS2 by a single-step hydrothermal process employing L-cysteine as a sulfur source. The composite materials showed enhanced photocatalytic CO2 reduction to hydrocarbons (CH4, CH3OH) as compared to individual g-C3N4 and SnS2 under visible-light irradiation. Moreover, X-ray photoelectron spectroscopy and density functional theory (DFT) studies proved that electrons transferred from g-C3N4 to SnS2, through the formation of interfacial internal electric fields (IEF) between the two semiconductors at equilibrium. Additionally, in situ Fourier-transform infrared spectroscopy (FT-IR) and energy levels of photoinduced electrons showed that formic acid (HCOOH) formed as an intermediate during CO2 conversion in the presence of light energy. The observed improved performance of g-C3N4/SnS2 can be ascribed to IEF-induced direct Z-scheme and increased CO2 adsorption capacity.107 The interfacial internal fields mainly influence the charge density at the surface of individual semiconductors, which stimulates the adsorption phenomena of the intermediate species on specific active sites and could be a decisive factor for selective product formation from CO2 reduction. Further theoretical study together with in situ experimental observations could extend the future scope of this research direction.
5.1.4. Composites with metal complexes: photosensitizer effects. In the past few decades, homogeneous molecular complexes have been introduced for photocatalytic CO2 reduction, but the major drawback is that they are not reusable. These molecular complexes can be transformed to heterogeneous systems as well as their photocatalytic activity improved by composite formation with graphitic materials. In this regard Kuriki et al. developed a Ru complex/C3N4 hybrid (Fig. 11), which exhibits excellent photocatalytic reduction of CO2 into HCOOH under visible light irradiation. Employing suitable reaction environment, the photocatalytic conversion of CO2 into HCOOH over the hybrid catalyst has been improved, with TON higher than 1000, and apparent quantum yields (AQYs) achieved of around 5.7% at 400 nm.108 Later on, they developed RuRu′/Ag/C3N4 composite photocatalyst taking Ru(II) binuclear complex (RuRu′) containing photosensitizer and C3N4 with Ag nanoparticle co-catalyst for photocatalytic CO2 reduction under visible light.109 The photocatalytic reaction of RuRu′/Ag/C3N4 selectively produced formic acid from CO2 reduction. In photocatalytic activity the resulting composite achieved a TON value around 33[thin space (1/6-em)]000, which is 30 times higher than previously reported for C3N4 modified with a mononuclear Ru(II) complex. This reported photocatalytic activity is the highest among the metal-complex/semiconductor hybrid systems. Based on the complete analyses of reaction, including emission decay measurements, and time-resolved infrared spectroscopy, it was proved that photoinduced electrons generated from C3N4 had several millisecond lifetimes on Ag nanoparticles and that electrons were transferred to the excited state of RuRu′ (i.e. two-step photoexcitation of C3N4 and RuRu′), consequently leading to improved catalytic performances.
image file: d0nr05884j-f11.tif
Fig. 11 Photocatalytic CO2 reduction employing a Ru complex/C3N4 hybrid. CB = conduction band; VB = valence band. Reprinted with permission from ref. 108. Copyright 2015 Wiley-VCH.

5.2. Graphene oxide-based materials

In the past few decades, graphene-based materials like graphene oxides (GOs), reduced graphene oxides (rGOs) and their nanocomposites have appeared as attractive materials in various applications.110 GOs are semiconductor-like wide band gap materials consisting of covalently bonded epoxide and hydroxyl groups at the basal plane and carboxyl functional groups at the edges. 2D layer structures of sp2 and sp3 bonded atoms in GOs are constructed with oxygenated functional groups, which can eventually tune the presence of a finite band gap due to isolated sp2 domains. GOs showed a tunable band structure like semiconductor materials by an adjustable ratio between the sp2 and sp3 fraction by reduction chemistry and can transform to a graphene-like semi-metal. Yeh et al. have introduced GOs for photocatalytic water splitting.111 Keeping this fact in mind, Hsu et al. first employed GOs for photocatalytic CO2 reduction to methanol under visible light irradiation (Fig. 12a). The modified Hummers’ method has been used to synthesize GO-based photocatalysts for moderately improved catalytic activity. The photocatalytic CO2 reduction and selective conversion to methanol on modified GOs were around six-fold higher compared to pure commercial TiO2 (P25).112 For improvement in photocatalytic CO2 reduction using GOs, several strategies have been considered. Among them, composite formation employing GOs with metals, metal oxides, and metal complexes is of paramount importance.
image file: d0nr05884j-f12.tif
Fig. 12 (a) The photocatalytic CO2 reduction mechanism on graphene oxide. Reprinted with permission from ref. 112. Copyright 2013, Royal Society of Chemistry. (b) The photocatalytic formation of solar fuels on Cu/GO. Reprinted with permission from ref. 113. Copyright 2014, American Chemical Society.
5.2.1. Composites with metal nanoparticles: increasing surface electron density. GOs suffer in terms of visible-light response due to their wide band gap, and many conventional methods such as doping with metal ions, preparing solid solutions and coupling with various narrow band gap semiconductors are critical approaches to transforming GOs into visible light active materials. Shown and co-workers prepared copper nanoparticles (Cu NPs) onto GO (Cu/GO) for enhanced photocatalytic CO2 reduction. The dispersed Cu NPs tuned the Fermi energy of GOs and influenced the onset reduction potential of CO2 for selective one-carbon-containing methanol and two-carbon-containing acetaldehyde formation. The optimum Cu NP-decorated GOs showed 60 times higher catalytic performance as compared to GOs under visible-light irradiation. The improved photocatalytic performances can be attributed to the suppression of electron–hole pair recombination along with change in GO's band gap and its work function.113 It has been demonstrated that charge carrier generation, charge carrier transfer and multi-electron chemical reduction for a particular CO2 onset reduction potential are solely associated with the photocatalytic CO2 reduction process. The initial photocatalytic step is mainly triggered by excitation with visible light, in which photons with higher energy than the band gap of the GOs generate electron–hole pairs on the Cu/GO photocatalyst surface. On the GO surface, Cu NPs play a pivotal role as an electron acceptor and the recombination of photoexcited electron–hole pairs was restricted, owing to the enhanced charge separation at the metal–semiconductor interfaces (Fig. 12b). Metal nanoparticles with typical size distributions possess unique electronic properties, which account for the phenomenal effect on the semiconductor surface during photocatalysis; thus use of a semiconductor with metal nanoparticle modification is a potential approach for CO2 reduction.
5.2.2. Composites with metal oxides: charge carrier separation and transport. Over the last four decades, the development of metal oxide-based catalysts has progressed, although at a slower pace. Inclusion of high surface area catalyst supports such as graphene and its functionalized derivatives has provided extraordinary utility in the development of stable efficient photocatalysts. An et al. synthesized Cu2O microparticles on rGO support through a facile one-step microwave-assisted chemical method. The Cu2O/rGO catalyst used for photocatalytic CO2 reduction under ambient conditions showed excellent catalytic activity, about six times and fifty times higher activity than Cu2O and Cu2O/RuOx, respectively. This observed increased catalytic activity and stability of Cu2O were mainly attributed to the efficient charge separation and transfer to rGO as well as the shielding effect of rGO.114 Later on, Huang and co-workers developed a series of zinc oxide/rGO nanocomposites (ZnO-rGO) using a one-step hydrothermal method which showed photocatalytic CO2 reduction to methanol. The composite catalyst exhibited five times higher activity than pure ZnO and excellent recycling property.115 Afterwards, Kumar et al. synthesized CuZnO@Fe3O4 microsphere wrapped with rGO (rGO@CuZnO@Fe3O4) through a solvothermal method and employed it for photocatalytic CO2 reduction to methanol under visible light irradiation. The rGO@CuZnO@Fe3O4 catalyst showed excellent photocatalytic activity as compared to CuZnO@Fe3O4 due to a synergistic effect of interfacial surface chemistry (Fig. 13a).116 Towards improvement of photocatalytic CO2 reduction, Xu and coworkers demonstrated a ternary Ag2CrO4/g-C3N4/graphene oxide nanocomposite for enhanced CO2 photoreduction activity. The interfaces of the Ag2CrO4/g-C3N4/GO potentially developed a direct Z-scheme by crystal strain-induced interfacial electric field. In this ternary nanocomposite, GO acts as a cocatalyst and facilitates an improved charge transfer and CO2 adsorption (Fig. 13b).117 That research work explored GO as a metal-free cocatalyst together with Ag-based oxide as a photosensitizer to improve the photocatalytic performance of g-C3N4.
image file: d0nr05884j-f13.tif
Fig. 13 (a) The possible mechanistic pathway for photocatalytic CO2 reduction using an rGO@CuZnO@Fe3O4 composite. Reprinted with permission from ref. 116. Copyright 2016, Elsevier. (b) A schematic diagram of the Z-scheme photocatalytic mechanism for an Ag2CrO4/g-C3N4/graphene oxide ternary nanocomposite. Reprinted with permission from ref. 117. Copyright 2018, Elsevier.

5.3. Other carbon-based materials

Goodenough and coworkers prepared carbon-coated indium oxide (In2O3) nanobelts for photocatalytic CO2 reduction with water as reductant and Pt as co-catalyst (Fig. 14). The catalyst triggered an enhanced photocatalytic reduction of CO2 to CO and CH4. It was demonstrated that the carbon coating can improve visible-light absorption and reduce electron–hole recombination by inducing excitation of electrons from the O 2p band of In2O3 to the carbon band within an electric double layer at the interface between the carbon layer and In2O3. Additionally, the carbon layer on In2O3 can increase the rate of CO2 reduction by suppressing H2 evolution from water splitting.118 We believe that the observed suppression of the water splitting reaction is one of the important factors to control CO2 reduction and selective multi-electron hydrocarbon formation. Recently, Shown et al. have developed carbon-doped SnS2 for photocatalytic CO2 reduction by an L-cysteine-assisted hydrothermal process.119 The obtained carbon-doped SnS2 exhibits selective reduction of CO2 to acetaldehyde with a moderately high photon-to-chemical quantum efficiency of around 0.7% under visible light. Through structural analysis, it was concluded that interstitial carbon doping in SnS2-C can induce micro-strains and change the electronic structure as well as optical properties. Additionally, DFT calculation proved carbon doping can increase CO2 molecule adsorption on the surface with a relatively small dissociation barrier, which is favorable for CO2 reduction.
image file: d0nr05884j-f14.tif
Fig. 14 The photocatalytic reduction of CO2 to CO and CH4 using indium oxide (In2O3) nanobelts coated with a thick carbon layer using Pt as co-catalyst. Reprinted with permission from ref. 118. Copyright 2017, American Chemical Society.

6. Metal organic frameworks

Metal–organic frameworks (MOFs) are a class of 3D crystalline inorganic–organic hybrid porous materials consisting of metal or metal cluster nodes interconnected with multi-dentate organic linkers. Such MOFs exhibit promising utility in various applications, including catalysis,120–122 separation,123–125 gas storage,126–128 carbon dioxide capture,129–131etc. Recently, porous MOFs and derivative materials were shown to exhibit excellent catalytic activity in different photochemical reactions like degradation of organic pollutants and water splitting reaction.132 Previous literature studies on Zn-containing MOF (MOF-5) reported that metal nodes in MOF can function as semiconductor quantum dots (QDs) with organic linkers as the antennas to sensitize those QDs in photocatalytic reactions.133 MOFs possess some intriguing characteristic features. (i) Introducing various metal ions and linkers in MOFs leads to tunable light absorption characteristics for a broad range of solar energy, (ii) charge transfer process from organic linker to metal can be easily achieved due to the crystalline nature of MOFs and (iii) the highly porous nature of MOF materials facilitates mass transport during reactions,134 which has garnered colossal interest in the scientific community making MOFs promising candidates for photocatalysis. In addition, the interconnected porous cage-like structure of MOF materials with active sites offers efficient CO2 adsorption and sequential transformation to value-added chemicals.135 In the past few years various research groups have been devoted to the design of different kinds of MOFs for photocatalytic CO2 reduction.136–138 Herein we discuss different metals (Ti, Zr, Fe, Cu, Ru, Co and Mg) contained in MOFs separately, which have been explored.

