Materials

For the present study, analytically pure M(NO₃)₂·6H₂O (M is Co, Mg, Ni, Zn, Cu), Cr(NO₃)₃·9H₂O, Na₂CO₃, NaOH, isopropanol, p-benzoquinone, ethylene diamine tetraacetic acid disodium salt, H₂O₂, and MB were purchased from Aladdin Reagent Co., Ltd (Shanghai, China). All solutions were prepared with deionized water.

Preparation of M₃Cr-CO₃-LDHs

A series of M ₃Cr-CO₃-LDHs was synthesized employing the co-precipitation method (Pan et al. Reference Pan, Ni, Cao and Li2012). Using a M/Cr molar ratio of 3.0, known amounts of M(NO₃)₂·6H₂O and Cr(NO₃)₃·9H₂O solid were dissolved in boiling water to be used as solution A. Then, 6.40 g of NaOH and 1.06 g of Na₂CO₃ were dissolved in boiling water as solution B. Under constant stirring, solutions A and B were added drop-wise to a beaker which was filled with deionized water. In this process, the mixed solution was adjusted using NaOH solution until the pH was between 9 and 10. The solution was then stirred continuously for 30 min. Finally, the slurry obtained was aged at 65°C for 24 h. The resulting precipitates were centrifuge washed with deionized water. After that, the samples were dried overnight at 65°C, and they were labeled using the format M ₃Cr-CO₃-LDH.

Structural Characterization

X-ray diffraction patterns of the samples were recorded an XD-6 instrument using CuKα radiation (Beijing Purkinje General Instrument Co., Ltd., Beijing, China). Infrared spectra were recorded using the KBr disc method on a Nicolet 5700 FTIR spectrophotometer (Thermo Scientific, Waltham, Massachusetts, USA). The thermal properties of the samples were measured using a CRY-2P instrument (Beijing Henven scientific instrument factory, Beijing, China); the heating rate was 10°C/min and the temperature range was 25–800°C. UV-Vis diffuse reflectance spectra (UV-Vis DRS) of the samples were tested using a UV4100 UV-Vis spectrophotometer (Thermo Scientific, Waltham, Massachusetts, USA); BaSO₄ was used as a reference standard. The time-dependent UV-Vis spectra of MB solutions were measured using a Shimadzu UV-2600 UV-Vis spectrophotometer (Tokyo, Japan).

Catalytic Experiments

The photocatalytic properties of ZnCr-CO₃-LDHs were determined by photodegradation of MB solution under visible light irradiation using a 150 W halogen lamp. A photocatalyst (0.5 g) was mixed with 25 mL of 5 mg/L MB aqueous solution with an additional 0.5 mL of H₂O₂. The mixed solution was stirred with a magnetic stirrer. A sample solution was taken at a given time and separated through centrifugation. The concentration of MB solution was evaluated by UV-Vis spectrophotometry at 662 nm. The degradation efficiency was calculated using the following equation:

$D=\left(C_0–C_t\right)/C_0\times 100=\left(A–A_t\right)/A\times 100\%$

where C ₀ is the initial concentration of MB solution after avoiding light, and C _t is the concentration of MB solution after a known photoreaction time, A is the initial absorbance of MB solution after avoiding light, and A _t is the absorbance of MB solution after a known photoreaction time.

XRD Characterization

The regularity of the crystal structure, crystallinity, and crystal-cell parameters of the samples was tested using XRD (Chen et al. Reference Chen, Bao and Wang2016). The XRD patterns of five kinds of M ₃Cr-CO₃-LDHs (Fig. 1) showed that the diffraction peaks were at 11.7, 23.6, and 34.7°2θ, which corresponded to the characteristic diffraction peaks of hydrotalcite; and the crystal face indexes were (003), (006), and (009), respectively. The above-mentioned results were consistent with published information (Prevot et al. Reference Prevot, Forano and Besse2005; Pan et al. Reference Pan, Cao, Ni, Li, Chen, Tang and Xu2011; Yang et al. Reference Yang, Qian, Chu and Zhang2018; Xu et al. Reference Xu, Pan, Meng, Guo, Wu and Chen2019), indicating that the five kinds of M ₃Cr-CO₃-LDHs had typical layered structures. The diffraction peaks of these five kinds of M ₃Cr-CO₃-LDHs were strong and sharp, which indicated that the synthesized M ₃Cr-CO₃-LDHs had fine crystallinity. In addition, the diffraction peak of Zn₃Cr-CO₃-LDHs was the strongest, which showed that Zn₃Cr-CO₃-LDHs formed the best crystals. The peak observed at d ₀₀₉ in Cu₃Cr-CO₃-LDHs, however, was unidentified and indicated less crystallinity. It may have been a consequence of incomplete washing.

