Removal of Arsenic from Wastewater by Using Nano Fe3O4/Zinc Organic Frameworks

Efficient removal of arsenic in wastewater is of fundamental importance due to the increasingly severe arsenic pollution. In this study, a new composite adsorbent (Fe3O4@ZIF-8) for As(V) removal from wastewater was synthesized by encapsulating magnetic Fe3O4 nanoparticles into metal organic frameworks. In order to evaluate the feasibility of Fe3O4@ZIF-8 as an adsorbent for As(V) removal, the adsorption properties of Fe3O4@ZIF-8 were systematically explored by studying the effects of dosage, pH, adsorption isotherm, kinetics, and thermodynamics. Additionally, the characterization of Fe3O4@ZIF-8 before and after adsorption was analyzed thoroughly using various tests including SEM-EDS, XPS, BET, XRD, TG, FTIR, and the properties and arsenic removal mechanism of the Fe3O4@ZIF-8 were further studied. The results showed that the Fe3O4@ZIF-8 has a specific surface area of 316 m2/g and has excellent adsorption performance. At 25 °C, the initial concentration of arsenic was 46.916 mg/L, and pH 3 was the optimum condition for the Fe3O4@ZIF-8 to adsorb arsenic. When the dosage of the Fe3O4@ZIF-8 was 0.60 g/L, the adsorption of arsenic by the Fe3O4@ZIF-8 can reach 76 mg/g, and the removal rate can reach 97.20%. The adsorption process of arsenic to the Fe3O4@ZIF-8 can be well described by the Langmuir isotherm model and the second-order kinetic equation. At pH 3 and temperature 298 K, the maximum adsorption capacity of arsenic by the Fe3O4@ZIF-8 was 116.114 mg/g. Through the analysis of thermodynamic parameters, it is proved that the adsorption process of arsenic by the Fe3O4@ZIF-8 is a spontaneous endothermic reaction. The Fe3O4@ZIF-8 has broad prospects for removing As(V) pollution in wastewater, because of its strong adsorption capacity, good water stability, and easy preparation.


Introduction
Human survival and development are inseparable from water resources. With the continuous expansion of human productivity and the continuous development of the manufacturing industry, the demand for mineral resources continues to increase, and the generated pollutants enter the water body, resulting in more and more serious water pollution [1]. Arsenic is a non-metallic element, which widely exists in nature [2]. At present, hundreds of arsenic minerals have been found. Arsenic and its compounds are used in pesticides, herbicides, insecticides, and various alloys [3,4]. In 2017, the World Health Organization's International Agency for Research on Cancer placed arsenic and inorganic arsenic compounds on the list of Class I carcinogens [5]. In 2019, arsenic and arsenic compounds were included in the "List of Toxic and Harmful Water Pollutants (First Batch)" by the Ministry of Ecology and Environment. The sources of arsenic pollution in water bodies are wastewater, waste gas and industrial waste discharged from industrial production, various organic arsenic pesticides used in agricultural production to avoid 3 of 15 China) were purchased from Guangzhou Chemical Reagent Factory. Deionized water is used throughout this study. The conductivity of H 2 O was less than 2 µs/cm. As(V) stock solution was prepared with 1001 µg/mL As(V) standard solution (H 3 AsO 4 , Beijing). The inductively coupled plasma atomic emission spectroscope (ICP-AES) was used to measure the As(V) concentration in the aqueous solutions.

Synthesis of Magnetic Fe 3 O 4 Nanoparticles
Magnetic Fe 3 O 4 nanoparticles were synthesized using a modified iron (II) and iron (III) co-precipitation method as described in the literature [21]. FeCl 2 ·4H 2 O (10 mmol) and FeCl 3 ·6H 2 O (10 mmol) were gradually poured into 30 mL deionized water and mixed by mechanical stirring, the solution was heated to 60 • C for 30 min in the ultrasonic cleaner. Then 40.00 mL of 3.50 mol/L ammonia solution was introduced into the mixed iron(II)/iron(III) solution with continuous stirring at 60 • C for 30 min. After the reaction was completed, the produced mixture was filtered and washed with deionized water until neutral and finally dried at 60 • C under the vacuum for 12 h to obtain Fe 3 O 4 nanoparticles.

