Ultrasound Assisted Adsorptive Removal of Cr, Cu, Al, Ba, Zn, Ni, Mn, Co and Ti from Seawater Using Fe₂O₃-SiO₂-PAN Nanocomposite: Equilibrium Kinetics

Abstract

This work reports the preparation and application of Fe₂O₃-SiO₂-PAN nanocomposite for the removal of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ from seawater. X-ray diffraction (XRD), scanning electron microscope/energy dispersive X-ray spectroscopy (SEM/EDS), transmission electron microscope (TEM), and Brunauer-Emmett-Teller (BET) characterized the synthesized composite. The following experimental parameters (Extraction time, adsorbent mass and pH) affecting the removal of major and trace metals were optimized using response surface methodology (RSM). The applicability of the RSM model was verified by performing the confirmation experiment using the optimal condition and the removal efficiency ranged from 90% to 97%, implying that the model was valid. The adsorption kinetic data was described by the pseudo-second order model. The applicability of the materials was tested on real seawater samples (initial concentration ranging from 0.270–203 µg L⁻¹) and the results showed satisfactory percentage efficiency removal that range from 98% to 99.9%. The maximum adsorption capacities were found to be 4.36, 7.20, 2.23, 6.60, 5.06, 2.60, 6.79, 6.65 and 3.00 mg g⁻¹, for Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti⁴⁺, respectively.

1. Introduction

Water crisis is the biggest problem faced by humanity and increased in water scarcity have a negative effect on economic development and human livelihoods [1]. The desire for clean water is caused by increase in global population, industrial activities, and development. Water scarcity can be resolved by means of building dams [2], use of ground water recharge [3], wastewater re-use [4] and desalination [5], among others, which are somewhat limited. However, the only endless resource that can produce high yield water is the ocean. The main concern about ocean water is that various types of contaminants around the globe are discharge in it. Then, this has affected the aquatic ecosystems and public health. This is motivating the search for a better technological solution to water shortage, whilst protecting ecosystem’s health. Potential toxic metals (PTMs) are some of the pollutants that are known to drive the reduction of marine life due to their toxic effects on living organisms [6]. This has also generated a huge interest amongst scientists and environmentalists in determining the global distribution of dissolved PTMs in the ocean.

Various methods such as chemical precipitation, ion exchange, biosorption, reverse osmosis and filtration have been used for removal of PTMs [7,8,9,10]. However, these methods are not commonly used because they are very expensive and their feasibility is extremely low [11,12]. On the other hand, adsorption technique remains an attractive method for removal of heavy metals due to its high removal efficiency and affordability [13,14].

Several scientists have conducted research on the development of cost effective and efficient use of the adsorbent for heavy metal removal. Therefore, metal oxides have been studied in depth for adsorptive removal of heavy metal. These include nanosized metal oxides (NMOs), which have good characteristics such as high surface areas and high activity [15,16,17]. Thus, α-Fe₂O₃ was found to be a more attractive alternative adsorbent to water treatment due to its cost-effectiveness and non-toxicity [18]. The nanostructures have a large surface area with excellent adsorption properties [19]. However, NMOs are unstable because of their nanoscale size, which leads to aggregation as a result of Van der Waals forces and later to the decrease in adsorption efficiency. In order to improve the stability state of NMO, their impregnation onto porous supports of natural materials, synthetic polymeric hosts and activated carbon has been reported [20]. These led to organic-inorganic polymer hybrids being used for its high transparency and excellent solvent-resistance [21].

Different methods have been used for preparation of different types of sorbents, these include the ex-situ and in-situ synthesis [22,23]. The use of in-situ synthesis has been the most used methods for the synthesis of polymers and nanoparticles to achieve homogeneous and well-dispersed material in polymer solution [24]. For this reason, preparation of composite materials such zeolitic imidazolate framework-8-PAN [23], sodium alginate-melamine sponge [25], poly (ether sulfone) (PES) and sulfonated poly (ether sulfone) (SPES) [26], and many others, were applied for heavy metal removal. Liu et al. (2011) showed that the uptake of As (III) was successfully accomplished using As (III) imprinted α-Fe₂O₃-impregnated chitosan beads [19]. Park et al. (2017) successfully impregnated Fe-Ti bimetal oxides into polymeric beads with the overall metal content of 4–6 wt.% [27].

