Layered Double Hydroxide Nanostructured ... - Wiley Online Library

www.advenergymat.de www.MaterialsViews.com

Yufei Zhao, Xiaodan Jia, Geoffrey I.N. Waterhouse, Li-Zhu Wu, Chen-Ho Tung, Dermot O’Hare, and Tierui Zhang* by the combustion of fossil fuels for electricity generation and transportation. There is an urgent need to develop new materials and renewable energy technologies to alleviate these concerns. These new technologies should be inexpensive, sustainable and “green”, using only materials, chemical reagents, or energy sources that are abundant, thereby eliminating secondary pollution production. Semiconductor photocatalysis is one of the most promising and practical solutions to address current and future global energy and environmental issues. By harvesting solar energy that is generally abundant everywhere on Earth,[1–4] semiconductor photocatalysts are expected to make an important contribution in areas such as: 1) photocatalytic water splitting to H2 and O2, 2) photoreduction of CO2 into hydrocarbon feedstocks or fuels, and 3) photocatalytic mineralization of organic pollutants.[5,6] Semiconductor photocatalysts with appropriate band gaps (Eg) and band energies (versus normal hydrogen electrode (NHE)) are capable of enhancing the rate of specific chemical reactions under light stimulation with E > Eg. TiO2-based photocatalysts have been the subject of detailed investigation, owing to their excellent activities and stabilities under UV excitation.[7] However, the wide band-gap of TiO2 (Eg = 3.0, 3.2, and 3.3 eV for rutile, anatase, and brookite polymorphs, respectively) limits its ability to efficiently utilize the solar spectrum (only 4% of the solar spectrum at the Earth’s surface has E > 3.0 eV). This represents a practical barrier to the widespread application of TiO2based photocatalysts in the energy sector. Hence, the discovery of novel and efficient visible-light-driven photocatalysts is essential in order to meet the predicted future energy and environmental challenges.[8–10] Amongst the alternative semiconductor photocatalysts that have been explored to date, layered double hydroxide (LDH)-based photocatalysts have emerged as a very promising candidate to replace TiO2, owing to their unique layered structure, tunable band gaps, low cost, ease of scale-up, and good photocatalytic activity for water splitting and other reactions.[11,12] LDHs are a family of 2D layered anionic clays with the gen-

An enormous research effort is currently being directed towards the development of efficient visible-light-driven photocatalysts for renewable energy applications including water splitting, CO2 reduction and alcohol photoreforming. Layered double hydroxide (LDH)-based photocatalysts have emerged as one of the most promising candidates to replace TiO2-based photocatalysts for these reactions, owing to their unique layered structure, compositional flexibility, controllable particle size, low manufacturing cost and ease of synthesis. By introducing defects into LDH materials through the control of their size to the nanoscale, the atomic structure, surface defect concentration, and electronic and optical characteristics of LDH materials can be strategically engineered for particular applications. Furthermore, through the use of advanced characterization techniques such as X-ray absorption fine structure, positron annihilation spectrometry, X-ray photoelectron spectroscopy, electron spin resonance, density-functional theory calculations, and photocatalytic tests, structure-activity relationships can be established and used in the rational design of high-performance LDH-based photocatalysts for efficient solar energy capture. LDHs thus represent a versatile platform for semiconductor photocatalyst development with application potential across the energy sector.

1. Introduction Over the past three decades, there have been growing concerns about fossil fuel depletion and environmental damage caused Dr. Y. Zhao, X. Jia, Prof. L.-Z. Wu, Prof. C.-H. Tung, Prof. T. Zhang Key Laboratory of Photochemical Conversion and Optoelectronic Materials Technical Institute of Physics and Chemistry Chinese Academy of Sciences Beijing 100190, P. R. China E-mail: [email protected] Dr. G. I. N. Waterhouse School of Chemical Sciences The University of Auckland Auckland 1142, New Zealand Prof. D. O’Hare Chemistry Research Laboratory Department of Chemistry University of Oxford Mansfield Road, Oxford OX1 3TA, UK

DOI: 10.1002/aenm.201501974

Adv. Energy Mater. 2015, 1501974

2+ M3x+ ( OH )2 ⎤⎦ ( A n − )q/n ⋅ yH2O , where typically eral formula ⎡⎣M1–x M2+ = Ca2+, Mg2+, Mn2+, Fe2+, Co2+, Ni2+, Cu2+, or Zn2+, M3+ = Al3+, Ga3+, Fe3+, q = x and An− are charge-balancing anions.

© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

q+

wileyonlinelibrary.com

(1 of 20) 1501974

REVIEW

Layered Double Hydroxide Nanostructured Photocatalysts for Renewable Energy Production

www.advenergymat.de

REVIEW

www.MaterialsViews.com

LDH compounds generally fall in the composition range 0.2 ≤ x ≤ 0.33, and are characterized by positively charged q+ 2+ M3x+ ( OH )2 ⎤⎦ layers with the anions and water located in ⎡⎣M1–x the interlayer region. The anions and divalent cations in these materials are often easily exchanged, which allows great flexibility in the composition and electronic structure of LDHs. By incorporating particular photoactive metal cations into LDHs, visible-light responsive photocatalysts with band gaps from 2.0 to 3.4 eV can be synthesized. Furthermore, by controlling the particle size, surface defects can be introduced that alter the LDH electronic structure and greatly enhance the efficiency of photogenerated charge separation and photocatalytic reactions rates. LDHs thus offer a robust structural platform for the development of novel semiconductor photocatalysts with high visible-light activity, motivating detailed experimental and theoretical investigations in this area. This review article focuses on the introduction of surface defects into LDH materials, as a means of systematically tailoring their electronic structures and enhancing their photocatalytic performance. Particular emphasis will be placed on structure-activity relationships and the versatility of LDHs for efficient solar energy capture, including photo/electrochemical water splitting and CO2 photoreduction. Recent progress in LDH characterization by X-ray absorption fine structure spectroscopy (XAFS), electron spin resonance (ESR), X-ray photoelectron spectroscopy (XPS) and density-functional theory (DFT) calculations will also be examined (Figure 1). In the near future, it is envisaged that LDHs will move from being viewed as simple low cost adsorbent materials to industrially-manufactured multifunctional photocatalytic materials for solar energy capture and fuel production.

2. Structures and Properties of LDH Nanostructured Photocatalysts The particle size and thickness of LDH materials strongly influence their properties and performance. When the size or thickness of LDH crystals are reduced to nanometer length scales, the average coordination number of metal cations in the LDH crystal is reduced, leading to different electronic properties and band gap energies.[13] Essentially, when the LDH thickness is on the order of 1 nm, surface phenomena dominate the properties of LDHs.[14,15] Correlations have been established between LDH particle size and properties using advanced characterization techniques such as XAFS, XPS, ESR, UV-Vis and positron annihilation measurements.[16] To further probe the electronic structure of LDH photocatalysts, DFT calculations are routinely performed using atomic structural parameters obtained by from XAFS measurements. Photoluminescence (PL) measurements also provide valuable information about the recombination rates of photogenerated electrons and holes in LDH nanomaterials, which is important for photocatalysis since reaction rates can typically be correlated with charge carrier lifetimes.

