An eco-friendly route for template-free synthesis of high specific surface area mesoporous CeO₂ powders and their adsorption for acid orange 7

Abstract

An eco-friendly route was developed for the synthesis of mesoporous CeO₂ powders without any additional template. The original cerium precursors were separated from Ce³⁺ aqueous solution by (NH₄)₂CO₃ or Na₂CO₃ via a chemical precipitation method, then H₂O₂ was introduced to induce the phase transformation from original cerium precursors to CeO₂ precursors with initial porous structures, finally the crystallinities of CeO₂ precursors were improved by a hydrothermal treatment, meanwhile the mesoporous structures of final CeO₂ powders were formed. The BET surface areas of mesoporous CeO₂ powders synthesized using (NH₄)₂CO₃ and Na₂CO₃ as precipitants were 106.1 and 76.9 m² g⁻¹, respectively. Moreover, a mesoporous CeO₂ sample with BET surface area of 100.0 m² g⁻¹ was also synthesized using commercial Ce₂(CO₃)₃·xH₂O as an existing cerium precursor under the same conditions as control, which could shorten experimental processes and reduce costs. The oxidation-induced phase transformation from original cerium precursors to CeO₂ precursors with initial porous structures was the precondition for further forming of mesoporous structures of final CeO₂ powders during the hydrothermal process. These mesoporous CeO₂ powders showed the rapid and effective adsorption for acid orange 7 dye from simulated wastewater without pH pre-adjustment at room temperature. Furthermore, the adsorption capacities of these mesoporous CeO₂ powders for removal of acid orange 7 dye were determined according to the Langmuir linear fits.

---

Introduction

Acid orange 7 (AO7) dye is one of the most common synthetic dyes in various industries ranging from dyeing to printing.^1–3 AO7 is considered toxic and could cause harmful health effects to human and aquatic organisms, such as skin diseases and carcinogenesis.^4,5 Moreover, it is difficult to biologically degrade AO7 in wastewaters because of its recalcitrant azo bond with an aromatic structure.⁶ Therefore, it is essential to treat the industrial wastewaters containing AO7.^7–9 To date, many approaches have been conducted to control organic pollutants, such as biodegradation,^10–12 photooxidation,^13,14 chemical oxidation,^15–17 electrochemistry,^18,19 ultrasonic destruction^20,21 and adsorption.^22–24 Among these techniques, adsorption using adsorbents is considered to be one of the most convenient and cost-efficient methods.²⁵ Huang et al. prepared a nitrilotriacetic acid anhydride modified ligno-cellulosic bio-adsorbent for removal of Cd²⁺ and Pb²⁺, the maximum sorption capacities for Cd²⁺ and Pb²⁺ could reach 143.4 and 303.5 mg g⁻¹ at 298.0 K, respectively.²⁶ Lu et al. reported the removal of acenaphthene by biochar and raw biomass, and investigated the effects of coexisting metal ions and organic compounds on their sorption performances.²⁷ Moreover, Wu et al. reviewed the role of biochar on composting of organic wastes and remediation of contaminated soils.²⁸ The mesoporous ceria (CeO₂) can serve as a promising candidate for removal of AO7 because of its high specific surface area and well-defined pore topology.

Generally, mesoporous CeO₂ powders are prepared by template methods with either surfactants as soft templates^29,30 or other porous material as hard templates.^31,32 However, the template methods require either additional procedures or high energy consumption in order to eliminate the hard or soft sacrificial templates, such as dissolution or heat treatment.^33–35 Moreover, the crystallinity of mesoporous CeO₂ even needs to be improved again by calcination, which easily causes the collapse of pore structures and thus reduces the specific surface area of CeO₂.³⁶ To date, there are limited reports for template-free synthesis of CeO₂ powders with mesoporous structures. For example, Wei et al. fabricated mesoporous CeO₂ nanoflowers with a BET surface area (S_BET) of 95.7 m² g⁻¹, however, polyvinylpyrrolidone (PVP) was introduced as a structure-directing agent to synthesize Ce(HCOO)₃ precursor in alcoholic solution, in which formic acid and ammonia solution were also added. Then, hydrogen peroxide was introduced as an oxidant to induce the phase transformation from Ce(HCOO)₃ to CeO₂ with inherited morphology. Finally, mesoporous CeO₂ nanoflowers were obtained by following solvothermal treatment at 150 °C for 6 h and drying at 70 °C for 10 h.³⁷ In another study, Xie et al. reported a template-free hydrothermal synthesis of flower-like CeO₂ powders, and its S_BET was 38.8 m² g⁻¹. The potassium chlorate and dimethyl formamide were employed, and the interaction effect of them played an important role in the formation of flower-like CeO₂.³⁸ Moreover, He et al. synthesized mesoporous CeO₂ colloidal spheres by the assembly of CeO₂ nanoparticles and nanocubes, respectively. The S_BET of mesoporous CeO₂ colloidal spheres assembled by nanoparticles and nanocubes were 114.3 and 122.5 m² g⁻¹, respectively. The whole process could be divided into three steps: the CeO₂ nanoparticles and nanocubes were first synthesized by a hydrothermal method and CO-assisted hydrothermal approach, respectively. Then, the CeO₂ nanocrystals self-assembled into colloidal spheres via an emulsion-based bottom-up self-assembling method. Finally, the colloidal spheres were obtained after following drying at 70 °C and calcination at 350 °C for 4 h.³⁹ From the above, one sample, mild, low-cost and environment-friendly route for template-free synthesis of mesoporous CeO₂ powders is desirable.

In the previous work, we presented a combined bottom-up and top-down route for template-free synthesis of mesostructured CeO₂ particles using Ce(NO₃)₃·6H₂O (cerium source), NH₄HCO₃ (precipitant), H₂O₂ (oxidant) and H₂O (solvent) as starting reagents, and its specific surface area was 166.5 m² g⁻¹.⁴⁰ In this work, (NH₄)₂CO₃ or Na₂CO₃ was employed in place of NH₄HCO₃ as a precipitant for separation of cerium precursors from Ce³⁺ aqueous solution. As an expansive research, commercial Ce₂(CO₃)₃·xH₂O powders were used as an existing precursor for synthesis of mesoporous CeO₂ powders. The roles of H₂O₂ were discussed, and the effects of calcination on the grain sizes and S_BET of mesoporous CeO₂ powders were investigated. Additionally, the absorption characteristics of these mesoporous CeO₂ powders for AO7 dye were investigated.

