Preparation of CoFe2O4@Carboxymethylcellulose/HZSM-5 as a Novel Nano Magnetic Adsorbent for Metronidazole Removal: Isotherm, Kinetic and Thermodynamic Study

In this research CoFe 2 O 4 @CMC/ HZSM-5 synthesized in the presence of Carboxymethylcellulose as a biopolymer in the green conditions as a new nanomagnetic adsorbent for metronidazole removal from aqueous media. Physical and chemical structure of adsorbent was investigated by FTIR, FESEM, EDS, Mapping, TEM, XRD, VSM, BET and TGA techniques. The results indicated that 94% of the metronidazole concentration removed by CoFe 2 O 4 @CMC/HZSM-5 in the optimal conditions including pH 6, temperature 20 o C, metronidazole concentration 50 mg/L, contact time 60 min and adsorbent dose 2 g/L. The resulting data from adsorption experimental experiments had better correlated with the Freundlich isotherm and pseudo -second order kinetic. Also, the thermodynamic study demonstrated that the adsorption process was an endothermic process and had a physical mechanism. As well as, the achieved findings demonstrated that following six adsorption runs, the adsorbent chemical structure had no change and the nanomagnetic adsorbent efficiency in the removal process had a slight decrease. CoFe 2 O 4 @CMC/HZSM-5 magnetic nanocomposite had effective adsorption capacity for metronidazole removal.


