Hydrazine-induced synthesis of CdS nanorings for the application in photodegradation

In this paper, CdS nanorings synthesized by a facile hydrazine-induced microwave method for the photodegradation of pollutants were reported for the first time. Different reaction method, microwave power, the category and dosage of pH regulating reagent, reaction temperature and reaction time were investigated. The formation of CdS nanorings from the self-assembly of nanoparticles was attributed to the coordination of hydrazine producing the dipole–dipole interaction among the uniform nanoparticles prepared by microwave method. The crystal phase, composition, morphology and surface property of CdS nanorings were characterized. The results showed that 100 nm-sized wurtzite CdS nanorings generated with the self-assembly of 5–8 nm nanoparticles, which presented mesoporous structures. To study the influence of ring-like structures on the photocatalysis, the photodegradation of rhodamine B (RhB) with CdS nanorings and nanoparticles was compared. The results showed that, CdS nanorings displayed higher photodegradation efficiency, which were originated from more favorable band edge potential and effective electron–hole separation producing more superoxide radical and holes as active specifies. The photodegradation path of RhB contained the process of the demethylation, the decarboxylation process, the chromophore cleavage and ring-open reactions. Finally, the available photodegradation of multiple pollutants and reusability of CdS nanorings were carried out.


Introduction
Rapid population growth and industrialization have arisen serious challenges related to environmental issues like the generation of the industrial wastewater with non-biodegradable organic pollutants. The most ideal and feasible strategy to solve the problem is to apply the semiconductor photocatalytic technology [1]. Previously, metal oxides with wide bandgaps such as TiO 2 [2] and ZnO [3] have been widely applied in the field of photodegradation, however, the photocatalysis is not ideal owe to their poor light-harvesting ability [1]. CdS is an important II-VI semiconductor with the narrow bandgap of 2.42 eV, and the electrons easily transit from valence band to conduction band under visible radiation for the application in the visible-light photocatalysis [4].
It's well-known that photocatalysis is strongly dependent on the surface property of catalysts such as the size and morphology [5]. Recently, hollow nanostructures as nanorings are attractive because of their high specific surface areas contributing to multiple reflection of light inside the hollow frameworks, moreover, the thin shells can shorten the transport distance of charge carriers to decrease the probability of charge recombination [6]. Previously, CdS nanorings [7] and Bi 3 O 4 Br nanorings [8] have been confirmed to display the faster degradation rate than nanoparticles. Generally, the synthesis of nanorings is either by template-assisted solution method or surfactant-assistant chemical transformation method. The template-assisted solution method is based on the reaction to form ring-like frameworks followed by the selective removal of the templates [9][10][11], however, it probably breaks the structural stability of the materials. The surfactant-assistant chemical transformation method such as microemulsion method [12] or solvothermal method [6,8,13] normally synthesizes the nanoplates firstly followed by etching the central of the nanoplates selectively, which requires for the expensive surfactant or organic solvent inevitably.
Self-assembly is a popular and facile approach for the preparation of special structural nanomaterials. Owe to the existence of specific interaction, the discrete nanoparticles can organize spontaneously [14], therefore, the key to the preparation of nanorings is to find out the effective additive which induces nanoparticles self-assembly simultaneously and promotes their in-situ linkage. Hydrazine is a weak electrolyte, it weakens the initial electrostatic repulsion among nanoparticles greatly and rebuilds the equilibrium of inter-nanoparticle interaction [15], promoting the assembly of nanoparticles. Previously, hydrazine played important roles in the self-assembly of nanoparticles to 1D nanostructures [15,16]. Therefore, it is expected to direct the self-assembly of CdS nanoparticles into CdS nanorings with the aid of hydrazine. To realize the self-assembly, the nanoparticles with uniform size and shape are requisite because of the facilitation of the nanoscale interaction forces in a liquid phase [17]. However, in conventional heating process such as solvothermal method or chemical bath deposition, crystals tend to nucleate on container walls or dust particles followed by the growth into large particles, resulting in the formation of different-sized nanoparticles assembled into uneven nanorings [7]. Compared to them, microwave irradiation method affords very fast and uniform reaction in the material synthesis because the direct coupling of microwave energy with reactant implements internal molecular heating without suffering thermal gradient effects [18]. In some local superheating zones, numerous "hot spots" result in the formation of massive seeds throughout the bulk solution with fast rate and high yield [19], contributing to the production of high-quality crystals with narrow size distribution.
In this paper, we reported a facile method to synthesize CdS nanorings. Microwave method afforded the efficient heating at the molecular level, contributing to the generation of uniform-sized nanoparticles. The hydrazine as pH regulating reagent strengthened the dipole-dipole interaction among adjacent nanoparticles, promoting the self-assembly of nanoparticles into nanorings. The as-synthesized nanorings displayed the favorable band edge potential and effective electron-hole separation for the generation of more active specifies. Noted that, although CdS was a common material, the synthesis of uniform CdS nanorings by such facile method was rarely reported. Previously, CdS nanorings were synthesized by solvothermal method in the presence of ammonium hydroxide [7]. However, the reaction time was up to 1 h and as-prepared nanorings were uneven with some nanochains and nanoparticles coexisted. Moreover, the photocatalysis was far lower than that of as-synthesized CdS nanorings in our paper. This paper may provide the possibility to synthesize CdS nanorings effectively as a potential photocatalysis material candidate in the long run.