Furthermore, it is worth noting that the mesoporous CeO₂ powders were synthesized in this work just using (NH₄)₂CO₃, Na₂CO₃, H₂O₂ and H₂O without any additional reagent and post-treatment. (NH₄)₂CO₃, Na₂CO₃ and H₂O₂ are accessible, cheap and safe chemistry reagents, which not only can save the cost, but also reduce the pollution degree to environment. Moreover, the route, using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor for synthesis of mesoporous CeO₂ powders, can shorten experimental processes and reduce costs.

Experimental

Materials

Cerium nitrate hexahydrate (Ce(NO₃)₃·6H₂O, 99.95%), ammonium carbonate ((NH₄)₂CO₃, 99.999%), sodium carbonate (Na₂CO₃, 99.5%), hydrogen peroxide (H₂O₂, 30 wt%) and commercial cerium carbonate hydrate (Ce₂(CO₃)₃·xH₂O, 99.9%) were supplied by Aladdin Co. Ltd. Acid orange 7 (AO7, >97.0%) was obtained from Tokyo Chemical Industry Co. Ltd.

Synthesis of mesoporous CeO₂ powders

As shown in Fig. 1, firstly, the original cerium precursors were separated from Ce³⁺ aqueous solution by (NH₄)₂CO₃ or Na₂CO₃ via a chemical precipitation method. Typically, 4 mmol Ce(NO₃)₃·6H₂O was dissolved into 28 mL distilled water to form a clear Ce³⁺ solution, and 16 mmol precipitant ((NH₄)₂CO₃ or Na₂CO₃) was added to the above solution under continuous stirring, immediately forming a white precipitate (labelled as Precursor 1 and Precursor 2, respectively). Meanwhile, as an extension experiment, commercial Ce₂(CO₃)₃·xH₂O powders were used as an existing precursor. The commercial Ce₂(CO₃)₃·xH₂O powders were dispersed in 28 mL distilled water, and the subsequent experimental steps were similar to that of the suspension of Precursor 1 and Precursor 2.

Fig. 1 Synthesis of mesoporous CeO₂ using (NH₄)₂CO₃, Na₂CO₃ as precipitants, and using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor in the presence of H₂O₂.

Then, H₂O₂ was introduced to induce the phase transformation from original cerium precursors to CeO₂ precursors. Typically, 7 mL H₂O₂ was added to the above suspension containing Precursor 1, Precursor 2 and commercial Ce₂(CO₃)₃·xH₂O, then stirring for 30 min and aging for 3 h. The as-prepared orange precipitates using (NH₄)₂CO₃, Na₂CO₃ as precipitants and using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor were labelled as Precursor 1-1, Precursor 2-1 and Precursor 3-1, respectively. Note that all operations were performed at room temperature.

The last step was the synthesis of mesoporous CeO₂ powders by a hydrothermal treatment. Typically, the above CeO₂ precursors in the total mother liquor were transferred into a 50 mL Teflon-lined stainless steel autoclave. After reacting at 200 °C for 24 h, the autoclave was cooled down. Then the resulting precipitates were washed with distilled water and ethanol, and dried at 60 °C for 24 h. These hydrothermally produced CeO₂ powders were labelled as Sample 1, Sample 2 and Sample 3, respectively.

For comparison, the samples were synthesized under the same conditions as control, however, in the absence of H₂O₂. Moreover, in order to investigate the effects of calcination on the grain sizes and S_BET of mesoporous CeO₂, these samples (Sample 1, Sample 2 and Sample 3) were treated by following calcination at 500 °C for 2 h, and their S_BET were also determined.

Characterization

The crystallographic phases of precursors and samples were characterized by X-ray diffraction (XRD, D/Max 2200PC). The microstructures of precursors and samples were evaluated by transmission electron microscopy (TEM, JEM-2100F). The specific surface areas, pore volumes and pore size distributions of mesoporous CeO₂ powders were obtained from nitrogen adsorption–desorption measurements (QuadraSorb SI).

Adsorption studies

About 10.3 g of AO7 powders (>97.0%) were dissolved in distilled water, and diluted to 100 mL with distilled water, the as-obtained concentration of AO7 solution was 10 g L⁻¹. The different concentrations of AO7 solution (20–120 mg L⁻¹) were obtained by pipetting varied volume of the above 10 g L⁻¹ AO7 solution into 100 mL volumetric flask and bringing to volume by distilled water. Subsequently, 0.2 g CeO₂ sample was dispersed into 100 mL of AO7 solution at varying initial concentrations (adsorbent dosage: 2.0 g L⁻¹) without pH pre-adjustments. The mixture was stirred at a constant speed (200 rpm) and temperature (298.0 K). Then, 4 mL suspension was withdrawn at regular intervals and centrifuged. The absorbance of supernatant was measured using an UV-2600 spectrophotometer.

The Beer–Lambert law is linear relationship between the absorbance and concentration of absorbing species.⁴¹ So, the concentration of AO7 dye can be converted from its absorbance based on Beer–Lambert law. The adsorption efficiency (η, %) and adsorption amount (q, mg g⁻¹) for AO7 dye were calculated using eqn (1) and (2), respectively.⁴²

where C₀ (mg L⁻¹) is the initial concentration of AO7 dye, C_t (mg L⁻¹) is the concentration of AO7 dye at a given time t (t = 0–60 min), C_e (mg L⁻¹) is the concentration of AO7 dye at equilibrium, m (g) is the mass of CeO₂ powders, and V (L) is the volume of AO7 solution.

Langmuir model as shown in eqn (3) was used to examine the adsorption characteristics of the as-obtained mesoporous CeO₂ powders.⁴³ And the saturated adsorption amount (q_m, mg g⁻¹) was obtained based on Langmuir linear fitting of adsorption isotherm curve.

where K_L (L mg⁻¹) is Langmuir constant. The eqn (3) can be rearranged to a linear form as shown in eqn (4). As observed, the plot of C_e/q against C_e can give a straight line with the slope of 1/q_m and intercept of 1/(K_Lq_m), and the values of q_m and K_L can be evaluated according to the slope and intercept.