6.1. Ti metal-containing MOFs: enhancing optical absorption and CO2 adsorption

Inspired by and as further development of conventional TiO2 semiconductors, MOFs based on Ti-oxo clusters show high potential for photochemical CO2 reduction application and dye degradation because of their tunable photophysical properties. Li and coworkers introduced for the first time a photoactive Ti-containing MOF, NH2-MIL-125(Ti), replacing 1,4-benzene dicarboxylate (BDC) linker by 2-aminoterephthalic acid (H2ATA) for the reduction of CO2 to HCOO under visible light irradiation.139 The MOF was synthesized via hydrothermal reaction of H2ATA and tetra-n-butyl titanate, Ti(OC4H9)4, in dimethylformamide (DMF) and dry MeOH solvent. UV/visible spectra revealed that NH2-MIL-125(Ti) (absorption edge at 550 nm) exhibited higher optical absorption due to the presence of amine groups in NH2-MIL-125(Ti) in comparison with MIL-125 (Ti) (absorption edge at 350 nm). CO2 adsorption isotherm study showed that the presence of amine functional groups contributed toward the higher CO2 adsorption capability in NH2-MIL-125(Ti) as compared to parent MIL-125(Ti). The liquid-phase photocatalytic CO2 reduction showed that CO2 selectively converted into formate ions (HCOO) in the presence of sacrificial agent triethanolamine (TEOA) under visible light irradiation. Additional 13CO2 based isotopic tracer analysis proved that the formate ion is formed from CO2. Electron spin resonance (ESR) and UV diffuse reflectance spectroscopy (DRS) study found that the Ti3+ moiety produced via ligand-to-metal charge-transfer (LMCT) mode in the presence of sacrificial donor TEOA is accountable for the formation of formate ions.

Later on, Sun et al. designed metal-doped (M = Pt and Au) NH2-MIL-125(Ti) by a wet impregnation method using PtH2Cl6·6H2O or AuHCl4·4H2O as metal precursors and utilized it in photocatalytic CO2 reduction using TEOA as sacrificial donor under visible light irradiation. It was reported that in photocatalytic CO2 reduction, along with formate ions, hydrogen was evolved over Pt- and Au-loaded NH2-MIL-125(Ti). The catalytic performance (i.e. in formate ion production) of Pt/NH2-MIL-125(Ti) is significantly higher than that of pure NH2-MIL-125(Ti), but Au/NH2-MIL-125(Ti) exhibited inferior performance (Fig. 15). In order to gain an insight, combination studies of ESR experiments along with DFT have been conducted. Interestingly, the combined ESR and DFT study concluded that hydrogen can overflow from Pt to the bridging oxygen linked to Ti atoms, forming Ti3+ which is thought to be responsible for the enhanced performances of Pt/NH2-MIL-125(Ti).140 Similarly, Zhu et al. developed Co-doped NH2-MIL-125(Ti) catalysts [Co/NH2-MIL-125(Ti)], which showed comparatively higher catalytic activity than NH2-MIL-125(Ti) for photochemical CO2 reduction under visible light. The enhanced catalytic activity may be ascribed to doping of Co nanoparticles onto NH2-MIL-125(Ti).141 The development of various transition metal-containing MOFs has some prospect towards photocatalytic CO2 reduction. Photocatalytic activity of Ti-MOF could be easily tuned through introducing other electron donor ligands or noble and non-noble metal doping as these factors not only increase visible-light absorption and CO2 adsorption but also induce more LMCT, which helps to improve the overall catalytic activity.

image file: d0nr05884j-f15.tif
Fig. 15 Photocatalytic performances over NH2-MIL-125(Ti), Pt/NH2-MIL-125(Ti), and Au/NH2-MIL-125(Ti): (a) hydrogen and (b) HCOO as the reduced products. Reprinted with permission from ref. 140. Copyright 2014, Wiley-VCH.

6.2. Zr metal-containing MOFs: high stability, optical absorption and charge transfer

In the progressive development of MOF-based photocatalysts, Zr-based MOFs are quite attractive and widely studied in photocatalysis due to their moderately high stability. UiO-66(Zr), a Zr-containing MOF, which is inert and robust in nature under a variety of chemical conditions, with hexameric Zr6O32 units and linker BDC142 exhibited photocatalytic activity after substitution of the BDC unit with H2ATA. Interestingly, this MOF exhibited an improved catalytic activity compared to Ti-containing MOF NH2-MIL-125(Ti). Like NH2-MIL-125(Ti), this MOF was also prepared following a hydrothermal technique with ZrCl4 and H2ATA in anhydrous DMF and deionized water. TEOA also plays a key role as electron donor and provides a basic environment to facilitate photocatalytic CO2 reduction. Based on photoluminescence studies, it is confirmed that photoinduced electron transfer took place from excited linker 2-aminoterephthalate (ATA) to Zr-oxo clusters in NH2-UiO-66(Zr). Through the ESR technique, production of Zr(III) has been trapped and its participation in photocatalytic CO2 reduction is verified. In addition the superior photocatalytic performance of MOF as synthesized using mixed ligands of ATA and 2,5-diaminoterephthalate could be connected with the increased absorption in the visible light region and increased CO2 adsorption. Meanwhile, Lee et al. observed that mixed MOF UiO-66 derivative Zr4.3Ti1.7O4(OH)4(C8H7O4N)5.17(C8H8O4N2)0.83 as derived by post-synthetic exchange (PSE) of Ti(IV) into a series of UiO-66 MOFs showed an improved photocatalytic efficiency. Comparative studies demonstrated that the mixed MOFs exhibited superior activity to others owing to a lowering of the electron accepting levels of secondary building block units like Zr6−xTix(Zr6−xTixO4(OH)4). Light absorption as well as charge transfer capability increased significantly after incorporating diamine-substituted ligands in the MOFs as established by UV-DRS and photoluminescence studies.143 With a combined ESR analysis and DFT study, Li and co-workers proposed that incorporated Ti moieties play a decisive role as electron mediators to promote electron transfer from the excited organic linker ATA to the Zr–O cluster to form Zr(III), the actual active species for CO2 reduction. Based on transient absorption spectroscopy it is proved that Ti acts as mediator in photoinduced electron transfer.144 Later on, Cohen and coworkers adopted a similar synthetic strategy where a catechol-functionalized organic linker (catbdc, referring to 2,3-dihydroxyterephthalic acid) was introduced into UiO-66 via a post-synthetic metalation method to prepare UiO-66-CAT. After that Cr(III) and Ga(III) metal ions were encaged into the catbdc sites to afford unprecedented Cr- and Ga-monocatecholato species in UiO-66.145 The photocatalytic CO2 reduction performance was assessed employing 1-benzyl-1,4-dihydronicotinamide and TEOA in solution. After reduction, mainly HCOOH (with negligible amounts of H2 and CO) was produced for both Cr- and Ga-containing catalysts. The calculated TON of HCOOH for UiO-66-CrCAT and UiO-66-GaCAT are 11.22 and 6.14, respectively. Based on photoluminescence studies, it can be emphasized that the presence of catbdc organic ligand assisted in visible light absorption and charge transfer between catbdc ligand and metals through a LMCT pathway in these M(III)-monocatecholato functionalized MOFs. Higher activity for UiO-66-GaCAT is ascribed to a higher redox potential of Ga(III)/Ga(II) as compared with that of Cr(III)/Cr(II) in UiO-66-CrCAT.

Lin et al. successfully introduced ReI(CO)3(dcbpy)Cl (dcbpy = (2,2′-bipyridine)-5,5′-dicarboxylic acid, H2L) into UiO-67,146 to furnish an isostructural Re-doped UiO-67 composite, which showed photocatalytic CO2 reduction activity under visible light. In this reaction mainly CO is formed with a TON of 5. Possibility of CO formation from decomposition of the Re complex has been ruled out because no CO was detected in the absence of CO2 for the same reaction conditions. CO2 reduction is triggered in the presence of [Re(dcbpy)(CO)3Cl] moiety in UiO-67, which unambiguously confirmed that no CO production was observed in the presence of parent UiO-67. Yaghi and coworkers developed a UiO-67 coated nanoparticle photocatalyst by covalently attaching ReI(CO)3(BPYDC)Cl (BPYDC = 2,2-bipyridine-5,5-dicarboxylate) to UiO-67 (Ren-MOF) and systematically controlling its density in the framework (n = 0, 1, 2, 3, 5, 11, 16, and 24 complexes per unit cell). Among all the reported Ren-MOF photocatalysts, Re3-MOF showed better photocatalytic activity for CO2 reduction. Thereafter, they developed Ag nanocube-decorated Re3-MOF (Ag@Re3-MOF) for enhanced photocatalytic activity. In the Ag@Re3-MOF photocatalyst, Re centers were spatially enclosed to the intensified near-surface electric fields on the surface of Ag nanocubes and improved the charge separation. Ag@Re3-MOF exhibited excellent catalytic activity with around 7-fold enhancement of CO2 to CO conversion under visible light along with long-term stability to 48 h due to synergistic effect between plasmonic Ag and photoactive Re centres.147 In addition to Ren-MOF photoactive complex, photoactive Rh and Ru complexes were also doped on MOF UiO-67 and used as efficient photocatalysts for CO2 reduction. Fontecave et al. have obtained HCOOH as main product in photocatalytic CO2 reduction with Cp*Rh@UiO-67, where Cp*Rh(bpydc)Cl2 (bpydc = 2,2′-bipyridine-5,5′-dicarboxylic acid) complex was integrated into UiO-67.148 Interestingly, it is reported that 10 mol% Rh loading in the complex photocatalyst leads to maximum HCOOH formation. However, a further increase of Rh loading leads to thermal decomposition of formate to H2. Later, Gao et al. compared two metal ion (Co2+, Re+)-doped UiO-67 catalysts named Co-UiO-67 and Re-UiO-67 in CO2 photoreduction. Their study showed that Co-UiO-67 exhibits better activity than Re-UiO-67 due to improved charge transport ability and higher CO2 adsorption capacity. Moreover, the DFT calculation proved that the energy barrier of Co-UiO-67 (0.86 eV) for the reaction is lower than that of Re-UiO-67 (0.92 eV), which also helps in the greater catalytic activity of the Co ion-doped MOF.149 Moreover, the photocatalytic activity of Zr-based MOFs could be further enhanced through various techniques like replacement of organic linker, producing a mixed metal containing MOF via post-synthetic strategy and immobilization of Re or Rh complex on MOF. Immobilization of metal complex approach for Zr-based MOFs reveals more promising activity because it reduces charge recombination and acts as a CO2 activation center. However, further study is needed to establish this approach for future application.

6.3. Fe metal-containing MOFs: dual excitation pathway under visible light

It has been reported that Fe-based MOF materials show excellent catalytic activity under visible light because of the presence of iron oxo clusters. Li and co-workers studied photocatalytic CO2 reduction over a series of Fe-containing MOFs, including MIL-101(Fe), MIL-53(Fe) and MIL-88B(Fe).150 Among these three Fe-MOFs, MIL-101(Fe) showed the highest catalytic performance and producing 59 mol formic acid in 8 h from CO2 during photocatalytic CO2 reduction. In situ FT-IR studies showed that the higher catalytic performance of MIL-101(Fe) as compared with MIL-53(Fe) and MIL-88(Fe) can be attributed to the coordination of bidentate carbonate to the metal center favoring direct adsorption of CO2 onto the Fe center. The presence of unsaturated coordination Fe site in MIL-101(Fe) mainly promotes the overall photocatalytic CO2 reduction reaction. Furthermore, the amine-functionalized Fe-MOF showed better catalytic activity than the parent un-functionalized MOF. The formic acid production rate over NH2-MIL-101(Fe) was 3-fold higher than that over the parent MIL-101(Fe) under similar conditions. This enhanced catalytic activity has been explained by a dual excitation pathway where an electron is transferred from O2− to Fe3+ in the Fe–O clusters upon visible-light irradiation followed by an electron transfer from the excited organic linker to the metal center to generate Fe2+ as shown in Fig. 16.
image file: d0nr05884j-f16.tif
Fig. 16 The dual-excitation mechanism pathway for photocatalytic CO2 reduction over amino-functionalized Fe-based MOFs. Reprinted with permission from ref. 150. Copyright 2014, American Chemical Society.