2.1. Compositional Structures As stated above, most LDHs have the generalized formula [M1–x2+Mx3+(OH)2]q+Ax/nn−·yH2O, where M2+ and M3+

1501974 (2 of 20)


Yufei Zhao is an assistant professor at the Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, in the lab of Prof. Tierui Zhang. He obtained his dual BS degrees in applied chemistry and computer science and technology at Shanxi University (2007), and his PhD degree in applied chemistry at Beijing University of Chemical Technology, under the supervision of Prof. Xue Duan and Prof. Min Wei in 2013. During 2011–2012, he visited Prof. Dermot O’Hare’s laboratory in the University of Oxford as a joint PhD student. His research focuses on layered double hydroxide nanostuctured photocatalysts for solar fuels.

Dermot O’Hare is currently Professor of Chemistry at Oxford University and Director of the SCG-Oxford Centre of Excellence for Chemistry. He obtained his D. Phil from Oxford University in 1985 under the supervision of Prof M. L. H. Green FRS. His interests are wide ranging, and include exploratory synthetic organometallic chemistry, intercalation chemistry, time-resolved, in situ diffraction studies and the synthesis of meso- and microporous solids.

Tierui Zhang is a full professor at the Technical Institute of Physics and Chemistry, Chinese Academy of Sciences. He obtained his PhD degree in Chemistry in 2003 at Jilin University, China. He then worked as a postdoctoral researcher in the labs of Prof. Markus Antonietti, Prof. Charl F. J. Faul, Prof. Hicham Fenniri, Prof. Z. Ryan Tian, Prof. Yadong Yin and Prof. Yushan Yan. His current scientific interests are focused on the surface and interface of micro/nanostructures, including photocatalysis, electrocatalysis, thermocatalysis, and the controllable synthesis, selfassembly, and surface modification of colloidal inorganic nanostructures.




REVIEW

Figure 1. Design and development of LDH nanostructured photocatalysts for solar energy conversion.

are divalent and trivalent cations, and A is the interlayer anion (Figure 2). However, LDH materials can also accommodate cations in other valence states, including Li+, Sn4+ or Ti4+, and have been widely studied as adsorbents and as catalyst supports due to the tunability of their compositions and morphologies. Bulk LDH materials are very easy to synthesize in the laboratory.[17] However, the synthesis of nanostructured LDH powders and films has recently come under the spotlight since the physicochemical properties of LDHs (such as phase purity, crystallinity, surface area and exposed crystal planes) vary dramatically with synthesis method.[18–21] Recently, O’Hare’s group surveyed methods for the synthesis of LDH ultrathin nanosheets, and

discussed the advantages and disadvantages of currently used delamination technologies for LDH nanosheet fabrication. LDH nanosheets were also classified by application into the following areas: polymer/LDH nanocomposites, thin films, LDH-based catalysts, LDH hybrid magnets and bioinorganic materials.[19,22,23] Li et al. reviewed the synthesis of LDHs, their transformation to mixed metal oxides by thermal treatment and/or reduction, and also their applications to heterogeneous catalysts.[24–26] The fabrication and applications of LDH-based nanocomposites represent an important new direction in the development of multifunctional materials,[20,27–29] with the obtained nanocomposites exhibiting enormous promise in

Figure 2. A) Polyhedral representation of the LDH structure showing MO6 octahedra along the c axis. Anions and water are present in the interlayer region. B) Side view and C) top view of an LDH layer.




(3 of 20) 1501974

www.advenergymat.de

REVIEW


applications such as catalysis, luminescence devices, sensors, and targeted drug delivery systems, amongst others.[30–35] 2.2. Atomic Structures X-ray absorption near-edge spectroscopy (XANES) is a particularly useful tool for studying the valence, coordination number, and degree of disorder of the metal cations in nanostructured LDH materials. Using ultrathin NiTi-LDH nanosheets as an example,[36] EXAFS spectra at the Ti K-edge (Figure 3A) show that the peak maximum for NiTi-LDH nanosheets is at a lower excitation energy compared with the bulk counterpart. Bulk NiTi-LDH contains almost exclusively Ti4+, whereas the NiTiLDH nanosheets contain a significant fraction of titanium cations in a lower oxidation state, which are responsible for the shift in the peak maximum to lower excitation energies. Fourier transforms of the Ti K-edge EXAFS spectra (Figure 3B) reveal severe structure distortions and lower coordination numbers about the Ti cations in the NiTi-LDH nanosheets compared with the bulk counterpart. Figure 3C provides evidence that the distance of the first Ti-O shell coordination in the NiTiLDH nanosheets was slightly decreased, and the degree of octahedral distortion about the Ti centers increased as revealed

by the Debye-Waller factor. This distorted structure endows ultrathin LDH nanosheets with an unusual electronic structure and enhanced electron transfer efficiency, which is expected to be beneficial for water splitting. In ultrathin ZnTi-LDH nanosheets systems, EXAFS data suggested a lower coordination number of the Ti, associated with a more serious structural distortion compared with that of the bulk counterpart.[37] In ZnCr-LDH/TiO2 nanosheet systems,[30] Zn, Cr and Ti K-edge XANES analyzes demonstrate that there are negligible changes in the crystal structure and electronic structure of ZnCr-LDH after the hybridization with TiO2 nanosheets (Figure 4). The ability of XANES and EXAFS to precisely probe atomic structure variations in ultrathin LDH nanosheets provides a firm foundation for establishing structure-property relationships. 2.3. Defect Structures During the formation of ultrathin LDH nanosheets, surface atoms can easily be removed or lost, leading to local structural distortions and the existence of vacancy defects. These defects can strongly impact the electronic structure and photocatalytic performance of LDH nanomaterials. To gain deeper insight into the type of defects formed, positron annihilation measurements,

Figure 3. A) Ti K-edge, XANES spectra and B) Fourier transforms of Ti K-edge EXAFS spectra for: a) LDH-nanosheet, b) LDH-bulk and c) Rutile TiO2. C) Local structure from EXAFS data. The supercell model of D) NiTi-LDH layer, and E) NiTi-LDH layer doped with Ti3+ defects (Green: Ti; Gray: Ni; Red: O; White: H). Reproduced with permission.[36] Copyright 2014, Royal Society of Chemistry.

1501974 (4 of 20)





REVIEW Figure 4. (Upper panels) Ti, Cr and Zn K-edge, XANES spectra of (a–c) ZnCr-LDH/layered titanate nanosheets with different cation ratios, d) protonated layered titanate/ZnCr-LDH/ZnCr-LDH, e) TiO2/Cr2O3/ZnO, and f) TiO2/CrO3. (Lower panels) HRTEM image of ZnCr-LDH/layered titanate nanosheets and structural model. Reproduced with permission.[30] Copyright 2011, American Chemical Society.

