Homologous Compounds ZnnIn2O3+n (n = 4, 5 ... - ACS Publications

Research Article www.acsami.org

Homologous Compounds ZnnIn2O3+n (n = 4, 5, and 7) Containing Laminated Functional Groups as Efficient Photocatalysts for Hydrogen Production Meilin Lv,† Gang Liu,‡ and Xiaoxiang Xu*,† †

Shanghai Key Lab of Chemical Assessment and Sustainability, School of Chemical Science and Engineering, Tongji University, 1239 Siping Road, Shanghai 200092, China ‡ Shenyang National laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Science, 72 Wenhua Road, Shenyang 110016, China S Supporting Information *

ABSTRACT: Strong visible light absorption and high charge mobility are desirable properties for an efficient photocatalyst, yet they are hard to be realized simultaneously in a single semiconductor compound. In this work, we demonstrate that these properties coexist in homologous compounds ZnnIn2O3+n (n = 4, 5, and 7) with a peculiar layered structure that combines optical active segment and electrical conductive segment together. Their enhanced visible light absorption originates from tetrahedrally or trigonalbipyramidally coordinated In atoms in Zn(In)O4(5) layers which enable p−d hybridization between In 4d and O 2p orbitals so that valence band minimum (VBM) is uplifted with a reduced band gap. Theoretical calculations reveal their anisotropic features in charge transport and functionality of different constituent segments, i.e., Zn(In)O4(5) layers and InO6 layers as being for charge generation and charge collection, respectively. Efficient photocatalytic hydrogen evolution was observed in these compounds under full range (λ ≥ 250 nm) and visible light irradiation (λ ≥ 420 nm). High apparent quantum efficiency ∼2.79% was achieved for Zn4In2O7 under full range irradiation, which is almost 5-fold higher than their parent oxides ZnO and In2O3. Such superior photocatalytic activities of these homologous compounds can be understood as layer-by-layer packing of charge generation/collection functional groups that ensures efficient photocatalytic reactions. KEYWORDS: homologous compounds, photocatalyst, water splitting, layered materials, orbital hybridizations

1. INTRODUCTION How to fulfill the ever-growing energy demand of our modern society has been an important task in the 21st century.1,2 Recent reports on global energy market have alerted that primary fossil fuels (oil, coal, and natural gas, etc.) will be deplenished shortly in this century, jeopardizing the sustainability of our fossil fuel based economy.3 More importantly, the increasing awareness of the environmental problems associated with the use of fossil fuels has led to the strong incentive for searching and developing alternative clean energy resources.4 In these contexts, photocatalytic water splitting into hydrogen and oxygen, driven by solar energy, has been considered as a promising scenario for establishing a renewable clean energy infrastructure.5−16 This is largely because hydrogen is a clean energy vector, and solar insolation is inexhaustible in nature with a wide distribution all over the world. The successful application of this appealing technique largely relies on the fundamental improvement of photocatalytic materials, which by far lies in the efficient conversion of solar energy into hydrogen fuel.17 This mainly involves a substantial absorption of solar spectrum and rapid trans© XXXX American Chemical Society

portation of photogenerated charges to the surface reaction sites as far as a semiconductor material is concerned.17−19 However, most stable semiconductors for photocatalysis are oxides whose band gaps are too large to warrant appreciable absorption of solar spectrum (e.g., TiO2 and SrTiO3).20 Reduction in the band gaps of oxide semiconductors can easily break the glass ceiling of their solar to hydrogen efficiency (STH). For instance, a decrease of band gap value from above 3.2 to 2.5 eV will enormously increase the theoretical maximal STH from less than 1% to nearly 10%.21 On the other hand, charge transport within the semiconductor is of critical importance for smooth photocatalytic reactions to proceed. Sluggish electron/hole movement would strongly aid charge recombination and prevent charges from reaching surface reaction sites.22−25 Thereby, an efficient semiconductor photocatalyst shall be both a good light absorber and a proper charge conductor.26 Very few compounds can meet these stringent Received: August 30, 2016 Accepted: October 10, 2016

A

DOI: 10.1021/acsami.6b10951 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

to previous literature: proper amounts of Ni(NO3)2 aqueous solution were impregnated into 100 mg sample powders under magnetic stirring. The resultant slurry was then heated up to 363 K until dry and calcined at 773 K in 5% H2/95% Ar for 2 h and then at 473 K in O2 for 1 h.23 A 500 W high-pressure mercury lamp (NBeT, Merc-500) was used as the light source. Visible light was generated by rectifying the light beams with a UV cutoff filter (λ ≥ 420 nm). The photon flux of the lamp is calibrated using a quantum meter (Apogee MP-300). The recorded photon flux is ∼1543.9 μmol/(m2 s) for full range irradiation (λ ≥ 250 nm) and ∼668.5 μmol/(m2 s) for visible light irradiation (λ ≥ 420 nm). A water jacket was applied to maintain the reactor temperature stabilized around 293 K. Gas composition in the reactor was analyzed by an on-line gas chromatograph (TECHCOMP, GC7900) with a TCD detector (5 Å molecular sieve columns and Ar carrier gas). The apparent quantum efficiency (AQE) is then calculated using the following equation: apparent quantum efficiency = 2 × moles of hydrogen production per hour/moles of photon flux per hour × 100%. 2.4. Theoretical Calculations. Theoretical calculations were carried out using density functional theory (DFT) implemented in the Vienna Ab initio Simulation Package (VASP).43 The Perdew, Burke and Ernzerhof (PBE) exchange-correlation functional within the generalized gradient approximation (GGA)44 and the projector augmented-wave pseudopotential were used.45 A unit cell (a = b = 3.3362 Å, c = 33.526 Å, α = β = 90°, and γ = 120°) with hexagonal symmetry was built for Zn4In2O7 simulations. All geometry structures were fully relaxed until the forces on each atom are less than 0.01 eV/ Å. Static calculations were done with a 6 × 6 × 2 Monkhorst−Pack kpoint grid.46