Introduction
About 30-90% of antibiotics are not metabolized in the animal and human bodies. Eventually, they enter the environment through the urine and stool in the form of active compounds. The hospital wastewater is the main source of antibiotics in the environment [1,2]. Medications and antibiotics are the most important groups of emerging pollutants. These organic compounds enter the water sources in two ways. One of these pollution sources is the pharmaceutical industry wastewater and the other one is excretion of the medicines compounds metabolites due to their incomplete metabolism during treatment. Because of the ability of these compounds to alter ecosystems and reduce biodiversity, removal of these compounds from the aquatic environments is so important [3].
Metronidazole is one of the nitroimidazole antibiotics group which has antibacterial and antiinflammatory properties and it can be used to treat diseases that caused by anaerobic bacteria, protozoa, and bacteroides. Metronidazole is one of the most commonly used antibiotics in the world which is in the WHO's list of essential medicines. Metronidazole, like other antibacterial and antioxidant medicines which have nitroimidazole ring structure, is suspicious to causes carcinogenic and mutagenic [4]. By reason of the non-biodegradability, toxicity, mutagenic and carcinogenic potential of metronidazole, the wastewater containing this antibiotic causes harmful effects on the humans and ecological environment so, removal of metronidazole has environmental, technical and economic importance [5].
The mentioned methods have some disadvantages. For example disadvantages of the membrane filtration process are the membrane blockage owing to the chemical sediment growth formation and the physicochemical properties changes of the membrane surface [17]. In the biological methods, during the treatment process, high concentrations antibiotics could make toxicity. The other problem is resistance of some antibiotics against biological treatment. In addition, the chemical methods because of sludge production and need to chemical materials are costly and not popular [18].
The adsorption process compared to the other mentioned methods has more high efficiency, simple design and operation, low cost and non-production of hazardous materials, which are basic features of the adsorption process. Therefore, many researchers focus on its use in the environmental pollutants removal [19].
Some of adsorbents such as activated carbon, ash, coal, metal oxides, agricultural wastes, kaolinite, bagasse, resins, and compost are used as adsorbents to removal of heavy metals and some organic compounds from the aquatic environments. But, some of these adsorbents show low adsorption efficiency since their low surface areas and adsorption capacities. Also, separate those from the environment by filtration and centrifugation methods is time-consuming. In most cases, the adsorbent passes through the filter and leads to the secondary pollution production in the environment [20][21][22][23][24].
To solve this problem, magnetic nanoadsorbents are used. One of the most widely used adsorbents is magnetic nanocomposites [25][26][27]. Magnetic nanocomposites due to their good magnetic property, high area surface and adsorption capacity have attracted the many researchers attention. In a lot of researches, nanocomposites which contain some of biopolymers such as cellulose, methylcellulose, carboxymethylcellulose, and chitosan in their structure, are used as bioadsorbents in water treatment processes [28,29].
The mentioned polysaccharide compounds owing to cross-links and functional groups such as OH, NH2, COOH, COONa in their structure can be easily dissolved in water and cause electrostatic interaction between the adsorbent and the contaminant [30]. These polysaccharide derivatives have some advantages such as availability, biodegradability and non-toxicity. Therefore, they are widely used as adsorbents for the pollutants removal from aquatic environments [31]. Cellulose derivatives such as Methylcellulose (MC), hydroxypropyl methylcellulose (HPMC), ethylcellulose (EC) and carboxymethylcellulose (CMC) are used in the various nanocomposites preparation for water treatment process [32]. Using raw cellulose because of some disadvantages such as lower solubility, inability to reuse, poor interaction with pollutants and low surface area is limited. So, to solve this problem, the nanocomposites combined with ionic functionalization, ester, and the etheric cellulose derivatives are used which can produce advanced material based on the cellulose derivatives [33,34]. Among these derivatives, CMC is a frequently used cellulose derivative and soluble in the water, which is used as an adsorbent in the water purification process. CMC as a biopolymer is sensitive to pH changes [35]. Use of these adsorbents could make some limitations such as filtration, dispersion, turbidity, secondary contamination and high cost of regeneration. Therefore, create conditions for optimal use of CMC and other nanoscale or powdered adsorbents could make magnetic property in them through compression with metal ferrite compounds [36]. Polysaccharide-based composites owing to their impressive properties such as high surface area, chemical stability and mechanical strength have attracted great attention to themselves [37].
In this research CMC that applied in the synthesis of the HZSM, played an essential role to improve the CoFe2O4@CMC/HZSM-5 nanoadsorbent structural features. CMC in the CoFe2O4@CMC/HZSM-5 composition caused increasing the adsorbent surface area. As well as, carboxyl and hydroxyl groups in CMC create the electrostatic attraction between metronidazole and CoFe2O4@CMC/HZSM-5 composite surface. CoFe2O4@CMC/HZSM-5 was prepared with highly efficient microwave-assisted co-precipitation method by iron and cobalt salts on CMC in the presence of HZSM-5 and was used in highly efficient adsorption of the metronidazole. CoFe2O4@CMC/HZSM-5 that was used as a nanomagnetic adsorbent in the adsorption process to remove contaminants was easily separated from aqueous solutions as well as recycled and reused. According to the literature, the purpose of this investigation is preparation and characterization of CoFe2O4@CMC/HZSM-5 as a new nanomagnetic adsorbent for metronidazole removal.

Materials and equipment
All the required materials for this study, including carboxymethylcellulose (CMC), iron chloride (FeCl3.6H2O), cobalt chloride (CoCl2.6H2O) and nitric acid (HNO3), were purchased from Merck (Germany). All used solutions in the study were prepared from high purity deionized water (DW). The solutions pH were adjusted using chlorinated hydrochloric acid (HCl) and 0.1 molar sodium hydroxide (NaOH) and were measured by using pH meter (HACH-HQ-USA). In order to adsorbent magnetic separation from the solution, a cubic magnet with a magnetic intensity of 1.3 Tesla was used. The metronidazole was bought from DarouPakhsh Holding Company (Iran).

Preparation of CoFe2O4@CMC/HZSM-5
The magnetic nanocomposite was prepared using iron chloride (FeCl3.6H2O), cobalt chloride (CoCl2.6H2O) salts, carboxymethylcellulose (CMC) and /HZSM-5 zeolite. At first, salts of iron chloride FeCl3.6H2O and cobalt chloride CoCl2.6H2O (2:1) were dissolved in 100 mL of distilled water (DW) then, 1 g of CMC and finally, 1 g of HZSM-5 zeolite were added to the solution. Subsequently, NaOH was added to the resulting suspension for one hour until the solution pH reached 11. The reaction container color changed and became black after one hour of stirring. Then, the reaction container was transferred to the microwave and was irradiated by microwave waves. The radiation process was performed in three periods of five minutes with 50% power of the microwave and 30 seconds with the off mode. The obtained product was a light and black powder that was washed many times using DW and dried in the oven (100 °C within 24 h) ( Figure 1).

