Amoxicillin and cefixime simultaneous adsorption by facile synthesized chitosan@polyacrylamide@ZIF-8: isotherm and kinetic study

Abstract

In recent years, taking medicine has been increasing around the world due to population growth and the spread of disease. Antibiotics as a kind of these medicines include about 10 to 15% of drug consumption. Studies show that antibiotics, which are not completely removed, have a permanent and destructive effect on the environment. In this study, the chitosan@polyacrylamide@ZIF-8 as an adsorbent was simultaneously used to remove cefixime and amoxicillin from the aqueous solution by adsorption methods. XRD and FTIR patterns were analyzed to investigate the surface crystallinity and the chemical properties of the adsorbent. In the following, SEM and TEM images were used for surface morphology study. Finally, a numerical investigation was done for predicting the adsorption effective parameters. The results show that pH 4, contact time 30 min, initial concentration 50 mg/L, temperature 25 °C and 15 mg of adsorbent in 10 mL of polluted water are optimal conditions at which a removal percentage of up to 90% was achieved. The Langmuir, Freundlich, Temkin, and Dubinin–Radushkevich isotherms were examined by experimental data and Langmuir model presented the best consistency with the experimental results and the maximum calculated adsorption capacity for amoxicillin and cefixime are 910 mg/g and 588 mg/g, respectively. Pseudo-first order, pseudo-second order and intraparticle diffusion were investigated for kinetic studies. The results of the calculations illustrated the pseudo-second order model has been dominant. Overall, this study expresses, chitosan@polyacrylamide coated by ZIF-8 can be used an appropriate, effective, and economical adsorbent for the adsorption of amoxicillin and cefixime.

Graphical Abstract

Introduction

Nowadays, water pollution has become a global concern. This causes access limitation to usable water. Therefore, scientists and researchers are seeking appropriate and efficient solutions for water treatment. Some sources that cause water pollution are drugs, even in low concentrations [1]. Among them, antibiotics play an extensive role in various medicines which have a widely endangered influence on public health [2]. The studies show antibiotics have a permanent and destructive effect on the environment which is not completely removed in refineries. [3]. Aquatic environments are influenced by different sources including direct discharge from municipal wastewater remedy plants, human waste, direct disposal of medical, veterinary and industrial squander. In order to eliminate the destructive effects of active bacteria in the treatment, it is necessary to remove the drugs before they enter the wastewater treatment plant [4]. Antibiotics are classified according to their structures; one of the important categories is the antibiotics with a beta-lactam ring in their structure. Accordingly, antibiotics are divided into beta-lactam and non-beta-lactam rings; 32.6% of them belong to the beta-lactam group (penicillin, amoxicillin, ampicillin and cefixime) [5, 6]. Amoxicillin (AMX) is one of the most important groups of antibacterial drugs, which is prescribed for the infections treatment of the body and as an antiviral drug for humans. Even very low concentrations of AMX can pollute the water sources. The molecular structure of AMX is shown in Fig. 1 [7].

Fig. 1

Molecular structure of AMX

Cefixime (CFX) is one of the most important antibiotics used to treat infections and also effectively used against various bacterial organisms and infections, including staphylococcus, hemophilia, Escherichia coli, streptococcal fever, tonsillitis and throat infections [8]. The CFX molecular structure is shown in Fig. 2 [9].

Fig. 2

Molecular structure of CFX

Numerous methods including, reverse osmosis [10], flocculation and coagulation [11], chemical oxidation [12], optical decomposition [13], biological treatment and adsorption have been developed for the treatment of wastewater containing antibiotics [14, 15]. Adsorption process due to low operating costs, high flexibility, insensitivity to toxic compounds and contaminants, lack of secondary material production in the system, high recovery, low consumption of adsorbent, high efficiency and specific surface area for removal, is a promising process to reduce pharmaceutical, heavy metal and dyes [16, 17]. Pham and et al. [18] investigated the adsorption of beta-lactam CFX onto strong polycation, polydiallyldimethylammonium chloride modified nanosilica, and also the adsorption of AMX was examined with MnO₂ nanosheets were in situ grown on biomass carbon microspheres. Also, Adriano and et al. investigated the adsorption of AMX on chitosan beads how have been determined in batch studies. [19]. Adsorption of CFX and lamotrigine on HKUST-1/ZIF-8 nanocomposite was investigated by Emami et al. [20].

