A family of solar light responsive photocatalysts obtained using Zn2+ Me3+ (Me = Al/Ga) LDHs doped with Ga2O3 and In2O3 and their derived mixed oxides: a case study of phenol/4-nitrophenol decomposition

Gabriela Carja a, Elena Florentina Grosu a, Mihaela Mureseanu b and Doina Lutic *c
aFaculty of Chemical Engineering and Environmental Protection, Technical University of Iaşi, 71A Bd. D. Mangeron, Iaşi, Romania. E-mail: carja@uaic.ro; elena_grosu89@yahoo.com
bFaculty of Chemistry, University of Craiova, 107 I Calea Bucureşti, 200478, Craiova, Romania. E-mail: mihaela_mure@yahoo.com
cFaculty of Chemistry, Alexandru Ioan Cuza University of Iaşi, 11 Bd. Carol I, 700506 Iaşi, Romania. E-mail: doilub@yahoo.com

Received 12th July 2017 , Accepted 14th August 2017

First published on 16th August 2017

Photocatalytic decomposition of dangerous organic contaminants using irradiation with sunlight is of great importance for environmental remediation and human health. In this paper Zn2+Me3+ (Me = Al/Ga) layered double hydroxides (LDHs) doped with gallium oxide (Ga2O3) and indium oxide (In2O3) and their derived mixtures of mixed oxides (MMOs) as novel solar light driven photocatalysts for degrading a mixture of phenol/4-nitrophenol (Ph–4NPh) are presented. The Ga2O3/ZnMeLDHs and In2O3/ZnMeLDHs were fabricated by exploiting the LDH capability to rebuild its structure, after being destroyed by calcination at 550 °C, in aqueous solutions of gallium sulfate [Ga2(SO4)3] and indium acetate [In(C2H3O2)3], respectively, while the corresponding MMOs were obtained after the calcination at 750 °C. The effects on the structure, surface characteristics and photo-absorption properties of the catalysts were assessed using X-ray diffraction, energy dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy, Fourier-transform infrared spectroscopy and ultraviolet-visible analyses. The findings demonstrate that both Ph and 4NPh are degraded by both the reconstructed LDHs and their derived MMOs in such a way that an increased efficiency is obtained for In/Zn3Ga_750 and Ga/Zn3Ga_750 which degraded almost 90% of the 4NPh. In comparison, for In/Zn3Ga and Ga/Zn3Ga catalysts, the degradation efficiency reached only 69% and 47%, respectively. Furthermore, in the family of ZnAlLDH derived catalysts, In/Zn3Al_750 is able to decompose 98% of 4-NPh and 77% of Ph, after 240 min of irradiation, whereas the efficiency of Ga/Zn3Al_750 reached only 57% for 4-NPh and 39% for Ph. The as-prepared LDHs, Zn3Al and Zn3Ga, are almost inactive for degradation of a Ph–4NPh mixture under solar light.

1. Introduction

With continuous expansion of industrial activity, environmental remediation is a worldwide concern. Because the amount of wastes from chemical processes continues to increase, photocatalysis based on semiconductors is an important and sustainable solution to remove dangerous contaminants.1 Several contaminants are of particular interest, such as phenol (Ph)/nitro-phenols (NPhs) which are commonly found in industrial wastewaters. In particular, the removal of Ph/NPh mixtures presents serious challenges because these molecules do not undergo direct photolysis by sunlight, and are known human carcinogens, thus, there is considerable concern about their effects on human health even at very low concentrations.2,3 What is accepted is that the participation in a subsequent redox reaction of photo-excited electrons and holes is the main process in photocatalysis.4 To achieve high photochemical conversion, it is necessary to maximize the high separation of the charge carriers.5 Recently, advances in photocatalysis have shown that in porous semiconductor matrices doped with semiconducting metal oxides, the recombination reaction is reduced and the photoresponse in the visible light region is extended. As a consequence, the photoresponsive performance of the matrix/metal oxide heterostructures is greater those of each component singly.6 Among the porous semiconducting matrices, layered double hydroxides (LDHs) are a class of heterobimetallic systems described by the general formula [MeII1−xMeIIIx(OH)2]x+·Anx/n·mH2O, where MeII and MeIII are cations in the layers (e.g., magnesium (Mg2+), zinc (Zn2+), nickel (Ni2+), copper (Cu2+), aluminum (Al3+), ferric iron (Fe3+) and so on) and An are interlayer anions. The cations are linked to each other by oxygen atoms, but the oxo-bridges, type Me2+–O–Me3+, may suppress the electron–hole recombination.7 Furthermore, LDHs have a self-repairing property to reconstruct their layered structure, called the LDH “structural memory effect”. This means that the LDH structure might be able to be reconstructed, after being destroyed by calcination, when the derived mixtures of mixed oxides (MMOs) are introduced into an aqueous solution containing anions.8