Results and discussion

Phase characterizations of precursors

The crystallographic phases of precursors after adding the precipitant and H₂O₂ were determined by XRD. Fig. 2a and b show the XRD patterns of Precursor 1 and Precursor 2 obtained by adding (NH₄)₂CO₃ and Na₂CO₃ to Ce³⁺ aqueous solution, respectively. As an verification and comparison, the XRD analysis of commercial Ce₂(CO₃)₃·xH₂O powders also were performed, and its XRD pattern was showed in Fig. 2c. As observed in Fig. 2a, the phase structure of Precursor 1 synthesized following adding (NH₄)₂CO₃ to Ce³⁺ aqueous solution was o-Ce(CO₃)OH (JCPDS no. 41-0013; density = 4.545 g cm⁻³). The XRD pattern of Precursor 2 in Fig. 2b was similar to that of commercial Ce₂(CO₃)₃·xH₂O in Fig. 2c, indicating its major phase of Ce₂(CO₃)₃·8H₂O (JCPDS no. 38-0377; density = 2.790 g cm⁻³). Moreover, the phase of precursor synthesized following adding NH₄HCO₃ to Ce³⁺ aqueous solution in our previous report⁴⁰ was similar to that of Precursor 2 in Fig. 2b and commercial Ce₂(CO₃)₃·xH₂O in Fig. 2c. However, the difference in phase structure of original cerium precursors will subtly affect the S_BET of final CeO₂ samples. Interestingly, the Precursor 1 synthesized following adding (NH₄)₂CO₃ to Ce³⁺ aqueous solution depended on the amount of (NH₄)₂CO₃. When the amount of (NH₄)₂CO₃ was less than 8 mmol, the major phase of as-obtained precursor was Ce₂(CO₃)₃·8H₂O. And the major phase structure was o-Ce(CO₃)OH when the amount of (NH₄)₂CO₃ was more than 10 mmol (see Fig. S1†).

Fig. 2 XRD patterns of (a) Precursor 1, (b) Precursor 2 and (c) commercial Ce₂(CO₃)₃·xH₂O.

Fig. 3a–c show the XRD patterns of precursors obtained following addition of H₂O₂ (Precursor 1-1, Precursor 2-1 and Precursor 3-1, respectively). As observed in Fig. 3a–c, the peaks related to Ce₂(CO₃)₃·8H₂O and o-Ce(CO₃)OH were no longer present. The XRD pattern of Precursor 1-1 in Fig. 3a displayed several relatively well-resolved peaks that could be indexed to the (111), (200), (220) and (311) planes of face-centred cubic CeO₂ (JCPDS no. 34-0394; density = 7.215 g cm⁻³). The XRD pattern of Precursor 2-1 in Fig. 3b and Precursor 3-1 in Fig. 3c all showed three broad featureless peaks centred at 2θ = 29, 47 and 56°, and the broad featureless peaks centred at 2θ = 29° was more easily observed than others. Compared with the XRD pattern in Fig. 3a, the broad featureless peaks centred at 2θ = 29° in Fig. 3b and c could be indexed to the (111) plane of CeO₂ phase, but with relatively low crystallinities. Combining with the XRD analyses in Fig. 2, we can derive a conclusion that H₂O₂ can induce the phase transformations from original cerium precursor (Ce₂(CO₃)₃·xH₂O or o-Ce(CO₃)OH) to CeO₂ precursor because of its oxidation.

Fig. 3 XRD patterns of (a) Precursor 1-1, (b) Precursor 2-1 and (c) Precursor 3-1.

Physical characterizations of the hydrothermally produced CeO₂ powders

Fig. 4a–c show the XRD patterns of the hydrothermally produced CeO₂ samples (Sample 1, Sample 2 and Sample 3, respectively). As observed, all the hydrothermally produced samples displayed several well-resolved peaks that indexed to the (111), (200), (220), (311), (222), (400) and (331) planes of face-centred cubic CeO₂ (JCPDS no. 34-0394), and no additional phases were observed. Moreover, the crystallinities of the hydrothermally produced CeO₂ in Fig. 4 were improved compared with that of the CeO₂ precursors in Fig. 3, which may be attributed to the rearrangement of CeO₂ grains with good crystallinities under certain temperatures and pressures during the hydrothermal process.⁴⁰ Combining with the results of XRD analyses in Fig. 3, we can draw a conclusion that the crystallinities of CeO₂ precursors could be improved and the pure CeO₂ samples can be obtained through a hydrothermal treatment.

Fig. 4 XRD patterns of (a) Sample 1, (b) Sample 2 and (c) Sample 3.

To understand the microstructures of the hydrothermally produced CeO₂ samples, TEM analyses were performed. Fig. 5a, c and e show the TEM images of the hydrothermally produced CeO₂ (Sample 1, Sample 2 and Sample 3, respectively). As observed in Fig. 5a, c and e, the porous structure of CeO₂ particles and the presence of pores around CeO₂ grains can be observed. In addition, the grain size of Sample 1 was obviously greater than that of Sample 2 and Sample 3. The corresponding high-magnification TEM images of Sample 1, Sample 2 and Sample 3 were showed in Fig. 5b, d and f, respectively. The porous structures of these CeO₂ particles could be further evidenced, and these CeO₂ particles consisted of aggregated grains. Moreover, these pores resulted from these aggregated grains, and the calculated grain sizes were about 7.7, 4.3 and 4.8 nm for Sample 1, Sample 2 and Sample 3, respectively. The existence of pore structure resulted from these CeO₂ particles possessing bigger specific surface area, consequently, more active sites can be provided for the adsorption of pollutants, which are beneficial to improving their capture capability.⁴⁴ The grain size of CeO₂ will have an impact on the pore diameter and pore volume of CeO₂ powders, and then affected their S_BET. Further analysis of S_BET was conducted by nitrogen adsorption–desorption experiments as discussed later.

Fig. 5 TEM images of (a) Sample 1, (c) Sample 2 and (e) Sample 3 ((b), (d) and (f) show the corresponding high-magnification TEM images, respectively).