These works indicated that using appropriate adsorbent is critical in removal processes, so synthesizing the new adsorbent to simultaneous removal of multi adsorbent is the key term. Metal–organic frameworks (MOFs) have become interesting and important adsorbent for the adsorption of medicines, heavy metals and dyes, mainly due to their high porosity surface area. The framework structure of ZIF-8 is intercalated into zinc metal and 2-methylimidazole ligands. ZIF-8 remarkably increases the adsorption efficiency due to its multifaceted and unique structure [21]. ZIF-8s are collected from tetrahedrally coordinated mutated metal ions such as: Co, Fe, Zn and Cu, coupled by imidazolate linkers [22]. Because AMX and CFX are widely used in the treatment of many diseases, the presence of these drugs in the solution is very destructive, so their removal from the solution before entering the environment is essential [23].

In this study, by adding ZIF-8 to the composite structure of chitosan@polyacrylamide, chitosan@polyacrylamide@ ZIF-8 (CPZ) was synthesized and used as a novel adsorbent to adsorption of AMX and CFX. As strong point, CPZ was used successfully for removing this antibiotics, simultaneously. Herein, we developed a facile and cost-effective strategy to synthesize a composite adsorbent (CP) modified by ZIF-8 (CPZ) that showed excellent adsorption properties. In addition, in examining the influence of effective parameters in the adsorption process, the performance of the CPZ composite was also superior to the CP. Physical properties of CP and CPZ, explored by XRD, FTIR, TEM and SEM. in order to specify optimum parameters, influence different factors such as: the quantity of adsorbent, initial concentration, temperature, contact time and pH in the adsorption process. Isotherm and kinetic models were investigated to determine the predominant equilibrium data and rate of process throughout the adsorption, respectively.

Materials and methods

Materials

Chitosan (medium molecular weight, 96%), acetic acid and amoxicillin provided by Sigma Aldrich, Switzerland. Acrylamide, ammonium persulfate, 99%, N,N,N,N-tetramethylethylenediamine (TEMED), methylenebisacrylamide, zinc nitrate, 99% and 2-methylimidazolate were purchased from Merck, Germany, and CFX become acquired from Acros. Other chemicals and analytic grade reagents were commercially available and used in their pure form. The stock solutions were prepared fresh for each experiment.

Preparation of polyacrylamide–chitosan (CS-PAA)

In order to prepare the chitosan/polyacrylamide adsorbent, the first 5 g of chitosan was added to 60 mL of distilled water and stirred for 15 min to obtain a homogeneous suspension. 10 g of acrylamide monomer was added to the suspension, 30 mL solution agitated for 5 min. 1 g of methylene bisacrylamide and 250 mg of ammonium persulfate were dissolved in 30 mL of distilled water, and added to suspension. Finally, 300 μL of tetramethylethylenediamine (TEMED) was added and polymerized at 25 °C. After the polymerization process is completed, the polyacrylamide–chitosan gel is washed with refined water until the outlet water reaches pH 7 and dried at room temperature. The synthesized adsorbent was analyzed in order to characterize the morphology of the composite, and then it was prepared to carry out the adsorption process.

Synthesis of chitosan@polyacrylamide@ZIF-8 (CS-PAA-ZIF-8)

1 g chitosan was added to 20 mL of 1% (v/v) acetic acid solution and blended for 1 h. 2 g of acrylamide was dissolved completely in 10 mL of deionized water, then added to the solution and mixed for 1 h. 400 mg of methylbisacrylamide and 80 mg of ammonium sulfate were dissolved in 10 mL of distilled water, then added to the original solution and stirred for 2 h. 0.86 g of zinc nitrate was dissolved in deionized water and it was added to the solution and mix for 1 h, and then dispersed for 20 min in the ultrasonic bath. 0.96 g of 2-methylimidazole was added into 10 mL of distilled water and after dissolving, was added to the suspension and mixed for 1 h and then dispersed for 15 min in the ultrasonic bath. After raising it with deionized water, the adsorbent (CPZ) was dried at 50 °C and prepared for the next steps of experiments. The main adsorbent (CPZ) was synthesized and then it was prepared for the adsorption process.

