1 Introduction

Nanotechnology is the science of studying materials and phenomena at the nanoscale. Fe3O4 NPs are smaller than 100 nm in at least one dimension. Fe3O4 NPs have extraordinary properties, compared to their classic counterparts, due to their high surface area to volume ratio [1]. Fe3O4 NPs can be fabricated using chemical, physical or a combination of both techniques, such as implantation, attrition, pyrolysis, sonochemical, hydrothermal, and co-precipitation. The co-precipitation technique is a conventional, facile, and environmentally friendly process producing various metal oxide nanoparticles, such as Fe3O4 NPs. This process does not produce toxic intermediates or dissolvents, does not need precursor complexes, and is produced at temperatures less than 100 °C. The type and ratio of salts used, process temperature, and pH are some important parameters that control the size and morphology of gained Fe3O4 NPs [2, 3]. Magnetic nanoparticles have exclusive properties, such as an extensive spontaneous magnetic field that can react quickly similar to huge paramagnetic atoms when placed in a magnetic field. Fe3O4 NPs are used more than other magnetic nanoparticles because they have lower toxicity than their metallic counterparts and have extraordinary paramagnetic behavior [4].

In addition to magnetic characteristics, the Fe3O4 NPs have high absorption capacity and may be used as a super-absorbent to absorb various water clouding agents such as natural organic matter (NOM), including hydrophobichumic and fulvic acids, hydrophilic acids, proteins, lipids, hydrocarbons and carbohydrates, and suspended solids. Thus, they can be used affectively in water treatment applications. Source waters having high concentrations of NOM have accordingly been treated using metal salts. The actual reason for removal of NOM in production of potable water is to stop disinfection by-products (DBPs) from being produced in the process of disinfection. The major clotting mechanisms involve charge neutralization of colloidal NOM, precipitation as humates or fulvates, and co-precipitation by adsorption on the metal hydroxide. The typical procedure in order to remove NOM in water has been clotting using iron and aluminum salts succeeded by a solid–liquid separation process such as sedimentation or dissolved air flotation [5,6,7,8]. Today, water treatment has received considerable attention, especially from economic view points [9, 10]. The practiced full-scale treatment sequence included coagulation, flocculation, clearing turbidity by flotation, disinfection with chlorine dioxide, activated carbon filtration, and post-chlorination. Among the water-treatment processes, coagulation is an important stage that can destabilize small suspended particles of water and convert them into a settable large mass that can be easily removed from the system by filtration or decantation. Generally, the absorbents are good coagulants, especially for removing or decreasing the turbidity of systems. The turbidants have almost negative charges that can be neutralized by absorbents and coagulated and precipitated effectively [11, 12]. Aluminum or iron salts are the most used and conventional absorbents for turbidity removal in water treatment processes. However, the large amount of sludge they produce causes environmental concerns [13]. The best way to improve the removal of NOM is to optimize coagulation conditions concerning pH and coagulant dose. The pH of coagulation is said to be the deciding factor for the removal of NOM. In most studies, the optimum values were between 5.2 and 5.8 for aluminum sulfate and from 4.5 to 5.0 for ferric sulfate. Moreover, optimizing the coagulant dose is important because coagulant over-dosing results in an extra amount of sedimentation. Thus, because the metal coagulant addition leads to a pH reduction, the addition of alkaline chemicals is necessary to maintain the coagulation pH within an effective range. On the other hand, under-dosing normally causes excess residual metal in treated water and can lead to an increase in processed water cloudiness. Moreover, the removal can be improved by a choice of a proper coagulant. The abundantly used coagulant in drinking water production is aluminum sulfate. Ferric salts have become more popular coagulants for many reasons, including health worries regarding the unconfirmed link between aluminum and Alzheimer disease. Ferric salts have also been found to remove NOM more actively at low temperature, especially in cold water (< 3 °C), and low turbidity waters [8]. The development of new coagulants is the probable solution to these problems. Recently, the relevant research in the coagulation flocculation field has converged on new materials as coagulants, such as ferrate salt, polymeric aluminum sulfate, cactus as a natural coagulant, ferromagnetic nanoparticle composited PACls, zirconium coagulant, modified clays, chitosan biopolymer, magnetic seeds together with polyaluminum chloride, and tannin-based coagulant [14,15,16,17,18,19,20,21]. Thus, the nano-absorbents, with lower cost, higher efficiency, and more enviromental compatibility, can be suggested as efficient alternatives for the mentioned salts [11, 22]. Due to the high surface area of nano-absorbents, they have a greater absorption rate and absorption efficiency. The van der Waals and ionic interactions are commonly the prevailing sorption mechanism of nano-absorbents, especially Fe3O4 NPs [23, 24].