requirements simultaneously, likely being the reason for the poor activity of many semiconductor photocatalysts. It has been realized that a series of homologous compounds with general formula ZnnIn2O3+n (n = 3, 4, 5, ...) have demonstrated interesting optical and conductive properties.27−30 Visible light absorbance (>400 nm) and high electrical conductivity (>2000 S/cm) are both achievable in these compounds.31,32 Their crystal structures can be roughly viewed as layer-by-layer alternate stacking of ZnO unit (wurtzite structure) and In2O3 unit (bixbyite structure) along the c direction.33−35 Such a laminate arrangement of photocatalytic active segment (ZnO) and electrical conductive segment (In2O3) brings these compounds promising applications in the field of photocatalysis and optoelectronics.27,30,36 During previous investigations, encouraging results in terms of photocatalytic water splitting have been reported among these compounds,37,38 yet detailed studies on the origin of their visible light absorption and associated photocatalytic activity are still lacking. In this work, we carried out an investigation on the optical and photocatalytic properties of three representative homologous compounds ZnnIn2O3+n (n = 4, 5, and 7).

2. EXPERIMENTAL SECTION 2.1. Material Synthesis. Homologous compounds ZnnIn2O3+n (n = 4, 5, and 7) were prepared by conventional solid state reactions: calculated amounts of ZnO (Aladdin, 99%) and In2O3 (Aladdin, 99.99%) were mixed using an agate mortar and pestle. Raw powders were pretreated in a muffle furnace at 500 °C for 3 h prior to weighing in order to remove moisture absorbed. In a typical synthesis of Zn4In2O7, 1.6442 g of ZnO and 1.3883 g of In2O3 were thoroughly blended and uniaxially pressed into pellets under a pressure of 5 tons. The pellets were then transferred into an alumina crucible and calcined at 1000 °C for 12 h and at 1300 °C for another 24 h. Intermediate grindings and recalcination were adopted until a single phase was reached. Pellets were finally ground into powders and collected for further measurements. 2.2. Materials Characterization. Phase purity and crystal structure were inspected using X-ray powder diffraction (XRD) techniques on a Bruker D8 Focus diffractometer. Incident X-ray radiation was Cu Kα1 (λ = 1.5406 Å) and Cu Kα2 (λ = 1.5444 Å). The step size was 0.01° for signal collection, and a collection time 0.1 s was applied at each step. General Structure Analysis System (GSAS) software package was applied to perform Rietveld refinement.39 A field emission scanning electron microscope (Hitachi S4800) and a transmission electron microscope (JEOL JEM-2100) were used to examine the microstructures of as-prepared samples. Diffuse reflectance spectra were collected on a UV−vis spectrophotometer coupled with integrating sphere (JASCO-750), and data were analyzed using JASCO software suite. BaSO4 was used as a reference nonabsorbing material.40 Surface conditions and valence band positions were analyzed using X-ray photoelectron spectroscopy (Thermo Escalab 250 with a monochromatic Al Kα X-ray source). All binding energies were referred to adventitious carbon C 1s peak at 284.7 eV.41 The surface area of as-prepared samples was analyzed on a Micromeritics instrument TriStar 3000 and was calculated based on the Brunauer−Emmett−Teller (BET) model. 2.3. Photocatalytic Activity. Photocatalytic properties of asprepared samples were evaluated by monitoring their hydrogen production in the presence of sacrificial agents under light illumination. The experiments were performed in a top-irradiationtype reactor connected to a gas-closed circulation and evacuation system (Perfect Light, Labsolar-IIIAG).42 The gas pressure inside the reactor is around 100 Pa. In a typical experiment, 0.1 g sample powders were ultrasonically dispersed into 100 mL Na2SO3 aqueous solution (0.05 M), which was then sealed and evacuated in the reactor. Na2SO3 was served as a hole scavenger to promote water reduction. 1 wt % NiOx was loaded onto sample powders as a cocatalyst according

3. RESULTS AND DISCUSSION 3.1. Phase Purity and Crystal Structure. The synthesis of homologous compounds ZnnIn2O3+n (n = 3, 4, 5, ...) generally involves high temperature calcination (>1000 °C) for a long period of time (>10 h).27,33 Several heating cycles at elevated temperatures (1300 °C) were adopted during our synthesis, and phase compositions were analyzed at the end of each cycle until a single phase compound was reached. Figure 1 displays the X-ray powder diffraction (XRD) patterns of as-prepared samples Zn4In2O7, Zn5In2O8, and Zn7In2O10. For sample Zn4In2O7, the XRD patterns can be indexed using a hexagonal

Figure 1. Observed and calculated X-ray diffraction patterns of asprepared sample powders. Rietveld refinement all converged with good χ2 factors (χ2 = 1.751, 1.735, and 1.889 for Zn4In2O7, Zn5In2O8, and Zn7In2O10, respectively). B


Research Article


Figure 2. Schematic representation of crystal structures from [010] direction for Zn4In2O7, Zn5In2O8, and Zn7In2O10; parent oxides ZnO and In2O3 are also shown for comparison. Unit cells are marked by the blue lines.

