Facile controllable hydrothermal route for a porous CoMn₂O₄ nanostructure: synthesis, characterization, and textile dye removal from aqueous media

Abstract

We herein report the synthesis of a pure sphere-like spinel CoMn₂O₄ nanostructure using a facile and surfactant-free hydrothermal approach followed by a thermal decomposition of the as-prepared CoCO₃/MnCO₃ composite precursor. Various factors affecting the hydrothermal reaction of cobalt chloride, manganese chloride, and ammonium hydrogen carbonate have been investigated to synthesize a pure CoCO₃/MnCO₃ composite precursor. Calcination of the CoCO₃/MnCO₃ composite (synthesized using 0.4Co²⁺ : 0.6Mn²⁺ molar ratio) at 550 °C for 1 h gave the pure sphere-like spinel CoMn₂O₄ nanostructure product (∼16 nm), but the other carbonate composites (synthesized using other molar ratios) did not generate pure spinel CoMn₂O₄ on calcination. The as-prepared products were identified employing XRD, FT-IR, TEM, EDS, FE-SEM, zeta potential, TG, and BET analyses. The as-prepared spinel CoMn₂O₄ product showed high adsorption capacity (132 mg g⁻¹) for the removal of Reactive Black 5 (RB5) dye from aqueous media. The pseudo-second-order kinetics and Langmuir isotherm models described well the experimental adsorption results. The adsorption of RB5 dye on the as-prepared adsorbent is an endothermic, spontaneous, and physisorption process according to the calculated thermodynamic constants: ΔH⁰ (22.144 kJ mol⁻¹), ΔG⁰ (from −4.321 to −6.990 kJ mol⁻¹), and E_a (20.916 kJ mol⁻¹), respectively. The as-prepared spinel CoMn₂O₄ adsorbent showed high stability, reusability, and high adsorption capacity implying its efficiency in removing the RB5 textile dye from aqueous media.

---

1. Introduction

Hazardous chemicals, in particular dyes, generated from various industries such as food, paper, leather, cosmetics, textile, and pharmaceutical cause serious problems to humanity and the environment.¹ Notably, one of the main sources of water pollution is the textile industry because of the release of textile dyes into the aquatic environment without treatment. Among these textile dyes, reactive dyes (e.g. Reactive Black 5 dye (RB5)) are dangerous dyes because of their toxicity and carcinogenicity due to the azo group, –N=N–, that these dyes have.² So, many research groups have directed much of their effort towards treatment of the wastewater before its discharge into the environment.^3,4

So far, various physical and chemical methods have been adopted for the removal of contaminants from wastewater such as reverse osmosis, filtration, precipitation, biological treatment, photodegradation, solvent extraction, chemical oxidation, membrane process, adsorption, and coagulation.^4–11 Nevertheless, due to the non-biodegradability and stability of most of the textile dyes (especially, the reactive dyes), adsorption is the most convenient technique for the removal of toxic dyes because of its simplicity, applicability, safety, and high efficiency.^3,12,13 Besides, it is still a big challenge to search for efficient and environmentally friendly adsorbents. Thence, different adsorbents such as activated carbons, sesame hulls, clay materials, and agricultural wastes, have been proposed for the removal of dyes from wastewater.^14–17 The necessity of having adsorbents with high efficiency and low-cost production stimulated researchers to direct their interest to nanotechnology.³

Therefore, several nanomaterials have been prepared and suggested as nano-adsorbents because of the high surface area, high adsorption capacity, and great number of active sites they have.^18–22 Even though Mn₂O₃ nanostructure is nontoxic and is proposed as a nano-adsorbent for the removal of some dyes, it shows low adsorption capacity.²³ Although manganese oxide composites have relatively high adsorption capacities,^4,23,24 CoMn₂O₄ nanostructure has not been investigated as an adsorbent. Besides, spinel CoMn₂O₄ nanostructure has drawn the research groups' attention due to its stability, low-cost production, environmental safety, and wide applications such as lithium-ion batteries, supercapacitors, and electro-catalysis.^25–27

In this vein, various routes have been developed for the synthesis of cobalt manganite (CoMn₂O₄) nanostructures such as precipitation, organometallic compound thermal decomposition, electrospinning, solvothermal, and spray pyrolysis method.^28–32 Despite the several advantages of the hydrothermal method including simplicity, availability of various factors to control (i.e. temperature, time, precursors, etc.), and low-cost, reports on the hydrothermal synthesis of CoMn₂O₄ are still limited. Besides, thermal decomposition of the hydrothermally synthesized metal carbonates proved its efficiency in producing various porous metal oxides attributing to the release of CO₂ gas on calcination.^3,23,33–36 Moreover, to the best of our knowledge, template-free hydrothermal synthesis of CoCO₃/MnCO₃ nanocomposite under benign conditions, subsequent production of CoMn₂O₄, and examination of the produced spinel as an adsorbent to eliminate the RB5 dye from wastewater have not been published so far.

Herein, we have developed a template-free hydrothermal synthesis of CoCO₃/MnCO₃ composite microspheres using NH₄HCO₃, CoCl₂(H₂O)₆, and MnCl₂(H₂O)₄. Various factors influencing the hydrothermal process, including reaction time and Co²⁺ : Mn²⁺ molar ratios, have been investigated. Subsequent calcination of the CoCO₃/MnCO₃ composites with various Co²⁺ : Mn²⁺ molar ratios gave interesting results, and only one molar ratio produced pure phase of spinel CoMn₂O₄ product. The adsorption study of the spinel CoMn₂O₄ nanoparticles exhibited a high adsorption capacity for the removal of Reactive Black 5 (RB5) dye (Scheme 1), as a model of a textile dye.

Scheme 1 Chemical molecular structure of Reactive Black 5 (RB5) dye.

2. Experimental

2.1. Materials and reagents

The chemicals employed in the present study: cobalt chloride (CoCl₂·6H₂O), manganese chloride (MnCl₂·4H₂O), ammonium hydrogen carbonate (NH₄HCO₃), and Reactive Black 5 dye (RB5) (C₂₆H₂₁N₅Na₄O₁₉S₆), were supplied by Sigma-Aldrich company. All other chemicals were of analytical grade and utilized as received without further purification.

2.2. Preparation of CoCO₃/MnCO₃ nanocomposite precursor

Under the optimized hydrothermal process: ammonium hydrogen carbonate (3.5 g, 44.3 mmol, 3 eq.), dissolved in distilled water (25 mL), was added to a stirring aqueous solution (35 mL) of manganese chloride (1.75 g, 8.86 mmol, 0.6 eq.) and cobalt chloride (1.40 g, 5.90 mmol, 0.4 eq.). The reaction blend was allowed to stir for 10 min. Afterward, the reaction mixture was transferred into a 100 mL-Teflon-lined autoclave, and the autoclave was then maintained in an oven at 120 °C for 3 h. The autoclave was then left to get the room temperature (25 °C) naturally, and the precipitated product was collected by centrifugation. The composite samples were washed with water and ethanol several times, and the products were dried in an oven at 60 °C overnight. The influence of Co²⁺ : Mn²⁺ molar ratio: (0.1 : 0.9), (0.2 : 0.8), (0.3 : 0.7), (0.4 : 0.6), (0.5 : 0.5), (0.6 : 0.4), (0.7 : 0.3), (0.8 : 0.2), and (0.9 : 0.1), on the hydrothermal treatment was investigated, and the samples were denoted as p19, p28, p37, p46, p11, p64, p73, p82, and p91, respectively. Additionally, the effect of the reaction time (0.5, 1, 2, and 3 h) on the hydrothermal treatment with 0.4Co²⁺ : 0.6Mn²⁺ molar ratio at 120 °C reaction temperature was studied.

2.3. Preparation of CoMn₂O₄ nanoparticles

Pure cobalt manganite (CoMn₂O₄) nanostructures were produced by calcination of the CoCO₃/MnCO₃ composite precursor (hydrothermally prepared using 0.4Co²⁺ : 0.6Mn²⁺ molar ratio) at 550 °C for 1 h. The particular 0.4Co²⁺ : 0.6Mn²⁺ molar ratio was chosen on the basis of investigation of the calcination products of the CoCO₃/MnCO₃ composite samples synthesized using various Co²⁺ : Mn²⁺ molar ratios. The calcined products were referred to as p19_550, p28_550, p37_550, p46_550, p11_550, p64_550, p73_550, p82_550, and p91_550, according to the corresponding used Co²⁺ : Mn²⁺ molar ratios and calcination temperature (550 °C).