Fig. 1 XRD patterns of Mg₃Cr-CO₃-LDHs (a), Co₃Cr-CO₃-LDHs (b), Ni₃Cr-CO₃-LDHs (c), Cu₃Cr-CO₃-LDHs (d), and Zn₃Cr-CO₃-LDHs (e)

The layer spacing of the diffraction peak corresponding to the d ₀₀₃ plane was calculated from the Bragg equation (nλ = 2dsinθ), giving values for CuCr-CO₃-LDHs, ZnCr-CO₃-LDHs, NiCr-CO₃-LDHs, MgCr-CO₃-LDHs, and CoCr-CO₃-LDHs of 0.747, 0.764, 0.773, 0.765, and 0.749 nm, respectively. These data are similar to the results reported in the literature of d ₀₀₃ spacing of carbonate hydrotalcite (Sels et al. Reference Sels, De Vos and Jacobs2001; Rives et al. Reference Rives, Prieto, Dubey and Kannan2003; Tichit et al. Reference Tichit, Gerardin, Durand and Coq2006; Yang et al. Reference Yang, Chu and Zhang2019). Comparison of the five d ₀₀₃ results revealed that the MgCr-CO₃-LDHs layer spacing was the largest, which might be due to the fact that Mg has the smallest atomic radius.

FTIR Spectroscopy Analysis

The FTIR patterns of the five kinds of M ₃Cr-CO₃-LDHs (Fig. 2) revealed five absorption peaks between 3470 and 3540 cm^–1 which were ascribed to the stretching vibration of the H₂O molecule; this was because the surface and the interlayer space of the LDHs adsorbed a certain number of H₂O molecules. Compared with the stretching vibration of free hydroxyl (at ~3650 cm^–1), the peak shifted to a lower wavenumber, which indicated that a hydrogen bond between layers of H₂O and the CO₃ ^2– ion and the hydroxyl group appeared. In addition, absorption peaks between 1650 and 1665 cm^–1 were caused by the bending vibration of –OH in the crystal water. The peak at 1384–1386 cm^–1 was ascribed to the anti-symmetric stretching vibration peak of CO₃ ^2–. Absorption peaks between 520 and 630 cm^–1 belonged to the characteristic peak of plane-bending vibration of C-O. All the patterns showed the characteristic peaks of hydrotalcite. IR spectra further confirmed the synthesis of M ₃Cr-CO₃-LDHs had typical structural characteristics of hydrotalcite.

Fig. 2 FTIR patterns of Mg₃Cr-CO₃-LDHs (a), Co₃Cr-CO₃-LDHs (b), Ni₃Cr-CO₃-LDHs (c), Cu₃Cr-CO₃-LDHs (d), and Zn₃Cr-CO₃-LDHs (e)

TG-DTA Characterization

The TG-DTA curves of Mg₃Cr-CO₃-LDHs, Zn₃Cr-CO₃-LDHs, Co₃Cr-CO₃-LDHs, Ni₃Cr-CO₃-LDHs, and Cu₃Cr-CO₃-LDHs (Fig. 3a) showed that during the heating process, the thermal behavior of these compounds could be divided into two stages. In the first stage, the weight loss was ~17–25%, and in the second stage, weight loss was ~17–23%. Compared with the five kinds of compounds, Mg₃Cr-CO₃-LDHs had the largest weight loss: 23%. The comparison of DTA curves showed that the thermal decomposition of the five kinds of M ₃Cr-CO₃-LDHs was due mainly to two obvious endothermic processes. An endothermic peak appeared at low temperature (i.e. 50–200°C), which corresponded to the removal of surface-adsorbed water and interlayer water. A peak also appeared at higher temperature (i.e. 265–450°C), indicating that the interlayer hydroxyl was dehydrated and interlayer CO₃ ^2– anions were decomposed or removed. These results were similar to the thermal decomposition of typical Mg-Al hydrotalcites (Hibino et al. Reference Hibino, Yamashita, Kosuge and Tsunashima1995).