Synthesis of Fe 3 O 4 @ZIF-8
The preparation of Fe 3 O 4 @ZIF-8 was improved according to reference [22]. Weighed 3.30 g of 2-methy imidazole and 0.10 g of magnetic nanometer Fe 3 O 4 , dissolved in 70.00 mL of methanol. The mixture was ultrasonicated for 15 min, 1.50 g of Zn(NO 3 ) 2 ·6H 2 O was weighed and dissolved in 70.00 mL of methanol, which was slowly added dropwise to the above-mentioned sonicated solution, mechanically stirred for 24 h, and then centrifuged. The obtained solid was washed with methanol at least three times, and then vacuum-dried at 80 • C for 12 h. It can be seen in Figure 1 that when the dosage of Fe3O4@ZIF-8 was 10 mg, the removal rate of As(V) reached 76.29%. When the dosage of Fe3O4@ZIF-8 was increased to 15 mg, the removal rate of As(V) reached 97.20%. When the dosage of Fe3O4@ZIF-8 was 20-35 mg, the removal rate of As(V) reached more than 99%, and the removal was basically complete. Continue to increase the dosage of Fe3O4@ZIF-8 to As(V) the removal rate did not improve much. Therefore, from the perspective of economy and resource saving, the optimal dosage of Fe3O4@ZIF-8 in this experiment is 15 mg/25 mL, which is 0.6 As illustrated in Figure 1, the equilibrium adsorption capacity of Fe 3 O 4 @ZIF-8 for As(V) shows decreasing trend with the dosage. It was because that the increase in the dosage of Fe 3 O 4 @ZIF-8, that the number of adsorption sites in water also increased, but the adsorption sites on the surface of Fe 3 O 4 @ZIF-8 had not reached saturation. Therefore, although the amount of adsorbed As(V) increased, the degree of saturation decreased, and Fe 3 O 4 @ZIF-8 the equilibrium adsorption capacity of As(V) decreased.

Results and Discussion
It can be seen in Figure 1 that when the dosage of Fe 3 O 4 @ZIF-8 was 10 mg, the removal rate of As(V) reached 76.29%. When the dosage of Fe 3 O 4 @ZIF-8 was increased to 15 mg, the removal rate of As(V) reached 97.20%. When the dosage of Fe 3 O 4 @ZIF-8 was 20-35 mg, the removal rate of As(V) reached more than 99%, and the removal was basically complete. Continue to increase the dosage of Fe 3 O 4 @ZIF-8 to As(V) the removal rate did not improve much. Therefore, from the perspective of economy and resource saving, the optimal dosage of Fe 3 O 4 @ZIF-8 in this experiment is 15 mg/25 mL, which is 0.6 g/L.

Effect of pH
Prepared 25 mg/L As(V) solution, adjusted the pH from 2 to 11 by NaOH solution and HNO 3 solution, pipetted 25 mL into several centrifuge tubes, and dosage of Fe 3 O 4 @ZIF-8 adsorbent was 10 mg. Shook on a shaker for 24 h at 25 • C. The adsorption capacity and removal rate of As(V) by Fe 3 O 4 @ZIF-8 are shown in Figure 2. It can be seen in Figure 1 that when the dosage of Fe3O4@ZIF-8 was 10 mg, the removal rate of As(V) reached 76.29%. When the dosage of Fe3O4@ZIF-8 was increased to 15 mg, the removal rate of As(V) reached 97.20%. When the dosage of Fe3O4@ZIF-8 was 20-35 mg, the removal rate of As(V) reached more than 99%, and the removal was basically complete. Continue to increase the dosage of Fe3O4@ZIF-8 to As(V) the removal rate did not improve much. Therefore, from the perspective of economy and resource saving, the optimal dosage of Fe3O4@ZIF-8 in this experiment is 15 mg/25 mL, which is 0.6 g/L.

Effect of pH
Prepared 25 mg/L As(V) solution, adjusted the pH from 2 to 11 by NaOH solution and HNO3 solution, pipetted 25 mL into several centrifuge tubes, and dosage of Fe3O4@ZIF-8 adsorbent was 10 mg. Shook on a shaker for 24 h at 25 °C. The adsorption capacity and removal rate of As(V) by Fe3O4@ZIF-8 are shown in Figure 2. From Figure 2a, it can be found that the adsorption amount of Fe3O4@ZIF-8 reached the maximum when pH = 3.0, and the removal rate of As(V) was the largest; when pH < 3.0, the adsorption of Fe3O4@ZIF-8 decreased rapidly. The amount of As(V) adsorption by Fe3O4@ZIF-8 decreased gradually with the increase of pH when pH > 3.0. When pH > 9.0, From Figure 2a, it can be found that the adsorption amount of Fe 3 O 4 @ZIF-8 reached the maximum when pH = 3.0, and the removal rate of As(V) was the largest; when pH < 3.0, the adsorption of Fe 3 O 4 @ZIF-8 decreased rapidly. The amount of As(V) adsorption by Fe 3 O 4 @ZIF-8 decreased gradually with the increase of pH when pH > 3.0. When pH > 9.0, the adsorption capacity and removal rate of As(V) decreased rapidly. This phenomenon can be explained by Figure 2b. The experimentally measured isoelectric point of Fe 3 O 4 @ZIF-8 was 9.46. Therefore, when pH < 9.46, the surface of Fe 3 O 4 @ZIF-8 was positively charged. According to the literature, when pH < 2, As(V) mainly exists in the form of H 3 AsO 4 in water. When pH is 2-7, As(V) mainly exists in the form of H 2 AsO 4 − . When pH > 7, As(V) It mainly exists in the form of H 2 AsO 4 − [23,24]. Combined with the experimental data, it can be seen that when pH = 2, the content of H 3 AsO 4 was relatively high, which was not conducive to the adsorption of Fe 3 O 4 @ZIF-8. When pH = 3-9, the positive charge of Fe 3 O 4 @ZIF-8 decreased with the increase of pH and adsorption capacity for H 2 AsO 4 − and HAsO 4 2− decreased. When pH > 9.46, the positive charge on the surface of Fe 3 O 4 @ZIF-8 changed to negative charge, which repelled each other with HAsO 4 , so the adsorption amount and removal rate of As(V) reduced rapidly.  Figure 3. At different temperatures, the adsorption isotherms of Fe 3 O 4 @ZIF-8 for As(V) were fitted by Langmuir and Freundlich models as shown in Figure 3. From the data in Table 1, it can be seen that the Langmuir adsorption isotherm equation was more suitable for describing the adsorption process of As(V) by Fe 3 O 4 @ZIF-8, and the calculated correlation coefficient was higher than that of the Freundlich adsorption isotherm equation.
for H2AsO4 − and HAsO4 2− decreased. When pH > 9.46, the positive charge on the surface of Fe3O4@ZIF-8 changed to negative charge, which repelled each other with HAsO4, so the adsorption amount and removal rate of As(V) reduced rapidly.

