Immobilization of cadmium ions to synthesis hierarchical flowerlike cadmium phosphates microspheres and their application in the degradation of organic pollutants under light irradiation

Abstract

The treatment of wastewater especially that contaminated by heavy metals and/or organic pollutants by a green, cost-efficient and robust route is highly desirable. Herein, by addition of phosphates to immobilize cadmium ions, hierarchical flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres were successfully prepared by nanosheet formation and following self-assembly at room temperature without additive assistance. The influence of experimental parameters such as pH value and raw materials (cadmium ions and phosphate ions) on the morphology and crystal structure of the products was studied. Hierarchical Cd₅(PO₄)₂P₂O₇ was readily obtained by calcination of the Cd₅H₂(PO₄)₄·4H₂O precursor. The microstructures of the products were characterized by X-ray power diffraction (XRD), field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), Fourier transform infrared (FTIR), N₂ adsorption–desorption (BET), and UV-vis diffuse reflectance spectroscopy (DRS). The as-prepared Cd₅H₂(PO₄)₄·4H₂O and Cd₅(PO₄)₂P₂O₇ possessed a wide band gap energy of 5.56 eV and ca. 5.00 eV, respectively. Owing to the strong oxidation/reduction ability and the unique hierarchical structure, hierarchical Cd₅(PO₄)₂P₂O₇ microspheres exhibited excellent photocatalytic activity and durability for the degradation of organic pollutants (rhodamine B, RhB) under UV light irradiation. The main reactive species and the photocatalytic mechanism of Cd₅(PO₄)₂P₂O₇ towards RhB degradation were also discussed.

---

Introduction

Photocatalytic oxidation technology has been proven to be potentially advantageous for environmental remediation such as wastewater treatment and air purification with solar energy.¹ Over the past several years, considerable efforts have been devoted to acquire high photocatalytic performance materials through exploring novel photocatalytic materials or modifying reported photocatalysts. More than 150 photocatalysts including oxides, hydroxides, sulfides, nitrides, and halides have been reported to be able to degrade pollutants under light irradiation.² Nevertheless, some photocatalysts show low activity and poorly structural stability, while others possess weak photooxidation ability toward the elimination of contaminants.³ Thus, developing new efficient and stable photocatalysts is still a great challenge in the photocatalysis field. Additionally, it has been shown that photocatalytic materials with three-dimensional (3D) hierarchical micro/nanostructures usually exhibit high efficiency and have specificity (i.e., good recyclability and high stability) for degradation of pollutants because of their unique dimensional and structural characteristics different from mono-morphological structures.^4–6 Up to now, remarkable progress in the controllable fabrication of hierarchical structures with different shapes has been achieved. Many hierarchical superstructures with different composition, including elements, metal oxides, and metal sulfides were successfully synthesized via various methods.^7–10 Although template-directed synthesis has been demonstrated as a versatile approach to fabricate hierarchical structures, the wealth of templates including hard templates and soft templates complicated the synthetic procedures and usually brought in harmful substrates to the environment and human health.¹¹ Thus, the development for template-free and environmentally benign methods to synthesize hierarchical structures is still an ongoing progress. More recently, self-assembly of low-dimensional building blocks into complex 3D uniform micro/nanostructures on the basis of some intrinsically physical phenomena, such as the Kirkendall effect, Ostwald ripening, and oriented attachment process, to fabricate hierarchical structures provides new alternative routes for template-free synthesis of hierarchical materials.^12–15

Metal phosphates, as one of important inorganic materials, have attracted the widespread attention recent decades because of their potential applications in the fields of ion-exchange, proton conductivity, biocompatibility, adsorption and catalysis.^16–18 Various hierarchical metal phosphates have been controllably prepared and shown interesting properties. For instance, hydroxyapatite (Ca(PO₄)₆(OH)₂), a typically bioactive and biocompatible material, was fabricated with various hierarchical microstructures via simple hydrothermal method in the absence of any surfactants.^19–21 The as-synthesized hydroxyapatite exhibited highly structural stability (up to 800 °C), high efficiency in protein/heavy metal adsorption, and enzymatic catalysis.^21–24 Wang et al. synthesized flowerlike titanium phosphate (Ti(HPO₄)₂·H₂O) crystals and investigated their properties in removal of lead ions.²¹ Hierarchical iron hydrogen phosphate (Fe₃H₉(PO₄)₆·6H₂O) and strontium flourapatite (Sr₅(PO₄)₃OH_1−xF_x) were also prepared and shown special application in wastewater treatment and H₂O₂ sensing.^25,26 In particular, metal phosphates have been found to be good photocatalysts with strong photocatalytic ability to decompose organic contaminants and split water due to the inductive effect of PO₄³⁻, which favors the separation efficiency of electrons–holes pairs in photocatalysis.²⁷ Numerous metal phosphate materials such as hydroxyapatite (Ca(PO₄)₆(OH)₂, TiO₂/Ca(PO₄)₆(OH)₂, Ag₃PO₄/Ca(PO₄)₆(OH)₂), BiCu₂PO₆, BiPO₄, Ag₃PO₄, Cu₂(OH)PO₄, and Ti₂O(PO₄)₂(H₂O)₂ have been explored as novel photocatalysts.^27–34 Among them, cadmium phosphate was found to possess excellent adsorption capacities for Pb(II) ions and novel catalytic property for the dehydrogenation of alcohols.^35,36 Its nucleation and crystal growth kinetics based on mass crystallization experiments was also investigated.³⁷ However, to the best of our knowledge, there have been seldom reports on the controllable preparation of hierarchical cadmium phosphate and their photocatalytic application.

