One-step preparation of a novel SrCO3/g-C3N4 nano-composite and its application in selective adsorption of crystal violet

A novel kind of nanoparticle SrCO3/g-C3N4 was prepared using strontium carbonate (SrCO3) and melamine (C3H6N6) as raw materials via one-step calcination. The formation of SrCO3/g-C3N4 was confirmed from the X-ray diffraction (XRD), Fourier transform infrared spectra (FT-IR), Scanning Electron Microscopy (SEM), Transmission Electron Microscopy (TEM), Brunauer–Emmett–Teller (BET) and X-ray photoelectron spectroscopy (XPS) analysis. Its selective adsorption performance was evaluated towards crystal violet (CV), rhodamine B (RhB) and methylene blue (MB). The results showed that the SrCO3/g-C3N4 had selective adsorption ability of CV. Furthermore, adsorption measurements of CV were conducted to investigate the influences of contact time, initial concentration, initial dye solution pH value and adsorbent dosage. The maximum removal rate of CV was 98.56% when the initial concentration was 1600 mg L−1. The kinetic study indicated the adsorption of CV followed the pseudo-second-second model well. The adsorption efficiency of SrCO3/g-C3N4 was greater (97.46%) than that of g-C3N4 (31.30%) and SrCO3 (17.30%). It could be deduced that the synergistic effect of conjugation interaction of g-C3N4 and the electrostatic attraction of SrCO3 might be the main driving force for the superb adsorption of CV.


Introduction
Nowadays, graphitic carbon nitride (g-C 3 N 4 ) has gained considerable research attention because it possesses excellent advantages such as high chemical and thermal stability under ambient temperature, cost-effective, non-toxic and simple preparation. 1,2 g-C 3 N 4 , as a new intriguing class of graphite analogue, consists of conjugated planes containing highly ordered tri-s-triazine (C 6 N 7 ) units. The layered structure of the tri-s-triazine is connected by weak forces-van der Waal forces. It has been widely used in the photocatalysis eld for it has an energy band gap of 2.7 eV which makes it is capable of the visible adsorption. [3][4][5][6] As a photocatalyst, it has been widely used in water splitting, 7,8 organic pollutants degradation, 9,10 CO 2 reduction, 11 and other elds. 12,13 Crystal violet (CV, C 25 H 30 ClN 3 , IUPAC name is N- [4-[bis][4dimethyl-amino]-phenyl]methylene]-2,5-cyclohexadien-1-ylidine]-n-methylmethanaminium chloride), a typical triphenylmethane dye, is widely used in cell biology, paper, leather and textile industry. [14][15][16] The wastewater containing CV is low biodegradability and high stability (complex aromatic structure) and it could be absorbed through the skin and causing the skin, eye, digestive irritation and even cancer. [17][18][19] From the aspect of environmental safety and health of life, it is vital to develop an effective way to abate CV in wastewater. Various methods have been adopted for eliminating dye pollution from water, including chemical oxidation, 20,21 photo-catalytic decomposition, 22,23 electro-catalytic degradation 24,25 and non-thermal plasma. [26][27][28] These methods usually have some defects, such as slow degradation rate, complex, heavy expenses and usually causing secondary pollution. The adsorption technique is gaining more attention for it is high efficiency, simple design, cost-effective and adaptable. [29][30][31][32][33][34][35] There have been various kinds of adsorbents developed, such as carbon materials, [36][37][38] natural clay minerals, [39][40][41][42] and bioadsorbents. [43][44][45] However, certain deciencies including costly, intricate pre-treatment and low adsorption capability limited the use of some certain adsorbents. [46][47][48][49][50][51][52][53] Due to the conjugated region, stacking structure and the weak forces of g-C 3 N 4 , it has the potential for aggregating functional groups or materials to form nanocomposite with multiform favorable properties. [54][55][56] This distinctive structure has drawn great interesting in improving the photocatalytic performance. 57,58 But the research of using modied g-C 3 N 4 as adsorbent for clearing up dye pollutant is rarely been reported. 59,60 As far as we know, there has been no report of using g-C 3 N 4 -based composite as an adsorbent in dye wastewater treatment. Strontium carbonate (SrCO 3 ), a typical alkaline earth metal carbonate, has been widely used as additives in industrial production. [61][62][63] Meanwhile, some research has been reported on its adsorption performance attributed to its various architectures. 64,65 In this work, we rstly induced SrCO 3 to incorporate with g-C 3 N 4 via one-step calcination method to fabricate a novel adsorbent SrCO 3 /g-C 3 N 4 . The morphology and structure of the composite was characterized by XRD, FT-IR, SEM, TEM, BET and XPS, and its adsorptive capacity and selectivity of CV in aqueous solution were investigated. To the best of our knowledge, this work represents the rst example employing g-C 3 N 4based composite for selective adsorption of CV.