In comparison with the parent LDHs, MMOs obtained by the calcination of the reconstructed LDHs have been much less explored for use as semiconductor photocatalysts. MMOs derived from LDH calcination can provide multifunctional properties including improved visible light absorption, higher carrier conductivity and efficient charge transfer, because of the existence of heterojunctions at the interface.9 The photoresponse properties of the Zn–Me3+-LDHs have been found to be useful for developing solar energy conversion systems and the concept of the “doped semiconductor based on the LDHs” was introduced recently by Gomes-Silva et al.,10 who prepared a series of hydrotalcite zinc oxides and then studied their activity for visible light photocatalytic oxygen (O2) generation. Furthermore, a Zn–chromium (Cr)LDH doped with titanium dioxide was reported by Gunjakar et al.11 as fairly active for visible light induced O2 generation from water, whereas Chen et al.12 fabricated an iron(II,III) oxide (Fe3O4)/Zn–Cr LDH composite using a hydrothermal method and reported the photocatalytic efficiency of Fe3O4/Zn–Cr LDH for methylene blue dye removal. Recent results have demonstrated that the assemblies of the porous matrices of ZnMe-LDH (Me = tin (Sn4+);13 Me = gallium (Ga3+);14 Me = Fe3+;15 Me = Al3+ (ref. 16 and 17)) with nanoparticles (NPs) of metals and/or metal oxides reveals synergistic semiconducting properties and their photocatalytic response under ultraviolet (UV) or visible (vis) irradiation that are better than those of each component singly.

To date, despite the significant interest in the doped semiconductor based on LDH photocatalysis, maximizing its efficiency under irradiation with solar light still remains a challenge. Because of the attractive structural features and synthetic strategies of the doped LDH semiconductors with specific metals and/or metal oxides, the main strategy in this research was to construct an accessible structure which was able to efficiently degrade Ph and NPh under solar light irradiation. In, this research simple self-assembly routes were investigated at room temperature to obtain novel gallium oxide (Ga2O3) and indium oxide (In2O3) on Zn3Me3+ (Me3+ = Ga or Al) LDHs, as novel photocatalysts for degrading a Ph/NPh mixture from an aqueous solution, under solar irradiation. Furthermore, the photocatalytic activity of the semiconducting mixtures of mixed oxides obtained after calcination of the above-mentioned solids at 750 °C were evaluated, and the results are presented in this paper. Compelling evidence is provided that shows that doping the surface of the parent LDHs with Ga2O3 or In2O3 was crucial for obtaining photocatalysts with good efficiency for removing a Ph/4-nitrophenol (4NPh) mixture from an aqueous solutions under solar irradiation.

2. Experimental

2.1. Catalyst synthesis

The LDH samples, denoted as Zn3Ga and Zn3Al, with a Zn2+/Me3+ ratio of 3 were prepared using a typical co-precipitation method. This involves the slow addition of mixed solutions of zinc nitrate hexahydrate/gallium nitrate octadecahydrate [Zn(NO3)2·6H2O/Ga(NO3)3·18 H2O] and zinc nitrate/aluminum nitrate [Zn(NO3)2·6H2O/Al(NO3)3·9H2O] (1 M in total) into a sodium carbonate/sodium hydroxide (Na2CO3/NaOH) solution, under vigorous stirring. During the synthesis, the pH of the reaction medium was kept constant at a value of 8.5 ± 0.2, by manipulating the flows of the metallic salts and Na2CO3/NaOH solutions. The resulting slurry was aged for about 24 h at room temperature, recuperated using filtration, washed several times with distilled water and then dried at room temperature under vacuum at 80 °C for several hours. The samples were calcined at 550 °C for 8 h, and were then denoted as Zn3Ga_MO and Zn3Al_MO, and were further introduced in to 0.1 M aqueous solutions of gallium sulfate octadecahydrate [Ga2(SO4)3·18 H2O, 99.9%, Sigma-Aldrich] and indium acetate [In(C2H3O2)3, 99.9%, Sigma-Aldrich], respectively, for 7 h at room temperature. When the previously discussed calcined clays were introduced into gallium (Ga) or indium (In) aqueous salt solutions, the reconstruction of the original layered structure, the formation of Ga or In derived species and their self-organization on the layered sheets were realized in a single synthetic step and the samples obtained were denoted as Ga/Zn3Ga and In/Zn3Ga, and Ga/Zn3Al and In/Zn3Al. The reconstructed samples were finally calcined at 750 °C and denoted as Ga/Zn3Ga_750, In/Zn3Ga_750, Ga/Zn3Al_750 and In/Zn3Al_750.

2.2. Characterization techniques

Structural characteristics, crystallinity and purity information were determined using X-ray diffraction (XRD) with a Shimadzu XRD-6100 diffractometer with monochromatic light (λ = 0.1541 nm), operating at 40 kV and 30 mA over a 2θ range from 5° to 80°. The structure constants for the rhombohedral symmetry were calculated using the relationship: a = 2d[110] and c = d[003], where the indices designate the orientation of the two characteristic planes from the XRD diffraction pattern. Transmission electron microscopy (TEM) imaging was performed on a Hitachi H-900 transmission electron microscope operating at an accelerating voltage of 200 kV, coupled with an energy dispersive X-ray (EDX) spectrometer. X-ray photoelectron spectroscopy (XPS) spectra were recorded using a PerkinElmer Model 5500-MT spectrometer equipped with Mg Kα radiation (1253.6 eV), operating at 15 kV and 20 mA and the binding energies (BE) were corrected by referencing the C1s peak to 284.8 eV. UV-vis absorption spectra were recorded on a Jasco V-550 spectrophotometer with an integration sphere. Fourier transform infrared (FTIR) spectra were collected on a PerkinElmer Spectrum 100 spectrophotometer in the wavenumber range 450–4000 cm−1, at a resolution of 4 cm−1, using potassium bromide pellets.