Characterization techniques of CoFe2O4@CMC/HZSM-5
Fourier-transform infrared spectroscopy (FTIR) was done using WQF-510 FT-IR spectrometer to determine the adsorbent's functional groups. Characterization of the specimens was done using a field-emission scanning electron microscope-energy dispersive X-ray spectroscopy-Mapping (FESEM-EDS-Mapping) (FE-SEM TESCAN MIRA3) and Transmission electron microscopy (TEM-Philips EM 208S). To find out the presence of the cobalt ferrite crystalline structure in the adsorbent, X-Ray diffraction (XRD) using Philips X-Pert device (the Netherlands) was employed. The magnetic property of the CoFe2O4@CMC/HZSM-5 were characterized by VSM (LakeShore Cryotronics-7404) at room temperature. To measure the porosity of the adsorbent surface area, Brunner-Emmett-Taller (BET) method and micrometrics model 021LN2 transfer device was used. Also, thermal gravimetric analysis (TGA) was applied to evaluate the adsorbent thermal resistance (TA Co. Model Q600-USA).

Adsorption Method
The metronidazole adsorption process experiments on the CoFe2O4@CMC/HZSM-5 nanocomposite were performed in the batch reactor and inside the Erlenmeyer flasks with a volume of 100 mL. At first, metronidazole stock solutions with a concentration of 1000 mg/L were prepared. Then, in order to investigate the synthesized adsorbent capabilities, 5 g of adsorbent was added to 50 mL metronidazole solution 50 mg/L and was placed in the shaker at 240 rpm for a certain period of time. After this period, by using magnetic separation, the adsorbent was separated from the solution and the metronidazole remaining concentrations were measured.
To recognize and measure the remained concentration of metronidazole, a HPLC device (YL 9100 Waters, USA) was used. Also, for adjusting the device Standard metronidazole with a purity of 99% was utilized. As well as, deionized water and acetonitrile were applied as mobile phases with volume ratios of 30:70. The utilized column was C18 with 5 μm particles, length of 250 mm, and internal diameter of 4.6 mm. By using a UV absorbance detector within the wavelength of 348 nm with an injection volume of 20 μL and flow rate of 1 mL/min, the metronidazole remained concentration was measured.
As control samples, the metronidazole samples with similar concentrations without adsorbent were utilized in the all steps of the experiment. The metronidazole initial concentration (50-300 mg/L), pH (3 -11), amount of adsorbent (0.5 -3 g/L), contact time (0 -300 min), and temperature (20-50oC) were investigated, to optimize the adsorption conditions and determine the optimal values of each of the adsorption process affecting factors. By solid addition method, the CoFe2O4@CMC/HZSM-5 point of zero charge (pHpzc) was evaluated. At first, 10 different pH (2)(3)(4)(5)(6)(7)(8)(9)(10)(11) values, 100 mL of KCl solution (0.1 mol/L) was prepared. Then, 0.01 g of the magnetic nanobiocomposite was added to each solution. After that, the obtained mixture was stored at room temperature overnight. In the next step, by using pH meter (Hanna), the initial and final pH (pHf) values of solutions were determined. To obtain the pHpzc, ΔpH = pHf -pHi and pHi plot was drawn. At last, the intersection point of ΔpH = 0 was taken as pHpzc. By using NaOH and HCl 0.1 N, the solutions pH was adjusted.
The adsorption experiments were repeated three times in each step and their mean values as the results were used in the calculations. At the end of each experiment, the amount of adsorbed metronidazole on adsorbent (qe) and the percentage of adsorption for each sample was calculated according to equations (Eq. 1) and (Eq. 2), respectively.
Where C0 (mg/L) is the adsorbate initial concentration (metronidazole), Ce (mg/L) is the pollutants equilibrium concentration and W is the adsorbent mass in the solution volume (g/L).