Characterization

In order to determine the surface morphology, scanning electron microscopy (SEM, SNE-4500M, accelerating voltage of 10–20 kv, SEC, South Korea) was utilized. The X-ray diffract meter (XRD, X’Pert Pro, Germany) was done in the range of 2θ = 0°–80° and operating conditions were 40 kV, λ = 1.54 oA, 30 Ma and 25 oC and used CuKa reflection as a radioactive resource. To identify the CP and CPZ functional groups, Fourier Transform Infrared spectrometry (FTIR, Spectrum 100, PerkinElmer CO.) analysis was done within the run of 500–4000 cm⁻¹ utilizing KBr strategy.

Batch experiments procedure

Co-adsorption method was used to adsorption of AMX and CFX from aqueous media. Influential parameters to achieve optimal conditions have been investigated for pH in range between 2 and 8, the amount of adsorbent range between 5 to 25 mg, initial concentration range between 25 and 200 mg/L, contact time 5 and 40 min and temperature between 25 and 65 °C. In similar and previous studies appropriate dosage was approximately as mentioned range, also many tests were carried out to obtain virtual ranges. All experiments were carried out with 10 mL of AMX and CFX solutions, and then adsorbents were added to the prepared solutions and centrifuged at 5000 rpm for 20 min. Finally, the obtained adsorption capacity calculated by following equation [24]:

$$ {\text{qe}} = \left( {{\text{Ci}} - {\text{Ce}}} \right) \times V/M, $$

(1)

where q_e (mg/g) is the capacity of adsorbent to adsorption of CFX and AMX at equilibrium, Ci and Ce (mg/L) are adsorbate concentrations at equilibrium and initial time, respectively. V(L) is volume of concentration and M(g) is mass of adsorbent.

Removal efficiency was determined as follows [25]:

$$ {\text{Removal efficiency}}\left( \% \right) = \left( {{\text{Ci}}{-}{\text{Ct}}} \right) \times 100/{\text{Ci,}} $$

(2)

where Ci is initial equilibrium concentrations and Ct is the equilibrium concentrations of adsorbate.

Results and discussion

Characterization study

Figure 3 shows the XRD pattern of CP and CPZ composite. The peak formed in position 2Ɵ = 20 indicates the presence of Chitosan in the initial composition. Also, the peaks in positions 28 and 35 demonstrate polyacrylamide in this composite. The XRD pattern of the modified adsorbent containing ZIF-8 is shown in Fig. 3. It can be realized by comparing the XRD pattern of CPZ and CP that, in addition to illustrated peaks in CP composite, the peaks have also are present in positions 10 and 13 of CPZ pattern. Obviously by adding a substance to the initial structure an amorphous structure is organized. The peak in the range of 17 to 25 is more broad in Fig. 3, Finally, according to XRD pattern, the present of ZIF-8 in CPZ adsorbent is confirmed [26, 27].