Adsorption Isotherm
Prepared 25 mL of As(V) solution with an initial concentration of 40-120 mg/L and pH = 3 and added 10 mg of Fe3O4@ZIF-8 adsorbent. Shook for 24 h on a shaker at 25 °C, 35 °C and 45 °C. The obtained adsorption isotherms are shown in Figure 3. At different temperatures, the adsorption isotherms of Fe3O4@ZIF-8 for As(V) were fitted by Langmuir and Freundlich models as shown in Figure 3. From the data in Table 1, it can be seen that the Langmuir adsorption isotherm equation was more suitable for describing the adsorption process of As(V) by Fe3O4@ZIF-8, and the calculated correlation coefficient was higher than that of the Freundlich adsorption isotherm equation.  This indicated that the adsorption of As(V) by Fe3O4@ZIF-8 was monolayer adsorption. Moreover, the RL value was between 0 and 1, which meant that the adsorption  This indicated that the adsorption of As(V) by Fe 3 O 4 @ZIF-8 was monolayer adsorption. Moreover, the R L value was between 0 and 1, which meant that the adsorption was easy to proceed. When the reaction temperature was 318 K, the maximum adsorption capacity of Fe 3 O 4 @ZIF-8 simulated by the Langmuir adsorption isotherm model was 125.628 mg/g. It can be found that Fe 3 O 4 @ZIF-8 has a good adsorption effect on As(V).

Kinetic Study
A 250 mL of As(V) solution with a concentration of 30 and 50 mg/L and a pH of 3 was prepared. Then, 0.1 g of Fe 3 O 4 @ZIF-8 adsorbent was added and shaken at 25 • C in a constant temperature shaker. The fitted kinetic equation of As(V) adsorption by Fe 3 O 4 @ZIF-8 is shown in Figure 4.
125.628 mg/g. It can be found that Fe3O4@ZIF-8 has a good adsorption effect on As(V).

Kinetic Study
A 250 mL of As(V) solution with a concentration of 30 and 50 mg/L and a pH of 3 was prepared. Then, 0.1 g of Fe3O4@ZIF-8 adsorbent was added and shaken at 25 °C in a constant temperature shaker. The fitted kinetic equation of As(V) adsorption by Fe3O4@ZIF-8 is shown in Figure 4. It can be seen in Figure 4 that the fitting effect of the second-order kinetic equation for the adsorption of As(V) to Fe3O4@ZIF-8 was obviously better than that of the first-order kinetic model. From the data in Table 2, the correlation coefficient R 2 of the second-order kinetic model was much larger than that of the first-order kinetic model, and both were higher than 0.999. Under different concentrations of As(V) solution, the equilibrium adsorption capacity of Fe3O4@ZIF-8 calculated by the second-order kinetic model was basically consistent with actual adsorption capacity. Therefore, the adsorption process of As(V) by Fe3O4@ZIF-8 satisfies the second-order kinetic model. This means that the adsorption of As(V) by Fe3O4@ZIF-8 is mainly a chemical adsorption process.

Thermodynamic Studies
The thermodynamic parameters of Fe3O4@ZIF-8 adsorption of As(V) at temperatures of 298 K, 308 K, and 318 K were calculated by formula, as shown in Table 3. Table 3. Parameters of thermodynamic for the adsorption of As(V) by Fe3O4@ZIF-8.