Recently, the use of ultrasound irradiation has gained more attention in various applications [16,28,29,30,31]. This is because ultrasound assists in speeding the chemical process through the formation of acoustic cavitation, which is due to the propagation of pressure waves through liquid [28]. The process creates growth and collapse of the micrometer scale bubbles formed by pressure wave, which helps in strengthening mass transfer process. This, in turn, facilitates the interaction between adsorbate (major and trace metal ions) and adsorbent, thus leading to enhanced adsorption process [30,31]. In addition, shock waves have the ability of forming microscopic turbulence within the interfacial films that surrounds the solid particle [29]. Furthermore, this helps in accelerating interaction between the adsorbent and adsorbate (major and trace metals), thus reducing the time required to reach the equilibrium process [30,31].

In this study, a Fe₂O₃-SiO₂-PAN nanocomposite was applied for the first time as an adsorbent for treatment of PTMs in synthetic saline water samples and seawater. Where the synthesis of Fe₂O₃-SiO₂-PAN composite material was achieved through an in-situ process. Iron oxide was selected for this study because it is a cheap material that can be employed for easy separation, with large surface area and specific affinity [32,33]. Meanwhile, mesoporous silica has recently attracted huge interest as a suitable adsorbent for the removal of various pollutants due to its unique physicochemical properties [34]. These include properties such as possible re-use, mechanical resistance, and easy modification. In addition, an inert silica coating on the surface of magnetite nanoparticles prevent their aggregation in liquid substances or matrix [35]. Therefore, many research studies have applied functionalized silica as an adsorbent in the analysis of various metals and compounds [36,37,38]. However, it has been reported that inorganic adsorbents tend to cause operational problems such as clogging of filter membranes [39]. Therefore, incorporation of polymer matrix in the inorganic adsorbents has been found to act as inert organic binder for the removal of PTMs in complex matrices [39]. In this study, the polymer of choice was polyacrylonitrile (PAN because of its cost effectiveness and attractive properties [40,41]. Some of its attractive properties include excellent molding to pellet property, low density, strong attractive forces with inorganic materials, and chemical and mechanical stability [39,42]. Thus, composite material Fe₂O₃-SiO₂ is an excellent candidate for producing composite fibers along with PAN due to their combined properties such as chemical stability and flexibility, among others. The combination of these materials presents a novel class of composite nanofibers that entails unique advantages as compared to other sorbents material used before. The following experimental parameters (Extraction time, adsorbent mass and pH) affecting the removal of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ were optimized using a multivariate approach, namely a response surface methodology (RSM) based on the Box-Behnken design.

2. Experimental

2.1. Materials and Reagents

Ultra-pure water (Direct-Q^® 3UV-R purifier system, Millipore, Merck, Germany) was used in these experiments. Tetraethyl orthosilicate (TEOS), ammonium hydroxide (NH₄OH) (25%, w/v), methanol (99.9%, HPLC grade), absolute ethanol, polyacrylonitrile (PAN, average Mw 150,000), N, N-dimethylformamide (DMF), tween-80 and sodium hydroxide, nitric acid (HNO₃) and ferric nitrate (Fe(NO₃)₃ 9H₂O) were purchased from Sigma-Aldrich (St. Louis, MO, United States). A multi element standard solution of 100 mg L⁻¹ containing the following elements of interest, (Al, Ba, Cd, Ca, Cr, Co, Cu, Fe, Pb, Mg, Mn, Ni, Na, Ti, and Zn) was supplied by Spex CertiPrep (Industrial Analytical (Pty) Ltd., Johannesburg, South Africa). The multi-elemental standard was also utilized for preparation of calibration standards. The pH of the model solutions was adjusted with 1.0 mol L⁻¹ HNO₃ and NH₄OH.

2.2. Instrumentation

The oven (CEM Corporation Mars 6, Matthews, NC, USA) was used as drying source for the synthesis of the material. The inductively coupled plasma atomic emission spectroscopy (ICP-OES) (iCAP 6500 Duo, Thermo Scientific, UK) was used for quantification of analytes in sample solutions. The crystallinity of the material was studied by X-ray diffraction (XRD) analysis using PANalytical X’Pert X-ray Diffractometer and Cu Kα radiation spectrometer and the scanning area covered the range 2-theta at start position 4.00–80.00. The scanning electron microscopy (SEM) (HITACHI COM-S-4200) was used to study the morphology of the material. The Energy Dispersive X-ray (EDX) spectroscopy was connected to the SEM for determination of the ratio of Si/Fe. Jeol JEM-2100F field emission electron microscopy instrument (JEOL Inc, Akishima, Japan) was used for transmission electron microscopy (TEM) studies. The preparation of TEM samples was done by putting a small quantity of synthesized sample that has been dispersed into copper grid with carbon film. The surface area was analyzed using BET micrometric ASAP 2020.