The majority of MOFs utilized for solution-phase photocatalytic CO2 reduction require a specific organic solvent which is not environmentally friendly. Therefore, scientists are looking for a solvent-free reaction route having an advantage like better solid–gas interaction, which would definitely accelerate the reaction. In this approach, Dao et al. tested solvent-free CO2 photoreduction using three classical Fe-based MOFs, NH2-MIL-53(Fe), NH2-MIL-88B(Fe), and NH2-MIL-101(Fe). Among the three MOFs, NH2-MIL-101(Fe) showed the best catalytic activity, which might be due to the unique structure with unsaturated coordination metal sites and convincing electron transfer.151 Later similarly, they developed NH2-MIL-101(Fe)/g-C3N4 composite for photocatalytic CO2 reduction via a solvent-free reaction. In the reported composite, the catalyst showed effective interfacial charge transfer between NH2-MIL-101(Fe) and g-C3N4, which improved the catalytic activity.152

6.4. Cu metal-containing MOFs: CO2 adsorption and high stability

Photocatalytic MOFs are usually fabricated from pure photocatalytic ligands responsible for delivering large amounts of photocatalytically active sites, with high light harvesting efficiency and catalytic activity. Porphyrins or metalloporphyrins have been well recognized as principal photoactive ligands. Custom-designed porphyrin ligands developed by a fabrication technique (Porph-MOFs) have been chosen as potential candidates in photocatalysis due to their rich photophysical/photochemical properties and tunable features. Liu et al. have synthesized Cu-based Porph-MOFs (SCu) adopting the metalation technique of Porph-MOFs via a PSE strategy, which attracted tremendous attention for both CO2 capture and photocatalytic CO2 conversion to methanol.153 The significant effect of introducing Cu on the photocatalytic CO2 reduction was prominent as the photocatalytic methanol evolution rate over SCu was enhanced by 7 times in comparison with Sp. In situ FT-IR study proved that chemical adsorption and activation of CO2 on the Cu site initiate the catalytic reaction by changing linear CO2 to bent CO2 and regulate the enhancement of the CO2 conversion efficiency (Fig. 17). MOF/TiO2 hybrid systems have been widely used for photocatalytic CO2 reduction mainly due to the advantages of high adsorption capacity along with high stability of MOFs toward CO2 and promising photocatalytic ability of TiO2. Cu3(BTC)2@TiO2 core@shell framework (Cu3(BTC)2 abbreviated as HKUST-1; BTC = benzene-1,3,5-tricarboxylate) was synthesized by coating a Cu3(BTC)2 core using a nanocrystalline TiO2 shell.154 The Cu3(BTC)2@TiO2 hybrid system showed fivefold enhancement in photocatalytic activity for CO2 conversion to CH4 along with significant increase of selectivity in comparison with pristine TiO2. The photoexcitation of the TiO2 shell produced photogenerated electrons, which effectively transferred to the Cu3(BTC)2 core, thereby relocating energetic electrons to reduce the adsorbed CO2 molecules on the Cu3(BTC)2 co-catalyst surface; this carrier dynamics was well proved by an additional ultrafast spectroscopy characterization technique.155 And it has been observed that photocatalytic CO2 reduction activity of Cu-based MOFs depends on the nature of Cu sites. Therefore, modification of Cu center through development of composite and ligand functionalization for Cu-MOFs and their detailed study could help us to develop next-generation photocatalysts for CO2 reduction.
image file: d0nr05884j-f17.tif
Fig. 17 (a) Photocatalytic CO2 reduction to methanol and (b) a CO2 adsorption study via in situ FT-IR spectroscopy over a copper porphyrin-based metal organic framework (“end-on” and “C-coordination” adsorption geometries of CO2 on SCu from experimental studies). Reprinted with permission from ref. 153. Copyright 2013, American Chemical Society.

6.5. Ru metal-containing MOFs: broad light absorption

Metal complexes are also useful as connectors for the construction of photoactive MOFs for the CO2 reduction process. Metalloligands such as [Ru(5,5′-dcbpy)3]4 ([Ru-L1]4, 5,5′-dcbpy = 2,2′-bipyridine-5,5′-dicarboxylate) and [Ru(4,4′-dcbpy)2(bpy)]2 ([Ru-L2]2, 4,4′-dcbpy = 2,2′-bipyridine-4,4′-dicarboxylate, bpy = 2,2′-bipyridine) are fruitfully employed as building blocks. Non-interpenetrated and interpenetrated structures of two Ru-polypyridine containing MOFs, {Cd3[Ru-L1]2·2(Me2NH2)·solvent}n (Ru-MOF-1) and {Cd[Ru-L2]·3(H2O)}n (Ru-MOF-2), have been rationally designed for photocatalytic CO2 reduction.156 The Ru-MOF-1 framework as synthesized from a mixture of Ru-L1 and Cd(ClO4)2·6H2O has predominately 1D channels with large void space. Previous studies revealed that the enclosed channel walls isolate the channels, thus creating a barrier to structural interpenetration. Similarly, they prepared Ru-MOF-2 with 2-fold interpenetrating structure from Ru(H2dcbpy)2Cl2, Cd(ClO4)2·6H2O, and bpy. [Ru–L2]2− metalloligand is linked to Cd(II) centers, creating a network with a large amount of 3D void space, which is responsible for the accommodation of a second framework via interpenetration as compared to Ru-MOF-1. In Ru-MOF-2 two interpenetrated frameworks are stabilized by a strong π-stacking interaction between the dcbpy2− ligands of the neighboring networks. Interestingly, broad absorption bands in the range of 400 and 650 nm in both MOFs indicate the singlet metal-to-ligand charge transfer of Ru metalloligands. In the process of photocatalytic CO2 reduction, both Ru-MOF-1 and Ru-MOF-2 produce formate under visible light in the presence of TEOA as a sacrificial agent. Moreover, Ru-MOF-2 revealed stable and repeatable photocatalytic activity as compared to Ru-MOF-1. Recently, Mahmoud et al. successfully synthesized another stable zirconium-based MOF through incorporation of photoactive bis(4′-(4-carboxyphenyl)terpyridine)ruthenium(II) (Ru(cptpy)2) unit, which exhibits excellent photocatalytic activity for CO2 reduction to formate having a conversion rate of 366 μmol g−1 h−1.157 Therefore, it could be concluded that through variation of photoactive Ru complex unit, the activity might be easily enhanced through broad light absorption.

6.6. Co metal-containing MOFs: charge separation and high quantum yields

Co-Zeolitic imidazolate framework-9 (Co-ZIF-9) is a porous MOF containing cobalt(II) ions and benzimidazolate as organic linker. Wang et al. first reported photocatalytic CO2 reduction employing Co-ZIF-9 as a co-catalyst, [Ru(bpy)3]Cl2·6H2O (bpy = 2,2′-bipyridine) as a photosensitizer and TEOA or trimethylamine (TEA) as a sacrificial electron donor.158 After reduction, mainly formic acid is produced along with H2. Co-ZIF-9 showed better photocatalytic activity as compared to other MOFs including Co-MOF-74, Mn-MOF-74, NH2-Uio-66(Zr) and Zn-ZIF-8 under similar conditions, due to the synergetic effect between cobalt and benzimidazolate entities in Co-ZIF-9. In the photocatalytic process, the used dye photosensitizer associated with Co-ZIF-9 photocatalyst leads to decreasing activity due to photobleaching phenomenon, which is considered as the main drawback. To overcome the above drawback Wang et al. extended the use of Co-ZIF-9 as co-catalyst for CO2 reduction by composite formation with different semiconductors like g-C3N4,159 CdS160 (Fig. 18) etc. The newly designed semiconductors/Co-ZIF-9 (CdS/Co-ZIF-9) hybrid composite exhibited higher catalytic activity and a high AQY (1.93% at 420 nm).
image file: d0nr05884j-f18.tif
Fig. 18 The photocatalytic CO2 reduction mechanism of a CdS-promoted zeolitic imidazolate framework. Reprinted with permission from ref. 160. Copyright 2015, Elsevier.

Later, Kong et al. synthesized CsPbBr3@ZIF composites via an in situ technique growing a ZIF coating on CsPbBr3 dot. Between the two composites CsPbBr3@ZIF-8 and CsPbBr3@ZIF-67, cobalt-based MOF CsPbBr3@ZIF-67 showed greater photocatalytic CO2 reduction activity because active Co centers in ZIF-67 mainly enhance the charge separation and create more active sites for CO2 adsorption.161 It was shown that Co-ZIF acts as a cocatalyst with n-type semiconductor and forms a Schottky junction at their interface.

7. Covalent organic frameworks

Covalent organic frameworks (COFs) represent an emerging class of crystalline porous materials whose backbone unit comprises entirely light elements (B, C, N, O, Si).162 COFs are constructed from various periodic organic building blocks such as phenyl, biphenyl, naphthalene, anthracene, pyrene, triazine, etc., via reversible covalent bond formation.163 COFs with 2D and 3D nano-architectures are mainly developed by using suitable multidentate building units with excellent chemical and thermal stabilities. Besides definite pore size, COFs with high surface areas have been achieved, to yield exciting applications in gas storage,164 catalysis,165 separation,166 energy storage,167 and proton conduction.168 Recently, COFs showed new promise for photocatalytic hydrogen production, which is encouraged by axial charge transport in the stacking direction by the overlap of π-orbitals along with in-plane π-electron conjugation.169,170 Moreover, these structurally diverse COFs contribute excellent activity towards CO2 adsorption, diffusion, and activation. The inherent advantages of COFs mean that they are excellent and promising CO2 reduction photocatalysts.171 In this respect, Yadav et al. were among the first to develop a COF-based photo-biocatalyst integrated system (shown in Fig. 19), wherein COF was synthesized via condensation polymerization between cyanuric chloride and 3,4,9,10-perylenetetracarboxylic diimide building units, and employed it for visible light-driven photocatalytic CO2 reduction.172 The excellent photocatalytic activity may be ascribed to the suitable band gap and highly ordered π-electron channel systems, which greatly facilitate charge carrier transport.
image file: d0nr05884j-f19.tif
Fig. 19 Photocatalytic CO2 reduction to formic acid employing a COF-based photocatalyst-enzyme coupled system. (Here Rhox = [Cp*Rh(bipy)H2O]2+, Rhred1 = Cp*Rh(bipy), Rhred2 = [Cp*Rh(bipy)H]+; Cp* = pentamethylcyclopentadienyl, bpy = 2,2′-bipyridine.) Reprinted with permission from ref. 172. Copyright 2016, Royal Society of Chemistry.

Later on, Yang et al. developed a 2D COF incorporated with Re complex (Re(bpy)(CO)3Cl) for photocatalytic CO2 reduction under visible light irradiation. This hybrid photocatalyst showed excellent catalytic activity and selectivity (90%) towards CO formation. Advanced transient optical, X-ray absorption and in situ diffuse reflectance spectroscopy results revealed charge separation, induction period and rate limiting step are the key responsible factors in the reported 2D COF for photocatalytic activity as shown in Fig. 20.173 In a similar way, Cao and coworkers synthesized the Re carbonyl complex Re(CO)3Cl grafted porous pyridine-based covalent triazine framework CTF (Re-CTF-py). The reported Re-CTF-py showed high catalytic efficiency in the photocatalytic CO2 reduction to CO (353.05 μmol g−1 h−1 within 10 hours) with a TON of around 4.8.174 As discussed, metal complex immobilization on the 2D COF exhibits greater photocatalytic CO2 reduction activity; therefore, in future, the development of more conjugated 2D COFs with different functionality is much needed to exploit this opportunity.

image file: d0nr05884j-f20.tif
Fig. 20 The possible mechanistic pathway for photocatalytic CO2 reduction to CO using a 2D COF with an incorporated Re complex (Re(bpy)(CO)3Cl). Reprinted with permission from ref. 173. Copyright 2018, American Chemical Society.