To further clarify the porous structures of the final hydrothermally produced CeO₂ samples (Sample 1, Sample 2 and Sample 3, respectively), nitrogen adsorption–desorption experiments were conducted to determine their S_BET, average pore sizes and pore volumes. Fig. 6a–c show the nitrogen adsorption–desorption isotherms of Sample 1, Sample 2 and Sample 3, respectively. From Fig. 6a–c, the recorded adsorption–desorption isotherms exhibited the hysteresis loops ranging from 0.4 to 1.0, suggesting their mesoporous structures.⁴⁵ Furthermore, the profiles of the nitrogen adsorption–desorption isotherms were similar to that of the mesoporous CeO₂ reported in previous literature.²⁹ The insets in Fig. 6a–c show the corresponding Barrett–Joyner–Halenda (BJH) pore size distribution curves. As observed the inset in Fig. 6a and b, BJH calculations for the pore size distributions presented a single distribution centred at about 7.8 and 3.4 nm for Sample 1 and Sample 2, respectively. By contrast, the BJH pore size distribution curves of Sample 3 presented two distributions centred at about 3.8 and 5.5 nm as observed the inset in Fig. 6c.

Fig. 6 Nitrogen adsorption–desorption isotherms of (a) Sample 1, (b) Sample 2 and (c) Sample 3 (the insets in (a–c) show the corresponding BJH pore size distribution curves).

The specific surface areas were determined using Brunauer–Emmett–Teller (BET) method, the average pore sizes and pore volumes were determined by BJH analysis, and these calculated textural parameters were compiled in Table 1. From Table 1, the S_BET of 106.1 and 76.9 m² g⁻¹ were obtained for Sample 1, Sample 2 and Sample 3, respectively, which had a lower S_BET than one using NH₄HCO₃ as a precipitant (166.5 m² g⁻¹) in our previous report.⁴⁰ The average pore size and pore volume were 7.8 nm and 0.19 cm³ g⁻¹ for Sample 1, while 3.4 nm and 0.05 cm³ g⁻¹ for Sample 2. Moreover, the mesoporous CeO₂ powders synthesized using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor (Sample 3) showed a S_BET of 100.0 m² g⁻¹, the average pore size and pore volume were 3.8 nm and 0.10 cm³ g⁻¹.

Table 1 Texture parameters of the hydrothermally produced CeO₂: Sample 1 and Sample 2 synthesized using (NH₄)₂CO₃, Na₂CO₃ as precipitants, and Sample 3 synthesized using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor in the presence of H₂O₂

Sample	Precipitant		Existing cerium precursor
	(NH4)2CO3	Na2CO3	Ce2(CO3)3·xH2O
	Sample 1	Sample 2	Sample 3
SBET (m^2 g^−1)	106.1	76.9	100.0
Pore diameter (nm)	7.8	3.4	3.8
Pore volume (cm^3 g^−1)	0.19	0.05	0.10

---

In summary, the presented route for template-free synthesis of mesoporous CeO₂ powders with different S_BET was feasible, in which (NH₄)₂CO₃ or Na₂CO₃ as a precipitant was used to separate original cerium precursors (Ce₂(CO₃)₃·8H₂O or o-Ce(CO₃)OH) from Ce³⁺ aqueous solution, H₂O₂ as an oxidant was introduced to induce the phase transformation from these original cerium precursors to CeO₂ precursors, finally the mesoporous CeO₂ were obtained by following hydrothermal treatment at 200 °C for 24 h. It is worth noting that (NH₄)₂CO₃, Na₂CO₃ and H₂O₂ are common, cheap, accessible and safe chemistry reagents, which not only can save the cost, but also can reduce the pollution degree to environment. Moreover, the route, using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor for synthesis of mesoporous CeO₂, can shorten experimental processes and reduce costs, and the S_BET of the as-obtained mesoporous CeO₂ powders was 100.0 m² g⁻¹. Inspired by the template-free synthesis of mesoporous CeO₂ powders using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor, the commercial Ce(CO₃)OH should be a feasible precursor for synthesis of mesoporous CeO₂ powders. However, it is with great regret that the existing Ce(CO₃)OH precursor cannot be obtained through purchase, so the experiment with commercial Ce(CO₃)OH as an existing precursor cannot be performed. Next, the effects of H₂O₂ on the phase structures and microstructures of samples will be investigated.

Role of H₂O₂

To clarify the effects of H₂O₂ on the phase structures of samples, the XRD analyses of samples synthesized in the absence of H₂O₂ was performed. Fig. 7 shows the XRD patterns of the hydrothermally produced samples synthesized under the same conditions as control, however, in the absence of H₂O₂. From Fig. 7a and c, the hydrothermally produced Sample 1 and Sample 3 obtained in the absence of H₂O₂ showed similar XRD patterns, and both consisted of h-Ce(CO₃)OH (JCPDS no. 32-0189) and CeO₂ (JCPDS no. 34-0394). From Fig. 7b, the hydrothermally produced Sample 2 obtained in the absence of H₂O₂ consisted of Ce(CO₃)₂ (JCPDS no. 22-0542), h-Ce(CO₃)OH and CeO₂. The results of XRD analyses in Fig. 7 indicates that the pure CeO₂ cannot be synthesized in the absence of H₂O₂, which can be attributed to the missing link of the oxidation-induced phase transformation from original cerium precursors to CeO₂ precursors (see Fig. 3). In other words, the phase transformations from original cerium precursor (Ce₂(CO₃)₃·8H₂O or o-Ce(CO₃)OH) to pure CeO₂ cannot be achieved by depending upon the following hydrothermal treatment only. Combining the XRD results in Fig. 2–4, it further indicates that the link of addition of H₂O₂ acts as a relay station for CeO₂ precursors from original cerium precursors that are then forwarded to the link of hydrothermal treatment for the formation of final CeO₂ products.

Fig. 7 XRD patterns of (a) Sample 1, (b) Sample 2 and (c) Sample 3 synthesized in the absence of H₂O₂.

To understand the effects of H₂O₂ on the microstructures of precursors obtained in the absence and presence of H₂O₂, TEM analyses were performed. Fig. 8a, c and e show the TEM images of Precursor 1-1, Precursor 2-1 and Precursor 3-1 synthesized in the absence of H₂O₂, respectively. As observed, all precursors synthesized in the absence of H₂O₂ were dense. In contrast, the TEM images of precursors synthesized in the presence of H₂O₂ in Fig. 8b, d and f revealed the porous structures. The area with lower contrast showed more and clearer pores compared to one with higher contrast, and the similar phenomenon could be observed in Fig. 5a, c and e. The formation of pore structures could be explained by the oxidation-induced phase transformation from original cerium precursor (Ce₂(CO₃)₃·8H₂O or o-Ce(CO₃)OH) to CeO₂ precursor that accompanied by the evolution of porous structure. It indicates that H₂O₂ plays a key role in the formation of initial pore structures of CeO₂ precursors, which provides the precondition for the further growth of pores during the hydrothermal process (see Fig. 5).