Temperature (K)
ΔG 0 (kJ/moL)  It can be seen in Figure 4 that the fitting effect of the second-order kinetic equation for the adsorption of As(V) to Fe 3 O 4 @ZIF-8 was obviously better than that of the first-order kinetic model. From the data in Table 2, the correlation coefficient R 2 of the second-order kinetic model was much larger than that of the first-order kinetic model, and both were higher than 0.999. Under different concentrations of As(V) solution, the equilibrium adsorption capacity of Fe 3 O 4 @ZIF-8 calculated by the second-order kinetic model was basically consistent with actual adsorption capacity. Therefore, the adsorption process of As(V) by Fe 3 O 4 @ZIF-8 satisfies the second-order kinetic model. This means that the adsorption of As(V) by Fe 3 O 4 @ZIF-8 is mainly a chemical adsorption process.

Thermodynamic Studies
The thermodynamic parameters of Fe 3 O 4 @ZIF-8 adsorption of As(V) at temperatures of 298 K, 308 K, and 318 K were calculated by formula, as shown in Table 3. It can be seen in Table 3 that the standard Gibbs free energy ∆G0 of Fe 3 O 4 @ZIF-8 adsorbing As(V) is less than 0 at different temperatures, which indicates that Fe 3 O 4 @ZIF-8 adsorbing As(V) is a spontaneous reaction, and with the increase of temperature, the absolute value of ∆G 0 increases, which indicates that the increase of temperature plays a role in promoting the adsorption reaction. In addition, the standard enthalpy changes ∆H 0 during the adsorption of As(V) by Fe 3 O 4 @ZIF-8 is greater than 0, indicating that the reaction is an endothermic reaction. The standard entropy change of Fe 3 O 4 @ZIF-8 adsorption of As(V) at different temperatures is ∆S 0 > 0, which indicates that the system disorder becomes larger after the adsorption reaction, which is an entropy-driven process. The above data can prove that the adsorption process of As(V) by Fe 3 O 4 @ZIF-8 is a spontaneous endothermic reaction.

Scanning Electron Microscopy (SEM)
From a-c in Figure 5, it can be seen that the prepared Fe 3 O 4 @ZIF-8 has irregular columnar, spherical and cubic structure particles with rough surfaces. It can be seen from the figure that the particles were adhered to each other and stacked to form many channels, and this agglomeration phenomenon provided more adsorption sites for Fe 3 O 4 @ZIF-8 to adsorb arsenic. From the SEM image, it can be known that the particle size of the prepared Fe 3 O 4 @ZIF-8 was about 100 nm. It can be seen from pictures d-f in Figure 5 that the prepared Fe 3 O 4 was a nanomaterial, which was approximately spherical particles, and the particles adhere to each other and agglomerate together. The particle size of the prepared magnetic nano Fe 3 O 4 was much smaller than that of the Fe 3 O 4 @ZIF-8 material.
adsorbing As(V) is less than 0 at different temperatures, which indicates that Fe3O4@ZIF-8 adsorbing As(V) is a spontaneous reaction, and with the increase of temperature, the absolute value of ΔG 0 increases, which indicates that the increase of temperature plays a role in promoting the adsorption reaction. In addition, the standard enthalpy changes ΔH 0 during the adsorption of As(V) by Fe3O4@ZIF-8 is greater than 0, indicating that the reaction is an endothermic reaction. The standard entropy change of Fe3O4@ZIF-8 adsorption of As(V) at different temperatures is ΔS 0 > 0, which indicates that the system disorder becomes larger after the adsorption reaction, which is an entropy-driven process. The above data can prove that the adsorption process of As(V) by Fe3O4@ZIF-8 is a spontaneous endothermic reaction.

Scanning Electron Microscopy (SEM)
From a-c in Figure 5, it can be seen that the prepared Fe3O4@ZIF-8 has irregular columnar, spherical and cubic structure particles with rough surfaces. It can be seen from the figure that the particles were adhered to each other and stacked to form many channels, and this agglomeration phenomenon provided more adsorption sites for Fe3O4@ZIF-8 to adsorb arsenic. From the SEM image, it can be known that the particle size of the prepared Fe3O4@ZIF-8 was about 100 nm. It can be seen from pictures d-f in Figure  5 that the prepared Fe3O4 was a nanomaterial, which was approximately spherical particles, and the particles adhere to each other and agglomerate together. The particle size of the prepared magnetic nano Fe3O4 was much smaller than that of the Fe3O4@ZIF-8 material. From Figure 6, it can be seen that the Fe3O4@ZIF-8 particle gap after adsorption of arsenic is reduced, which indicates that the arsenic in the water is adsorbed to the surface of Fe3O4@ZIF-8, thus filling the pores. From Figure 6, it can be seen that the Fe 3 O 4 @ZIF-8 particle gap after adsorption of arsenic is reduced, which indicates that the arsenic in the water is adsorbed to the surface of Fe 3 O 4 @ZIF-8, thus filling the pores.
It can be seen from Figure 6 that the Fe 3 O 4 @ZIF-8 particle agglomeration phenomenon after adsorption is more obvious, which reflects the strong adsorption capacity of the material. The particle shape of Fe 3 O 4 @ZIF-8 after adsorption was changed and the particle size became smaller than that before adsorption.  It can be seen from Figure 6 that the Fe3O4@ZIF-8 particle agglomeration phenomenon after adsorption is more obvious, which reflects the strong adsorption capacity of the material. The particle shape of Fe3O4@ZIF-8 after adsorption was changed and the particle size became smaller than that before adsorption.