8. Conjugated porous polymers: towards metal-free photocatalysts

Towards metal-free photocatalyst systems for sustainable energy application, in recent years conjugated porous polymers (CPPs) have been shown as potential candidates for various photocatalysis processes. A CPP is a crosslinked high surface area polymer containing extensive tunable π-conjugation with tunable electron acceptor and donor characteristics. MOF, COF, and CPP building blocks possess strong chemical and thermal stability due to their covalent linkage during the polymerization process.175 In the last few years, CPP-based photocatalysts have been developed with a tunable band structure and pore size for a range of photocatalytic reactions like oxidation, coupling, dye degradation, photopolymerization, hydrogen production and CO2 reduction. The CPPs are synthesized via the polymerization reaction of a bifunctional (bridge) to a multifunctional monomeric unit (cross-linker) containing conjugated donors and acceptors. Recently, CPPs have been reported as emerging photocatalysts with highly tunable optoelectronic properties due to a wide range of functional variation in organic conjugation.176 Liao et al. synthesized a series of nitrogen-rich CPP-based photocatalysts, of which the higher nitrogen content had a great influence on the CO2 adsorption characteristics,177 which is highly beneficial for the photocatalytic CO2 conversion process. In another synthetic approach, introducing nitrogen-containing carbazole in CPP photocatalyst systems led to excellent photocatalytic activity under visible light.178,179 Cooper and coworkers developed various chain-extended biphenyl analogs and also planar spirobifluorene-based CPP photocatalysts.180 The planarization in the CPPs effectively influenced the optoelectronic characteristics and improved the photocatalytic performance as compared with linear nonporous phenylene-based CPPs. Subsequently, Xiao et al. developed Fe-porphyrin-based interconnected metallo CPP photocatalyst for dye degradation.181 These macrocyclic transition metal-containing CPPs as shown in Fig. 21 have opened a promising route for the development of a selective CO2 reduction photocatalyst system with tunable transition metal center. Moreover, the different structural modifications with donor–acceptor unit in CPP systems are promising approaches towards tailor-made photocatalyst development for effective CO2 conversion to solar fuel. From the mechanistic viewpoint, such a tunable approach enables a controllable surface polarity of the CPP photocatalysts, resulting in selective adsorption and desorption of CO2 molecules on the photocatalyst surface and helping CO2 conversion towards higher hydrocarbons.
image file: d0nr05884j-f21.tif
Fig. 21 Selective structural conjugated porous polymers (CPPs): (a) aniline-based,177 (b) carbazole-based,178 (c) BODIPY-based,179 (d) biphenyl-based,180 and (e) Fe-porphyrin-based CPPs.181

Over the last few years, CPPs have been increasingly utilized in the area of photocatalytic CO2 reduction processes; however, the CPP-based photocatalyst systems are still at an early stage of development in photocatalytic CO2 reduction research. In 2018, Yang et al. developed triazine-based CPPs to capture, activate and reduce CO2 to CO with visible light. The triazine-based CPPs containing benzothiadiazole porous photocatalyst effectively converted CO2 to CO with an AQY of 1.75% at 405 nm as shown in Fig. 22.182 That work revealed that different kinds of donor–acceptor groups in the conjugated polymeric structure tuned the electronic band structure as well as controlled the selectivity and efficiency of the photocatalytic CO2 reduction process using visible light. Recently, Zhong et al. demonstrated covalent triazine-based CCPs for visible light-driven CO2 reduction. That work showed that triphenylamine- and triazine-containing extended π-conjugated CPP system well separated the photogenerated carriers and efficiently utilized visible light during the CO2 conversion process.183 The tailor-made donor–acceptor dyad concept in a porous conjugated macromolecular system represents a promising and versatile strategy of utilizing CPPs as efficient visible-light photocatalysts for CO2 reduction in the near future.

image file: d0nr05884j-f22.tif
Fig. 22 (a) Triazine-based 1,4-phenylenediboronic acid (CPs-B)-, 2,5-thiophenediboronic acid (CPs-Th)- and 2,1,3-benzothiadiazole-4,7-bis(boronic acid pinacol ester) (CPs-BT)-containing conjugated polymer photocatalysts. (b) The optical band gaps and corresponding band structures of three triazine-based conjugated polymer photocatalysts. (c) The wavelength dependence of AQY for H2 and CO production for the photocatalyst CPs-BT. Reprinted with permission from ref. 182. Copyright 2018, Wiley-VCH.

9. Layered double hydroxide-based inorganic-hybrid photocatalysts

Synthetic inorganic layered double hydroxides (LDHs) have become emerging and highly tunable layered structure materials, possessing a wide range of poly types and variable degree of stacking disorder with different interlayer distance from the same compositions by changing various anions. Recently, LDHs have attracted great attention in the area of photocatalysis, mainly toward the photo-oxidation of pollutants, photocatalytic water splitting, and CO2 photoreduction.184,185 LDH is a class of ionic solids mostly comprising earth-abundant elements (such as Mg2+, Al3+, Ni2+, Cu2+, Zn2+, etc.) with a chemical formula of [M2+1−xM3+x(OH)2](An)(x/nmH2O, where M2+ represents divalent cations and M3+ represents trivalent cations such as Al3+, while An represents anion species intercalating the interlayer space and m corresponds to the water amount through the natural electrostatic stacking between metal oxide layer structure bearing positive charge and counter-balancing negatively charged anion interlayers such as carbonate, hydroxyl group during the reaction.185–188 In recent years, several LDH materials have been widely utilized for photocatalytic CO2 reduction processes. In 2012, Teramura and coworkers introduced various LDHs (M2+ = Mg2+, Zn2+, Ni2+; M3+ = Al3+, Ga3+, In3+) for photocatalytic CO2 reduction.189 It was shown that among all the LDHs, Mg–In hydrotalcite has promising photocatalytic CO2 reduction activity. Interestingly they reported only selective CO and O2 formation without evolution of hydrogen, which reveals that surface base sites on LDHs have high water tolerance and superior activity for the photocatalytic conversion of CO2 like natural photosynthesis. We believe that the different layer type configuration has variable surface energy due to the edge effect, which lowers the ideal conduction band potential for hydrogen evolution reaction and suppresses the hydrogen evolution reaction. The edge sites of the LDH mainly provide the binding sites for CO2 adsorption and activation sites for photocatalytic CO2 reduction.

Zn–Cr LDH (Zn–Cr LDH) and noble metal (Pt, Pd, Au)-loaded Zn–Cr LDHs have been reported for promising photocatalytic conversion of CO2 into CO under UV irradiation.190 The 0.1 mass% Pt-loaded LDH exhibited 13 times higher photocatalytic performance than pristine Zn–Cr LDH. In 2016, Saliba et al. synthesized cadmium–aluminum LDH (Cd–Al LDH) microspheres by using the reaction–diffusion framework process for photocatalytic CO2 reduction.191 The prepared Cd–Al LDH exhibited a promising CO2 photoreduction performance and additional palladium nanoparticle-decorated counterparts showed a significant improvement as compared with CdAl LDH. Recently, Fujishima and co-workers have reported a series of Cu2O-loaded Zn–Cr LDHs for photoreduction of CO2.192 In the Zn–Cr LDHs, dispersed Cu2O nanoparticles mainly promoted the charge separation and provided the active sites for CO2 reduction and enhanced the photocatalytic performance. In aqueous-phase photocatalytic reaction, 0.1Cu2O@Zn1.8Cr LDH exhibited optimal activity for the conversion of CO2 into CO as compared with ternary Cu–Zn–Cr LDH and pristine Zn2Cr LDHs. Notably, Ahmed et al. reported copper-modified LDH as an active photocatalyst for conversion of CO2 in the presence of H2 gas.193 Based on the commercial P25 (TiO2: anatase and rutile), which has been widely used as a wide band gap semiconductor photocatalyst, a P25@CoAl-LDH heterojunction nanocomposite photocatalyst was synthesized by a hydrothermal process.194 The p-type Co–Al LDHs and n-type P25 hierarchical heterostructure showed efficient activity and selectivity (>90%) for CO2 to CO (Fig. 23). The reported low-cost P25@CoAl-LDH nanocomposites provided apparent quantum efficiency and stability for CO production.

image file: d0nr05884j-f23.tif
Fig. 23 (a) A P25@CoAl-LDH nanocomposite heterostructure photocatalyst and (b) aqueous-phase CO2 photoreduction to CO and H2 over P25 and Co Al-LDH references and P25@CoAl-LDH nanocomposites under UV–visible irradiation for 4 h. Reprinted with permission from ref. 194. Copyright 2015, Elsevier.

10. Summaries, strategies and future prospects

The carbon-based, MOF, COF, and LDH hybrid nanostructured photocatalysts have been widely explored for photocatalytic CO2 reduction, showing several specific advantages towards promising catalytic activity; however, they have a few limitations and some disadvantages. These advantages and disadvantages are summarized in Table 3. The selective ternary, binary and hybrid nanostructured photocatalytic performances clearly show that CO2 can be reduced to single-carbon (CO, HCOOH, CH4, CH3OH) and two-carbon (CH3CHO) hydrocarbons. Although the mechanisms of the photocatalytic CO2 reduction product formation and selectivity are not clear yet, several mechanism pathways have been proposed based on theoretical simulations and spectroscopic studies. These proposed mechanisms and the various hydrocarbon product formations are highly dependent on reaction intermediate formation and surface energy of the photocatalyst. However, detection of the reactive photocatalytic intermediates is highly challenging. Recently, several in situ IR spectroscopy studies clearly identified some of the promising photocatalytic CO2 reduction intermediates like CO*, CHO*, HCOO* and CH3* ions or radicals. However, more specific rigorous study is needed to establish these. Table 4 summarizes the photocatalytic CO2 reduction activity and different product formation from selected photocatalyst systems. It clearly shows that different light energy has been utilized for photocatalytic reduction processes and the product formation yield is not normalized in a uniform manner. In the photocatalytic CO2 reduction process, calculating the solar fuel formation in a universal way like AQY or photochemical quantum efficiency is highly desirable. Moreover, nanostructure-based photocatalysis is mainly a heterogeneous process and not providing specific catalytic sites. Therefore, TON or turnover frequency calculations are not feasible. We believe that establishing a typical solar fuel yield calculation protocol based on the total photon flux with respect to the evolved electrons is more promising. Additionally, considering carbon balance is another potential approach to understand the hidden mechanistic pathways.
Table 3 Advantages and disadvantages of selected nanostructured photocatalyst systems for photocatalytic CO2 reduction
Advantages Disadvantages
Carbon-based nanoarchitecture photocatalysts
• Semiconductor-like properties of 2D g-C3N4 and GO. • Low specific surface areas and rapid recombination of charge carriers of bare g-C3N4 and GO.
• Low cost and high chemical stability. • Unclear photocatalytic mechanism for these heterogeneous systems.
• Tunable band structures by controlling C and O ratio in GO. • In liquid-phase catalysis, tertiary amines (e.g., TEOA and TEA) are added as hole scavengers to improve the activity, which is neither economical nor green.
• Triazine functional groups in g-C3N4 exhibit strong ability for CO2 adsorption.
• Excellent activity through hybridization with other semiconductors (metal oxides and metal sulfides) or metals as co-catalysts to improve charge separation and for the better utilization of visible light.
MOF-based photocatalysts
• Inherent porosity and specific CO2 adsorption capacity. • Comparatively lower productivity and selectivity for CO2 photoreduction.
• Encapsulation of metal ions gives active metal centers as specific active sites. • Photocatalytic activity improved by introducing sacrificial hole scavengers like tertiary amines (e.g., TEOA or TEA), which is neither economical nor green.
• Wide range of different novel MOF photocatalysts via varying the metal clusters and organic linkers. • MOF-based semiconductor photocatalysts: unstable in aqueous solution or under UV light.
• Optical absorption can be tuned via inorganic metal clusters or organic linkers. • Low MOF yields limit the application of these materials on a large scale.
• Efficient charge separation via mesoporosity.
COF-based photocatalysts
• High CO2 adsorption capacity. • Poor photogenerated carrier lifetimes.
• Large surface areas and porosity facilitate CO2 adsorption, diffusion, and activation. • CO2 photoreduction: low efficiency and selectivity.
• Organic building blocks are connected through covalent bonds. • Innovative approaches are highly needed to improve catalytic efficiency.
• Stable photocatalytic materials with extended π-conjugation, which can act as photosensitizers.
CPP-based photocatalysts
• High surface areas and tunable pore sizes. • Non-persuasive mechanism for photocatalytic CO2 reduction over CPPs.
• Facilitate CO2 adsorption, diffusion, and activation. • Poor carrier separation.
• Metal-free tunable visible-light-active photocatalysts. • Less efficient and selective for CO2 photoreduction in terms of solar-to-chemical energy conversion.
• Conjugated frameworks facilitate electron transport.
LDH-based inorganic-hybrid photocatalysts
• High visible light absorption capacity of LDHs. • Lack of efficiency and stability.
• Not dependent on co-catalysts. • Unique synthetic limitations for divalent–trivalent LDHs.
• Tunable sandwich-like 2D homogeneous nanosheets. • Poor crystallinity and impure phases.
• Perfect structures for photogenerated carrier separation and transfer.