Fig. 8 TEM images of Precursor 1-1, Precursor 2-1 and Precursor 3-1 synthesized in the absence (a, c and e) and presence (b, d and f) of H₂O₂.

From the above, it can be found that H₂O₂ plays an indispensable role in the development of pure CeO₂, which induces the phase transformation from original cerium precursors to CeO₂ precursors with initial pore structures in the aqueous solution. Interestingly, the initial pore structures are the pre-requisite for formation of final mesoporous CeO₂ products during the hydrothermal process. From a chemical perspective, the formation mechanism of the original cerium precursors with dense structures and the CeO₂ precursors with pore structures are summarized as eqn (5)–(8). In eqn (5) and (6), the original precipitate (Ce₂(CO₃)₃·8H₂O or Ce₂(CO₃)₃OH) is obtained upon the addition of (NH₄)₂CO₃ or Na₂CO₃ to Ce³⁺ aqueous solution, respectively (see Fig. 2). After adding H₂O₂, the original precipitates are oxidized, and CeO₂ precursors with low crystallinities are formed (see eqn (7) and (8)), which supported by the XRD analyses in Fig. 3. At the same time, the by-products of H₂O and CO₂ are produced. So, the phase transformation from original cerium precursors to CeO₂ precursors could be due to the oxidation of H₂O₂, while the initial pores on CeO₂ precursors (see Fig. 8b, d and f) could be attributed to the density difference between the original cerium precursors and CeO₂ precursors and the loss of by-products (H₂O and CO₂). Above all, the formation of pore structures could be essentially ascribed to the oxidation-induced phase transformation from original cerium precursors to CeO₂ precursors that accompanied by the evolution of porous structures. After addition of H₂O₂, cerium precursors are oxidized into CeO₂ and simultaneously with the formation of by-products H₂O and CO₂ as shown in eqn (7) and (8). The difference in density between cerium precursors (Ce₂(CO₃)₃·8H₂O (2.790 g cm⁻³) and o-Ce(CO₃)OH (4.545 g cm⁻³)) and CeO₂ (7.215 g cm⁻³) is the main cause for the formation of pore structures of CeO₂, while the by-product CO₂ bubbles play a stirring role, which are beneficial to the process of oxidation reaction and the homogeneity of CeO₂ particles. Moreover, the crystallinities of CeO₂ precursors could be improved and the pores grow further by following hydrothermal treatment, which supported by the XRD analyses in Fig. 4 and TEM analyses in Fig. 5.

2Ce(NO₃)₃ + 3(NH₄)₂CO₃ + H₂O = 2Ce(CO₃)OH + CO₂↑ + 6NH₄NO₃ (5)

2Ce(NO₃)₃ + 3Na₂CO₃ + 8H₂O = Ce₂(CO₃)₃·8H₂O + 6NaNO₃ (6)

Ce₂(CO₃)₃·8H₂O + H₂O₂ = 2CeO₂ + 3CO₂↑ + 9H₂O (7)

2Ce(CO₃)OH + H₂O₂ = 2CeO₂ + 2CO₂↑ + 2H₂O (8)

The S_BET of final mesoporous CeO₂ powders not only relates to the difference in density between cerium precursors and CeO₂, but also to the particle size of original cerium precursors. The phase transformation from original cerium precursors to CeO₂ precursors under the stimulation of H₂O₂ could be considered to be a diffusion process of H₂O₂. The surface of cerium precursors is first oxidized to CeO₂, these original CeO₂ grains have the tendency to aggregate with time to decrease their energy, and the hole between the grains are consider as the initial porous structures, which was the precondition for further forming of mesoporous structures of final CeO₂ powders during the hydrothermal process. However, the content of H₂O₂ decreases as the reaction progress, and the framework of cerium precursor is filled by the aqueous solution or by-product CO₂ bubbles, which could influence the diffusion of H₂O₂ from the surface to the inside of the cerium precursor framework, and then will result in the lesser porosity (see the darker areas in Fig. 8b, d, f and 5a, c, e). Moreover, the small particle sizes of cerium precursor are favorable to the diffusion of H₂O₂ from its surface to internal framework. The greater the difference in density, and the smaller its particle size, the more its S_BET. So, the S_BET of final CeO₂ products is the outcome of both the difference in density between cerium precursors and CeO₂ and the diffusion of H₂O₂ from surface to internal framework of cerium precursors. This can be used to explain why Sample 1, Sample 2, Sample 3 in this work and the CeO₂ sample in our previous report (ref. 40) possess different S_BET, even if some CeO₂ powders are synthesized with same phase of precursor.

Effect of calcination on S_BET of mesoporous CeO₂

In order to investigate the effect of calcination on the grain sizes and S_BET of samples, the hydrothermally produced mesoporous CeO₂ powders were furthermore treated by following calcination at 500 °C for 2 h, and the grain sizes were estimated using Scherrer's formula. Fig. 9a shows the effect of calcination on the grain sizes of mesoporous CeO₂ (Sample 1, Sample 2 and Sample 3, respectively). As observed, the mean grain sizes were 9.0, 4.9 and 5.7 nm for Sample 1, Sample 2 and Sample 3, respectively. After calcination, the mean grain sizes of samples increased by 14.4, 125.5 and 78.9%, which implied that the high temperature could cause the grains to grow. In addition, the hydrothermally produced CeO₂ using Na₂CO₃ as a precipitant (Sample 2) treated by calcination showed the biggest change in grain size, which could ascribed to the minimum grain size in all hydrothermally produced CeO₂ samples. Fig. 9b shows the effect of calcination on the S_BET of mesoporous CeO₂ (Sample 1, Sample 2 and Sample 3, respectively). As observed, the S_BET of samples decreased by 18.7, 43.8 and 39.4% after calcination for Sample 1, Sample 2 and Sample 3, respectively. Moreover, the hydrothermally produced CeO₂ powders using NH₄HCO₃ as a precipitant in our previous report (ref. 40) were also treated by calcination, the mean grain size increased from 5.4 to 10.9 nm with a gain of 101.8%, and the S_BET decreased from 166.5 to 105.9 m² g⁻¹ with a gain of 36.4%. The reduction of S_BET could be explained by the growing of grains or the collapse of pores during the calcination process. Obviously, the subsequent post-calcination treatment could lead to the growth of CeO₂ grains, which in turn reduced the S_BET of mesoporous CeO₂ powders.

