Fabrication of Nano Zero valent Iron/Biopolymer Composite with Antibacterial Properties for Simultaneous Removal of Nitrate and Humic Acid: Kinetics and Isotherm Studies

This investigation compared the adsorption behavior of humic acid (HA) on cellulose, chitosan and nano zerovalent iron/chitosan (nZVI/chitosan). Results show that nZVI/chitosan is very effective in the adsorption of HA from aqueous media. The feasibility of using nZVI/chitosan as an adsorbent for the simultaneous removal of nitrate and HA from aqueous media was also studied. Structural analyses of the samples were identified by TEM, FT-IR, EDX, XRD and N2 isotherms. The effects of pH, amount of composite, nitrate concentration, HA concentration and contact time and their interactions on responses were explored by central composite design (CCD) and response surface methodology (RSM). The optimal conditions of pH (5.5), adsorbent amount (0.098 g), reaction time (27 min) and initial concentrations (110 mg/L for nitrate and 30 mg/L for HA) were obtained from the desirability function. The adsorption properties of the resulting nanocomposite toward nitrate and HA were investigated through kinetic and isotherm studies. The adsorption kinetics was found to fit the pseudo-second order model. The obtained results indicate that nitrate uptake fitted well with Langmuir model while Freundlich isotherm was the best model for describing the multilayer uptake of HA from aqueous solutions. Moreover, nZVI/chitosan nanocomposite illustrates a very high antibacterial activity against pathogen bacteria strains such as Staphylococcus aureus ATCC 25935, ATCC 25923, and Pseudomonas aeruginosa ATCC 27853. The findings reported in this investigation highlight the potential of using nZVI/chitosan as a promising adsorbent for the simultaneous removal of nitrate and HA from aqueous solutions.


Introduction
In the process of agricultural and industrial development, environmental and microbial pollutions have become the most universal serious problems, so quality and security of drinking water are the tacit requisites for human society [1][2][3]. An extensive range of environmental pollution such as phenolic compounds, HA, polycyclic aromatic hydrocarbons, nitrate, and heavy metals (lead, arsenic…) in water have become a serious concern for aquatic life and public health. Among them, nitrate and HA are regarded as perilous contaminants in the drinking water [4][5][6][7][8]. Nitrates surely cause harm to the ecosystem and the various aquatic organisms leading to many epidemic diseases including methemoglobinemia, stomach cancer, diabetes, eutrophication, and so on [4,6,9]. As a result of the primary problems that increased nitrate concentrations in water makes, the US Environmental Protection Agency (USEPA) has set the maximum contaminant level (MCL) of nitrate to 10 mg/L in drinking water [4,5].
HA is a sort of organic macromolecule that exists in soil and sources of drinking water storage and has a negative surface charge due to carboxylic (-COOH) and phenolic (-OH) groups which are linked to aromatic rings [7,8,[10][11][12]. The presence of HA in the water can lead to a yellowish to brown color, problems in taste and odor that are aesthetically unpleasant and can also lead to Blackfoot diseases, cancer, and goiter. Furthermore, HA can be assumed to have effects on disinfection efficiency. It can react with chlorine during water treatment and make strong carcinogenetic disinfection byproducts (DBPs) such as trihalomethanes [3,11,13]. Therefore, because of the negative effects of HA, the USEPA has established the maximum contaminant level (MCL) of 2 mg/L for HA in drinking water. On the other hand, HA also acts as a layer for bacterial growth in the water distribution system so the microbial pollution is another obstacle caused by pathogenic Gram-negative and positive microorganisms like Staphylococcus aureus (S. aureus), Escherichia coli (E. coli) as well as Pseudomonas aeruginosa (P. aeruginosa). Therefore, the removal of nitrate and HA from polluted water is very important to restore water quality [3,13].
Various treatment methods have been applied to remove nitrate and HA from water including coagulation [3,14], ion-exchange [14,15], adsorption [15][16][17] and membrane technology [16,18]. Most of these methods are not suitable due to the production of secondary waste and high cost [16].
In the past few years, nanoparticles have been used to remove a wide range of pollutants owing to their high adsorption capacity, large surface area, and other physicochemical properties. Nanoparticles such as zero valance iron (nZVI) have potential advantages over existing nanoparticles such as their high removal efficiency as well as their antibacterial properties. Moreover, these nanoparticles have magnetic properties that can easily and rapidly remove pollutants from water [16,[18][19][20].
However, the bare nZVI tends to accumulate and oxidize, thereby reduces the reactivity surface area, and performance. As a result, its use in water treatment causes problems. Recently biopolymer-based coatings are one of the most commonly used methods to overcome clumping agglomeration of nanoparticles and to enhance their separation [21].
Various biopolymers as iron nanoparticle coatings are used such as starch [22], alginate, cellulose, and chitosan [23,24]. Among these biopolymers, cellulose and chitosan have recently received much attention due to their high uptake performance, excellent biodegradability, good biocompatibility, non-toxicity, low price, high abundance, as well as strong affinity towards certain contaminants to outfit their removal [25][26][27]. Ahmadi et al. indicated that the coating of nZVI with chitosan resulted in increased stability and adsorption capacity [23]. Furthermore, biopolymers such as cellulose and chitosan are effective adsorbents for nitrate and HA adsorption due to excellent functional groups (hydroxyl and/or amine groups) in their chains [28,29]. In fact, these biopolymers have been widely applied both as adsorbents of these contaminants and support for nZVI in treatment processes [9,[30][31][32][33][34]. However, to the best of our knowledge, there is no report about removing simultaneous HA and nitrate using a low-cost impressive, antibacterial composite system of nZVI/chitosan. In this current study a comparison between the adsorption behaviors of cellulose, chitosan and nZVI/chitosan toward HA was performed. Also, magnetic nanocomposite combined with the specific surface area of chitosan, the antibacterial capacity of chitosan and nano zero valent iron, and magnetic separation property of nano zero valent iron is synthesized for simultaneous removal nitrate and HA. The surface physicochemical properties and presence of elements and functional groups of the biopolymers and magnetic nanocomposite are explored by different methods. A CCD was chosen to check the effects of five different factors (pH, contact time, composite dose, nitrate concentration, and HA concentration) on the simultaneously removal of nitrate and HA by nZVI/chitosan. To achieve the controlling mechanism of the uptake system, the adsorption kinetics and isotherms were assessed.
Furthermore, the antibacterial activity of the nZVI/chitosan composites against the E. coli, S. aureus and P. aeruginosa was investigated. Finally, the feasible antibacterial mechanism, reusability and reproducibility of the nZVI/ chitosan biocomposite was also assessed.