Table 4 A summary of the different types of photocatalysts for CO2 reduction discussed in this review
No. Catalyst Light source Phase Reaction conditions Products Ref.
1 Zn2GeO4 nanobelt 300 W xenon arc lamp Gas 0.1 g of catalyst, 230 ml glass reactor, ambient pressure CH4 74
2 ZnO/g-C3N4 500 W xenon arc lamp Liquid 0.01 g of catalyst, 132 ml SS reactor, 4 ml of DI water, CO2 pressure at 0.4 MPa CO, CH3OH, CH4 and C2H5OH 94
3 CeO2/g-C3N4 300 W xenon arc lamp Gas 0.05 g of catalyst, glass reactor (500 ml) CO and CH4 95
4 g-C3N4-N-TiO2 300 W xenon arc lamp Gas 0.1 g of catalyst, glass reactor (780 ml) CO and CH4 96
5 BiOI/g-C3N4 300 W xenon arc lamp Gas 0.1 g of catalyst, 180 ml stainless steel reactor CO, CH4, H2 and O2 97
6 WO3/g-C3N4 435 nm wavelength LED Liquid 0.003 g of catalyst and 5 ml of ion-exchanged water saturated with CO2 (Liquid: CH3OH and trace amounts of HCOOH) (Gas: H2, CO, and CH4) 98
7 g-C3N4/NaNbO3 nanowire 300 W xenon arc lamp Gas 0.005 g of catalyst, 230 ml Pyrex glass reactor CH4 99
8 C3N4/SnS2 300 W xenon arc lamp Gas 0.05 g of catalyst, 200 ml Pyrex glass vessel reactor, at ambient temperature and atmospheric pressure CH4 and CH3OH 107
9 Graphene oxide (GO) 300 W halogen lamp Gas 0.2 g of catalyst, 300 ml SS reactor CH3OH 112
10 Cu-NPs/GO 300 W halogen lamp Gas 0.1 g of catalyst, 300 ml SS reactor CH3OH and CH3CHO 113
11 Cu2O/rGO 150 W Xe lamp Liquid 0.5 g of photocatalysts and 3 ml of DI water in a CO2-purged 120 ml glass chamber, sodium sulfite as a hole scavenger CO 114
12 rGO@CuZnO@Fe3O4 20 W white cold LED light Liquid 0.1 g of catalyst, 100 ml glass vessel, DMF[thin space (1/6-em)]:[thin space (1/6-em)]water (45[thin space (1/6-em)]:[thin space (1/6-em)]5 v/v) CH3OH 116
13 Carbon-doped SnS2 (SnS2-C) 300 W halogen lamp Gas 0.1 g of photocatalysts, 300 ml stainless steel CH3CHO 119
14 Ti-Containing MOF (NH2-MIL-125(Ti)) 500 W Xe lamp with UV- and IR-cutoff filter Liquid 0.05 g of photocatalyst, MeCN[thin space (1/6-em)]:[thin space (1/6-em)]TEOA = 5[thin space (1/6-em)]:[thin space (1/6-em)]1 (v/v), alcohols as electron donor HCOOH 139
15 UiO-66 derivative (Zr4.3Ti1.7O4(OH)4(C8H7O4N)5.17(C8H8O4N2)0.83) 300 W Xe lamp Liquid 0.005 g of catalyst, 5 ml of a mixed solution of 4[thin space (1/6-em)]:[thin space (1/6-em)]1 (v/v) MeCN[thin space (1/6-em)]:[thin space (1/6-em)]TEOA (as a sacrificial base), and 0.1 M 1-benzyl-1,4-dihydronicotinamide (as a sacrificial reductant) HCOOH 143
16 Fe-Based MOFs (MIL-101(Fe), MIL-53(Fe), and MIL-88B(Fe)) and their amino-functionalized derivatives (NH2-MIL-101(Fe), NH2-MIL-53(Fe), and NH2-MIL-88B-(Fe)) 300 W Xe lamp with a UV- and IR-cutoff filter Liquid 0.05 g of catalyst, MeCN and TEOA solution (60 ml, 5/1 v/v) (TEOA as a sacrificial base) HCOOH 150
17 Cu-Porphyrin-based MOF 300 W Xe arc lamp Liquid 0.03 g of catalyst, 100 ml of water containing 1 ml of TEA, CO2 gas flow rate of 0.2 ml min−1 CH3OH 153
18 Ru-Polypyridine-containing MOFs: {Cd3[Ru-L1]2·2(Me2NH2)·solvent}n (Ru-MOF-1) and {Cd[Ru-L2]·(H2O)}n (Ru-MOF-2) 500 W Xe lamp (UV- and IR-cutoff filter) Liquid 0.04 g of catalyst, mixture of MeCN and TEOA HCOOH 156
19 [Ru(bpy)3]Cl2·6H2O(bpy = 2,2-bipyridine) and Co-ZIF-9 as a photosensitizer Xe lamp (420 nm cut-off) Liquid [Ru(bpy)3]Cl2·6H2O (10.0 mmol), Co-ZIF-9(0.8 mmol, activated), solvent (5 ml, acetonitrile/H2O = 4[thin space (1/6-em)]:[thin space (1/6-em)]1), TEOA (1 ml) as sacrificial electron donor, CO2 (1 atm) CO and H2 158
20 Re-COF 225 W Xe lamp (420 nm cut-off) Liquid 0.001 mg of catalyst, 11 ml septum sealed glass vials, 3 ml of CH3CN, and 0.2 ml of TEOA CO 173
21 Re(CO)3Cl in pyridine-based covalent triazine framework (Re-CTF-py) 300 W Xe lamp Gas 0.002 g of catalyst, 350 ml reactor, 2 ml of MeCN/TEOA (4/1), 0.8 atm of pure CO2 CO 174

Nanotechnology is helping to develop various nanostructured semiconductors or nanostructure-supported hybrid photocatalysts for CO2 reduction. Potentially, these nanostructures facilitate photon absorption, followed by carrier generation and separation, reduce the carrier recombination and improve carrier transport, as well as enhance the CO2 adsorption of the photocatalysts. A 1D nanostructure semiconductor shows distinct chemical and structural behavior and greater chemical reactivity than the bulk structure due to large surface-to-volume ratio and two-dimensional confinement. The narrow bandgap semiconductor heterostructures are bringing a revolution to heterogeneous photocatalysis. However, most of these crystalline semiconductor photocatalyst systems suffer due to their inherently low efficiency carrier transportation and non-optimum light energy utilization. We believe that the controlled morphology and components of hybrid semiconductor nanocatalysts with oxygen or sulfur vacancies are beneficial for carrier separation and transfer processes. Moreover a nanostructure template-supported semiconductor system possesses more amorphous characteristics, which reveals better gas adsorption and longer band tail for improved heterogeneous photocatalysis. However, a single component photocatalyst possesses insufficient association between the strong reduction and oxidation ability with sufficient absorption range. Therefore, development of highly efficient heterogeneous photocatalysts is one of the key challenges.

To improve the solar-to-fuel conversion efficiency, a deeper insight into the light–matter–CO2 interactions, photogenerated charge kinetics and reaction mechanisms is critical. Hence, nanostructuring a semiconductor to gain control over photogenerated charge pathway and specific defects is the major strategy to design a highly functional heterostructure photocatalyst system. Manipulating the photogenerated charge carriers between two different redox potential semiconductors via direct Z-scheme is highly desirable for constructing efficient heterostructure photocatalysts due to the advantages over conventional Type-II heterostructures. Hybrid heterostructures of two semiconductors often exhibit an interfacial lattice mismatch inherently, which induces a little strain at the interface region between two semiconductors. Through the development of proper interfacial mismatch by choosing suitable semiconductors to build strain-induced interfacial electric field, one might control photogenerated charge separation and create separate redox active sites at the heterostructure photocatalyst as shown in Fig. 24a. Among all the reported semiconductor photocatalysts, few of them selectively possesses as a strong a CO2 reduction and water oxidation ability based on their potential energy as those called reductive photocatalyst (PC-1, mainly metal sulfides) and oxidative photocatalyst (PC-2, mainly metal oxides). Moreover, the band configuration of PC-1 with lower work function and PC-2 with higher work function makes them favorable for reduction reaction with higher CB position and oxidation reaction with lower VB position. Therefore, in the progressive development of heterogeneous photocatalysts for improved CO2 reduction efficiency, tuning the structure, morphology and composition of 0D, 1D, 2D and 3D nanostructures with PC1 and PC-2 is one of the key strategies to regulate charge carrier separation, photocorrosion and transfer kinetics. Some possible nanostructured hybrid photocatalysts are shown in Fig. 24b.

image file: d0nr05884j-f24.tif
Fig. 24 (a) The formation mechanism of a direct Z-scheme semiconductor heterostructure photocatalyst using interfacial crystal strain. (b) Various hybrid nanostructure photocatalysts.

It was reported that the reduced band gaps and strong light–matter interactions in sulfides are favorable for harvesting solar energy. Moreover, high absorption coefficient allowing maximum solar energy utilization within the submicron length scale, as well as an efficient charge separation are also important factors for the high activity of a photocatalyst. Therefore, controlling the carrier diffusion pathways and the defects in the bulk or at the interfaces and surfaces of a nanostructure semiconductor are the key influencing factors for designing a highly efficient photocatalyst system. Transition metal dichalcogenides without dangling bonds are another potential factor for effective charge transfer characteristic. Therefore, we believe that owing to all the specific characteristics, transition metal dichalcogenides hold promise for the improvement of heterogeneous photochemical CO2 reduction. Therefor more research needs to focus on the design principles of versatile 2D materials together with heteroatom doping, semiconductor couple, sensitizer, co-catalyst and nano-architectures. It is expected that the defect manipulation at a 2D nanostructure interface could synergistically enhance the photocatalytic performance. Nevertheless, the formation of integrated 2D–2D semiconductor heterostructures is also highly promising to improve light absorption and photogenerated carrier separation at the interface for enhanced photocatalysis as shown in Fig. 25a.195 However, more detailed studies will be needed to overcome other challenges in the next step of development of photochemical CO2 reduction. So far, very few studies have focused on the effect of morphology in interfacial charge transfer of heterogeneous photocatalysts on photocatalytic CO2 reduction. Therefore, for better understanding of the nanostructure-based photocatalysis mechanism, development of some facile in situ photochemical approach is required to study effective electron transfer to surface-active sites for multi-step CO2 reduction.

image file: d0nr05884j-f25.tif
Fig. 25 (a) The carrier diffusion pathway of 2D–2D and 2D–0D heterostructures. (b) A hybrid design for gas-phase photocatalytic CO2 reduction.

Gas-phase photocatalytic CO2 reduction is a more feasible process as compared with the aqueous slurry-based reaction because the gas-phase heterogeneous process can easily suppress the hydrogen evolution reaction and overcome the CO2 solubility issue. However, the reduction and oxidation reactions on the same catalyst surface spontaneously affect the photocatalytic reaction due to inevitable unwanted reactions. Besides, the radical-based reaction mechanism in gas-phase photocatalytic CO2 reduction is highly favorable for the formation of higher carbon containing products. Therefore, in order to perform an isolated photo-oxidation and photoreduction reaction we propose the design of macrostructure photocatalyst systems where a hybrid heterojunction light absorber is sandwiched between separate reductive (SC1) and oxidative (SC2) catalyst as schematically illustrated in Fig. 25b. The separated redox reaction utilizing the higher CB position and lower VB position for individual semiconductor with highly efficient charge kinetics would synergistically enhance the efficiency of multistep CO2 reduction.

11. Conclusions

Nanostructured semiconductors or nano-architecture-supported photocatalysts have drawn significant scientific and technological attention in the field of photocatalysis due to their unique beneficial optoelectronic properties. Among conventional strategies, like doping, sensitization, and heterostructuring, nanostructuring has unprecedented potential for the further enhancement of the existing semiconductor photocatalyst performance for CO2 reduction to solar fuels. Using the nanostructuring process, semiconductor photocatalysts with different nanoscale structures (e.g., nanopores, nanotubes, nanorods, nanosheets, etc.) have been developed; this mainly helps to improve surface-active areas and optimize the charge separation process, and directly affects the photocatalysis process. The superior optoelectronic performance of narrow band gap semiconductor nanotubes or nanorods has been attributed to 1D channel-assisted charge separation and transport, leading to appreciably reduced electron–hole recombination. Semiconductor carrier diffusion length tuning by the length scale of 1D or 2D nanostructures together with the length and separation between the tubes is effective for controlling the recombination of photogenerated carriers for heterogeneous photocatalytic CO2 reduction. The advantages of nanostructured materials are many, ranging from increasing the surface areas of semiconductor photocatalysts to enhancing light absorption and promoting gas molecule adsorption during heterogeneous photocatalysis. This review, though mainly focused on the development of a few nano-architectured semiconductor photocatalysts, will progressively open up new opportunities for photocatalytic CO2 reduction with industrial applications.