Fig. 9 Effects of calcination on the (a) grain sizes and (b) S_BET of the hydrothermally produced mesoporous CeO₂ powders: Sample 1, Sample 2 and Sample 3 in the presence of H₂O₂ (calcination condition: 500 °C; 2 h; in air).

Adsorption properties

AO7 dye was selected as a model target to evaluate the adsorption abilities of mesoporous CeO₂ powders. Fig. 10a–c depicts the time-dependence of adsorption profiles of AO7 dye on mesoporous CeO₂ powders synthesized in the presence of H₂O₂ (Sample 1, Sample 2 and Sample 3, respectively). As observed, the adsorption efficiencies of AO7 dye achieved within 60 min were 94.2, 83.4 and 89.3% for Sample 1, Sample 2 and Sample 3, respectively. Furthermore, the adsorption of AO7 dye was rapid at the early stages, and the adsorption process was mostly completed within 40 min of reaction. The rapid adsorption of these mesoporous CeO₂ powders for AO7 dye could be ascribed to their high S_BET and pore structures. The high S_BET could provide numerous adsorption sites for AO7 molecules, while the pore structures were conducive to the transportation of AO7 molecules to CeO₂ framework and increasing the effective contact areas between CeO₂ and AO7 molecule. Interestingly, CeO₂ also can serve as an alternative photocatalyst for degradation of dye.⁴⁶ The high S_BET of mesoporous CeO₂ powders contribute to providing more active adsorption and photocatalytic reaction sites, which favor the augmentation of photocatalytic performance.⁴⁷ So, the proposed mesoporous CeO₂ powders in this work have potential to photodegrade high density dye and dye intermediate from industrial effluents. The concentrations of dyes are reduced rapidly though the rapid and remarkable adsorption of mesoporous CeO₂ powders. The reduced concentrations of dyes are benefit to increase the transmission of exciting lights, and thus enhance the intensity of the exciting lights reached the surface of CeO₂, which could improve the photocatalytic activity of CeO₂.

Fig. 10 Time-dependence of adsorption profiles of AO7 dye on mesoporous CeO₂: (a) Sample 1, (b) Sample 2 and (c) Sample 3 synthesized in the presence of H₂O₂ (T = 25 °C; [AO7] = 40 mg L; [CeO₂] = 2.0 g L; V = 100 mL; in the dark; no pH pre-adjustments).

Table 2 shows the adsorption efficiencies from the recent literatures on CeO₂ development for the adsorption of AO7 dye.^48–56 By comparing the adsorption efficiencies of CeO₂ in these reported literatures, we can find the mesoporous CeO₂ in this work showed stronger adsorption ability and achieved the absorption equilibrium more quickly, which ascribed to the higher S_BET of mesoporous CeO₂ in this work. The adsorption mode of AO7 on CeO₂ could be described as a Lewis acid–base reaction between the electron-rich groups (sulfonate group, SO³⁻) of AO7 and empty 4f orbital of cerium ion on the surface of CeO₂, which eventually formed an inner–sphere complex.^48,57 Moreover, CeO₂ could serve as an excellent adsorbent for the adsorption of other azo dyes, such methyl orange,⁵⁸ congo red,⁵⁹ alizarin red S and eriochrome black-T,⁶⁰ and the adsorption of the azo dyes onto CeO₂ was solely associated with the oxygen atoms of SO³⁻ group.⁵⁷

Table 2 Recent literatures on CeO₂ development for the adsorption of AO7 dye

Authors	Operating conditions	Adsorption efficiencies (%)	SBET (m^2 g^−1)
Cai^48 et al.	[CeO2] = 0.5 g L^−1; [AO7] = 35 mg L^−1; V = 50 mL; T = —; in the dark; no pH pre-adjustments; t = 2 h	∼23	67
Hu^49 et al.	[CeO2] = 1.0 g L^−1; [AO7] = 35 mg L^−1; V = 60 mL; at room temperature; in the dark; no pH pre-adjustments; t = 1 h	∼40	63
Arul^50,51 et al.	[CeO2] = ∼0.67 g L^−1; [AO7] = ∼105 mg L^−1; V = 15 mL; T = —; in the dark; no pH pre-adjustments; t = 10 h	Almost zero	52
Wang^52 et al.	[CeO2] = 0.5 g L^−1; [AO7] = 35 mg L^−1; V = 50 mL; T = —; in the dark; pH = 6.35; t = 1 h	44–56	40–46
Ge^53 et al.	[CeO2] = 0.5 g L^−1; [AO7] = 35 mg L^−1; V = 50 mL; T = —; in the dark; pH = 4.0; t = ∼27 h	∼50	57.5
Yao^54 et al.	[CeO2] = 8.0 g L^−1; [AO7] = 60 mg L^−1; V = 25 mL; T = 25 °C; in the dark; pH = —; t = 1 h	∼13.3	54.58
Wen^55 et al.	[CeO2] = 0.5 g L^−1; [AO7] = 40 mg L^−1; V = 20 mL; T = —; in the dark; pH = 5.0; t = 1 h	∼20	<67.8
Zang^56 et al.	[CeO2] = 0.5 g L^−1; [AO7] = 40 mg L^−1; V = 50 mL; T = 313 K; in the dark; no pH pre-adjustments; t = 1 h	12.5–37.5	—

---

The effects of AO7 initial concentration on the AO7 adsorption amount and adsorption efficiency are shown in Fig. 11. For all samples, the adsorption amount increased with increasing AO7 initial concentrations until [AO7] = 100 mg L⁻¹. In contrast, the removal efficiency decreased with increasing AO7 initial concentrations. More specifically, the removal efficiencies could reach 99.6, 99.2 and 99.5% at [AO7] = 20 mg L⁻¹ for Sample 1, Sample 2 and Sample 3 synthesized in the presence of H₂O₂, respectively.

Fig. 11 Effects of AO7 initial concentration on the AO7 adsorption efficiency and adsorption amount measured in the dark and presence of mesoporous CeO₂: (a) Sample 1, (b) Sample 2 and (c) Sample 3 synthesized in the presence of H₂O₂.