ESR and XPS measurements are routinely applied. In Ti-containing LDH systems,[38] defects such as monovacancies or oxygen vacancies, oxygen-vacancy clusters and micropores can be probed by positron annihilation experiments through analysis of positron lifetimes (τ) and relative intensities (I). Generally, monovacancies or oxygen vacancies (τ1) exist in the bulk of materials, whereas oxygen-vacancy clusters (τ2) have a higher probability of being located on the surface of the samples. As shown in Table 1, the ratio of I2/I1 for the LDH nanosheets is 1.63, much larger than the same values determined for layered oxides and hydroxides. The data clearly indicate the predomination of surface defects (I2) for the LDH materials. Positron annihilation experiments thus give important information about how defects are distributed in LDH materials. These defects, especially the surface defects, could serve as trapping sites for

electrons and thus improve e-h separation efficiency for photocatalytic reactions. However, positron annihilation only provides a ratio of the defects that exist in a photocatalyst. To identify the true state of the defect, including metal ions and their oxidation state, additional techniques are required.[13] Detailed information about the defects in ultrathin LDHs can be obtained using ESR and XPS. For ZnTi-LDH nanosheets, ESR spectroscopy reveals two sharp signals with g values of 1.996 and 2.030 (Figure 5A), which can be assigned to Ti3+ and O− species, respectively.[37] The presence of Ti3+ is due to the formation of oxygen defects. The oxidation state of Ti in LDH nanosheets is readily determined by XPS. The Ti 2p XPS spectrum for ZnTi-LDH nanosheets contains contributions from both Ti3+ and Ti4+ (Figure 5B). With a decrease of ZnTi-LDH particle size from 2 µm to 30 nm, the ratio of Ti3+/Ti increases

Table 1. Positron lifetimes and relative intensities of NiTi-LDH nanosheets and counterpart layered oxide (K2Ti4O9) and Ni(OH)2. Reproduced with permission.[38] Sample

τ1 [ps]

τ2 [ps]

τ3 [ns]

I1 [%]

I2 [%]

I3 [%]

I2/I1

LDH

183.6

367.1

1.888

36.97

60.36

2.67

1.63

K2Ti4O9

206.9

338.7

2.011

45.80

52.80

1.40

1.15

Ni(OH)2

231.6

365.5

2.206

53.20

45.30

1.50

0.85




(5 of 20) 1501974

www.advenergymat.de

REVIEW


Figure 5. A) ESR spectra for freshly prepared ZnTi-LDH nanosheets, a) before, and b) after 20 minutes of visible light exposure at 110 K under a N2 atmosphere. B) Ti 2p XPS spectra for: a) TiO2, b) K2Ti4O9, c) ZnTi-LDH-Bulk, and d) ZnTi-LDH nanosheets. ESR spectra of: C) ZnAl-LDH nanosheets, and D) bulk ZnAl-LDH. Each plot C, D contains spectra for: a) fresh catalyst, b) catalyst after 20 min of visible light irradiation (λ > 400 nm), and c) catalyst after 20 min of UV-Vis light irradiation. a,b) Reproduced with permission.[37] Copyright 2013, Royal Society of Chemistry. c,d) Reproduced with permission.[40]

from 14.7% to 27.0%, accompanied with a concomitant decrease in the Ti4+ signal. The greatly increased concentration of surface Ti3+ sites in ultrathin LDH nanosheets can be attributed to the high concentration of oxygen defects. The large proportion of paramagnetic Ti3+ centers modifies the electronic and optical properties of ZnTi-LDH considerably. Ti3+ species (g = 1.996) were also detected for NiTi-LDH nanosheets[36] and also a nanohybrid of LDH and reduced graphene oxide (rGO), and further corroborated by XPS.[39] In the nanohybrid case, the presence of Ti3+ was not only due to particle size effects, but also electron transfer from LDH to rGO. Oxygen defects have also been detected in other LDH systems, such as ultrathin ZnAl-LDH nanosheets, which can be prepared with a platelet size of 40 nm and a thickness of only 2.7 nm.[40] Figure 5Ca shows an ESR signal at g = 1.998 for ZnAl-LDH nanosheets. This feature is assigned to an electron trapped around a Zn-O site. However, no signals were detected for defect-free bulk ZnAl-LDH (Figure 5Da). This g-value of observed for ZnAl-LDH nanosheets is in good agreement with ESR data previously reported by Chen et al. that the Zn+ in Zn-ZSM-5 exhibited an ESR signal around g = 1.998.[41] In the ZnAl-LDH system, the Vo defect is likely located next to Zn cations, leading to the existence of coordinatively unsaturated Zn ions in the form of Zn+-Vo complexes. Ti-containing LDH compounds contain a M2+-O-Ti4+ network, in which MO6 and TiO6 octahedra are bonded to each other through metal oxo bridging. By controlling the particle

1501974 (6 of 20)


size of LDH on the nanoscale, oxygen defects are introduced leading to the formation of Ti3+ states. Occasionally these defects can also affect the charge on the M cations. In the monolayered NiTi-LDH nanosheets,[42] ESR and XPS provided evidence that Ni3+ states are formed due to high concentrations of oxygen vacancy defects, with Ni3+ existing in the form of NiOOH. Monolayered NiTi-LDH nanosheets exhibit increased electrical conductivity and more efficient charge transfer compared with bulk NiTi-LDH. Hence, by adjusting the metal ions in LDH nanosheets and using particle size to control defect levels, it is possible to introduce intragap levels in the band gap of LDH, thereby allowing a better visible-light response and improved photocatalytic performance under direct solar irradiation. To summarize, recent advances in the synthesis of ultrathin LDH nanosheets allow the fabrication of novel semiconductor photocatalysts with a high concentration of surface active sites, specifically defects containing low coordination number metal centers. Such ultrathin LDH nanosheets typically possess different electronic structures and reactivities compared to their bulk counterparts, which can be highly beneficial for photocatalysis, as is examined in detail below. 2.4. Electronic Properties It is widely accepted that photocatalytic performance is intimately connected to the electronic properties of a semiconductor




Moreover, by incorporating defects in LDH systems, dramatic electronic structure changes can be realized in comparison with defect-free bulk LDH. As mentioned above, by controlling the particle size of Ti-containing LDH to around 40 nm, Ti3+ sites were created due to the formation of oxygen defects.[36,37] Ti3+ gives rise to interband states in the band gap of LDH, which narrows the bandgap to ≈2.3 eV compared with 3.0 eV for the defective-free Ti-containing LDH (Figure 7). The narrower band gap allows a wider visible light absorption range and improved water splitting performance under direct sunlight. DFT calculations have also been used to analyze the band structure of hybrid systems such as TiO2@CoAl-LDH.[46] DOS results reveal a band gap of 2.2 eV for CoAl-LDH and 3.2 eV for TiO2. However, for the TiO2@CoAl-LDH interface, the electronic structure shows a significant change due to strong donor-acceptor coupling. Band structure calculations confirmed that electrons photoexcited in CoAl-LDH can inject into the CB of TiO2, resulting in a longer hole lifetimes in LDH. Hence the hybrid system is effective in facilitating the separation of photoinduced charge carriers (Figure 8). Recently, Wei and co-workers used DFT calculations to probe the electronic properties of Mg/Ni/Zn/Co-Al/Ga LDHs, a promising family of photocatalysts for water splitting.[47] The valence band maximum (VBM) and conduction band minimum (CBM) of the LDHs are shown in Figure 9. Co/Ni/Zn-LDH and NiGa-LDH were identified as particularly promising candidates for water splitting due to the energetically favorable positions of their VBM and CBM, and their modest band gaps allow visible light excitation. In general, DFT calculations provide a robust theoretical framework for correlating the atomic structure, defect type, and the physical properties of LDH photocatalysts. It is expected that DFT calculations will be used increasingly in the future for both photocatalyst design and performance optimization.

Figure 6. A) Band structure, B) TDOS and PDOS for CuCr-NO3-LDH. Reproduced with permission.[38] Copyright 2012, Wiley.