Preparation of nZVI/Chitosan Composite
The nZVI/Chitosan was prepared using a chemical reduction technique (reducing Fe 2+ to Fe 0 using NaBH 4 ) [23,35]. At the beginning, a solution of chitosan (0.5%) in 2% acetic acid was prepared. Due to the negligible solubility of chitosan, the solution was blended for 4 h with stirring until it quite dissolved. In the following, 10 mmol (2.78 g) of FeSO 4 ·7H 2 O was added to 50 ml of chitosan aqueous solution (0.5%) under the N 2 -purged atmosphere and the solution is rapidly stirred for thirty minutes. Afterward, 50 ml of an aqueous solution containing 30 mmol (1.14) of NaBH 4 in the form of drops was added into the above solution under stirring and nitrogen atmosphere. The chemical reduction reaction is exhibited in Eq. (4): After 60 min, the black solid is separated by magnetism. At first, it is washed with deoxygenated water and then with pure ethanol and acetone, respectively. The nZVI/chitosan was dried under a vacuum situation at 70 °C.

Characterization
FTIR spectra were registered on a Fourier transform infrared spectrometer over the wavenumber range of 4000-400 cm −1 (Shimadzu, 8400S). The core-shell structure of chitosan-nZVI supported was investigated by model transmission electron microscopy (TEM, EM208S, Philips). Furthermore, X-ray energy dispersive spectroscopy was used to analyze the composition of the composite (TESCAN-VEGA III). The crystal structure of cellulose, chitosan and nZVI/chitosan were obtained by X-ray diffraction (XRD, X, per PRO model-Dron-8 diffractometer). According to the N 2 adsorption-desorption isotherm (Micromeritics, ASAP 2020), the pore size distribution was achieved by using the Barrett-Joyner-Hallenda (BJH) method and the specific surface area was obtained by using the Brunauer-Emmett-Teller (BET) method. The magnetic property of nZVI/chitosan was distinguished using Vibrating Sample Magnetometer (VSM, 7400, Lakeshore).
The concentration of nitrate in a solution containing natural organic matter samples was colorimetrically determined by using a UV-Vis spectrophotometric in which nitrate was analyzed using the Vanadium reduction method at 540 nm [36]. Moreover, the concentration of HA was obtained via a UV-Vis spectrophotometer at 420 nm [32].