2.4. Materials characterization

The phase structure, purity, and morphology of the as-prepared products were investigated employing X-ray diffractometer (Bruker, model D8 Advance) with Cu-Kα radiation (λ = 1.54178 Å), a field emission scanning electron microscope (FE-SEM; JEOL, model JSM-6390), and a high-resolution transmission electron microscope (HR-TEM; model JEM-2100) with an accelerating voltage of 200 kV equipped with energy-dispersive X-ray spectrometer (EDS spectrometer). The chemical compositions of the products were further examined using FT-IR spectra measured on an FT-IR spectrometer (Thermo Scientific, model Nicolet iS10) in the frequency range of 4000–400 cm⁻¹. The adsorption studies were performed using a UV-visible spectrophotometer (Jasco, model v670). Thermal stability of the as-synthesized CoCO₃/MnCO₃ composite was carried out using a thermal analyzer instrument (Shimadzu; model TA-60WS) under nitrogen gas atmosphere at 15 °C min⁻¹ heating rate. The BET (Brunauer–Emmet–Teller) surface area and pore size of the as-produced CoMn₂O₄ nanoparticles were estimated using N₂ adsorption isotherms on Quantachrome analyzer (Nova 2000 series, USA) at 77 K. The isoelectric point, IEP, of the as-prepared CoMn₂O₄ nanoparticles, was determined by means of Zeta-sizer nano series meter (Nano ZS, Malvern, UK) and the zeta potential measurements were performed in NaCl solutions (0.01 M) with different pH values (2–10). The UV-Vis diffuse reflectance spectrum of the spinel CoMn₂O₄ was measured using UV-visible spectrophotometer (Jasco, model v670) equipped with an integral sphere (Jasco, model ISN-723). The chemical stability of the as-synthesized CoMn₂O₄ nanoparticles was examined by determining the concentration of the released cobalt and manganese ions in solutions with pH 1.5 from CoMn₂O₄ for 24 h, using an inductively coupled plasma-optical emission spectrometer (ICP-OES; Optima 7000 DV, Perkin-Elmer, USA).

2.5. Adsorption studies

To study the adsorption properties of the as-produced CoMn₂O₄ nanoparticles, different batch experiments were performed in the dark using Reactive Black 5 (RB5) dye as an example for a textile dye. Typically, in an Erlenmeyer flask, 0.05 g of CoMn₂O₄ nano-adsorbent was added to 25 mL of Reactive Black 5 dye (RB5) with an initial concentration of 50 mg L⁻¹ at specific pH (adjusted using HCl and/or NaOH solutions (0.1 M)) and constant temperature. And the suspension was stirred (400 rpm) for a pre-defined time (t). At pre-determined intervals (t), an aliquot was withdrawn and centrifuged to separate the suspension. The concentration of the dye (C_t) in the supernatant was estimated by measuring its absorbance at λ_max = 598 nm and employing a pre-constructed calibration graph for the RB5 dye. The adsorption capacity (q_t, mg g⁻¹) of dry CoMn₂O₄ nanoparticles for RB5 dye at t time and the RB5 dye removal efficiency (%R) were estimated employing eqn (1) and (2), respectively.where, C₀ is the initial dye concentration (mg L⁻¹), C_t is the remained dye concentration in the supernatant (mg L⁻¹), m is the weight of the dry adsorbent (g), and V is the volume of the dye solution (L). Effects of different factors affecting the adsorption of RB5 dye on the nano-adsorbent of interest have been studied such as pH (1–10), contact time (0–60 min), KCl concentration (0.05–0.55 g), temperature (298–328 K) and initial RB5 dye concentration (50–600 mg L⁻¹). The equilibrium adsorption capacity, q_e (mg g⁻¹), was calculated using eqn (3).where, C_e is the dye concentration remained at equilibrium in the supernatant solution (mg L⁻¹). The remaining symbols: V, m, and C₀, have the same aforementioned meaning.

3. Results and discussion

3.1. Hydrothermal synthesis and characterization of CoCO₃/MnCO₃ composite microspheres

3.1.1. XRD analysis. In our earlier study, the hydrothermally prepared metal carbonates were found to be efficient for the preparation of nanostructures of various simple metal oxides which were investigated as adsorbents for water treatment.^3,23,33 Therefore, we have developed, in the present study, a two-step synthetic method for the synthesis of pure spinel CoMn₂O₄ nanostructure. Firstly, we have hydrothermally prepared pure CoCO₃/MnCO₃ composite via a benign hydrothermal treatment of cobalt chloride, manganese chloride, and ammonium hydrogen carbonate without any other additives. This was performed by studying the factors affecting the hydrothermal reaction to reach the optimum hydrothermal conditions. Secondly, thermal decomposition of the as-prepared carbonate composite precursor at 550 °C for 1 h produced pure spinel CoMn₂O₄ product. The as-produced spinel oxide was then examined as an adsorbent for the removal of a textile dye, Reactive Black 5 (RB5) dye, as it will be shortly described.

Fig. 1(a) displays the XRD pattern of the pure CoCO₃/MnCO₃ composite precursor (p46) hydrothermally synthesized under the optimum conditions: 0.4Co²⁺ : 0.6Mn²⁺ : 3HCO₃⁻ molar ratio, at 120 °C reaction temperature, and for 1 h reaction time. All the diffraction peaks can be indexed well to a mixture of cobalt carbonate of a rhombohedral phase which is consistent with the standard XRD reflections of pure CoCO₃ (space group R3̄c, JCPDS card no. 78-0209),^36,37 and manganese carbonate with a hexagonal phase which agrees well with the standard XRD pattern of pure MnCO₃ (space group R3̄c, JCPDS card no. 44-1472).^23,38 Notably, other impurities have not been observed. By applying the Debye–Scherrer equation (eqn (4)), the average crystallite size (D, nm) of the as-synthesized CoCO₃/MnCO₃ composite nanostructure was found to be 32.5 nm:³⁹

D = 0.9λ/β cos θ_B (4)

where, D is the crystallite size (nm), λ is the X-ray radiation wavelength (nm), β is the XRD line full width at half maximum (FWHM) of the diffraction lines, and θ_B is the Bragg diffraction angle. Fig. 1(b) manifests the XRD pattern of the as-produced CoMn₂O₄ nanostructure. All the reflections indicate the complete transformation of the carbonate composite precursor to pure spinel CoMn₂O₄ because the XRD pattern of the product corresponds well to the standard XRD pattern of the tetragonal spinel (Co, Mn)₃O₄ (space group R3̄c, JCPDS card no. 18-0408).^40,41 The calculated crystallite size of the as-prepared spinel CoMn₂O₄ using the Debye–Scherrer equation was 15.3 nm.

Fig. 1 XRD patterns of the as-synthesized CoCO₃/MnCO₃ composite precursor (a) and spinel CoMn₂O₄ product (b).