Herein, hierarchical flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres assembled with nanosheets were successfully prepared by addition of phosphates to cadmium ions solution at room temperature without additives assistant. Various cadmium ions including CdCl₂, Cd(NO₃)₂, and Cd(CH₃COO)₂ could be immobilized by such route to form hierarchical structures. Cd₅(PO₄)₂P₂O₇ with similar hierarchical structures was readily obtained by calcination of Cd₅H₂(PO₄)₄·4H₂O and was firstly applied as novel photocatalyst for dye wastewater treatment. Such hierarchical Cd₅(PO₄)₂P₂O₇ showed excellent photocatalytic efficiency and structural stability during the photocatalytic process. The main active radicals during dye degradation and the possible photocatalytic mechanism were also investigated.

Experimental section

Preparation

Hierarchical cadmium phosphate precursor, Cd₅H₂(PO₄)₄·4H₂O, was fabricated via a simple precipitation method without the assistance of any template or surfactant. All used analytical grade materials were purchased from the Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China) and were used without further purification. In a typical process, 1.8 g Na₂HPO₄·12H₂O and 1.6 g Cd(CH₃COO)₂·2H₂O were dissolved in 100 mL of deionized water, respectively. Then the Na₂HPO₄ aqueous solution was added dropwise to the Cd(CH₃COO)₂ aqueous solution under stirring within 30 min. The mixture was stirred for 1 h and the resultant white precipitates were collected by filtration, washed with distilled water repeatedly, and dried at 60 °C overnight. The as-obtained Cd₅H₂(PO₄)₄·4H₂O sample was obtained and labeled as T0. The Cd₅H₂(PO₄)₄·4H₂O sample was heated in a tube furnace at 400, 500 and 700 °C for 4 h in air to yield Cd₅(PO₄)₂P₂O₇, which was respectively labeled as T400, T500, and T700.

Sample characterizations

The crystal phase of the as-synthesized samples was examined by X-ray diffraction (XRD) on a Rigaku MinFlex II equipped with Cu Kα irradiation (λ = 0.15418 nm). Morphologies of the samples were observed by field emission scanning electron microscope (FESEM, JSM-6700F). Selected area electron diffraction (SAED) and transmission electron micrograph (TEM) were recorded on a JEM-2010. Fourier transform infrared spectra (FTIR) of the samples were recorded on a Perkin-Elmer IR spectrophotometer using a KBr pellet technique. The thermal behavior of the samples was investigated by thermogravimetric analysis using a Netzsch thermoanalyzer STA449C at a heating rate of 15 °C min⁻¹ in air atmosphere. The UV-vis reflection spectra (DRS) were recorded on a Varian Cary 500 UV-vis spectrophotometer using BaSO₄ as a reference and converted from reflection to absorbance by the Kubelka–Munk method. The specific surface areas of the samples were measured by nitrogen adsorption–desorption isotherms at 77 K using an automatic ASAP-2020 micromeritics analyzer. A desorption isotherm was used to determine the pore size distribution using the Barrett–Joyner–Halenda (BJH) method. Before each measurement, each sample was degassed for 6 h in helium at 333 K. The Cd²⁺ content of the residual solution after photocatalytic reaction was carried out by inductively coupled plasma (ICP) spectrometry (Ultima2).

Evaluation of photocatalytic activity

For UV-light photocatalytic activity, four UV lamps (Philips, TUV 4W/G4 T5) with wavelength centers at 254 nm were used. 100 mg powder was suspended in 100 mL rhodamine B (RhB) aqueous solution (10 ppm). Prior to irradiation, the suspensions were magnetically stirred in the dark for 30 min to attain the adsorption/desorption equilibrium between RhB and catalysts at ambient conditions. At varied irradiation time intervals, 3 mL suspension was collected and analyzed on a Perkin-Elmer UV WinLab Lambda 35 spectrophotometer. The degradation percentage is reported as C/C₀, where C₀ is the initial concentration of RhB, and C represents the corresponding concentration at a certain time interval. Photocatalytic stability was performed as follows: after each degradation experiment, the suspension was filtered and the solids were washed with distilled water and dried at 60 °C in air. Then the regenerated product was employed to degrade a new dye aqueous solution for another test under the same UV light irradiation. The generation of hydroxyl radicals (˙OH) was investigated using the PL (F-4600 type) technique, in which a basic terephthalic acid (TA) solution including 5 × 10⁻³ M TA and 0.01 M NaOH was added to the reactor. The excitation wavelength was 312 nm. The detection of ˙O₂⁻ radicals was performed by a nitroblue tetrazolium (NBT) probe method according to the literature.³⁸

Results and discussion

The synthesis of Cd₅H₂(PO₄)₄·4H₂O hierarchical architectures was achieved by simply mixing cadmium ions solution and phosphate ions solution at room temperature without additives assistant. The morphology and composition of the as-prepared samples were firstly examined by SEM and TEM. The low-magnification SEM image in Fig. 1a shows that the Cd₅H₂(PO₄)₄·4H₂O products are mainly composed of numerous of flowerlike 3D microspheres with diameters ranging from 3 to 6 μm. Closer observations by the higher magnification SEM image in Fig. 1b and TEM in Fig. 1c and d expose that each flowerlike microsphere is composed of dozens of nanosheets with thickness of about 30 nm. These nanosheets as building blocks are connected to each other to build the flowerlike architectures. The selected-area electron diffraction (SAED) pattern (Fig. 1e) taken on a single nanosheet suggests the single-crystalline nature of the nanosheet unit.

Fig. 1 (a and b) FESEM images (c and d) TEM images with different magnifications, and (e) SAED pattern of the as-synthesized Cd₅H₂(PO₄)₄·4H₂O precursor.