Fig. 3 TG curves of of M ₃Cr-CO₃-LDHs (a), and DTA curves of M ₃Cr-CO₃-LDHs (b)

Comparison of second-stage DTA curves (Fig. 3b) indicated the thermal decomposition temperature of Cu₃Cr-CO₃-LDHs was the highest, which showed that the thermal stability was the best. The relative thermal stabilities of the other four kinds of hydrotalcites were in the order: Mg₃Cr-CO₃-LDHs > Ni₃Cr-CO₃-LDHs > Zn₃Cr-CO₃-LDHs ≈ Co₃Cr-CO₃-LDHs, consistent with the result reported previously (Pan et al. Reference Pan, Ni, Cao and Li2012), i.e. the order of thermal stability of hydrotalcites obtained from the second law of crystal chemistry.

UV-Vis DRS Analysis

The UV-Vis DRS spectra of LDHs (Fig. 4) showed that Mg₃Cr-CO₃-LDHs and Ni₃Cr-CO₃-LDHs (curve a and curve c) contained obvious absorption peaks at ~333 and 480 nm. Co₃Cr-CO₃-LDHs (curve b) had a weak absorption peak in the visible region (λ > 400 nm), but Cu₃Cr-CO₃-LDHs (curve d) had almost no absorption peaks, while Zn₃Cr-CO₃-LDHs (curve e) had obvious absorption peaks at ~410 and 560 nm. In order to utilize these kinds of catalysts effectively in visible light (λ > 400 nm), UV-Vis DRS patterns of the five kinds M ₃Cr-CO₃-LDHs were compared and analyzed. The absorption region of Zn₃Cr-CO₃-LDHs was widest, next was Ni₃Cr-CO₃-LDHs, and Mg₃Cr-CO₃-LDHs and Co₃Cr-CO₃-LDHs had narrower spectra, while the absorption region of Cu₃Cr-CO₃-LDHs was narrowest. With the various divalent metal cations, the final products were different colors (Zhao et al. Reference Zhao, Zhang, Li, Yan, He, Tian and Duan2011), meaning that their absorption of visible light spanned different wavelengths and provided an incentive to explore their visible-light catalytic properties.

Fig. 4 UV-Vis DRS patterns of Mg₃Cr-CO₃-LDHs (a), Co₃Cr-CO₃-LDHs (b), Ni₃Cr-CO₃-LDHs (c), Cu₃Cr-CO₃-LDHs (d), and Zn₃Cr-CO₃-LDHs (e)

Photocatalytic Properties of ZnCr-CO₃-LDHs

In order to investigate the catalytic activity of ZnCr-CO₃-LDHs, MB solution was selected as the substrate for catalytic oxidation. The removal rate of MB was only 5% in the presence of ZnCr-CO₃-LDHs without visible light (Fig. 5), which was attributed to the adsorption capability of ZnCr-CO₃-LDHs (for MB). In the presence of ZnCr-CO₃-LDHs exposed to visible light, the removal rate was 10.06% after 140 min, indicating that ZnCr-CO₃-LDHs is a photocatalyst for MB degradation. When both ZnCr-CO₃-LDHs and H₂O₂ were involved in an unilluminated reaction, the removal rate of MB was 46.36% after 140 min. However, MB decreased by 49.08% when using only H₂O₂ for 140 min under visible light irradiation, which might be due to oxidation of H₂O₂. When MB solution was exposed to ZnCr-CO₃-LDHs, H₂O₂, and visible light irradiation together, >90.67% of MB was removed after 140 min. The above phenomena indicated that MB was most effectively degraded by ZnCr-CO₃-LDHs/H₂O₂ under visible light irradiation.

Fig. 5 Photocatalytic degradation of MB under various conditions

Reuse Performance

Repeated experiments on photodegradation of MB using ZnCr-CO₃-LDHs/H₂O₂ were performed. After each run, the photocatalysts were collected by centrifugation; distilled water was added to the centrifuge tube and stirred until the catalyst was washed washed thoroughly, and then centrifugation was continued. The process was repeated three times. After the ZnCr-CO₃-LDHs were reused three times, the degradation rates of MB solution were still >90% (Fig. 6), indicating that ZnCr-CO₃-LDHs were stable and not photocorroded during photocatalytic reactions. In addition, when the photocatalysts were used for the second or third times, the MB was degraded rapidly. The photocatalytic degradation rate of MB was 42.65% (second attempt) and 42.14% (third attempt) after 20 min, which was faster than that of the first time of photocatalytic degradation of MB (17.76%).