3.1.1.1. Optimization of the hydrothermal synthesis of CoCO₃/MnCO₃ composite precursor. A facile hydrothermal synthesis of CoCO₃/MnCO₃ composite microspheres was achieved by studying the various factors influencing the hydrothermal reactions of cobalt chloride, manganese chloride, and ammonium hydrogen carbonate such as Co²⁺ : Mn²⁺ molar ratio and reaction time. The hydrothermal treatment was first carried out at 120 °C for 3 h while the concentration of ammonium hydrogen carbonate was remained constant (3 equivalents), and various Co²⁺ : Mn²⁺ molar ratios (0.1 : 0.9, 0.2 : 0.8, 0.3 : 0.7, 0.4 : 0.6, 0.5 : 0.5, 0.6 : 0.4, 0.7 : 0.3, 0.8 : 0.2, and 0.9 : 0.1) were used. It is noteworthy that the overall molar ratio of the metal cations remained constant (1 equivalent) so that the molar ratio of the metal cations (i.e. molar ratio of Co²⁺ + molar ratio of Mn²⁺) to HCO₃⁻ was 1 : 3, respectively. Fig. 2 shows the XRD patterns of the composite precursors: p19, p28, p37, p46, p11, p64, p73, p82, and p91, for the corresponding employed molar ratios, respectively. Interestingly, inspection of the XRD patterns of Fig. 2 exhibits the formation of pure CoCO₃/MnCO₃ composite irrespective of the used molar ratio. However, Co²⁺ : Mn²⁺ molar ratio of 0.4 : 0.6 was selected as the optimum molar ratio because it produced a pure carbonate precursor (p46) which generated pure spinel CoMn₂O₄ product on calcination as it will be explained later. Afterward, we investigated the effect of reaction time (0.5–3 h) on the hydrothermal products, and the XRD patterns of the products were displayed in Fig. 3. The results revealed that the time factor influenced the crystallinity of the composite product. The optimum reaction time for the hydrothermal treatment of interest was found to be 1 h because this time (1 h) was enough to produce a pure composite product with a moderate crystallinity while 0.5 h reaction time gave poor crystalline carbonate composite product. Hence, the hydrothermal treatment of cobalt chloride, manganese chloride, and ammonium hydrogen carbonate with a molar ratio of 0.4 : 0.6 : 3, respectively, for 1 h at 120 °C produced pure CoCO₃/MnCO₃ composite precursor (p46).

Fig. 2 XRD patterns of the as-prepared CoCO₃/MnCO₃ composite precursors: p19 (a), p28 (b), p37 (c), p46 (d), p11 (e), p64 (f), p73 (g), p82 (h), and p91 (i), under hydrothermal conditions with various Co²⁺ : Mn²⁺ molar ratios: 0.1 : 0.9, 0.2 : 0.8, 0.3 : 0.7, 0.4 : 0.6, 0.5 : 0.5, 0.6 : 0.4, 0.7 : 0.3, 0.8 : 0.2, and 0.9 : 0.1, respectively, at 120 °C for 3 h.

Fig. 3 XRD patterns of the as-prepared CoCO₃/MnCO₃ composites under hydrothermal conditions: Co²⁺ : Mn²⁺ : HCO₃⁻ molar ratio of 0.4 : 0.6 : 3, respectively, at 120 °C, for reaction times of 0.5 (a), 1 (b), 2 (c), and 3 h (d).

3.1.2. Morphology investigation. The morphology of the as-synthesized CoCO₃/MnCO₃ precursor (p46) was examined using a field emission scanning electron microscope as shown in Fig. 4(a and b). The low-magnification FE-SEM image (Fig. 4(a)) reveals that the CoCO₃/MnCO₃ precursor is composed of two kinds of microspheres with average diameters of ca. 0.5 and 2.5 μm. Besides, inspection of the high-magnification FE-SEM image (Fig. 4(b)) of the composite exhibits that the composite microspheres consist of aggregates of sphere-like nanoparticles with an average diameter of ca. 15 nm indicating the formation of a porous structure precursor.

Fig. 4 FE-SEM micrographs of the as-prepared CoCO₃/MnCO₃ composite precursor (a and b), and spinel CoMn₂O₄ product, p46_550 (c and d); where (a) and (c) are low-magnification images, as well as (b) and (d) are high-magnification images.

3.1.3. FT-IR investigation. The chemical structure of the as-prepared CoCO₃/MnCO₃ composite precursor (p46) was further examined using the FT-IR spectroscopy. The FT-IR spectrum of the carbonate composite, Fig. 5(a), shows vibrational peaks which confirm that the composite precursor p46 is composed of a mixture of CoCO₃ and MnCO₃ compounds. In this connection, vibrational peaks appeared at 732, 862, 1399, and 2484 cm⁻¹ correspond to CoCO₃, and this is consistent with the published data.³⁶ Additionally, Fig. 5(a) also shows vibrational absorptions at 732, 862, 1399, and 1792 cm⁻¹ attributing to the presence of MnCO₃ compound in the composite precursor (p46), and this is in good agreement with the reported data.²³ The FT-IR spectrum (Fig. 5(a)) also reveals two vibrational peaks appeared at 1599 and 3391 cm⁻¹ correspond to the bending and stretching frequencies of the adsorbed water molecules, respectively.^42,43

Fig. 5 FT-IR spectra of the as-synthesized CoCO₃/MnCO₃ composite precursor (p46) (a), CoMn₂O₄ (p46_550) (b), and RB5 dye-loaded CoMn₂O₄ (c).

3.1.4. Thermal behavior investigation. Thermal behavior of the as-synthesized CoCO₃/MnCO₃ composite precursor was investigated by means of the TG technique, Fig. 6. To further prove the chemical structure of the cobalt–manganese carbonate precursor (p46), the thermogravimetric profile (TG curve) of the CoCO₃/MnCO₃ composite product (p46), shown in Fig. 6, revealed two distinct weight loss steps in the temperature range of 60–130 °C and 130–600 °C, respectively. The first weight loss stage of 1.29% (calcd 1.27%) can be attributed to the elimination of the physically adsorbed water molecules (0.25 mole of H₂O per one mole of the composite precursor). The second weight loss step of 33.05% (calcd 32.84%) can be assigned to the decomposition of the carbonate precursor into CO₂ and CO gasses (two moles of CO₂ and one mole of CO per one mole of the composite precursor) leaving CoMn₂O₄ as a residue with a weight percent of 65.66% (calcd 65.89%) which is compatible with the XRD results.

Fig. 6 TG analysis of the as-synthesized CoCO₃/MnCO₃ composite precursor under N₂ gas.

3.2. Preparation and characterization of CoMn₂O₄ nanostructure

Based on the thermal analysis results, the as-prepared carbonate composite precursors were calcinated at 550 °C for 1 h and the XRD patterns of the products were displayed in Fig. 7(a–i). Interestingly, the XRD results exhibited that the CoCO₃/MnCO₃ composite synthesized using 0.4Co²⁺ : 0.6Mn²⁺ molar ratio was the only sample (p46) which produced pure spinel CoMn₂O₄ product (p46_550) on calcination, Fig. 7(d). On the other hand, calcination of the carbonate composite samples: p19, p28, and p37, synthesized using higher molar ratios of manganese gave CoMn₂O₄ and CoMnO₃ mixture as presented in Fig. 7(a–c), respectively. Furthermore, calcination of the samples prepared using higher molar ratios of cobalt (p55, p64, p73, and p82) produced only Co₃O₄ on calcination (Fig. 7). However, the manganese oxide phase has not been observed in the calcined products by the XRD instrument. This may be due to the amorphous nature of the products which cannot be detected with the XRD, and this is one of the intrinsic limitations of the XRD technique.^44,45 Additionally, the morphology of the as-prepared spinel CoMn₂O₄ product (p46_550) was investigated, using FE-SEM and TEM, and presented in Fig. 4 and 8, respectively. Inspection of the low-magnification FE-SEM image (Fig. 4(c)) of the oxide product showed that the spherical morphology of the carbonate precursor remained almost unchanged on calcination, and the carbonate precursor generated micro-spherical spinel CoMn₂O₄ particles with 1.4 μm diameter. However, the spinel CoMn₂O₄ product is more porous, and the spherical surfaces of the product particles contain more voids with larger volumes compared to the carbonate precursor as shown in the high-magnification FE-SEM image (Fig. 4(d)). This can be attributed to the release of the CO₂ and CO gasses on calcination of the CoCO₃/MnCO₃ composite precursor. The high-magnification FE-SEM image (Fig. 4(d)) also displays that the spinel CoMn₂O₄ microspheres are composed of spherical and peanut-like particles. Additionally, TEM image of the as-prepared spinel CoMn₂O₄ reveals that the product consists of spherical-like agglomerates of irregular, square, and spherical particles with an average size of ca. 16.1 nm, as displayed in Fig. 8(a), and this is consistent with the XRD results. Interestingly, close inspection of the TEM image (Fig. 8(b)) of the as-synthesized spinel CoMn₂O₄ product manifested the porous structure of the product. Besides, Fig. 8(c) shows the EDS analysis results which confirm that the as-synthesized spinel CoMn₂O₄ is composed solely of cobalt, manganese, and oxygen elements. However, Fig. 8(c) also exhibited peaks due to the presence of carbon and copper elements in the EDS spectrum and this owing to that the sample was placed on a carbon coated copper grid for TEM and EDS analyses.