XRD, shown in Fig. 2a, was used to study the crystal structure and phase composition of the obtained precursor. According to the main diffraction peak locations, the sample could be easily indexed to Cd₅H₂(PO₄)₄·4H₂O, which matched well with the standard pattern (JCPDS card no. 14-0400). The low-angle reflection at 10.6° with strong diffraction peak confirms the presence of interlayer spacing in the lamellar structure.¹⁰ In addition, the FTIR of the Cd₅H₂(PO₄)₄·4H₂O precursor is shown in Fig. 2b. The broad band from 3000 to 3500 cm⁻¹ can be assigned to the O–H stretching vibrations while the sharp peak located at 1625 cm⁻¹ corresponds to the bending mode vibration of water molecules.³⁹ The observation of two shoulders at about 2440 and 1237 cm⁻¹ can be associated to the (P)–O–H stretching modes and the deformation of P–O–(H) in the plane, respectively.^35,39 Considering the studies of other phosphate compounds, the vibrations of the P–O bonds are also observed from about 900 to 1200 cm⁻¹ (ν_as(P–O)) and from 520 to 600 cm⁻¹ (δ_as(P–O)).³⁹ As for the peaks at 587, 671, and 735 cm⁻¹, they can be related to the stretching vibration of Cd–O bonds, which show obvious shifts as compared to the Cd–O bonds in pure CdO.⁴⁰ The shift of these characteristic bands confirms the incorporation of phosphorus into the framework in the form of Cd–O–P bonds.⁴¹

Fig. 2 (a) XRD pattern and (b) FTIR spectrum of the as-synthesized Cd₅H₂(PO₄)₄·4H₂O precursor.

To understand the formation mechanism of the hierarchical Cd₅H₂(PO₄)₄·4H₂O microspheres, time-dependent experiments were performed, in which intermediate products were collected at different intervals once Na₂HPO₄ was added into the Cd(CH₃COO)₂ solution media. All of the intermediate products were monitored by SEM and the results are shown in Fig. 3. At the initial stage (1 min) with little Na₂HPO₄ adding, the obtained precursors were ultrafine nanoparticles with size lower than 100 nm (Fig. 3a). As the reaction proceeded (5 min) and more Na₂HPO₄ was added, the primary nanoparticles tended to aggregate together to form microscaled spheres (Fig. 3b). It was found that some hierarchical structures assembled with connected nanosheets were observed when Na₂HPO₄ was completely added after 30 min of reaction (Fig. 3c). However, the hierarchical structures in this case were partial or incomplete. Uniformly flowerlike structures were completely formed after 3 h of reaction (Fig. 3d). For clarity, the morphological evolution process of the flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres is illustrated in Fig. 3e, which involves a nucleation–aggregation–ripening process.¹⁵ Ultrafine Cd₅H₂(PO₄)₄·4H₂O nanoparticles were produced by the reaction between HPO₄²⁻ and Cd²⁺ at room temperature. Meanwhile, the freshly formed monomers were unstable due to their high surface energy and aggregated together gradually, resulting in the microscaled spheres. These monomers located in the Cd₅H₂(PO₄)₄·4H₂O microspheres further underwent a dissolution–recrystallization (Ostwald ripening) process to form the nanosheets, which were further assembled into hierarchical Cd₅H₂(PO₄)₄·4H₂O microspheres.

Fig. 3 SEM images of the nanostructures collected at different reaction time once the Na₂HPO₄ was added: (a) 1 min, (b) 5 min, (c) 30 min, and (d) 3 h. (e) A schematic illustration of the formation process for hierarchical Cd₅H₂(PO₄)₄·4H₂O microspheres.

Control experiments were conducted to reveal suitable reaction condition for the preparation of flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres. As shown from Fig. 4a and b, flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres could be prepared when other cadmium ions such as Cd(NO₃)₂ or CdCl₂ were used as starting material instead of Cd(CH₃COO)₂. The XRD patterns of the collected samples show that they have the composition of Cd₅H₂(PO₄)₄·4H₂O. However, when NaH₂PO₄ was replaced by Na₂HPO₄ or Na₃PO₄ (Fig. 4c and d), novel hierarchical Cd₅H₂(PO₄)₄·4H₂O microspheres and irregular nanoparticles with amorphous phase were observed, respectively. The change of morphology and crystal phase of the final products may be explained by the release of H⁺ ions to solution from various phosphate anion species H₂PO₄⁻, HPO₄²⁻ and PO₄³⁻, while PO₄³⁻ ions were consumed for cadmium phosphate formation.^42,43 Thus, we also investigated the influence of pH value on the final products. It was found that the optical pH values for the formation of flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres were between 5 and 7. A slightly decrease in pH value to about 4 by addition of a small amount of HNO₃ solution contributed to Cd₅H₂(PO₄)₄·4H₂O microspheres assembled by large nanoplates with width of ca. 300 nm (Fig. 4e). These large nanoplates were arranged compactly to form the 3D microspheres and no interleaved pores were constructed in this condition. On the other hand, when the pH value was increased above 9 by addition of NaOH solution, the obtained products were irregular nanoparticles with amorphous characteristics (Fig. 4f), supported by the weak diffraction peaks as shown in the XRD pattern. From the above results, it can be concluded that by simply tuning the phosphates or the pH value, hierarchical Cd₅H₂(PO₄)₄·4H₂O microspheres could be easily fabricated via in situ immobilization of cadmium ions in solution, which provides a simple and environmental attractive remediation method for wastewater contaminated by cadmium ions.