Therefore, this research was aimed to synthesize modified Fe3O4 NPs at high efficiency through a co-precipitation technique in order to absorb ionic contaminants, organic matters, and soil in the coagulation flocculation stage of water treatment. After the coagulation process, the synthesized Fe3O4 NPs can be simply collected by a magnetic process and reused and minimum water sludge remains in this process.

2 Materials and Methods

2.1 Materials

Ferrous chloride (FeCl2.4H2O, 198.83 g/mol) and ferric chloride (FeCl3.6H2O, 270.30 g/mol) were purchased from Darmstadt, Germany. Sodium hydroxide was obtained from Merck, Germany.

2.2 Synthesis of Fe3O4 NPs

The Fe3O4 NPs were synthesized from FeCl2.4H2O and FeCl3.6H2O using the co-precipitation method. To 100 mL of deionized water, 0.1 M FeCl2.4H2O and 0.2 M FeCl3.6H2O were added and retained using a mechanical laboratory stirrer (Heidolph, PZR 2020, USA) until a homogeneous solution was formed. The solution was covered and placed in a water bath at 60 °C and warmed for 15–20 min. Then 14 mL of 25% sodium hydroxide was added to the solution and the formed black precipitate was centrifuged (BADR, Model 8153, USA) for 15 min at 5000 rpm. The Fe3O4 NPs were completely separated after centrifugation from the liquid phase and deposited. The supernatant was discarded and washed several times with deionized water, and the precipitate was dried at 60 °C to get the powder form of iron oxide Fe3O4 NPs [25]. The Fe3O4 NPs were obtained with a dark brown to black color.

2.3 Characterization

The Fe3O4 NPs characteristics were evaluated as follows: magnetic properties were measured using a vibrating sample magnetometer (VSM, MDKF, Iran), surface area and pore size using Brunauer–Emmett–Teller (BET, BelSorp mini II, Japan), crystal phase using X-ray diffraction (XRD, Siemens D500 X-ray, Germany), and morphology using field emission scanning electron microscopy (FESEM, MIRA3 FEG-SEM, Tescan, Czech).

2.4 Coagulation Experiments

To determine the effectiveness of synthesized Fe3O4 NPs in removal of water turbidity, they were applied to water samples that originated from the Zarrineh-roud River (East Azerbaijan, Islamic republic of Iran). For this purpose, the upper limit and lower limit for the three-year average were determined. The turbidity of the Zarineh-roud River for three years is shown in Table 1. The highest turbidity limit was 65 NTU, and the lowest turbidity limit was 17 NTU. Source water was collected at 24–26 °C, pH was 8–8.2, EC was 737–1358 micro-Siemens per centimeter (μs/cm), and turbidity was 17–394.5 NTU.

Table 1 Water turbidity of the Zarineh-roud River for three years
Full size table

Various dosages of Fe3O4 NPs (0.015–0.25 g) were added to 200 mL of surface water samples. Then the water sample was mixed using a mechanical stirrer at 200 rpm for 2 min and stirring was continued at 40 rpm for 10 min. Then the system was placed on the super magnet (1.2 T, 10 × 2 cm) for 5 min, in order to collect Fe3O4 NPs, and 20 mL of the suspension supernatant was transferred to another container in order to measure the turbidity [6]. The turbidity of each sample was measured by a portable turbidity meter (HACH Company, 2100P, USA; Range 0–1000 NTU). The experiment was repeated for pH = 6, 7, and 8. The pH of the suspension was adjusted by adding 0.1 M NaOH or 0.1 M HCl using EC and a pH-meter (WTW, 750 pH/ION/Cond, Germany). The temperature of all the experiments was set at 25 ± 1 °C.

2.5 Statistical Analysis

Analysis of variance (ANOVA) comparison of data and optimization was performed using the Minitab statistical package (Minitab v.17 Inc. PA, USA).