Fig. 7 XRD patterns of the calcination products: p19_550 (a), p28_550 (b), p37_550 (c), p46_550 (d), p11_550 (e), p64_550 (f), p73_550 (g), p82_550 (h), and p91_550 (i), of the corresponding CoCO₃/MnCO₃ composite precursors at 550 °C for 1 h.

Fig. 8 TEM images (a and b), and EDS spectrum (c) of the as-prepared spinel CoMn₂O₄ product, p46_550.

Additionally, Fig. 5(b) displays the FT-IR spectrum of the as-prepared spinel CoMn₂O₄. The FT-IR spectrum, Fig. 5(b), shows two distinct absorption bands at 502 and 613 cm⁻¹ confirming the spinel structure of the product. The first absorption peak can be assigned to the stretching vibration of Co²⁺–O and Mn²⁺–O at the tetrahedral sites, and the second one can be attributed to the stretching vibration of Mn³⁺–O at the octahedral sites. These results are consistent with the reported data.^46–48 The absorption band appeared at 401 cm⁻¹ can be ascribed to the vibration of Co–O–Mn.⁴⁶ Finally, Fig. 5(b) also shows two bands at 1616 and 3385 cm⁻¹ corresponding to the bending and stretching vibrations of the adsorbed water molecules on the spinel nanoparticles, respectively.⁴⁹ Furthermore, by using the Kubelka–Munk function (eqn (5)), the optical band gap energy (E_g) of the spinel product can be estimated employing the UV-Vis diffuse reflective spectrum (Fig. 9(a)).^50,51 Fig. 9(a) exhibits three absorption bands at 318, 616, and 700 nm corresponding to surface plasmons excitation in the spinel product. Also, an absorption band appeared at ca. 868 nm which might be attributed to the coupling between the plasmon modes of the neighboring nanoparticles. These results are compatible with the published data.⁴⁶

where, R corresponds to the percentage reflectance. The direct band gap energy of the spinel CoMn₂O₄ product is estimated by plotting [F(R) × hν]² against hν (Fig. 9(b)) where hν is the energy of the incident photon in eV. And, from the extrapolation of the linear section of the plot at [F(R) × hν]² = 0, the band gap energy value (E_g) of the as-prepared spinel CoMn₂O₄ product was found to be 1.3 eV which was slightly lower than the reported one.⁴⁶ This might be due to the larger crystallite size of the as-prepared CoMn₂O₄ product in the current research.

Fig. 9 Diffuse reflectance spectrum (a), and plot [F(R) × hν]² against hν (b) of the as-prepared CoMn₂O₄ product.

For further characterization of the as-prepared spinel CoMn₂O₄ product, its specific surface and porosity were estimated using the BET method. The N₂ adsorption–desorption isotherm of the spinel CoMn₂O₄ product (Fig. 10(a)) revealed a hysteresis loop of the type-IV isotherm referring to the mesoporosity (i.e. 2–50 nm pore size) characteristics of the product.⁵² Besides, the CoMn₂O₄ spinel BET surface area was estimated to be 110 m² g⁻¹ and its BJH (Barret–Joyner–Halenda) pore volume was found to be 0.222 cm³ g⁻¹. Fig. 10(b) reveals the distribution of the pore size, and it is clear that the pore volume of the product is mostly filled with one group of pores with a diameter of ca. 3.1 nm.

Fig. 10 N₂-adsorption–desorption isotherm (BET) (a) and BJH pore size distribution (b) of the as-prepared spinel CoMn₂O₄ product, p46_550.

To gain more information about the as-prepared spinel CoMn₂O₄ product for subsequent adsorption application, determination of its pH_pzc (point of zero charge) and IEP (isoelectric point) were essential. The pH_pzc value of the spinel CoMn₂O₄ product was estimated using the pH drift method.⁵³ Hence, suspensions of CoMn₂O₄ in 0.01 M NaCl solutions of various initial pH values (pH_initial) in the range of 2–10 were employed, as explained in details elsewhere.³³ The values of the pH_initial were drawn against the values of the final pH (pH_final) to obtain a pH_initial–pH_final curve, and the intersection of the obtained curve with the pH_initial = pH_final line marked the pH_pzc value of the spinel CoMn₂O₄ product, as presented in Fig. 11(a). Accordingly, the pH_pzc value of the spinel CoMn₂O₄ product was estimated to be 4. Furthermore, the IEP value of the as-prepared CoMn₂O₄ was determined using a zeta-potential analyzer at 25 °C. Thus, portions of CoMn₂O₄ product suspended in 0.01 M NaCl solutions of different initial pH values (2–10) were used in this process as explained extensively elsewhere.³³ Moreover, by employing the zeta potential–pH_initial plot (Fig. 11(b)), the isoelectric point (IEP) of the product could be determined, and it was found to be 4.2.

Fig. 11 The pH_initial against pH_final graph for pH_pzc determination (a), the zeta potential versus pH curve for IEP estimation (b) for CoMn₂O₄ product (p46_550), and the effect of initial pH on the RB5 dye removal percentage using CoMn₂O₄ adsorbent (c).

3.3. Adsorption properties of CoMn₂O₄ nanostructure

The adsorption efficiency of the as-synthesized CoMn₂O₄ nano-adsorbent for the removal of RB5 dye from aqueous media was investigated. In this regard, it was reported by Nassar and others that the adsorption of dyes by nano-adsorbents could be confirmed by employing the FT-IR technique.^23,33 Fig. 5(b and c) depicts the FT-IR spectra of the bare CoMn₂O₄ nano-adsorbent and RB5 dye-loaded CoMn₂O₄ (i.e. CoMn₂O₄ after adsorption), respectively. Inspection of Fig. 5(b and c) shows the differences clearly between the two FT-IR spectra. The vibrational absorptions observed at 401 and 613 for the bare CoMn₂O₄ nano-adsorbent were shifted to higher frequencies, 423 and 619 cm⁻¹, respectively. On the contrary, the frequencies of the bare nano-adsorbent appeared at 502, 1616, and 3385 cm⁻¹ were shifted to lower vibrational absorptions: 455, 1610, and 3293 cm⁻¹, respectively. The bands observed at 1610 and 3293 cm⁻¹ looked stronger and broader due to the contribution from the vibrations of the OH groups of the adsorbed dye and water molecules, confirming the adsorption process. Furthermore, some new vibrational frequencies appeared at 882, 1002, 1179, and 1364 cm⁻¹ (Fig. 5(c)) also prove the adsorption of the RB5 dye on the nano-adsorbent.^23,33