2.3. Photocatalytic tests

The photocatalytic activity was tested for the photodegradation of 4-NPh and a mixture of Ph + 4-NPh) in aqueous solutions. For this, appropriate amounts of catalyst powder were dispersed in the pollutant solution with an initial concentration of 0.025 g L−1 for each pollutant, while maintaining the solid/liquid ratio at 1 g L−1. Before starting the irradiation, the mixtures were stirred in the dark to realize the adsorption–desorption equilibrium between the catalytic surface and the molecules of the phenol compounds. The mixtures were stirred in dark until no modification of the UV-vis absorption spectra was observed. Afterwards, the solutions were irradiated in an Unnasol US800 solar simulator, power consumption of 180 W, equipped with UV and visible light block filters. The photocatalytic activity of all catalysts was tested under simulated solar light irradiation. The reactions were carried out in 200 mL reactor with water cycling to avoid evaporation and to keep the system at a constant volume and temperature. The photodegradation profiles of the tested organic compounds were monitored by following the characteristics of the UV-vis absorption spectra and by measuring the total organic carbon (TOC) in the solution. The measurements on the pollutant solutions at the end of the catalytic runs, were performed using a Shimadzu TOC analyzer. The decrease in the concentration of the phenolic substrates and the formation of intermediates was estimated using high-performance liquid chromatography (HPLC) with a C18 column and a Shimadzu SPD-20A UV-detector. Benzoquinone, catechol, muconic acid and resorcinol standards were used to identify the possible intermediate products obtained.

3. Results and discussion

3.1. XRD, EDX, FTIR and XPS analyses

The powder XRD method was used to investigate the structure type of the samples obtained after the doping procedure and to evaluate at a qualitative level the degree of crystallinity of the samples. The XRD patterns of Zn3Al, Zn3Ga and of the reconstructed Ga/Zn3Al, Ga/Zn3Ga, In/Zn3Al and InZn3Ga LDHs are displayed in Fig. 1.
image file: c7cy01404j-f1.tif
Fig. 1 XRD patterns of the initial reconstructed LDHs.

For the “as synthesized” LDHs the XRD profiles indicate a regular LDH structure defined by the diffraction peaks common at 2θ = 11.8°, 23.6°, 34.1°, 34.8°, 37.5°, 39.4°, 44.2°, 47.0°, 53.2°, and 56.6° assigned to [003], [006], [101], [009], [104], [015], [107], [018], [100], and [1/2 11] (=[0111]) diffractions planes, respectively, for the regular layered structure of the LDHs. Furthermore, after the reconstruction in Ga2(SO4)3 and In(C2H3O2)3 aqueous solutions, the layered LDH structure is fully recovered as demonstrated by the sharp and symmetric basal reflections of the [00[small script l], [small script l] = 3, 6, 9] planes although broader and much less intense reflections were observed for the nonbasal [01[small script l], [small script l] = 2, 5, 8] planes which indicates the lower crystallinity and/or reduced particle size of the crystallites for the hybrid LDHs.18–20 No peaks characteristic of Ga or In phases can be seen, however, it is possible that these are organized as small NPs that were not detected by the XRD analyses.13,14

Furthermore, after reconstruction, the [003] peak is much broader where it is shifted to lower 2θ values, with reflection from 11.67° for Zn3Al and 11.7° for Zn3Ga to 7.1° for In/Zn3Al and 7.03° for In/Zn3Ga, whereas the values were 9.85° and 9.92° for Ga/Zn3Ga and Ga/ZnAl, respectively. This indicates that the reconstruction in indium triacetate [In(OAc)3] and Ga2(SO4)3 aqueous solutions gives an expanded LDH structure, in which the intercalation of acetate and sulfate anions into the interlayers is clearly seen by virtue of the increase in the interlayer distance (see Table 1).

Table 1 Diffraction angles for the [003] and [110] planes and the calculated structure constants for the as-synthesized LDHs
Sample 2θ [003], ° d [003], Å 2θ [110], ° d [110], Å a, Å c, Å
Zn3Ga 11.7 7.56 60.34 1.53 3.07 22.67
Ga/Zn3Ga 9.85 8.97 60.26 1.53 3.07 26.92
In/Zn3Ga 7.10 12.44 60.90 1.52 3.04 37.32
Zn3Al 11.62 7.61 60.12 1.54 3.08 22.83
Ga/Zn3Al 9.92 8.91 60.24 1.54 3.07 26.73
In/Zn3Al 7.03 12.56 60.40 1.53 3.06 37.69

The structural reconstruction after calcination comprises the layered stacking reformation at the expense of cations from the mixed oxide and anions from the aqueous medium (sulfate ions from the Ga salt and acetate ions from In salt). The LDH structural parameters “a” and “c”, corresponding to a rhombohedral arrangement, are listed in Table 1.

