Nitrate removal from aqueous solutions by cobalt ferrite nanoparticles synthesized by co-precipitation method: isotherm, kinetic and thermodynamic studies

There are usually several nitrogen compounds, such as ammonia, nitrite and nitrate, in drinking water (Chabani et al. 2006). Here, nitrate (NO₃⁻) is a polluting agent of growing concern affecting water quality globally (Oyarzun et al. 2018). Pollution of water resources by nitrate has become a significant problem in recent decades. The widespread use of organic and mineral fertilizers in agriculture, the entry of animal wastes and leakage from sewage systems are among the most significant sources of water pollution with nitrate. The nitrate ion is highly soluble in water and can easily enter groundwater and surface water (Divband Hafshejani et al. 2016; Kalaruban et al. 2016).

The presence of high concentrations of nitrate in drinking water can cause methemoglobinemia (blue-baby syndrome) in babies, carcinogenic nitrosamine and stomach cancer. Moreover, nitrate causes the spread of infectious diseases such as cancer of the alimentary canal and cyanosis among children (Rajeswari et al. 2016). Additionally, the high concentrations of nitrate in aquatic systems can be poisonous and cause various problems, like eutrophication in the environment (Keränen et al. 2015; Ghasemi & Sillanpää 2015). This affects fish and other aquatic organisms as well as recreational use of water (lakes and rivers) (Kalaruban et al. 2016).

Using nitrate-contaminated drinking water is hazardous in terms of health and many countries have set strict regulations for controlling this contaminated water source (Alimohammadi et al. 2018; Radfarda et al. 2019). The World Health Organization (WHO) has classified nitrate in a special group of hard-to-purify pollutants (Masoumbeigi et al. 2012). WHO has determined the allowed nitrate in drinking water as 50 mg/L as NO₃⁻, and 15 mg/L as NO₃⁻ for children (Govindan et al. 2015). Moreover, the US Environmental Protection Agency has announced the maximum nitrogen concentration as 10 mg/L as N (44 mg/L as NO₃⁻) (Divband Hafshejani et al. 2016). Hence, the waters with nitrate more than these values should be treated appropriately to reduce the concentration of nitrate and nitrite to the standard level.

Various methods have been used to remove nitrate from aqueous solutions, such as biological treatment (Park & Yoo 2009), electro-dialysis (Elmidaoui et al. 2001), reverse osmosis (Schoeman & Steyn 2003), ion exchange (Samatya et al. 2006), and adsorption (Zhao et al. 2018). Technologies such as reverse osmosis and catalytic and biological processes are relatively expensive. Adsorption is one of the best methods for nitrate removal because of its simplicity, sludge-free operation, ease of use and availability of various adsorbents (Song et al. 2016).

Different adsorbents such as activated carbon (Demiral & Gündüzoğlu 2010), bamboo powder charcoal (Mizuta et al. 2004), nano-alumina (Bhatnagar et al. 2010), iron nanoparticles (Kassaee et al. 2011), SiO₂-FeOOH-Fe nanocomposites (Ensie & Samad 2014), and magnetic nanocomposite Fe₃O₄/ZrO₂/chitosan (Jiang et al. 2013) have been used for the removal of nitrate.

Among various adsorbents, nanomaterials have received great attention in pollutant adsorption and environmental degradation given their high surface area and adsorption capacity (Zhang et al. 2010; Dehghani et al. 2020; Khan et al. 2020). One of the problems with nanomaterials is their separation. The traditional method for separating is nanomaterial filtration. The main problem of this method is the filter clogging, constraining its uses in many areas (Ai et al. 2010). In this regard, magnetic separation has been developed to ease the accumulation of nanomaterials. Magnetic nanomaterials have many advantages, like high adsorption, rapid separation and ease of use, and their compatibility with environmental purification (Zhou et al. 2014). Magnetite nanoparticles have advantages such as high adsorption capacity and easy separation by a strong magnetic field, which is useful for recycling and reuse (Babaei et al. 2012). Among different magnetic nanomaterials, MFe₂O₄ type ferrites (M, divalent metallic cation) are magnetic materials with a cubic spinel structure used extensively in different technological applications in recent decades (Ai et al. 2010). Among the ferrites, cobalt ferrite (CFO) is an interesting magnetic material with a cubic spinel structure, which can be used to adsorb various pollutants due to moderate saturation magnetization, high chemical stability, and mechanical hardness (Ai et al. 2010; Zhou et al. 2014).