The adsorption experiments of AO7 dye at varying initial concentrations onto mesoporous CeO₂ powders were performed, and the saturated adsorption amount of AO7 dye was obtained according to Langmuir linear fits. Fig. 12a–c shows the Langmuir linear fits of experimental data of adsorption of AO7 dye onto mesoporous CeO₂ powders, and the resulting isotherm constants and correlation coefficients are presented in Table 3. From Table 3, we can see that the saturated adsorption amounts (q_m) are 378.8, 261.1 and 332.2 mg g⁻¹, and Langmuir adsorption constants (K_L) are 0.4740, 0.3460 and 0.3830 for Sample 1, Sample 2 and Sample 3, respectively. In addition, all associated correlation coefficients (R²) are greater than 0.9920, confirming that Langmuir isotherm model is a good fit for modelling the adsorption of AO7 dye onto mesoporous CeO₂ surface. The results indicate that the proposed route for template-free synthesis of mesoporous CeO₂ powders is one marker of success to effectively and rapidly remove AO7 dye.

Fig. 12 Langmuir linear fits of AO7 dye adsorbed onto mesoporous CeO₂: (a) Sample 1, (b) Sample 2 and (c) Sample 3 synthesized in the presence of H₂O₂.

Table 3 Relevant parameters of Langmuir fitting for mesoporous CeO₂: Sample 1 and Sample 2 synthesized using (NH₄)₂CO₃, Na₂CO₃ as precipitants, and Sample 3 synthesized using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor in the presence of H₂O₂

Sample			Langmuir isotherm model
			qm (mg g^−1)	KL (L mg^−1)	R^2
Precipitant	(NH4)2CO3	Sample 1	378.8	0.4740	0.9929
	Na2CO3	Sample 2	261.1	0.3460	0.9951
Existing precursor	Commercial Ce2(CO3)3·xH2O	Sample 3	332.2	0.3830	0.9937

---

Conclusions

The accessible, cheap and safe chemistry reagents (NH₄)₂CO₃, Na₂CO₃ and H₂O₂ were employed for template-free synthesis of mesoporous CeO₂ powders with high BET surface areas. (NH₄)₂CO₃ or Na₂CO₃ as a precipitant was used to separate original cerium precursors from Ce³⁺ aqueous solution, while H₂O₂ served as an oxidant to induce the phase transformation from original cerium precursors to CeO₂ precursors with initial porous structures, which was the precondition for the formation of final CeO₂ phase and mesoporous structures during the following hydrothermal process at 200 °C for 24 h. The BET surface areas of mesoporous CeO₂ powders synthesized using (NH₄)₂CO₃ and Na₂CO₃ as precipitants were 106.1 and 76.9 m² g⁻¹. Moreover, another route, using commercial Ce₂(CO₃)₃·xH₂O as existing precursor for synthesis of mesoporous CeO₂ powders with a BET surface area of 100.0 m² g⁻¹, can shorten experimental processes and reduce costs. These mesoporous CeO₂ powders could be used as a suitable sorbent for rapid and effective removal of AO7 dye. Moreover, the saturated adsorption amounts could reach up to 378.8, 261.1 and 332.2 mg g⁻¹ without pH pre-adjustments for these mesoporous CeO₂ powders using (NH₄)₂CO₃, Na₂CO₃ as precipitants and using commercial Ce₂(CO₃)₃·xH₂O as an existing precursor, respectively. Prompted by the high BET surface area, low cost, environmental friendliness and omissible calcination process, these mesoporous CeO₂ powders synthesized with the routes in this work could be a promising candidate for practical application. In subsequent study, the optimization of experimental parameters will be explored, such as the additive amount of H₂O₂, hydrothermal treatment temperature and time, and so on.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors appreciate the financial support from Leshan Normal University Research Program, China (No. Z16024), Science and Technology Bureau of Leshan city, China (No. 17GZD051).