Antibacterial Testing
The antimicrobial activities of the chitosan and nZVI/chitosan were determined using the 1.510 8 colony-forming units (CFU)/ml of E. coli and S. aureus in an agar medium. Then, a disc with a 1 cm diameter of chitosan and nZVI/ chitosan (0.08 g) was placed in the culture medium containing the bacteria. The media containing bacteria and the disc was located at 4 °C for 2 h to complete the diffusion of the disc. Followed by the culture mediums were incubated at 37 °C for 24 h.

Analytical Methods
To investigate the simultaneous removal of HA and nitrate by nZVI/chitosan, a binary nitrate-HA system was used in the experiments. Therefore, the stock solutions of the nitrate and HA were prepared by the dissolution of analytical-grade KNO 3 and HA in distilled water, respectively. All experiments were carried out in various conditions, according to the designed experiments, in a 50 ml flask on a shaker (150 rpm) to find the optimum initial nitrate and HA concentrations, composite dose, contact time, and pH. Furthermore, all pHs were adjusted with Britton-Robinson (BR) buffer solution. CCD experiments surveyed and appraised the effect of individual variables as well as their conceivable interactions on the removal percentage of nitrate and HA as responses. The equilibrated samples were taken out and the concentration of nitrate and HA in the solution was simultaneously checked by UV-Vis spectrophotometer. Moreover, the adsorption kinetic and isotherm ware investigated for co-adsorption in binary systems.The removal percentage and adsorption capacity of nitrate and HA were determined with Eqs. (5) and (6), respectively [37].
where C 0 (mg/L) and C e (mg/L) are the initial and final concentrations of concentration (i.e., HA and nitrate) in the aqueous solution, respectively. q e (mg/g) is the equilibrium amount of nitrate and HA adsorbed per unit mass of adsorbent, m (g) is the dose of adsorbent and V (L) is the volume of nitrate and also HA solution.

Experimental Design
RSM consists of a group of statistical and mathematical techniques that are effective in developing, improving, and optimizing a process. CCD is the most popular RSM which optimizes the effective experimental variables with a minimum number of experimental runs, in addition to analyzing the interaction between variables [38][39][40][41][42].
In this study, a CCD was used for RSM to obtain the emphasis of the influences of the variables (i.e., nitrate concentration, HA concentration, pH, contact time, composite dose) on the responses (i.e., the percentage removal of nitrate and HA) by nZVI/chitosan.
Five numerical factors were studied in five levels including two levels for axial point (± α), two levels for high/low levels (± 1), and one level for the center point. Six replicates were made for center point and 21 for not center, which made a CCD with overall 27 experimental runs. These five variables together with their respective ranges were chosen based on the literature and our preliminary studies. The experimental data were analyzed and validated for the removal percentage of nitrate and HA on the nZVI/chitosan, and each response was correlated with the most suitable model developed from the quadratic model (Eq. 6): where Y is the predicted response (removal percentage of either HA and nitrate), xi's are the input variables that are studied for every experimental run. The parameters b 0 is the model constant, b i (i = 1, 2,…, n) is the linear coefficient, b ii (i = 1, 2,…, n) is the quadratic, b ij (i = 1, 2,…, n, j = 1, 2,…, n) is interaction coefficient. Design Expert 11.0.3.0 software was applied for the actuarial analysis of results attained from the experimental design.