(7 of 20) 1501974

REVIEW

photocatalyst, such as the electronic band gap, charge separation, and charge mobility, which all impact photocatalytic performance. Understanding electronic structure-performance relationships allows the rational design of photocatalysts with improved performance. In this regard, ultrathin LDH nanosheets represent near ideal model compounds for exploring the electronic structure-performance relationships, since both the composition and defect types within the layered LDH structure can be easily controlled and manipulated. Furthermore, the performance of ultrathin LDH nanosheets and their bulk counterparts can easily be compared. The electronic structure of bulk and ultrathin LDH nanosheets, including the band structure, total densities of states (TDOS) and partial densities of states (PDOS), can be modeled by DFT calculations. Garcia et al. reported that the ZnCr-LDH exhibits excellent water splitting performance compared with ZnTi LDH and ZnCe LDH.[43] It is unclear from that study how Cr ions enhance the water splitting process, as the electronic structure of ZnCr-LDH was unknown. Zhao and co-workers used DFT calculations to probe the electronic structure of CuCr-LDH photocatalysts.[44,45] Figure 6 shows the band structure calculated for CuCr-LDH, which reveals a band gap of about 1.6 eV, with the compound behaving as a direct band gap semiconductor that allows easy electron-hole separation under visible light. In addition, the DOS results show that the valence band (VB) maximum is dominated by occupied Cr 3d orbitals (A2g), and the conduction band (CB) minimum consists of unoccupied Cr 3d orbitals (T2g + T1g). The electronic structure calculations also indicated two different band energies, consistent with the optical absorption bands at around 420 and 570 nm observed experimentally. From the above DFT results, the CrO6 octahedra in the LDH layer were identified as the key photoactive site for visible-light-driven photocatalysis. Ti-containing LDH compounds have also been investigated by DFT.[38] The band gap of the NiTi-LDH was determined to arise through covalent interactions between TiO6 and NiO6 octahedra.

www.advenergymat.de

REVIEW


Figure 7. TDOS and PDOS for A, B) an ideal ZnTi-LDH system, and C, D) Ti3+-doped ZnTi-LDH. Reproduced with permission.[37] Copyright 2013, Royal Society of Chemistry.

2.5. Optical Properties A bottleneck for the practical application of many semiconductor photocatalysts is their narrow excitation range and low charge separation efficiency. Bulk doping oxide semiconductors with metal or non-metal ions is an effective approach for improving the visible-light response of wide band gap semiconductors such as TiO2, though often the introduction of dopant ions creates electron-hole pair recombination sites that are highly detrimental to photocatalyst activity. Recently, metal-tometal charge transfer (MMCT) in heterobimetallic systems has emerged as a new class of high-performance photocatalysts, circumventing many of the historical limitations of doped semiconductor systems.[48] LDHs, by virtue of their layered structure and versatility in chemical composition, are now being widely used as photocatalysts. By adjusting the metal ions in the LDH

1501974 (8 of 20)


layer, the absorption response can be tuned to almost any desired visible wavelength. As examples, ZnTi-LDH shows an intense absorption band around 310 nm, and ZnCe-LDH exhibits an absorption band at 280 nm, whilst ZnCr-LDH gives the two absorption maxima at 410 and 570 nm (Figure 10A). This strong visible light adsorption makes Cr-LDH a promising material for visible-lightdriven photocatalysis. Various ions such as Cu2+, Ni2+, Mg2+, Zn2+, In3+, and Y3+[49] have been intercalated into Cr/Ti-containing LDH as a means of further modifying the visible-light response and photocatalytic properties (Figure 10B).[43,50,51] Recently, 2D nanosheets, including graphene and TiO2 nanosheets, have attracted significant research interest owning to their excellent electronic properties and charge transfer efficiencies. Accordingly, these nanosheets are finding increasing use in electronics and catalysis.[12,52] Hwang and co-workers




REVIEW Figure 8. A) Optimized structural geometry of TiO2@CoAl-LDH, and B–D) the corresponding TDOS. E) Schematic illustration of the TiO2@CoAl-LDH for water splitting. Reproduced with permission.[46] Copyright 2015, Wiley.

reported that mesoporous nanohybrids consisting of positively charged ZnCr-LDH nanosheets and negatively charged TiO2 nanosheets or graphene showed two strong absorption peaks in the visible light range, which can be attributed to the d-d transitions of the Cr ions in ZnCr-LDH nanosheets.

Hence, efficient electronic coupling between the component nanosheets was achieved.[30] Other LDH-based composite photocatalysts developed to date include LDH/POM,[53] LDH/ CdS,[54–57] TiO2@LDH,[46] Fe2O3@LDH[58] and LDH/Ag,[59] all of which feature an improved visible light absorption and

Figure 9. Band edge positions for Mg/Ni/Zn/Co-Al/Ga-LDHs. The two dashed red lines represent the reduction potential of H2 and the oxidation potential of O2 at pH 7. Reproduced with permission.[47] Copyright 2015, American Chemical Society.




(9 of 20) 1501974

www.advenergymat.de

REVIEW


Figure 10. Diffuse reflectance UV-Vis spectra of Zn/Cr-containing LDH. Left panel reproduced with permission.[43] Copyright 2009, American Chemical Society. Right panel reproduced with permission.[44]

increased photocatalytic performance compared to the nonhybridised LDH material. PL spectroscopy is commonly used to explore the separation and recombination of the photoinduced charge carriers in semiconductor photocatalysts, including LDH-based photocatalysts. The PL signal of ZnCr-LDH/GO nanohybrids is strongly attenuated compared to that of ZnCr-LDH (Figure 11B), indicating that GO can accept electrons photoexcited in ZnCr-LDH. This suppression of electron-hole pair recombination in ZnCr-LDH increases the number of charge carriers available for photoreactions and enhances reaction rates.[60] The layered structures of both ZnCr-LDH and GO are important for achieving efficient contact and electron transfer between the two photocatalyst components. Following photoexcitation, NiTi-LDH nanosheet/ graphene[39] also displays a weaker PL signal compared to bare NiTi-LDH nanosheets,[36] providing further evidence of the

advantages of hybridizing semiconductors with conductive 2D materials as a means of facilitating charge separation. Defects alter the electronic structure and photocatalytic performance of LDHs. PL spectra reveal that the emission from the LDH nanosheets is significantly weaker than that of bulk LDH, presumably reflecting a lower efficiency of electronhole recombination (Figure 12). Moreover, the defect (Ti3+) doped LDH nanosheets possess a narrower band gap and a red shifted UV-Vis absorption spectrum compared with the bulk counterpart, both of which can be attributed to Ti3+ defects that create an impurity level below the CBM. Thus, the narrower band structure for the Ti3+ doped LDH not only allows ultrathin defective LDH to capture more solar energy but also facilitates photoinduced charge transport, thus enhancing the water splitting activity of Ti3+ doped LDH compared to its bulk counterpart.[36]

3. Photocatalytic Water Splitting The search for efficient and stable semiconductor photocatalysts for water splitting into H2 and O2 using solar energy is one of the most important areas of modern materials science research. The activities of various LDH photocatalysts for water splitting are summarized and discussed in this section.[16,30,31,36,43,61,62] Data for other important water splitting photocatalysts are also provided for comparison.[63] 3.1. LDHs for Photocatalytic Oxygen Generation

Figure 11. Diffuse reflectance UV-Vis spectra (A), and PL spectra (B) of self-assembled ZnCrLDH/GO and ZnCr-LDH. Reproduced with permission.[60] Copyright 2013, Royal Society of Chemistry.