2.3. Synthesis of Fe₂O₃-SiO₂ by Sol Gel Method

The preparation, of Fe₂O₃-SiO₂ was adopted from Ref [43] with some modifications. Mesoporous Fe₂O₃-SiO₂ composite was prepared by TEOS and ferric nitrate Fe(NO₃)₃·9H₂O respectively. The 300 mL of deionized water was mixed with 380 mL of absolute ethanol and stirred for 15 min for the preparation of mesoporous Fe₂O₃-SiO₂ composite. The 23.5 g of TEOS (98%) was added to the resulting solution and vigorously stirred for 30 min. Then, to the above clear solution, 4.6 g of Fe(NO₃)₃·9H₂O (Si/Fe = 50, respectively) was added at once and stirred for 30 min. For gelation to take place 115 mL of ammonium hydroxide was added and the formed precipitate was stirred for another 30 min and aged for 24 h at 25 °C. The material was then dried in an oven at a temperature of 60 °C for 24 h and calcined at 550 °C for 4 h in a furnace. The final product was cooled at room temperature and stored for further use.

2.4. Preparation of Fe₂O₃-SiO₂-PAN Nanocomposite

Fe₂O₃-SiO₂-PAN adsorbent was prepared following the procedure reported by İnan and Altaş [39]. Fe₂O₃-SiO₂ hydrous oxide powders synthesized were used in the experiment as inorganic active ion exchangers in the organic-inorganic composite beads. The prepared Fe₂O₃-SiO₂-PAN was composed of about 65 wt.% of Fe₂O₃-SiO₂ on a PAN polymeric support. A mass (10 g) of Fe₂O₃-SiO₂ hydrous powder was mixed with 50 mL of DMF (N, N-dimethylformamide) and a few drops of Tween-80 surfactant was stirred at a temperature of 50 °C for 2 h to form homogeneous solution. Then 2 g of PAN were added in the stirring solution and temperature was kept at 50 °C for 2 h to obtain homogeneous solution of the composite dope. Ultra-pure water/methanol alcohol mixture at a ratio of 2:1 was used as a gelation agent. The gelled composite beads were left 24 h for aging and washed using ultra-pure water. Modification was done on the surface of the spheres by 1 M NaOH and then washed and air-dried at 70 °C for 2 days to remove the solvent. The adsorbent was characterized using-ray diffraction (XRD), scanning electron microscope/energy dispersive X-ray spectroscopy (SEM/EDS), transmission electron microscope (TEM), and Brunauer-Emmett-Teller (BET).

2.5. Optimization of the Adsorption Batch Method

The optimisation method of the experimental conditions was performed using the Box-Behnken design matrix on the following factors: pH, extraction time (ET) and mass of adsorbent (MA). The minimum and maximum levels of the factors were generated and are shown in

Table 1. Variables and levels used in Box–Behnken design.

Variables	Low Level (−)	Central Points (0)	High Level (+)
pH	3.00	6.00	9.00
Extraction time (ET) (min)	5.00	17.5	30.0
Mass of adsorbent (MA) (mg)	100	200	300

. The results were evaluated using the recovery of Co and similar results were obtained for all the metals.

2.6. Ultrasound Assisted Adsorptive Removal

In these experiments, 20 mL of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ solution contained in 100 mL plastic bottles were contacted with 100 to 300 mg of Fe₂O₃-SiO₂-PAN nanocomposite adsorbent. The latter were then placed in a sample rack that was then dipped in an ultrasonic water bath at a temperature of 25 °C. Equilibrium adsorption studies on removal of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ metals ions using ultrasonic assisted adsorptive removal method was performed in an ultrasonic bath for 5–20 min. The appropriate amount of supernatant was collected from the sonicated samples, filtered and analyzed using ICP-OES. This same procedure was performed for removal and treatment of PTMs in real samples. The adsorption capacity (Qe, mg g⁻¹) was calculated using Equation (2).