The unit cell parameter “a”, which gives information about the cation–cation distance in the brucite-like sheets, remains unchanged when Ga3+ is introduced into the LDH layer, even though Ga3+ is bulkier than Al3+ (ionic radius 0.62 Å versus 0.53 Å). The unchanged value of the “a” parameter when Ga3+ replaces Al3+ in the LDH layers is related to the stacking disorders arising out of the random intergrowth of different polytypes during the synthesis processes and this was previously reported in the literature.21–23

The value of “a” is almost the same for all the samples, confirming that the reconstruction of the ordered structure and the initial order of the brucite-like cationic layer was achieved after the reconstruction.

The interlayer space parameter “c” was highly modified, and this varied according to the nature of the anions of the reconstruction solutions. The values obtained agree with previously reported results for samples containing SO42− and/or CH3COO in the interlayers.24–27

Upon calcination at 750 °C, the XRD pattern (Fig. 2) becomes quite complex. For the Zn3Ga_750 and Ga/Zn3Ga_750 samples, the main component is the zinc gallate (ZnGa2O4) spinel, as shown by the strongest peaks at 30.4°, 35.7°, 43.4°, 54°, 57.4° and 63°, corresponding to the [220], [311], [400], [422], [511] and [440] planes.28–30 Some zinc oxide (ZnO) is also present, as shown by the maxima at 31.8°, 34.3° and 36.4°. Calcination of the In doped samples gives rise to less crystalline solids, especially for the preparation using Zn3Al_MO as a precursor (see Fig. 3). The ZnGa2O4 phase was also formed, together with the ZnO phase, whereas the In2O3 is confirmed as an individualized cubic phase by the maxima at 21.6°, 33.5° and 51°.20

image file: c7cy01404j-f2.tif
Fig. 2 XRD patterns of Zn3Ga_750 and Ga/ZnGa_750.

image file: c7cy01404j-f3.tif
Fig. 3 XRD patterns of In/Zn3Ga_750 and In/Zn3Al_750.

The EDX results gave information about the composition of the surface of the samples. In the EDX spectra, the Lα transitions of Ga and Zn have close values of energy 1.012 eV (Zn) and 1.098 eV (Ga),31 therefore the corresponding peaks are not well resolved (Fig. 4). The surface compositions of some of the samples are given in Table 2.

image file: c7cy01404j-f4.tif
Fig. 4 The EDX spectra of the selected catalysts using (a) Zn3Ga, (b) Zn3Al as precursors.
Table 2 Quantification of the composition of the selected samples using EDX spectra (weight%)
Sample Zn Ga Al In S
Zn3Ga 68 32
Zn3Ga_750 77 23
Ga/Zn3Ga 47 44 9
Ga/Zn3Ga_750 57 43
In/Zn3Ga 61 35 4
Zn3Al 66 34
Ga/Zn3Al 5 57 31 7
Ga/Zn3Al_750 52 11 37
In/Zn3Al 64 32 4
In/Zn3Al_750 61 34 5

In2O3 was detected as an important phase (Fig. 4b) in the In/Zn3Ga samples, despite it not showing a high concentration in the element mapping results (Table 2). This is an indication that In2O3 had grown as a distinctive phase on the surface of the Zn3Al-derived samples. This result is in agreement with the In2O3 cubic phase detected using XRD. For the samples treated with Ga2(SO4)3 solution the calcination, after the reconstruction process, led to a slight enrichment of the surface in gallium so, probably, a gallium rich ZnGa2O4 phase is formed at the surface.34,35

The FTIR analysis was used to investigate the nature of the functional groups from the samples and to confirm further the insertion of the anions in the interlayer space of the reconstructed solids. The FTIR spectra of samples Zn3Ga, Ga/Zn3Ga and In/Zn3Al are presented in Fig. 5.

image file: c7cy01404j-f5.tif
Fig. 5 FTIR spectra of the selected catalysts.

According to previous results,32 a distinction can be made between the groups from the LDH structure giving signals in the infrared (IR). The broad peak around 3500 cm−1 is because of the stretching absorption of OH groups, at around 1600 cm−1, there is a peak which is because of the water adsorption, the sharp peaks between 1400–1600 cm−1 come from the stretching absorption of the CO and C[double bond, length as m-dash]O bonds in CO32− and the peaks below 1000 cm−1 correspond to the metal–OH stretching.

Further, the peaks from the IR spectrum of the Zn3Ga sample indicate the presence of a CO32− ion (1365 cm−1), and water (shoulder at 1621 cm−1). In the spectrum of Ga/Zn3Al, the strong peak at 1112 cm−1 indicates the presence of SO42−. Finally, the In/Zn3Ga sample exhibits two strong peaks at 1406 cm−1 and 1361 cm−1, which are characteristic of CH3COO, and a weak peak at 1362 cm−1 shows that small amounts of CO32− contaminating ions are present.33 Thus, the FTIR results further confirms the presence of SO42− and CH3COO after the reconstruction in the aqueous solutions of Ga2(SO4)3 and In(C2H3O2)3, respectively, as shown by the XRD analysis.

The XPS analysis was used to obtain information about the surface compositions of the samples tested, and the aim was to obtain information about the states of the Ga, Zn and In in the solids. In Fig. 6 the overall XPS spectra and the Ga 2p1/2 region (1116–1120 eV) for Ga-doped Zn3Ga-based samples, are displayed.

image file: c7cy01404j-f6.tif
Fig. 6 XPS spectra of Zn3Ga-based samples and details of the Ga 2p1/2 region (inset).