EDS Analysis
The EDS analysis of Fe3O4@ZIF-8 before and after arsenic adsorption are shown in Figure 7 and Table 4. Through EDS analysis of Fe3O4@ZIF-8, we can find that As element was not detected in Fe3O4@ZIF-8 before adsorption, while As element was detected in Fe3O4@ZIF-8 after adsorption. This proved that Fe3O4@ZIF-8 has adsorption capacity for arsenic. The Fe3O4@ZIF-8 before adsorption detected C, N, O, Fe and Zn elements, which were consistent with the elemental composition of the Fe3O4@ZIF-8 material, which proved that the metal-organic framework composite was successfully synthesized [25].

EDS Analysis
The EDS analysis of Fe 3 O 4 @ZIF-8 before and after arsenic adsorption are shown in Figure 7 and Table 4. Through EDS analysis of Fe 3 O 4 @ZIF-8, we can find that As element was not detected in Fe 3 O 4 @ZIF-8 before adsorption, while As element was detected in Fe 3 O 4 @ZIF-8 after adsorption. This proved that Fe 3 O 4 @ZIF-8 has adsorption capacity for arsenic. The Fe 3 O 4 @ZIF-8 before adsorption detected C, N, O, Fe and Zn elements, which were consistent with the elemental composition of the Fe 3 O 4 @ZIF-8 material, which proved that the metal-organic framework composite was successfully synthesized [25].

Thermogravimetric Analysis (TG)
In order to evaluate the stability of Fe3O4@ZIF-8 material, Fe3O4@ZIF-8 was heated from 30 °C to 800 °C under N2, and the stage of material weight loss was analyzed. As shown in Figure 9, it can be observed that Fe3O4@ZIF-8 was in a state of gradual weight loss at 30-264.5 °C , with a weight loss of 8.5%. This can be attributed to desorption of adsorbed water or solvent. At 264.5-800 °C, the weight loss rate of Fe3O4@ZIF-8 increased and remained in a weightless state, which may be due to the decomposition of the ligand ZIF-8. Comparing the weight loss changes of Fe3O4@ZIF-8 before and after adsorption, Fe3O4@ZIF-8 after adsorption of arsenic was heated from 30 °C to 516.5 °C, and the weight loss was 15.2%. It was obvious from Figure 9 that Fe3O4 after the adsorption of arsenic, the weight loss rate of @ZIF-8 was significantly slowed down. At 800 °C, the weight loss of Fe3O4@ZIF-8 before adsorption reached 66.4%, and the weight loss of Fe3O4@ZIF-8 after adsorption was 36.8%. This shows that Fe3O4@ZIF-8 has good thermal stability after adsorbing arsenic, and arsenic is adsorbed to the surface of Fe3O4@ZIF-8, and it is not easy to desorb and decompose. Comparing the prepared magnetic nano Fe 3 O 4 and Fe 3 O 4 standard cards, it is found that the characteristic peak positions of the standard cards were the same, which can prove the successful preparation of magnetic nano Fe 3 O 4 . By comparing the standard cards of Fe 3 O 4 @ZIF-8, ZIF-8, and Fe 3 O 4 , it can be found that the characteristic peaks are basically the same as those of Fe 3 O 4 and standard cards, and there were diffraction peaks consistent with the standard pattern of ZIF-8, which can prove that Fe 3 O 4 @ZIF-8 synthesis was relatively successful. Comparing the XRD patterns of Fe 3 O 4 @ZIF-8 before and after adsorption, it can be found that the position of the diffraction peak of Fe 3 O 4 @ZIF-8 does not change after adsorption of arsenic, and the relative intensity of the peak increases, which indicates that Fe 3 O 4 @ZIF-8 is adsorbing arsenic. The arsenic process did not destroy the structure and properties. Therefore, the material can exist stably in an aqueous solution.