2.7. Application to Real Samples

Adsorption of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺ and Ti⁴⁺ in seawater was investigated and, experiments were conducted using the optimum conditions obtained from the RSM. Real water (seawater) samples collected from Durban, South Africa, were used to evaluate the applicability of adsorption method. The collected seawater samples were stored in lab plastic bottles for further analysis at a temperature of 4 °C, whilst pH and conductivity were found to be 8.3 and 47.7 mS/cm, respectively. The batch adsorption experiments of seawater samples were carried out using the optimum conditions and the procedure elaborated in Section 2.6 was used.

2.8. Data Analysis

The adsorption efficiency was calculated using Equation (1).

$$Adsorption efficiency=\frac{C_o-C_e}{C_o}\times 100$$

(1)

where C_o is the initial concentration in mg L⁻¹ and C_e is the final concentration mg L⁻¹.

The adsorption capacity that is the amount of metal ions adsorbed per gram adsorbent (mg g⁻¹). Its equation can be written as follows:

$$Adsorption capacity=\frac{C_o-C_e}{m}\nu$$

(2)

where C_o is the initial concentration in mg L⁻¹, C_e is the final concentration mg L⁻¹, m is the amount of the adsorbent in grams (g) and v is the volume of the sample solution measured in liters (L).

3. Results

3.1. Surface Identification and Characterisation

Prior to analysis, the prepared samples were finely crushed to powder, mounted in the sample holder and loaded in the sample rack analysis. The XRD analysis of Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN are demonstrated in Figure 1. X-ray diffraction patterns ware analyzed by scanning from 4.00–80.00° 2-theta range. Figure 1a shows XRD patterns of SiO₂-Fe₂O₃ with crystalline structure of Fe₂O₃ at 2-theta values 35.3°, 44.7°, 56.1°, and 63.8°. These major characteristic peaks can be indexed as 104, 113, 116, and 300 according to the peak list obtained from the XRD report data. According to Debye-Scherer’s equation the particle sizes for Fe₂O₃-SiO₂ nanocomposite ranged from 50–110 nm. These results are in agreement with the pattern reported by Panda et al. [43]. In addition, this crystalline structure was found to be monoclinic. When the polymer was introduced on the same material, an amorphous structure can be observed on the XRD pattern of Fe₂O₃-SiO₂-PAN (Figure 1b). Liu et al. [19] reported these observations.

Figure 2 shows the SEM images of Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN composite that was studied through the SEM/EDX. The SEM image in Figure 2b shows the random distribution of large sizes that have irregular shapes through the encapsulation of Fe₂O₃-SiO₂ particles by PAN. Bhaumik et al. [44], reported results with the same resemblance. These observations were different from the images observed in Figure 2a, which confirms the incorporation of PAN.

The TEM images of Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN are shown in Figure 3. The Fe₂O₃-SiO₂ (Figure 3a) shows structures like hexagonal in shape. It is clear that Fe₂O₃-SiO₂ nanoparticles are incorporated on the surface of PAN. This is evident based on the distinctive film of PAN surrounding black spots that represent Fe₂O₃-SiO₂ nanocomposite. These results were further confirmed by EDX mapping densely covered Fe₂O₃-SiO₂ spheres. Setshedi et al. [45] and Teo et al. [46] reported similar observations.

Figure 4 shows the SEM-EDS spectra of the prepared Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN composites. The prepared composite material shows the presence of carbon (38%) compared to Fe₂O₃-SiO₂. The appearance of carbon in the spectrum of Fe₂O₃-SiO₂-PAN composite (Figure 4B) confirms the presence of PAN. The presence of carbon in Fe₂O₃-SiO₂ nanocomposite (Figure 4A) and Au in Fe₂O₃-SiO₂-PAN composite (Figure 4B) was a result of carbon and gold coating.

The dispersion of C, N, Fe, Si and O atoms was further investigated by EDX-mapping analysis (Figure 5). As seen in Figure 5, the composite composed of the expected elements that is C, N, Fe, Si, and O. The Fe₂O₃-SiO₂ nanoparticles were uniformly deposited on the surface of PAN.

The physical property analyses of Fe₂O₃-SiO₂ nanocomposite gave some understanding on the effect of cross-linking reaction to the chemical and physical properties of Fe₂O₃-SiO₂-PAN. The BET results are very important in explaining adsorption capacity of the adsorbents towards adsorbates. Data obtained from BET surface area analysis on Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN were presented in

Table 2. Summary of BET analysis.

