Materials

For this work, Mg(NO₃)₂·6H₂O, La(NO₃)₃·6H₂O, Al(NO₃)₃·9H₂O, NaNO₃, NaOH, isopropanol, p-benzoquinone, ethylene diamine tetraacetic acid disodium salt, H₂O₂, and MB of analytical purity grade were purchased from Aladdin Reagent Co., Ltd (Shanghai, China). All solutions were prepared with distilled water.

Synthesis of MgAlLa-LDHs

A series of MgAlLa-LDHs with various molar ratios of Al(III)/La(III) were synthesized by the co-precipitation method. At a molar ratio M(II)/M(III) of 3.0, specific quantities of Mg(NO₃)₂·6H₂O, Al(NO₃)₃·9H₂O, and La(NO₃)₃·6H₂O solid were dissolved in distilled water, as solution A. Then NaOH and NaNO₃ were dissolved in distilled water; this was marked as solution B. With constant stirring, solutions A and B were added dropwise into a beaker, which was filled with distilled water. In this process, the mixed solution was adjusted with NaOH solution until the pH was between 9 and 10, after which the solution was stirred continuously for a further 30 min. Finally, the slurry obtained was subsequently aged at 65°C for 24 h. The resulting precipitates were collected by centrifugal separation and washed thoroughly with distilled water. After that, the as-obtained solid was oven-dried at 60°C overnight to obtain MgAlLa-LDHs precursors with various Mg(II)/Al(III)/La(III) (Mg/Al/La) molar ratios (i.e. 3:1:0, 3:0.8:0.2, 3:0.6:0.4, 3:0.5:0.5, 3:0.2:0.8, and 3:0:1); these precursors were denoted as MgAlLa-x-L, where x stands for the La(III) molar ratio. For example, the precursor with a Mg(II)/Al(III)/La(III) molar ratio of 3:1:0 was denoted as MgAl-L, that with a Mg(II)/Al(III)/La(III) molar ratio of 3:0.5:0.5 was specifically denoted as MgAlLa-0.5-L, and that with a Mg(II)/Al(III)/La(III) molar ratio of 3:0:1 was specifically denoted as MgLa-1-L.

Synthesis of mixed-metal oxides

The MgAlLa-LDH precursor was calcined in air at 600°C for 5 h, and allowed to cool to room temperature to obtain Mg/Al/La mixed-metal oxides, which were named MgAlLa-M. The mole ratios of Mg(II)/Al(III)/La(III) were 3:1:0, 3:0.8:0.2, 3:0.6:0.4, 3:0.5:0.5, 3:0.2:0.8, and 3:0:1, respectively. In the present study, the mixed-metal oxide with a Mg(II)/Al(III)/La(III) mole ratio of 3:0.8:0.2 was denoted specifically as MgAlLa-0.2-M.

Characterizations

The morphology of the samples was examined using a S-3400N scanning electron microscope (Hitachi, Tokyo, Japan). Elemental analysis was obtained using a Model 550i, energy dispersive X-ray fluorescence (EDX) (Model 550i, IXRF System, Austin, Texas, USA). The crystal phases in the samples were identified by XRD (CuKα, λ = 0.1545 nm, beam voltage 36 kV, beam current 20 mA) (Beijing Purkinje General Instrument Co., Ltd., Beijing, China). Infrared (FTIR) spectra were recorded using the KBr disc method on a Nicolet 5700 FTIR spectrophotometer (Thermo Scientific, Waltham, Massachusetts, USA). The sample was dried at 120°C for 1 h and then the sample was mixed with KBr at a mass ratio of 1:100 for grinding and finally pressed into thin sheets for measurement. UV-Vis diffuse reflectance spectra (DRS) of the samples were obtained using an ISA-220 UV-Vis spectrophotometer (Thermo Scientific, Waltham, Massachusetts, USA). The time-dependent UV-Vis spectra of MB solutions were measured using a UV-2600 UV-vis spectrophotometer (Shimadzu, Tokyo, Japan).