Figure 3. Field emission scanning electron microscopy images of Zn4In2O7 (a), Zn5In2O8 (b), and Zn7In2O10 (c); a digital photograph of asprepared sample powders with their parent oxides ZnO and In2O3 (d).

(Figure 2). Ordering of In atom in the ZnO slab (i.e., “zigzag fringe”) has been observed in previous studies and is believed to be thermodynamically more stable.48−51 The random configuration of In in ZnO slab in our case can be understood due to the high temperature calcination that ease atom interchange as well as a rapid cooling rate (∼10 °C) that “frozen” atomic arrangement at the initial state.36,49 More importantly, the refined crystal structures of homologous compounds clearly indicate a strong correlation with their parent oxides ZnO and In2O3.33 Layers of wurtzite type and bixbyite type structure can be easily identified according to their distinct coordination numbers (CN). Wurtzite type structure similar to ZnO is largely maintained in these homologous

symmetry while for Zn5In2O8 and Zn7In2O10, a rhombohedral symmetry is needed to include all the diffraction peaks. These observations are consistent with previous reports that ZnnIn2O3+n crystallize in space group R-3m for odd n values and P63/mmc for even n values.27,47 We then performed a Rietveld refinement on the XRD data in accord with the space group reported, and the results are illustrated in Figures 1 and 2. Reasonable goodness-of-fit parameters were only achieved with the constraints that both Zn and In atoms occupy some of the crystallographic positions (Wyckoff site 2c, 4e, and 4f for Zn4In2O7 and 6c for Zn5In2O8 and Zn7In2O10). In other words, incorporation of In into ZnO slab is random, as can be seen from the schematic representation of refined crystal structures C


Research Article

ACS Applied Materials & Interfaces Table 1. Unit Cell Parameters, BET Surface Area, and Band Gap Value of As-Prepared Samples sample

space group

a/Å

c/Å

V/Å3

BET surface area/(m2/g)

band gap/eV

Zn4In2O7 Zn5In2O8 Zn7In2O10

P63/mmc R-3m R-3m

3.3361(2) 3.3330(3) 3.3107(2)

33.544(2) 58.275(5) 73.730(4)

323.33(4) 560.65(9) 699.88(8)

2.830 2.834 3.884

2.57 2.61 2.67

Figure 4. High-resolution transmission electron microscopy image of Zn4In2O7 showing the (002) lattice fringes; a refined crystal structure is illustrated on the right for visual inspections.

Figure 5. (a) UV−vis light absorption spectra (converted from diffuse reflectance spectra) of as-prepared samples and (b) Kubelka−Munk transformation of diffuse reflectance data.

be envisaged according to their surface terrace (indicated by the yellow arrow on the SEM images). Microstructures of homologous compounds were further analyzed using transmission electron microscopy (TEM). Figure 4 displays a typical high-resolution TEM image of sample Zn4In2O7 showing the (002) lattice fringes. The HRTEM image is characterized by a periodic separation of five fringes with one heavier dark fringe (marked by yellow arrow in Figure 4), which closely matches with the refined crystal structure along [001] direction (Figure 4, right side), confirming the correctness of refinement. The heavier dark fringe therefore represents InO6 octahedral layer, and the other fringes belong to Zn(In)O4(5) layers.33 Disordering of In in the Zn(In)O4(5) layer is supported by the absence of “zigzag fringe” within these layers, suggesting a random substitution of Zn with In. 3.3. Visible Light Absorption. The color of all homologous compounds synthesized is yellow (Figure 3d),

compounds although deviation from tetrahedral coordination (CN = 4) to trigonal-bipyramidal coordination (CN = 5) occurs. This is mainly due to the off-centering of Zn/In atoms in the tetrahedral site toward the center plane of the slab. The thickness of the wurtzite layer is governed by the number of n in ZnnIn2O3+n, and these wurtzite layers are then sandwiched between two layers of In2O3 containing edge-shared InO6 octahedrons (CN = 6). 3.2. Microstructures. The microstructure of freshly prepared sample powders were then inspected under fieldemission scanning electron microscopy (FESEM) conditions. Figure 3 displays the typical SEM images of Zn4In2O7, Zn5In2O8, and Zn7In2O10. Very large bulky particles with the size of several micrometers can be seen in all cases, arising from repeated calcination of sample powders at high temperatures. This is consistent with their poor surface area that is generally less than 4 m2/g (Table 1). Nevertheless, all their particles are characterized by the sharp edge, and their laminal structure can D


Research Article


Figure 6. (a) XPS valence band scan of Zn4In2O7, In2O3, and ZnO, regions for valence band maximum are enlarged in the inserted images. (b) Schematic representation of p−d hybridization among constituent elements of Zn4In2O7, In2O3, and ZnO. p−d hybridization is symmetry allowed in Zn4In2O7 and ZnO (same t2 symmetry for p and d orbitals in Td point group) whereas it is symmetry forbidden in In2O3 (different symmetry for p (t1) and d (t2) orbitals in Oh point group).

