Fabrication of Magnetic Fe3O4/Polypyrrole/Carbon Black Nanocomposite for Effective Uptake of Congo Red and Methylene Blue Dye: Adsorption Investigation and Mechanism

In most wastewater systems, the existence of more than one dye is a reality. Herein, mesoporous iron oxide/polypyrrole/carbon black (Fe3O4/PPy/C) nanocomposite with high saturation magnetization (45.2 emu/g) has been prepared by encapsulating very fine Fe3O4 nanoparticles (4–7 nm) in PPy/C for the first time. The superior adsorption performance of the developed Fe3O4/PPy/C nanocomposite for the removal of basic dye (methylene blue, MB) and acidic dye (Congo red, CR) has been studied in single and binary systems. XRD, FESEM, EDX, HRTEM, FT-IR, VSM, BET, and zeta potential analyses were used for adsorbent characterization. XRD, EDX, and FT-IR findings confirmed the successful inclusion of Fe3O4 into the polymer matrix. The XDR and HRTEM results revealed a significant reduction in the size of the grown Fe3O4 with the addition of PPy/C, which acts as a particle growth inhibitor, and thus the nanocomposite has a greater surface area reached 341.01 m2/g relative to 210.66 m2/g and 277.59 m2/g for pure Fe3O4 and PPy/C, respectively. The Langmuir isotherm matched the experimental data quite well, with R2 and adsorption capacity of 0.994 and 500 mg/g for CR and 0.995 and 90.9 mg/g for MB, respectively. The effectiveness of Fe3O4/PPy/C nanocomposite to remove both CR and MB from binary solution reached 96.9% for CR and 95.9% for MB. In kinetic investigations, the sorption of both dyes obeyed the pseudo-second order model, where R2 equals 0.999 and 0.998 for CR and MB, respectively. Desorption tests revealed the easy separation and good efficiency of the nanocomposite after four cycles. Finally, the sorption mechanism was discussed in detail based on FTIR analysis. Therefore, the developed nanocomposite is a bi-functional, recyclable, and efficient adsorbent for environmental applications.