Synthesis of CdS nanorings
10 mM Cd(NO 3 )·4H 2 O which was adjusted to pH = 11 by N 2 H 4 and 30 mM thiourea were transferred to XH-MC-1 microwave synthesis reactor at 800 W, 80 °C for 20 min. After the reaction completed, the products were centrifuged at 10,000 rpm for 10 min followed by wash and dry to obtain yellow powders as CdS nanorings. As a comparison, other morphologies of CdS were synthesized as shown in Table 1, containing different reaction method as chemical bath deposition (CBD), solvothermal and microwave method (No.1), different microwave power as 0, 400 and 800 W

Photocatalysis experiments
CdS nanorings (or CdS nanoparticles which synthesized at 800 W, 90 °C for 20 min) were used to carry out the photocatalysis experiments (S1 in Supplementary Material). To detect active species in the photodegradation of RhB, IPA (20 mM), TEOA (20 mM) and AA (8 mM) as the scavengers for •OH, h + and •O 2 − were added in the system, respectively.

Characterization
The morphology of products was carried out by scanning electron microscopy (FEI Sirion 200) and transmission electron microscopy (JEOL-2010). The crystal phase was measured by X-ray diffractometer (MXP18AHF). The surface valence state and the chemical composition were performed on X-ray photoelectron spectrometer (ESCALAB 250). The specific surface areas were examined on the instrument of NOVA 2000e with Brunauer-Emmett-Teller (BET) method, and the pore-size distribution was estimated by Barrett-Joyner-Halenda (BJH) method. The measurement of zeta potential was performed on Brookhaven 90 plus zeta potential analyzer. The optical measurements were recorded on UV-vis diffused reflectance spectra (UV 2550 spectrophotometer) and fluorescence spectra (F−4500 spectrofluorometer) with

3
Hydrazine-induced synthesis of CdS nanorings for the… excitation wavelength at 380 nm. The photoelectrochemical experiments, including the transient photocurrent-time (I-t) curves and electrochemical impedance spectroscopy (EIS) were carried out using an electrochemical workstation (CHI-660C, Chenhua Instrument, China) (S2 in Supplementary Material). The photodegradation products were identified by Shimadzu LCMS-8030 liquid chromatography-mass spectrometry (LC-MS) with similar measured condition as before [20].