Experimental
2.1 Synthesis of SrCO 3 /g-C 3 N 4 SrCO 3 , melamine and CV are all AR grade and purchased from Chengdu Kelong Chemical Agents (China), without any further purication. SrCO 3 /g-C 3 N 4 was synthesized by one-step calcination process in a muffle furnace. SrCO 3 and melamine (mass ratio ¼ 1 : 1) were dissolved with deionized water in alumina crucible, and then the mixed solution was dispersion with ultrasonic irradiation for 20 min under ambient temperature. The nal solution was transferred into muffle furnace and maintained at 600 C for 4 hours at heating rate of 5 C min À1 to obtain the prepared nanocomposite. SrCO 3 and melamine were also treated as the forward route for making comparison.

Characterization
X-ray diffraction (XRD, Shimadzu, XRD-6100) patterns were detected with Cu Ka radiation (40 kV, 30 mA, 2q ¼ 10-80 ) to investigate the crystal structures of the samples. Fourier transform infrared spectra (FT-IR, Shimadzu, IR Prestige-21), were recorded in the range of 4500-400 cm À1 , using KBr technique to analyze the functional groups on the surface of the composite. Scanning electron microscope (SEM, JEOL, JSM-7800F) and Transmission Electron Microscopy (TEM, JEOL, JEM-2100) were used to observe the morphologies and the microstructures of the samples. The N 2 adsorption apparatus (Micromeritics, ASAP 2020) were used to obtain the Brunauer-Emmett-Teller (BET) surface area of the samples. X-ray photoelectron spectroscopy (XPS, Thermo Scientic, ESCALAB 250xi) was used to determine the binding energy.

Adsorption experiments
The concentration of CV in the solution was determined at the maximum absorbance (l max ¼ 580 nm) by UV-vis spectrophotometer (UV-vis spectrometer, Tianmei, UV1102). For high concentration, the samples were diluted before measurements. And the initial pH of the dye solution was measured by pH meter (pH meter, Sartorius, PB10).
The adsorption experiments were conducted by a batch method. All experiments were conducted at 7.0 pH value, except those that investigated the effect of initial pH of dye solution.
And 0.08 g adsorbent was dispersed in 80 mL CV solution, except that were used to study the effect of the dosage. Similarly, the initial concentration is 1600 mg L À1 , except those that investigate the same parameters. The pH of the initial dye solution was adjusted with HCL (0.1 mol L À1 ) and NaOH (0.1 mol L À1 ). The kinetic experiments were carried out at the initial concentration of 500, 1000, 1200, 1400, and 1600 mg L À1 . It performed on an air bathed shaker. The solution was separated by centrifugation at 5000 rpm for 5 min.
The adsorption capacity at time t q t (mg g À1 ) and removal rate (%) were calculated using the following equation: where C 0 and C t are the liquid-phase concentrations of the CV (mg L À1 ) at initial time and time t, respectively. V is the volume of the solution (mL) and m is the mass of the used adsorbent (mg).