Introduction
It was reported that higher than 650 different contaminants were found in wastewater, including phenols, dyes, pesticides, heavy metals, and pharmaceuticals [1]. Specifically, dyes are organic chemicals that are commonly used to colour paper, plastic, ink, cosmetic, paint, textile, and food [2]. These companies pollute the environment by releasing coloured waste into the environment. Out of 7 × 10 5 tonnes of commercially used dyes annually generated, around 2.8 × 10 5 tonnes are dumped into water streams each year [1,[3][4][5]. Dyes in the outflow inhibit photosynthetic activity and, at extremely low concentrations (less than 1 ppm), pose a major harm to the aquatic environment [5][6][7]. Some pigments induce mutations, skin, lung, and cancer when inhaled [1,[3][4][5]8]. Dyes have thermal, optical, and physicochemical stability due to their complex aromatic structures, making degradability difficult [9]. As a result, dye removal from water sources is critical, and effluent from diverse businesses must be treated before being discharged into the aquatic environment.
There are different kinds of synthetic dyes in waste water. Dyes are classified according to ionic nature, into cationic, anionic, and non-ionic dyes, based on the charge that the dye acquired in its aqueous medium [10]. Congo red (CR) is one of anionic acidic dyes that has high solubility in aqueous environments and thermal stability. In addition, it exhibits resistance against light and biodegradation [5]. The textile, paper, and leather industries are the most common users of this dye. CR is extremely toxic, causing irritation to the skin and eyes, breathing problems, as well as stomach discomfort and blood clotting. It transforms into the carcinogen benzidine during metabolism. This dye was chosen for the for the present investigation because to its danger, strong persistence in the aquatic environment, and the necessity of elimination [1,3,5,11]. Methylene blue (MB) is a cationic soluble dye that widely known as basic blue 9. Like other organic dyes, MB has a variety of negative effects on humans; inhalation can cause difficulty breathing, while injection in the mouth causes burning sensations, vomiting, nausea, diarrhoea, and gasestrics, as well as painful micturation. Because of its wide speared usage as a dye in the textile industry, biological stain, and a range of other uses, in addition to the presence of large amounts of MB in industrial waste [5,12], MB was chosen as a pollutant in this study.
Carbon materials, such as activated carbon (AC), carbon nanotubes, graphene and its carbonaceous composites, have attracted researchers' interest as prospective sorbents owing to their large active surface area, high reactivity, porous structure as well as high mechanical and thermal strength and p-p electrostatic interactions [1,3,4,25,26]. In recent years, the use of carbon nanotubes and graphene oxide showed superior adsorption properties for dye removal when compared to activated carbon [27]. Mishra et al. [28] prepared multiwalled carbon nanotubes by chemical vapour deposition for the adsorption of CR dye. In another study, adsorption of CR was studied by using graphene oxide [29]. Despite the characteristics of such nanomaterials, they have high manufacturing costs and poor separation performance, restricting their scalability [27]. Owing to these disadvantages, researchers are actively looking for low-cost nanocarbonaceous materials with similar features, such as black carbon (CB). CB is mainly composed of elemental carbon (90%), with the rest being oxygen and hydrogen. It has various functional groups, including carbonyl (C=O), hydroxyl (OH), and carboxyl (COOH) [30]. Gaminian and Montazer [31] demonstrated the influence of CB as a filler on the properties of the obtained carbon nanofibers (CFs) derived from cellulose. They demonstrated that the synthesized CB/ CFs possess enhanced carbonization yield, thermal stability, electrical conductivity, surface area, and dye adsorption capacity in comparison with CFs obtained from neat cellulose. However, the difficulty in regeneration has not been solved yet. Thus, it is necessary to develop new materials exhibiting strong adsorption and regeneration ability, which is a critical step in the adsorption process.
Magnetic nanoparticles (MNPs) include magnetite (Fe 3 O 4 ), maghemite (γ-Fe 2 O 3 ), and hematite (α-Fe 2 O 3 ). Magnetite is the most common magnetic core employed in adsorption investigations, however, it soon oxidizes to maghemite, which is the more stable magnetic oxide in aqueous solution [20]. They possess unique properties, including low toxicity, small particle size, high surface area, biocompatibility, chemical stability, high saturation magnetization, and low synthesis cost [32][33][34]. The magnetic properties of iron oxide based adsorbents enable them to be separated from water after adsorption using gradient magnetic separation technique. As a result, magnetic based sorbents have gained high interest in adsorption because of its remarkable characteristics, including high reusability, easy separation, as well as preserving high efficiency after 3-5 cycles of regeneration [11,20]. Raw MNPs can be quickly oxidized by the environment and it could leach into low pH conditions, becoming poisonous. In addition, because of their large surface area and energy, they are subject to agglomeration in solutions, which reduces their sorption capacity and limits their use [20]. To compensate for these limitations, much focus has been given to surface modification of MNPs with organic polymer [35], clay [36,37], surfactant [38], silica [12], activated carbon [39,40], and etc. For the synthesis of magnetic-based nanocomposites, co-precipitation was recommended as being the most simple, fast, effective, and low-cost technique [37].
Polyaniline (PANI), polypyrrole (PPy), polyethylenimine, and polyacrylonitrile, are nitrogen-containing polymers and possess positive charged sites in which anionic dyes (like CR) can be adsorbed [41]. Among these polymers, PPy is the most frequently applied in various technologies owing to its multifunctional properties, including porous structure, excellent electrical conductivity, great thermal and environmental stability, low toxicity, ease of synthesis, non-localized π-electron conjugate system, and remarkable redox characteristics [18,42]. As a consequence, it can interact with the target molecule via a variety of intermolecular interactions, including hydrogen bonding, acid-base interactions, and pollutant-polymer exchange [43]. PPy as well as its modified forms have shown great effectiveness in eliminating several common contaminants, as reported in numerous publications. For instance, Xin and co-workers [44] prepared PPy nanofibers with high specific surface area of 86 m 2 /g and outstanding adsorption capacity (q max ) 1 3 for methyl orange (MO) dye reached 169.55 mg/g. However, according to Langmuir model q max of PPy for MB was only 49.7 mg/g as reported by Li et al. [45]. Bhaumik et al. [46] prepared PPy/PANI nanofibres for adsorption of CR (q max = 222.22 mg/g). Shahnaz et al. [47] modified nanocellulose with PPy for elimination of Cr (VI) and CR dye in binary mixture. Kamal et al. [48] developed a ternary nanocomposite composed of chitosan, graphene oxide, and PPy to eliminate heavy metals (Cd 2+ , Cu 2+ ) and CR dye. Wang et al. [49] prepared hyper-cross-linked polymer/PPy for efficient phenol adsorption. Nevertheless, separating adsorbent material from aqueous medium is a major source of consternation, restricting how effective adsorbents may be. As a result, the incorporation of magnetic NPs enables magnetic decantation to easily retrieve the nanocomposite from an aqueous solution.
A variety of magnetic polypyrrole based composites have been used for environmental applications. Magnetic NPs can be coated or integrated into a polymer matrix to give colloidal stability and enough protection across a wide range of pH values [50]. Recently, Marghaki et al. [18] studied the isotherms, kinetics, and thermodynamics for the adsorption of Cr (VI) using microbial cellulose/Fe 3 O 4 /PPy. Falahian et al. [51] used Fe 3 O 4 /PPy composite for Hg 2+ elimination. The removal of methyl orange and crystal violet dyes was studied by using polypyrrole modified chitosan-based magsorbent [52]. Recently, Han et al. [53]  Because both the physicochemical characteristics of the as-prepared adsorbent and the physicochemical properties of the target adsorbate influence adsorption performance, preparing an adsorbent with outstanding properties is a need for success. Competitive adsorption of several types of organic pollutants, such as anionic and cationic dyes, is the normal situation in practical applications. Numerous materials have exhibited good adsorption characteristics in single dye system, however, few only were examined for efficient sorption of different kinds of dyes in binary systems. As a result, we developed new adsorbent with improved adsorption capacity, stability, and recyclability by integrating the advantages of magnetic Fe 3 O 4 , along with organic polymer (PPy), and carbon based material for the effective uptake of basic dye (MB) and acidic dye (CR) from single and binary system. Until now, no research has been conducted on the utilization of Fe 3 O 4 /PPy/C nanocomposite as an adsorbent for environmental applications, as will be presented in this work. XRD, FTIR, FESEM, EDX, HRTEM, VSM, zeta potential, and BET were used to characterize the samples. The influence of different parameters, including pH, adsorbent dosage, contact time, and initial dye concentration were discussed. The adsorption kinetics and isotherms have been investigated. The reusability potential of Fe 3 O 4 /PPy/C nanocomposite is also established. The high adsorptive capability of the nanocomposite is governed by its large surface area, mesoporous structure, high surface roughness, and different function groups. Finally, the sorption mechanism of MB and CR onto the surface of magnetic Fe 3 O 4 /PPy/C nanocomposite was discussed in detail based on the FTIR results before and after adsorption of CR and MB dye.

Materials
Polypyrrole doped-20 wt. % composite with carbon black (PPy/C) was purchased from Sigma-Aldrich. Ferrous sulphate and ferric chloride were obtained from Loba Chemie. Congo red (CR) and methylene blue (MB) were supplied by S D Fine-Chem limited and their characteristics are summarized in Table 1.