Figure 7. (a) Temporal photocatalytic hydrogen production for sample powders (100 mg) under full range irradiation (λ ≥ 250 nm) in sodium sulfite aqueous solution (0.05 M). (b) Average photocatalytic hydrogen production rate under full range irradiation (λ ≥ 250 nm) and visible light irradiation (λ ≥ 420 nm).

compared to their parent oxides (Figure 5b and Table 1).27 Band gap reduction of these compounds has been recognized during previous studies, yet the origin of this reduction has not been fully investigated or discussed.27,52 Here we tentatively interpret the cause of band gap reduction with the aid of XPS valence band scan and molecular orbital (MO) theory. Figure 6a displays XPS valence band scan data of homologous compound Zn4In2O7 and its parent oxides ZnO and In2O3. Several core-level states can be easily identified: peaks around 18 and 10 eV typically belong to In 4d10 and Zn 3d10 states, respectively, whereas broad peaks around 5 eV are ascribed to O 2p6 valence state due to small XPS cross section of O 2p states.41 The position of valence band maximum (VBM) was evaluated by taking a linear extrapolation of the onset of

strikingly different from their parent oxides ZnO (white) and In2O3 (light yellow). The strong color of these compounds indicates substantial visible light absorptions. This is confirmed from their UV−vis light absorption spectra in comparison with ZnO and In2O3 (Figure 5). The absorption spectrum of pristine ZnO is characterized by a steep absorption edge below 400 nm, confirming its wide band gap nature. The light yellow color of In2O3 can be ascribed to the small absorption tail extending up to 450 nm. Thereby, neither ZnO nor In2O3 demonstrates significant absorption in the visible light region. However, substantial visible light absorption can be realized in homologous compounds according to their absorption spectra, which clearly show absorption edges approaching 500 nm. Band gaps of homologous compounds are thus greatly reduced E


Research Article


as layered perovskite compound Sr2TiO4 and spinel compound Zn2TiO4 which exhibit AQE around 0.16% and 0.74% under the same conditions,16,56 respectively, these homologous compounds show a much higher activity. More importantly, hydrogen production under visible light irradiation (λ ≥ 420 nm) was also achieved for homologous compounds. The highest average hydrogen production rate ∼5.4 μmol/h was also found for Zn4In2O7 under visible light irradiation, corresponding to AQE ∼ 0.16%. On the contrary, negligible amounts of hydrogen were produced for ZnO and In2O3 under the same conditions, highlighting the benefits of constituting homologous compounds by intergrowth between ZnO and In2O3. On the other hand, there is a clear correlation between hydrogen production and slab thickness of homologous compounds, as an increase of n in ZnnIn2O3+n significantly decreases the activity (Figure 7b). This observation is consistent with previous report37 and probably not a result of surface properties of these compounds as their surface area and surface state are more or less the same (Table 1 and Figure S1). It is worth mentioning that the surface area of these homologous compounds is quite small, which implies that their photocatalytic activity can be further enhanced simply by enlarging their surface area. Strategies for increasing surface area includes ball-milling of sample powders or seeking other synthetic methods using low calcination temperatures and will be our future study. 3.5. Theoretical Calculations. For better understanding the photocatalytic properties of homologous compounds, especially the structural dependence of catalytic activity, we performed theoretical calculations on the electronic structure of Zn4In2O7. The calculated results are displayed in Figure 8

valence band emission, which is read to be 2.03, 2.29, and 2.69 eV for Zn4In2O7, In2O3, and ZnO, respectively. Apparently, there is an uplift of VBM for homologous compound Zn4In2O7 in comparison with parent oxides. According to MO theory, close lying of In 4d10 and Zn 3d10 states to O 2p6 states may suggest possible hybridization of these states (Figure 6b). This is true for ZnO where Zn is tetrahedrally coordinated to O (Td point group, no inversion center); therein direct p−d hybridization is symmetry allowed,53 while p−d hybridization is symmetry forbidden for In2O3 as all In is octahedrally coordinated to O (Oh point group, with inversion center). In the case of Zn4In2O7, however, half of In is accommodated in the ZnO slab where In is either tetrahedrally or trigonalbipyramidally coordinated to O (Td or D3h point group, no inversion center); p−d hybridization becomes symmetry allowed between In and O (Figure 6b). It is known p−d hybridization has important consequences that O 2p6 states are raised to higher energy due to the antibonding contribution from Zn 3d10 and In 4d10 states (i.e., p−d repulsion).53,54 The additional hybridization between In 4d10 and O 2p6 states brings Zn4In2O7 a higher VBM compared to In2O3 and ZnO; thereby, a smaller band gap provided that conduction band minimum (CBM) is not severely altered. These considerations therefore well explain the observations in XPS valence band scan and UV−vis light absorption spectra. Moreover, the slight increase in the band gap value from Zn4In2O7 to Zn7In2O10 can be easily deduced from the decreased In content in the structure that contributes less p−d hybridization to the VBM. 3.4. Photocatalytic Properties. The photocatalytic properties of freshly prepared samples were then investigated by monitoring hydrogen evolution in the presence of sacrificial agent (Na2SO3, 0.05 M). 1 wt % NiOx was loaded onto sample powders as a cocatalyst to promote photoreduction reactions. This cocatalyst is composed of both metallic Ni and NiO therefore is denoted as NiOx (0 ≤ x ≤ 1).10 According to previous analysis, x roughly equals 0.85 under the experimental conditions in this study.55 Control experiments in the absence oflight irradiation or photocatalysts induce no hydrogen evolution, thus precluding any spontaneous reactions that lead to hydrogen production. Immediate hydrogen evolution was detected once sample powders were irradiated, indicating real photocatalytic reactions. The temporal photocatalytic hydrogen production for all samples is presented in Figure 7a, along with their parent oxides ZnO and In2O3 for comparison. It is clear from the figure that all homologous compounds Zn4In2O7, Zn5In2O8, and Zn7In2O10 demonstrate a much higher activity than their parent oxides under full range irradiation (λ ≥ 250 nm). The highest activity belongs to Zn4In2O7, where more than 660 μmol of hydrogen was produced within 3 h irradiation time. Repeated photocatalytic reactions on Zn4In2O7 for another two cycles indicated no apparent degradation of catalytic activity, and XRD analysis precluded structural changes before and after reactions (Figures S2 and S3). Considering the much less amounts of sample powders used (∼84 μmol of Zn4In2O7) and vigorous bubbling at their surface under irradiation, these homologous compounds are highly active photocatalysts. The average hydrogen production rates of all samples are summarized in Figure 7b. Zn4In2O7 gives the highest average hydrogen production rate ∼219.1 μmol/h, corresponding to apparent quantum efficiency (AQE) ∼ 2.79%. This value is almost 5 times higher than pristine ZnO and In2O3, which only give AQE ∼ 0.56%. Compared with other photocatalysts investigated recently, such