Physicochemical Characterization of Adsorbents
FTIR measurements were carried out to illuminate the stabilization mechanism and gain better insight into the interactions between the different functional groups of chitosan and the nZVI particles. Figure 1 showed the FT-IR spectra of cellulose, chitosan and nZVI/chitosan. In the cellulose (( Fig. 1a), the absorption band at 1650 cm −1 is associated with -OH bending of the adsorbed water. The band at 1160 cm −1 was attributed to C-O stretching in the acetyl group. The characteristic band was located at 1055 cm −1 , assigned to the pyranose ring skeletal vibration of the C-O-C. The principal bonds in the IR spectra of chitosan (Fig. 1b) can be seen as follows [43]: a broad and strong overlapped band around 3446 cm −1 (O-H and N-H stretch); a weak band at 2873 cm −1 (C-H stretch), 1645 cm −1 (N-H bending vibration), 1386 cm −1 (-C-O stretching of the primary alcoholic group), 1083 cm −1 (C-O stretching vibrations) [35,[44][45][46]. According to Liang's studies [47], the adsorption of chitosan molecules on the surfaces of iron nanoparticles causes considerable changes in the tensile frequency of the chitosan functional groups.
It can be seen from Fig. 1c that several remarkable changes have occurred in the composite spectrum compared to the chitosan spectrum, which may confirm the accuracy of the composite synthesis. The stretching bands of the hydroxyl and amino groups shift from 3446 to 3426 cm −1 for nZVI/chitosan showing that the N-H and O-H vibration was affected because of the iron attachment. Moreover, the N-H bending vibration shifts from 1645 to 1627 cm −1 and accompanied by a decrease in intensity. The position of C-O peaks shift from 1386 and 1083 cm −1 to 1300 and 1056 cm −1 in the composite, respectively [48]. On the other hand, the peak around 613 cm −1 in the nZVI/chitosan, was attributed to the Fe-O stretching vibration implying that the nano zerovalent iron was successfully prepared and introduced into the chitosan [23].
The N 2 isotherms of cellulose, chitosan and nZVI/chitosan are demonstrated in Fig. 2. The BET specific surface area for cellulose was equal to 0.85 m 2 /g and the average pore diameter was 89.16 nm. The BET surface areas of chitosan and nZVI/chitosan were found to be 0.97 and 63 m 2 /g, respectively. The average pore diameters of chitosan and nZVI/chitosan were found to be 90.22 and 13.29 nm, respectively. It was obvious that the surface area of chitosan increased when it was coated on the nZVI. The BET surface area of the nZVI/chitosan was greater than that of chitosan, most likely due to the fact that the nanoparticles of nZVI had a high surface area.
The XRD patterns of samples in the 2θ range of 5-70° are shown in Fig. 3. Three major peaks in the XRD pattern of cellulose were observed at 2θ = 15°, 24° and 35°, reflecting the low degree of crystallinity of cellulose. Figure 3b shows the XRD data of the chitosan. It is apparent from the XRD results that chitosan displays typical peaks at 2θ = 11° and 23°. As shown in Fig. 3c, nZVI/chitosan indicates the main peaks at the 2θ of 46.7° and 66.25° due to the presence of metallic Fe [23,35,49].  The elemental composition of cellulose, chitosan and nZVI/chitosan were evaluated using EDX spectrum analysis and results were depicts in Fig. 4. The EDX spectra displayed the presence of C and O in the cellulose and the presence of C, O, and N in the chitosan. The spectrum of nZVI/chitosan in Fig. 4 illustrates peaks of C, O, and N in addition to Fe which is the result of successful modification of Fe nanoparticles. In addition, the EDX elemental mapping image of synthesized nanocomposite reveals that Fe species are incorporated uniformly in the synthesized nZVI/chitosan. TEM analysis was performed to accurately investigate morphology. Figure 5 illustrates that almost spherical Fe nanoparticles are formed inside the chitosan templates and the nZVI because the presence of chitosan which can be well diffused. Furthermore, the core-shell structure of the composite is well indicated, the polymeric layer of chitosan is the clear space as the shell and the dark space corresponding to the iron nanoparticles as the core [44].
A magnetic hysteresis curve of nZVI/chitosan composite at room temperature and in the fields from −10,000 to 10,000 Oersted is shown in Fig. 6. The saturation magnetization (Ms) of nZVI/chitosan was found to be 35.11 emu/g. As can be seen, the nZVI/chitosan displays superparamagnetic characteristic behavior at room temperature. When an outer magnetic field was placed under the aqueous solution including the nZVI/chitosan composite, the substance can be quickly attracted and separated from the solution, therefore preventing environmental pollution. Similar results for the superparamagnetic behavior of nZVI were reported in the literature [15].
The above results show that zero valent iron nanoparticles have been successfully loaded onto chitosan and the nZVI/chitosan composite could provide a high surface area and accessible for simultaneous removal of nitrate and HA. Furthermore, nZVI/chitosan could be applied as a magnetic composite to remove contaminants from an aqueous solution leading to preventing secondary pollution.