Synthesis of Fe 3 O 4 and Fe 3 O 4 /Polypyrrole/Carbon Black (Fe 3 O 4 /PPy/C) Nanocomposite
Iron oxide (Fe 3 O 4 ) was prepared by co-precipitation method under nitrogen gas. Prepare 2:1 M solution of Fe 3+ and Fe 2+ ions and then an ammonia solution was added to it to raise pH up to 11 at a temperature of 70 °C and continue stirring for four hours. The resulting black precipitate was centrifuged, washed several times with deionized water, and then dried in a vacuum oven at 50 °C.
The synthesis of Fe 3 O 4 /PPy/C nanocomposite was carried out by dispersing 0.4 g of powdered polypyrrole/carbon black (PPy/C) in Fe 3+ /Fe 2+ solution and stirring it for 1 h. Then, follow the same steps as in the case of pure iron oxide synthesis. Scheme 1 revealed the synthesis steps of pure

Characterization
Fourier transform infrared (FTIR) spectra were measured to identify the function groups in the prepared samples by using Bruker instrument, Germany. At 77 K, the nitrogen adsorption/desorption isotherm was measured using Quantachrome apparatus, USA. The specific surface area, average pore diameter, and pore volume were evaluated using Brunauer-Emmett and Teller (BET) method. The surface microstructure and the chemical composition of Fe 3 O 4 , PPy/C, and Fe 3 O 4 /PPy/C nanocomposite were identified by using field emission scanning electron microscopy (FESEM, Quanta 250 FEG FEI, United States) and the energy dispersive X-ray spectroscopy (EDX), respectively. X-ray diffraction (XRD) examination employing PAN analytical X-ray diffractometers was used to determine phase purity, structural type, and crystallite size of the materials. The particle size and shape of the samples were determined using high resolution transmission electron microscopy (HR-TEM) utilizing the Jeol JEM2100 series, Japan. At room temperature, the saturation magnetization as well as the coercivity were measured using a vibration sample magnetometer (VSM, MicroMag 3900 series). The applied magnetic field ranged from − 20 to 20 KOe. Zeta potential values (ζ) of samples was measured using a nano Zetasizer analyzer (Malvern, UK) to determine the charge on their surfaces.

Adsorption Experiments
In the batch mode, CR and MB adsorption tests were carried out individually or in mixed solution using a shaker at 200 rpm as follows; 20 mg of Fe 3 O 4 /PPy/C nanocomposite was added to 40 mL of dye solution and, at various intervals of time, part of the supernatant was taken and centrifuged. The residual concentration of dye was determined by using spectrophotometer (2401 Shimadzu UV-vis, Japan). The two most important parameters of adsorption are adsorption capacity (q max ) and removal efficiency (RE), which indicate how successful an adsorbent is for removing pollutants. Adsorption capacity is a characteristic of an adsorbent toward the adsorbate, whereas removal efficiency is influenced by the concentration and dose of the adsorbate [20]. Using Eqs. 1 and 2, the dye removal efficiency and the amount of dye adsorbed, respectively, were estimated.
where, C o -initial concentration of each dye (mg/L), C tthe equilibrium concentration of each dye (mg/L), W-the mass of the nanocomposite (g), and V-the volume of dye solution (L). The influence of different parameters, including pH, adsorbent dosage, contact time, and initial dye concentration were discussed. The kinetics of adsorption were investigated. To explore the detail adsorption behavior and study adsorption isotherms, multiple starting concentrations of CR solution ranged from 40 to 140 mg/L and MB solution (20-60 mg/L) were utilized.

Reusability
The use of magnetic adsorbents is characterized by the simplicity be recovered after use just by applying external magnetic field to nanocomposite-dye suspension solution. The durability studies were conducted by immersing the resultant dye-adsorbed nanocomposite overnight in 0.1 mol/L solution of NaOH and HCl to release CR and MB, respectively, and then washing with distilled water for usage in the subsequent adsorption process. Four rounds of regeneration tests were carried out. Figure 1 displays the XRD patterns of PPy/C, Fe 3 O 4 , and Fe 3 O 4 /PPy/C nanocomposite. Pure PPy/C in Fig. 1a has an amorphous broad peak at 25° of 2θ, similar to other common polymers. This peak is attributed to amorphous polypyrrole [51,59]. The weak peak in the range of 40°-45° is related to amorphous carbon of carbon black [60].

HRTEM Analysis
For more clarity, the morphology of Fe 3 O 4 /PPy/C nanocomposite in comparison with pure Fe 3 O 4 was examined by HRTEM. The synthesized Fe 3 O 4 nanoparticles have regular spherical shape with diameter ranged from 9 to 14 nm, Fig. 4a, b, as estimated by using software ImageJ. Pure PPy/C displays spongy node-like structures that are highly interconnected (not shown) [65]. The Fe 3 O 4 /PPy/C nanocomposite, Fig. 4c, d, is composed of black spots grouped in some regions in the PPy/C gray area, indicating the growth of Fe 3 O 4 particles that are impeded and strong interacted with the polymer chains. Nevertheless, the presence of PPy/C during the synthesis of Fe 3 O 4 may act as a surfactant or particle growth inhibitor, which causes a significant decrease in the size of the grown Fe 3 O 4 particles to 4-7 nm as revealed by comparing Fig. 4b, d. This phenonium will have an impact on the surface area of the obtained Fe 3 O 4 /PPy/C nanocomposite, as will be illustrated in the surface area section. The very common strategy conducted by numerous researches [55,58] to synthesize magnetic-polypyrrole based composites is that polymerization of pyrrole in the presence of the powder of inorganic nanomaterial in the polymer forming solution, which resulted in an increase in the diameter of the nanoparticles as well as a decrease in the surface area of the obtained composite. Whereas, our modified method, which is based on encapsulation of the powdered polymer in the Fe 3 O 4 synthesis solution, led to a clear decrease in nanoparticle size in addition to an increase in the nanocomposite surface area.   [5,42,55]. The bending mode of adsorbed water molecules appeared at 1630 cm −1 [67]. Stretching vibration of C = C appeared at 1542 cm −1 [35]. The 1458 cm −1 band might be attributed to symmetric stretching of pyrrole ring as well as conjugated C-N stretching [35,68]. The peaks at 1376 and 1165 cm −1 are attributed to the in-plane deformation and bending of C-H of the pyrrole ring, respectively, [69,70]. At 1041 cm −1 , the N-H in-plane deformation appeared [68]. The spectrum of Fe 3 O 4 /PPy/C nanocomposite demonstrated the Fe 3 O 4 and PPy/C characteristic peaks. The pyrrole ring peak at 1542 cm −1 is enhanced. The Fe-O stretching peak shifted to 592 cm −1 . The peak at ~ 628 cm −1 could be correlated to Fe-O stretching in the Fe 3 O 4 [3,33]. These results confirmed that Fe 3 O 4 has been successfully introduced into polymer matrix.