Characterization of the samples
The XRD patterns of the prepared samples are shown in Fig. 1.
The g-C 3 N 4 displayed a typical diffraction peaks at 27.46 and 12.96 , corresponding to the (002) and (100) diffraction planes, which represented the interlayer stacking reection and inplane structure of aromatic system, respectively. 66 (111) diffraction of g-C 3 N 4 in the synthesized composite, the diffraction peaks were mostly the same as the SrCO 3 , but the feature peaks positions of SrCO 3 shied slightly toward a lower diffraction angle. And the impure peaks of the composite might be the carbonization of the raw materials or the melamine did not condense completely. The FT-IR spectra of the prepared materials are shown in Fig. 2. For the calcined melamine, the typical adsorption peaks in the 1200-1700 cm À1 range and at 808 cm À1 of g-C 3 N 4 could be observed. They are assigned to the typical skeletal stretching vibrations of armomatic C-N heterocycles and the out-of-plane bending vibration of tri-s-triazine rings, respectively. 70,71 The peaks at 698, 858, 1070, 1458 and 1774 cm À1 corresponding to the CO 3 2À of the calcined SrCO 3 were observed. 50 In case of the SrCO 3 /g-C 3 N 4 , the typical skeletal stretching vibrations of tri-striazine were hardly observed; it might be the adsorption intense of SrCO 3 was so strong in order to impede the peaks of g-C 3 N 4 . Meanwhile, the peak at 808 cm À1 presented in g-C 3 N 4 shied to 821 cm À1 , the peak at 856 cm À1 presented in SrCO 3 shied to 858 cm À1 , the peaks at 1070 cm À1 and 1774 cm À1 were disappeared, these might due to the strong interactions between the carbonate of SrCO 3 and tri-s-triazine rings of g-C 3 N 4 . Furthermore, the new peak at 2112 cm À1 corresponding to C^N groups appeared aer calcination indicating that the incorporation of SrCO 3 has intense interaction during the condense process of melamine. 72 To make further investigation of the constitution of the prepared composite, the calcined product was treated with HCl (0.5 mol L À1 ) and deionized water, because g-C 3 N 4 possessed fascinating acid stability. The treatment was terminated until the pH value was neutral, and then the treated nanocomposite was dried at 60 C for 24 h. The XRD patterns and FT-IR spectra of the HCl-treated product are characterized in Fig. 3. The feature diffraction peak of g-C 3 N 4 , indicating the interlayer stacking, changed from 27.46 to 27.92 is explicitly observed in Fig. 3a. The change of the diffraction peak was corresponded to the stacking distance reduced from 0.325 to 0.319 nm. The above results implied that the interlayer stacking order was improved. And FT-IR spectra of the product were well matched with the pure g-C 3 N 4 except the weak intensity (Fig. 3b). So, it could be indicated that the SrCO 3 /g-C 3 N 4 was formed with the raw materials SrCO 3 and melamine.
SEM and TEM images of the calcined samples are displayed in Fig. 4. It revealed the morphology and microstructure of SrCO 3 /g-C 3 N 4 . In Fig. 4a, it could be seen the calcined melamine was predominantly composed of plate-like sheets, 73 and the morphology of calcined SrCO 3 was irregular polyhedrons with smooth surface is shown in Fig. 4b. In Fig. 4c, it could be seen clearly that the g-C 3 N 4 covered on the SrCO 3 incompletely. Further structural details of SrCO 3 /g-C 3 N 4 are shown in Fig. 4d. There were some step edges of the layered g-C 3 N 4 and rods-like SrCO 3 with some parts overlapping. It might be the amount of g-C 3 N 4 was not enough to disperse on the entire surface of SrCO 3 ; and some areas cannot be wrapped by the g-C 3 N 4 . The SEM and TEM image could reveal that there existed some interactions between these two materials. Meanwhile, the specic surface areas of the clained samples were measured. As shown in Table 1, compared with g-C 3 N 4 the composite BET surface area gets smaller with the addition of SrCO 3 , this might be the SrCO 3 interacted with the stacking structure of g-C 3 N 4 . These results  To further study the chemical state of SrCO 3 /g-C 3 N 4 , the XPS measurements were carried out. Fig. 5a is the general XPS spectra of calcined materials, indicating the presence of C, N, O and Sr in SrCO 3 /g-C 3 N 4 . The XPS spectra of C 1s are shown in Fig. 5b. For pristine g-C 3 N 4 the peaks at 284.2 eV, 287.7 eV were corresponding to sp 2 -hybridized carbon in C-C group or the adventitious hydrocarbon 74 and sp 2 -bond carbon in form of C-C]N. 75 But aer the blended calcinations, the peak at 284.2 eV shied to 284.6 eV. The N 1s spectra are shown in Fig. 5c. The peak at 398.2 eV was attributed to sp 2 -hybridized nitrogen in N atom aromatic rings in form of C-N]C, 76 and the nearly peaks at 399.3 eV and 401.0 eV were regarded as tertiary nitrogen (N-(C) 3 ) and amino functional groups, respectively. 77 Aer incorporation, the binding energy peaks slightly shied. The peaks at 531.1 eV and 533.2 eV of O 1s spectra corresponding to the SrCO 3 have no signicant change in Fig. 5d. The XPS spectra of Sr 3d at 133.1 eV and 134.6 eV for calcined SrCO 3 were observed in Fig. 5e. These peaks was corresponding to the Sr 3d5/2 and Sr 3d3/2, and the peak at 134.6 eV shied to 134.7 eV aer calcination. All these changes aer co-calcination suggested that there were strong interactions between the two raw materials. Therefore, according to the analysis of XPS and the above results of XRD, FT-IR, SEM, TEM and BET measurements, it could be concluded that the SrCO 3 /g-C 3 N 4 was synthesized using SrCO 3 and melamine by one-step calcination.