Study on HA Adsorption by Adsorbents
The adsorption isotherms of HA adsorption on the cellulose, chitosan and nZVI/chitosan are displayed in Fig. 7. These adsorbents all demonstrated enhanced HA uptake with the increase of initial concentrations of HA solutions and gradually reached saturation. It is clear from Fig. 7 that the chitosan possessed a higher adsorption performance than that of the cellulose. The possible adsorption mechanisms of HA on chitosan involved different kinds of interactions. The adsorption process involved electrostatic attraction between the Other mechanisms such surface adsorption and van der Waals interaction also played a key role during the adsorption process. The nZVI/chitosan displays much further HA uptake than the pristine chitosan. The composition of chitosan by the nZVI increased the specific surface area, which facilitated the fast mass transfer of HA inside the nZVI/chitosan structure and improved the contact between HA and binding sites. The nZVI/chitosan was used as a suitable adsorbent for further investigation (Table 1).

RSM Model Fitting
The experimental results of nitrate and HA removal percentages with independent variables [A = pH, B = contact time, C = nitrate concentration, D = HA concentration, E = nanocomposite dose] were established in Table 2.
The results were collected using the standard polynomial regression method and make it conceivable to obtain the following Eqs. (7 and 8) for nitrate and HA, respectively: Tables 3 and 4 show the results of the analysis of variance (ANOVA) of nitrate and HA, respectively. The results of the ANOVA test for response nitrate removal efficiency have indicated that the model was of great significance (F value = 3444.28, P value < 0.0001). The adaptability of the multinomial model equation for nitrate ions fitting was declared the coefficient of designation predicted R 2 = 0.9728 and adjusted R 2 = 0.9996. The high adjusted (8) ANOVA data of HA (Table 4) indicates P value < 0.0001 and F value 87.88 that establish the high yield and suitability of the model for appropriate and interpretation of empirical data, while the coefficients of definition for its multinomial model equation are predicted R 2 = 0.8426 and adjusted R 2 = 0.9858.