The spectra of the three samples are quite similar, indicating the same Ga and Zn bonds, even after the reconstruction and calcination procedures. The signal positions from the analyzed samples are shown in Table 3.

Table 3 XPS results of gallium characteristic BE (eV)
Sample Ga 2p1/2 Ga 2p3/2 Ga 3d3/2 O 1s
Zn3Ga 1145.2 1118.3 20.2 531.8
Ga/Zn3Ga 1145.3 1118.3 20.36 531.7
Ga/Zn3Ga_750 1145.0 1118.1 20.12 531.1
Results from the literature Oxide 1145 (ref. 31) 1117.4 ± 0.5 (ref. 32) 20.4 ± 0.5 (ref. 32) 531.96 ± 0.17 (ref. 33)
Metal 1143 (ref. 31) 18.6 ± 0.3 (ref. 32)

According to results found in the literature,37 Ga2O3 signals are situated between 1118–1120 eV for Ga 2p3/2, at 1145 eV for Ga 2p1/2 and at 20.5 for Ga 3d3/2. The characteristic O 1s signal for Ga-O binding is found at 531.96 ± 0.17 eV. Metallic Ga could be identified by the signals 2p1/2 at 1143 eV, 2p3/2 from 1117 eV and 3d 5/2 at 18.6 eV. In these samples, all the peaks are symmetric and well-centered around the maximum, indicating that Ga is present only as Ga2O3. Therefore, it can be concluded that the Zn3Ga-based solids are made up of Ga3+ ions.

Fig. 7 shows the spectrum of In/Zn3Ga (a) and the details for the characteristic peaks of Ga (b, c), In (d) and O 1s (e). The Ga 2p1/2 and Ga 3d3/2 peaks are slightly asymmetrical and can be deconvoluted into two components, as shown in Fig. 7b and c.

image file: c7cy01404j-f7.tif
Fig. 7 XPS spectra of In/Zn3Ga sample, a overall spectrum; b–e detailed regions.

In the samples In/Zn3Ga, Ga is found as Ga3+ and probably also as metal Ga or Ga suboxide (Ga–O–Ga),36 as shown by the deconvolution of Ga 2p and Ga 3d signals. In the In3d energy range (460–440 eV), the two peaks from 445.33 and 452.91 eV are symmetrical and large enough to indicate the presence of In2O3.38 The O 1s further confirms the bonds from indium and gallium oxides in the solids, through the presence of a large and asymmetrical peak between 535–525 eV, which can be deconvoluted defining two maxima at 531.96 eV from the Ga2O3 and at 530.78 from the In2O3.36

3.2. Photoabsorption properties of the materials

An important parameter used for the characterization of the semiconductor materials is the band gap value between the energies of the conduction band (CB) and of the valence band (VB). The radiation (visible or UV light) able to promote an electron from the VB to the CB should have a higher energy than that of the band gap.

The value of the band gap can be easily measured experimentally by interpreting the diffuse reflectance (DR) spectra of the material traced in the visible and UV range of the spectrum. The wavelength corresponding to the energy band gap value can be obtained by extrapolating the linear portion of the spectrum. A more accurate method to calculate the band gap is by applying the Tauc plot.39

Fig. 8 shows the UV-vis DR spectra of the Zn3Ga-derived samples and Table 4 shows the values of the band gap determined from the Tauc plots.

image file: c7cy01404j-f8.tif
Fig. 8 UV-vis DR spectra of the Zn3Ga-based samples.
Table 4 Band gap values obtained by applying the Tauc plots (eV)
Sample Band gap, eV
Zn3Ga 3.95
Ga/Zn3Ga 2.83
Ga/Zn3Ga_750 2.75
In/Zn3Ga 2.92
In/Zn3Ga_750 2.81
Zn3Al 3.57
Ga/Zn3Al 2.92
Ga/Zn3Al_750 2.84
In/Zn3Al 3.10
In/Zn3Al_750 2.74

The wavelength where the linear portion begins is strongly shifted towards the visible range of the spectrum, when Zn3Ga is doped with Ga or In. This behavior suggests that the light wavelength is able to encourage the electrons to jump over the band gap shifts from the UV range to the visible part of the spectrum. This fact is confirmed by the band gap values listed in Table 4.

The results show that the initial LDH Zn3Ga is activated only by the UV light with a wavelength lower than 313 nm, which corresponds to its band gap of 3.95 eV. The value of 3.1 eV corresponds to the limit between the UV and visible light range, 400 nm. All the modifications brought about by the reconstruction with Ga or In, yields materials with lower band gaps, which are able to be activated even by visible light. However, Wenderich and Mul proposed recently, that in semiconductor-NP composites, the changes in the absorption characteristics should not be automatically assigned to the alteration of the band-gap values, deduced from the Tauc plot, but were ascribed to a physical phenomenon, known as band-bending.39 Because of the band-bending process, the changes in the DR absorption spectra, after NPs deposition, are a consequence of the absorption characteristics of the metals or metal oxides NPs that superimposed on the initial spectrum. Because the NPs of In2O3 and Ga2O3 were directly obtained on larger LDHs, it is difficult to distinctly differentiate between the manifestations of the band-bending and/or band-gapping processes in their self-assemblies. Therefore, the values of the band gaps listed in Table 4 are rather useful for a qualitative comparison of the semiconducting properties of the catalysts studied.