Fig. 4 SEM images and XRD patterns of the obtained products collected from various conditions: (a) using CdCl₂ as cadmium source and Na₂HPO₄ as phosphate source, (b) using Cd(NO₃)₂ as cadmium source and Na₂HPO₄ as phosphate source, (c) using Cd(CH₃COO)₂ as cadmium source and NaH₂PO₄ as phosphate source, (d) using Cd(CH₃COO)₂ as cadmium source and Na₃PO₄ as phosphate source, (e) using Cd(CH₃COO)₂ as cadmium source and Na₂HPO₄ as phosphate source at pH value of 4, and (f) using Cd(CH₃COO)₂ as cadmium source and Na₂HPO₄ as phosphate source at pH value of 9.

As a typical metal hydrogen phosphate, Cd₅H₂(PO₄)₄·4H₂O may also be written as Cd₅(PO₄)₂(HPO₄)₂·4H₂O, in which HPO₄ and H₂O entities are highly temperature-dependent, suffering from condensation and dehydration reaction, respectively.⁴⁴ Thus, the effect of thermal treatment on microstructures of Cd₅H₂(PO₄)₄·4H₂O precursor was further investigated. The TG-DSC curve of Cd₅H₂(PO₄)₄·4H₂O is initially shown in Fig. 5a. It displays two main steps of weight loss in the range of 30–900 °C. The first weight loss of 7.4% below 300 °C can be attributed to the loss of four molecule crystal water of Cd₅H₂(PO₄)₄·4H₂O leading to Cd₅H₂(PO₄)₄. An addition weight loss of 1.8% is observed in the range of 300–600 °C, corresponding to the loss of one molecule of water from the condensation reaction of HPO₄ groups, which seems to take place leading to the formation of Cd₅(PO₄)₂P₂O₇, analogous to U₂(PO₄)₂HPO₄·H₂O.⁴⁴ No significant mass loss was detected above 600 °C. The XRD patterns (Fig. 5b) indicate that the resulted samples after calcination at 400, 500, and 700 °C are pure Cd₅(PO₄)₂P₂O₇ (JCPDS card no. 14-0399). With increasing the calcination temperature, XRD patterns become sharper, which indicates the increased crystallinity and grain growth as well. The morphology of the obtained Cd₅(PO₄)₂P₂O₇ samples was examined by SEM and TEM. As seen from Fig. 6a–d, the hierarchical structure was still retained upon calcination at 400 °C. However, due to the condensation and dehydration process, the single-crystalline nanosheets converted into many smaller nanoparticles along with numerous pores with size of 10–150 nm were generated within the nanosheets. The corresponding SAED pattern (Fig. 6e) depicts that the nanosheet assembled from nanoparticles is polycrystalline in nature. Further increasing calcination temperature to 500 °C caused the grain growth and particle aggregation (Fig. 6f). When the temperature increased to 700 °C, the nanosheets were completely converted into large nanoparticles with size of ca. 200 nm. The hierarchical flowers finally changed to irregular spheres and no pores were observed due to the aggregation of particles (Fig. 6g).

Fig. 5 (a) TG-DSC curves of the Cd₅H₂(PO₄)₄·4H₂O precursor, and (b) XRD patterns of the obtained Cd₅(PO₄)₂P₂O₇ samples.

Fig. 6 SEM (a and b), TEM (c and d) and SAED pattern (e) of the Cd₅(PO₄)₂P₂O₇ samples (T400); SEM image (f) of the Cd₅(PO₄)₂P₂O₇ sample (T500); and SEM image (g) of the Cd₅(PO₄)₂P₂O₇ sample (T700).

The optical properties of the Cd₅H₂(PO₄)₄·4H₂O precursor and the calcined Cd₅(PO₄)₂P₂O₇ were measured using UV-vis diffuse reflectance spectra. Fig. 7a shows the UV absorption spectra of the samples, indicating that the maximal absorbance wavelengths of Cd₅H₂(PO₄)₄·4H₂O and Cd₅(PO₄)₂P₂O₇ (T400, T500, and T700) are approximately 233 nm, 255 nm, 266 nm and 268 nm, respectively. The band gap energy (E_g) was evaluated using the equation, α(hν) = A(hν − E_g)^n/2, where α, h, E_g, and A are the absorption coefficient, light frequency, band gap energy, and a constant, respectively, and n is determined by the type of optical transition in the semiconductor. The band gap energy (E_g values) of these samples was estimated from a plot of α(hν)² as a function of the photon energy (hν) (Fig. 7b) to be approximately 5.56 eV, 5.04 eV, 4.97 eV, and 4.95 eV for Cd₅H₂(PO₄)₄·4H₂O and Cd₅(PO₄)₂P₂O₇ samples (T400, T500, and T700), respectively.

Fig. 7 (a) UV-vis diffuse reflection spectra of the as-synthesized Cd₅H₂(PO₄)₄·4H₂O precursor (T0) and the calcined Cd₅(PO₄)₂P₂O₇ samples (T400, T500, and T700), and (b) the plots of (αhν)² as a function of photon energy (hν).