Fig. 6 Catalyst recycling study of ZnCr-CO₃-LDHs for photocatalytic degradation of MB

The quicker degradation might be due to the transition of electrons in the valence band when using ZnCr-CO₃-LDHs in the presence of light, resulting in the formation of photogenerated electron-hole pairs on the inside and on the surface of ZnCr-CO₃-LDHs. Electrons and holes reacted with H₂O₂ to produce active species that effectively prevented recombination of electron-hole pairs, and effectively transferred electrons and holes, resulting in a series of redox reactions. In the process of reuse, visible light irradiation may activate ZnCr-CO₃-LDHs to produce photogenerated electron-hole pairs. Therefore, the reaction of photogenerated electrons and holes with H₂O₂ was accelerated and, ultimately, the degradation of MB was also accelerated.

The XRD patterns of ZnCr-CO₃-LDHs after different cycles (Fig. 7) showed that the diffraction peaks of ZnCr-CO₃-LDHs were unchanged and the peaks were similar in shape when ZnCr-CO₃-LDHs were used two or three times. The results indicated that the crystal structure of hydrotalcites did not change obviously after cyclic use, and also proved that ZnCr-CO₃-LDHs had good stability.

Fig. 7 XRD patterns of ZnCr-CO₃-LDHs after various cycle times

Catalyst Types

The photocatalytic degradation of MB with different metal ions substituted in MCr-CO₃-LDHs (M = Zn, Mg, Co, Ni, or Cu) (Fig. 8) showed that, under the same experimental conditions, the degradation of MB with CoCr-CO₃-LDHs after 60 min was nearly 100%, while the degradation with MgCr-CO₃-LDHs and CuCr-CO₃-LDHs was >90%. But the degradation efficiencies of ZnCr-CO₃-LDHs and NiCr-CO₃-LDHs were worse; the degradation of MB was 55.78% and 48.14%, respectively, under the same conditions. When M was Mg, Co, or Cu, MB degraded rapidly, and a large number of bubbles was produced during the reaction. When M was Zn or Ni, fewer bubbles were formed. The photocatalytic activity of five kinds of M ₃Cr-CO₃-LDHs was in the order: Co₃Cr-CO₃-LDHs > Mg₃Cr-CO₃-LDHs > Cu₃Cr-CO₃-LDHs > Zn₃Cr-CO₃-LDHs > Ni₃Cr-CO₃-LDHs.

Fig. 8 Effect of catalyst types on degradation of MB

Photocatalytic Mechanism

The intermediate products of MB degradation were tested (Fig. 9); the maximum absorption peak (λ = 664 nm) of MB solution decreased gradually with the prolongation of reaction time, and the maximum absorption peak shifted to λ = 651 nm. The blue shift (to lower wavelengths) of the maximum absorption wavelength of MB occurred, possibly due to the demethylation reaction found when photocatalytic degradation of MB takes place, and it removed a methyl to form azure B. According to the literature (Zhang et al. Reference Zhang, Oyama, Aoshima, Hidaka, Zhao and Serpone2001), absorption bands at 654–648, 620, and 610 nm are due to MB molecular transformations to azure A, azure B, and azure C, respectively. The results of photocatalytic degradation of MB agreed with those reported in the literature. In addition, along with the prolonged irradiation time, a new absorption peak appeared in the UV region (λ = 372 nm); the absorption peak was enhanced as illumination time increased, suggesting that new substances appeared during degradation of MB.

Fig. 9 UV-Vis absorption spectra of MB with different reaction times in the presence of ZnCr-CO₃-LDHs/H₂O₂ under visible light irradiation.