Materials	Surface Area (m2 g−1)	Pore Volume (cm3 g−1)	Pore Size (nm)
Fe2O3-SiO2 PAN	253 32.0	0.96 0.26	14.4 35.4
Fe2O3-SiO2-PAN	158	0.53	22.1

. The surface areas of Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN were 253 and 158 m² g⁻¹, respectively. The reduced surface area on the composite may be due to the polymer layer shrinking around the nano-metal oxides matrix (Figure 3b).

Liu et al. (2011) showed that the internal pore structure of each material plays an important role in the adsorption performance of different adsorbate [19]. For this reason, the average pore diameter of Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN were investigated and results were shown in Table 2. The pores are divided in comprehensive terms according to the size of their diameter (d) (IUPAC classification). Results show that both Fe₂O₃-SiO₂ and Fe₂O₃-SiO₂-PAN are mesopores in nature (2 < d < 50 nm). Therefore, Fe₂O₃-SiO₂-PAN composite could be a suitable adsorbent for the removal of PTMs. In addition, Munonde et al. [47], reported that nanocomposite material has a variety of metal oxides with different shapes and sizes, which ensures more active sites due to more atoms on the surface and edges of the composite.

3.2. Optimization of the Adsorption Batch Method

The central composite design (CCD) matrix and the experimental data of Co are tabulated in

Table 3. Box–Behnken design matrix and analytical response.

	pH	ET	MA	%Re
1	3.00	5.00	200	25.5
2	9.00	5.00	200	91.7
3	3.00	30.0	200	43.5
4	9.00	30.0	200	92.6
5	3.00	17.5	100	14.1
6	9.00	17.5	100	87.4
7	3.00	17.5	300	76.6
8	9.00	17.5	300	92.4
9	6.00	5.00	100	48.0
10	6.00	30.0	100	40.6
11	6.00	5.00	300	72.2
12	6.00	30.0	300	89.2
13	6.00	17.5	200	51.1
14	6.00	17.5	200	79.3
15	6.00	17.5	200	70.8

; similar results were obtained for the other metals. The experimental results were statistically analyzed by means of analyses of variance (ANOVA), which is presented in a form of a Pareto chart (Figure 6). The parameters that were more influential on the adsorption process were pH, adsorbent mass and the interaction of pH and adsorbent mass. This can be noticed by passing the 95% confidence level. This implied that the parameters that are responsible for quantitative removal of target analytes in synthetic samples are pH and adsorbent mass.

The quadratic models of the RSM was used to construct the response surface plots that were used to investigate the interactive effect of two independent factors and their interactions on the amount of trace metal adsorbed. The 3D plot of combined effect of pH with extraction time and adsorbent mass are shown in Figure 7. The maximum percentage recovery of the above 95% was obtained when pH was at the range of 8–8.3 and MA of 300–330 mg (Figure 7). Based on the RSM model the optimum condition were found to be MA 330 mg, sample pH 8.3 and extraction time of 24 min.

3.3. Confirmatory Experiments and Adsorption Capacity

The analytical data obtained from the RSM model under the optimized condition (pH 8.3, adsorbent mass of 330 mg and extraction time of 24 min) were validated by performing confirmatory experiments. According to the results given by the RSM model, the predicted response for sorption of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ ranged from 95% to 97%. The experimental results ranged from 94% to 98% and these results were in close agreement with the predicted response.

3.4. Adsorption Kinetics

The adsorption kinetics of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ were studied using the following kinetic models; pseudo-first order, pseudo-second order and the intra particle diffusion models. The equation for pseudo-first order is as follows:

$$\ln (q_e-q_t)=\ln q_e-k_{1}t$$

(3)

where q_t is the amount of adsorbate, adsorbed (mg g⁻¹) at time t, q_e is equilibrium adsorption capacity (mg g⁻¹), k₁ is the rate constant (min⁻¹) The first order rate constant can be calculated from the intercept and slope of the plot [48,49]. Pseudo second order equation is as follows:

$$\frac{t}{q_t}=\frac{1}{k_2^{}{q}_e{}^2}+\frac{1}{q_e}t$$

(4)

where the equilibrium sorption capacity (q_e), q_t is the amount of adsorbate, adsorbed (mg g⁻¹) at time t and the second-order constant k₂ (g mg⁻¹ min) (

Table 4. Kinetic parameters for pseudo-first order and second order model.