Figure 8. Calculated band structure, total density of states (DOS), and partial density of states (PDOS) of constituent elements for Zn4In2O7. The Fermi level is marked by the dotted orange line.

which covers the energy window close to Fermi level. The semiconductivity of Zn4In2O7 is confirmed by a calculated direct band gap of 0.13 eV at Γ point. Although this value is much smaller than the experimental one (2.57 eV), likely being the drawbacks of the generalized gradient approximation (GGA) method for underestimating band gaps, 57 the calculation offers qualitative predictions. Close examination of the band structure reveals an anisotropic phenomenon of this compound. Take the conduction band (CB) for instance; the band dispersion from Γ to M governs the electron behavior in F


Research Article


Figure 9. Schematic representation of photocatalytic mechanisms in homologous compound Zn4In2O7 with different functional parts.

In light of the above analysis, it is now worthwhile to discuss the superior photocatalytic activity of homologous compounds from the viewpoint of their peculiar layered structures, as can be seen from Figure 9. Zn(In)O4(5) layers contain tetrahedrally or trigonal-bipyramidally coordinated In atoms and thus own strong light absorption due to In/O p−d hybridization that reduces the band gap. Efficient charge generation can be envisaged within these layers that involve substantial amounts of In atoms. InO6 octahedral layers, on the other hand, serve as charge collectors due to their high charge mobility, which provide fast charge transportation pathways to the surface reaction sites. The intimate contact between Zn(In)O4(5) layers and InO6 layers guarantees a rapid delivery of photogenerated charges to the surface, and each pair of InO6−Zn(In)O4(5) layers can be treated as an independent photoelectrochemical cell. The superior photocatalytic activity of these homologous compounds can then be understood as layer-by-layer compaction of individual charge generation/collection functional group that establishes an efficient network for photocatalytic reactions.

the [100] crystallographic direction and is characterized by a wide energy variations of nearly 4 eV. Contrarily, all bands are nearly pinned at constant energy levels from K to H point, corresponding to electron behavior along the [001] direction. It is well-known that the effective mass (m*) is proportional to the second derivatives of of E versus k curve (i.e., wide band gives small m*):58,59 ⎛ d2E ⎞−1 m* = ℏ2⎜ 2 ⎟ ⎝ dk ⎠

Thereby, electrons therefore are facile to move along the [100] direction (small m* suggesting high mobility) but are forbidden to transport along the [001] direction (large m* suggesting poor mobility). Examination on other directions reveals that relatively small m* occurs for all directions that are perpendicular to the c-axis; therefore, intralayer electron mobility is very high rather than interlayer one. A similar situation is also realized in the valence band, implying that homologous compound shall be highly anisotropic in electrical conductivity. This was confirmed by previous conductivity measurements that in-plane conductivity is about 2 orders of magnitude larger than that along the c-axis.31 On the other hand, DOS and PDOS (Figure 8 and Figure S2) reveal that In 5s orbitals have a major contribution to the conduction band (CB) and are in conformity with previous calculations that CBM tends to locate on the In−O networks. This explains the high conductivity of homologous compounds and the general trend for a better conductivity at lower n, as In 5s orbitals have a much wider dispersion than Zn 4s orbitals and In2O3 has a much higher electron mobility than ZnO.27,30,60−62 Thereby, fast electron transport is more likely to occur in InO6 octahedral layers rather than Zn(In)O4(5) layers. In addition, contributions from In 4d orbitals to the valence band (VB) is clearly seen in the DOS and PDOS (Figure 8 and Figure S2 for enlarged plot), in agreement with previous analysis that p−d hybridization between In and O is symmetry allowed in tetrahedral or trigonal-bipyramidal coordination (i.e., in Zn(In)O4(5) layers). Thereby, band gap reduction and visible light absorption occur mainly in Zn(In)O4(5) layers rather than InO6 octahedral layers.

4. CONCLUSIONS We have prepared three homologous compounds ZnnIn2O3+n (n = 4, 5, and 7) through conventional solid state reactions. Their crystal structures and other physicochemical properties were investigated by various analytic techniques. In particular, the origin of band gap reduction and improved visible light absorption in comparison with parent oxides ZnO and In2O3 was studied. Homologous compounds ZnnIn2O3+n contain tetrahedrally or trigonal-bipyramidally coordinated In atoms which enable p−d hybridization between In 4d and O 2p orbitals. Valence band minimum (VBM) is then raised to a higher energy level due to the antibonding character of p−d hybridization, therefore reducing the band gap. All three homologous compounds demonstrate much improved photocatalytic activities over parent oxides ZnO and In2O3. The highest activity achieved belongs to Zn4In2O7 with average hydrogen production rate ∼219.1 μmol/h under full range irradiation (λ ≥ 250 nm), corresponding to AQE ∼ 2.79%. This value is almost 5-fold higher than parent oxides. More G