The specific surface area and porosity of the hierarchical Cd₅H₂(PO₄)₄·4H₂O and Cd₅(PO₄)₂P₂O₇ were measured by N₂ adsorption–desorption isotherms. As shown in Fig. 8a, the isotherm of the Cd₅H₂(PO₄)₄·4H₂O precursor exhibits a hysteresis loop in a relative pressure range of 0.6 to 1.0, implying the presence of mesopores in Cd₅H₂(PO₄)₄·4H₂O hierarchical structures. The corresponding pore-size distribution (insert in Fig. 8a) shows that most of the pores fall into the size region from 5 to 80 nm. The mesopores are formed most likely due to the intercrossing of nanosheets.⁴⁵ The BET surface area of the Cd₅H₂(PO₄)₄·4H₂O precursor was determined to be 7.7 m² g⁻¹. Note that after 400 °C calcination, similar porosity and more broader pore-size distribution (from 5 to 130 nm) were observed on Cd₅(PO₄)₂P₂O₇ (T400). The BET surface area of the Cd₅(PO₄)₂P₂O₇ sample (T400) was calculated to be 7.4 m² g⁻¹. The broad pore-size distribution for Cd₅(PO₄)₂P₂O₇ (T400) could be attributed to the generated pores within the original single-crystalline nanosheets (Fig. 6a–d). However, since higher temperature caused the growth and aggregation of particles (Fig. 6f and g), no obvious pores were found in the corresponding Cd₅(PO₄)₂P₂O₇ samples (T500 and T700) (Fig. S1†). The BET surface area of the Cd₅(PO₄)₂P₂O₇ samples (T500 and T700) was also reduced to 4.5 and 1.6 m² g⁻¹, respectively.

Fig. 8 Nitrogen adsorption–desorption isotherms of (a) Cd₅H₂(PO₄)₄·4H₂O (T0) and (b) Cd₅(PO₄)₂P₂O₇ (T400). The inset is the corresponding pore-size distribution.

Degradation of RhB was examined as a probe reaction to evaluate the photocatalytic activities of the as-prepared samples under UV light irradiation. The degradation curves of RhB in the presence/absence of photocatalysts as a function of irradiation time are plotted in Fig. 9a. The blank experiment without photocatalyst shows that about 36% of RhB was decomposed within 100 min by the photolysis due to the strong energy of UV light. The addition of Cd₅H₂(PO₄)₄·4H₂O precursor into RhB system gave rise to a negligible enhancement in RhB degradation, suggesting that Cd₅H₂(PO₄)₄·4H₂O with band gap of 5.56 eV can not be excited by the used UV light to generate photoinduced electron–hole pairs. However, Cd₅H₂(PO₄)₄·4H₂O can act as supporting materials by grafting active semiconductor (such as Ag₃PO₄) to show superior photocatalytic property under the irradiation of visible light (Fig. S2†). In contrast, Cd₅(PO₄)₂P₂O₇ sample after 400 calcination exhibited an excellent photocatalytic activity for RhB degradation under UV light irradiation. After 100 min irradiation, about 93% of RhB was decomposed and the mineralization yield of RhB reached 100% from TOC measurement. It has also been found that the photocatalytic degradation of RhB in the presence/absence of different photocatalysts followed the pseudo-first-kinetics model, ln(C_t/C₀) = −kt, where C₀ and C_t are the initial concentration of the dye solution and the concentration at time t, respectively, and k is the kinetic constant. As displayed in Fig. 9b, the Cd₅(PO₄)₂P₂O₇ (T400) has much higher rate constant (ca. 0.02729 min⁻¹) than Cd₅H₂(PO₄)₄·4H₂O (ca. 0.006 min⁻¹) and is approximately 6.4 times higher than that of the photolysis (ca. 0.00427 min⁻¹). Considering the perniciousness of Cd²⁺ in Cd₅H₂(PO₄)₄·4H₂O and Cd₅(PO₄)₂P₂O₇, we also tested the Cd²⁺ content of the residual solution after photocatalytic reaction by ICP. It is found that no Cd²⁺ was dissolved out from these two materials, suggesting the high stability of these hierarchical cadmium phosphates.

Fig. 9 (a) Photocatalytic degradation curves of RhB as a function of the irradiation time in the presence/absence of different photocatalysts, (b) the kinetics over different photocatalysts.

For comparison, the photocatalytic activity of the hierarchical flowerlike Cd₅(PO₄)₂P₂O₇ was also compared with other reported phosphate photocatalysts such as BiPO₄, Ca(PO₄)₆(OH)₂ and Ti₂O(PO₄)₂(H₂O)₂. As shown in Fig. 10, the activity of Cd₅(PO₄)₂P₂O₇ was much higher than that of Ca(PO₄)₆(OH)₂ and Ti₂O(PO₄)₂(H₂O)₂ and comparable to that of BiPO₄. The inductive effect of PO₄³⁻ playing in photocatalysis has been well proposed.²⁷ Cd₅(PO₄)₂P₂O₇ with mixed PO₄³⁻ and P₂O₇⁴⁻ groups may have larger inductive effect than Ca(PO₄)₆(OH)₂, Ti₂O(PO₄)₂(H₂O)₂, and BiPO₄ possessed PO₄³⁻ group only, and therefore a higher photocatalytic activity. Besides, the hierarchical structure of Cd₅(PO₄)₂P₂O₇ also benefits the activity. We also compared the activity of flowerlike Cd₅(PO₄)₂P₂O₇ with irregular nanoparticles of Cd₅(PO₄)₂P₂O₇ (SEM was given in Fig. S3†). As presented in Fig. 10, Cd₅(PO₄)₂P₂O₇ with hierarchical structure exhibited superior activity to that of the nanoparticles, which might be due to the hierarchical structure benefiting the light utilization efficiency and dye adsorption property. The detailed relationship between microstructures and photocatalytic activity should be further investigated.

Fig. 10 Comparative photocatalytic activity of hierarchical flowerlike Cd₅(PO₄)₂P₂O₇ (T400) with other phosphate photocatalysts and Cd₅(PO₄)₂P₂O₇ (T400) with other morphology (irregular nanoparticles) under UV (254 nm) light irradiation.