Fig. 3

XRD pattern of CP and CPZ

The FTIR spectrum of the CP and CPZ is indicated in Fig. 4. As shown in Fig. 4a, the presence of a peak in 1068 cm⁻¹ of the C–O group tensile vibration in chitosan is shown. As shown in Fig. 4a, a peak in cm⁻¹1068 indicates the presence of the tensile vibration of the C–O group related to chitosan. The adsorption band at 12,925 cm⁻¹ illustrates C–H aliphatic tensile group in acrylamide and a band at 3400 cm⁻¹ corresponds the N–H and O–H groups in chitosan and the amide group in acrylamide. Peak appearance at 1700–1600 cm⁻¹ illustrates the overlap of symmetric stretching C=O of chitosan and acrylamide and also adsorption band at 3190 cm⁻¹ indicates NH₂ bonding in acrylamide [28]. Asymmetric stretching band of CH₃ related to chitosan appeared at1400 cm⁻¹. According to Fig. 4b, the peaks in 757 cm⁻¹ and 699 cm⁻¹ show Zn–O and Zn–N bonds, respectively, which confirm that ZIF-8 was loaded in the CPZ structure. The peak at 3134 cm⁻¹ corresponding to C–H bond of aromatic and aliphatic tensile of imidazole ring and also, the peak at 1110 cm⁻¹ regarding to the tensile vibration of ZIF-8, has not been seen in the CS-PAA spectrum, so it confirms the presence of ZIF-8 in the composite clearly [29, 30].

Fig. 4

FTIR spectra of Ch-PAA (a) and Ch-PAA-ZIF-8 (b)

Figure 5 shows the SEM image of the CP (a, b) and CPZ (c, d). Both of synthesized adsorbents have rough and non-smooth surface. CPZ has a rougher surface than CP which increases the specific surface area of the CPZ adsorbent compared to CP. Furthermore, porosity can also be seen in CPZ surface which has improved the adsorbent properties. Another factor that observable in the SEM review is the proper distribution of ZIF-8 in the absorbent structure and there is not phase separation among the ZIF-8 and composite. Finally, the existence of ZIF-8 in CPZ structure led to adsorption performance enhancement by effective parameters such as specific surface area and porosity of the final composite [31].

Fig. 5

SEM image of CP (a, b) and CPZ (c, d)

TEM images related to CP and CPZ are shown in Fig. 6. In Fig. 6a, the physical accumulation of chitosan and polyacrylamide nanoparticles, suitable dispersion and distribution of particles related to chitosan@polyacrylamide is obvious. Dark spots and regular particle size in Fig. 6b indicate the presence of ZIF-8 in the chitosan–polyacrylamide polymer matrix. The regular distribution of ZIF-8 particles is seen in the composite structure and in addition, significant aggregation is not seen in the image [32].

Fig. 6

TEM image of CP (a) and CPZ (b)

pH influence on AMX and CFX removal

By examining the pH in the range of 2–8, removal of AMX and CFX with chitosan@polyacrylamide@ZIF-8 was determined and reported and the results are shown in Fig. 7. With increasing pH between 2 to 4, the percentage removal was enhanced and up to pH 4, it was decreased. Experiments illustrated the highest removal of AMX and CFX occurred at pH 4, it occurs because of, at this pH, adsorbates are ionized and take on negative charge [33]. Therefore, due to the positive charge of chitosan and polyacrylamide [amine groups (NH₂) of chitosan, polyacrylamide and ZIF-8 take on positive charge and released NH₃⁺ and Zn²⁺ in the aqueous solution] and electrostatic interaction is established between the species and the adsorbent. Also, at pH < pk_a values, the drugs become protonated and due to the positive charge of the adsorbents, a repulsive force is created, so the adsorption is reduced (pk_a,_CFX = 3.45 and pk_a,_AMX = 4.18) [34, 35]. The experimental data show that in all of the pH values, the adsorbent contain ZIF-8 had more removal percentage than the CP. Asadi et al. investigated the pH of CFX removal from wastewater using UVC/sodium persulphate system in the range of 3–11 and optimum was reported pH = 3 [36].

Fig. 7

pH influence on the percentage of AMX and CFX removal (initial concentration: 50 mg/L, amount of adsorbent: 15 mg, 40 min and 25 °C)

Effect of adsorbent dosage

Amount of CP and CPZ as adsorbent for removal of AMX and CFX is shown in Fig. 8. The results indicated that with increasing the amount of CP and CPZ to 15 mg the removal percent is increased and then remained constant. With increasing the amount of adsorbent, the adsorbent surface and the active site for adsorption increases, so the removal percentage of AMX and CFX raised. Being constant or even decreasing the adsorption capacity of AMX and CFX with increasing the amount of CP and CPZ more than 15 mg is mainly because of the persistence of unsaturated sites during the process and also, overlap of active sites on the adsorbent. So 15 mg of CPZ and CP determined as an optimum amount and was utilized to continue the adsorption process. According to Fig. 8, CPZ had higher efficiencies than CP in all values [37, 38].