Thermogravimetric Analysis (TG)
In order to evaluate the stability of Fe 3 O 4 @ZIF-8 material, Fe 3 O 4 @ZIF-8 was heated from 30 • C to 800 • C under N 2 , and the stage of material weight loss was analyzed. As shown in Figure 9, it can be observed that Fe 3 O 4 @ZIF-8 was in a state of gradual weight loss at 30-264.5 • C, with a weight loss of 8.5%. This can be attributed to desorption of adsorbed water or solvent. At 264.5-800 • C, the weight loss rate of Fe 3 O 4 @ZIF-8 increased and remained in a weightless state, which may be due to the decomposition of the ligand ZIF-8. Comparing the weight loss changes of Fe 3 O 4 @ZIF-8 before and after adsorption, Fe 3 O 4 @ZIF-8 after adsorption of arsenic was heated from 30 • C to 516.5 • C, and the weight loss was 15.2%. It was obvious from Figure 9 that Fe 3 O 4 after the adsorption of arsenic, the weight loss rate of @ZIF-8 was significantly slowed down. At 800 • C, the weight loss of Fe 3 O 4 @ZIF-8 before adsorption reached 66.4%, and the weight loss of Fe 3 O 4 @ZIF-8 after adsorption was 36.8%. This shows that Fe 3 O 4 @ZIF-8 has good thermal stability after adsorbing arsenic, and arsenic is adsorbed to the surface of Fe 3 O 4 @ZIF-8, and it is not easy to desorb and decompose.
In order to evaluate the stability of Fe3O4@ZIF-8 material, Fe3O4@ZIF-8 was heated from 30 °C to 800 °C under N2, and the stage of material weight loss was analyzed. As shown in Figure 9, it can be observed that Fe3O4@ZIF-8 was in a state of gradual weight loss at 30-264.5 °C , with a weight loss of 8.5%. This can be attributed to desorption of adsorbed water or solvent. At 264.5-800 °C, the weight loss rate of Fe3O4@ZIF-8 increased and remained in a weightless state, which may be due to the decomposition of the ligand ZIF-8. Comparing the weight loss changes of Fe3O4@ZIF-8 before and after adsorption, Fe3O4@ZIF-8 after adsorption of arsenic was heated from 30 °C to 516.5 °C, and the weight loss was 15.2%. It was obvious from Figure 9 that Fe3O4 after the adsorption of arsenic, the weight loss rate of @ZIF-8 was significantly slowed down. At 800 °C, the weight loss of Fe3O4@ZIF-8 before adsorption reached 66.4%, and the weight loss of Fe3O4@ZIF-8 after adsorption was 36.8%. This shows that Fe3O4@ZIF-8 has good thermal stability after adsorbing arsenic, and arsenic is adsorbed to the surface of Fe3O4@ZIF-8, and it is not easy to desorb and decompose.

FTIR Analysis
From the infrared spectrum of magnetic nano Fe3O4 in Figure 10, it can be seen that absorption bands corresponding to the carboxylate groups on the surface of Fe3O4 NPs appeared at 1384.25 cm −1 and 1628.70 cm −1 [27]. At 590.88 cm −1 and the peak at 631.52 cm −1 can be attributed to the Fe-O bond of Fe3O4 [28,29]. Analysis of the infrared spectrum of

FTIR Analysis
From the infrared spectrum of magnetic nano Fe 3 O 4 in Figure 10, it can be seen that absorption bands corresponding to the carboxylate groups on the surface of Fe 3 O 4 NPs appeared at 1384.25 cm −1 and 1628.70 cm −1 [27]. At 590.88 cm −1 and the peak at 631.52 cm −1 can be attributed to the Fe-O bond of Fe 3 O 4 [28,29]. Analysis of the infrared spectrum of Fe 3 O 4 @ZIF-8 shows that the peaks at 3179.34 cm −1 and 2926.14 cm −1 can correspond to the characteristic peaks of =C−O bond and C−H bond with ZIF-8 structure, respectively [30,31]. The peak of C=N bond in the imidazole ring is 1569.76 cm −1 . The absorption peak of CN bond appeared at 1143.10 cm −1 and 994.82 cm −1 , and the vibration peak of functional group in Zn-N appeared at 423.37 cm −1 . At 591.03 cm −1 and 628.44 cm −1 the peaks appearing at −1 can be assigned to Fe-O peaks [32][33][34]. The analysis of Fe 3 O 4 @ZIF-8 by FTIR can prove that the experiment successfully complexes Fe 3 O 4 with ZIF-8. Comparing the infrared spectrum of Fe 3 O 4 @ZIF-8 after arsenic adsorption, 3359.15 cm −1 corresponds to the stretching vibration region of O−H and N−H. A new strong band is observed near 427.90 cm −1 , which can be attributed to the Zn-O-As vibration, which means the formation of a new inner spherical complex [35][36][37]. A new peak appears at 828.21 cm −1 , which may be due to the formation of As-O groups [38] showing that arsenic was bound to Fe 3 O 4 @ZIF-8.