Results and discussion
The formation of nanorings Figure 1 displayed SEM pattern of the products synthesized by different reaction methods. With the chemical bath deposition (CBD) method, small-sized nanoparticles were stacked irregularly (as shown in Fig. 1a). With the solvothermal method, similar nanoparticles were observed with some of them assembling into ring-like structures (Fig. 1b). Figure 1c displayed uniform nanorings with the size of 100-200 nm assembled by nanoparticles, implying nanorings were synthesized by microwave method successfully. Next, we observed the influence of microwave power on the morphology of products. When the microwave power was 0 W which was actually identified with the CBD method, small-sized nanoparticles were stacked irregularly (Fig. 2a). With the power increasing to 400 W, most of the nanoparticles assembled into ring-like structures with the residual nanoparticles aggregating into chain-like structures (Fig. 2b). Furthermore, the increased power of 800 W led to the formation of 100-200 nm uniform nanorings completely (Fig. 2c). Figure 3 displayed the influence of different pH regulating reagents. The reagent as triethanolamine (TEOA) only promoted the generation of extremely tiny nanoparticles (Fig. 3a). The presence of NH 3 ·H 2 O resulted in the formation of slightly large nanoparticles with some of them assembled into ring-like structures (Fig. 3b). Figure 3c displayed SEM of the products synthesized with the mixture of NH 3 ·H 2 O and N 2 H 4 as pH regulating reagents. More ring-like structures formed with most of the residual nanoparticles forming chain-like structures. Finally, with N 2 H 4 as pH regulating reagent, 100-200 nm uniform nanorings could be seen clearly (Fig. 3d).

Influence of different factors on the morphology
In addition, the volume of N 2 H 4 also affected the morphology of products (as shown in Fig. 4). With 4 mL N 2 H 4 in the system, the ring-like structures and chainlike structures coexisted (Fig. 4a). With 7.5 mL N 2 H 4 , more ring-like structures formed with less chain-like structures remained (Fig. 4b). With V N 2 H 4 increasing to 10 mL, 100-200 nm uniform nanorings generated completely (Fig. 4c). Continuously increased V N 2 H 4 to 13 mL made the nanoparticles begin to fill in the cavities of the nanorings (Fig. 4d). More nanoparticles were filled into the cavities to form nanorings with thicker walls when V N 2 H 4 was 16 mL (Fig. 4e). Finally, in the presence of 20 mL N 2 H 4 , almost all of the cavities were filled completely to form nanodisks (Fig. 4f). Figure 5 displayed SEM of the products synthesized at different temperature. When the reaction occurred at 70 °C, most of nanoparticles were stacked to form chain-like structures with bits of them assembled into ring-like structures (Fig. 5a). With the temperature at 80 °C, the assembly of nanoparticles into uniform nanorings was achieved (Fig. 5b). However, the temperature up to 90 °C resulted in irregular aggregation of nanoparticles (Fig. 5c). Figure 6 displayed SEM of the products synthesized for different reaction time. At the time of 5 min, most of nanoparticles were stacked irregularly with some chain-like and ring-like structured products (Fig. 6a). When the time increased to 10 min, lots of chain-like structured products formed with bits of irregular nanoparticle aggregation remained (Fig. 6b). The increased time to 20 min contributed to the formation of uniform nanorings (Fig. 6c). However, some of the nanorings began to decompose into the nanochains at the time of 30 min (Fig. 6d). Finally, nanochains further decomposed to nanoparticles after 40 min (Fig. 6e).