1501974 (10 of 20)


The formation of oxygen via water splitting is challenging since the process requires four positive holes and several mechanistic steps to form O–O bonds. Although TiO2 has been widely study for water splitting, the




REVIEW Figure 12. A) Fluorescence spectra of: a–c) NiTi-LDH-nanosheets with different particle size, and d) NiTi-LDH bulk. B) UV-Vis diffuse reflectance spectra of a) NiTi-LDH nanosheets, b) NiTi-LDH bulk, and c) TiO2. The inset in B shows the direct band gap. C) The energy states for the Ti3+ self-doped NiTi-LDH. D) Schematic illustration of the NiTi-LDH nanosheet under visible-light irradiation for water oxidation. Reproduced with permission.[36] Copyright 2014, Royal Society of Chemistry.

UV response of the semiconductor severely restricts its further application. There are many ways to extend the activity of TiO2 into the visible region, though generally methods such as bulk doping with metal cations or electron negative anions (e.g., F−) yield photocatalysts with poor photocatalytic activities. A more effective approach is to incorporate Ti cations into compounds such as LDHs, which allows controlled doping of Ti-based semiconductors with different metals without the detrimental charged imbalances which plague doped TiO2 photocatalysts such as TiOxNy. The atomic ratio of divalent and tri/tetravalent metal ions in LDH materials can be varied over a wide range from 1:1 to 5:1 without changing the layer structure, and the “dopant” metal would be located in a well-defined position in the LDH structure. Garcia et al. first reported that the LDH could serve as band gap controlled semiconductor for the water splitting into oxygen.[43] However, the photocatalytic properties of LDH materials in water splitting were largely ignored for a long period of time, and only recently have LDH photocatalysts again attracted attention for this application. Inspired by the good performance of ZnO-based photocatalysts, the initial LDH photocatalysts were designed by fixing Zn as the divalent ions in the layer, and selecting Cr, Ti, and Ce as tri/tetravalent metal ions. ZnCr-LDHs are especially active for visible light-driven


photocatalytic O2 generation from water with an apparent quantum yield of 60.9% at 410 nm. This result demonstrates the great potential of LDHs as visible-light-driven photocatalysts for water splitting. However, Zn-containing LDH photocatalysts, like ZnO, are susceptible to photocorrosion which reduces their effectiveness. Hence, research emphasis is now being directed towards the development of alternative, non-Zn containing LDH photocatalysts. Ti is the logical candidate to incorporate into LDH photocatalyst systems, since modified TiO2 photocatalysts are active for water splitting. Ti-based LDH was successfully synthesized by adding Ni and Cu as divalent ions to form the LDH structure.[64] By this approach, good visible light activity is achievable through metal-to-metal charge transfer (Ni/Cu-O-Ti). This covalent anchoring of Ni or Cu centers, which introduces new energy levels inside the TiO2 band gap, extends the photoactive region well into the visible range. NiTi-LDH shows an oxygen evolution activity of 40 µmol g−1 h−1 when AgNO3 is used as the sacrificial agent under visible light irradiation. CoFe-LDH is an even more efficient photocatalyst for water oxidation under visible light, affording a rate of 300 µmol g−1 h−1.[65] However, the activities of these LDH photocatalysts are still quite low due to the larger particle size (several micron) and long charge migration paths from the bulk to the surface.



(11 of 20) 1501974

www.advenergymat.de

REVIEW


Accordingly, the photocatalytic properties of nanosized LDH materials have attracted significant attention as they possess higher specific surface areas, a higher proportion of surface active sites and shorter charge diffusion paths. Zhao et al. reported the successful synthesis of ultrathin Ti-containing LDH nanosheets with thickness ≈2.0 nm and a particle of 30 nm by incorporating M2+ cations (such as Ni2+ or Zn2+) into a LDH matrix.[36] Due to the high surface areas and exposed Ti3+ active sites, the ultrathin LDH nanosheets showed a high O2 evolution rate (2148 µmol g−1 h−1) compared with the bulk counterpart (267 µmol g−1 h−1) under visible-light irradiation. An apparent quantum yield of 65% was obtained for the LDH nanosheets at 400 nm. Furthermore, the ultrathin LDH nanosheets showed good recyclability and stability during testing, further emphasizing the promise of LDH nanosheet photocatalysts for water splitting. Another effective strategy for improving the photocatalytic efficiency of LDH nanosheets is to couple them with other functional semiconductor materials. The resulting nanohybrid materials often possess a wider light absorption range and more efficient charge transfer separation. Through electrostatic self-assembly, various nanohybrids of LDH with negatively charged nanosheets have been synthesized (e.g., ZnCr-LDH and TiO2 nanosheets, ZnCr-LDH and rGO, NiTi-LDH and rGO), and their performance was evaluated for visible-lightdriven water splitting. Inspired by the above, negatively charged polyoxometalate (POM) nanoclusters have also been successfully self-assembled with LDH to generate LDH/POM hybrids for water splitting into oxygen (Figure 13). Table 2 summarizes the properties and performance of such LDH-based nanohybrid photocatalysts for water splitting. Data for other widely used semiconductor photocatalysts are also provided for comparison. Although the hybrids show excellent photocatalytic activity, they generally suffer from poor control over particle size and crystalline orientation. Furthermore, severe aggregation can occur during photoreactions, which limits the extent of any activity improvements expected by coupling LDH materials with other 2D semiconductors. In order to overcome the aforementioned issues, hierarchical nanostructured LDHs with well-defined morphologies and high

dispersion have been developed for water splitting and other applications.[29,66] TiO2@CoAl-LDH core-shell nanospheres are a good example of such hierarchical systems,[46] where the CoAl-LDH nanocrystals act as the oxygen evolution site whilst the TiO2 core accepts photoexcited electrons from CoAl-LD, thereby facilitating charge carrier separation (Figure 14). The structure exhibits high O2 generation rates of 2.24 mmol g−1 h−1 under visible light. UV-Vis absorption spectra, PL spectra and DFT calculations confirm the suitable band gap matching of the TiO2 and LDH, which together with the intimate contact of the two components facilitate effective separation of the charges following photoexcitation, resulting in enhanced activity. Potentially, a wide range of similar smart hierarchical structures could be developed, suggesting a great future potential of LDH-based photocatalysts for water splitting to O2.

3.2. LDHs for Photocatalytic Hydrogen Generation LDH photocatalysts can also be engineered for H2 evolution. As with oxygen evolution by water splitting, this requires careful selection of the metal ions in LDH layer. Fe-containing LDHs were one of the first such systems studied for water splitting into hydrogen. By intercalation of Fe into MgAl-LDHs, photocatalysts with H2 production activities as high as 493 µmol g−1 h−1 were obtained under visible light excitation.[67] Incorporation of FeO6 octahedral units in the LDH structure enhances the photocatalytic activity for H2 evolution, since the CB energy becomes more negative (versus NHE) than that of the H2O/H2 redox couple. However, excess Fe ions led to the formation of amorphous Fe2O3, which was highly detrimental to the charge transfer processes responsible for H2 production. Many Ticontaining LDHs also show good activity for H2 generation.[38] Using lactic acid as a sacrificial agent, Ni/Zn/MgAlTi-LDHs with different composition ratios yielded H2 production rates as high as 314 µmol g−1 h−1, approximately 18 times higher than that of layered Ti oxides (e.g., K2Ti4O9). This excellent performance is ascribed to the very high dispersion of the photoactive component (TiO6 octahedra) in the LDH matrix, compared with the serious aggregation of the active sites encountered for

Figure 13. Left) HRTEM images of ZnCr-LDH and POM nanohybrids with schematic representations of their layered structures. Right) Time-dependent photoproduction of O2 under visible light. Reproduced with permission.[53] Copyright 2013, Nature Publishing Group.