Photocatalysis

The photocatalytic activities of the samples were assessed by photodegradation of MB solution under visible light irradiation from a 150 W halogen lamp, which was preheated for 30 min. Photocatalyst (0.05 g) was mixed with 25 mL of MB (5 mg/L) aqueous solution, to which was added 0.5 mL of H₂O₂ (30% w/w). For each run, 0.05 g of catalyst was added to a 25 mL MB solution and stirred vigorously for 30 min in the dark to establish an adsorption/desorption equilibrium. Subsequently, the solution was stirred under visible light irradiation. The catalyst was kept suspended by a magnetic stirrer. Samples for analysis were extracted using a pipette every 10 min and centrifuged immediately. After that, the filtrates were tested by measuring the absorbance at 664 nm using 752 UV-Vis spectroscopy to determine the concentration of MB. The effects of light and the La(III) content on the catalytic degradation of MB were investigated. Blank experiments, i.e. without catalyst or in the dark, were carried out under the same conditions.

X-ray Diffraction

The powder XRD patterns of as-synthesized MgAlLa-L precursors with variable Mg(II)/Al(III)/La(III) molar ratios (Fig. 1) revealed that, as for the MgAl-L sample (Fig. 1a), the diffraction peaks at 11.7, 22.5, 34.5, 60.2, and 62.4^o2θ could be assigned to the (003), (006), (009), (110), and (113) reflections of LDH, respectively, characteristic of a layered structure (Ferreira et al. Reference Ferreira, Alves, Gouveia, Filho, Paiva and Filho2004; Carvalho et al. Reference Carvalho, Ferreira, Filho, Ferreira, Soares and Oliveira2015). They were consistent with the standard card JCPDS 40-0216. Almost identical, typical structures were observed in the XRD patterns of the MgAlLa-0.2-L, MgAlLa-0.4-L, and MgAlLa-0.5-L precursors. Compared with MgAl-L, they had the characteristics of a layered structure. Besides, the diffraction peaks at 15.5, 27.3, 27.9, 39.4, and 48.6^o2θ could be indexed to La(OH)₃ (JCPDS 36-1481). In other words, when the initial La(III) molar ratio was in the range 0.2 to 0.5, La(OH)₃ and LDHs coexisted in MgAlLa-L.

Fig. 1 XRD patterns of MgAlLa-L precursors: (a) MgAl-L, (b) MgAlLa-0.2-L, (c) MgAlLa-0.4-L, (d) MgAlLa-0.5-L, (e) MgAlLa-0.8-L, and (f) MgLa-1-L.

However, the characteristic diffraction peak of hydrotalcite did not appear in MgAlLa-0.8-L and MgLa-1-L, indicating that the materials synthesized were not hydrotalcite when the Al(III)/La(III) molar ratio was 0.2:0.8 or when Al(III) was completely replaced by La(III). Besides the diffraction peak of La(OH)₃, the reflections at 18.4, 38.0, 50.8, and 58.6°2θ could be indexed to Mg(OH)₂ (JCPDS 44-1482), revealing the co-existence of La(OH)₃ and Mg(OH)₂ in MgLa-L.