Adsorption kinetic equations
The kinetic equations are used to describe the behavior of the adsorbate molecules transfer per time. The adsorption process maybe takes place in several stages with different speeds. In the present study, four kinetic models, Lagergren (pseudo-first order) (Eq. 3), HO (pseudosecond order) (Eq. 4), Elovich model (Eq. 5) and intraparticle diffusion model (Eq. 6) were used.

Adsorption equilibrium equations
In the adsorption systems design, the equilibrium isotherms are used as important parameters. In fact, they describe relationship between the adsorbent adsorption capacity and adsorbate concentration. In the present study, two common models, the Freundlich (Eq. 7) and Langmuir (Eq. 8) model, have been used for this purpose. log qe= log kf + 1/n log Ce (Eq. 7) Ce /qe = Ce /q0 + 1/klq0 (Eq. 8) kF (mg/g(L/mg) and n are the Freundlich constants that 1/n is intensity of the adsorption process and kF is adsorbent adsorption capacity. As well as, kL (L/mg) is the Langmuir constant and q0 (mg/g) is the maximum amount of adsorbed materials per unit mass of adsorbent. RL is

FESEM-EDS-Mapping and TEM of CoFe2O4@CMC/HZSM-5
The surface morphology of the adsorption material was assessed by FESEM and TEM. The morphological result of CoFe2O4@CMC/HZSM-5 magnetic nanocomposite is illustrated in Fig. 3a-c. The CMC attendance in the CoFe2O4@CMC/HZSM-5 synthesis causes the smoothly, uniformly, and loosely aggregated sphere-shaped of magnetic nanocomposite form. The average particle size of CoFe2O4@CMC/HZSM-5 is about 27 nm.

(Figure 4)
Mapping is the observation of how the elements are distributed at high resolution. It was used to investigate the CoFe2O4@CMC/HZSM-5 elements distribution. Based on the obtained results which is showed in Fig. 5, Al, Si, Co, Fe, O, and C had a homogeneous distribution that indicates high uniformity of the synthesized CoFe2O4@CMC/HZSM-5 (Fig. 5).

XRD
The XRD patterns were prepared separately from CMC, HZSM-5, CoFe2O4, and CoFe2O4@CMC@HZSM-5 and were compared with each other (Figure 6). In the CMC XRD pattern, the index peak can be seen in the area of 18  -900-5813). The presence of sharp and strong peaks in this pattern and comparison of peak locations with reference data indicated the crystalline structure of cobalt ferrite with complete crystallization that has preserved the crystalline structure of CoFe2O4@CMC/HZSM-5, after being composite with carboxymethylcellulose and zeolite. The results revealed that the CoFe2O4@CMC/HZSM-5 magnetic nanocomposite had a ferromagnetic effect with elevated saturation magnetization and high magnetic coercive force. It could be quickly separated and retrieved from the reaction container using a magnet and could be reused in subsequent periods. (Figure 7)

BET
The obtained information from the BET test showed that surface area, pore volume, and pore size equaled 235.21 m²/g, 0.207 cm³/g, and 3.52 nm, respectively. The BET analysis results of CoFe2O4@CMC/HZSM-5 magnetic nanocomposite can be considered as a mesoporous material 3.1.6. TGA TGA was used to evaluate CoFe2O4@CMC/HZSM-5 thermal decomposition. The results are demonstrated in Fig. 8. TGA analysis was done in the room temperature to 600 ˚C, to evaluate the CoFe2O4@CMC/HZSM-5 nanocomposite thermal stability (Fig. 8). The mass loss was happened in the temperature range of 20-100 ˚C which is linked with the adsorbed water loss. Due to the breakdown of CMC glycosidic bonds and also, its decomposition into lower fatty acids, the weight loss was reported at 100-600 ˚C . According to the results, the adsorbent total weight loss was seen 10.18% until 600 ˚C which shows the CoFe2O4@CMC/HZSM-5 magnetic nanocomposite high thermal stability (Figure 8)