1501974 (12 of 20)





Photocatalyst

Synthesized method

Size [nm]

Mass [mg]

Light source

Incident light [nm]

Sacrificial agent/ Cocatalyst

Amounts [µmol g−1 h−1]

Q.Y. [%]

Ref

ZnCr-LDH

Co-precipitation

>1000

45

200 W Xe

>400

AgNO3

O2 1073

60.9

[43]

ZnTi-LDH

Co-precipitation

>1000

45

200 W Xe

>400

AgNO3

O2 268.3

–

[43]

ZnCe-LDH

Co-precipitation

>1000

45

200 W Xe

>400

AgNO3

O2 626.1

–

[43]

NiTi-LDH

Co-precipitation

≈600

200

300W Xe

700–400

AgNO3

O2 50

–

[64]

CuTi-LDH

Co-precipitation

–

200

300W Xe

700–400

AgNO3

O2 30

–

[64]

ZnCr/layered TiO2

Layer by layer

500–800

10

450W Xe

>420

AgNO3

O2 1180

–

[30]

ZnCr-LDH/RGO

Self-assembly

500

10

–

>420

AgNO3

O2 1200

61.0

[60]

NiTi-LDH/RGO

In situ growth

30–100

50

300 W Xe

>400

AgNO3

O2 1968

61.5

[39]

ZnCr-LDH/POM

Self-assembly

30–100

10

450 W Xe

>420

AgNO3

O2 2400

75.2

[53]

CoAl-LDH

Hydrothermal

–

20

300 W Xe

>420

AgNO3

O2 950

–

[46]

CoFe-LDH

Hydrothermal

50

50

300 W Xe

700–400

AgNO3

O2 300

TiO2@CoAl-LDH

In situ growth

200

20

300 W Xe

>420

AgNO3

O2 2240

–

[46]

NiTi-LDH nanosheet

Microemulsion

30

50

300 W Xe

>400

AgNO3

O2 2148

65.0

[36]

[16]

FeMgAl-LDH

Co-precipitation

≈100

20

125 W Hg

>420

CH3OH/None

H2 493

–

[67]

Zn/Ni/MgAl-Ti-LDH

Co-precipitation

100–500

100

300 W Xe

whole range

Lactic acid/Pt

H2 314/153/49

–

[38]

Ni-Zn/Cr-LDH

Co-precipitation

≈30

20

125 W Hg

>400

CH3OH/None

H2 1915

Memory effect

≈110

100

1300 W cm−2

whole range

CH3OH

H2 181

–

[69]

cm−2

whole range

CH3OH

H2 132

–

[69]

>420

Na2SO4+Na2S/Pt

H2 1560

–

[54]

Au/ZnCeAl-LDH

[68]

Au/ZnAl-LDH

Memory effect

≈110

100

CdSe/LDH

Self-assembly

≈100

20

Exfoliation-restacking

–

100

300 W Xe

>420

Na2SO3+Na2S/None

H2 374

42.6

[70]

Impregnation

>200

30

125 W Hg

>420 >400

AgNO3, CH3OH/None

O2 14767 H2 24800

–

[71]

Solid-state reaction

100–700

1000

400 W Hg

whole range

AgNO3, CH3OH/Pt

O2 9700

56

[72]

Thermal polycondensation

–

100

300 W Xe

0.1

[73]

Photoreduced graphene oxide

–

30

300 W Xe

(Ga1–xZnx)(N1–xOx)/ RhCrOy

NH3 reduction

–

100

300 W Xe

Co-Silicon| NiMoZnSilicon cell

Self-assembly -Deposition

≈1000

CdS/ZnCr-LDH g-C3N4/NiFe-LDH NiO/NaTaO3:La

1300 W

450 W Xe

H2 19800

whole range g-C3N4

BiVO4/GRO/ Ru/SrTiO3:Rh

85–200 1 sun nm thick (100 mW cm−2)

>420

AgNO3/RuO2

O2 ≈9

>420

Triethano- lamine/Pt

H2 ≈104

>420 >420

Overall water splitting in H2SO4

>400

Overall water splitting

O2 ≈1.7 and H2 ≈3.4

2.5

[75]

whole range

Photoelectro chemical

5% solar to fuels efficiency

–

[76]

O2 ≈167 and 0.97/1.03 H2 ≈10

[74]

Figure 14. Schematic illustration of a TiO2@CoAl-LDH hierarchical nanostructure for water splitting. Reproduced with permission.[46] Copyright 2015, Wiley.




(13 of 20) 1501974

REVIEW

Table 2. Summary of LDH photocatalysts and other widely used visible-light-driven photocatalysts for water splitting to oxygen or hydrogen.

www.advenergymat.de

REVIEW


layered Ti oxides. Transient-absorption and PL spectra measurements suggested that electron-hole recombination is seriously suppressed in Ni/Zn/MgAlTi-LDHs due to the presence of the surface defects. This strategy of incorporating atomic dispersed photoactive sites (FeO6, TiO6, CrO6, etc.) in a LDH matrix is therefore very effective in increasing charge separation, and is expected to be exploited in the future in photocatalysis, electrochemistry, and energy conversion. In order to further enhance charge transfer efficiencies, LDH nanosheets can be functionalized with other semiconductor cocatalysts (e.g., CdSe quantum dots (QDs)).[54] The strong electronic coupling that exists in such nanohybrid systems leads to dramatically improved water splitting performance. As an example, a LDH-CdSe nanohybrid realized a H2 evolution rate of 1.56 mmol g−1 h−1 under visible light, approximately 5 times higher than pristine CdS QD (0.3 mmol g−1 h−1) under the same test conditions. Furthermore, the localized surface plasmon resonance (LSPR) of supported Au and Ag nanoparticles (NPs) can be exploited to enhance the visible light activity of LDH-based photocatalysts. Au/LDH (LDH = ZnAl, ZnAlCe) nanocomposites were reported to be effective for H2 generation in water-methanol solvent mixtures. The existence of Ce in the LDH layer may cause oxidation of supported Au0 to Au3+, resulting in electron injection into the conduction band of LDH, thereby increasing the H2 evolution during light irradiation (Figure 15).[62] Alternatively, stimulation of the Au LSPR by visible light could induce the injection of electrons from Au into the LDH conduction band, with methanol then acting as a sacrificial electron donor to keep the supported Au NPs in the metallic state. 3.3. LDHs for Photoelectrochemical Water Splitting Photoelectrochemical (PEC) water splitting to hydrogen and oxygen is an exciting strategy for capturing and storing solar energy. Developing efficient photoelectrodes for PEC water splitting is challenging. To date, Zn-containing LDH-based photoelectrodes (ZnCr-mixed metal oxide (MMO)[77] and N-doped ZnAl-MMO[78] have been used in PEC water oxidation, with Ni-containing LDHs (NiFe[79] and NiCo-LDH[63] proving effective as cocatalysts to suppress electron-hole pair recombination and accelerate reaction kinetics. Lee et al. reported the in situ