Research Article


(11) Wang, X. C.; Maeda, K.; Thomas, A.; Takanabe, K.; Xin, G.; Carlsson, J. M.; Domen, K.; Antonietti, M. A Metal-Free Polymeric Photocatalyst for Hydrogen Production from Water under Visible Light. Nat. Mater. 2009, 8 (1), 76−80. (12) Xu, X. X.; Randorn, C.; Efstathiou, P.; Irvine, J. T. S. A Red Metallic Oxide Photocatalyst. Nat. Mater. 2012, 11 (7), 595−598. (13) Lu, L. W.; Lv, M. L.; Wang, D.; Liu, G.; Xu, X. X. Efficient Photocatalytic Hydrogen Production over Solid Solutions Sr1‑xBixTi1‑xFexO3 (0 ≤ x ≤ 0.5). Appl. Catal., B 2017, 200, 412−419. (14) Xu, X. X.; Liu, G.; Azad, A. K. Visible Light Photocatalysis by in situ Growth of Plasmonic Ag Nanoparticles upon AgTaO3. Int. J. Hydrogen Energy 2015, 40, 3672−3678. (15) Xu, X. X.; Azad, A. K.; Irvine, J. T. S. Photocatalytic H2 Generation from Spinels ZnFe2O4, ZnFeGaO4 and ZnGa2O4. Catal. Today 2013, 199, 22−26. (16) Sun, X. Q.; Wang, S. W.; Shen, C.; Xu, X. X. Efficient Photocatalytic Hydrogen Production over Rh-Doped Inverse Spinel Zn2TiO4. ChemCatChem 2016, 8, 2289−2295. (17) Osterloh, F. E. Inorganic Materials as Catalysts for Photochemical Splitting of Water. Chem. Mater. 2008, 20 (1), 35−54. (18) Miseki, Y.; Kato, H.; Kudo, A. Water Splitting into H2 and O2 over Niobate and Titanate Photocatalysts with (111) Plane-type Layered Perovskite Structure. Energy Environ. Sci. 2009, 2 (3), 306− 314. (19) Wu, F. F.; Lv, M. L.; Sun, X. Q.; Xie, Y. H.; Chen, H. M.; Ni, S.; Liu, G.; Xu, X. X. Efficient Photocatalytic Oxygen Production over Nitrogen-Doped Sr4Nb2O9 under Visible-Light Irradiation. ChemCatChem 2016, 8 (3), 615−623. (20) Scaife, D. E. Oxide Semiconductors in Photoelectrochemical Conversion of Solar-Energy. Sol. Energy 1980, 25 (1), 41−54. (21) Chen, Z. B.; Jaramillo, T. F.; Deutsch, T. G.; KleimanShwarsctein, A.; Forman, A. J.; Gaillard, N.; Garland, R.; Takanabe, K.; Heske, C.; Sunkara, M.; McFarland, E. W.; Domen, K.; Miller, E. L.; Turner, J. A.; Dinh, H. N. Accelerating Materials Development for Photoelectrochemical Hydrogen Production: Standards for Methods, Definitions, and Reporting Protocols. J. Mater. Res. 2010, 25 (1), 3− 16. (22) Xie, Y. H.; Wang, Y. W.; Chen, Z. F.; Xu, X. X. Role of Oxygen Defects on the Photocatalytic Properties of Mg-doped Mesoporous Ta3N5. ChemSusChem 2016, 9 (12), 1403−1412. (23) Lv, M. L.; Wang, Y. W.; Lu, L. W.; Wang, R. N.; Ni, S.; Liu, G.; Xu, X. X. Structural Dependence of the Photocatalytic Properties of Double Perovsite Compounds A2InTaO6 (A = Sr or Ba) Doped with Nickel. Phys. Chem. Chem. Phys. 2016, 18, 21491−21499. (24) Lv, M. L.; Ni, S.; Wang, Z.; Cao, T. C.; Xu, X. X. Cation Ordering/Disordering Effects upon Photocatalytic Activity of CrNbO4, CrTaO4, Sr2CrNbO6 and Sr2CrTaO6. Int. J. Hydrogen Energy 2016, 41, 1550−1558. (25) Chen, H. M.; Xie, Y. H.; Sun, X. Q.; Lv, M. L.; Wu, F. F.; Zhang, L.; Li, L.; Xu, X. X. Efficient Charge Separation Based on Type-II gC3N4/TiO2-B Nanowire/Tube Heterostructure Photocatalysts. Dalton T. 2015, 44 (29), 13030−13039. (26) Khaselev, O.; Turner, J. A. A Monolithic PhotovoltaicPhotoelectrochemical Device for Hydrogen Production via Water Splitting. Science 1998, 280 (5362), 425−427. (27) Moriga, T.; Edwards, D. D.; Mason, T. O.; Palmer, G. B.; Poeppelmeier, K. R.; Schindler, J. L.; Kannewurf, C. R.; Nakabayashi, I. Phase Relationships and Physical Properties of Homologous Compounds in the Zinc Oxide-Indium Oxide System. J. Am. Ceram. Soc. 1998, 81 (5), 1310−1316. (28) Hopper, E. M.; Zhu, Q. M.; Song, J. H.; Peng, H. W.; Freeman, A. J.; Mason, T. O. Electronic and Thermoelectric Analysis of Phases in the In2O3(ZnO)(k) system. J. Appl. Phys. 2011, 109 (1), 013713. (29) Peng, H. W.; Song, J. H.; Hopper, E. M.; Zhu, Q. M.; Mason, T. O.; Freeman, A. J. Possible n-type Carrier Sources in In2O3(ZnO)(k). Chem. Mater. 2012, 24 (1), 106−114. (30) Walsh, A.; Da Silva, J. L. F.; Yan, Y. F.; Al-Jassim, M. M.; Wei, S. H. Origin of Electronic and Optical Trends in Ternary In2O3(ZnO)(n) Transparent Conducting Oxides (n = 1,3,5): Hybrid Density