The XRD patterns of MgAlLa-x-L calcined at 600°C (Fig. 2) revealed that the reflections corresponding to LDHs had disappeared completely and new reflections had appeared after calcination of MgAlLa-LDHs. The reflections at 42.8 and 62.3°2θ could be indexed to MgO (JCPDS No. 45-0946), whereas those at 31.3, 36.8, 44.8, 59.4, and 65.2°2θ could be indexed to MgAl₂O₄ (JCPDS No.21-1152). The MgAl₂O₄ spinel formed during the calcination process, similar to the work of Ahmed et al. (Reference Ahmed, Talib, Hussein and Zakaria2012). With an increased La(III) content, the reflections at 26.1, 29.9, 39.5, 46.0, 52.1, and 55.4°2θ could be indexed to La₂O₃ (JCPDS No. 05-0602), whereas those at 11.4, 18.3, 27.6, 29.3, 33.4, 45.6, and 47.1°2θ could be indexed to La₁₀Al₄O₂₁ (JCPDS No.39-0009). The La₁₀Al₄O₂₁ spinel occurred at MgAlLa-0.4-M. These results may be ascribed to the homogeneous distribution of metal ions within MgAlLa-LDHs, which facilitated the formation of well dispersed and crystalline mixed-metal oxides through a topotactic process during calcination (Zhao et al. Reference Zhao, Zhang, Xu, Evans and Duan2010). No signal corresponding to an Al₂O₃ phase was detected, implying that the Al₂O₃ was amorphous. This was because, following the collapse of the MgAlLa-LDHs layered structure upon calcination, generation of MgO (La₂O₃) and Al₂O₃ subsequently gave rise to the formation of spinel, MgAl₂O₄ (La₁₀Al₄O₂₁) (Zhao et al. Reference Zhao, Zhang, Xu, Evans and Duan2010).

Fig. 2 XRD patterns of MgAlLa-M: (a) MgAl-M, (b) MgAlLa-0.2-M, (c) MgAlLa-0.4-M, (d) MgAlLa-0.5-M, (e) MgAlLa-0.8-M, and (f) MgLa-1-M.

FTIR spectroscopy analysis

The FTIR spectra of as-synthesized LDH samples (Fig. 3) were very similar to those reported generally for hydrotalcite-like compounds. The absorption peak between 3470 and ~3640 cm^–1 in the IR spectra was ascribed to the stretching vibration of the OH group with hydrogen bonding from interlayer water molecules (Yan et al. Reference Yan, Yang, Shan, Yan, Wei, Yu, Yu and Du2015). Similarly, the band observed at 1630 cm^-1 was attributed to the bending mode of crystalline water. An absorption peak at 1368 cm^–1 was attributed to the telescopic vibration peak of NO₃ ^- (Scavetta et al. Reference Scavetta, Ballarin, Corticelli, Gualandi, Tonelli, Prevot, Forano and Mousty2012). The bands at ~846 and 602 cm⁻¹ arose from the metal-oxygen bonds (M-O, M-O-M, and M-OH) vibration in the LDHs. In the case of MgAlLa-0.4-L, MgAlLa-0.5-L, MgAlLa-0.8-L, and MgLa-1-L, a new band was observed at 3603 cm⁻¹, which corresponded to non-bonded –OH group vibrations for La(OH)₃. An absorption peak at 3696 cm⁻¹ corresponds to non-bonded –OH group vibrations for Mg(OH)₂. Compared with uncalcined samples, the differences in infrared spectra might be due to the structural difference in compounds formed with different amounts of La. The above results were consistent with XRD analysis.

Fig. 3 FTIR spectra of MgAlLa-L precursors: (a) MgAl-L, (b) MgAlLa-0.2-L, (c) MgAlLa-0.4-L, (d) MgAlLa-0.5-L, (e) MgAlLa-0.8-L, and (f) MgLa-1-L.

Morphology

The morphology of the MgAlLa-L with various molar ratios before calcination was characterized using SEM (Fig. 4). The MgAl-L synthesized showed many wrinkles on the surface (Fig. 4A1). MgAlLa-0.2-L and MgAlLa-0.4-L also displayed wrinkles on the surface (Fig. 4A2 and A3). Some fragments of MgAlLa-0.4-L also appeared. The morphology of MgAlLa-0.5-L (Fig. 4A4) changed significantly, and many irregular sheets aggregated. Needle-like and granular aggregates appeared for MgAlLa-0.8-L (Fig. 4A5) and MgLa-L (Fig. 4A6). With the increase in La(III) content in MgAlLa-L, the morphologies of the synthesized samples varied greatly.

Fig. 4 SEM images of MgAl-L (A1), MgAlLa-0.2-L (A2), MgAlLa-0.4-L (A3), MgAlLa-0.5-L (A4), MgAlLa-0.8-L (A5), and MgLa-L (A6).