growth of ZnCr-LDH nanosheets from Zn metal-coated FTO glass substrates. Following calcination to form Cr-doped ZnO, the resulting photoelectrodes exhibited a photocurrent density of 73 µA cm−2 under visible light, which is an order of magnitude higher that of the ZnCr-MMO powders (2 µA cm−2).[77] Furthermore, the photoelectrodes gave a higher incident-photon-tocurrent conversion efficiency (IPCE) in the visible light range because of Cr doping into ZnO, whereas conventional ZnO electrodes exhibited no response in the visible region (Figure 16). The photoanodes also demonstrated good stability after 2 hours of photoelectrochemical testing. In addition to transition metal doping in UV responsive semiconductors like ZnO, non-metal doping with C or N has also been reported in LDH-based photoelectrodes. Lee et al. reported that intercalation of terephthalate into ZnAl-LDH, followed by thermal ammonoloysis (i.e., heat treatment in NH3 vapor), resulted in C,N-doped MMO with strong visible-light absorption characteristics and a higher photocurrent density than pure MMO, with an IPCE of 3% achieved under 400 nm excitation.[78] Recently, Ta-based oxide, oxynitride and nitride photoelectrodes have attracted great attention because of their excellent performance in PEC water oxidation. However, a key challenge to overcome is their severe photocorrosion during photochemical reactions. Although modification of Ta3N5 with IrO2 or Co(OH)x can reduce photocorrosion, the stability of the obtained Ta3N5-based photoelectrodes is generally too low for practical applications. Wang et al. reported that the deposition of NiFe-LDH on the surface of Ta3N5 leads to greatly enhanced stability for PEC water splitting, maintaining a very high photocurrent (≈90% of the initial value) after 2 h irradiation (Figure 17).[79] Electrochemical characterization studies established that the decoration of Ta3N5 with LDH-cocatalyst decreased the photoinduced transfer resistance, allowing photoexcited holes generated in Ta3N5 to migrate onto the LDHcocatalyst (thereby preventing photocorrosion caused by the oxidation of N3− to N2), whilst also enhancing the photocurrent. Furthermore, the core-shell structure increased charge transfer and the kinetics of PEC water splitting. Shao and co-workers reported an excellent example of the fabrication of nanowire arrays based on a ZnO core and NiCo-LDH shell for PEC water splitting.[63] The intimate chemical bonding interaction at the ZnO/NiCo-LDH interface allowed fast transfer of charge (mediated by the Co(II)/Co(III) redox couple), and increased the PEC water splitting efficiency.

4. Photovoltaiacs

Figure 15. Schematic for water splitting into H2 using Au/ZnAlCe-LDH under visible light. After absorption of light, Au NPs inject electrons into the LDH, promoting H2 evolution. Reproduced with permission.[62] Copyright 2013, Royal Society of Chemistry.

1501974 (14 of 20)


Dye-sensitized solar cells (DSSCs) transform solar energy directly into electricity and are expected to be an integral part of future renewable energy infrastructures. However, improvements in solar-to-electrical conversion efficiencies are needed before widespread commercial deployment. By virtue of the wide composition ranges and electronic properties achievable in LDH materials, it is anticipated that LDH-based semiconductors could eventually replace TiO2 and other oxide systems as the working electrode semiconductor element in DSSCs. As an example, calcination of ZnTi-LDH[80] yields highly dispersed ZnO and TiO2 nanocrystals (i.e., a MMO). In order




REVIEW Figure 16. A) I-V curves for a ZnCr:MMO photoelectrode under visible-light irradiation. B) I-V curves for ZnCr:MMO with and without CoOx-deposited photoelectrode under visible light irradiation and no-bias. C) IPCE measurements for ZnO and ZnCr:MMO photoelectrodes carried out at 1.23 V vs. RHE. D) I-t curves of ZnCr:MMO. Reproduced with permission.[77] Copyright 2014, Royal Society of Chemistry.

to produce higher surface area materials to increase light capture, the interlayer space of the LDH was first intercalated with dodecyl sulfate (DS) anions. The performance of the obtained MMOs was evaluated in a DSSC, and exhibited a reasonable photovoltaic efficiency (0.64%) (Figure 18). However, this efficiency was still quite low compared with the widely used P25 TiO2 (2.08%) under the same testing conditions. Thus, the engineering of LDH-based semiconductors needs to be improved and optimized to improve their photovoltaic efficiency. In DSSCs, hybrid inorganic semiconductors and organic sensitizer systems are becoming popular because of their low

cost, ease of production and high efficiency. However, selecting suitable inorganic materials to increase the thermal or optical stability of the cells is challenging.[28] Kim reported a new hybrid consisting of anthraquinone sulfonate anion (AQS) and MgAl-LDH nanosheets (Figure 19).[81] Tyndall light scattering demonstrated the formation of the exfoliated LDH nanosheets in solution, and the red color of the suspension is ascribed to the reduced state of AQS2−. The obtained LDH-AQS nanosheets exhibit an overall photovoltaic efficiency of 0.20%, which improved the conversion efficiency up to 160% over an AQS-sensitized cell. This work represents the first example of a LDH nanosheet-organic hybrid light sensitizer in DSSC

Figure 17. A) Schematic of the fabrication of vertically aligned Ta3N5 nanorod arrays for PEC water splitting. B) Steady-state photocurrents of NiFe-LDH/Ta3N5 nanorod arrays. Reproduced with permission.[79] Copyright 2015, American Chemical Society.




(15 of 20) 1501974

www.advenergymat.de

REVIEW


Figure 18. A) I-V plot for DSSCs constructed using calcination products of ZnTi-LDH (a), ZnTi/DS-LDH (b) and P25 (c). B) Photocurrent spectrum of ZnTi-LDH solar cell. Reproduced with permission.[80] Copyright 2010, Royal Society of Chemistry.

devices. However, the power conversion efficiency of the LDHbased materials is still too low to justify practical application, though the combination of photoactive LDH nanosheets and a highly efficient organic sensitizer remains a promising route for fabricating DSSC cells with high conversion efficiencies and stabilities.

5. Photocatalytic CO2 Reduction Photocatalytic reduction of CO2 to CO, methane or another organic species is an attractive way to meet the growing demand for clean energy and to conserve fossil fuel resources.[82,83] LDH-based photocatalysts are attracting increasing attention for CO2 photoreduction due to their relatively low cost and specific advantages over other metal oxide photocatalysts, which include: 1) strong sorption capacity for CO2 in the interlayer space, and 2) tunable semiconductor properties (photocatalytic active sites, band gap, etc.) as a result of the choice of metal cations (such as Mg2+, Ni2+, Zn2+, In3+) in the layer structure.[84–89] Izumi et al. first reported that LDH can be used as a photocatalyst for CO2 reduction under UV-Vis light.[86] Among the various metal ions, divalent cations such as Mg, Co, Ni, Cu, Zn, and

trivalent cations such as Al or Ga are generally used due to their good CO2 sorption ability and potential for CO2 reduction. For example, Zn-containing LDH compounds show good promise for CO2 reduction to CO evolution (rates of 0.62 µmol g−1 h−1 have been reported), whereas Cu-containing LDH is effective for CO2 reduction to methanol. In order to improve the photoactivity of CO2 reduction, WO3 and CuZnGa-LDH were successfully integrated into a reverse polymer electrolyte photofuel cell (Figure 20).[90] Under visible light excitation, WO3 photoxidised H2O to O2, and CuZnGa-LDH reduced CO2 to CH3OH without any further requirement to separate the products. Pt, Pd and Au have been identified as efficient cocatalysts for the photoreduction of CO2 by ZnCr-LDH.[84,88,91] Besides the aforementioned metals (Cu, Zn, Ga, Al), the incorporation of Mg, Ni and In into LDH compounds is also effective in enhancing the photoreduction of CO2 into CO. CO evolution rates in the range of 0.6 to 3.6 µmol g−1 h−1 were reported for the Ni/Mg/Zn-Al/Ga/In-LDH.[92] Interestingly, Mg(OH)2 and In(OH)3, whose structure is very similar to MgIn-LDH did not show any obvious activity for CO evolution, which can be rationalized in terms of the inherent LDH structure and its ability to intercalate ions. However, detailed information about the effects of LDH particle size, surface area

Figure 19. A) Schematic showing LDH-AQS nanosheets, photographs of the LDH-AQS powder, and the exfoliated LDH-AQS nanosheet suspension. B) I-V curves of LDH-AQS and AQS. Reproduced with permission.[81] Copyright 2010, Wiley.