The effect of calcination temperature on the photocatalytic performance of the resulted Cd₅(PO₄)₂P₂O₇ samples was also investigated in the present work. As shown in Fig. 11a, Cd₅(PO₄)₂P₂O₇ sample after 400 °C calcination (T400) possessed the highest photocatalytic performance (93%) for RhB degradation. Increasing the calcination temperature from 400 °C to 500 °C resulted in a 16% decrease in removal rate of RhB. The slight decrease may be due to its lower surface area and growth of particle size, though the degree of crystallinity for T500 is much higher than that for T400. Further calcination at 700 °C, a very low removal rate (only 43%) was obtained. Such a great decrease of activity should be attributed to the agglomeration of large particle, the reduced surface area and the destruction of unique hierarchical structure at higher temperature.

Fig. 11 (a) Photocatalytic degradation curves of RhB as a function of the irradiation time over Cd₅(PO₄)₂P₂O₇ samples calcined at different temperatures, (b) the recyclability of the photocatalytic degradation of RhB in the presence of the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres (T400), (c) XRD patterns of the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres (T400) before and after five runs of photocatalytic reaction, and (d) SEM image of the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres (T400) after five runs of photocatalytic reaction.

Except for photocatalytic efficiency, stability is another critical factor for the wide application of photocatalysts, especially for those noxious element-containing photocatalysts like CdSe, Pb₃O₄ Ag₃AsO₄, and etc.^46–48 Thus, the stability of the hierarchical Cd₅(PO₄)₂P₂O₇ was further investigated by recycling the degradation of RhB under the same conditions. As shown in Fig. 11b, the hierarchical Cd₅(PO₄)₂P₂O₇ sample (T400) did not show any apparent decrease in RhB removal even after five successive operations. Both XRD and SEM examinations on Cd₅(PO₄)₂P₂O₇ further confirm that there were no obvious changes in the crystal structure or in the flowerlike architecture of the catalysts before and after photoreaction (Fig. 11c and d). All these results demonstrate that the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres were highly stable and are promising candidates for wastewater treatment in solar-driven applications.

The photocatalytic mechanism of hierarchical Cd₅(PO₄)₂P₂O₇ during RhB degradation was further investigated. It is generally accepted that the dyes can be degraded by a large number of reactive species including photoinduced holes (h⁺), hydroxyl radicals (˙OH), and superoxide anions (˙O₂⁻) involved in the photocatalytic oxidation process. Therefore, the reactive species scavengers, including ammonium oxalate (AO), tert-butyl alcohol (TBA), and benzoquinone (BQ), were employed to investigate the corresponding effect of h⁺, ˙OH, and ˙O₂⁻ on the degradation activity of RhB. From Fig. 12a, it was found that the addition of BQ greatly reduced the photocatalytic efficiency of RhB from 93% to about 44%. This indicates that the ˙O₂⁻ is the key factor affecting the photocatalytic performance of the Cd₅(PO₄)₂P₂O₇ microspheres. The presence of ˙O₂⁻ radicals can be proved by a nitroblue tetrazolium (NBT) probe method.³⁸ As shown from Fig. 12b, it was observed that the maximum absorption peak at 259 nm decreased with the prolonging irradiation time, indicating the specific reaction between NBT and ˙O₂⁻ radicals. When TBA and AO were added into the reaction system, the photocatalytic efficiency of RhB also decreased to 28 and 66%, respectively. This suggested that the photocatalytic oxidation of RhB may involve the direct oxidation by photoinduced h⁺ and the indirect oxidation by ˙OH radicals. The ˙OH radicals were also examined by a photoluminescence (PL) technology with terephthalic acid (TA) as a probe molecule. The fluorescence intensity was found to increase steadily with irradiation time (Fig. 12c), implying that ˙OH radicals can be generated in the Cd₅(PO₄)₂P₂O₇ suspensions under UV light irradiation. In summary, the main reactive species involved in the photocatalytic degradation of RhB over Cd₅(PO₄)₂P₂O₇ are ˙O₂⁻, h⁺, and ˙OH.

Fig. 12 (a) Effects of scavengers on the degradation efficiency of RhB, (b) UV-vis absorption spectra of NBT in Cd₅(PO₄)₂P₂O₇ (T400) suspension under UV irradiation, (c) ˙OH-trapping photoluminescence spectra of Cd₅(PO₄)₂P₂O₇ (T400) under UV irradiation, and (d) the possible mechanism of the RhB degradation over Cd₅(PO₄)₂P₂O₇.

Since the generation of reactive species involved in the photocatalytic reaction requires the band structure of semiconductor to meet the thermodynamic potential for the reaction, it is highly necessary to investigate the oxidizing or reducing ability of photogenerated carries. As for Cd₅(PO₄)₂P₂O₇, the band position was calculated using the empirical equation, E_CB = X − E^e + 0.5E_g, where E_CB is the conduction band edge potentials, X is the electronegativity of the semiconductor, which is the geometric mean of the electronegativity of the constituent atoms, E^e is the energy of the free electrons on the hydrogen scale (approximately 4.5 eV), E_g is the band gap energy of the semiconductor, and E_VB can be determined by E_VB = E_CB + E_g.²⁷ The X value for Cd₅(PO₄)₂P₂O₇ is 6.44 eV, and the E_g of Cd₅(PO₄)₂P₂O₇ was set as 5.0 eV from the DRS results. Thus, the E_CB and E_VB of Cd₅(PO₄)₂P₂O₇ are estimated to be −0.56 eV (vs. NHE) and 4.44 eV (vs. NHE), respectively. According to the band gap structures of Cd₅(PO₄)₂P₂O₇ and the effects of scavengers on the photocatalytic reaction, a possible mechanism for the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres is proposed in Fig. 12d. Cd₅(PO₄)₂P₂O₇ can be efficiently excited to generate photoinduced electron–hole pairs under UV irradiation. The photogenerated electrons in the CB of Cd₅(PO₄)₂P₂O₇ can reduce O₂ to yield ˙O₂⁻ as its potential is more negative than E(O₂/O₂˙⁻) (−0.33 eV vs. NHE). On the other hand, because the photogenerated holes in the VB is more positive than E(˙OH/OH⁻) (2.38 eV vs. NHE),⁴⁹ it becomes more favorable for the holes to react with OH⁻ or H₂O, producing active ˙OH radicals. The formed ˙O₂⁻ radicals, ˙OH radicals as well as the photoinduced holes with high oxidation power will decompose the RhB molecules and contribute to the high photocatalytic efficiency for Cd₅(PO₄)₂P₂O₇.