pH effect
Environmental acidity where the adsorption process takes place is largely effective in adsorption and removal efficiency. A part of this issue is related to hydrogen ions (H + ), which adsorbed on the adsorbent surface. The contaminant adsorption on the adsorbent surface leads to reduce the positive ions adsorption. The other part refers to the pH effect, which is effective in some compounds ionization. In order to evaluate effect of pH changes on the metronidazole removal by the CoFe2O4@CMC/HZSM-5 nanocomposite, 50 mL solution containing 2 g/L adsorbent and the metronidazole initial concentrations 50 mg/L were used. The results are presented in figure 9. The obtained results showed that the pollutant removal efficiency increased with increasing pH from 2 to 6 and decreased at pH values more than 6. The highest metronidazole removal efficiency took place at pH 6 and was equal to 94 %.

(Figure 9)
The effect of pH could be explain in terms of the adsorbent pHzpc and MNZ pKa. pHzpc of CoFe2O4@CMC/HZSM-5 was determined by solid addition method and was obtained 6.8. At pHs less than pHzpc, the CoFe2O4@CMC/HZSM-5 surface charge was positive; thereby the chance of MNZ ions adsorption increased. At pHs more than pHzpc, the surface charge of the CoFe2O4@CMC/HZSM-5 was negative. As a result, the repulsion between the interfaces increased in the solution [54]. On the other hand, high pH leads to higher concentrations production of -OH in the solution and also, inhibit of MNZ ions scattered, thereby the chance of MNZ adsorption decreased [54]. Therefore, pH 6 was selected as the optimum pH.

Equilibrium time
The equilibrium time for the metronidazole removal process was evaluated over a period of 300 min, metronidazole initial concentrations 50 mg/L and at the optimum pH ( Figure 10). The above figure shows that the adsorbates adsorption rate in the initial times was much higher than the next times or in other words, by increasing contact time, the adsorption capacity increased. But after 60 minutes, no significant changes were observed in adsorption capacity. For example, by increasing the contact time from 60 to 300 minutes, the metronidazole adsorption capacity increased from 22.5 and 23.5 mg/g to 22.6 and 23.7 mg/g, respectively. These results indicate that the metronidazole removal process is balanced in 60 minutes and therefore, this time was considered as the equilibrium time for the adsorption experiments continuation. Significant increase in adsorption capacity at the beginning of the contact time ( Figure 10) could be relate to the presence of many active and unsaturated sites on the CoFe2O4@CMC/HZSM-5 external surfaces. However, by increasing contact time, the adsorption capacity is stabilized. The stabilization of the adsorption capacity or reaching the equilibrium point in more than 60 minutes may be due to the active sites occupation on the adsorbent surface or the adsorbent capacity completion. In general, by increasing contact time, the adsorbat ions access to the empty active sites on the adsorbent surface is reduced and eventually these sites are saturated and the process reaches equilibrium.
( Figure 10) Figures 11a and b are showing the results of the adsorbent (CoFe2O4@CMC/HZSM-5) and adsorbate different amounts concentration (metronidazole) effects under optimal conditions, respectively. As shown in figure 11a, by increasing the amount of adsorbent from 0.5 to 3 g/L, the metronidazole adsorption increased from 73 to 94%. Therefore, considering that in the amount of 2g/L of adsorbent, the removal efficiency was 94%, this value was selected as the adsorbent optimal dose. But, in figure 11b, it can be seen that the changes in the initial concentration had a negative effect on the adsorption rate, such that by increasing the metronidazole initial concentration from 50 to 300 mg/L, the removal efficiency decreased from 95 to 59%, respectively. The removal efficiency increasing in the study of the amount of adsorbent changes (Figure 11a) maybe due to the adsorbent surface increasing or increase the access of the adsorbate molecules to the pores on the magnetic nanocomposite surface. On the other hand, figure 11b shows that by increasing the metronidazole initial concentrations, the removal efficiency decreased. Probably, constant numbers of active sites on the adsorbent against the increasing in the number of adsorbate molecules or the saturation of the adsorbent surface at high concentrations of contaminants was the main reason for the decrease in removal efficiency by increasing the initial contaminant concentration. (Figure 11)