VSM Measurement
For wastewater treatment, Fe 3 O 4 nanoparticles serve as effective magnetic carriers for absorbents or photocatalysts, enabling recyclability, reusability, nontoxicity, and low-cost materials [71]. The magnetic behavior of PPy/C, Fe 3 O 4 , and Fe 3 O 4 /PPy/C nanocomposite at room temperature was investigated using a vibrating sample magnetometer (VSM) as shown in Fig. 6. The values of the saturation magnetization (M s ), coercivity (H c ), and remanent magnetization (M r ) for all samples are summarized in Table 2. The magnetization curve of pristine PPy/C, Fig. 6a, revealed its diamagnetic behavior [72] with very small saturation magnetization (M s ) of 0.176 emu/g. This result confirming the nonmagnetic character of polypyrrole. However, both Fe 3 O 4 and Fe 3 O 4 /PPy/C possess a ferromagnetic behavior as presented by the hysteresis loops in the inset of Fig. 6b, c, [59]. However, the applied co-precipitation method is easier, faster, and more scalable than solvothermal method. Loading the non-magnetic PPy/C on the surface of magnetic nanoparties declines the magnetization to 45.19 emu/g while increasing the coercivity to 53.65 Oe owing to surface heterogeneity and external morphological alteration [73]. Various studies revealed that the sensitization of magnetic nanoparticles with PPy decreased their magnetic features [53,59,73]. Despite the fact that the prepared Fe 3 O 4 /PPy/C nanocomposite has a higher magnetic value than the majority of previous publications on magnetic Fe 3 O 4 /PPy composites [51,58,59,74], an outside magnetic field may separate it from water fast. The relatively low Hc value (< 100) proved that Fe 3 O 4 and Fe 3 O 4 /PPy/C possess soft magnetic characters [75].

Surface Area Analysis
Adsorption process is strongly influenced by the adsorbent's surface chemistry. The particle size, specific surface area, pore volume, and pore size of an adsorbent all have a role for efficient adsorption of contaminants [20]. Figure 7A shows the nitrogen adsorption isotherms of Fe 3 O 4 , PPy/C, and Fe 3 O 4 /PPy/C nanocomposite. According to Fig. 7a, the specific surface area of the synthesized Fe 3 O 4 was around 210.66 m 2 /g. When compared to the reported value for Fe 3 O 4 (33.3 m 2 /g), which was prepared using a solvothermal method, this result is obviously superior [59]. Also, Anjana et al. (Anjana et al., 2018) prepared Fe 3 O 4 by a co-precipitation method; however, the BET surface area was only 48.29 m 2 /g. The raw PPy/C has 277.59 m 2 /g specific surface area. This value is eight times higher than the surface area of polypyrrole alone (34.7 m 2 /g) [58]. Following the addition of PPy/C in the Fe 3 O 4 synthesis solution, the surface area of Fe 3 O 4 /PPy/C nanocomposite dramatically increased up to 341.01 m 2 /g. This might be attributed to the Fe 3 O 4 's smaller nanoparticle size, as well as the presence of PPy/C inside the obtained composite, acting as a surfactant or particle growth inhibitor. This value is much higher than the reported surface area of Fe 3 O 4 /PPy composites (19.8-10.3 m 2 /g), which were prepared by oxidative polymerization of polypyrrole in the presence of Fe 3 O 4 nanoparticles [58]. The isotherm of Fe 3 O 4 /PPy/C nanocomposite exhibited type IV with H3 hysteresis loop, indicating the formation of highly porous materials [76]. In addition, the pore size distribution curve using the DFT model is displayed in Fig. 7B. PPy/C has small pores of size 5.44 nm, whereas pure Fe 3 O 4 contains large pores with a size of 11.98 nm. The pore volume is 0.5502 cm 3 /g for PPy/C and 0.9055 cm 3 /g for Fe 3 O 4 . Relative to pure Fe 3 O 4 , the incorporation of polymer in the iron oxide synthesis solution to get Fe 3 O 4 /PPy/C nanocomposite lowered the pore size to 6.75 nm and pore volume to 0.8623 cm 3 /g. This behavior is owing to the fact that PPy/C was attached to the surface of Fe 3 O 4 , so many pores were partly or entirely occupied. The summary results regarding surface area, pore size, and total pore volume are seen in Table 2.

Adsorption Performance of Fe 3 O 4 /PPy/C Nanocomposite Toward Anionic Dye (CR), Cationic Dye (MB), and CR-MB Mixed Solution
To evaluate the adsorption activity of Fe 3 O 4 /PPy/C nanocomposite toward anionic and basic dyes, Congo red (CR) and methylene blue (MB), respectively, were chosen. The adsorption experiments were carried out using CR, MB, and their binary mixture. Figure 8 shows the UV-Vis absorption spectra of CR, MB, and CR-MB mixture solution in the presence of Fe 3 O 4 /PPy/C nanocomposite. The characteristic peak of CR dye, with initial concentration of 120 mg/L, Fig. 8a, diminished with the passage of time. The CR removal percentage is 94.9% after 180 min. The experimental adsorption capacity is 227.7 mg/g. Obviously, the adsorption for MB (40 mg/L) is faster relative to CR adsorption, as shown in Fig. 8b. In this case, the removal efficiency reached 98.2% within only 120 min. The experimental adsorption capacity for MB is 78.6 mg/g. The simultaneous sorption of anionic dye (CR) and basic dye (MB) using the developed Fe 3 O 4 /PPy/C nanocomposite was confirmed by testing the nanocomposite in CR-MB binary mixture as shown in Fig. 8c. The nanocomposite can achieve complete removal of both dyes. At 240 min, the Fe 3 O 4 /PPy/C nanocomposite revealed 96.9 ± 2% of decolorization for CR and 95.9 ± 1% elimination for MB.