Notes and references

1. K. Vinodgopal, J. Peller, O. Makogon and P. V. Kamat, Water Res., 1998, 32, 3646–3650.
2. L. Y. Liu, M. Pu, L. Yang, D. G. Evans and J. He, Mater. Chem. Phys., 2007, 106, 422–427.
3. S. Momeni and D. Nematollahi, Sci. Rep., 2017, 7, 41963.
4. E. Daneshvar, M. Kousha, N. Koutahzadeh, M. S. Sohrabi and A. Bhatnagar, Environ. Prog. Sustainable Energy, 2013, 32, 285–293.
5. K. Vinodgopal, J. Peller, O. Makogon and P. V. Kamat, Water Res., 1998, 32, 3646–3650.
6. H. Zollinger, Color Chemistry: Synthesis, Properties and Applications of Organic Dyes and Pigments, VCH Publishers, New York, 1987.
7. J. Li, P. Ye, J. Fang, M. Wang, D. Wu, A. Xu and X. Li, Appl. Surf. Sci., 2017, 422, 754–762.
8. W. Wang, G. Huang, C. An, X. Xin, Y. Zhang and X. Liu, Appl. Surf. Sci., 2017, 405, 119–128.
9. X. Li, W. Guo, Z. Liu, R. Wang and H. Liu, Appl. Surf. Sci., 2016, 369, 130–136.
10. R. G. Saratale, G. D. Saratale, J. S. Chang and S. P. Govindwar, J. Taiwan Inst. Chem. Eng., 2011, 42, 138–157.
11. W. E. Thung, S. A. Ong, L. N. Ho, Y. S. Wong, F. Ridwan, H. K. Lehl, Y. L. Oon and Y. S. Oon, Chem. Eng. J., 2018, 336, 397–405.
12. X. Wu, H. He, W. L. Yang, J. Yu and C. Yang, Appl. Microbiol. Biotechnol., 2018, 102, 7597–7610.
13. K. M. Reza, A. Kurny and F. Gulshan, Appl. Water Sci., 2017, 7, 1569–1578.
14. S. L. Lee, L. N. Ho, S. A. Ong, Y. S. Wong, C. H. Voon, W. F. Fhalik, N. A. Yusoff and N. Nordin, Chemosphere, 2018, 209, 935–943.
15. S. Wu, H. Li, X. Li, H. He and C. Yang, Chem. Eng. J., 2018, 353, 533–541.
16. M. Minière, O. Boutin and A. Soric, Can. J. Chem. Eng., 2018 DOI:10.1002/cjce.23195.
17. C. Zhang, W. Chen, J. Xian and D. Fu, RSC Adv., 2018, 8, 3934–3940.
18. Q. Qiao, S. Singh, S. L. Lo, Y. Li, J. Jin and L. Wang, J. Taiwan Inst. Chem. Eng., 2018, 84, 110–122.
19. J. Li, H. Lin, K. Zhu and H. Zhang, Chemosphere, 2017, 188, 139–147.
20. G. Li, W. Zhao, B. Wang, Q. Gu and X. Zhang, Desalin. Water Treat., 2016, 57, 2167–2174.
21. H. Oualid and M. Slimane, Water Environ. Res., 2017, 89, 250–259.
22. M. Yusuf, M. A. Khan, M. Otero, E. C. Abdullah, M. Hosomi, A. Terada and S. Riya, J. Colloid Interface Sci., 2017, 493, 51–61.
23. W. Wang, G. Huang, C. An, S. Zhao, X. Chen and P. Zhang, J. Cleaner Prod., 2018, 172, 1986–1997.
24. X. Inthapanya, S. Wu, Z. Han, G. Zeng, M. Wu and C. Yang, Environ. Sci. Pollut. Res., 2019, 26, 5944–5954.
25. M. L. Ma, J. H. Qin, C. Ji, H. Xu, R. Wang, B. J. Li and S. R. Batten, J. Mater. Chem. C, 2014, 2, 1085–1093.
26. Y. Huang, C. Yang, Z. Sun, G. Zeng and H. He, RSC Adv., 2015, 5, 11475–11484.
27. L. Lu, Y. Lin, Q. Chai, S. He and C. Yang, Colloids Surf., A, 2018, 558, 103–109.
28. S. Wu, H. He, X. Inthapanya, C. Yang, L. Lu, G. Zeng and Z. Han, Environ. Sci. Pollut. Res., 2017, 24, 16560–16577.
29. Q. Wei, Q. Ma, P. Zuo, H. Fan, S. Qu and W. Shen, Chemcatchem, 2018, 10, 1019–1026.
30. H. Zhu, Y. Chen, Z. Wang, W. Liu and L. Wang, RSC Adv., 2018, 8, 14888–14897.
31. K. Li, Y. Zhao, C. Song and X. Guo, Appl. Surf. Sci., 2017, 425, 526–534.
32. L. Wang and H. Liu, Catal. Today, 2018 DOI:10.1016/j.cattod.2018.04.015.
33. S. Zhan, H. Zhang, Y. Zhang, Q. Shi, Y. Li and X. Li, Appl. Catal., B, 2017, 203, 199–209.
34. Y. Wang, D. Gao, C. Li, C. Li, F. Rosei, D. Ma and G. Chen, Part. Part. Syst. Charact., 2018, 35, 1700367.
35. C. Ni, X. Li, Z. Chen, H.-Y. H. Li, X. Jia, I. Shah and J. Q. Xiao, Microporous Mesoporous Mater., 2008, 115, 247–252.
36. N. Nabih, R. Schiller, I. Lieberwirth, E. Kockrich, R. Frind, S. Kaskel, C. K. Weiss and K. Landfester, Nanotechnology, 2011, 22, 135606.
37. J. Wei, Z. Yang, H. Yang, T. Sun and Y. Yang, CrystEngComm, 2011, 13, 4950–4955.
38. A. Xie, W. Liu, S. Wang, X. Liu, J. Zhang and Y. Yang, Mater. Res. Bull., 2014, 59, 18–24.
39. Y. He, S. Du, J. Li, R. Zhang, X. Liang and B. Chen, Chemcatchem, 2017, 9, 4070–4082.
40. Y. Xu and R. Li, RSC Adv., 2015, 5, 44828–44834.
41. D. F. Swinehart, J. Chem. Educ., 1962, 39, 333–335.
42. S. Duran, D. Şolpan and O. Güven, Nucl. Instrum. Methods, 1999, 151, 196–199.
43. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361–1403.
44. Y. Lin, S. Wu, C. Yang, M. Chen and X. Li, Appl. Catal., B, 2019, 245, 71–86.
45. J. C. Groen, L. A. A. Peffer and J. Pérez-Ramírez, Microporous Mesoporous Mater., 2003, 60, 1–17.
46. D. Sun, M. Gu, R. Li, S. Yin and T. Sato, Appl. Surf. Sci., 2013, 280, 693–697.
47. Y. Lin, S. Wu, X. Li, X. Wu, C. Yang, G. Zeng, Y. Peng and L. Lu, Appl. Catal., B, 2018, 227, 557–570.
48. W. Cai, F. Chen, X. Shen, L. Chen and J. Zhang, Appl. Catal., B, 2010, 101, 160–168.
49. S. Hu, F. Zhou, L. Wang and J. Zhang, Catal. Commun., 2011, 12, 794–797.
50. N. S. Arul, D. Mangalaraj, P. C. Chen, N. Ponpandian, P. Meena and Y. Masuda, J. Sol-Gel Sci. Technol., 2012, 64, 515–523.
51. N. S. Arul, D. Mangalaraj, T. W. Kim, P. C. Chen, N. Ponpandian, P. Meena and Y. Masuda, J. Mater. Sci.: Mater. Electron., 2013, 24, 1644–1650.
52. Y. Wang, X. Shen and F. Chen, J. Mol. Catal. A: Chem., 2014, 381, 38–45.
53. L. Ge, C. Zang and F. Chen, Chin. J. Catal., 2015, 36, 314–321.
54. H. Yao, X. Ding, Z. Wang, F. Zhang, Y. Wang and G. Luo, RSC Adv., 2016, 6, 112413–112419.
55. T. Wen, Y. Tang, F. Chen and B. Mo, Arch. Environ. Prot., 2016, 42, 12–19.
56. C. Zang, X. Zhang, S. Hu and F. Chen, Appl. Catal., B, 2017, 216, 106–113.
57. P. Ji, J. Zhang, F. Chen and M. Anpo, Appl. Catal., B, 2009, 85, 148–154.
58. Z. Cui, J. Zhang, Y. Xue and H. Duan, Langmuir, 2018, 34, 3197–3206.
59. Z. Yang, J. Wei, H. Yang, L. Liu, H. Liang and Y. Yang, Eur. J. Inorg. Chem., 2010, 2010, 3354–3359.
60. S. Mishra, S. Soren, A. K. Debnath, D. K. Aswal, N. Das and P. Parhi, Optik, 2018, 169, 125–136.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ra02294e

---