3.3.1. pH influence. pH of the adsorption media is one of the essential parameters which can influence both the adsorbent and the dye; hence, it may affect the adsorption efficiency of the adsorbent under study. This may be attributed to the effect of pH on the surface charges of the nanoparticles as well as the degree of dissociation and speciation of the dye. Consequently, the pH value affects the kind of interaction between the adsorbent and the adsorbate molecules. Hence, the effect of the initial pH (1–10) of the adsorption media on the adsorption of RB5 dye with an initial concentration of 50 mg L⁻¹ on a fixed amount (0.05 g) of the CoMn₂O₄ nano-adsorbent has been examined at 25 °C for 24 h, as shown in Fig. 11(c). It could be clearly seen that the removal efficiency of the RB5 dye increased from 94% to 95.1% when pH changed from 1 to 1.5. Afterward, the removal efficiency reduced slowly with increasing the pH from 1.5 to 4 then decreased sharply after pH 4. It is reported that metal oxide aqueous dispersions have an outermost layer of hydroxyl groups (M–OH) attributing to the reaction of the surfaces of the metal oxide particles with water molecules.⁵⁴ The surface charges of metal oxide particles depend on the solution pH. At pH < pH_pzc, the particle will be positively charged (M–OH₂⁺) because of the presence of high concentration of H⁺ ions. And at higher pH values (i.e. pH > pH_pzc), the surface OH groups will undergo de-protonation which results in negatively charged particles (M–O⁻).⁵⁴ Hence, the surface charge of the adsorbent particles is controlled by the pH value of the adsorption media and the pH_pzc value of the adsorbent. Moreover, it is known that the RB5 dye has low pK_a value owing to the sulfonic groups that its chemical structure has; consequently, the RB5 dye molecules will be ionized at low pH values. Based on this, the obtained adsorption behavior of the RB5 dye (Fig. 11(c)) can be explained. The removal efficiency reached the maximum at pH 1.5, and this may be due to the strong electrostatic attractions constructed between the negatively charged dye ions (generated from the ionization of the dye at this pH value) and the positively charged CoMn₂O₄ particles because of the high concentration of protons. Additionally, this pH value is lower than the pH_pzc value of the CoMn₂O₄ nano-adsorbent. As the pH increases till pH 4, the dye removal percentage decreases slowly. This may be attributed to the reduction in the accumulated positive charges on the surfaces of the CoMn₂O₄ particles; hence, the electrostatic attractions between the positively charged nano-adsorbent particles and the negatively charged dye molecules will be weaker which will result in lower adsorption capacity. On the contrary, when the pH value increases above ca. 4 (i.e. pH > pH_pzc), the CoMn₂O₄ particles will be charged with negative charges. This will result in an electrostatic repulsion between the negatively charged nanoparticles and the anionic dye molecules, and this will produce very low dye removal percentage. Similar behavior was also reported for the adsorption of RB5 dye on other nano-adsorbents.^3,23 However, since the dye removal percentage did not drop to the zero value and it remained constant with low percent value at higher pH values (i.e. pH > 4), it could be concluded that the adsorption was not only controlled by the electrostatic mechanism but also controlled by other mechanisms such as hydrophobic–hydrophobic and hydrogen bonding.^54–56

3.3.2. Initial dye concentration influence. At pH 1.5 (the optimum pH value) and 25 °C temperature, the effect of the initial RB5 dye concentration was studied in the range of 50–600 mg L⁻¹ in the presence of 0.05 g CoMn₂O₄ adsorbent. Fig. 12(a) reveals the effect of the initial RB5 dye concentration on the adsorption capacity of the CoMn₂O₄ nano-adsorbent. It is observed that increasing the initial RB5 dye concentration from 50 to 400 mg L⁻¹ causes the enhancement of the adsorption capacity of the CoMn₂O₄ nano-adsorbent from 23 to 132 mg g⁻¹, respectively. Thus, this value (i.e. the maximum adsorption capacity, q_m,exp) has remained constant with increasing the initial concentration over 400 mg L⁻¹ and the plateau behavior is attained. This behavior can be explained based on that the driving force (required to overcome the adsorbate mass transfer resistance which exists between the liquid and solid phases⁵⁷) strengthens with increasing the dye concentration until it reaches maximum at C₀ = 400 mg L⁻¹. And at this point, the number of the active sites of the adsorbent particles has been completely occupied with the adsorbed dye molecules so the plateau behavior is reached.

Fig. 12 Effects of the initial concentration of the RB5 dye (a) and ionic strength (b) on the adsorption of RB5 dye on CoMn₂O₄ adsorbent.

3.3.3. Ionic strength influence. Textile dye industries charge large quantities of salts during the dyeing process so that it is important to study the effect of the presence of KCl salt on the adsorption process. Fig. 12(b) shows the influence of the presence of KCl in the range of 0.05–0.6 g on the adsorption of RB5 dye (25 mL) with an initial concentration of 50 mg L⁻¹ using 0.05 g of CoMn₂O₄ adsorbent at pH 1.5 and 25 °C temperature. The results exhibit that increasing the ionic strength (i.e. increasing the KCl concentration) decreases the removal efficiency of the RB5 dye. This trend can be attributed to the competition between the Cl⁻ anions and the negatively charged dye molecules; hence, blocking some of the adsorption active sites of the CoMn₂O₄ nano-adsorbent with some Cl⁻ anions which results in a decrease in the RB5 dye adsorption.⁵⁸

3.3.4. Contact time influence and adsorption kinetics investigation. To investigate the influence of the contact time on the adsorption of the RB5 dye on the CoMn₂O₄ nano-adsorbent, adsorption experiments were carried out under the adsorption conditions: 0.05 g portions of CoMn₂O₄ adsorbent, initial RB5 dye concentrations in the range of 50–300 mg L⁻¹ at 25 °C, pH 1.5 with contact time in the range of 0–60 min. The obtained results are displayed in Fig. 13(a). The results revealed that the adsorption of the RB5 dye was fast in the initial stages; then, it was enhanced slowly in the vicinity of the equilibrium stage. The high rate of adsorption at the beginning of the adsorption process is attributed to the availability of the active sites for adsorption and as time elapses the number of the active sites available for adsorption decreases. Besides, Fig. 13(a) revealed that the time taken by RB5 dye with an initial RB5 dye concentration of 50 mg L⁻¹ to attain the equilibrium was 15 min; however, for all other initial dye concentrations the equilibration time was 20 min.

Fig. 13 Influence of time (a), pseudo-second-order model (b), and intra-particle diffusion model (c) for the adsorption of RB5 dye on CoMn₂O₄ adsorbent.

Furthermore, the kinetics and mechanism of the adsorption of the RB5 dye on the CoMn₂O₄ nano-adsorbent were investigated by applying the linearized forms of the pseudo-first-order (eqn (6)), pseudo-second-order (eqn (7)), and intra-particle diffusion (eqn (8)) models to simulate the adsorption process.^3,23,33,59

q_t = k_it^0.5 + C (8)

where, q_e (mg g⁻¹) is the adsorption capacity of the CoMn₂O₄ nano-adsorbent at equilibrium, q_t (mg g⁻¹) is the adsorption capacity of the nano-adsorbent at t time (min), k₁ (min⁻¹) is the pseudo-first-order constant of the adsorption process, k₂ (g mg⁻¹ min⁻¹) is the pseudo-second-order of the adsorption process, k_i (mg g⁻¹ min^−0.5) is the intra-particle diffusion rate constant, and C (mg g⁻¹) is the intercept for eqn (8) which gives an indication of the boundary layer thickness. The values of k₁ and q_e(cal) in (eqn (6)) were evaluated from plotting of log(q_e − q_t) against t (not presented), values of k₂ and q_e(cal) were estimated employing the plot of t/q_e versus t (Fig. 13(b)), and finally the values of k_i and C were determined from the plot of q_t against t^0.5 (Fig. 13(c)). The constants of the three kinetic models are listed in Table 1.

Table 1 Adsorption kinetics constants for the adsorption of RB5 dye on CoMn₂O₄ adsorbent at different initial dye concentrations

C0, mg L^−1	qe(exp), mg g^−1	Pseudo-first-order model			Pseudo-second-order model				Intra-particle diffusion model
		qe(cal), mg g^−1	k1, min^−1	r^2	qe(cal), mg g^−1	k2, g mg^−1 min^−1	r^2	h, g mg^−1 min^−1	ki1 g mg^−1 min^−0.5	ki2 g mg^−1 min^−0.5
50	23	7.14	0.094	0.632	23.97	0.0169	0.996	9.72	6.42	0.104
100	49.5	0.80	0.032	0.019	51.95	0.0091	0.998	24.61	8.22	0.001
150	74	7.71	0.125	0.402	77.03	0.0069	0.998	41.19	11.87	0.017
200	99	13.03	0.140	0.476	102.77	0.0056	0.999	59.66	14.09	0.002
300	126.3	30.16	0.131	0.433	130.19	0.0056	0.999	94.17	20.07	0.059

---

The values of the correlation coefficients (r²), evaluated from the pseudo-first-order kinetic model for the initial RB5 dye concentrations: 50, 100, 150, 200, and 300 mg L⁻¹, were found to be 0.63, 0.019, 0.402, 0.476, and 0.433, respectively. On the other hand, the correlation coefficients (r²) determined by applying the pseudo-second-order kinetic model were in the range from 0.995 to 0.999 for the initial concentrations of C₀ = 50–300 mg L⁻¹, respectively. Based on this, it could be concluded that the adsorption of RB5 dye on the CoMn₂O₄ adsorbent followed the pseudo-second-order kinetic model. Besides, as listed in Table 1, the equilibrium adsorption capacities (q_e(cal)) calculated using the pseudo-second-order kinetic for the different used initial dye concentrations are consistent with the experimental values (q_e(exp)) which further confirm that the adsorption process of interest follows the pseudo-second-order kinetic model. Moreover, feeding the pseudo-second-order constant into eqn (9), one can determine the rate of the initial sorption (h).⁶⁰ The values of the initial adsorption rate at different initial dye concentrations, as provided in Table 1, show that increasing the initial RB5 dye concentrations results in increasing the rates of the initial adsorption.