Influence of Solution pH
The surface charge and functional groups of the adsorbent material could be influenced by the pH of dye solution, resulting in a direct effect on the overall adsorption efficiency. CR and MB removal by Fe 3 O 4 /PPy/C nanocomposite in a single dye system were carried out in the pH range of 4 to 10 in order to find the ideal pH. Adsorbent dose was 0.5 g/L and initial concentration of CR and MB was 120 and 40 mg/L, respectively. To alter the pH of dye solutions, NaOH and HCl were used. Figure 9a depicts the percentage of CR and MB removed by Fe 3 O 4 /PPy/C nanocomposite at various pH values. As pH rises from 4 to 7, CR removal percentage slightly decreased from 97.5% to 94.9%, whereas the adsorption efficiency significantly decreased to 89.4% at pH 10. According to the previous literature, the sulfonate moiety in CR dye dissociates into negative sulfonic groups (− SO 3 − ) at acidic conditions. As a result electrostatic interaction between positively charged adsorbent surface and dye molecules was induced, consequently enhances the adsorption process. Nevertheless, at high pH, extra OHions compete with negatively charged dye molecule for adsorption sites causing a decrease in dye removal percentage. As previously mentioned, pH 7 was the most effective for CR sorption onto graphene oxide [79], montmorillonite [80], and aminefunctionalized mesoporous hollow shells [81]. Thus, for the rest of our studies on CR adsorption, we chose to start with a solution with a pH of 7, which is the natural pH for dye solution in addition it is less expensive to study at neutral pH than at acidic or basic pH. For MB removal, Fig. 9a, demonstrated a clear improvement in removal efficiency as pH increases. Changes in surface electrical characteristics may be responsible for the phenomenon. At low pH values, the composite would protonated with H + , resulting in a decrease in negative charge. The declined electronegative composite would inhibit the adsorption of additional electropositive MB molecules. According to these findings, the MB adsorption significantly improved from pH 4 to 7 and remained stable from pH 7 to 10. Therefore, a pH of 8 was selected for the further MB adsorption experiments. This result is in agreement with the previous study on the removal of MB on PEO/Bentonite/PANI [82].

Influence of Adsorbent Dose
To investigate the effect of adsorbent dose, a series of adsorption tests were performed using adsorbent doses ranging from 0.1 to 0.7 g/L while maintaining a constant initial an optimum removal percentage value of 95% and 98% was obtained for CR and MB, respectively. After that, it nearly reaches equilibrium, resulting in a much smaller increase in dye removal efficiencies. This could be due to the fact that more active surface sites are available for dye adsorption at higher adsorbent dosages. Therefore, other adsorption experiments were conducted with adsorbent dose of 0.5 g/L. These findings agree with the recently reported [83,84].

Influence of the Initial Dye Concentration
The predominant trend reported in most of the papers is that increasing the concentration of pollutant declines the removal efficiency. The adsorption efficiency of CR was calculated as a function of increasing dye concentration from 40 to 160 mg/L, as presented in Fig. 9c. The CR elimination percentage stepwise decreased from 99.1% to 83.7% with raising the initial dye concentration. For adsorption of MB the dye concentration changed from 20 to 60 mg/L as shown in Fig. 9c. The MB dye adsorption efficiency declined from 99.6 to 75.6% as the initial concentration of dye increased. These findings are consistent with other studies [4,85,86]. According to the researchers, the drop in removal % with rising dye concentration was due to a decrease in the amount of active sites at high initial dye concentration in a given dose of adsorbent material [87,88]. Consequently, 120 mg/L and 40 mg/L were used in all experiments as the optimal initial concentration of CR and MB, respectively.

Influence of the Interaction Time
In the adsorptive removal of contaminants from aqueous solutions, contact time is crucial. It offers an estimate of how long it takes to reach equilibrium and also defines the kinetics of contaminant removal by nano-adsorbents. Whenever somebody defines the adsorbent's economic value and practical relevance in water treatment, contact time is essential. Figure 9d shows the adsorption percentage of CR and MB with initial concentration of 120 and 40 mg/L, respectively, as a function of contact time (0-240 min) in the presence of Fe 3 O 4 /PPy/C nanocomposite. The elimination % improved gradually with increasing contact time in case of CR dye. The same trend was observed for the MB dye. At an early stage, the catalyst surface available to interact with the dye molecules, and hence the dye removal percentage rose dramatically with the passage of time until reaching saturation/ equilibrium. After that, the dye removal does not vary much with increasing the contact duration [17,68]. This behavior is consistent with that being previously observed for the removal of various pollutants such as phenol [49], Pb(II) and Cd(II) [4], Cr (IV) and CR dye [39,89] as well as MB dye [4,90].