3.3. Photocatalytic efficiency

The photocatalytic degradation of phenolic compounds was performed first in a series of experiments using 4-NPh solution. The measurement of the concentrations was performed using spectrophotometry, and was based on the intensity of the characteristic peak of 4-NPh at 400 nm. The changes of the spectra over time when Ga/Zn3Ga_750 was used as photocatalyst are displayed in Fig. 9.
image file: c7cy01404j-f9.tif
Fig. 9 Evolution of the 4-NPh spectrum during the photodegradation on Ga/Zn3Ga_750.

The photocatalyst action induces the strong decrease of the main peak, while the attraction of the UV region of the spectrum remains almost unchanged and the intensity of the maximum near 200 nm only weakly changes during the photodegradation. This is an indication that the organic compound is gradually fragmented and slowly mineralized.

The overall photocatalyst performance and the comparison between the Zn3Ga-based catalysts can be highlighted by representing the variation of the relative absorbance (Fig. 10a) and the intensity of the main maximum on the series of catalysts (Fig. 10b).

image file: c7cy01404j-f10.tif
Fig. 10 (a) Evolution of the relative absorbance at 400 nm over time, (b) UV-vis spectra registered after 130 min of irradiation on the Zn3Ga-based photocatalysts.

The activity of the catalysts strongly depends on their composition and the doping procedure. The Zn3Ga precursor is practically inactive, whereas the doping procedure and then the calcination cause an important increase of the degradation yields of 4-NPh at over 90% in two hours of exposure to light. The performance of the reconstructed powders, Ga/Zn3Ga and In/Zn3Ga, record an important improvement compared to the parent precursor, but they are less active than the corresponding calcined materials. Apart from the evident big difference in the absorbance values at 400 nm which is attributed to the 4-NPh molecule splitting, the mineralization of the fragments resulting from the first reactions is highlighted by the significant decrease of the peaks in the far UV region of the spectra (Fig. 10b). Also, it is worth noting that the doping with indium generated more active photocatalysts, even if the band gap values are quite similar in all the series of samples (see Table 3). It means that the presence of In has, apart from the ability to absorb the photon energy, some structural peculiarities making it more efficient in this respect.

According to data in the literature,20,41 In seems to be responsible for an important stabilization of the electron–hole pair in the In–Zn binary oxide. The top levels of the VB and the bottom levels of the CB are positioned at −7.39 eV and −4.19 for ZnO, and at −6.68 eV and −3.88 for In2O3, respectively, with respect to the absolute vacuum scale.20 When a heterostructure containing both oxides is formed, the photons absorbed induce the formation of electrons and holes pairs. The recombination of both electrons and holes is avoided by the possibility of electrons migrating from the In2O3 CB to the ZnO CB, and by the migration of holes from the ZnO VB to the In2O3 VB. Also a p–n junction can form at the interface between ZnO and In2O3, allowing both ions to diffuse into the lattice of each other.

Murat and Medvedeva40 and Peelaers et al.41 simulated the structure of In–Ga–Zn and In–Ga oxides, to determine the influence of oxygen defects in the formation energy of the framework, as well as to calculate the band alignment and the band gas values for different forms of In and Ga oxides and for the InGaZnO4 spinel. The energy values depend to a certain extent on the detailed structure of the oxide (for In2O3, the values vary between 2.78–3.04, for Ga2O3 they are between 4.57–5.0, and for InGaZnO4, 3.39 eV).

The interesting behavior of the Zn3Ga-based solids, especially the In doped samples, obtained on the 4-NPh test molecule, motivated the testing of the Zn3Al solids in the photocatalytic transformation of the 4-NPh and Ph mixture. In the measurement of the yields based on spectrophotometry it needs to be noted that both species absorb strongly at 200 nm and that in the UV-vis spectrum of phenol, there is an absorption maximum at 270 nm. The decrease of the peak at 400 nm can clearly be assigned to the 4-NPh conversion, whereas the one at 270 nm is mainly because of the Ph conversion, but 4-NPh also absorbs significantly at this wavelength. In order to make an evaluation of the contribution of each species in the absorbance value possible, the dependence of the absorbance at 400 nm (A400) against the absorbance at 270 nm (A270) for the series of samples discussed previously (Fig. 11) was represented. Except for the Zn3Ga sample, which is basically inactive, the variations were mostly linear, with a slope between 11.8–15.7% for all the samples. Therefore, a contribution of 14% of 4-NPh in the decrease of the 270 nm peak intensity and a 86% contribution because of Ph transformation was assumed to be reasonable.

image file: c7cy01404j-f11.tif
Fig. 11 Correlations between A400 nm and A270 nm on various samples.

The Zn3Al-based samples were tested under the same experimental conditions and the reaction time was prolonged up to 240 min and the results are shown in Fig. 12.

image file: c7cy01404j-f12.tif
Fig. 12 The conversion yields of 4-NPh (a) and of Ph (b) from their mixture on Zn3Al-derived photocatalysts.

The results are similar to those for the corresponding Zn3Ga series. The parent material Zn3Al is almost inactive, the activity increases for the reconstructed samples and the increase is even higher for the calcined form after the reconstruction. Indium is also a better dopant than Ga for both 4-NPh and Ph conversion and the transformation of 4-NPh is almost total, whereas the Ph conversion value reaches 80%.