The formation mechanism of nanorings
From above results, it can be concluded that microwave irradiation as the heating method and N 2 H 4 as pH regulating reagent played important roles in the formation of nanorings. Different from TEOA and NH 3 ·H 2 O, N 2 H 4 containing two -NH 2 groups saturated Cd dangling bonds through the coordination [15]. With the hydrolysis of thiourea to release H 2 S, Cd 2+ was attacked by H 2 S to form CdS crystal nucleus. Next, numerous "hot spots" as the center for the nucleation produced, contributing to the generation of massive and uniform crystal nucleus followed by the formation of nanoparticles with Ostwald ripening [19]. To decrease the high surface energy, CdS nanoparticles tent to aggregate rapidly. The presence of N 2 H 4 brought the nanoparticles closer through dipole-dipole interaction for the self-assembly into nanochains, similar to the assembly of CdTe nanoparticles into nanochains in previous report [21]. The nanochains elongated gradually during the successive selfassembly process, which was disadvantageous to the structural stability [7]. Therefore, nanochains inclined to bend and link into nanorings for the structural stability. The formation mechanism is shown in Scheme 1.
During the process, microwave irradiation afforded the efficient heating at molecular level to form massive and uniform nanoparticles instantaneously, contributing to the formation of nanorings (Fig. 1). The increase in microwave power was beneficial to the generation of uniform nanoparticles for the assembly of nanorings (Fig. 2). Compared to other ammoniacal pH regulating reagents, N 2 H 4 adjusted the interaction of interparticles and promoted the dipole-dipole attraction for the formation of nanorings (Fig. 3). The amount of N 2 H 4 affected the dipole interaction directly, resulting in the appearance of different morphologies (Fig. 4). Temperature is another important factor in chemical reactions. The increased temperature promoted Ostwald ripening with oriented assembly of the nanoparticles. However, high temperature led to drastic thermal motion, therefore, the increased probability of the collision among discrete nanoparticles damaged the dipole-dipole attraction inevitably (Fig. 5). Finally, the increase in reaction time showed the aggregation, assembly and decomposition clearly (Fig. 6).

Crystal phase and composition
The crystal phase of as-synthesized nanorings was characterized by XRD (as shown in Fig. 7a). All the diffraction peaks can be indexed to pure wurtzite hexagonal CdS (JCPDS 41-1049, a = b = 4.141 Å and c = 6.720 Å) without the presence of byproduct peaks, suggesting as-synthesized products were CdS with pure wurtzite structure.
To investigate the surface valence state and the chemical composition of CdS nanorings, X-ray photoelectron spectroscopy (XPS) was carried out. The survey spectrum of XPS (Fig. 7b) displayed not only the element of cadmium and sulphur, but also those of oxygen and carbon which were originated from adsorbed O 2 and CO 2 molecules in the atmosphere [20,22]. Cd spectrum revealed two peaks located at 405.25 eV and 411.95 eV assigning to the binding energy of Cd3d 3/2 and Cd3d 5/2 (as shown in Fig. 7c), and S spectrum presented two weak peaks centered at around 161.5 eV and 162.6 eV related to the binding energy of S2p 3/2 and S2p 1/2 (as shown in Fig. 7d), which confirmed the existence of Cd 2+ and S 2− species [23]. Based on the intensity of Cd and S peaks, the obtained surface stoichiometry of Cd:S was 1.00:1.06, in accordance with the stoichiometric composition of CdS. Figure 8a was TEM image of CdS nanorings, which showed 100 nm hollow ring with 50 nm cavity. Nanoring was assembled by 5-8 nm aggregated nanoparticles (Fig. 8b). The lattice plane spacing was 0.208 nm and 0.128 nm, corresponding to (110) and (114) crystal planes of wurtzite CdS, respectively (Fig. 8c). Selected area electron diffraction (SAED) presented the polycrystalline ring, relating to (002), (110), (112), (103) and (114) crystal planes [24,25] (as shown in Fig. 8d). Figure 9 displayed N 2 adsorption-desorption isotherm of CdS nanorings, which was exploited to determine the specific surface area and pore size distribution. The adsorption curve had a hysteresis loop in the range of 0.86-0.99 P/P 0 , which was type IV isotherm identified as the characteristic of mesoporous materials from the intervals among nanorings [26]. The pore size distribution curve (using BJH analysis) displayed wide range from 5 to 130 nm with the maximum at 52.68 nm (inset to Fig. 9). Quantitative calculation showed that BET surface area was 35.61 m 2 /g. The large specific surface areas can provide more active sites for sufficient contact between photocatalysts and pollutants [27].