In this study, CFO magnetic nanoparticles were synthesized and the effects of different parameters like initial pH, adsorbent dose, contact time, initial concentration and temperature on nitrate removal from aqueous solutions were examined.

All chemicals used in this study were of laboratory purity. The chemicals such as potassium nitrate (KNO₃), cobalt nitrate (Co(NO₃)₂.6H₂O), iron nitrate (Fe(NO₃)₃. 9H₂O), hydrochloric acid (HCl), sodium hydroxide (NaOH), and ethanol were obtained from Merck Co., Germany. The instruments used in this research were a spectrophotometer (PG Instrument Ltd. spectrophotometer T80+ UV/VIS), pH meter (Hach, HQ411d, USA), shaker (multi-shaker, model NB-101MT, Korea), shaker incubator (SI-100R, Korea), and magnetic stirrer (heating magnetic stirrer, HM-101, Italy).

Iron nitrate (Fe(NO₃)₃.9H₂O) and cobalt nitrate (Co(NO₃)₂.6H₂O) were used for CFO synthesis. First, the above materials were dissolved in deionized water and added to each other. Then, sodium hydroxide 1 M was added to the above solution and placed at 80 °C for 2 hours. After the formation of CFO nanoparticles, these nanoparticles were separated by a magnet (Saffari et al. 2014). Then, scanning electron microscopy (SEM), transmission electron microscopy (TEM), Fourier transform infrared (FTIR) spectroscopy, X-ray diffraction (XRD) and vibrating-sample magnetometry (VSM) analyses were used to determine the synthesized CFO specification.

In this equation, q_e is the adsorption capacity of CFO (mg/g), C₀ is the initial concentration of nitrate (mg/L) and C_e is the concentration of the residual nitrate (mg/L), m is the adsorbent dose (g), and V is the sample volume (L) (Song et al. 2016).

In this stage, the equations presented in Table 1 were used for the isotherm, kinetics and thermodynamics study of nitrate adsorption by CFO (Foo & Hameed 2010; Moradi et al. 2013; Baghani et al. 2017; Sharifi et al. 2019; Yousefi et al. 2019).

In this study, the morphology of CFO nanoparticles was examined using SEM and TEM analyses (Figure 1). As shown, the diameter of these nanoparticles was about 76 nm. Moreover, the accumulation of these nanoparticles is clearly seen.

FTIR analysis was used to determine the molecular structure of synthesized CFO nanoparticles. Figure 2 shows the results of this analysis. By examining this spectrum, it was found that the peak at 3,406 cm⁻¹ is specific to the O-H group. The peak at 1,625 cm⁻¹ indicates the C=C group. The peaks at 898 and 963 cm⁻¹ are related to the O–H out-of-plane vibration group, and peaks at 600 and 410 cm⁻¹ correspond to Fe-O and Co-O groups (Li et al. 2010).

In Figure 3, CFO nanoparticle XRD pattern is specified. As shown, the positions and relative intensity of the peaks show the crystalline structure and the cubic spinel structure of synthesized CFO nanoparticles, which is similar to the literature with JCPDS card number 1121-01 (Saffari et al. 2014).

The magnetization curves measured for CFO are shown in Figure 4. As shown in this figure, the saturation magnetization value of the CFO was 42 emu/g.