Conclusions

Hierarchical flowerlike Cd₅H₂(PO₄)₄·4H₂O microspheres have been simply prepared by addition of phosphates to immobilize various cadmium ions including CdCl₂, Cd(NO₃)₂, and Cd(CH₃COO)₂ in aqueous solution. Such hierarchical Cd₅H₂(PO₄)₄·4H₂O microspheres assembled with nanosheets were fabricated via an Ostwald ripening process in the absence of any template or surfactant. Hierarchical flowerlike Cd₅(PO₄)₂P₂O₇ microspheres, which were achieved by condensation and dehydration of Cd₅H₂(PO₄)₄·4H₂O, exhibited excellent photocatalytic performance for the degradation of RhB under UV light irradiation due to their hierarchical structure, numerous intra- and inter-pores, and strong oxidation ability. During successive operations for treatment of dye wastewater, the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres showed not only durable photocatalytic degradation toward RhB, but also highly structural stability the catalyst itself. Except for dye wastewater treatment, the hierarchical Cd₅(PO₄)₂P₂O₇ microspheres with strong oxidation capability may also have potential applications in environmental purification, especially in efficient removal of persistent volatile organic pollutants such as aromatic hydrocarbons and polychlorinated biphenyls.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (no. 21103193), Doctoral Foundation of Shandong Province (BS 2013NJ013) and Scientic Research Foundation of Qufu Normal University (BSQD20110116).