Fig. 8

Amount of adsorbent efficacy on the percentage of AMX and CFX removal (initial concentration: 50 mg/L, pH = 4, 40 min and 25 °C)

Adsorption experimental results in AMX removal used magnetic adsorbent by Jafari and et al. [39]. Research indicated that the removal efficiency reaches its maximum using 0.05 g/L of the adsorbent.

Effect of initial concentration

The influence of initial concentrations 25 to 200 mg/L on the adsorption rate was investigated. According to Fig. 9, with increasing the initial concentration, the removal percentage increased until 50 mg/L then it reached to the optimal adsorption point and decreased. At low concentrations, due to the low driving force the removal percentage was low, with increasing the initial concentration to 50 mg/L, the adsorbent is almost saturated with the drugs and with increasing the concentrations, active site was saturated by adsorbates, so it caused reduction in the removal rate. Overall, the results demonstrate that CPZ has higher removal of AMX and CFX than CP, indicating success in modifying CP adsorbent by adding ZIF-8 [40]. Initial concentration effect of CFX removal by GO–chitosan composite in batch media was investigated by Zeinep and et al. [41]; in the range of 20–100 mg/L and optimum was obtained 42 mg/L.

Fig. 9

Effect of initial concentration on AMX and CFX removal (amount of adsorbent: 15 mg, pH = 4, 30 min and 25 °C)

Effect of temperature

As shown in Fig. 10, with increasing temperature in all samples, decreasing in adsorption was achieved. In general, CPZ illustrated better adsorption for AMX and CFX than CP in all temperatures. The reduction in the rate of adsorption for CPZ was lower than CP, which proves the success of adsorbent modification and also improving in adsorption properties. In all tests, with increasing temperature the removal rate was reduced, so it can be found that AMX and CFX adsorption on CP and CPZ are exothermic [42]. Removal of AMX via hollow fiber was investigated by Pirom and et al. [43] and it was found that by increasing the temperature of the system from 278.15 to 318.15 K, extraction of AMX increased from 81.81 to 89.65%

Fig. 10

Effect of temperature on the removal of AMX and CFX (amount of adsorbent = 15 mg, initial concentration: 50 mg/L in 5 mL of solutions, pH = 4, 30 min)

Effect of contact time

The variation of contact time for adsorption of AMX and CFX simultaneously by CPZ is shown in Fig. 11. Over time up to 1 h, about 90% adsorption of CPZ adsorbent is take place for both AMX and CFX. As can be seen in the figures, at first 10 min the removal rate was increased rapidly and the adsorption process was completed until 30 min, then equilibrium state was achieved.

Fig. 11

Effect of contact time on AMX and CFX removal (initial concentration: 25 to 200 mg/L, amount of adsorbent: 15 mg in 5 mL of solutions, pH = 4, and T: 25 °C)

Isotherm study

Experimental data for the adsorption of AMX and CFX on chitosan @ polyacrylamide@ZIF-8 were plotted by linear regressions, such as Langmuir, Freundlich, Temkin and Dubinin–Radushkevich models. Langmuir model is based on monolayer and homogeneous surface and the adsorption energy, remains constant during reaction. Langmuir model expressed by following equations:

$$ q_{{\text{e}}} = \frac{{q_{{\text{m}}} \cdot k_{{\text{l}}} \cdot C_{{\text{e}}} }}{{1 + k_{{\text{l}}} \cdot C_{{\text{e}}} }}. $$

(3)

The above equation linearized by following [44]:

$$ \frac{{C_{{\text{e}}} }}{{q_{{\text{e}}} }} = \frac{1}{{k_{{\text{l}}} \cdot q_{{\text{m}}} }} + \frac{1}{{q_{{\text{m}}} }} \cdot c_{{\text{e}}} , $$

(4)

where k_L (L/mg) is the Langmuir constant, q_m (mg/g) and q_e (mg/g) are maximum capacity and is equilibrium capacity of adsorption, respectively, C_e (mg/L) is the equilibrium AMX and CFX concentrations in residual.