Investigation of Interaction of Parameters with 3D Graphs
Figure 8a-f displays the simultaneous effect of different parameters on the percentage removal of nitrate and HA.
As shown in Fig. 8a, the percentage removal of nitrate in solution decreased with increasing pH from 3 to 8 and enhanced with the amount of composite from 0.05 to 0.1 g. The enhancement of nitrate removal in acidic pH could be due to the following reasons: (1) Acidic condition helps to eliminate the iron oxide formed in the surface during nZVI oxidation and constantly makes the fresh surface of composite exposed to the solution [60]. (2) Acidic solution is favorable for nZVI corrosion, so the oxidation process of zero-valent iron is dramatically increased and the electrons more readily available for nitrate reduction [60,61].     Figure 8b demonstrates that the percentage removal of HA increased to pH about 7 after that it is almost constant, while the dose of composite in ranging from 0.05 to 0.1 g. Generally, the adsorption behavior of HA on the nZVI/chitosan depends on the surface charge of HA and the nZVI/ chitosan at different pHs. The pH pzc of the nZVI/chitosan and HA are 7.83 and 1.6, respectively. Accordingly, HA removal could occur through two processes: (1) At pHs less than 7.83, the surface of the nZVI/chitosan is positively charged and the adsorption of HA mainly occurs via electrostatic interaction.
(2) At greater pH than 7.83, due to the negatively charged surface of the composite, the electrostatic adsorption was very limited and adsorption occurs through complex formation between phenolic and carboxyl groups of HA and iron oxide on surface nZVI [61,63], as shown in the following: For better investigation of the interaction between HA and nZVI/chitosan in the presence and absence of nitrate, FT-IR spectra of the composite after reaction with HA and the HA/  55 26 nitrate binary system were studied under acidic conditions (Fig. 9). The FTIR spectrum of HA has been reported by the previous literature [64]. Figure 9a illustrates the FTIR spectrum of nZVI/chitosan after adsorption of HA. Based on the observations, the C-O peak at 1056 cm −1 shifted to 1040 cm −1 compared to the HA. Furthermore, the displacement of the N-H peak from 1627 to 1600 cm −1 and the change in its intensity can indicate electrostatic interaction. These findings suggested that carboxylic or phenolic functional groups of HA might be attracted through electrostatic by the surface of composite or be complexed with iron oxides. Figure 9b indicates the IR spectrum of the composite after simultaneous reaction with nitrate and HA. By comparing the two spectra ( Fig. 9a and b), it can be seen that the presence of nitrate reduced the rate of HA removal in acidic conditions due to competition between two contaminations for the adsorption on composite active sites. Figure 8c and d presents the removal percentage of nitrate and HA increased with increasing the dose of composite from 0.05 to 0.1 g and with reaction time ranging from 25 to 60 min. In general, with increasing the dose of nZVI/ chitosan, the number of active sites for reaction increases, resulting in higher removal efficiency. Moreover, because of the high kinetics of the reaction, the contact time did not have much effect. Figure 8e and f displays the percentage of nitrate and HA removal by nZVI/chitosan as a function of the concentration of these two pollutants in the binary system. As shown in Fig. 8e with increasing the concentration of HA, the percentage of nitrate removal is reduced owing to the accumulation of iron oxides-HA complexes formed on the nZVI/chitosan surface that prevents the nitrate mass transfer [63]. On the other hand, Fig. 8f depicts that with an enhancement of the nitrate concentration, because of the enhancement possibility of iron oxides-HA complexes formation, the percentage of HA removal increases [61,63].

Response Optimization
The process variables were optimized to obtain the maximum removal percentage of nitrate and HA simultaneously by nZVI/chitosan with a quadratic model in the studied   (Fig. 11).
According to the previous reports, the properties of antibacterial chitosan depended on the molecular weight and solubility in water [50,51]. Chitosan alone has weak antibacterial properties, so the addition of certain metals such as iron improves the antibacterial properties of chitosan [52][53][54]. Therefore, this antimicrobial activity of nZVI/chitosan is effectively relevant to the presence of loaded nano zero-valent iron nanoparticles. Three possible mechanisms have been suggested for the nZVI/chitosan based on the release of iron ions from the composite (Fig. 12). In the first mechanism, iron can be adsorbed by the electronegative molecules of the bacterial wall, resulting in bacterial destruction [55]. The second mechanism is based on oxidative stress generated by reactive oxidative species (ROS) [56]. The reaction of zero-valent iron nanoparticles with intracellular oxygen leads to ROS generation and eventually disrupting cell membranes [56][57][58]. The third mechanism involves the passage of iron ions through the bacterial cell and interaction with DNA [55,59]. Accordingly, our findings illustrate that nZVI/chitosan indicated antibacterial properties which would kill the harmful bacteria present in the contaminated water.

Adsorption Kinetics
The adsorption rate is a vital parameter to assess the efficiency of an adsorbent for the removal of contaminates. The adsorption of nitrate and HA onto the nZVI/chitosan was explained through interparticle diffusion, pseudofirst-order and pseudo-second-order kinetic models. The linearized forms of mentioned models are as follows, respectively [65, 66]: 1 3 Fig. 10 Antibacterial activities of nZVI/chitosan Fig. 9 FT-IR spectrums of nZVI/chitosan after reaction with HA (a) and the HA/nitrate (b) binary system at acidic conditions where K i (g/mg min 0.5 ), K 1 (l/min) and K 2 (g/mg min) are the rate constants of intraparticle diffusion, pseudo-first and pseudo-second order models, respectively; q t (mg/g) is the time dependent amount of nitrate and HA adsorbed per unit mass of nZVI/chitosan.
The obtained parameters with correlation coefficients (R 2 ) of three kinetic models are given in Table 5. The pseudo-second-order model seemed to describe the kinetic data better as compared with the intraparticle diffusion 1 q e t and pseudo-first-order models for fitting the kinetic data of nitrate and HA uptake, which implied that the nitrate and HA removal are a chemical adsorption process.