Pseudo-First Order
Ions	qeexp	k1 (min−1)	qe (mg g−1)	R2
Al3+	2.23	0.26	098	0.564
Ba2+	6.60	0.15	1.20	0.606
Cr3+	4.36	0.01	0.90	0.227
Cu2+	7.20	0.02	3.40	0.389
Co2+	6.65	0.23	1.30	0.899
Mn2+	6.79	0.31	0.76	0.910
Ti3+	3.00	0.05	1.40	0.874
Ni2+	2.60	0.06	0.58	0.305
Zn2+	5.06	0.01	0.72	0.824
Pseudo-Second Order
Ions	qe (mg g−1)	k2 (g mg−1 min−1)	t1/2 (min)	R2
Al3+	2.16	1.00	0.48	0.999
Ba2+	6.80	0.03	4.90	0.918
Cr3+	4.34	7.50	0.03	0.999
Cu2+	7.35	0.42	0.32	0.999
Co2+	7.04	0.02	8.71	0.800
Mn2+	3.22	0.02	0.31	0.611
Ti3+	2.86	0.13	2.63	0.998
Ni2+	2.51	0.18	2.17	0.994
Zn2+	5.10	0.69	0.28	0.999

) can be determined experimentally from the slope and intercept of plot of t q_t⁻¹ versus t). Figure 8 shows the representative graphs for pseudo-second order equations. Kinetics were studied using the optimum parameters obtained from the RSM method; pH 8.3, adsorbent mass of 330 mg and the concentration of 200 mg L⁻¹. It should be noted that for simplicity reasons only four graphs were presented. In addition, the initial sorption rate (h) and the half-adsorption times were calculated from Equations (5) and (6).

$$h=k_2{q}_e^2$$

(5)

$$t_{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}=\frac{1}{k_2{q}_e}$$

(6)

Table 4 shows results for pseudo-first and second-order. The correlation co-efficient (R²) of pseudo-second order gives the best fit (R² ≥ 0.99) for sorption of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ onto Fe₂O₃-SiO₂-PAN together with R² ≥ 0.8 for Co²⁺. The pseudo-first order was followed by Mn²⁺ PTMs. The correlation co-efficient on this model was higher than the ones obtained in pseudo-first order model. Moreover, the q_e values that were calculated in pseudo-second order are in agreement with the experimental values obtained. This suggested that the adsorption process was chemisorption. Furthermore, the half-adsorption time is the time required in the removal of half of the amount of the analyte of interest at equilibrium [50]. The results show that the affinity was high between the adsorbent and metal ions; this can be seen through the short half-adsorption times achieved in most of the metals.

In order to get the information on the rate-limiting step, intraparticle diffusion was calculated and results were reported in

Table 5. Kinetic parameters for intra particle diffusion.

Ions	kid (g/mg min1/2)	Qe (mg g−1)	R2
Al	0.05	2.07	0.987
Ba	1.13	1.16	0.832
Cr	0.03	4.43	0.753
Cu	0.32	6.85	0.515
Co	1.01	1.20	0.908
Mn	1.38	6.27	0.949
Ti	0.23	5.10	0.951
Ni	1.00	1.64	0.660
Zn	0.03	4.91	0.907

. The step that limits the rate is either the boundary layer, which is the (film), or the intra particle (pore) diffusion of solute from the bulk solution to the adsorbent surface [51]. To investigate the chance of having intraparticle diffusion, Equation (7) was also used [51].

$$q_t=k_{id}{t}^{0.5}+C$$

(7)

Adsorption capacity (q_t) is calculated at any time t, the k_id is the constant for intra particle diffusion (mg g⁻¹ min^1/2) and C is the intercept. The experimental data of q_t versus t^1/2, was plotted and it was observed that a relatively good linear correlation existed between q_t and t^1/2. The intraparticle diffusion plots for adsorption of major and trace metals by the adsorbent showed that the regression lines did not pass the origin because C is non-zero. This implied that intraparticle diffusion was not the only rate-determining step [51,52,53,54]. This then suggested that both film diffusion and intraparticle diffusion influence the adsorption process. In addition, it is evident that stage 1 was influenced by electrostatic attraction between the external surface of the adsorbent the metal ions.

3.5. Application of Fe₂O₃-SiO₂-PAN in Real Samples

To evaluate the applicability of the optimized adsorption method, the synthesized adsorbent was used for the removal of PTMs from seawater collected from Durban, South Africa.