Equilibrium Adsorption Isotherm Studies
To understand the interaction between the adsorbed molecules, including CR and MB dye, and the sorbent surface (Fe 3 O 4 /PPy/C nanocomposite), isotherm investigations were carried out using Langmuir, Freundlich, Temkin, and Dubinin-Radushkevich (D-R) isotherms as shown in Fig. 10. The parameters of various isotherms were summarized in Table 3.
For monolayer sorption processes, the Langmuir equation is used, and the resultant Eq. 3 [91]. The Langmuir model postulates that all sorption sites are identical, so sorption occurs at discrete/homogeneous sites within the sorbent and the highest adsorption is comparable to a saturated monolayer of adsorbate molecules on the adsorbent's surface. C e and q e is the equilibrium dye concentration in mg/L and its corresponding adsorption capacity, respectively. q m is the evaluated Langmuir adsorbing capacity in mg/g and K l represents the Langmuir adsorption constant in L/mg. In the Langmuir adsorption isotherm plot, Fig. 10a, Ce/qe is drawn versus Ce to can determine the values for q m and K l . As revealed in Table 3, the Langmuir isotherm has the highest correlation coefficient value, 0.994 for CR and 0.995 for MB, when compared to other isotherm models. This proposes that the adsorption of CR and MB by Fe 3 O 4 /PPy/C nanocomposite was mostly monolayer coverage. The highest adsorption capacity (q m ) for CR was 500 mg/g and MB was 90.9 mg/g. The Langmuir model revealed the highest regression coefficient for the adsorption of CR onto magnetic poly(glycidyl methacrylate)-grafted-PPy microspheres [92]. Also, this model was the best to describe the adsorption of MB dye onto pure PPy [45] and PPy/sodium alginate [4].
The sorption type or nature is determined by estimating the separation factor (R L ) value using Eq. 4.
It is preferable adsorption when the value of R L is between 0 and 1, undesirable at for R L > 1, reversible for R L = 1, and irreversible for R L = 0 [93,94]. At initial concentration of 120 mg/L and 40 mg/L of CR and MB, respectively, the R L value was 0.028 and 0.004, indicating favorable adsorption of both dyes onto the nanocomposite surface.
Freundlich isotherm designates to multilayer sorption process. The equation of Freundlich model in its linear form is given below, Eq. 5 [95]. The Freundlich model describes the adsorbent's surface heterogeneity.
Freundlich constant is K F ((mg/g)(L/mg) 1/n ) and n (L/g) is adsorption intensity, which value provides us an idea about the affinity of the process. Their values were determined by plotting lnqe versus lnCe as shown Fig. 10b. When 1/n is less than one, the adsorption is favourable, but when 1/n is greater than one, it is unfavourable; when 1/n is equal to one, the sorption is homogeneous [96,97]. From Table 3, 1/n is less than 1 for both dyes indicating favourable adsorption. Temkin's equation has the following linear form [98], Eq. 6. This model assumes that the heat of adsorption of all molecules reduces linearly with increasing adsorption on the adsorbent's surface in addition the adsorption happens with a homogeneous distribution of binding energies until the maximum binding energy is attained [99].
R is the universal gas constant, b t related to the heat of adsorption, and K T denotes equilibrium binding constant (L/ mg) that corresponds to the maximum binding energy [100]. As shown in Fig. 10c, we can estimate the values of b T and  K T by plotting qe against lnCe. Table 3 demonstrates that the Temkin model is not the best suited model, since its R 2 value was 0.976 and 0.886 for CR and MB, respectively. The calculated value of binding energy (bT) was 30.5 J/mol for CR and 292.5 J/mol for MB. These values correspond to the physical characteristics of the sorption process. The linear form of the Dubinin-Radushkevich (D-R) isotherm is presented as follows [97], Eq. 7. It assumed that adsorption occurs on both homogeneous and heterogeneous surfaces. q m is the estimated adsorption capacity by D-R model and β is the D-R isotherm constant. ε is the Polanyi adsorption, which is calculated using Eq. 8. This model helps to evaluate mean sorption energy (E) using Eq. 9.
The type of adsorption is determined according to the mean sorption energy value, where the E is less than 8 kJ/mol suggesting that the sorption of CR and MB is a physical process [4].

Adsorption Kinetics
The reaction-limiting stage of the sorption process is determined by the adsorption kinetics. Kinetics of dye adsorption were performed using CR solution with initial concentrations ranged from 80 to 140 mg/L and MB solution separately with initial concentrations of 20-60 mg/L. The experimental results were fitted to the pseudo first order (PFO) (Eq. 10), pseudo second order (PSO) (Eq. 11), Intra-particle diffusion (IPD) (Eq. 12), and Elovich model (Eq. 13) and their equations are given as follows.
where K 1 , K 2 , and K i , is the PFO, PSO, and IPD rate constant, respectively. q e -the equilibrium adsorption capacity Elovich q t = 1 ln + 1 ln t and q t -adsorption capacity at time t. C-the boundary diffusion influence [101]. α and β are constants of the Elovich model. Figure 11 shows the kinetic models, including PFO, PSO, IPD, and Elovich, for sorption of CR and MB onto Fe 3 O 4 /PPy/C nanocomposite. Table 4 displays the values of kinetic parameters as well as rate constants. The high correlation coefficient (R 2 ) of the PSO model (0.999 for CR and 0.998 for MB) reflects that the kinetics of CR and MB adsorption on nanocomposite were better expressed by PSO model. Furthermore, the calculated adsorption cabacity from this kinetic model is very close to the experimental values. Generally, the pseudo-second-order (PSO) model is the most commonly used model for describing the sorption kinetics of pollutants [17]. The results suggest that chemisorption phenomena are involved in the whole adsorption process for CR and MB dye indicating electron exchange between adsorbate and adsorbent molecules [102]. PSO had a calculated adsorption capacity of 333.3 mg/g for CR and 83.3 mg/g for MB. The calculated k 2 of MB adsorption was higher than that of CR adsorption, indicating the faster MB adsorption onto the surface of Fe 3 O 4 /PPy/C than CR, which agrees with the obtained results. Yu et al. [92] recently reported that the PSO model best describes CR adsorption onto magnetic poly(glycidyl methacrylate)-grafted-PPy microspheres. Also, studying the kinetics of the adsorption of CR dye onto γ-Fe 2 O 3 /PPY indicated that it is a chemisorption process [35]. For the PFO model, the calculated adsorption capacities not comparable with the experimental one in addition the relatively low R 2 value that indicating the adsorption processes not obey the pseudo-first order. From the data in Table 4 for IPD model, it was found that C ≠ 0 for adsorption of both CR and MB onto Fe 3 O 4 /PPy/C nanocomposite. This revealed that intraparticle diffusion isn't the sole rate controlling step and the dye adsorption processes not fitted the IPD model. The Elovich model assumes that the active sites of a sorbent are heterogeneous. The constants of Elovich model α and β, respectively, represent the initial adsorption rate and the desorption constant. The obtained R 2 values indicated that the Elovich model didn't accurately describe the adsorption processes on the nanocomposite.