The morphology of the MgAlLa-M with various molar ratios (Fig. 5) showed that MgAl-M after calcination was made up of irregular, multilateral sheets and some particles. In contrast, the multilateral fragments on the surface of MgAlLa-0.2-M and MgAlLa-0.4-M were more obvious (Fig. 5B2 and B3). The morphology of MgAlLa-0.5-M collapsed from a specific structure and became regular, agglomerated platelet-like particles after calcination at 600°C; the platelets had a side length of 0.08 μm (Fig. 5B4). Fine particles appeared for MgAlLa-0.8-M (Fig. 5B5) and MgLa-M (Fig. 5B6). The change in morphology seemed to be the result of MgAlLa-M having a large specific surface area. The EDX pattern (Fig. 5B7) further confirmed the presence of Mg, Al, La, and O in MgAlLa-0.5-M (Fig. 4F); the Mg(II)/Al(III)/La(III) molar ratio was close to 3:0.5:0.5.

Fig. 5 SEM images of MgAl-M (B1), MgAlLa-0.2-M (B2), MgAlLa-0.4-M (B3), MgAlLa-0.5-M (B4), MgAlLa-0.8-M (B5), and MgLa-M (B6), along with EDX measurement of MgAlLa-0.5-M (B7).

Optical properties

The UV-Vis absorption spectra of the as-prepared samples (Fig. 6A and C) revealed that the MgAlLa-M had a better absorbance stability than MgAlLa-L, and the absorption edge was extended to ~800 nm covering the full visible light spectrum, compared with the absorption edges of MgAlLa-L. The band-gap energies of the samples before and after calcination were calculated using the onset of the UV–Vis spectra of the absorption values (Eq. 1) (Hussain et al., Reference Hussain, Russo and Saracco2011)

(1) ${\left(ahv\right)}^2=K\left(hv-\mathrm{Eg}\right)$

Fig. 6 UV-Vis absorption spectra of the MgAlLa-L (A) and MgAlLa-M (C), and the plotting of (ahν)² vs. hν based on the direct transition (B, D). MgAlLa-L (A, B): (a) MgAl-L, (b) MgAlLa-0.2-L, (c) MgAlLa-0.4-L, (d) MgAlLa-0.5-L, (e) MgAlLa-0.8-L, and (f) MgLa-1-L. MgAlLa-M (B, D): (a) MgAl-M, (b) MgAlLa-0.2-M, (c) MgAlLa-0.4-M, (d) MgAlLa-0.5-M, (e) MgAlLa-0.8-M, and (f) MgLa-1-M.

Where a is the absorption coefficient, ν represents the light frequency, Eg is the band gap energy, K is a proportionality constant, and h is Planck’s constant. The normalized graphs of (ahν)² for the photon energy of the MgAlLa-L and MgAlLa-M samples (Fig. 6B and D) showed that all MgAlLa-M samples exhibited relatively lower band gap energy and were expected to show better photocatalytic activity compared with MgAlLa-L. The smaller band-gap might be attributed to the composite oxides formed by calcination, and the band-gap values of MgAlLa-M samples (3.11–3.35 eV) were smaller than those of MgAl-M (3.36 eV). The MgAlLa-M samples were thus expected to show better photocatalytic activity than MgAl-M under visible light. This result could be attributed to the coupling interaction among the MgO, La₂O₃, MgAl₂O₄, and La₁₀Al₄O₂₁ phases, the optimizing content of MgO, La₂O₃, and the form of heterojunction structure, which led to the enhancement of utilization of light and photocatalytic activity (Zhang et al., Reference Zhang, Dai, Zhang, Liu and Yan2016).