1501974 (16 of 20)





6. Conclusions and Perspectives

Figure 20. Schematic of the reverse photofuel cell and the electron transfer between WO3 and LDH in CO2 reduction. Reproduced with permission.[90] Copyright 2014, Royal Society of Chemistry.

and surface defects on CO2 reduction rates are not yet available, prompting further research. Recently, a self-assembled hybrid of carbon nitride and MgAl-LDH was successfully synthesized. After deposition of a Pd co-catalyst, the Pd/MgAl-LDH-C3N4 hybrid showed excellent photocatalytic activity for CO2 reduction to CH4, approximately 2.6 times higher than that observed for Pd/C3N4 (Figure 21A).[93] The reason for this improvement was the high affinity of the LDH layer for CO32−. Replacing MgAl-LDH with other more photoactive LDH materials such as NiAl-LDH, ZnAl-LDH or ZnCr-LDH did not enhance the CO2 reduction rate, indicating the catalytic or light absorbing properties of the LDH itself were not the key factor, but rather that performance was strongly connected to CO2 sorption capacity. Figure 21B shows that amongst LDHs, MgAl-LDH exhibits the highest absorption ability for CO2, explaining the improved CO2 reduction ability. Inspired by the enhanced CO2 absorption of MgAl-LDH, a MgO-Al2O3 MMO derived from MgAl-LDH was synthesized and grafted with TiO2 cuboids to form a hybrid photocatalyst for CO2 photoreduction.[94] The obtained photocatalyst achieved a CO2 reduction rate of 1.5 µmol g−1 h−1 at 323 K.

This review highlights recent advances in the application of LDH nanostructures to solar energy conversion. LDH materials clearly demonstrate great potential in this regard, due to compositional flexibility and also particle size effects that allow the atomic structure, surface defect concentration, and electronic and optical characteristics of LDH materials to be fine-tuned for specific applications. Using advanced characterization techniques, such as XAFS, ESR, XPS, and positron annihilation measurements, surface defects such as oxygen vacancies and low metal valence centers (e.g., Ti3+) associated with structural distortion in LDH nanomaterials can be clearly identified and quantified. Such information is critical for understanding electron transfer efficiency in photocatalysis. DFT calculations allow full exploration of structure-property relationships by modeling the band gap and the densities of states of metal ions in the host layer as well as the existence of surface defects. The various studies highlighted here confirm that nanostructured LDH materials with highly exposed active sites and large surface areas possess unique electronic structures that can enhance the visible light absorption range and charge transfer efficiencies, resulting in improved photocatalytic performance for water splitting as well as CO2 photoreduction. Despite the many recent advances described here, opportunities and challenges remain in the design, synthesis, and applications of nanostructured LDH-based photocatalysts: 1) The synthesis methods used to prepare nanostructured LDH photocatalysts can generally be described as “bottom-up” (reverse microemulsion methods), “top-down” (exfoliation

Figure 21. A) CH4 production, and B) room temperature CO2 adsorption isotherms over different catalysts. Reproduced with permission.[93] Copyright 2014, Wiley.




(17 of 20) 1501974

REVIEW

Table 3 summarizes the activity of various LDH photocatalysts for the photoreduction of CO2 to CO, CH3OH, or CH4. These data suggest that LDHs are a vibrant platform for the development of CO2 reduction photocatalysts, with or without the addition of noble metal cocatalysts. It should be noted that the CO2 photoreduction efficiency of LDH is still quite low (0.5–5 µmol g−1 h−1) except under strong UV light at this stage.[84,85,89] However, with the rapid growth in research interest in LDH materials, improved materials and composites for the photoreduction of CO2 are expected in the near future.

www.advenergymat.de

REVIEW

www.MaterialsViews.com Table 3. Summary of LDH-based photocatalysts for CO2 reduction to CO and hydrocarbons. Photocatalyst

Synthesized method

Mass [mg]

Co-catalyst

Reactant

Light source

Temperature [K]

Major products

Amounts [µmol g−1 h−1]

Ref

CO2

Reducing agent

NiM-LDH(M = Al, Ga, In)

Co-precipitation

100

None

500 µmol

Water (0.4mL)

200 W Hg-Xe lamp

RT

CO

1.9, 3.1, 3.6

[92]

MgM-LDH(M = Al, Ga, In)

Co-precipitation

100

None

500 µmol

Water (0.4mL)

200 W Hg-Xe lamp

RT

CO

2.4,2.6,1.6

[92]

ZnM-LDH(M = Al, Ga, In)

Co-precipitation

100

None

500 µmol

Water (0.4mL)

200 W Hg-Xe lamp

RT

CO

1.9, 0.7, 0.6

[92]

Zn3Al-LDH

Co-precipitation

100

None

180 µmol

H2 (1.7 mmol)

500 W Xe arc

305–313

CO, methanol

0.62, 0.039

[86]

ZnCuAl-LDH

Co-precipitation

100

None

180 µmol

H2 (1.7 mmol)

500 W Xe arc

305–313

CO, methanol

0.37, 0.13

[86]

ZnGa-LDH

Co-precipitation

100

None

180 µmol

H2 (1.7 mmol)

500 W Xe arc

305–313

CO, methanol

0.08, 0.051

[86]

ZnCuGa-LDH

Co-precipitation

100

None

180 µmol

H2 (1.7 mmol)

500 W Xe arc

305–313

CO, methanol

0.079, 0.17

[86,87]

ZnCr-LDH

Co-precipitation

50

Pt, Pd, Au

–

Water vapor

200 W Hg-Xe lamp

RT

CO

7.6, 4.7, 3.4

[91]

Hydrothermal

100

None

CO2

Water vapor

185 nm Hg

–

CH4, CO

77, 740

[95]

CuZnGa-LDH/WO3

Co-precipitation

45

None

CO2 in He

Water vapor

500 W Xe arc

323K

methanol

0.045

[90]

C3N4/MgAl-LDH

Co-precipitation /self-assmbled

200

Pd

200 torr

Water solution

500 W Hg Xe lamp

RT

CH4, CO

0.77, 0.20

[93]

MgAl-MMO/TiO2 cuboids

Hydrothermalcoprecipitation

100

None

CO2

Water

100 W Hg

323

CO

1.5

[94]

ZnAl-LDH nanosheet

Microemulsion hydrolysis

100

None

CO2

Water

300 W Xe

323

CO

7.6

[40]

ZnTi-LDH

methods), or self-assembly methods. Each of these methods has inherent limitations in terms of the size and size distribution of LDH particles that can be synthesized. There exists a pressing need to develop new synthesis methods that yield better control of both particle size and size distributions, as well as improved methods for the synthesis of core-shell, hybrid, or highly dispersed LDH nanostructures. The lateral size of most LDH nanomaterials reported to date is in the range 20–300 nm, with layer thicknesses ranging from a few layers to several hundred layers. Research emphasis needs to be directed at the synthesis of nanostructured LDHs with both nanosized lateral dimensions (