In order to evaluate the role of these active oxidants, scavengers were added to the photocatalytic system. These were isopropanol (IPA) for •OH, p-benzoquinone (BQ) for •O₂ ^– (Yuan et al. Reference Yuan, Wei, Hu, Yao, Jiang, Fang and Chu2015), and ethylene diamine tetraacetic acid disodium salt (EDTA-2Na) for h ⁺ (Xiong et al. Reference Xiong, Chen, He, Sun and Wang2012; Pan et al. Reference Pan, Wang, Sun, Xu and Wang2013; Chen et al. Reference Chen, He, Lv, Liu, Wang and Lv2014). The molar ratio of initial concentration of these three kinds of scavengers in the reaction system to MB was 200:1. The photodegradation of MB was slightly inhibited by the addition of IPA (Fig. 10). The results indicated that MB degradation was driven mainly by the participation of •OH radicals. But the reaction was accelerated when EDTA-2Na and BQ were added. The degradation rates of MB exceeded 90% after irradiation for 60 min.

Fig. 10 Effect of various scavengers on degradation of MB

Based on the band theory of solids and the mechanism of semiconductor photocatalysis, researchers speculated on the possible mechanism of photocatalytic degradation of MB. Photoinduced e ^–/h ⁺ pairs were generated on the surface of the photocatalyst ZnCr-CO₃-LDHs during irradiation (Eq. 1). H₂O₂ molecules oxidized by photoinduced h ⁺ were adsorbed into •OH species on the photocatalyst surface (Eq. 2), while the photoinduced e ^–-activated H₂O₂ and O₂ molecules were adsorbed into •OH, OH^– species on the catalyst surface (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 (Eq. 6). •OH and O₂ species are known to be the main active species in the photocatalytic reaction (Ishibashi et al. Reference Ishibashi, Fujishima, Watanabe and Hashimoto2000; Xiang et al. Reference Xiang, Yu and Wong2011). The MB molecules were oxidized by •OH and O₂ species into small molecules (Eq. 7 and Eq. 8).

(1) $\text{ZnCr}-{\mathrm{CO}}_3-\text{LDHs}+hv\to e^{–}+h^{+}$

(2) $h^{+}+H_2{O}_2\to 2•\mathrm{OH}$

(3) $e^{–}+H_2{O}_2\to •\mathrm{OH}+{\mathrm{OH}}^{-}$

(4) $e^{–}+O_2\to •{O_2}^{–}$

(5) $H_2{O}_2+hv\to 2•\mathrm{OH}$

(6) $•{O_2}^{–}+H_2{O}_2\to •\mathrm{OH}+{\mathrm{OH}}^{-}+O_2$

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

(8) $O_2+\mathrm{MB}\to \text{Degradation products}$

When the EDTA-2Na was added to the reaction system, the h ⁺ was captured, which caused Eq. 1 to move to the right, so that a large amount of photoinduced e^– formed on the catalyst surface and inside. Photoinduced e^– deoxidizes H₂O₂ and O₂ molecules to produce •OH and •O₂ ^– species on the photocatalyst surface according to Eq. 3 and Eq. 4. The •OH could oxidize MB effectively, while •O₂ ^– and H₂O₂ generated •OH and O₂ according to Eq. 6. Therefore, in the process of photocatalysis of MB, a large number of bubbles were produced in the solution, and the MB was degraded rapidly. When p-benzoquinon was added, •O₂ ^– in the solution was trapped, which caused Eq. 4 to move to the right. Subsequently, Eq. 1 and Eq. 2 were accelerated, speeding up the generation of •OH and accelerating the degradation of MB. However, one could speculate that the addition of p-benzoquinon did not capture •O₂ ^– rapidly, so that part of •O₂ ^– could react with H₂O₂ (Eq. 6). This concept is consistent with the fact that a large number of bubbles was produced in the solution during the experiment.

Based on the above discussion, combined with the analysis of UV-Vis absorption spectra, the possible mechanism of ZnCr-CO₃-LDHs photocatalytic degradation of MB was considered (Fig. 11). Under visible light irradiation, ZnCr-CO₃-LDH was excited, and its valence band electrons were transferred to the conduction band to produce photogenerated electrons (e ^–) and holes (h ⁺). These e ^– and h ⁺ could move freely in the catalyst. When they moved to the catalyst surface, e ^– reacted with O₂ and H₂O₂ in the solution to produce •O₂ ^– and •OH, and h ⁺ in the valence band reacted with H₂O₂ to generate •OH. These active species then reacted with MB, causing MB to undergo demethylation, and also oxidized MB to small molecules, thus achieving the purpose of decolorization and degradation.

Fig. 11 Possible mechanism of photocatalytic degradation of MB using ZnCr-CO₃-LDHs