Adsorption kinetic
The kinetic parameters amounts of the metronidazole removal process are presented in table 1. The computational adsorption capacity values (qe, cal) for pseudo-first order and pseudosecond order models were obtained 22 and 24.2 mg/g, respectively. In addition, the experimental adsorption capacity (qe, exp) from the obtained experiments was 23.5 mg/g. The lowest correlation coefficient was obtained for the Elovich equation and intraparticle diffusion model. The curves of the studied metronidazole removal kinetic models compared with the obtained experimental data from the experiments are presented in figure 12. The results of the metronidazole removal kinetic study by CoFe2O4@CMC/HZSM-5, according to the table 1 showed that based on the calculated correlation coefficients, the adsorption behavior per unit time follows pseudo-first order and pseudo-second order models. However, it is observed that the metronidazole correlation coefficient in the pseudo-second order kinetic was 0.997 that is higher than the obtained values for the pseudo-first order kinetic model. Therefore, it can be admitted that the model in the present study has more validity than the other three models. Tang et al. (2012) [55] in the study on aniline and caprolactam simultaneous adsorption by the granular activated carbon found that the pollutants adsorption behavior per unit time follows the pseudo-second order model. As well as, valderrama in the study of the phenol and aniline simultaneous adsorption by the activated carbon, reported both pseudo-first order and pseudosecond order kinetic models as suitable models for the data describing. According to table 1, the boundary layer thickness values (Ci) in the intraparticle diffusion model were not equal to zero, that indicates the intraparticle diffusion model was not the only controlling step of the process and perhaps the external adsorption mechanism may also had been effective as a control factor in the adsorption process.  Table 1) Table 2 shows the Langmuir and Freundlich equilibrium isotherms values for the metronidazole adsorption in the temperature range of 20-50 ° C. The above table indicates that the correlation coefficient in Langmuir and Freundlich models is R 2 >0.97. In this research, the maximum adsorption capacity (q0) based on the Langmuir model at 50 °C was obtained 206.6 mg/g. In figure 13, the curves of the Langmuir and Freundlich equilibrium isotherm models with experimental data were compered. In the metronidazole adsorption isotherms examination ( Table 2) it can be seen that the obtained regression coefficients in Freundlich and Langmuir models for all three temperatures are more than 0.97. In fact, these results indicate that both models are suitable for describing experimental test data. In the above table, it is also noteworthy that the pollutant maximum adsorption capacity values based on the Langmuir isotherm model for the increased by increasing the solution temperature. For example, by increasing temperature from 20 to 50 o C, the maximum adsorption capacity increased from 201.1 to 206.6 mg/g. This trend indicates that the increase in the temperature had a positive effect on the adsorption capacity. In addition, the Langmuir model RL values of the adsorbat were studied at all temperatures and was obtained less than one (1<RL<0). Therefore, it is concluded that the CoFe2O4@CMC/HZSM-5 had high efficiency in the metronidazole removal. Table 2 also shows that the 1/n values related to the Freundlich isotherm were less than one in all three temperatures, which indicate the high adsorption of the pollutant.  Table 2)