Influence of light

The influence of light on the photocatalytic activity of MgAlLa-0.5-M was determined by degradation of MB (Fig. 7). The residual amount of MB was almost unchanged with no light, which indicated that the adsorption of MB by MgAlLa-0.5-M was weak. Another test without light demonstrated that MB could not be degraded under MgAlLa-0.5-M/H₂O₂, because the residual level of MB was 91.65% after reaction for 60 min. When the MB solutions were light irradiated for 60 min, however, the residual MB was only 0.11% under MgAlLa-0.5-M/H₂O₂. Meanwhile, the color of the MB solution was close to colorless, which led to the conclusion that photocatalytic degradation of MB in the MgAlLa-0.5-M and H₂O₂ system is accelerated by light irradiation.

Fig. 7 Effect of light on photocatalytic degradation of MB with MgAlLa-0.5-M.

Effect of catalyst

Under illumination only, the residual amount of MB was 84% after 1 h, and the rate of degradation was very slow (Fig. 8). When 0.5 mL H₂O₂ was added, the degradation rate of MB increased, and the residual amount was 70.1% after 1 h. The photocatalytic degradation efficiency of MB with 50 mg of MgAl-M was almost the same as that by MB only. The degradation efficiency of MgAlLa-0.5-M was more obvious, and the residual MB was 36.23% after 60 min. The degradation rate of MB increased significantly when the catalyst and H₂O₂ acted simultaneously. The results above showed that the catalytic degradation of MB by a single catalyst was not ideal, only when MgAlLa-M and H₂O₂ were used in concert was the best level of photocatalytic degradation achieved; photocatalytic degradation efficiency of MB dye with MgAlLa-0.5-M under visible light irradiation for 1 h was 99.89%, which exceeded the binary MgAl-M (84.06%) under the same conditions. This might be attributed to the introduction of La(III) to improve the photocatalytic activity of mixed-metal oxides.

Fig. 8 Effect of catalyst on photocatalytic degradation of MB.

Effect of MgAlLa-L/H₂O₂

The photocatalytic effect of MgAlLa-L with a molar ratio of Mg(II)/Al(III)/La(III) (Fig. 9) showed that MgAlLa-L possessing different molar ratios exhibited a strong catalytic degradation ability for MB, and that the La(III) content affected the photocatalytic degradation rate. The photocatalytic effect of MgAlLa-L/H₂O₂ increased when the La(III) content decreased. When Al(III) was completely replaced by La(III), however, the photocatalytic activity was best. According to XRD analysis, MgLa-1-L does not have a hydrotalcite structure, so the basic substances such as Mg(OH)₂ and La(OH)₃ in the MgLa-1-L possibly played a photocatalytic role.

Fig. 9 Photocatalytic degradation of MB with MgAlLa-L/H₂O₂.

Effect of MgAlLa-M/H₂O₂

The photocatalytic effect of La₂O₃ on MB (Fig. 10) was not obvious; after 60 min the residual MB was >60% (contrast experiment), which indicated that La₂O₃ did not play a major role in the photocatalytic process. When Mg-Al hydrotalcite was calcined to form mixed-metal oxides, the residual MB was 15.9% after photocatalytic degradation for 60 min. While Mg(II), Al(III), and La(III) coexisted in the composite oxides (the molar ratio of La(III) being from 0.2 to 0.8), the MB residue decreased and the catalytic activity of MgAlLa-M increased. When the molar ratio of Mg(II):Al(III):La(III) was 3:0.5:0.5, the photocatalytic activity of MgAlLa-0.5-M was better than that of MgAl-M and the residual MB was only 0.11% after 60 min. The photocatalytic MB effect of the mixed-metal oxides that formed when Al(III) was replaced completely by La(III) was most obvious, and the degradation was completed within 30 min. Irradiation at ~150 W from a halogen lamp was less bright than with a xenon lamp. In addition, the reaction time was 30–60 min, which shortened the illumination time and improved the photocatalytic efficiency.

Fig. 10 Photocatalytic degradation of MB with MgAlLa-M/H₂O₂.