Conflicts of interest

There are no conflicts to declare.


S. C. S. wishes to gratefully acknowledge the Council of Scientific and Industrial Research CSIR, New Delhi, for his senior research fellowship. I. S. acknowledges financial startup grant support from Amrita Centre for Nanosciences and Molecular Medicine (ACNSMM), Amrita Vishwa Vidyapeetham, Kochi, India. J. M. gratefully acknowledges financial support from the Department of Science and Technology, India, for DST-INSPIRE Faculty Research project grant (GAP-0522) and INSA-JSPS Fellowship at CSIR-IICT, Hyderabad. We acknowledge DKIM of IICT (Division of Knowledge and Information Management) for plagiarism checking and providing us with the manuscript communication number: IICT/Pubs./2019/203. This study was financial supported by the Ministry of Science and Technology (MOST) in Taiwan, under the Academic Summit Project 107-2745-M-002-001-ASP and the Science Vanguard Project 108-2119-M-002-030. Financial support from the i-MATE program in Academia Sinica, and the Center of Atomic Initiative for New Materials (AI-Mat), National Taiwan University (107L9008 and 108L9008), from the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan is also acknowledged.


  1. IEA report CO2 emissions from fuel combustion 2018, 2018. ISBN 978-92-64-27819-6.
  2. IEA World Energy Outlook 2017, 2017. ISBN 978-92-64-28230-8.
  3. S. Das, J. P. Ramírez, J. Gong, N. Dewangan, K. Hidajat, B. C. Gates and S. Kawi, Chem. Soc. Rev., 2020, 49, 2937–3004 RSC.
  4. T. F. Hurst, T. T. Cockerill and N. H. Florin, Energy Environ. Sci., 2012, 5, 7132–7150 RSC.
  5. T. Xia, R. Long, C. Gao and Y. Xiong, Nanoscale, 2019, 11, 11064–11070 RSC.
  6. L. Wei, J. Lin, S. Xie, W. Ma, Q. Zhang, Z. Shena and Y. Wang, Nanoscale, 2019, 11, 12530–12536 RSC.
  7. S. Sultana, C. P. Sahoo, S. Martha and K. Parida, RSC Adv., 2016, 6, 44170–44194 RSC.
  8. A. E. Ashley, A. L. Thompson and D. O'Hare, Angew. Chem., 2009, 121, 10023–10027 CrossRef.
  9. J. A. Rodriguez, P. Liu, D. J. Stacchiola, S. D. Senanayake, M. G. Whiteand and J. G. Chen, ACS Catal., 2015, 5, 6696–6706 CrossRef CAS.
  10. S. G. Jadhav, P. D. Vaidya, B. M. Bhanage and J. B. Joshi, Chem. Eng. Res. Des., 2014, 92, 2557–2567 CrossRef CAS.
  11. F. Li, D. R. MacFarlane and J. Zhang, Nanoscale, 2018, 10, 6235–6260 RSC.
  12. H. Zhang, Y. Zhang, Y. Li, S. Ahn, G. T. R. Palmore, J. Fu, A. A. Peterson and S. Sun, Nanoscale, 2019, 11, 12075–12079 RSC.
  13. Y. Bao, J. Wang, Q. Wang, X. Cui, R. Long and Z. Li, Nanoscale, 2020, 12, 2507–2514 RSC.
  14. C. Lu, D. S. Itanze, A. G. Aragon, X. Ma, H. Li, K. B. Ucer, C. Hewitt, D. L. Carroll, R. T. Williams, Y. Qiu and S. M. Geyer, Nanoscale, 2020, 12, 2987–2991 RSC.
  15. F.-Y. Fu, I. Shown, C.-S. Li, P. Raghunath, T.-Y. Lin, T. Billo, H.-L. Wu, C.-I. Wu, P.-W. Chung, M.-C. Lin, L.-C. Chen and K.-H. Chen, ACS Appl. Mater. Interfaces, 2019, 11, 25186–25194 CrossRef CAS.
  16. G. Chen, R. Gao, Y. Zhao, Z. Li, G. I. N. Waterhouse, R. Shi, J. Zhao, M. Zhang, L. Shang, G. Sheng, X. Zhang, X. Wen, L.–Z. Wu, C.–H. Tung and T. Zhang, Adv. Mater., 2018, 30, 1704663 CrossRef.
  17. Y. Zhao, G. I. N. Waterhouse, G. Chen, X. Xiong, L.-Z. Wu, C.-H. Tung and T. Zhang, Chem. Soc. Rev., 2019, 48, 1972–2010 RSC.
  18. T. Billo, I. Shown, A. K. Anbalagan, T. A. Effendi, A. Sabbah, F.-Y. Fu, C.-M. Chu, W.-Y. Woon, R.-S. Chen, C.-H. Lee, K.-H. Chen and L.-C. Chen, Nano Energy, 2020, 72, 104717 CrossRef CAS.
  19. S. Wu, H. Pang, W. Zhou, B. Yang, X. Meng, X. Qiu, G. Chen, L. Zhang, S. Wang, X. Liu, R. Ma, J. Ye and N. Zhang, Nanoscale, 2020, 12, 8693–8700 RSC.
  20. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38 CrossRef CAS.
  21. T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature, 1979, 277, 637–638 CrossRef CAS.
  22. S. Navalón, A. Dhakshinamoorthy, M. Álvaro and H. Garcia, ChemSusChem, 2013, 6, 562–577 CrossRef.
  23. P. Kar, S. Farsinezhad, X. Zhang and K. Shankar, Nanoscale, 2014, 6, 14305–14318 RSC.
  24. A. Manzi, T. Simon, C. Sonnleitner, M. Döblinger, R. Wyrwich, O. Stern, J. K. Stolarczyk and J. Feldmann, J. Am. Chem. Soc., 2015, 137, 14007–14010 CrossRef CAS.
  25. E. A. Kozlova, M. N. Lyulyukin, D. V. Markovskaya, D. S. Selishchev, S. V. Cherepanova and D. V. Kozlovab, Photochem. Photobiol. Sci., 2019, 18, 871–877 RSC.
  26. C. M. Mömming, E. Otten, G. Kehr, R. Fröhlich, S. Grimme, D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2009, 48, 6643–6646 CrossRef.
  27. M. D. Rakowski and D. L. Dubois, Acc. Chem. Res., 2009, 42, 1974–1982 CrossRef.
  28. S. Li, K. Ji, M. Zhang, C. He, J. Wang and Z. Li, Nanoscale, 2020, 12, 9533–9540 RSC.
  29. T. Wu, C. Zhu, D. Han, Z. Kang and L. Niu, Nanoscale, 2019, 11, 22980–22988 RSC.
  30. H. Takeda, C. Cometto, O. Ishitaniand and M. Robert, ACS Catal., 2017, 7, 70–88 CrossRef CAS.
  31. H. Takeda, H. Koizumi, K. Okamoto and O. Ishitani, Chem. Commun., 2014, 50, 1491–1493 RSC.
  32. T. Morimoto, C. Nishiura, M. Tanaka, J. Rohacova, Y. Nakagawa, Y. Funada, K. Koike, Y. Yamamoto, S. Shishido, T. Kojima, T. Saeki, T. Ozeki and O. Ishitani, J. Am. Chem. Soc., 2013, 135, 13266–13269 CrossRef CAS.
  33. G. Sahara and O. Ishitani, Inorg. Chem., 2015, 54, 5096–5104 CrossRef CAS.
  34. H. Takeda, H. Kamiyama, K. Okamoto, M. Irimajiri, T. Mizutani, K. Koike, A. Sekine and O. Ishitani, J. Am. Chem. Soc., 2018, 140, 17241–17254 CrossRef CAS.
  35. K. Li, B. Peng and T. Peng, ACS Catal., 2016, 6, 7485–7527 CrossRef CAS.
  36. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Chem. Rev., 1995, 95, 69–96 CrossRef CAS.
  37. S. Xie, Q. Zhang, G. Liu and Y. Wang, Chem. Commun., 2016, 52, 35–59 RSC.
  38. K. Li, X. An, K. H. Park, M. Khraisheh and J. Tang, Catal. Today, 2014, 224, 3–12 CrossRef CAS.
  39. J. Mao, K. Li and T. Peng, Catal. Sci. Technol., 2013, 3, 2481–2498 RSC.
  40. Y. Izumi, Coord. Chem. Rev., 2013, 257, 171–186 CrossRef CAS.
  41. J. Yang, D. Wang, H. Han and C. Li, Acc. Chem. Res., 2013, 46, 1900–1909 CrossRef CAS.
  42. J. Hong, W. Zhang, J. Ren and R. Xu, Anal. Methods, 2013, 5, 1086–1097 RSC.
  43. Y. Chen, D. Wang, X. Deng and Z. Li, Catal. Sci. Technol., 2017, 7, 4893–4904 RSC.
  44. R. Li, W. Zhang and K. Zhou, Adv. Mater., 2018, 30, 1705512 CrossRef.
  45. A. Kidanemariam, J. Lee and J. Park, Polymers, 2019, 11, 2090 CrossRef CAS.
  46. J. Ozdemir, I. Mosleh, M. Abolhassani, L. F. Greenlee, R. R. Beitle and M. H. Beyzavi, Front. Energy, 2019, 7, 77 CrossRef.
  47. H. Shen, T. Peppel, J. Stunk and Z. Sun, Sol. RRL, 2020, 1900546 CrossRef CAS.
  48. S. Kumar, S. Karthikeyan and A. F. Lee, Catalysts, 2018, 8, 74 CrossRef.
  49. T. Xia, R. Long, C. Gao and Y. Xiong, Nanoscale, 2019, 11, 11064–11070 RSC.
  50. S. R. Lingampalli, M. M. Ayyub and C. N. R. Rao, ACS Omega, 2017, 2, 2740–2748 CrossRef CAS.
  51. E. Kalamaras, M. M. M. Valer, M. Shao, J. Xuan and H. Wang, Catal. Today, 2018, 317, 56–75 CrossRef CAS.
  52. K. Sivula and R. v. d. Krol, Nat. Rev. Mater., 2016, 1, 15010 CrossRef CAS.
  53. S. Castro, J. Albo and A. Irabien, ACS Sustainable Chem. Eng., 2018, 6, 15877–15894 CrossRef CAS.
  54. A. U. Pawar, C. W. Kim, M.-T. N. Le and Y. S. Kang, ACS Sustainable Chem. Eng., 2019, 7, 7431–7455 CrossRef CAS.
  55. J. T. Ren, Y.-L. Zheng, K. Yuan, L. Zhou, K. Wua and Y.-W. Zhang, Nanoscale, 2020, 12, 755–762 RSC.
  56. M. Nasr, C. Eid, R. Habchi, P. Miele and M. Bechelany, ChemSusChem, 2018, 11, 3023–3047 CrossRef CAS.
  57. S. Sato, T. Morikawa, S. Saeki, T. Kajino and T. Motohiro, Angew. Chem., Int. Ed., 2010, 49, 5101–5105 CrossRef CAS.
  58. L. L. Tan, S. P. Chai and A. R. Mohamed, ChemSusChem, 2012, 5, 1868–1882 CrossRef CAS.
  59. T. Arai, S. Tajima, S. Sato, K. Uemura, T. Morikawa and T. Kajino, Chem. Commun., 2011, 47, 12664–12666 RSC.
  60. Y. S. Chaudhary, T. W. Woolerton, C. S. Allen, J. H. Warner, E. Pierce, S. W. Ragsdale and F. A. Armstrong, Chem. Commun., 2012, 48, 58–60 RSC.
  61. M. Tahir, J. CO2 Util., 2020, 37, 134–146 CrossRef CAS.
  62. S. Feng, J. Zhao, Y. Bai, X. Liang, T. Wang and C. Wang, J. CO2 Util., 2020, 38, 1–9 CrossRef CAS.