The mineralization of the pollutants was also confirmed using TOC analyses. An almost 87% disappearance of the TOC could be observed after 4 h for In/Zn3Al750 and almost 60% for Ga/Zn3Al750 confirming not only the degradation of the pollutants but also their mineralization.

The identification of the intermediates that appeared in the HPLC chromatograms, see Fig. S1 (ESI), confirmed the formation of the hydroxyl radicals that interacted with the phenolic substrates of both Ph and 4-NP giving rise to an indirect ring cleavage process to produce hydroxylated phenolic intermediates as shown in Fig. 13.

image file: c7cy01404j-f13.tif
Fig. 13 The proposed scheme for the catalytic degradation/mineralization of Ph/4-NPh under solar light irradiation, for the tested catalysts.

4. Conclusions

Zn2+Me3+ (Me = Al/Ga) LDHs, with a ratio of Zn2+/Me3+ = 3/1, were prepared using a coprecipitation method and then calcined at 550 °C. The LDH structure was further examined in Ga2(SO4)3 and In(C2H3O2)3 aqueous solutions, by exploiting the manifestation of the LDH's structural memory. So Zn2+Me3+ (Me = Al/Ga) LDHs doped with Ga2O3 and In2O3 were obtained. Furthermore, the calcination at 750 °C gave rise to complex mixtures of mixed oxides. The typical LDH structure was confirmed using XRD for both initial and reconstructed LDHs. After the calcination, spinel-like structures were identified as ZnGa2O4 and InGaZnO4. FTIR results showed the presence of SO42− and CH3COO as interlayer anions for the LDHs reconstructed in the Ga2(SO4)3 and In(C2H3O2)3 solutions, respectively. The XPS analysis showed that in the solids derived from Zn3Ga and Zn3Al LDHs using Ga2(SO4)3, the surface Ga exists in the form of Ga2O3. For the catalysts reconstructed in In(C2H3O2)3 solutions, the In bonds reveal the presence of oxide, whereas part of the Ga species from the In/Zn3Ga sample shows Ga–O–Ga bonds.

The photocatalytic activity was confirmed by measuring the bad gap values, which were between 2.74–3.1 eV, which allows the activation of an electron from the VB to the CB by the visible light. The photodegradation of 4-NPh and Ph (0.025 g l−1 of each) at a catalyst dose of 1 g L−1 proved that the spinel structures were very active. The conversion yields were over 90% on both spinels, after 130 min of irradiation, for the transformation of 4-NPh on Ga/Zn3Ga_750 and In/Zn3Ga_750 samples. On the In/Zn3Al, the conversion yields reached 98% for 4-NPh and 80% for Ph.

Conflicts of interest

There are no conflicts to declare.


The authors are grateful for the financial support from the Romanian National Authority for Scientific Research and the Romanian Ministry of Research, Project Number: PN-III-P2-2.1-PED-2016-0473, acronym ELECTROPHOTO.