h = k₂q_e² (9)

Moreover, the intra-particle diffusion model (eqn (8)) can be used to know deeply about the adsorption mechanism, and whether the rate determining step of the adsorption of RB5 dye on the CoMn₂O₄ nano-adsorbent is a sole diffusion mechanism or not. Close inspection of Fig. 13(c), reveals that all curves are multi-linear graphs and the first-stage linear parts of the curves do not pass the origin regardless of the used initial dye concentration. Consequently, it can be concluded that the adsorption mechanism and hence the rate-determining step is not only controlled by the intra-particle diffusion mechanism but also controlled by other adsorption mechanisms including film diffusion and bulk diffusion mechanisms.^19,61

3.3.5. Adsorption isotherm investigation. To further understand the adsorption mechanism, including the interaction between the adsorbate molecules and the CoMn₂O₄ adsorbent at equilibrium, some batch experiments were carried out employing a fixed weight of CoMn₂O₄ adsorbent (0.05 g) and various initial RB5 dye concentrations (50–600 mg L⁻¹) at pH 1.5 and 25 °C temperature for 20 min equilibrium time. The gained adsorption results were examined using two well-recognized isotherm models: Langmuir isotherm equation (i.e. homogeneous surface for adsorption, eqn (10)) and Freundlich isotherm equation (i.e. heterogeneous surface for adsorption, eqn (11)).where, C_e (mg L⁻¹) is the RB5 dye equilibrium concentration, q_e (mg g⁻¹) is CoMn₂O₄ equilibrium adsorption capacity, q_m (mg g⁻¹) is the CoMn₂O₄ maximum adsorption capacity (i.e. the quantity of the adsorbate required to form a monolayer coverage), K_L (L mg⁻¹) is the binding constant of the Langmuir equation, n is the Freundlich constant indicating intensity of the adsorption, and K_F is a Freundlich constant referring to the adsorption capacity. The experimental adsorption data are fitted to the Langmuir and Freundlich isotherms as displayed in Fig. 14, and the calculated isotherm parameters are listed in Table 2. Based on the Langmuir and Freundlich isotherm correlation coefficients (Table 2), it can be observed that the adsorption of RB5 dye on the CoMn₂O₄ nano-adsorbent is better described by the Langmuir isotherm model implying the homogeneous nature of the CoMn₂O₄ nano-adsorbent surface and the monolayer coverage of the dye molecules on the nano-adsorbent.⁶² Although the maximum adsorption capacity (q_m) can be determined usually using the Langmuir model, this value can be also estimated according to eqn (12).⁶³

Fig. 14 Langmuir isotherm plot (a) and Freundlich isotherm plot (b) for the adsorption of RB5 dye on CoMn₂O₄ adsorbent.

Table 2 Langmuir and Freundlich isothermal parameters for the adsorption of RB5 dye on CoMn₂O₄ adsorbent

Adsorption isotherm model	Constants	Value
Langmuir	KL (L mg^−1)	0.199
	qm(cal) (mg g^−1)	134.8
	r1^2	0.998
	RL	0.0099–0.091
	qe(exp) (mg g^−1)	132
Freundlich	KF [(mg g^−1) (L mg^−1)^1/n]	50.42
	qm(cal) (mg g^−1)	146.7
	r2^2	0.451
	n	5.6
	qe(exp) (mg g^−1)	132

---

However, the CoMn₂O₄ maximum adsorption capacity (q_m,cal = 134.8 mg g⁻¹) calculated using the Langmuir model is more consistent with the experimental value (q_e,exp = 132 mg g⁻¹) in comparison to the one determined from the Freundlich isotherm model (q_m,cal = 146.7 mg g⁻¹ for C₀ = 400 mg L⁻¹). This also further supports the suitability of the Langmuir model to describe the adsorption of RB5 dye on CoMn₂O₄ nano-adsorbent.

Furthermore, the favorability of the RB5 dye adsorption on CoMn₂O₄ adsorbent can be predicted utilizing a dimensionless constant (R_L, eqn (13)) obtained from the Langmuir isotherm model.

where, C₀ (mg L⁻¹) is the initial concentration of the RB5 dye, and K_L (L mg⁻¹) is the Langmuir constant. On the basis of the R_L values, one can predict whether the adsorption process is favorable or not. If the R_L value obeys the following: R_L = 0, 0 < R_L < 1, R_L = 1, or R_L > 1, then the adsorption process of the RB5 dye on the CoMn₂O₄ adsorbent will be irreversible, favorable, linear, or unfavorable, respectively.⁶⁴ It was found that the calculated R_L values verified the relationship; 0 < R_L < 1, for various initial dye concentrations (50–600 mg L⁻¹); therefore, the adsorption of the RB5 dye on the CoMn₂O₄ adsorbent is favorable.

3.3.6. Adsorption thermodynamics investigation. The influence of the temperature factor on the adsorption process was investigated by carrying out the adsorption experiments at the initial RB5 dye concentration of 50 mg L⁻¹ for 20 min equilibrium time at the temperature range of 298–328 K. The results presented in Fig. 15(a) show a slight increase in the adsorption capacity of the adsorbent with increasing the temperature indicating the endothermic nature of the adsorption process.

Fig. 15 The influence of temperature (a), plot of ln K_c versus 1/T (b), and plot ln(1 − θ) versus 1/T (c) for the adsorption of RB5 dye on CoMn₂O₄ adsorbent.

The adsorption thermodynamics was further explored by the calculation of some thermodynamic constants such as the Gibbs free energy change; ΔG⁰, entropy change; ΔS⁰, enthalpy change; ΔH⁰, to gain in-depth knowledge about the energetic changes associated with the RB5 dye adsorption. Eqn (14) and (15) were employed to calculate the thermodynamic constants.³³

ΔG⁰ = ΔH⁰ − TΔS⁰ (15)

where, K_c (L g⁻¹) is the thermodynamic equilibrium constant (K_c = q_e/C_e), R (8.314 × 10⁻³ kJ mol⁻¹ K⁻¹) is the universal gas constant, and T (K) is the absolute temperature of the dye solution. From eqn (14), it becomes clear that plotting of ln K_c versus 1/T results in a straight line, and from the slope and intercept ΔH⁰ and ΔS⁰ can be determined, as shown in Fig. 15(b). Besides, by employing the calculated values of ΔH⁰ and ΔS⁰ as well as eqn (15), ΔG⁰ can be calculated. The estimated thermodynamic constants are listed in Table 3. The calculated thermodynamic results (Table 3) provide some aspects about the adsorption process. Firstly, the positive ΔH⁰ and negative ΔG⁰ values mean that the adsorption of the RB5 dye on CoMn₂O₄ is an endothermic and spontaneous process, respectively. Secondly, increasing the negative values of the ΔG⁰ with increasing the temperature indicates that the adsorption process is more thermodynamically preferable at higher temperatures. Notably, the adsorption process is classified as a physisorptive process if the values of ΔG⁰ and ΔH⁰ are in the range from −20 to 0 kJ mol⁻¹ and <40 kJ mol⁻¹, respectively. Consequently, the adsorption of the RB5 dye on the CoMn₂O₄ nano-adsorbent is a physisorption since it fulfills those conditions. However, as the ΔH⁰ (22.144 kJ mol⁻¹) lies in the range of 2–40 kJ mol⁻¹ noticed for hydrogen bonds, this indicates that the hydrogen bonding mechanism may have some contribution to the adsorption process along with the main contribution that comes from the electrostatic interactions.^56,65

Table 3 Thermodynamic constants for the adsorption of RB5 dye on CoMn₂O₄ adsorbent