The catalytic activity of MgAlLa-M for MB was obviously better than that of MgAlLa-L (Fig. 11). This was because the mixed-metal oxides contain MgO, La₂O₃, MgAl₂O₄, and La₁₀Al₄O₂₁, which had a relatively smaller band gap than MgAlLa-L, thereby improving the visible light absorption of the mixed-metal oxides. On the other hand, the effects of photocatalytic degradation of MB by MgAlLa-L or MgAlLa-M was irregular as the La content increased, which might be related to the composition and morphology of the composite oxides.

Fig. 11 Comparison of photocatalytic activity of MgAlLa-L and MgAlLa-M with various La(III) contents.

The interface heterostructures in MgAlLa-M were different. Heterojunctions with good electron (e ^–) and hole (h ⁺) transport channels could be formed at the interface of MgAlLa-M with appropriate molar ratios of Mg(II), Al(III), and La(III). Moreover, the key factor of high photocatalytic activity was the effective separation of photogenerated e ^– and h ⁺ pairs. Consequently, when the molar ratio of Mg(II), Al(III), and La(III) was appropriate in MgAlLa-M, MgAlLa-M displayed excellent photocatalytic performance. The photocatalytic activity of MgAlLa-0.5-M on MB was slightly better than that of MgLa-1-M, and the photocatalytic performance of MgAlLa-M was better than that of MgAl-M, inferring that hydrotalcite-based mixed-metal oxides prepared with MgAl-LDHs as precursors, which added La(III), improves the degradation ability of the catalyst toward MB. Interestingly, in contrast with MgLa-0.5-M, MgLa-1-M had a higher ratio of La(III) but showed no improvement in photocatalytic activity, suggesting that the amount of La(III) was the crucial factor.

UV-Vis absorption spectra of MB

The UV-Vis absorption spectra (Fig. 12) showed a strong absorption peak at 664 nm which belonged mainly to the conjugated system of aromatic rings in the MB structure. With the prolongation of photocatalytic reaction time, the maximum absorption peak of MB solution weakened gradually. This was due to the destruction of display groups, which consist of chromophore and cochromophore in MB during the photocatalytic reaction. When photocatalysis lasted for 60 min, the absorption peak was no longer obvious, which was attributed to the interruption of the unsaturated conjugate bond of the chromophore group during the photocatalytic degradation of MB.

Fig. 12 The UV-Vis absorption spectra of MB with MgAlLa-0.5-M/H₂O₂.

Photocatalytic mechanism

To further elucidate the reaction mechanism, the reactive species were determined in the presence of isopropanol (IPA), p-benzoquinone (BQN), and disodium ethylenediamine tetraacetic acid (EDTA) as scavengers of hydroxyl radicals (•OH), superoxide radicals (•O₂ ^–), and photoexcited holes (h⁺), respectively (Yan et al. Reference Yan, Zhu, Li, Li, Liu and Wang2013). So, in the reaction system, the mole ratio of MB to the capture agent was 1:200.

When the MB solutions (Fig. 13) were irradiated for 60 min, the residual amounts of MB were 33.78%, 10.63%, and 2.92% over BQN, EDTA, and IPA, respectively. A significant decrease of photocatalytic performance in the presence of BQN was observed, implying that •O^2– played a major role in the photocatalytic reaction. Moreover, the photocatalytic performance decreased slightly in the presence of IPA and EDTA, suggesting that the •OH and h⁺ were the less important reactive species for the decomposition of MB.

Fig. 13 Active species trapping experiments under visible light irradiation.