Adsorption Isotherms
Adsorption isotherms illustrate the correlation between the adsorbent and adsorbates, which is decisive for optimizing the adsorption procedure [16]. In order to attain the uptake performance of nZVI/chitosan for nitrate and HA in adsorption systems, Langmuir, Freundlich, and Temkin isotherm models were employed in this work [18]. The linear forms of the Langmuir, Freundlich, and Temkin isotherms are illustrated by the following Equations, respectively. [16,[19][20][21].
where q m (mg/g) and K L (l/mg) are the maximum monolayer adsorption capacity and Langmuir constant, respectively; R L is the separation factor; n and K F (l/g) are constants representing the heterogeneity factor and the adsorbent capacity, respectively; B T and A T are the Temkin constant, respectively. The results of adsorption isotherm for nitrate and HA in aqueous systems are shown in Table 6. It can be observed from Table 6 that R 2 obtained from the Langmuir model was greater than Temkin and Freundlich models for nitrate in aqueous systems. According to the results, nitrate uptake on nZVI/chitosan mainly took place in a monolayer uptake pattern. As for HA, the high value of R 2 supports the best and successful applicability of Freundlich isotherm for fitting equilibrium data related to HA adsorption onto nZVI/chitosan. The result suggested a heterogeneous distribution of adsorption sites on the nZVI/chitosan surface.
(14) C e q e = 1 K Lq m + 1 q m C e R L = 1 1 + K L C 0 (15) lnq e = ln K F + 1 n lnC e (16) q e = B T ln A T + B T lnC e Fig. 11 Diameter values of inhibition for nZVI/chitosan against pathogenic bacteria Fig. 12 The possible mechanism antibacterial of nZVI/chitosan The heterogeneity was caused by the attendance of numerous functional groups (derived from HA) on the adsorbent surface, and also by the diverse mechanism of liquid-solid interaction. As a result, it allowed multi-layer uptake of HA. The obtained R L values were between 0 and 1 that illustrates desirable adsorption. Furthermore, the magnitude of 1/n in the Freundlich model is a measure of adsorption intensity, displays desirable adsorption when 1/n is < 1. Both of the 1/n values were < 1 for nitrate and HA in the adsorption system, indicating a beneficial uptake for nitrate and HA on nZVI/chitosan.

Regeneration and Reusability of nZVI/Chitosan Composite
To evaluate the possibility of recycling and reuse of nZVI/ chitosan nanocomposite, regeneration experiments were performed for 4 cycles. After each cycle, the composite was collected magnetically from the solutions and were washed using NaOH (0.1 M) for 30 min. The removal efficiency of nitrate and HA by regenerated nZVI/chitosan composites in each cycle is shown in Fig. 13. It can be seen that the removal percent decreased moderately from 90.0 to 65.6% and 98.1 to 80.5% for nitrate and HA after 4 cycles, respectively. The good reusability demonstrates that nZVI/chitosan has a great potential in practical application.

Conclusions
In summary, here we presented for the first time, nZVI/ chitosan composite as an antibacterial adsorbent for simultaneous removal of nitrate and HA. The synthesized composite has a core-shell structure with high surface area and inhibition of growth S. aureus ATCC 25935, ATCC 25923, and P. aeruginosa ATCC 27853. Besides, the VSM result indicated that the nZVI/chitosan is easily collected by an external magnetic field and separates the contaminants from the aqueous media. CCD was utilized to identify the optimal conditions of the simultaneous removal process and obtain maximum efficiency. To achieve maximum removal percentage (90.0% for nitrate and 98.1% for HA), from optimization of process modeling, the optimum reaction time, adsorbent amount, pH, initial nitrate concentration, and initial HA concentration were found to be 27 min, 0.098 g, 5.5, 110 mg/L, and 30 mg/L respectively. On the basis of kinetic studies, the pseudo-second-order model could describe the nitrate and HA uptakes on nZVI/chitosan well, which refers to the chemisorption mechanism. The isotherm equilibrium data fitted well with Langmuir and Freundlich models for nitrate and HA uptake, respectively. Consequently, our laboratory studies suggest that this antibacterial composite with fast separation has great potential for the simultaneous removal of nitrate and HA.