Temperature (K)	Kc	ΔG^0 (kJ mol^−1)	ΔS^0 (J mol^−1 K^−1)	ΔH^0 (kJ mol^−1)	Ea (kJ mol^−1)	S*
298	5.75	−4.336	0.0888	22.144	20.916	1.734 × 10^−5
308	7.83	−5.214
318	9.5	−6.102
328	13.39	−6.990

---

Moreover, the activation energy (E_a) of the adsorption of the dye on the as-prepared nanoparticles was estimated employing a modified Arrhenius equation (eqn (16)) associated with surface coverage (θ).^66,67

where, S* (0 < S* < 1) is an adsorbate/adsorbent function (sticking probability), and it depends on the temperature, θ = [1 − C_e/C₀], C₀ and C_e have the aforementioned meaning. Eqn (17) can be obtained if we substitute θ with [1 − C_e/C₀] in eqn (16).

ln(1 − θ) = ln S* + E_a/RT (17)

Plotting of ln(1 − θ) against 1/T produces a straight line (Fig. 15(c)), and from its slope the activation energy (E_a) of the adsorption process can be calculated. Interestingly, the E_a value is indicative, and it indicates whether the adsorption process is a chemisorption or physisorption based on whether its value is in the range of 40–800 or 5–40 kJ mol⁻¹, respectively.^3,68 In the current case, the activation energy (E_a) value for the adsorption of RB5 dye on the CoMn₂O₄ nano-adsorbent is estimated to be 20.92 kJ mol⁻¹ confirming the physisorption nature of the adsorption process, and this conclusion is consistent with the earlier one concluded from the ΔG⁰ and ΔH⁰ values.

3.3.7. Reusability investigation of the CoMn₂O₄ nano-adsorbent. The regeneration and reusability behavior of the adsorbents is an essential aspect of their industrial applications. Fig. 16(a and b) exhibits the FE-SEM images of the bare CoMn₂O₄ and the RB5 dye-loaded CoMn₂O₄ nano-adsorbent, respectively. It is obvious that the dye molecules did not mainly influence the morphology of the CoMn₂O₄ nanoparticles on adsorption, and only some microspheres have been ruptured. Additionally, one of the major differences in the CoMn₂O₄ morphology on adsorption of the dye is the presence of some darken areas on the surfaces of the CoMn₂O₄ nano-adsorbent, and this is obvious in the SEM image (Fig. 16(b)) of the dye-loaded CoMn₂O₄. To determine the regeneration and recycling availability of the as-prepared CoMn₂O₄ nano-adsorbent, regeneration experiments were performed on the dye-loaded adsorbent by its ignition at 500 °C for 30 min. Thus, five successive adsorption–desorption cycles were carried out, and the results were displayed in Fig. 16(c). Fig. 16(c) shows clearly that the CoMn₂O₄ nano-adsorbent adsorption capacity (q_m) maintained up to 128 and 125 mg g⁻¹ for the fourth and fifth cycles, respectively. Thus, the results demonstrate the stability and reusability of the as-prepared CoMn₂O₄ nano-adsorbent without a remarkable reduction in its adsorption efficiency for the removal of RB5 dye pollutant from aqueous media.

Fig. 16 FE-SEM images of the as-prepared CoMn₂O₄ adsorbent before adsorption (a) and after adsorption (b); and the reusability of CoMn₂O₄ adsorbent for the removal of RB5 dye (c).

3.3.8. Comparison between the adsorbent efficiencies of CoMn₂O₄ and other adsorbents. The maximum adsorption capacity (q_m) value of the as-prepared CoMn₂O₄ adsorbent for RB5 dye was compared with those reported for other adsorbents and tabulated in Table 4. Inspection of Table 4 reveals clearly that the as-prepared CoMn₂O₄ adsorbent has the superiority over the all tabulated adsorbents except graphite oxide/chitosan composite since our adsorbent has q_m value of 132 mg g⁻¹, and the q_m value of that composite is 277 mg g⁻¹. Though, generally, some adsorbents may have higher q_m values for the removal of RB5 dye, some features of the as-synthesized CoMn₂O₄ adsorbent including its stability, ca. 100% desorption ratio, and five-time reusability make it still a promising and suitable adsorbent for the removal of RB5 dye from textile industry wastewater. Additionally, the leaching of Co and Mn into the dye solution from the CoMn₂O₄ adsorbent under the optimum adsorption conditions was investigated. In this regard, the concentration of the dissolved metal ions (i.e. Co and Mn ions) under the optimum conditions (i.e. at pH 1.5) was determined using an inductively coupled plasma-optical emission spectrometer, and the leached Co and Mn concentrations were found to be ca. 0.056 and 0.009 mg L⁻¹, respectively. The relatively low concentrations of the leached ions would not result in environmental metal pollution, indicating good chemical stability of the CoMn₂O₄ adsorbent.

Table 4 Maximum adsorption capacities of CoMn₂O₄ adsorbent and other different adsorbents toward RB5 dye

Adsorbent	Maximum adsorption capacity, qm (mg g^−1)	References
Graphite oxide/chitosan composite	277	69
Spinel CoMn2O4	132	Present work
Modified clay (sepiolite)	120.5	70
Chitin	92	7
ZnO nanoparticles	80.9	3
Carbon from sugar beet pulp	80	71
Brown seaweed (acid treated)	73.2	72
Chitin : graphene oxide (3 : 1)	70	7
Modified zeolites	61	73
Activated carbon	59	74
Powdered activated carbon	58.8	75
Bagasse fly ash	16.42	76
Afsin–Elbistan fly ash	7.936	75
Biomass fly ash	4.38	77
Sunflower seed shells	1	78
Eichhornia/chitosan composite	0.606	79

---

3.3.9. Proposed mechanism of RB5 adsorption on CoMn₂O₄ nano-adsorbent. Based on the obtained experimental results and the published data, we have proposed the probable mechanisms contributing and controlling the adsorption of RB5 dye on CoMn₂O₄ nano-adsorbent at pH 1.5. The proposed mechanisms are (i) electrostatic interaction between the positively charged particles of CoMn₂O₄ generated from the protonation of the oxygen atoms at the CoMn₂O₄ surfaces (i.e. M–OH₂⁺) at lower pH values (i.e. pH < pH_pzc), and the negatively charged dye molecules containing SO₃⁻ groups, and this is the main controlling mechanism; (ii) hydrogen bonding between O, N, and S atoms of the functional groups of the RB5 dye molecules and OH groups (M–OH) produced from the reaction of the metal oxide surfaces with the adsorbed water molecules,⁵⁴ as noted previously, (i.e. M–OH₂⁺, M–OH, and M–O⁻); and (iii) hydrophobic–hydrophobic mechanism due to the interaction between the hydrophobic parts of the adsorbent and the RB5 dye molecules. Moreover, the proposed mechanism may be further confirmed from the FR-IR spectra of the adsorbent before and after the dye adsorption (Fig. 5(b and c)) since there are many differences between their FT-IR spectra. After the adsorption process, the frequencies of the bare CoMn₂O₄ appeared at 1616, and 3385 cm⁻¹ were shifted to lower vibrational frequencies: 455, 1610, and 3293 cm⁻¹, respectively, and this reduction might be attributed to the formation of the hydrogen bonds between the surface OH of the adsorbent and O-, N-, and S-containing functional groups of the RB5 dye molecules. The results are also compatible with the calculated ΔH⁰ value of the adsorption process and the published data.⁵⁴ Moreover, the decrease in the adsorption capacity of the adsorbent after its recycling due to the ignition of the RB5 dye-loaded CoMn₂O₄ indicating that the structural OH groups of the CoMn₂O₄ adsorbent are the main contributors to the adsorption mechanism of RB5 dye molecules.⁵⁶

4. Conclusions

In summary, mesoporous sphere-like spinel CoMn₂O₄ nanostructure was successfully synthesized via a template-free hydrothermal approach. In this respect, we have developed a facile hydrothermal synthesis of CoCO₃/MnCO₃ composite microspheres by the hydrothermal reaction of CoCl₂(H₂O)₆, MnCl₂(H₂O)₄ and NH₄HCO₃ in aqueous media. Various experimental parameters influencing the hydrothermal reaction were examined. Interestingly, the hydrothermally synthesized CoCO₃/MnCO₃ composite with a sole 0.4Co²⁺ : 0.6Mn²⁺ molar ratio produced pure CoMn₂O₄ nanostructure (∼16 nm) on calcination at 550 °C for 1 h. However, the composites with other Co²⁺ : Mn²⁺ molar ratios gave impure products on calcination. The produced CoMn₂O₄ nanostructure revealed high adsorption capacity (∼132 mg g⁻¹) for the removal of Reactive Black 5 (RB5) dye from aqueous media. The adsorption process could be explained well using the pseudo-second-order kinetics and the Langmuir isotherm model. Moreover, the adsorption of the dye is an endothermic, spontaneous, and physisorption process. Consequently, the as-prepared product can be efficient for the removal of RB5 dye from wastewater. Besides, the proposed synthetic route will hopefully open a channel to synthesize other mixed metal oxides.