The selective adsorption of CV
In our previous study, cationic dyes MB, RhB and CV were chosen as the target dye to conduct the adsorption experiments. We found that SrCO 3 /g-C 3 N 4 appeared efficient adsorption capacity for CV except for the other dye solutions. To make further investigation of the selectivity of SrCO 3 /g-C 3 N 4 , mixed solutions of CV/MB, CV/RhB and CV/MB/RhB with equal volume ratio were prepared. The concentration of each dye solution was 10 mg L À1 , and dosage ratio of SrCO 3 /g-C 3 N 4 and mixed solutions was 1 g L À1 . The adsorption experiments were carried out under ambient temperature and different pH value for 120 min.
The optical pictures and UV-vis measurements of the adsorption process are shown in Fig. 6. It could be apparently seen that CV was decolorized aer the test (Fig. 6b), but in the MB and RhB groups the color slightly changed (Fig. 6a-c). In Fig. 6d, the color of CV/MB mixed solution was changed from dark blue to incipient blue (the color corresponding to the MB). For the CV/RhB group, the end color is very close to RhB   ( Fig. 6e). It suggested that CV in these two mixtures was selectively captured. However, the color in group CV/MB/RhB did not change, because the color of MB and RhB mixed solution is the same as that of CV. To further investigate the selective adsorption of CV, the UV-vis spectra were conducted for the tested groups. In the case of CV/MB, the UV-vis adsorption peaks of CV and MB were at 580 nm and 664 nm before adsorption, respectively. Aer adsorption, the peak of CV was dropped down drastically while the MB peak was almost unchanged (Fig. 6f). The same situation happened in the CV/RhB group (Fig. 6g).
Although the optical of CV/MB/RhB group did not change, but the UV-vis adsorption curve appeared the same situation with the other two mixed groups (Fig. 6h). It could be concluded that the SrCO 3 /g-C 3 N 4 performed selective adsorption of CV from CV/MB/RhB. Furthermore, we also evaluated the effect of the initial pH of the dye solution. Fig. 7 shows the tested results. It could be seen that the initial pH (ranged from 4 to 10) of mixed dye solution did not signicantly inuence the adsorption process in each tested group.
3.2.1 Effect of contact time. Fig. 8 shows the effect of different contact time (0-160 min) in the adsorption process. The initial concentration of CV is 1600, 1800 and 2000 mg L À1 and the dosage ratio of SrCO 3 /g-C 3 N 4 and CV solution was 1 g L À1 . For all concentrations, the removal rate of CV became constant aer 120 min. Meanwhile, the group of 1600 mg L À1 reached the equilibrium rstly and the removal rate was up to the other concentration. So, the contact time of 120 min and the initial concentration of CV of 1600 mg L À1 were chosen for the following investigation.   Fig. 9. From the obtained results, it was found that the removal efficiency kept steadily before the initial dye concentration up to 1800 mg L À1 . This might be attributed to the higher initial concentration offered more numbers of dye molecules that could contact easily with SrCO 3 /g-C 3 N 4 . However, the removal rate was reduced at higher initial concentration indicating the saturation of SrCO 3 /g-C 3 N 4 .
3.2.3 Effect of adsorbent dosage. The dosage of the adsorbents is one of the major parameter which inuences the adsorption process; an appropriate amount of adsorbents is in favor of the real industrial treatment. Fig. 10 shows the effect of the adsorbent dosage (0.25, 0.50, 0.75, 1.00, 1.25, 1.50 g L À1 ) of the CV. The removal rate increased with the increased amount of the adsorbent. When the dosage ups to 1.00 g L À1 , the removal efficiency did not increase with the increasing dosage. It might due to the binding of almost CV ions onto SrCO 3 /g-C 3 N 4 , which made the equilibrium between solution and adsorbents. So the amount of adsorbent 1.00 g L À1 was the suitable dosage for the adsorption of CV.

Effect of pH.
The pH of initial dye solution is a key factor in adsorption process, which could inuence the interaction between adsorbent and dye. In order to nd out the pH effect, the adsorption test was conducted at initial solution pH ranged from 4 to 10. The results in Fig. 11 show that the initial pH of CV solution had little effect on the adsorption capacity. This observation could demonstrate that SrCO 3 /g-C 3 N 4 is suitable for removing CV in wastewater in a wide pH range, and there is no need to adjust the initial solution pH before treatment.