  63. B. Khan, F. Raziq, M. B. Faheem, M. U. Farooq, S. Hussain, F. Ali, A. Ullah, A. Mavlonov, Y. Zhao, Z. Liu, H. Tian, H. Shen, X. Zu, S. Li, H. Xiao, X. Xiang and L. Qiao, J. Hazard. Mater., 2020, 381, 120972 CrossRef CAS.
  64. J. Low, J. Yu and W. Ho, J. Phys. Chem. Lett., 2015, 6, 4244–4251 CrossRef CAS.
  65. L. Liu, H. Zhao, J. M. Andino and Y. Li, ACS Catal., 2012, 2, 1817–1828 CrossRef CAS.
  66. W. Jiao, L. Wang, G. Liu, G. Q. Lu and H. M. Cheng, ACS Catal., 2012, 2, 1854–1859 CrossRef CAS.
  67. J. Yu, J. Low, W. Xiao, P. Zhou and M. Jaroniec, J. Am. Chem. Soc., 2014, 136, 8839–8842 CrossRef CAS.
  68. T. Billo, F. Y. Fu, P. Raghunath, I. Shown, W. F. Chen, H. T. Lien, T. H. Shen, J. F. Lee, T. S. Chan, K. Y. Huang, C. I. Wu, M. C. Lin, J. S. Hwang, C. H. Lee, L. C. Chen and K. H. Chen, Small, 2018, 14, 1702928 CrossRef.
  69. S. Sun, M. Watanabe, J. Wu, Q. An and T. Ishihara, J. Am. Chem. Soc., 2018, 140, 6474–6482 CrossRef CAS.
  70. J. Jin, J. Yu, D. Guo, C. Cui and W. Ho, Small, 2015, 11, 5262–5271 CrossRef CAS.
  71. J. Low, B. Dai, T. Tong, C. Jiang and J. Yu, Adv. Mater., 2019, 31, 1802981 CrossRef.
  72. F. Xu, J. Zhang, B. Zhu, J. Yu and J. Xu, Appl. Catal., B, 2018, 230, 194–202 CrossRef CAS.
  73. W. Tu, Y. Zhou and Z. Zou, Adv. Mater., 2014, 26, 4607–4626 CrossRef CAS.
  74. Q. Liu, Y. Zhou, J. Kou, X. Chen, Z. Tian, J. Gao, S. Yan and Z. Zou, J. Am. Chem. Soc., 2010, 132, 14385–14387 CrossRef CAS.
  75. S. Yan, L. Wan, Z. Li and Z. Zou, Chem. Commun., 2011, 47, 5632–5634 RSC.
  76. A. Pant, Z. Mutlu, D. Wickramaratne, H. Cai, R. K. Lake, C. Ozkan and S. Tongay, Nanoscale, 2016, 8, 3870–3887 RSC.
  77. K. Sasan, Q. Lin, C. Maob and P. Feng, Nanoscale, 2016, 8, 10913–10916 RSC.
  78. X. Jiao, Z. Chen, X. Li, Y. Sun, S. Gao, W. Yan, C. Wang, Q. Zhang, Y. Lin, Y. Luo and Y. Xie, J. Am. Chem. Soc., 2017, 139, 7586–7594 CrossRef CAS.
  79. X. Li, Y. Sun, Ji. Xu, Y. Shao, J. Wu, X. Xu, Y. Pan, H. Ju, J. Zhu and Y. Xie, Nat. Energy, 2019, 4, 690–699 CrossRef CAS.
  80. S. Wang, B. Y. Guan, Y. Lu and X. W. Lou, J. Am. Chem. Soc., 2017, 139, 17305–17308 CrossRef CAS.
  81. B. Su, L. Huang, Z. Xiong, Y. Yang, Y. Hou, Z. Ding and S. Wang, J. Mater. Chem. A, 2019, 7, 26877–26883 RSC.
  82. Z. Xiong, L. Huang, J. Peng, Y. Hou, Z. Ding and S. Wang, ChemCatChem, 2019, 11, 5513–5518 CrossRef CAS.
  83. L. Huang, B. Li, B. Su, Z. Xiong, C. Zhang, Y. Hou, Z. Ding and S. Wang, J. Mater. Chem. A, 2020, 8, 7177–7183 RSC.
  84. Y. Gong, J. Wang, Z. Wei, P. Zhang, H. Li and Y. Wang, ChemSusChem, 2014, 7, 2303–2309 CrossRef CAS.
  85. X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. Hou, X. Fu and M. Antonietti, J. Am. Chem. Soc., 2009, 131, 1680–1681 CrossRef CAS.
  86. Y. Gong, M. Li and Y. Wang, ChemSusChem, 2015, 8, 931–946 CrossRef CAS.
  87. F. Dong, L. W. Wu, Y. J. Sun, M. Fu, Z. B. Wu and S. C. Lee, J. Mater. Chem., 2011, 21, 15171–15174 RSC.
  88. L. Ge, F. Zuo, J. Liu, Q. Ma, C. Wang, D. Sun, L. Bartels and P. Feng, J. Phys. Chem. C, 2012, 116, 13708–13714 CrossRef CAS.
  89. X. Bai, S. Yan, J. Wang, L. Wang, W. Jiang, S. Wu, C. Sun and Y. Zhu, J. Mater. Chem. A, 2014, 2, 17521–17529 RSC.
  90. Z. Sun, W. Fang, L. Zhao and H. Wang, Appl. Surf. Sci., 2020, 504, 144347 CrossRef CAS.
  91. Y. Xia, K. Xiao, B. Cheng, J. Yu, L. Jiang, M. Antonietti and S. Cao, ChemSusChem, 2020, 13, 1730–1734 CrossRef CAS.
  92. R. Neumann, H. Yu and E. Haviv, Angew. Chem., Int. Ed., 2020, 59, 6219–6223 CrossRef.
  93. W. J. Ong, L. K. Putri, L. L. Tan, S. P. Chai and S. T. Yong, Appl. Catal., B, 2016, 180, 530–543 CrossRef CAS.
  94. Y. He, Y. Wang, L. Zhang, B. Teng and M. Fan, Appl. Catal., B, 2015, 168, 1–8 Search PubMed.
  95. M. Li, L. Zhang, M. Wu, Y. Du, X. Fan, M. Wang, L. Zhang, Q. Kong and J. Shi, Nano Energy, 2016, 19, 145–155 CrossRef CAS.
  96. S. Zhou, Y. Liu, J. Li, Y. Wang, G. Jiang, Z. Zhao, D. Wang, A. Duan, J. Liu and Y. Wei, Appl. Catal., B, 2014, 158, 20–29 CrossRef.
  97. J. C. Wang, H. C. Yao, Z. Y. Fan, L. Zhang, J. S. Wang, S. Q. Zangand and Z. J. Li, ACS Appl. Mater. Interfaces, 2016, 8, 3765–3775 CrossRef CAS.
  98. T. Ohno, N. Murakami, T. Koyanagi and Y. Yang, J. CO2 Util., 2014, 6, 17–25 CrossRef CAS.
  99. H. Shi, G. Chen, C. Zhang and Z. Zou, ACS Catal., 2014, 4, 3637–3643 CrossRef CAS.
  100. Y. He, L. Zhang, M. Fan, X. Wang, M. L. Walbridge, Q. Nong, Y. Wu and L. Zhao, Sol. Energy Mater. Sol. Cells, 2015, 137, 175–184 CrossRef CAS.
  101. H. Guo, M. Chen, Q. Zhong, Y. Wang, W. Ma and J. Ding, J. CO2 Util., 2019, 33, 233–241 CrossRef CAS.
  102. J. Low, J. Yu, M. Jaroniec, S. Wageh and A. A. Al-Ghamdi, Adv. Mater., 2017, 29, 1601694 CrossRef.
  103. M. Grätzel, Nature, 2001, 414, 338–344 CrossRef.
  104. Q. Xu, L. Zhang, B. Cheng, J. Fan and J. Yu, Chem, 2020, 6, 1543–1559 CAS.
  105. Z. Wang, Y. Chen, L. Zhang, B. Cheng, J. Yu and J. Fan, J. Mater. Sci. Technol., 2020, 56, 143–150 CrossRef.
  106. F. He, B. Zhu, B. Cheng, J. Yu, W. Ho and W. Macyk, Appl. Catal., B, 2020, 272, 119006 CrossRef CAS.
  107. T. Di, B. Zhu, B. Cheng, J. Yu and J. Xu, J. Catal., 2017, 352, 532–541 CrossRef CAS.
  108. R. Kuriki, K. Sekizawa, O. Ishitani and K. Maeda, Angew. Chem., Int. Ed., 2015, 54, 2406–2409 CrossRef CAS.
  109. R. Kuriki, H. Matsunaga, T. Nakashima, K. Wada, A. Yamakata, O. Ishitani and K. Maeda, J. Am. Chem. Soc., 2016, 138, 5159–5170 CrossRef CAS.
  110. D. Wu, F. Zhang, H. Liang and X. Feng, Chem. Soc. Rev., 2012, 41, 6160–6177 RSC.
  111. T. F. Yeh, J. M. Syu, C. Cheng, T. H. Chang and H. Teng, Adv. Funct. Mater., 2010, 20, 2255–2262 CrossRef CAS.
  112. H. C. Hsu, I. Shown, H. Y. Wei, Y. C. Chang, H. Y. Du, Y. G. Lin, C. A. Tseng, C. H. Wang, L. C. Chen, Y. C. Lin and K. H. Chen, Nanoscale, 2013, 5, 262–268 RSC.
  113. I. Shown, H. C. Hsu, Y. C. Chang, C. H. Lin, P. K. Roy, A. Ganguly, C. H. Wang, J. K. Chang, C. I. Wu, L. C. Chen and K. H. Chen, Nano Lett., 2014, 14, 6097–6103 CrossRef CAS.
  114. X. An, K. Liand and J. Tang, ChemSusChem, 2014, 7, 1086–1093 CrossRef CAS.
  115. L. Zhang, N. Li, H. Jiu, G. Qi and Y. Huang, Ceram. Int., 2015, 41, 6256–6262 CrossRef CAS.
  116. P. Kumar, C. Joshi, A. Barras, B. Sieber, A. Addad, L. Boussekey, S. Szunerits, R. Boukherroub and S. L. Jain, Appl. Catal., B, 2017, 205, 654–665 CrossRef CAS.
  117. D. Xu, B. Cheng, W. Wang, C. Jiang and J. Yu, Appl. Catal., B, 2018, 231, 368–380 CrossRef CAS.
  118. Y.-X. Pan, Y. You, S. Xin, Y. Li, G. Fu, Z. Cui, Yu.-L. Men, F.-F. Cao, S.-H. Yu and J. B. Goodenough, J. Am. Chem. Soc., 2017, 139, 4123–4129 CrossRef CAS.
  119. I. Shown, S. Samireddi, Y.-C. Chang, R. Putikam, P. H. Chang, A. Sabbah, F. Y. Fu, W. F. Chen, C. Wu, T. Y. Yu, P. W. Chung, M. C. Lin, L. C. Chen and K. H. Chen, Nat. Commun., 2018, 9, 169 CrossRef.
  120. J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450–1459 RSC.
  121. C. Y. Sun, S. X. Liu, D. D. Liang, K. Z. Shao, Y. H. Ren and Z. M. Su, J. Am. Chem. Soc., 2009, 13, 1883–1888 CrossRef.
  122. P. Koley, S. C. Shit, B. Joseph, S. Pollastri, Y. M. Sabri, E. L. H. Mayes, N. Lingaiah, J. Tardio and J. Mondal, ACS Appl. Mater. Interfaces, 2020, 12, 21682–21700 CrossRef CAS.
  123. J. R. Li, J. Yu, W. Lu, L. B. Sun, J. Sculley, P. B. Balbuena and H. C. Zhou, Nat. Commun., 2013, 4, 1538 CrossRef.
  124. J. R. Li, J. Sculley and H. C. Zhou, Chem. Rev., 2011, 112, 869–932 CrossRef.
  125. J. R. Li, R. J. Kuppler and H. C. Zhou, Chem. Soc. Rev., 2009, 38, 1477–1504 RSC.
  126. N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim, M. O'Keeff and O. M. Yaghi, Science, 2003, 300, 1127–1129 CrossRef CAS.
  127. J. L. Rowsell and O. M. Yaghi, Angew. Chem., Int. Ed., 2005, 44, 4670–4679 CrossRef CAS.
  128. S. Ma and H. C. Zhou, Chem. Commun., 2010, 46, 44–53 RSC.
  129. K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T. H. Bae and J. R. Long, Chem. Rev., 2011, 112, 724–781 CrossRef.
  130. A. R. Millward and O. M. Yaghi, J. Am. Chem. Soc., 2005, 127, 17998–17999 CrossRef CAS.
  131. J. M. Simmons, H. Wu, W. Zhou and T. Yildirim, Energy Environ. Sci., 2011, 4, 2177–2185 RSC.
  132. S. C. Shit, I. Mondal, S. Pendem, L. Bai, J. Y. Park and J. Mondal, ChemElectroChem, 2018, 5, 2842–2849 CrossRef CAS.