importantly, visible light activity was also realized in these homologous compounds. The highest activity under visible light irradiation (λ ≥ 420 nm) was also found for Zn4In2O7 with average hydrogen production rate ∼5.4 μmol/h, corresponding to AQE ∼ 0.16%. Theoretical calculations reveal the anisotropic features of these layered compounds, particularly the different functions of Zn(In)O4(5) layers and InO6 layers within the structure, being for charge generation and charge collection, respectively. The superior photocatalytic activity of these homologous compounds can be understood as layer-by-layer packing of charge generation/collection functional groups that facilitate efficient charge utilizations.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b10951. XPS of all samples, repeated temporal photocatalytic hydrogen production of Zn4In2O7, XRD before and after photocatalytic reactions, enlarged DOS and PDOS (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]; Tel +86-21-65986919 (X.X.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Young Scientists Fund of the National Natural Science Foundation of China (Grant No. 21401142) for funding and Recruitment Program of Global Youth Experts (1000 plan). The work was supported by Shanghai Science and Technology Commission (14DZ2261100) and the Fundamental Research Funds for the Central Universities.

■

REFERENCES

(1) Nocera, D. G. Living Healthy on a Dying Planet. Chem. Soc. Rev. 2009, 38 (1), 13−15. (2) Lewis, N. S.; Nocera, D. G. Powering the Planet: Chemical Challenges in Solar Energy Utilization. Proc. Natl. Acad. Sci. U. S. A. 2006, 103 (43), 15729−15735. (3) Dudley, B. BP Statistical Review of World Energy June 2014, 2014. (4) Armstrong, R. C. The Frontiers of Energy. Nat. Energy 2016, 1, 15020. (5) Hoffmann, M. R.; Martin, S. T.; Choi, W. Y.; Bahnemann, D. W. Environmental Applications of Semiconductor Photocatalysis. Chem. Rev. 1995, 95 (1), 69−96. (6) Chen, X. B.; Shen, S. H.; Guo, L. J.; Mao, S. S. SemiconductorBased Photocatalytic Hydrogen Generation. Chem. Rev. 2010, 110 (11), 6503−6570. (7) Kudo, A.; Miseki, Y. Heterogeneous Photocatalyst Materials for Water Splitting. Chem. Soc. Rev. 2009, 38 (1), 253−278. (8) Gratzel, M. Photoelectrochemical cells. Nature 2001, 414 (6861), 338−344. (9) Maeda, K.; Teramura, K.; Lu, D. L.; Takata, T.; Saito, N.; Inoue, Y.; Domen, K. Photocatalyst Releasing Hydrogen from Water Enhancing Catalytic Performance Holds Promise for Hydrogen Production by Water Splitting in Sunlight. Nature 2006, 440 (7082), 295−295. (10) Zou, Z. G.; Ye, J. H.; Sayama, K.; Arakawa, H. Direct Splitting of Water under Visible Light Irradiation with an Oxide Semiconductor Photocatalyst. Nature 2001, 414 (6864), 625−627. H