The possible mechanism of photocatalytic degradation of MB is as follows. Photo-induced e ^– /h ⁺ pairs were generated on the surface of the photocatalyst MgAlLa-M during irradiation (Eq. 1). The photoinduced h ⁺ -oxidized H₂O₂ molecules that were adsorbed on the photocatalyst surface into •OH species (Eq. 2), while the photoinduced e ^– -activated H₂O₂ and O₂ molecules adsorbed on the catalyst surface into •OH, OH ^– species (Eq. 3) and •O₂ ^– species (Eq. 4), respectively. In addition, •OH species were also generated during irradiation of H₂O₂ (Eq. 5). H₂O₂ oxidized the active •O₂ ^– species into •OH, OH ^– , and O₂ species (equation 6). The •OH and O₂ are the main active species in the photocatalytic reaction (Ishibashi et al. Reference Ishibashi, Fujishima, Watanabe and Hashimoto2000; Liu & Sun Reference Liu and Sun2011; Xiang et al. Reference Xiang, Yu and Wong2011). The MB molecules were oxidized by •OH and O₂ species into small molecules (Eqs. 7 and 8).

(2) $\text{MgAlLa}-M+hv\to e^{–}+h^{+}$

(3) $h^{+}+{\text{H}}_2{\text{O}}_2\to 2•\text{OH}$

(4) $e^{–}+{\text{H}}_2{\text{O}}_2\to •\text{OH}+{\text{OH}}^{–}$

(5) $e^{–}+{\text{O}}_2\to •{{\text{O}}_2}^{–}$

(6) $H_2{O}_2+hv\to 2•\text{OH}$

(7) $•{{\text{O}}_2}^{–}+{\text{H}}_2{\text{O}}_2\to •\text{OH}+{\text{OH}}^{–}+{\text{O}}_2$

(8) $•\text{OH}+\mathrm{MB}\to \text{Degradation products}$

(9) ${\text{O}}_2+\mathrm{MB}\to \text{Degradation products}$

Based on the above discussions and the experimental results of MB degradation under different capture agents, when BQN was added to the reaction system, the •O₂ ^– was captured, so reactions 6, 7, and 8 were hindered, which would hinder the degradation process of MB. On the other hand, reaction 4 proceeded to the right while trapping •O₂ ^– . As a result, the formation of h ⁺ and •OH was accelerated. But the degradation of MB had not accelerated, implying that •O₂ ^– played a major role in the photocatalytic reaction.

When EDTA was added, the h ⁺ was captured, and reaction 2 was hindered and the amount of •OH produced was reduced, which would diminish the degradation of MB. Although the h ⁺ was captured, reaction 1 was pushed to the right, which led to the production of a large amount of e ^– at the surface and inside the catalyst. The production of e ^– and H₂O₂ and O₂ in the solution is enabled by reactions 3 and 4 to produce •OH and •O₂ ^– . The •OH could effectively oxidize MB, while •O₂ ^– and H₂O₂ could react according to reaction 6 to produce •OH and O₂, which could oxidize MB (Liu & Sun Reference Liu and Sun2011). The experimental results showed, however, that the degradation of MB slows down after adding EDTA, so one could speculate that reaction 2 was the key step in the degradation of MB.

In addition, when IPA was present, •OH in the solution was captured, which hindered the progress of reaction 7, but enabled reactions 2, 3, and 6 to proceed to the right. At the same time, reaction 1 was accelerated. A large number of e ^– and h ⁺ were produced in the solution, which accelerated the formation of •O₂ ^– and promoted the degradation of MB. In fact, in the photocatalytic degradation of MB, many bubbles were produced in the solution, leading to speculation that MB was degraded according to reaction 8. This was consistent with the results of accelerating the degradation of MB under isopropanol.

Based on the aforementioned experimental results, a possible mechanism for the enhanced photoactivity of MgAlLa-0.5-M/H₂O₂ is proposed. Under visible-light irradiation, photogenerated e ^– and h ⁺ is formed at the inside of MgAlLa-M after irradiation. These e ^– and h ⁺ move freely within the MgAlLa-M. When they move to the surface, e ^– interacts with O₂ and H₂O₂ in the solution to form •O₂ ^– and •OH, and h ⁺ reacts with H₂O₂ to form •OH, which together promotes the oxidative decomposition of MB. In addition, in the reaction system of MgAlLa-0.5-M/H₂O₂, •O₂ ^– plays a major role and the h ⁺ and •OH are the less important reactive species for the decomposition of MB.