  133. H. Yoneyama, Catal. Today, 1997, 39, 169–175 CrossRef CAS.
  134. T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature, 1979, 277, 637 CrossRef CAS.
  135. H. Takeda, K. Koike, H. Inoue and O. Ishitani, J. Am. Chem. Soc., 2008, 130, 2023–2031 CrossRef CAS.
  136. J. Mao, K. Liand and T. Peng, Catal. Sci. Technol., 2013, 3, 2481–2498 RSC.
  137. L. Zhao, Z. Zhao, Y. Li, X. Chu, Z. Li, Y. Qu, L. Bai and L. Jing, Nanoscale, 2020, 12, 10010–10018 RSC.
  138. J.-T. Ren, Y.-L. Zheng, K. Yuan, L. Zhou, K. Wua and Y.-W. Zhang, Nanoscale, 2020, 12, 755–762 RSC.
  139. Y. Fu, D. Sun, Y. Chen, R. Huang, Z. Ding, X. Fu and Z. Li, Angew. Chem., Int. Ed., 2012, 51, 3364–3367 CrossRef CAS.
  140. D. Sun, W. Liu, Y. Fu, Z. Fang, F. Sun, X. Fu, Y. Zhang and Z. Li, Chem. – Eur. J., 2014, 20, 4780–4788 CrossRef CAS.
  141. Y. Fu, H. Yang, R. Du, G. Tu, C. Xu, F. Zhang, M. Fan and W. Zhu, RSC Adv., 2017, 7, 42819–42825 RSC.
  142. J. H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga and K. P. Lillerud, J. Am. Chem. Soc., 2008, 130, 13850–13851 CrossRef.
  143. Y. Lee, S. Kim, J. K. Kang and S. M. Cohen, Chem. Commun., 2015, 51, 5735–5738 RSC.
  144. A. S. Portillo, H. G. Baldovi, M. T. G. Fernandez, S. Navalon, P. Atienzar, B. Ferrer, M. Alvaro, H. Garcia and Z. Li, J. Phys. Chem. C, 2017, 121, 7015–7024 CrossRef.
  145. Y. Lee, S. Kim, H. Fei, J. K. Kangand and S. M. Cohen, Chem. Commun., 2015, 51, 16549–16552 RSC.
  146. C. Wang, Z. G. Xie, K. E. deKrafft and W. B. Lin, J. Am. Chem. Soc., 2011, 133, 13445–13454 CrossRef CAS.
  147. K. M. Choi, D. Kim, B. Rungtaweevoranit, C. A. Trickett, J. T. D. Barmanbek, A. S. Alshammari, P. D. Yang and O. M. Yaghi, J. Am. Chem. Soc., 2017, 139, 356–362 CrossRef CAS.
  148. M. B. Chambers, X. Wang, N. Elgrishi, C. H. Hendon, A. Walsh, J. Bonnefoy, J. Canivet, E. A. Quadrelli, D. Farrusseng, C. M. Draznieks and M. Fontecave, ChemSusChem, 2015, 8, 603–608 CrossRef CAS.
  149. X. Gao, B. Guo, C. Guo, Q. Meng, J. Liang and J. Liu, ACS Appl. Mater. Interfaces, 2020, 12, 24059–24065 CrossRef CAS.
  150. D. K. Wang, R. K. Huang, W. J. Liu, D. R. Sun and Z. H. Li, ACS Catal., 2014, 4, 4254–4260 CrossRef CAS.
  151. X. Y. Dao, J. H. Guo, Y. P. Wei, F. Guo, Y. Liu and W. Y. Sun, Inorg. Chem., 2019, 58, 8517–8524 CrossRef CAS.
  152. X. Y. Dao, X. F. Xie, J. H. Guo, X. Y. Zhang, Y. S. Kang and W. Y. Sun, ACS Appl. Energy Mater., 2020, 3(4), 3946–3954 CrossRef CAS.
  153. Y. Liu, Y. Yang, Q. Sun, Z. Wang, B. Huang, Y. Dai, X. Qin and X. Zhang, ACS Appl. Mater. Interfaces, 2012, 5, 7654–7658 CrossRef.
  154. R. Li, J. Hu, M. Deng, H. Wang, X. Wang, Y. Hu, H. L. Jiang, J. Jiang, Q. Zhang, Y. Xie and Y. Xiong, Adv. Mater., 2014, 26, 4783–4788 CrossRef CAS.
  155. X. He, Z. R. Gan, S. Fisenko, D. W. Wang, H. M. El-Kaderi and W. N. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 9688–9698 CrossRef CAS.
  156. S. Q. Zhang, L. N. Li, S. G. Zhao, Z. H. Sunand and J. H. Luo, Inorg. Chem., 2015, 54, 8375–8379 CrossRef CAS.
  157. M E. Mahmoud, H. Audi, A. Assoud, T. H. Ghaddar and M. Hmadeh, J. Am. Chem. Soc., 2019, 141, 7115–7121 CrossRef.
  158. S. Wang, W. Yao, J. Lin, Z. Ding and X. Wang, Angew. Chem., Int. Ed., 2014, 53, 1034–1038 CrossRef CAS.
  159. S. Wang, J. L. Lin and X. Wang, Phys. Chem. Chem. Phys., 2014, 16, 14656–14660 RSC.
  160. S. Wang and X. Wang, Appl. Catal., B, 2015, 162, 494–500 CrossRef CAS.
  161. Z. C. Kong, J. F. Liao, Y. J. Dong, Y. F. Xu, H. Y. Chen, D. B. Kuang and C. Y. Su, ACS Energy Lett., 2018, 3, 2656–2662 CrossRef CAS.
  162. P. J. Waller, F. Gándaraand and O. M. Yaghi, Acc. Chem. Res., 2015, 48, 3053–3063 CrossRef CAS.
  163. N. Huang, P. Wang and D. Jiang, Nat. Rev. Mater., 2016, 1, 16068 CrossRef CAS.
  164. H. Furukawa and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 8875–8883 CrossRef CAS.
  165. H. Xu, X. Chen, J. Gao, J. Lin, M. Addicoat, S. Irle and D. Jiang, Chem. Commun., 2014, 50, 1292–1294 RSC.
  166. H. Oh, S. B. Kalidindi, Y. Um, S. Bureekaew, R. Schmid, R. A. Fischerand and M. Hirscher, Angew. Chem., Int. Ed., 2013, 52, 13219–13222 CrossRef CAS.
  167. C. R. Deblase, K. E. Silberstein, T.-T. Truong, H. D. Abruña and W. R. Dichtel, J. Am. Chem. Soc., 2013, 135, 16821–16824 CrossRef CAS.
  168. S. Chandra, T. Kundu, S. Kandambeth, R. BabaRao, Y. Marathe, S. M. Kunjir and R. Banerjee, J. Am. Chem. Soc., 2014, 136, 6570–6573 CrossRef CAS.
  169. L. Stegbauer, K. Schwinghammerand and B. V. Lotsch, Chem. Sci., 2014, 5, 2789–2793 RSC.
  170. X. Ding, J. Guo, X. Feng, Y. Honsho, J. Guo, S. Seki, P. Maitarad, A. Saeki, S. Nagase and D. Jiang, Angew. Chem., Int. Ed., 2011, 50, 1289–1293 CrossRef CAS.
  171. Z. Fu, X. Wang, A. M. Gardner, X. Wang, S. Y. Chong, G. Neri, A. J. Cowan, L. Liu, X. Li, A. Vogel, R. Clowes, M. Bilton, L. Chen, R. S. Sprick and A. I. Cooper, Chem. Sci., 2020, 11, 543–550 RSC.
  172. R. K. Yadav, A. Kumar, N. J. Park, K. J. Kong and J. O. Baeg, J. Mater. Chem. A, 2016, 4, 9413–9418 RSC.
  173. S. Yang, W. Hu, X. Zhang, P. He, B. Pattengale, C. Liu, M. Cendejas, I. Hermans, X. Zhang, J. Zhang and J. Huang, J. Am. Chem. Soc., 2018, 140, 14614–14618 CrossRef CAS.
  174. R. Xu, X. S. Wang, H. Zhao, H. Lin, Y. B. Huang and R. Cao, Catal. Sci. Technol., 2018, 8, 2224–2230 RSC.
  175. S. C. Shit, S. Khilari, I. Mondal, D. Pradhan and J. Mondal, Chem. – Eur. J., 2017, 23, 14827–14838 CrossRef CAS.
  176. C. Ayed, L. C. da-Silva, D. Wanga and K. A. I. Zhang, J. Mater. Chem. A, 2018, 6, 22145–22151 RSC.
  177. Y. Liao, Z. Cheng, W. Zuo, A. Thomas and C. F. J. Faul, ACS Appl. Mater. Interfaces, 2017, 44, 38390–38400 CrossRef.
  178. C. Su, R. Tandiana, B. Tian, A. Sengupta, W. Tang, J. Su and K. P. Loh, ACS Catal., 2016, 6, 3594–3599 CrossRef CAS.
  179. J. M. Tobin, J. Liu, H. Hayes, M. Demleitner, D. Ellis, V. Arrighi, Z. Xu and F. Vilela, Polym. Chem., 2016, 7, 6662–6670 RSC.
  180. R. S. Sprick, B. Bonillo, M. Sachs, R. Clowes, J. R. Durrant, D. J. Adams and A. I. Cooper, Chem. Commun., 2016, 52, 10008–10011 RSC.
  181. Z. Xiao, Y. Zhou, X. Xin, Q. Zhang, L. Zhang, R. Wang and D. Sun, Macromol. Chem. Phys., 2016, 217, 599–604 CrossRef CAS.
  182. C. Yang, W. Huang, L. C. Da-Silva, K. A. I. Zhang and X. Wang, Chem. – Eur. J., 2018, 24, 17454–17458 CrossRef CAS.
  183. H. Zhong, Z. Hong, C. Yang, L. Li, Y. Xu, X. Wang and R. Wang, ChemSusChem, 2019, 12, 4493–4499 CrossRef CAS.
  184. N. Ahmed, M. Morikawa and Y. Izumi, Catal. Today, 2012, 185, 263–269 CrossRef CAS.
  185. L. Mohapatra and K. Parida, J. Mater. Chem. A, 2016, 4, 10744–10766 RSC.
  186. S. Iguchi, Y. Hasegawa, K. Teramura, S. Kidera, S. Kikkawa, S. Hosokawa, H. Asakura and T. Tanaka, Sustainable Energy Fuels, 2017, 1, 1740–1747 RSC.
  187. M. S. Cantú, M. E. H. Torres, A. C. Navarro, E. C. Torres, E. R. Rosas, J. M. G. Jiménez and F. Tzompantzi, Appl. Clay Sci., 2017, 135, 1–8 CrossRef.
  188. C. G. Silva, Y. Bouizi, V. Fornés and H. García, J. Am. Chem. Soc., 2009, 131, 13833–13839 CrossRef.
  189. K. Teramura, S. Iguchi, Y. Mizuno, T. Shishido and T. Tanaka, Angew. Chem., Int. Ed., 2012, 51, 8008–8011 CrossRef CAS.
  190. K. I. Katsumata, K. Sakai, K. Ikeda, G. Carja, N. Matsushita and K. Okada, Mater. Lett., 2013, 107, 138–140 CrossRef CAS.
  191. D. Saliba, A. Ezzeddine, R. Sougrat, N. M. Khashab, M. Hmadeh and M. A. Ghoul, ChemSusChem, 2016, 9, 800–805 CrossRef CAS.
  192. H. Jiang, K. I. Katsumata, J. Hong, A. Yamaguchi, K. Nakata, C. Terashima, N. Matsushita, M. Miyauchi and A. Fujishima, Appl. Catal., B, 2018, 224, 783–790 CrossRef CAS.
  193. N. Ahmed, Y. Shibata, T. Taniguchi and Y. Izumi, J. Catal., 2011, 279, 123–135 CrossRef CAS.
  194. S. Kumar, M. A. Isaacs, R. Trofimovaite, L. Durndell, C. M. A. Parlett, R. E. Douthwaite, B. Coulson, M. C. R. Cockett, K. Wilson and A. F. Lee, Appl. Catal., B, 2017, 209, 394–404 CrossRef CAS.
  195. D. Jiang, J. Li, C. Xing, Z. Zhang, S. Meng and M. Chen, ACS Appl. Mater. Interfaces, 2015, 7, 19234–19242 CrossRef CAS.


Contributed equally as 1st author.

This journal is © The Royal Society of Chemistry 2020