Reusability Study
The reusability of Fe 3 O 4 /PPy/C nanocomposite as an adsorbent for CR and MB was studied for individual dye solution to assess its economic efficiency. The adsorbent dosage was 0.5 g/L and the used concentration of CR and MB was 120 and 40 mg/L, respectively. After the first adsorption cycle, external magnetic field was used to collect the dye adsorbed nanocomposite. 0.1 M solution of NaOH and HCl was used to liberate CR and MB dye, respectively. Then wash with distilled water for usage in the subsequent adsorption 1 3 procedure. Four rounds of regeneration tests were carried out and the corresponding removal efficiency was estimated for each dye as presented in Fig. 9e. According to the results, after the fourth cycles, the nanocomposite still exhibited more than 85% removal efficiency. This indicates that the developed magnetic nanocomposite has good reusability and stability.

Proposed Adsorption Mechanism
FTIR analysis was conducted to understand the adsorption mechanism since it reveals the functional groups in the adsorbent material and their interaction locations. Figure [63,103]. The characteristic peaks of polypyrrole in the range of 1560 and 1350 cm −1 are obviously broadened, confirming incorporation of the pyrrole ring with the adsorbed dyes, which may be due to π-π interaction between the aromatic ring PPy and that of adsorbed dyes. Also, Bhaumik et al. [46] reported that after CR adsorption, the peaks of PPy/PANI are slightly moved towards a lower wavenumber and their intensity is lowered. In comparison to fresh nanocomposite, the FT-IR spectrum of CR or MB after adsorption over Fe 3 O 4 /PPy/C nanocomposite showed minor shifts in band locations and a decrease in intensity of the Fe-O stretching peaks 592 and 628 cm −1 [3,33]. This demonstrates involvement of Fe 3 O 4 in the dye adsorption process, which could be through surface complexation. These results confirmed our proposed adsorption mechanism for CR and MB on the surface of Fe 3 O 4 / PPy/C nanocomposite, Scheme 2.
Physical adsorption or chemical adsorption or a combination of both, generally, is the possible adsorption strategies based on the type of interaction between the adsorbent and adsorbate [104]. Adsorption mechanism of CR and MB onto the surface of Fe 3 O 4 /PPy/C nanocomposite may be attributed to a variety of interactions, such as π-π interaction, hydrogen bonding, electrostatic interaction, and surface complexation [1,3,45,57,103], as presented in Scheme 2. (i) π-π interaction may occur between the aromatic ring of CR or MB dye with that of PPy in the nanocomposite [17,57,104]. (ii) The N (nitrogen) atom of the dye (C-N group), which is highly electronegative and has lone pair of electrons, can act as a hydrogen bonding acceptor and form hydrogen bonds with the H-atom of the hydroxyl groups of the PPy/C backbone. Owing to the difference in the chemical structure of each dye, CR dye can form more hydrogen bonds than MB dye, leading to higher adsorption affinity toward CR dye. (iii) The sulphonate group in CR dye is negatively charged and thus can interact with metallic cationic moieties (ζ of Fe 3 O 4 = + 24.7), leading to electrostatic attraction between nanocomposite and CR dye [35,92]. Also, there is electrostatic attraction between the positively charged MB dye and the net negative charge on the surface of the nanocomposite (ζ of Fe 3 O 4 /PPy/C = − 20.6). (IV) Finally, surface complexation between nanocomposite and CR dye. It was reported that the task of Fe 3 O 4 for the sorption of CR in Fe 3 O 4 -based composites is that Fe 3 O 4 helps in the creation of a complex between the molecules of CR dye and the composite [103]. Also, the existence of Fe(II) assists chemisorption sites, consequently facilitating bonding of anionic dye (CR) [1,103]. In addition, Afkhami and Moosavi [105] attributed the CR adsorption onto γ-Fe 2 O 3 to the coordination interaction between metal ions and amine groups of CR molecules. Accordingly, different forces are involved in the adsorption mechanism of CR and MB onto the Fe 3 O 4 /PPy/C nanocomposite as displayed in Scheme 2.
The proposed mechanism is supported by the dye adsorption findings. (1) The experimental dye adsorption results indicated that the synthesized nanocomposite exhibited a higher adsorption affinity toward CR than MB dye. This agrees with the proposed adsorption mechanism where the higher combination interaction boosts CR adsorption more than in the case of MB dye. (2) According to adsorption kinetic studies, the PSO model exhibited the highest correlation coefficient (R 2 ) (0.999 for CR and 0.998 for MB). This indicates that chemisorption phenomena are involved in the whole adsorption process for both CR and MB dye, in agreement with the proposed adsorption mechanism. (3) The value of the PSO rate constant (k 2 ) of MB adsorption was higher than that of CR adsorption, indicating the faster MB adsorption onto the surface of Fe 3 O 4 /PPy/C than CR in accordance with adsorption results, Fig. 8. (4) The interpretation of isotherm results indicated that the adsorption of CR and MB by Fe 3 O 4 / PPy/C nanocomposite follows Langmuir isotherm owing to the highest correlation coefficient value, 0.994 for CR and 0.995 for MB. The the sorption type is determined by estimating the separation factor (R L ) value, which was 0.028 and 0.004 for CR and MB, respectively, indicating favorable adsorption of both dyes onto the nanocomposite surface. (5) The mean sorption energy (E D-R ), which determined from the Dubinin-Radushkevich (D-R) isotherm, for the sorption of CR and MB is less than 8 kJ/ mol suggesting that the adsorption takes place physically [4]. According to the zeta potential results, the surface of nanocomposite is negatively charged thus it can be predicted that the nanocomposite exhibits higher adsorption properties toward positively charged MB dye than  On the other hand, experimentaly the nancomposite showed larger adsorption capacity for anionic CR dye revealing that the electrostatic attractive forces not playing a prominent role in the adsorption process. There are many adsorption processes follwed the PSO kinetic model eventhough E D-R is less than 8 kJ/mol. For example, the adsorption of MB dye on sodium alginate/polypyrrole composite followed the Langmuir isotherm model and PSO kinetic model, however the E D-R value was 0.24 kJ/mol [4]. The authors implied that the adsorption of MB onto the adsorbent might take place via a variety of mechanisms, such as Van der Waals, hydrogen bonding, electrostatic interactions, and π-π interactions. Recently, Sarojini et al. [1] studied the adsorption on CR dye on polypyrrole/iron oxide/seaweed (PPy/Fe 3 O 4 /SW) nanocomposite. They found that the PSO model fits the kinetics of CR adsorption, the CR adsorption followed the Langmuir isotherm, and the calculated E D-R was very low, E = 0.853 kJ/mol. In addition, Ekka et al. [21] found that the Cr(VI) adsorption was well fitted with the Langmuir isotherm model and E D-R = 2.64 kJ/mol. However, PSO was the best to descripe the kinetics of adsorption, R 2 = 0.999. Similar findings were reported by other authors [22,[106][107][108][109]. To sum up, the adsorption characteristics of the prepared Fe 3 O 4 /PPy/C nanocomposite more likely be controlled by a number of different interactions cumulatively apart from electrostatic attractive forces there are hydrogen bonding, π-π interaction, and surface complexation may also contribute to the adsorption features of the nanocomposite. Table 5 compares the characteristics of the developed Fe 3 O 4 / PPy/C nanocomposite with various adsorbents in terms of saturation magnetization, specific surface area, and the adsorption capacity toward CR and/or MB dye. These adsorbents are classified into three groups as follows. The adsorbents in group A can uptake CR dye only. The adsorbents in group B can remove both CR and MB dye. For group C, these materials can remove MB only. Relative to these adsorbents, it is clear that the prepared Fe 3 O 4 /PPy/C nanocomposite has higher adsorption capacity. There are some adsorbents have comparable adsorption capacities of the nanocomposite, but the Fe 3 O 4 /PPy/C exhibited more enhanced magnetic properties compared to these adsorbents. In conclusion, the advantages of the Fe 3 O 4 /PPy/C nanocomposite in comparison with other adsorbents are large surface Scheme 2: Proposed adsorption mechanism of CR and MB on Fe 3 O 4 /PPy/C nanocomposite area, mesoporous structure, high adsorption capacity, sorption ability toward cationic and anionic dyes, as well as simple separation owing to high magnetization. Obviously, the synthesized nanocomposite has potential application for treating wastewater.