To determine the pH of point of zero charge (pH_pzc), 50 mL of distilled water was poured into six Erlenmeyer flasks (Erlenmeyer volume 100 mL) and the pH of the solutions was adjusted by using HCl and NaOH 1 and 0.1% N solutions. Then, 0.025 g of CFO nanoparticles was added to each the Erlenmeyer flasks. After 24 hours, the final pH of each Erlenmeyer flask was read and the pH_pzc graph was plotted (Figure 5). As is seen in the figure, pH_pzc of the CFO nanoparticle is about 7.3.

Examining the effect of solution pH on nitrate adsorption by CFO showed better adsorption in acidic pH. With increase in pH from 3 to 11, adsorbent adsorption capacity decreased and the maximum value was at pH = 3, which was about 28.9 mg/g (Figure 6).

Given the negative charge of nitrate ions, pH can have a significant effect on its ion adsorption. According to Figure 6, at pH = 3, the highest adsorption was observed for CFO. Then, with increase in pH from 3 to 11, the adsorption decreased. To explain the reason for this, according to the adsorbent pH_pzc (pH_pzc = 7.3), the adsorbent surface charge is positive at pH lower than 7.3 and the adsorbent surface charge is negative at pHs higher than 7.3 (Yaghmaeian et al. 2016). As a result, at a pH of 3, due to the fact that the nitrate charge is negative and the adsorbent surface charge is positive, the highest amount of nitrate adsorption has occurred. Similar results were obtained by Rajeswari and colleagues for the removal of nitrate ions using chitosan/polyethylene glycol and chitosan/polyvinyl alcohol polymer composites (Rajeswari et al. 2016). Moreover, another study found that the highest nitrate adsorption was at pH = 3 (Ghanizadeh et al. 2015).

The results of the effect of adsorbent dose with the values 0.2, 0.4, 0.6 and 0.8 g/L at optimal pH are shown in Figure 7. According to Figure 7, adsorption capacity decreased with increase in adsorbent dose. Thus, 0.2 g/L was considered as the optimal dose. At this dose, the adsorption capacity is 38.8 mg/g.

As already stated, adsorption capacity decreased with increase in adsorbent from 0.2 to 0.8 g/L. This can be due to the non-saturation of adsorbent active positions during the adsorption process (Yaghmaeian et al. 2016). In other words, with increasing the adsorbent dose, the active adsorbent sites are more than the saturation threshold adsorption points. As a result, only a few active adsorbent sites are filled with pollutants, leading to a reduction in adsorption capacity (Wu et al. 2013a). Similar results are reported for nitrate removal by nano zero-valent iron in a study by Yaghmaeian and colleagues. In this study, with increase in adsorption capacity from 1.25 to 2 g/L, adsorption capacity decreased from 61.7 to 43.8 mg/g (Yaghmaeian et al. 2016).

In Figure 8, the result of the effect of contact time and initial nitrate concentration on adsorption of nitrate by the CFO is shown. This result showed that with increase in initial concentration of nitrate, CFO adsorption capacity increased. Moreover, at the initial moments of the reaction (up to 20 minutes), the value of adsorption reached its maximum, and then reached approximately equilibrium. Hence, the increase in contact time up to 120 minutes did not have much effect on nitrate adsorption by CFO. Moreover, the adsorption capacity increased with increase in the initial concentration of nitrate.

According to Figure 8, the maximum adsorption capacity was obtained at the concentration of 200 mg/L and the contact time of 20 minutes, and was equal to 99.1 mg/g. With increase in the concentration of nitrate, the adsorption capacity increased as well. Moreover, in the first 20 minutes, the slope of the nitrate adsorption curve by CFO was sharp, and then with increase in contact time, its slope is almost constant. This can be due to more access of connecting sites near the adsorbent surface (Chatterjee & Woo 2009) In other words, in the early stages, due to the higher driving force that causes the rapid transfer of nitrate to the adsorbent surfaces and also the availability of a large number of active sites in the adsorbent, the adsorption capacity is high (Wu et al. 2013b). Saturation of the adsorbent surface after 20 minutes and the lack of nitrate access to sufficient adsorbent sites can be due to other reasons (Chatterjee & Woo 2009).