Surface property
As another critical factor influencing the adsorptive behavior of materials, zeta potential is normally deemed as an indicator for the surface charge of solid particles. We measured zeta potential of CdS nanorings to obtain the value of 16.84 ± 0.75 eV, implying the positive charges were taken on the surface. Hydrazine-induced synthesis of CdS nanorings for the…

Photodegradation of RhB
CdS nanorings were applied in the photodegradation of RhB (as shown in Fig. 10). Before the photodegradation, the suspension was kept in the dark for 60 min. The concentration decrease (characterized by C t /C 0 ) of RhB during this period was extremely low, indicating the adsorption of RhB on the surface of CdS was negligible. With light on, RhB only displayed 2.03% absorption attenuation without CdS, however, it obtained the final photodegradation efficiency of 98.16 ± 0.15% with CdS nanorings, which was much higher than previous report [7]. As a comparison, CdS nanoparticles (the morphology as Fig. 5c) were also used to photodegrade RhB, only obtaining the final efficiency of 83.20 ± 1.12%. Note that, we afforded the averaged photodegradation data of three individual tests for every sample and conducted a thorough error analysis on them. Minor standard deviation indicated the photocatalysis displayed the excellent reproducibility. Next, we measured the kinetics parameters by fitting approximately with the liner equation of ln (C t /C 0 ) = − kt (the inset to Fig. 10). R 2 values were both more than 0.9, suggesting two reactions meet the pseudo-first-order kinetics. k values were easily got from the slope of the equation to be 3.26 × 10 -2 min −1 and 1.31 × 10 -2 min −1 for the nanorings and nanoparticles, respectively. Clearly, the photodegradation of RhB with nanorings was more effective than that of nanoparticles. Hydrazine-induced synthesis of CdS nanorings for the…