Adsorption thermodynamic
The results of measuring Lnkc values versus temperature changes based on the Vant Hoff curve to determine the values of the metronidazole adsorption process thermodynamic parameters are shown in figure 14. Also, table 3 presents the obtained values of the thermodynamic parameters. Table 3 demonstrates that the standard enthalpy (∆H o ) and standard entropy values (∆S o ) are positive and the standard gibbs free energy (∆G o ) is negative.
Due to the presented results in table 3, it can be seen that the standard enthalpy values (∆H o ) is positive, which indicates that the metronidazole removal process is endothermic and adsorption at the higher temperatures is desirable. Standard entropy value (∆S o ) was obtained positive which indicates the adsorbent tendency to adsorb the adsorbate and increase the efficiency by increasing the temperature in the solid-liquid interface during the adsorption process. Nevertheless, the Gibbs free energy values (∆G o ) were negative in all three studied temperatures. These negative values indicate that the metronidazole removal by CoFe2O4@CMC/HZSM-5 is spontaneous reaction. Since the metronidazole ∆H o value was less than 40 kJ/mol, it can be concluded that the adsorption process had a physical mechanism. In addition, the ∆G o obtained values at all the studied temperatures were less than -20 kJ/mol, which confirm that the adsorption process was a physical type.

Effect of CMC on the removal efficiency
The obtained metronidazole removal efficiencies by CoFe2O4, CMC, HZSM-5, and CoFe2O4@CMC/HZSM-5 were compared ( Figure 15). Based on the findings, the removal efficiencies by using HZSM-5, CMC, CoFe2O4, and CoFe2O4@CMC/HZSM-5 were 75%, 83%, 87%, and 94%, respectively. CoFe2O4@CMC/HZSM-5 showed a significantly higher removal efficiency than the other adsorbents after 60 min contact time. It can be concluded that CoFe2O4@CMC/HZSM-5 composed of carboxymethylcellulose as a polysaccharide with high carboxyl and hydroxyl groups. Consequently, the presence of these functional groups causes the interaction of electrostatic attraction between the adsorbent surface and metronidazole [56,57]. As well as, CMC plays an important role in improving the structural properties of CoFe2O4@CMC/HZSM-5 magnetic nanocomposite. The presence of CMC during the preparation of CoFe2O4@CMC/HZSM-5, increases the adsorbent surface area and provides more active sites for the metronidazole removal. (Figure 15)

Mechanism
The metronidazole adsorption efficiency depends upon to the surface groups of the magnetic nanoadsorbent protonation and non-protonation status. The pH of solution is one of the most important parameters which can qualifies the CoFe2O4@CMC/HZSM-5 charge and has effect on the interaction between metronidazole molecules. To determine the amount of CoFe2O4@CMC/HZSM-5 magnetic nanoadsorbent surface charge and its type, at different pHs, pHZPC was investigated. The synthesized magnetic nanoadsorbent surface charge was negative at pHs more than 6.8, although at the lower pHs, the adsorbent surface charge was positive. Metronidazole due to its electron-rich aromatic rings has desire to adsorb on the adsorbent positively charged surface. Metronidazole pKa was equal to 2.55, it created an electrostatic gravity energy which leaded to more metronidazole removal by the CoFe2O4@CMC/HZSM-5. The amount of metronidazole molecules which have negatively charged increased slowly by increasing the pH value in the solution [58]. With attention to the CoFe2O4@CMC/HZSM-5 adsorption behavior at different pH values and its characteristics, the main mechanism for the metronidazole removal perhaps makes a complex between metronidazole anions and the adsorbent surface. The metronidazole removal mechanism by CoFe2O4@CMC/HZSM-5 is shown in figure 16.