Moreover, the increase in the nitrate concentration from 50 to 200 mg/L increases CFO adsorption capacity (Figure 8). This behavior can be explained by increasing the concentration gradient driving force with increase in the initial nitrate concentration The increase in adsorption capacity with increasing nitrate concentration can be due to the high probability of a collision between nitrate and the adsorbent surface (Bhatnagar et al. 2010). Similar results were obtained by Ganesan colleagues regarding nitrate removal by graphene (Ganesan et al. 2013).

Identifying adsorption isotherms is to better understand adsorbent surface properties and to express how adsorption is done (Nemati Sani et al. 2014). Table 2 shows the information on nitrate adsorption isotherms by CFO. Given the regression coefficients of different isotherms, the nitrogen adsorption reaction by CFO follows Freundlich and Langmuir models (R² = 0.99). However, as R_L in the Langmuir isotherm is between 0 and 1, it shows that the adsorption process is optimal and more consistent with the Langmuir isotherm. The basis of the Freundlich isotherm is multilayer and heterogeneous adsorption. However, unlike the Freundlich model, the Langmuir isotherm is based on single-layer and homogeneous adsorption (Nemati Sani et al. 2014).

Table 3 shows the results of the traditional study of nitrate adsorption by CFO. The calculations obtained from kinetic equations show that the kinetics of nitrate removal using CFO follows the pseudo-second-order kinetics. The results of this stage are similar to those of Nemati Sani et al. (2014).

The effect of temperature increase on the nitrogen adsorption capacity of CFO is shown in Figure 9. According to this figure, with increase in temperature from 283 to 313 K, adsorption capacity increased from 74.4 to 107.8 mg/g. As the adsorption capacity of the adsorbent increases with temperature, one can state that the adsorption of nitrate by CFO is an endothermic process. Moreover, with increase in temperature, the interaction forces between solute and solvent are weaker than those that exist between the solvent and the adsorbent (Ganesan et al. 2013).

Additionally, the nitrate adsorbed at different temperatures (283–313 K) by CFO was used to obtain the thermodynamic parameters of the adsorption process (Table 4). The positive values of Gibbs free energy variation (ΔG) show that nitrate adsorption by CFO was non-spontaneous. The positive values of enthalpy change (ΔH) for adsorbents show the nature of the endothermic process, which confirms the increase of nitrate adsorption with an increase in its temperature. The positive value of adsorption entropy change (ΔS) shows the tendency of nitrate ions to adsorb and increase the value of adsorbent along the solution during adsorption (Gueu et al. 2007; Ganesan et al. 2013).

The results of the effect of initial pH on the removal of nitrate by CFO showed that the highest nitrate adsorption occurred at pH = 3. Also, the effect of adsorbent doses in removal of nitrate showed that the increase in the dose of adsorbent reduces the adsorption capacity. In general, the results of this research indicated that the maximum adsorption capacity was 107.8 mg/g (optimum condition pH = 3, adsorbent dosage: 0.2 g/L, nitrate concentration: 200 mg/L, contact time: 20 min and temperature: 313 K). Adsorption isotherm information indicated that the nitrate adsorption process matched Langmuir (R² = 0.99) and Freundlich (R² = 0.99) models. Moreover, the kinetics of the adsorption process follows pseudo-second-order kinetics. The results of the thermodynamics of the nitrate adsorption process by CFO showed that the values of ΔG, ΔH and ΔS were positive. Therefore, this process was endothermic and non-spontaneous.

This study was carried out in the form of a research project approved by Birjand University of Medical Sciences. We are grateful to the never-ending support of this university for this research.

All relevant data are included in the paper or its Supplementary Information.