Research Article

ACS Applied Materials & Interfaces Functional Theory Calculations. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 79 (7), 073105. (31) Malochkin, O.; Seo, W. S.; Koumoto, K. Thermoelectric Properties of (ZnO)(5)In2O3 Single Crystal Grown by a Flux Method. Jpn. J. Appl. Phys. 2 2004, 43 (2A), L194−L196. (32) Moriga, T.; Kammler, D. R.; Mason, T. O.; Palmer, G. B.; Poeppelmeier, K. R. Electrical and Optical Properties of Transparent Conducting Homologous Compounds in the Indium-Callium-Zinc Oxide System. J. Am. Ceram. Soc. 1999, 82 (10), 2705−2710. (33) Cannard, P. J.; Tilley, R. J. D. New Intergrowth Phases in the ZnO-In2O3 System. J. Solid State Chem. 1988, 73, 418−429. (34) Schinzer, C.; Heyd, F.; Matar, S. F. Zn3In2O6 - Crystallographic and Electronic Structure. J. Mater. Chem. 1999, 9 (7), 1569−1573. (35) Na, C. W.; Bae, S. Y.; Park, J. Short-Period Superlattice Structure of Sn-Doped In2O3(ZnO)(4) and In2O3(ZnO)(5) Nanowires. J. Phys. Chem. B 2005, 109 (26), 12785−12790. (36) Wen, J.; Zhang, X. T.; Gao, H. Effects of the Slab Thickness on the Crystal and Electronic Structures of In2O3(ZnO)(m) Revealed by First-Principles Calculations. J. Solid State Chem. 2015, 222, 25−36. (37) Kudo, A.; Mikami, I. New In2O3(ZnO)(m) Photocatalysts with Laminal Structure for Visible Light-induced H2 or O2 Evolution from Aqueous Solutions Containing Sacrificial Reagents. Chem. Lett. 1998, 27, 1027−1028. (38) Narendranath, S. B.; Yadav, A. K.; Bhattacharyya, D.; Jha, S. N.; Devi, R. N. Photocatalytic H-2 Evolution from Water-Methanol System by Anisotropic InFeO3(ZnO)(m) Oxides without Cocatalyst in Visible Light. ACS Appl. Mater. Interfaces 2014, 6 (15), 12321−12327. (39) Larson, A. C.; Von Dreele, R. B. GSAS-Generalised Crystal Structure Analysis System. Los Alamos National Laboratory Report No. LA-UR-86-748, 1994. (40) Kortum, G.; Braun, W.; Herzog, G. Principles and Techniques of Diffuse-Reflectance Spectroscopy. Angew. Chem., Int. Ed. Engl. 1963, 2 (7), 333−341. (41) Van der Heide, P. X-ray Photoelectron Spectroscopy - An Introduction to Principles and Practices; John Wiley & Sons, Inc.: Hoboken, NJ, 2012. (42) Lv, M. L.; Xie, Y. H.; Wang, Y. W.; Sun, X. Q.; Wu, F. F.; Chen, H. M.; Wang, S. W.; Shen, C.; Chen, Z. F.; Ni, S.; Liu, G.; Xu, X. X. Bismuth and Chromium Co-Doped Strontium Titanates and Their Thotocatalytic Properties under Visible Light Irradiation. Phys. Chem. Chem. Phys. 2015, 17 (39), 26320−26329. (43) Kresse, G.; Furthmuller, J. Efficient Iterative Schemes for ab initio Total-Energy Calculations Using a Plane-Wave Basis Set. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54 (16), 11169−11186. (44) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77 (18), 3865− 3868. (45) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59 (3), 1758−1775. (46) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B 1976, 13 (12), 5188−5192. (47) Yoshinari, A.; Ishida, K.; Murai, K. I.; Moriga, T. Crystal and Electronic Band Structures of Homologous Compounds ZnkIn2Ok+3 by Rietveld Analysis and First-Principle Calculation. Mater. Res. Bull. 2009, 44 (2), 432−436. (48) Goldstein, A. P.; Andrews, S. C.; Berger, R. F.; Radmilovic, V. R.; Neaton, J. B.; Yang, P. D. Zigzag Inversion Domain Boundaries in Indium Zinc Oxide-Based Nanowires: Structure and Formation. ACS Nano 2013, 7 (12), 10747−10751. (49) Zhang, X. T.; Lu, H. Q.; Gao, H.; Wang, X. J.; Xu, H. Y.; Li, Q.; Hark, S. K. Crystal Structure of In2O3(ZnO)(m) Superlattice Wires and Their Photoluminescence Properties. Cryst. Growth Des. 2009, 9 (1), 364−367. (50) Da Silva, J. L. F.; Yan, Y. F.; Wei, S. H. Rules of Structure Formation for the Homologous InMO(3)(ZnO)(n) Compounds. Phys. Rev. Lett. 2008, 100 (25), 255501. (51) Li, C. F.; Bando, Y.; Nakamura, M.; Onoda, M.; Kimizuka, N. Modulated Structures of Homologous Compounds InMO3(ZnO)(m)

(M = In, Ga; m = integer) Described by Four-Dimensional Superspace Group. J. Solid State Chem. 1998, 139 (2), 347−355. (52) Hiramatsu, H.; Seo, W. S.; Koumoto, K. Electrical and Optical Properties of Radio-Frequency-Sputtered Thin Films of (ZnO)(5)In2O3. Chem. Mater. 1998, 10 (10), 3033−3039. (53) Wei, S. H.; Zunger, A. Role of Metal d States in II-VI Semiconductors. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37 (15), 8958−8981. (54) Hu, C. C.; Lee, Y. L.; Teng, H. S. Influence of Indium Doping on the Activity of Gallium Oxynitride for Water Splitting under Visible Light Irradiation. J. Phys. Chem. C 2011, 115 (6), 2805−2811. (55) Domen, K.; Kudo, A.; Onishi, T.; Kosugi, N.; Kuroda, H. Photocatalytic Decomposition of Water into H2 and O2 over NiOSrTiO3 Powder. 1. Structure of the Catalyst. J. Phys. Chem. 1986, 90, 292−295. (56) Sun, X. Q.; Xie, Y. H.; Wu, F. F.; Chen, H. M.; Lv, M. L.; Ni, S.; Liu, G.; Xu, X. X. Photocatalytic Hydrogen Production over Chromium Doped Layered Perovskite Sr2TiO4. Inorg. Chem. 2015, 54 (15), 7445−7453. (57) Xiao, H.; Tahir-Kheli, J.; Goddard, W. A. Accurate Band Gaps for Semiconductors from Density Functional Theory. J. Phys. Chem. Lett. 2011, 2 (3), 212−217. (58) Kittel, C. Introduction to Solid State Physics, 7th ed.; Wiley: New York, 1996; p xi, 673 pp. (59) Mizoguchi, H.; Woodward, P. M. Electronic Structure Studies of Main Group Oxides Possessing Edge-Sharing Octahedra: Implications for the Design of Transparent Conducting Oxides. Chem. Mater. 2004, 16 (25), 5233−5248. (60) Hamberg, I.; Granqvist, C. G. Evaporated Sn-doped In2O3 Films: Basic Optical Properties and Applications to Energy Efficient Windows. J. Appl. Phys. 1986, 60, R123−R159. (61) Alfaramawi, K. Electric Field Dependence of the Electron Mobility in Bulk Wurtzite ZnO. Bull. Mater. Sci. 2014, 37 (7), 1603− 1606. (62) OhhaTa, Y.; Shinoki, F.; Yoshida, S. Optical Properties of R.F. Reactive Sputtered Tin-doped In2O3 Films. Thin Solid Films 1979, 59, 255−261.

I


Homologous Compounds ZnnIn2O3+n (n = 4, 5 ... - ACS Publications

Recommend Documents