  1. L. F. Liotta, M. Gruttadauria, G. Di Carlo, G. Perrini and V. Librando, J. Hazard. Mater., 2009, 162, 588 CrossRef CAS PubMed .
  2. V. M. Brown, D. H. M. Jordan and B. A. Tiller, Water Res., 1967, 1, 587 CrossRef CAS .
  3. J. C. K. Bubacz, D. Dolat and W. Morawski, Pol. J. Environ. Stud., 2010, 19, 685 Search PubMed .
  4. M. Morikawa, Y. Ogura, N. Ahmed, S. Kawamura, G. Mikami, S. Okamoto and Y. Izumi, Catal. Sci. Technol., 2014, 4(6), 1644 CAS .
  5. H. G. Baldovi, F. Albarracin, P. Atienzar, B. Ferrer, M. Alvaro and H. Garcia, ChemPhysChem, 2015, 16(2), 335 CrossRef CAS PubMed .
  6. G. Carja, L. Dartu, K. Okada and E. Fortunato, Chem. Eng. J., 2013, 222, 60 CrossRef CAS .
  7. S. J. Kim, Y. Lee, D. K. Lee, J. W. Lee and J. K. Kang, J. Mater. Chem. A, 2014, 2, 4136 CAS .
  8. G. Carja, E. Husanu, C. Gherasim and H. Iovu, Appl. Catal., B, 2011, 107, 253 CrossRef CAS .
  9. G. Mikami, F. Grosu, S. Kawamura, Y. Yoshida, G. Carja and Y. Izumi, Appl. Catal., B, 2016, 199, 260 CrossRef CAS .
  10. C. Gomes-Silva, Y. Bouizi, V. Fornes and H. Garcia, J. Am. Chem. Soc., 2009, 131, 13833 CrossRef CAS PubMed .
  11. J. L. Gunjakar, T. W. Kim, H. N. Kim, I. Y. Kim and S. J. Hwang, J. Am. Chem. Soc., 2011, 133, 14998 CrossRef CAS PubMed .
  12. D. Chen, Y. Li, J. Zhang, J. Z. Zhou, Y. Guo and H. Liu, Chem. Eng. J., 2012, 185–186, 120 CrossRef CAS .
  13. E. M. Seftel, M. C. Puscasu, M. Mertens, P. Cool and G. Carja, Appl. Catal., B, 2015, 164, 251 CrossRef CAS .
  14. S. Kawamura, M. C. Puscasu, Y. Yoshida, Y. Izumi and G. Carja, Appl. Catal., A, 2015, 504, 238 CrossRef CAS .
  15. E. M. Seftel, M. Puscasu, M. Mertens, P. Cool and G. Carja, Catal. Today, 2015, 252, 7 CrossRef CAS .
  16. G. Mikami, F. Grosu, S. Kawamura, Y. Yoshida, G. Carja and Y. Izumi, Appl. Catal., B, 2016, 199, 260 CrossRef CAS .
  17. G. Carja, A. Nakajima, S. Dranca, C. Dranca and K. Okada, J. Phys. Chem. C, 2010, 114, 14722 CAS .
  18. N. Fukuda, Y. Watanabe, S. Uemura, Y. Yoshida, T. Nakamura and H. Ushijima, J. Mater. Chem. C, 2014, 13, 2448 RSC .
  19. V. B. R. Boppana, D. J. Doren and R. F. Lobo, J. Mater. Chem., 2010, 20, 9787 RSC .
  20. Z. Wang, B. Huang, Y. Dai, X. Qin, X. Zhang, P. Wang, H. Liu and J. Yu, J. Phys. Chem. C, 2009, 113, 4612 CAS .
  21. E. Angelescu, O. D. Pavel, R. Bîrjega, M. Florea and R. Zăvoianu, Appl. Catal., A, 2008, 341, 50–57 CrossRef CAS .
  22. G. S. Thomas and P. V. Kamath, Solid State Sci., 2006, 8, 1181–1186 CrossRef CAS .
  23. M. Bellotto, B. Rebours, O. Clause, J. Lynch, D. Bazin and E. Elkaïm, J. Phys. Chem., 1996, 100, 8527–8534 CrossRef CAS .
  24. E. M. Seftel, P. Cool and D. Lutic, Mater. Sci. Eng., C, 2013, 33, 5071 CrossRef CAS PubMed .
  25. P. S. Braterman, Z. P. Xu and F. Yarberry, Layered Double Hydroxides (LDHs), in Handbook of Layered Materials, ed. S. M. Auerbach, K. A. Carrado and P. K. Dutta, CRC Press, 2004, pp. 373–474 Search PubMed .
  26. A. Halajnia, S. Oustan, N. Najafi, A. R. Khataee and A. Lakzian, Appl. Clay Sci., 2013, 80–81, 305–312 CrossRef CAS .
  27. J. T. Kloprogge, D. Wharton, L. Hickey and R. L. Frost, Am. Mineral., 2002, 87, 623 CrossRef CAS .
  28. L. Zou, X. Xiang, M. Wei, F. Li and D. G. Evans, Inorg. Chem., 2008, 47, 1361 CrossRef CAS PubMed .
  29. J. F. D. Figueiredo, V. Bouquet, S. Députier, O. Merdrignac-Conanec, I. Peron, E. P. Marinho, A. G. Souza, I. M. G. Santos, M. Guilloux-Viry and I. T. Weber, Ceramica, 2013, 59, 269 CrossRef CAS .
  30. J. Prince, F. Tzompantzi, G. Mendoza-Damián, F. Hernández-Beltranc and J. S. Valentec, Appl. Catal., B, 2015, 163, 352 CrossRef CAS .
  31. Energy table for EDS analysis, available at: https://www.unamur.be/services/microscopie/sme-documents/Energy-20table-20for-20EDS-20analysis-1.pdf (accessed May 16th, 2017) .
  32. R. Valdez, D. B. Grotjahn, D. K. Smith, J. M. Quintana and A. Olivas, Int. J. Electrochem. Sci., 2015, 10, 909 CAS .
  33. http://xpssimplified.com/periodictable.php (accessed May 16th, 2017) .
  34. S. C. Ghosh, M. C. Biesinger and R. R. LaPierre, J. Appl. Phys., 2007, 101, 114322 CrossRef .
  35. I. Lopez, A. D. Utrilla, E. Nogales, B. Mendez, J. Piqueras, A. Peche, J. Ramírez-Castellanos and J. M. Gonzalez-Calbet, J. Phys. Chem. C, 2012, 116, 3935 CAS .
  36. D. Kim, P. Thissen, G. Viner, D.-W. Lee, W. Choi, Y. J. Chabal and J. B. Lee, ACS Appl. Mater. Interfaces, 2013, 5, 179 CAS .
  37. http://www.xpsfitting.com/search/label/Indium (accessed May 16th, 2017) .
  38. J. Tauc, Mater. Res. Bull., 1968, 3, 37–46 CrossRef CAS .
  39. K. Wenderich and G. Mul, Chem. Rev., 2016, 116, 14587–14619 CrossRef CAS PubMed .
  40. A. Murat and J. E. Medvedeva, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 86, 085123 CrossRef .
  41. H. Peelaers, D. Steiauf, J. B. Varley, A. Janotti and C. G. Van de Walle, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 92, 085206 CrossRef .


Electronic supplementary information (ESI) available. See DOI: 10.1039/c7cy01404j

This journal is © The Royal Society of Chemistry 2017