Adsorption kinetics
In order to study the characteristics of the adsorption process, pseudo-rst-order and pseudo-second-order kinetics models were carried out with different concentration (500, 1000, 1200, 1600 mg L À1 ). The pseudo-rst-order and pseudo-second-order are expressed by eqn (3) and (4). 78 ln(q e À q t ) ¼ ln q e À k 1 t (3)    where k 1 (min À1 ) and k 2 (g (mg À1 min À1 )) are the rate constant of pseudo-rst-order and pseudo-second-order model, respectively; q e (mg g À1 ) and q t (mg g À1 ) are the adsorbed amount at equilibrium time and time t.
The rate constant of adsorption could be calculated from the slope of the plot in Fig. 12 and was listed in Table 2. R 2 of pseudo-second-order model is much higher than the other, indicating that the pseudo-second order equation is more suitable to describe the CV adsorption with SrCO 3 /g-C 3 N 4 . So the chemisorptions or chemical sorption was the ratecontrolling step for the adsorption process. 79

Desorption and reuse
In order to evaluate the reusability of SrCO 3 /g-C 3 N 4 , the regeneration process was performed. Adsorption experiment was conducted at the optimal condition as ascribed above in a 250 mL CV solution rstly, aer equilibrium the used adsorbent was regenerated by centrifugation and washed several times with methanol and water at room temperature. Finally, the regenerated composite was dried for 24 h at 60 C for reusing.
It could be seen that the removal rate gradually decreased aer regeneration in Fig. 13. It might conclude that the adsorption mechanism of this process is not due to the physical adsorption.

Adsorption mechanism of crystal violet
To further study the adsorption performance of CV, the adsorption experiments were conducted with the calcined SrCO 3 , g-C 3 N 4 and SrCO 3 /g-C 3 N 4 rstly. 0.08 g tested materials were dispersed in 80 mL CV solution (10 mg L À1 ), respectively. Aer 2 h adsorption process, the removal rate of CV for tested materials is exhibited in Fig. 14. It could be seen obviously that the removal ability of g-C 3 N 4 and SrCO 3 to CV is much smaller than that of the nanocomposite. The removal rate of g-C 3 N 4 , SrCO 3 and SrCO 3 /g-C 3 N 4 were 31.30%, 17.30% and 97.46%, respectively. The adsorption mechanism of g-C 3 N 4 might be attributed to p-p interaction between the g-C 3 N 4 and CV, and that for the SrCO 3 , the electrostatic attraction between the carbonate generated from the hydrolytic of SrCO 3 and the cationic chromogenic groups of CV. And for the SrCO 3 /g-C 3 N 4 , the synergistic effect of p-p interaction and electrostatic attraction highly improved the adsorption properties of the  Paper nanocomposite on CV. 50 According to the above results and the crystal structures, we could deduce that superb adsorption capacity of CV might attribute to the synergistic interaction of the conjugation of g-C 3 N 4 and the electrostatic attraction of SrCO 3 . Furthermore, the FT-IR spectra of SrCO 3 /g-C 3 N 4 before and aer CV adsorption were recorded. As shown in Fig. 15, the peak at 2112 cm À1 assigned to the nitrile groups shied to the 2107 cm À1 and the peak at 1319 cm À1 assigned to aromatic C-N heterocycles observed clearly. These results indicate that the previously mentioned groups are involved in the adsorption process. Moreover, the intensity of the peak 1458 cm À1 corresponding to C-N heterocycle aromatic became stronger, that was for the CV was adsorbed onto the SrCO 3 /g-C 3 N 4 . All the changes of the FT-IR spectra of the SrCO 3 /g-C 3 N 4 before and aer adsorption could indicate that there have been intense interactions between dye molecules and SrCO 3 /g-C 3 N 4 . Furthermore, the result of desorption test might indicate that there has been intense chemical adsorption between the composite and CV. It could deduced that SrCO 3 /g-C 3 N 4 performed selective adsorption of CV might due to the chemical structure of CV is more symmetrical and conjugation degree is higher than MB and RhB. On the basis of the above analysis, the conjugation and electrostatic interaction would be the driving force for the selective adsorption of CV.

Conclusions
In summary, a new SrCO 3 /g-C 3 N 4 composite adsorbent was rstly synthesized with SrCO 3 and melamine via a simple onestep calcination method. The SrCO 3 /g-C 3 N 4 performed high and specic selective adsorption for CV. The experiment data was well described by the pseudo-second-order model. According to the crystal structural feature of SrCO 3 /g-C 3 N 4 and the FT-IR analysis of SrCO 3 /g-C 3 N 4 before and aer adsorption of CV, it could be deduced that the synergies of p conjugation and electrostatic interaction would be the mechanism for the selective adsorption of CV. Our ndings indicate that g-C 3 N 4 based composite SrCO 3 /g-C 3 N 4 could be a promising adsorbent which can potentially be applied for the removal of CV pollutants from aqueous solution.

Conflicts of interest
There are no conicts to declare.