Conclusions
In this paper, the newly Fe 3 O 4 /PPy/C nanocomposite was prepared by encapsulating very fine Fe 3 O 4 nanoparticles (4-7 nm) in PPy/C polymer matrix. XRD and HRTEM proved that the inclusion of PPy/C in the iron oxide synthesis solution causes a clear decrease in the size of the grown Fe 3 O 4 nanoparticles. It is suggested that the polymer works as a surfactant or particle growth inhibitor. Magnetic analyses demonstrate that saturation magnetization (M s ) and coercivity (H c ) values are 45.201 emu/g and 54.03 Oe for Fe 3 O 4 /PPy/C nanocomposite and 73.365 emu/g and 47.4 Oe for uncoated Fe 3 O 4 , respectively. In comparison with pure Fe 3 O 4 and PPy/C, the developed Fe 3 O 4 /PPy/C nanocomposite has higher surface roughness (R a = 145.0 nm) and specific surface area (341.01 m 2 /g). The adsorption ability of the nanocomposite was evaluated using CR, MB, and CR-MB binary mixture solution. In single dye solution, the effect of different adsorption parameters such as pH, contact time, adsorbent dosage, and initial dye concentration as well as the sorption kinetics and isotherms were investigated. It was found that the Langmuir model and the pseudo-second order model well described the sorption of organic dyes onto the nanocomposite surface. The as-prepared nanocomposite had a high adsorption capacity, reaching 500 mg/g for CR and 90.9 mg/g for MB. The reusability potential of Fe 3 O 4 /PPy/C nanocomposite is also established. The proposed adsorption Table 5 Comparison of the saturation magnetization (M s ), surface area, and adsorption capacity toward CR and/or MB dye of the synthesized nanocomposite with those of numerous adsorbents mechanism for the uptake of CR and MB is based on the FTIR results after dye adsorption. The adsorption of CR on Fe 3 O 4 /PPy/C nanocomposite is due to π-π stacking, hydrogen bonding, electrostatic attraction, and surface complexation, whereas there are only three types of interaction that are involved in the MB adsorption, including π-π stacking, hydrogen bonding, and electrostatic attraction. This may explain the higher removal efficiency for anionic CR dye than MB. Considering the characteristics of Fe 3 O 4 /PPy/C, including large surface area, high rough surface, mesoporous structure, rich functional groups, and high magnetization, it may be used to efficiently remove various pollutants.
Acknowledgements Prof. Zahra Saleh, Central Laboratories Network and The Centers of Excellence, NRC, is acknowledged for her valuable support and help.
Funding Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB). This work wasn't funded by any funding agency.

Data Availability
The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

Conflict of interest
The authors declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.