Analysis on photocatalysis
Energy level To obtain the information of energy level for CdS, we investigated UVvis DRS of two samples firstly (Fig. 11a). They both showed visible-light absorption with the absorption band edge at 595-615 nm, similar to previous report [23]. Nanorings displayed slightly increased absorbance with blue-shift absorption band edge. The slightly increased absorbance was attributed to the increased light reflection and scattering ability inside the cavities of nanorings. On the basis of DRS result, the bandgap of two CdS was calculated with Tauc plot method [28], obtaining the values of 2.25 eV and 2.15 eV for nanorings and nanoparticles, respectively (inset to Fig. 11a). Next, we calculated E VB (valence band edge potential) and E CB (conduction band edge potential) according to the empirical equation [29,30] (S 3 in Supplementary Materials). E VB and E CB for nanorings were 1.805 and − 0.445 eV, while E VB and E CB for nanoparticles were 1.755 and − 0.395 eV, respectively. Obviously, compared to nanoparticles, nanorings displayed higher E VB and lower E CB , indicating stronger oxidizability of holes and stronger reducibility of electrons.
Charge separation PL spectra resulted from the recombination of photoexcited charge carriers are also used to reflect charge separation ability. Figure 11b showed PL spectra of two samples, which displayed obvious emission peaks located at 560-565 nm, similar to the peak at 565 nm in literature [27]. Compared to nanoparticles, nanorings displayed lower PL intensity, indicating that the electrons and holes were uneasy to recombine [31].  Figure 11c was the transient I-t curves of two samples. Generally, strong photocurrent response under the illumination implies high separation efficiency of charge carrier. Under illumination, two samples responded to get relatively high photocurrent, once the light was switched off, the photocurrent came back nearly to zero rapidly, suggesting the rapid photoresponse and recovery properties [32]. Nanorings exhibited higher photocurrent density, implying it was more efficient in separating the photogenerated electron-hole pairs. It's noted that the photocurrent is slightly attenuated during the increased on-off cycle, originating from the minor recombination of holes and electrons owe to ineffective transfer of the electron to the electrode in time [33].
EIS was carried out to evaluate the charge transfer resistance of two samples (as shown in Fig. 11d). The diameter of the semicircle in the Nyquist plots affords the information on the charge transfer process, and a small semicircle diameter represents low charge transfer resistance (R ct ), contributing to fast charge immigration across the electrode-electrolyte interface [34]. Obviously, nanorings showed small semicircle diameter, indicating fast charge transfer rate from the photocatalysts to the electrolyte for effective electron-hole separation.
Active species It's well-known that the photocatalysis reaction depends on the participation of active species as hydroxyl radical (•OH), superoxide anion radical (•O 2 − ) and hole (h + ). To detect the active species in the reaction, isopropyl alcohol (IPA), ascorbic acid (AA) and triethanolamine (TEOA) as the scavengers for •OH, •O 2 − and h + were added in the system, respectively. Figure 12a displayed the change in the photodegradation of RhB with CdS nanorings, not only TEOA and AA suppressed the photodegradation greatly, but also IPA slightly inhibited the reaction. Obviously, Fig. 12 The photodegradation of RhB with CdS nanorings (a) and CdS nanoparticles (b) in the presence of different radical scavengers and the illustration of the photocatalytic mechanism over CdS nanorings and nanoparticles under illumination (c) h + and •O 2 − were the main active specifies, and •OH also precipitated in the photodegradation. The potential of h + (+ 1.805 eV) was more negative than φ (•OH/H 2 O) (+ 2.68 eV) or φ (•OH/OH − ) (+ 1.99 eV) [29], which was unable to oxidize H 2 O or OH − to •OH. However, the potential of e − (− 0.445 eV) was more negative than φ [30], which was effective to make O 2 into •O 2 − (as shown in Fig. 12c). As a result, h + and •O 2 − were the main active specifies in the system of nanorings. Although h + could not oxidize H 2 O to •OH directly, the protonation of •O 2 − into H 2 O 2 followed by its decomposition generated bits of •OH inevitably [35]. Compared to Fig. 12a, Fig. 12b displayed obviously decreased rate in the presence of TEOA, slightly decreased rate with AA. Therefore, h + was the main active specifies with bits of •O 2 − participating in the reaction. The potential of e − in CdS nanoparticles (− 0.395 eV) was nearly to φ (•O 2 − /O 2 ) (as shown in Fig. 12c), which displayed poor reducibility to generate •O 2 − . Therefore, in the system of nanoparticles, only h + was the main active specifies.
On the whole, the photocatalysis of CdS nanorings was superior than CdS nanoparticles, which was attributed to the reasons as followings. Firstly, the hollow structure increased light reflection and scattering ability (Fig. 11a) for lightharvesting [36]. Secondly, the higher E VB and lower E CB in nanorings contributed to stronger oxidizability of holes and stronger reducibility of electrons (Fig. 12c). Most importantly, nanoparticles based on quantum dot scale were easy to agglomerate, therefore, the lifetime of photogenerated electrons and holes was transient with the inclination to recombining. Nanorings with hollow structures provided large specific surface areas, which not only adsorbed more dye molecules at active sites, but also increased the effective electron-hole separation for the generation of more •O 2 − and h + as the active specifies. As a result, CdS nanorings displayed better photocatalysis than nanoparticles. Previous reports also confirmed similar results, for example, the photodegradation of MB with Cu 2 S micron-ring was superior to that of microsphere [37] and the photodegradation of Cr (IV) with SnS 2 nanorings excelled than that of nanoflowers [38].