3 Results and Discussion

3.1 Characterizing Synthesized Fe3O4 NPs

3.1.1 X-Ray Diffraction (XRD)

X-ray diffraction pattern was gained using Cuka, the X-ray diffraction intensity was recorded as a function of Bragg’s 2θ in the angular range of 5–70°. IR spectra were recorded between 500 and 4000 cm−1. Figure 1 represents the Fe3O4 magnetic nanoparticles XRD pattern, the XRD peak 2θ values such as 18.3, 35.5, 43.1, 57.0, and 62.6 correspond to (111), (311), (400), (511), and (440) hkl values, respectively, and the JCPDS-ICDD reference pdf No is 11-614. Comparing the peaks, it is clear that the peaks of the synthesized Fe3O4 NPs correspond to the standard sample peaks and the presence of Fe3O4 NPs is established. The sharpness of the peaks indicate that the synthesized nanoparticles are crystallite and spherical. The average particle sizes of Fe3O4 NPs can be judged using the Debye–Scherrer equation, which gives a relationship between peak broadening in XRD and particle size that is shown by the following equation: d = k λ/ (β·cos θ), where d is the particle size of the crystal, k is Sherrer constant (0.9), λ is the X-ray wavelength (0.15406 nm), β is the width of the XRD peak at half-height, and θ is the Bragg diffraction angle. Using the Debye–Scherrer equation, the average crystallite sizes of the magnetic Fe3O4 NPs are discovered to be in the range of 16–19 nm. The XRD results show smaller measurement compared with the FESEM results. This can be due to the fact that XRD indicates the crystallite size (crystalline domains) according to the Debye–Scherrer formula, which highlights the limit between crystallographic planes of the particles, and FESEM indicates the particle size as an agglomeration of many crystallites. In this way, our results show a good correlation between the SEM and XRD investigations concerning the particle sizes. Liu et al. [26] have also synthesized Fe3O4 NPs through the co-precipitation technique. According to their XRD result, 2θ tops 30.1°, 35.5°, 43.1°, 53.4°, 57.0°, and 62.6°, which are distributed to the (220), (311), (400), (422), (511), and (440) hkl value estimations of magnetite stages.

Fig. 1
figure 1

The XRD pattern of the synthesized Fe3O4 NPs

Full size image

3.1.2 Field Emission Scanning Electron Microscope (FESEM)

Figure 2 shows the FESEM image of synthesized Fe3O4 NPs as adsorbent. As depicted in Fig. 2a, the Fe3O4 NPs crystals have a smooth surface. Figure 2b proves the almost spherical morphology of nanoparticles with diameter ranged from 22 to 32 nm. The FESEM image also confirms that the Fe3O4 NPs formed are spherical. It is considerable that in practical application nanoparticles with diameters less than 10–20 nm will have a reduced effect, due to their lower physical stability and higher susceptibility to agglomeration. Moreover, very small metallic nanoparticles have lower chemical stability and can be rapidly oxidized in the presence of oxygen. Previous studies also showed that the approximately 30 nm magnetic Fe3O4 NPs have more desired characteristics, such as more electron conductivity and enhanced absorption capability, especially for enzyme immobilization uses, compared to smaller ones [27].

Fig. 2
figure 2

a & b The FESEM images of the synthesized Fe3O4 NPs

Full size image

3.1.3 Brunauer–Emmett–Teller (BET) Surface Area Analysis

The pore size, as well as volume and regional distribution can be achieved using BET analysis of mesoporous or macroporous materials. In general, a smaller pore diameter leads to the synthesized particle specific surface area [28]. Figure 3 shows the BET surface area and total pore volume of synthesized Fe3O4 NPs. It can be concluded that surface area and total pore volume of gained Fe3O4 NPs are 99. 6 m2 g−1 and 22.88 cm3 g−1, respectively, which were obtained using nitrogen as an adsorptive agent, at an adsorption temperature of 77 K and activation time of 180 min. It should be noted that when Po/P is in the range of 0.3–0.05 multilayer adsorption does not occur in most cases and the adsorption data show good agreement with the BET eq. [29]. Some isotherms have a hysteresis curve, which indicates the presence of mesopores in the material, and can provide information on the geometry of the pores [30]. As can be seen in Fig. 4, the adsorption/desorption isotherm is type IV, and thus the porosity of the synthesized samples is thin and capillary. Consequently, the adsorption rate for Fe3O4 NPs is significantly higher with concentrated adsorbed material on the absorbent surfaces. The hysteresis isotherm showed that the diagram of the porosity is cylindrical and the relationship between the sample surface and the adsorbent is relatively strong. The mean pore diameter is 10.6 nm and according to the IUPAC classification, the porosity is categorized in the mesopores group. The Barrett, Joyner, and Halenda (BJH) method is used to calculate the diameter of the pores and the regional distribution of the pores. In general, the smaller the pore diameter, the greater the specific surface area of the particles [28]. Figure 5 indicated that the distribution of porosity was homogenous and the diameter peak was 10.7 nm.

Fig. 3
figure 3

The BET plot of the synthesized Fe3O4 NPs

Full size image
Fig. 4
figure 4

Adsorption/desorption isotherm of the synthesized Fe3O4 NPs

Full size image
Fig. 5
figure 5

a & b BJH plot of the synthesized Fe3O4 NPs

Full size image

3.1.4 Vibrating Sample Magnetometer (VSM)

In most materials, the spin moments are small and aligned randomly causing para magnetism. In some materials, specifically transition metals such as iron, the spin moments are large and aligned in parallel or ferro magnetically. This causes a net spontaneous magnetic moment in the material [31, 32]. Figure 6 for Fe3O4 NPs shows a phenomenal paramagnetic behavior without any obstruction or inclination. The range of the applied field with an intensity of 10,000 Oe, for the maximum magnetic saturation is 66.21 emu/g. A hysteresis loop with magnetization of this value demonstrates that it has a super paramagnetic nature [33].