XPS Analysis
Through XPS analysis of Fe 3 O 4 @ZIF-8 before and after arsenic adsorption in Figure 11, it can be found that the full spectrum in Figure 11 of Fe 3 O 4 @ZIF-8 before adsorption can detect the peaks of Zn, Fe, O, N, C, which proves to be in line with the preset Chemical composition of Fe 3 O 4 @ZIF-8 material. Fe 3 O 4 @ZIF-8 after adsorption detected the characteristic peak of As, which proved that the material successfully combined with arsenic and had an adsorption effect.
FTIR can prove that the experiment successfully complexes Fe3O4 with ZIF-8. Comparing the infrared spectrum of Fe3O4@ZIF-8 after arsenic adsorption, 3359.15 cm −1 corresponds to the stretching vibration region of O−H and N−H. A new strong band is observed near 427.90 cm −1 , which can be attributed to the Zn-O-As vibration, which means the formation of a new inner spherical complex [35][36][37]. A new peak appears at 828.21 cm −1 , which may be due to the formation of As-O groups [38] showing that arsenic was bound to Fe3O4@ZIF-8.

XPS Analysis
Through XPS analysis of Fe3O4@ZIF-8 before and after arsenic adsorption in Figure  11, it can be found that the full spectrum in Figure 11 of Fe3O4@ZIF-8 before adsorption can detect the peaks of Zn, Fe, O, N, C, which proves to be in line with the preset Chemical composition of Fe3O4@ZIF-8 material. Fe3O4@ZIF-8 after adsorption detected the characteristic peak of As, which proved that the material successfully combined with arsenic and had an adsorption effect.

XPS Analysis
Through XPS analysis of Fe3O4@ZIF-8 before and after arsenic adsorption in Figure  11, it can be found that the full spectrum in Figure 11 of Fe3O4@ZIF-8 before adsorption can detect the peaks of Zn, Fe, O, N, C, which proves to be in line with the preset Chemical composition of Fe3O4@ZIF-8 material. Fe3O4@ZIF-8 after adsorption detected the characteristic peak of As, which proved that the material successfully combined with arsenic and had an adsorption effect. From Figure 11b, it can be observed that the positions of the peaks of Zn 2p1/2 and Zn 2p3/2, before and after adsorption, move to the position with higher binding energy, which means that the electronegativity of the surrounding atoms of Zn increases, resulting in Zn element. The binding energy also increases [39][40][41]. By fitting the peaks of As 3d after adsorption, it can be judged that after As(V) is adsorbed by Fe3O4@ZIF-8, a part of As(V) is reduced to As(III) by Fe3O4@ZIF-8.

Surface Area Analysis (BET)
The porosity and specific surface area of the prepared Fe3O4@ZIF-8 composites were tested by N2 adsorption-desorption isotherms at the test temperature of 77 K. From Figure   Figure 11. XPS spectra of Fe 3 O 4 @ZIF-8: (a) Wide scan; (b) Zn 2p core level. (c) As 3d core level after adsorption.
From Figure 11b, it can be observed that the positions of the peaks of Zn 2p1/2 and Zn 2p3/2, before and after adsorption, move to the position with higher binding energy, which means that the electronegativity of the surrounding atoms of Zn increases, resulting in Zn element. The binding energy also increases [39][40][41]. By fitting the peaks of As 3d after adsorption, it can be judged that after As(V) is adsorbed by Fe 3 O 4 @ZIF-8, a part of As(V) is reduced to As(III) by Fe 3 O 4 @ZIF-8.

Surface Area Analysis (BET)
The porosity and specific surface area of the prepared Fe 3 O 4 @ZIF-8 composites were tested by N 2 adsorption-desorption isotherms at the test temperature of 77 K. From Figure 12, it can be observed that the specific surface area of Fe 3 O 4 @ZIF-8 calculated by the BET method is 316.3593 m 2 /g, and the total pore volume measured by the single-point method is 0.224 cm 3 /g, and the micropore volume measured by the t-plot method is 0.097 cm 3 /g. The pore size is 2.83 nm, and the average mesopore size is 33.59 nm. The specific surface area of Fe 3 O 4 @ZIF-8 after adsorption of arsenic is 95.3942 m 2 /g, the measured total pore volume is 0.182 cm 3 /g, and the micropore volume is 0.017 cm 3 /g, the average pore size is 7.62 nm, and the average mesopore size is 34.35 nm. By comparison, it can be found that the specific surface area, total pore volume and micropore volume of Fe 3 O 4 @ZIF-8 after adsorption of arsenic are significantly reduced, indicating that arsenic is adsorbed to the surface of the material.