Photodegradation path
MS was carried out to determine the product in the photodegradation of RhB with CdS nanorings at different moment (Fig. 13a). It displayed the peak of RhB matrix (m/z = 443) at the initial moment. With the illumination time increasing to 30 min, the abundance of RhB matrix decreased, more peaks occurred with increased abundance in the range of m/z = 100-400. Next, the abundance in the range of m/z = 300-400 decreased, while the abundance in the range of m/z ≤ 250 increased the illumination time of 60 min. Furthermore, the abundance of m/z ≈ 100 obtained maximum with the illumination time of 120 min. Obviously, with irradiation time increased, more and more small molecules with higher abundance generated, indicating RhB was photodegraded but not adsorbed by CdS nanorings.
Next, we speculated the possible photodegradation path from RhB matrix to small molecular products. On one hand, RhB experienced a series of demethylation reactions, such as from m/z = 443 to m/z = 415, 387 and 359 [39,40]. On the other hand, the demethylation and decarboxylation process occurred simultaneously 1 3 to form the product of m/z = 343 [41]. Next, the intermediates of m/z = 359 and m/z = 343 transformed into the product of m/z = 279 after the simultaneous reaction of decarboxylation and chromophore cleavage [42], or to that of m/z = 190 after demethylation process and cleavage of the chromophore [43]. Then the intermediate of m/z = 279 continued demethylating to obtain the product of m/z = 222, followed by the composite reaction of demethylation with the addition of -OH to get the product of m/z = 198 [44]. Finally, all of these intermediates kept on decomposing to produce small molecules (m/z = 149 [45], m/z = 138 [46], m/z = 116 [47], m/z = 97 [48], m/z = 78, m/z = 74, m/z = 65 [43]) by ring-open reactions. According to above results, we deduced the possible photodegradation path (as shown in Fig. 13b).

Availability and reusability of CdS nanorings
To probe the superior photocatalysis of CdS nanorings, multiple pollutants with different structures were photodegradated by CdS nanorings (Fig. 14a). The suspensions consisted of CdS and pollutants were kept in the dark for 60 min followed by the illumination for 120 min. In the dark, CdS nanorings displayed strong adsorption to MB, CR and CPDX molecules, which were attributed to different reasons. The presence of sulphonate anions in MB and CR molecules contributed to strong electrostatic interaction between them and positive-charged CdS nanorings for strong adsorption. However, CPDX molecules contained conjugated aromatic rings with nearly planar structure [49], and CdS nanorings were comprised of large lamellar structures, thus they were easy to contact though the interaction of π-π stacking, resulting in strong adsorption [50]. After light on, CdS nanorings displayed excellent photocatalysis to almost all of the pollutes, containing cationic dyes (as RhB, CV, MG and MB), anionic dyes (as CR) and antibiotics (as TC and CPFX). Obviously, CdS nanorings can photodegradate different pollutants effectively, indicating their availability in practical application.
As a catalyst, long-term stability is required to be considered. As shown in Fig. 14b, the reusability of CdS nanorings was measured with the photodegradation of RhB in sequential four cycles. After 120 min irradiation, the photodegradation efficiency decreased slightly from 98.16%, 95.09%, 95.30 to 94.88%. Obviously, CdS nanorings as the photocatalysts displayed excellent reusability, the slightly decreased efficiency after 4 cycles possibly resulted from the mass loss in the recycle  [51]. Moreover, XRD and SEM measurements of reused CdS nanorings were carried out, with almost no change in crystal structure (Fig. 14c) and morphology (inset to Fig. 14c), indicating CdS nanorings as the photocatalysts were structural stability.

Conclusions
100-200 nm CdS nanorings which were self-assembled by 5-8 nm nanoparticles were synthesized successfully. N 2 H 4 as pH regulating reagent with pH = 11 and onepot microwave method with the power of 800 W at 90 °C for 20 min implemented the effective preparation of CdS nanorings. The microwave heating method promoted the formation of uniform nanoparticles. The coordination of N 2 H 4 forming longrange and substantially strong driving force among the nanoparticles contributed to the self-assembly of nanoparticles to the nanorings. Compared to nanoparticles, nanorings displayed favorable band edge potential and high electron-hole separation contributing to the generation of more active specifies for high photocatalysis. The decomposition of RhB experienced the deethylation, decarboxylation, chromophore cleavage and ring-open reactions up to producing small molecules. Moreover, CdS nanorings displayed excellent photodegradation of multiple pollutants, and four sequential recycle photodegradation of RhB confirmed its excellent reusability.

Data availability
The authors confirm that the data supporting the findings of this study are available within the article [and/or] its supplementary materials.

Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Ethical approval This research does not include experiments involving human tissue and does not contain any studies with human participants or animals performed by any of the authors.