Table 6. Application of Fe₂O₃-SiO₂-PAN nanocomposite for removal of trace elements in real sample (n = 6 replicates).

Analytes	Initial Concentration (µg L−1)	Final Concentration (µg L−1) a	%RE
Al3+	203	2.79	98.6
Ba2+	0.991	0.003	99.7
Cr3+	0.270	0.002	99.2
Cu2+	17.2	0.077	99.6
Co2+	2.17	0.002	99.9
Mn2+	1.49	0.002	99.9
Ti3+	9.55	0.010	99.9
Ni2+	65.3	1.36	98.0
Zn2+	34.8	0.086	99.8

^a Obtained by subtracting the concentration of metals after desorption from the intimal concentration.

, shows the analytical results before adsorption and after adsorption as well as percentage removal efficiencies. As seen from Table 6, the percentage removal efficiencies ranged from 98 to 99.9 suggesting that the Fe₂O₃-SiO₂-PAN was suitable for adsorptive removal of Al³⁺, Ba²⁺, Cr³⁺, Cu²⁺, Co²⁺, Mn²⁺, Ti³⁺, Ni²⁺, and Zn²⁺ from complex matrix such as seawater. It should be noted that the concentration of the metals adsorbed on the adsorbent were desorbed using 1.5 mol L⁻¹ nitric acid. This was done in order to find the concentration after adsorption.

The Fe₂O₃-SiO₂-PAN composite was comparable with previous adsorbents reported in the literature for the removal of trace elements from seawater [52,53,54]. Chelating resins were applied for the removal of Ni, Cu, Zn, Cd, Pb, Co, Cr, and Mn from seawater. Results showed that chelating resin was able to remove about 80% to 104% of the trace elements from seawater [55,56]. Therefore, the performance of Fe₂O₃-SiO₂-PAN composite was comparable with other studies and it removed trace metals from the seawater better than the reported adsorbents from the literature (

Table 7. Comparison of heavy metal removal using other adsorbents.

Analytes	Adsorbents	Removal Efficiency (%)	Ref.
Pb (II), Cu (II), Cr (II), Cd (II)	Mabamboo activated carbon	99.9, 100, 96.4, 98.2	[57]
Zn (II)	Clinoptilolite	100	[58]
Ni (II)	Clinoptilolite	93.6	[59]
Cu (II), Cr (II), Ni (II)	Eryngium campestre	98.9, 98.2, 93.4	[60]
Fe, Pb, Cd, Cu, Ni	Fly ash	86.8, 76.1, 73.5, 98.6, 96.0	[61]
Cd, Cr, Mn, Cu, Ni, Pb, Zn, Fe	Aquatic plants	61.1, 69.2, 68.0, 79.1, 74.9, 62.1, 63.0, 81.2	[62]
Cr3+ Cu2+, Al3+, Ba2+, Zn2+, Ni2+, Mn2+, Co2+, Ti3+	Fe2O3-SiO2-PAN	99.2, 99.6, 98.6, 99.7, 99.8, 98.0, 99.9, 99.9, 99.9	This work

).

4. Conclusions

In this study, the application on the Fe₂O₃-SiO₂-PAN adsorbent was executed for the removal of major and traces metals; Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ from synthetic brine and seawater samples. The prepared Fe₂O₃-SiO₂-PAN material was characterized by SEM, EDX, TEM, XRD, and BET surface area. The transmission electron image of the composite material shows a core-shell structured material that was formed on the surface of PAN. These results were further confirmed by EDX mapping that has densely covered Fe₂O₃-SiO₂ spheres. The removal of Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ was optimized using the Box-Behnken design matrix on pH, extraction time and adsorbent mass. Kinetic studies were investigated by fitting adsorption data on pseudo-first order, pseudo-second order and intraparticle diffusion. The adsorption data was best described by pseudo-second-order kinetic model and the intraparticle diffusion was not the rate-limiting step. The maximum percentage removal efficiency of metal ions Cr³⁺, Cu²⁺, Al³⁺, Ba²⁺, Zn²⁺, Ni²⁺, Mn²⁺, Co²⁺, and Ti³⁺ ion seawater samples ranged between 98% to 99.9%. These results demonstrated that Fe₂O₃-SiO₂-PAN composite is a suitable material for the removal of trace elements from seawater when compared with other reported studies.