Fig. 6
figure 6

The VSM of the synthesized Fe3O4 NPs

Full size image

3.2 Evaluation of the Coagulation Ability of Fe3O4 NPs

The effect of nanoparticle concentration and initial turbidity and pH was analyzed using the generalized linear model (GLM). The analysis of variance in evaluation of the effects of Fe3O4 NPs, initial turbidity, and pH on their turbidity reduction ability are depicted in Table 2. As can be seen, the concentration of nanoparticles and turbidity has significant (p < 0.05) effects. Although according to a previous study, with an extensive range, pH is the main factor affecting adsorption processes [6], pH was not shown to influence the adsorption of nanoparticles significantly (as indicated by the obtained p-value). This could be due to the smaller pH range (pH 6–8) examined in this study. Thus, under the examined pH ranges, the interaction of nanoparticles and H+ ions could not alter the nanoparticles’ electrostatic forces, and consequently, their adsorption capacity.

Table 2 Results of ANOVA (F-value and p-value)
Full size table

As shown in Fig. 7, by increasing the quantity of nanoparticles, the amount of turbidity decreases, and by reducing turbidity, the efficiency of absorption was improved. This result is very proficient for purifying low turbidity water. In low turbidity water, the number of impure particles is small and the distribution is relatively uniform. Colloidal particles have small negative potential and strong inter-particle rejection. Accordingly, colloids are difficult to remove and the aggregation effect is poor. Thus, the colloids are difficult to coagulate when an electrolyte or salt is used as coagulant [34]. The turbidity removal ability of Fe3O4 NPs is shown in Fig. 8, the turbidity removal efficiency of Fe3O4 NPs can be increased effectively by increasing the nanoparticles’ dosage. For instance, at a 0.015 g dosage of Fe3O4 NPs, the turbidity was reduced by approximately half and by increasing the Fe3O4 NPs dosage to 0.035 g, turbidity reached below 10 NTU. For instance, using 0.15 g of Fe3O4 NPs approximately 96.15% and 88.82% turbidity can be removed for primary water turbidity of 65 and 17 NTU, respectively. The single optimization results (Fig. 8) indicate that the highest turbidity reduction would occur using 0.2 g of Fe3O4 NPs, primary turbidity of 17, and pH of 6. The results of this study are almost compatible with the results of Hatamie et al. [6]. They could remove 95.8% and 86% turbidities of samples with an initial water turbidity of 60 and 24 NTU, respectively, using 0.15 g of 30–50 nm Fe3O4 NPs. In other research, 94% of turbidities were removed from the water with an initial turbidity of 80 NTU, using 0.32 g of graphene oxide based nano-absorbents with 10 min contact time [35]. Thus, the Fe3O4 NPs are more efficient than graphene oxide based nano-absorbents with similar contact time. Fe3O4 NPs can also remove water turbidity in smaller dosages. Fe3O4 NPs showed a turbidity removal tendency at neutral pHs, and consequently, they can treat greater amounts of water without any pH regulations. Thus, Fe3O4 NPs are appropriate for elimination of several types of ionic and organic contaminants and turbidants from water in a very short period of time [36, 23].

Fig. 7
figure 7

The effect of nanoparticle concentration and initial turbidity and pH

Full size image
Fig. 8
figure 8

The effect of the Fe3O4 NPs dosage on turbidity removal of 200 mL water with an initial turbidity of (a) 65 NTU and (b) 17 NTU

Full size image

4 Conclusions

Fe3O4 NPs with high stability were successfully synthesized using a simple co-precipitation method in 23–32 nm mean sizes, in order to absorb and remove the water turbidants and clarify water resources perfectly. The characteristics of the obtained nanoparticles were then evaluated in terms of VSM, BET, XRD, and FESEM to measure magnetic properties, surface area, crystal phase, and morphology, respectively. The synthesized Fe3O4 NPs were without any impurity in the sizes of 22–32 nm, surface area of 99.6 m2/g, pores’ mean diameter size of approximately 10.6 nm, and maximum magnetic saturation of 66.21 emu/g. The effects of the synthetic Fe3O4 NPs, initial NTU, and pH were then evaluated on turbidity reduction of gained nanoparticles. Using 0.2 g of synthetic Fe3O4 NPs with an initial turbidity of 17 NTU and pH of 6 could reduce all water turbidity (100%). Thus, it seems that using Fe3O4 NPs as an absorbent in the water treatment process can be an efficient and economical alternative due to their cheap raw materials, facile synthesizing and recovering processes, as well as the ability to reuse them several times.