The maximum adsorption capacity of AMX and CFX were calculated 910 and 588 mg/g, respectively (Table 1).

Table 1 The values of parameters for isotherm models at 25 °C

The Freundlich model expresses that the adsorption process occurs on heterogeneous surfaces. It can be found from the equation that the adsorption energy is decreased exponentially during the adsorption. The Freundlich model is expressed by the following equations:

$$ q_{{\text{e}}} = k_{{\text{f}}} \cdot C_{{\text{e}}}^{{{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 n}}\right.\kern-0pt} \!\lower0.7ex\hbox{$n$}}}} . $$

(5)

It can be linear by following [45]:

$$ \log q_{{\text{e}}} = \log k_{{\text{f}}} + \left( \frac{1}{n} \right)\log C_{{\text{e}}} , $$

(6)

where K_f is the distribution coefficient and n is a correction factor and their values were predictable by Lnq_e vs. LnC_e plot, are indicated in Table 1. As it is known, the value of n in this model was 2.28, which indicates that the desired adsorption has occurred.

Temkin isotherm assumes that the heat adsorption is linear and ignores very low and high concentrations and considers a homogeneous boundary energy distribution up to a portion of the maximal bonding energy.

Its expresses by equation below [46]:

$$ q_{{\text{e}}} = B\left( {{\text{ln}}A} \right) + B\left( {{\text{lnCe}}} \right). $$

(7)

The correlation coefficient of the Temkin model is lower than other isotherms, which means experimental data do not justify the Temkin isotherm (Table 1).

The Dubinin–Radushkevich model expresses information about the adsorption type by the following equation:

$$ {\text{Ln}}q_{{\text{e}}} = {\text{ln}}q_{{\text{m}}} - \beta \varepsilon^{2} , $$

(8)

where q_m(mg/g) is adsorption capacity, β and ε are D–R constant and Polanyi potential, respectively, that are dependent on the equilibrium concentration [47, 48]:

$$ \varepsilon = {\text{RTln}}\left( {{1} + {1}/{\text{Ce}}} \right). $$

(9)

Results are presented in Table 1. The average free energy (E) in this model for AMX and CFX is calculated at 0.707 and 0.408 kJ/mol, respectively. If the calculated E is less than 8 kJ/mol, adsorption is physical. Therefore, physical adsorption is achieved in this work [49]. According to Table 1, the lines plotted for the Langmuir model had better conformity to experimental data of CFX and AMX, so the Langmuir model was superior.

As seen in Table 1, the lines plotted for the Langmuir model had better conformity to experimental data of CFX and AMX so the Langmuir mod was superior.

Maximum adsorption capacity of AMX and CFX on different adsorbents is listed in Table 2.

Table 2 Maximum adsorption capacity of various adsorbents used in the removal of CFX and AMX from aqueous solutions

The results in Table 2 revealed chitosan@polyacrylamide@ZIF-8 has a more adsorption capacity than other adsorbents in the removal of AMX and CFX.

Adsorption kinetics

Being aware of the dynamic reaction and investigation of adsorption conditions are important in the kinetic process, several kinetic models for experimental data have been utilized to determine the adsorption mechanism and control step. To determine the kinetic type equation governing the adsorption process, intraparticle diffusion, pseudo-first-order and pseudo-second-order kinetic models were used. Weber and Morris provide the intraparticle diffusion velocity based on the function t^0.5 equation related to the Intraparticle diffusion model are expressed as follows [55]:

$$ q_{t} = k_{p} \cdot t^{0.5} + C. $$

(10)

K_p and C are intraparticle diffusion velocity constant parameters that depend on the kinetics of the boundary layer. The calculated values with the correlation coefficient are presented in Tables 3 and 4 for AMX and CFX, respectively, demonstrated the intraparticle diffusion model is one of the steps determinative adsorption rates of CFX and AMX on CPZ.