Regeneration, reusability and chemical stability of CoFe2O4@CMC/HZSM-5
Regeneration and recycling are two important items in the adsorbent functional. After metronidazole removal, CoFe2O4@CMC/HZSM-5 magnetic nanocomposite, were recycled for six consecutive adsorption runs. To regenerate CoFe2O4@CMC/HZSM-5, this magnetic nanocomposite, following each cycle, was separated from the solution by a magnet and was washed with HCl/DW for several times. At last, the magnetic nanoadsorbent was dried in an oven at 60°C. The dried magnetic nanoadsorbent was reused to start the next cycle of metronidazole adsorption. Base on the figure 17a, the metronidazole removal efficiency after six recycling cycles, reduced from 94% to 86%. On the other hand, the pollutant adsorption percentage from the first to last cycles demonstrated solely 12% reduction that can be because of the magnetic nanocomposite mass over the several cycles. Based on the obtained results from the CoFe2O4@CMC/HZSM-5 regeneration, a slight decrease in the removal efficiency can be related to the occupation of active adsorbent sites by the metronidazole and decreasing in the amount of CoFe2O4@CMC/HZSM-5 during the washing and regeneration processes. Therefore, due to the high recyclability of CoFe2O4@CMC/HZSM-5, its application is economical in terms of operational costs. The CoFe2O4@CMC/HZSM-5 stability was investigated by FESEM (Fig. 17b) band XRD (Fig. 17c) analysis. Indicates the CoFe2O4@CMC/HZSM-5 XRD pattern after the sixth run of the magnetic nanoadsorbent recycling. The XRD analysis demonstrated that the CoFe2O4@CMC/HZSM-5 magnetic nanoadsorbent, did not have obvious changes except in the intensity of diffraction peaks following the sixth run of magnetic nanoadsorbent recycling. Also, the morphology of the magnetic nanadosorbent has been preserved and did not have significantly change. The results showed the chemical stability of the magnetic nanoadsorbent after sixth run recycling.

Removal of metronidazole from real wastewater
The optimal conditions were determined. After that, the experiment was done on the campus wastewater at Kerman University of Medical Sciences. The real sample physicochemical characterizations were obtained and its specifications were achieved COD (415 mg L -1 ), BOD (247 mg L -1 ), TSS (20 mg L -1 ), TDS (545 mg L -1 ), TKN (78 mg L -1 ), Phosphate (16.7 mg L -1 ), Nitrate (16.5 mg L -1 ), Sulfate (352 mg L -1 ), and Metronidazole (45 mg L -1 ). The metronidazole maximum removal efficiency was 85% in the adsorbent 2 g/L, pH 6, initial metronidazole concentration 45 mg/L and contact time 60 min. Due to the presence of cations, anions, TDS and TSS in the real wastewater, the metronidazole removal efficiency decreased.
The performance of CoFe2O4@CMC/HZSM-5 magnetic nanoadsorbent in the pharmaceutical compounds removal was compared with the other magnetic nanoadsorbent. It can be concluded that CoFe2O4@CMC/HZSM-5 magnetic nanoadsorbent in comparison with the other magnetic nanoadsorbent, had a higher pollutants removal efficiency in a shorter time, a lower dose of adsorbent and at the relatively neutral pH. Also, CoFe2O4@CMC/HZSM-5 magnetic nanoadsorbent had a high specific surface area and the higher adsorption capacity.

Conclusion
The results of the present study showed that the using magnetic nanoparticles with the aim of magnetic property induction in widely used adsorbents, especially CMC, is a simple, fast and practical method due to the lack of problems belong to separation, filtration and centrifugation. CMC plays an important role in improving the structural properties of CoFe2O4@CMC/HZSM-5 magnetic nanocomposite. The presence of CMC during the preparation of this nanocomposite, increased the adsorbent surface area and provided more active sites for the metronidazole removal. The optimal conditions in the present study were obtained at pH 6, contact time 60 min, adsorbent 2 g/L and temperature 20 °C. The metronidazole removal efficiency and maximum adsorption capacity based on the Langmuir equilibrium model were obtained 94 and 206.6 mg/g, respectively. The adsorption efficiency increased by increasing contact time, the adsorbent amount and solution temperature, while the metronidazole initial concentrations had an indirect effect on the adsorption efficiency. Magnetized nanocomposite due to its high specific surface area can be used as an effective and cost-effective adsorbent to remove environmental pollutants from water sources. Therefore it is expected it use for quickly and easily separate different types of adsorbents from solutions. According to the obtained results, it could be concluded that the adsorption process was endothermic and had a physical mechanism. Finally, the chemical stability and reusability of the synthesized CoFe2O4@CMC/HZSM-5 magnetic nanoadsorbent were evaluated in the metronidazole removal. The achieved findings demonstrated that following six adsorption runs, he adsorbent chemical structure had no change and the magnetic nanoadsorbent efficiency in the removal process had a slight decrease.