Notes and references

1. X. B. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891–2959.
2. J. X. Wang, X. J. Wei, J. Y. Shen, X. M. Lu, J. M. Xie and M. Chen, Prog. Chem., 2014, 26, 1460–1470.
3. C. Wen, A. Y. Yin and W. L. Dai, Appl. Catal., B, 2014, 160, 730–741.
4. H. X. Li, Z. F. Bian, J. Zhu, D. Q. Zhang, G. S. Li, Y. N. Huo, H. Li and Y. F. Lu, J. Am. Chem. Soc., 2007, 129, 8406–8407.
5. J. H. Huang, R. Ma, Y. Ebina, K. Fukuda, K. Takada and T. Sasaki, Chem. Mater., 2010, 22, 2582–2587.
6. F. K. Chen, J. T. Zai, M. Xu and X. F. Qian, J. Mater. Chem. A, 2013, 1, 4316–4323.
7. W. A. Lopes and H. M. Jaeger, Nature, 2001, 414, 735–738.
8. G. J. D. Soler-Illia, C. Sanchez, B. Lebeau and J. Patarin, Chem. Rev., 2002, 102, 4093–4138.
9. R. M. Xing, Y. Xue, X. Q. Liu, B. S. Liu, B. J. Miao, W. Z. Kang and S. H. Liu, CrystEngComm, 2012, 14, 8044–8048.
10. H. X. Zhong, Y. L. Ma, X. F. Cao, X. T. Chen and Z. L. Xue, J. Phys. Chem. C, 2009, 113, 3461–3466.
11. M. Shang, W. Z. Wang, J. Ren, S. M. Sun and L. Zhang, CrystEngComm, 2010, 12, 1754–1758.
12. X. F. Chen, J. B. Liu, H. Wang, Y. L. Ding, Y. X. Sun and H. Yan, J. Mater. Chem. A, 2013, 1, 877–883.
13. Z. Wang, J. G. Hou, C. Yang, S. Q. Jiao, K. Huang and H. M. Zhu, Phys. Chem. Chem. Phys., 2013, 15, 3249–3255.
14. J. B. Fei, Y. Cui, X. H. Yan, W. Qi, Y. Yang, K. W. Wang, Q. He and J. B. Li, Adv. Mater., 2008, 20, 452–456.
15. B. Jiang, P. Zhang, Y. Zhang, L. Wu, H. X. Li, D. Q. Zhang and G. S. Li, Nanoscale, 2012, 4, 455–460.
16. C. N. R. Rao, S. Natarajan, A. Choudhury, S. Neeraj and A. A. Ayi, Acc. Chem. Res., 2001, 34, 80–87.
17. R. Murugavel, A. Choudhury, M. G. Walawalkar, R. Pothiraja and C. N. R. Rao, Chem. Rev., 2008, 108, 3549–3655.
18. P. Olivera-Pastor, P. Maireles-Torres, E. Rodriguez-Castellon, A. Jimenez-Lopez, T. Cassagneau, D. J. Joner and J. Roziere, Chem. Mater., 1996, 8, 1758–1769.
19. H. Yang, H. J. Zeng, L. J. Hao, N. R. Zhao, C. Du, H. Liao and Y. J. Wang, J. Mater. Chem. B, 2014, 2, 4703–4710.
20. X. L. Wang, J. F. Shi, Z. Li, S. H. Zhang, H. Wu, Z. Y. Jiang, C. Yang and C. Y. Tian, ACS Appl. Mater. Interfaces, 2014, 6, 14522–14532.
21. X. J. Wang, X. L. Yang, J. H. Cai, T. T. Miao, L. H. Li, G. Li, D. R. Deng, L. Jiang and C. R. Wang, J. Mater. Chem. A, 2014, 2, 6718–6722.
22. R. X. Sun, K. Z. Chen, Z. M. Liao and N. Meng, Mater. Res. Bull., 2013, 48, 1143–1147.
23. X. Y. Zhao, Y. J. Zhu, F. Chen, B. Q. Lu and J. Wu, CrystEngComm, 2013, 15, 206–212.
24. S. D. Jiang, Q. Z. Yao, G. T. Zhou and S. Q. Fu, J. Phys. Chem. C, 2012, 116, 4484–4492.
25. T. B. Zhang, Y. C. Lu and G. S. Luo, ACS Appl. Mater. Interfaces, 2014, 6, 14433–14438.
26. N. Niu, D. Wang, S. H. Huang, C. X. Li, F. He, S. L. Gai, X. B. Li and P. P. Yang, CrystEngComm, 2012, 14, 1744–1752.
27. C. S. Pan and Y. F. Zhu, Environ. Sci. Technol., 2010, 44, 5570–5574.
28. A. M. Hu, M. Li, C. G. Chang and D. L. Mao, J. Mol. Catal. A: Chem., 2007, 267, 79–85.
29. X. T. Hong, X. H. Wu, Q. Y. Zhang, M. F. Xiao, G. L. Yang, M. R. Qiu and G. C. Han, Appl. Surf. Sci., 2012, 258, 4801–4805.
30. Y. Yang, Y. Murakami, A. Y. Nosaka and Y. Nosaka, AZojomo, 2008, 4, 1–5.
31. Z. G. Yi, J. H. Ye, N. Kikugawa, T. Kako, S. X. Ouyang, H. Stuart-Williams, H. Yang, J. Y. Cao, W. J. Luo, Z. S. Li, Y. Liu and R. L. Withers, Nat. Mater., 2010, 9, 559–564.
32. I.-S. Cho, D. W. Kim, S. Lee, C. H. Kwak, S.-T. Bae, J. H. Noh, S. H. Yoon, H. S. Jung, D.-W. Kim and K. S. Hong, Adv. Funct. Mater., 2008, 18, 2154–2162.
33. G. Wang, B. B. Huang, X. B. Ma, Z. Y. Wang, X. Y. Qin, X. Y. Zhang, Y. Dai and M.-H. Whangbo, Angew. Chem., Int. Ed., 2013, 52, 4810–4813.
34. S. Y. Guo and S. Han, J. Power Sources, 2014, 267, 9–13.
35. P. Yin, R. J. Qu, X. G. Liu, X. Q. Dong and Q. Xu, Food Chem., 2014, 148, 307–313.
36. F. Nozaki and Y. Iimori, Bull. Chem. Soc. Jpn., 1976, 49, 567–568.
37. H. E. L. Madsen, J. Cryst. Growth, 2004, 263, 564–569.
38. X. L. Xu, X. Duan, Z. G. Yi, Z. W. Zhou, X. M. Fan and Y. Wang, Catal. Commun., 2010, 12, 169–172.
39. V. Brandel, N. Clavier and N. Dacheux, J. Solid State Chem., 2005, 178, 1054–1063.
40. H. Gülce, V. Eskuizeybek, B. Haspulat, F. Sarl, A. Gülce and A. Avcl, Ind. Eng. Chem. Res., 2013, 52, 10924–10934.
41. J. C. Yu, L. Z. Zhang, Z. Zheng and J. C. Zhao, Chem. Mater., 2003, 15, 2280–2286.
42. M. Manechi, P. A. Maurice and S. J. Traina, Soil Sci., 2000, 165, 920–933.
43. K. J. Zhu, K. Yanagisawa, A. Onda and K. Kajiyoshi, J. Solid State Chem., 2004, 177, 4379–4385.
44. V. Brandel, N. Clavier and N. Dacheux, J. Solid State Chem., 2005, 178, 1054–1063.
45. S. P. Hu, C. Y. Xu, F. X. Ma, L. Cao and L. Zhen, Dalton Trans., 2014, 8439–8445.
46. F. A. Frame, E. C. Carroll, D. S. Larsen, M. Sarahan, N. D. Browning and F. E. Osterloh, Chem. Commun., 2008, 2206–2208.
47. Y. G. Zhou, J. L. Long, Q. Gu, H. X. Lin, H. Lin and X. X. Wang, Inorg. Chem., 2012, 51, 12594–12596.
48. J. T. Tang, Y. H. Liu, H. Z. Li, Z. Tan and D. T. Li, Chem. Commun., 2013, 49, 5498–5500.
49. Z. H. Li, T. T. Dong, Y. F. Zhang, L. Wu, J. Q. Li, X. X. Wang and X. Z. Fu, J. Phys. Chem. C, 2007, 111, 4727–4733.

---

Footnote

† Electronic supplementary information (ESI) available: N₂ adsorption–desorption isotherms and pore-size distribution of Cd₅(PO₄)₂P₂O₇ (T500 and T700), comparative photocatalytic activity of samples under the irradiation of visible light (λ > 400 nm), SEM image of Cd₅(PO₄)₂P₂O₇ (T400) irregular nanoparticles. See DOI: 10.1039/c5ra07224g

---