Table 3 Values of kinetic models parameters at different concentrations for AMX

Table 4 Values of kinetic models parameters at different concentrations for CFX

The pseudo-first-order kinetic model is presented by the following equation [56]:

$$ {\text{Ln}}\left( {q_{{\text{e}}} - q_{{\text{t}}} } \right) = {\text{ln}}\left( {q_{{\text{e}}} } \right) - k_{{\text{l}}} \cdot t. $$

(11)

Linear plotting Ln(q_e − q_t) in terms of t, and the results are also collected in Tables 3 and 4, for AMX and CFX. The pseudo-first-order kinetic model indicates the rate of occupancy of the adsorbent active sites is proportional to the number of vacancies.

As can be seen in Tables 3 and 4, the large difference between the equilibrium adsorption capacity values (q_e) calculated from the pseudo-first order kinetic model (q_cal) and the experimental values (q_exp) for AMX and CFX indicates an inconformity between the results and experimental data were from pseudo-first-order kinetic model.

Pseudo-second-order model suggests the surface adsorption rate control phase from share or exchange of electrons between the adsorbent and the adsorbate. The linearized form of the pseudo-second-order equation is indicated by the equation below [57, 58]:

$$ \frac{t}{q} = \frac{1}{{k_{{\text{h}}} \cdot q_{{\text{e}}}^{2} }} + \frac{t}{{q_{{\text{e}}} }}. $$

(12)

k_h is a constant of the pseudo-second-order (g/mg min). The adsorption of AMX and CFX on CPZ was plotted and the values of q_e and k_h were calculated by obtaining the slope and intercept of the lines drawn in different concentrations are given in Tables 3 and 4.

The low difference between the adsorption capacity of experimental data (q_exp) and calculated from the pseudo-second-order kinetic model (q_cal), for both AMX and CFX and also, higher correlation coefficient than others kinetic models, indicated the pseudo-second-order equation is an appropriate model for describing kinetics process of AMX and CFX on CPZ. Therefore, among all studied kinetic models in this work, the pseudo-second-order is predominant and the data have followed this model. Fakhri and et al. [59] investigated the adsorption characteristics of cephalosporins antibiotics and results showed the pseudo-second-order kinetic model with the activation energy of 7.6682 and 5.3745 kJ/mol for cephalexin and cefixime, respectively.

Conclusion

In the present study, the modified chitosan–polyacrylamide adsorbent was prepared by ZIF-8 for the simultaneous adsorption of AMX and CFX from aqueous solution. FTIR, XRD, TEM and SEM analysis revealed ZIF-8 was well placed in the structure of chitosan–polyacrylamide and the modified adsorbent had good porosity for adsorption. Various tests were performed to achieve the optimal parameters which: pH = 4, contact time 30 min, initial concentration 50 mg/L, temperature 25 °C and 15 mg of adsorbent are optimal values for both AMX and CFX adsorbents were determined. The linear isotherm models such as Langmuir, Freundlich, Temkin and Dubinin–Radushkevich were studied and the Langmuir isotherm prevailed in the comparison of data. Accordingly, maximum adsorption capacity for AMX and CFX with chitosan@polyacrylamide@ZIF-8 obtained 910 and 588 mg/g, respectively, which indicates the physicality adsorption process. Adsorption process kinetic was utilized to evaluate the reaction rate by kinetic models. The results illustrate that the experimental data had a better agreement with the pseudo-second-order kinetic model. The investigation of different parameters in all experiments revealed, chitosan@polyacrylamide@ZIF-8 had more removal percent than chitosan@polyacrylamide. Therefore, chitosan@polyacrylamide composite modified by ZIF-8 can be chosen as a proper adsorbent for adsorption of drugs due to its high efficiency.