Conclusions
Characterization analysis shows that the Fe3O4@ZIF-8 materials prepared in this study have different shapes and small particle sizes, and the specific surface area can reach 316 m 2 /g. Through various characterization analyses, it can be proved that the metalorganic framework composite material was successfully prepared in the experiment; comparing the characterization results of Fe3O4@ZIF-8 before and after adsorption, it can be found that the material has a good adsorption effect on arsenic. At 25 °C , the initial concentration of arsenic is 46.916 mg/L, and the initial pH is 3. When the dosage of adsorbent is 0.4 g/L, the adsorption rate at equilibrium is 76.29%. When the dosage increases to 1.0 g/L, the adsorption rate at equilibrium increases to 99.29%. However, when the dosage of adsorbent increases to 0.6 g/L, the equilibrium adsorption rate can reach 97.20%, which is not much different from the equilibrium adsorption rate when the dosage is 1.0 g/L. Therefore, from the economic benefit to the perspective of resource saving, the optimal dosage is 0.6 g/L. When the pH value is 3, the adsorption effect of Fe3O4@ZIF-8 on arsenic can reach the best, the adsorption amount reach 59.80 mg/g, and the removal rate is 99.93%. When pH > 3, the equilibrium adsorption capacity of Fe3O4@ZIF-8 to arsenic decreases with the increase of pH, and when pH > 9.46, the decreasing rate is faster. The adsorption process of Fe3O4@ZIF-8 for arsenic has the best fit with the Langmuir adsorption isotherm equation. When the temperature is 298 K, the maximum adsorption capacity of Fe3O4@ZIF-8 for arsenic can reach 116.114 mg/g. For the adsorption kinetics, the second-order kinetic equation has the best fitting result for the adsorption kinetics of arsenic on Fe3O4@ZIF-8, with R 2 > 0.999. By analyzing the fitting curve of the second-order kinetics, it can be found that the adsorption process of Fe3O4@ZIF-8 to arsenic-containing wastewater mainly occurs through chemical adsorption. From the analysis of thermodynamic parameters, it can be concluded that the adsorption process of Fe3O4@ZIF-8 to arsenic is a spontaneous endothermic reaction. Magnetic Fe 3 O 4 nanoparticles have been used for the treatment of arsenic from wastewater. Ref. [42] reported the maximum adsorption capacity of magnetite Fe 3 O 4 nanoparticles occurred at pH 2, with a value of 3.70 mg/g for As (V). Due to aggregation effect, magnetic Fe 3 O 4 nanoparticles are difficult to use in continuous flow systems [43]. Consequently, some researchers had encapsulated nanoparticles into metal organic frameworks to resolve above problem. Iron and 1,3,5-benzenetricarboxylic (Fe-BTC) used for As(V) removal from waters, and its adsorption capacity can reach 12.3 mg/g, more than 6 times that of iron oxide nanoparticles with a size of 50 nm [19]. Studies have shown that ZIF-8 demonstrated an adsorption amount of 60.03 mg/g [44] in comparison to that of MIL-53(Fe) (21.27 mg/g) [45]. Furthermore, the activated indium MOF (AUBM-1) was applied for As(V) removal from water and showed a high arsenic uptake capacity of 103.1 mg/g at a neutral pH. In this study, the maximum adsorption capacity of Fe 3 O 4 @ZIF-8 for As(V) can reach 116.114 mg/g; the result implies that the Fe 3 O 4 @ZIF-8 could be as a potential candidate for removing As(V) pollution in wastewater.

Conclusions
Characterization analysis shows that the Fe 3 O 4 @ZIF-8 materials prepared in this study have different shapes and small particle sizes, and the specific surface area can reach 316 m 2 /g. Through various characterization analyses, it can be proved that the metal-organic framework composite material was successfully prepared in the experiment; comparing the characterization results of Fe 3 O 4 @ZIF-8 before and after adsorption, it can be found that the material has a good adsorption effect on arsenic. At 25 • C, the initial concentration of arsenic is 46.916 mg/L, and the initial pH is 3. When the dosage of adsorbent is 0.4 g/L, the adsorption rate at equilibrium is 76.29%. When the dosage increases to 1.0 g/L, the adsorption rate at equilibrium increases to 99.29%. However, when the dosage of adsorbent increases to 0.6 g/L, the equilibrium adsorption rate can reach 97.20%, which is not much different from the equilibrium adsorption rate when the dosage is 1.0 g/L. Therefore, from the economic benefit to the perspective of resource saving, the optimal dosage is 0.6 g/L. When the pH value is 3, the adsorption effect of Fe 3 O 4 @ZIF-8 on arsenic can reach the best, the adsorption amount reach 59.80 mg/g, and the removal rate is 99.93%. When pH > 3, the equilibrium adsorption capacity of Fe 3 O 4 @ZIF-8 to arsenic decreases with the increase of pH, and when pH > 9.46, the decreasing rate is faster. The adsorption process of Fe 3 O 4 @ZIF-8 for arsenic has the best fit with the Langmuir adsorption isotherm equation. When the temperature is 298 K, the maximum adsorption capacity of Fe 3 O 4 @ZIF-8 for arsenic can reach 116.114 mg/g. For the adsorption kinetics, the second-order kinetic equation has the best fitting result for the adsorption kinetics of arsenic on Fe 3 O 4 @ZIF-8, with R 2 > 0.999. By analyzing the fitting curve of the second-order kinetics, it can be found that the adsorption process of Fe 3 O 4 @ZIF-8 to arsenic-containing wastewater mainly occurs through chemical adsorption. From the analysis of thermodynamic parameters, it can be concluded that the adsorption process of Fe 3 O 4 @ZIF-8 to arsenic is a spontaneous endothermic reaction.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.