Acknowledgements

Benha University (Egypt), Prof. Alaa S. Amin (Chem. Dept., Faculty of Science, Benha University), and Prof. Ibrahim S. Ahmed (Chem. Dept., Faculty of Science, Benha University) are acknowledged by the first author (M. Y. Nassar) for their support.

References

1. T. A. Khan, S. Dahiya and I. Ali, Use of kaolinite as adsorbent: Equilibrium, dynamics and thermodynamic studies on the adsorption of Rhodamine B from aqueous solution, Appl. Clay Sci., 2012, 69, 58–66.
2. N. F. Cardoso, R. B. Pinto, E. C. Lima, T. Calvete, C. V. Amavisca, B. Royer, M. L. Cunha, T. H. M. Fernandes and I. S. Pinto, Removal of remazol black B textile dye from aqueous solution by adsorption, Desalination, 2011, 269, 92–103.
3. M. Y. Nassar, M. M. Moustafa and M. M. Taha, Hydrothermal tuning of the morphology and particle size of hydrozincite nanoparticles using different counterions to produce nanosized ZnO as an efficient adsorbent for textile dye removal, RSC Adv., 2016, 6, 42180–42195.
4. W. Wang, T. Jiao, Q. Zhang, X. Luo, J. Hu, Y. Chen, Q. Peng, X. Yan and B. Li, Hydrothermal synthesis of hierarchical core–shell manganese oxide nanocomposites as efficient dye adsorbents for wastewater treatment, RSC Adv., 2015, 5, 56279–56285.
5. J. Hu, J. Men, Y. Liu, H. Huang and T. Jiao, One-pot synthesis of Ag-modified LaMnO₃-graphene hybrid photocatalysts and application in the photocatalytic discoloration of an azo-dye, RSC Adv., 2015, 5, 54028–54036.
6. M. Y. Nassar, I. S. Ahmed and I. Samir, A novel synthetic route for magnesium aluminate (MgAl₂O₄) nanoparticles using sol–gel auto combustion method and their photocatalytic properties, Spectrochim. Acta, Part A, 2014, 131, 329–334.
7. J. A. González, M. E. Villanueva, L. L. Piehl and G. J. Copello, Development of a chitin/graphene oxide hybrid composite for the removal of pollutant dyes: Adsorption and desorption study, Chem. Eng. J., 2015, 280, 41–48.
8. J. García-Montaño, X. Domènech, J. A. García-Hortal, F. Torrades and J. Peral, The testing of several biological and chemical coupled treatments for Cibacron Red FN-R azo dye removal, J. Hazard. Mater., 2008, 154, 484–490.
9. Y. Xing, X. Chen and D. Wang, Electrically Regenerated Ion Exchange for Removal and Recovery of Cr(VI) from Wastewater, Environ. Sci. Technol., 2007, 41, 1439–1443.
10. B. R. Singh, M. Shoeb, W. Khan and A. H. Naqvi, Synthesis of graphene/zirconium oxide nanocomposite photocatalyst for the removal of rhodamine B dye from aqueous environment, J. Alloys Compd., 2015, 651, 598–607.
11. T. Jiao, H. Guo, Q. Zhang, Q. Peng, Y. Tang, X. Yan and B. Li, Reduced Graphene Oxide-Based Silver Nanoparticle-Containing Composite Hydrogel as Highly Efficient Dye Catalysts for Wastewater Treatment, Sci. Rep., 2015, 5, 11873.
12. K. B. Tan, M. Vakili, B. A. Horri, P. E. Poh, A. Z. Abdullah and B. Salamatinia, Adsorption of dyes by nanomaterials: Recent developments and adsorption mechanisms, Sep. Purif. Technol., 2015, 150, 229–242.
13. T. Jiao, Y. Liu, Y. Wu, Q. Zhang, X. Yan, F. Gao, A. J. P. Bauer, J. Liu, T. Zeng and B. Li, Facile and Scalable Preparation of Graphene Oxide-Based Magnetic Hybrids for Fast and Highly Efficient Removal of Organic Dyes, Sci. Rep., 2015, 5, 12451.
14. Y. Feng, F. Yang, Y. Wang, L. Ma, Y. Wu, P. G. Kerr and L. Yang, Basic dye adsorption onto an agro-based waste material – Sesame hull (Sesamum indicum L.), Bioresour. Technol., 2011, 102, 10280–10285.
15. E. R. Cornelissen, N. Moreau, W. G. Siegers, A. J. Abrahamse, L. C. Rietveld, A. Grefte, M. Dignum, G. Amy and L. P. Wessels, Selection of anionic exchange resins for removal of natural organic matter (NOM) fractions, Water Res., 2008, 42, 413–423.
16. R.-k. Xu, S.-c. Xiao, J.-h. Yuan and A.-z. Zhao, Adsorption of methyl violet from aqueous solutions by the biochars derived from crop residues, Bioresour. Technol., 2011, 102, 10293–10298.
17. B. H. Hameed, Equilibrium and kinetics studies of 2,4,6-trichlorophenol adsorption onto activated clay, Colloids Surf., A, 2007, 307, 45–52.
18. H. Guo, T. Jiao, Q. Zhang, W. Guo, Q. Peng and X. Yan, Preparation of Graphene Oxide-Based Hydrogels as Efficient Dye Adsorbents for Wastewater Treatment, Nanoscale Res. Lett., 2015, 10, 1–10.
19. C. Luo, Z. Tian, B. Yang, L. Zhang and S. Yan, Manganese dioxide/iron oxide/acid oxidized multi-walled carbon nanotube magnetic nanocomposite for enhanced hexavalent chromium removal, Chem. Eng. J., 2013, 234, 256–265.
20. M. Y. Nassar, A. S. Attia, K. A. Alfallous and M. F. El-Shahat, Synthesis of two novel dinuclear molybdenum(0) complexes of quinoxaline-2,3-dione: New precursors for preparation of α-MoO₃ nanoplates, Inorg. Chim. Acta, 2013, 405, 362–367.
21. M. Y. Nassar, T. Y. Mohamed and I. S. Ahmed, One-pot solvothermal synthesis of novel cobalt salicylaldimine–urea complexes: A new approach to Co₃O₄ nanoparticles, J. Mol. Struct., 2013, 1050, 81–87.
22. E. Wang, L. Hu, S. Lei, S. Zhang, S. Zhang and W. Gong, Influence of calcination atmosphere on adsorptive performance of composite minerals materials, Appl. Clay Sci., 2015, 118, 138–150.
23. M. Y. Nassar, A. S. Amin, I. S. Ahmed and S. Abdallah, Sphere-like Mn₂O₃ nanoparticles: Facile hydrothermal synthesis and adsorption properties, J. Taiwan Inst. Chem. Eng., 2016, 64, 79–88.
24. M. Abdel Salam, Synthesis and characterization of novel manganese oxide nanocorals and their application for the removal of methylene blue from aqueous solution, Chem. Eng. J., 2015, 270, 50–57.
25. G. Yang, X. Xu, W. Yan, H. Yang and S. Ding, Single-spinneret electrospinning fabrication of CoMn₂O₄ hollow nanofibers with excellent performance in lithium-ion batteries, Electrochim. Acta, 2014, 137, 462–469.
26. F. Yunyun, L. Xu, Z. Wankun, Z. Yuxuan, Y. Yunhan, Q. Honglin, X. Xuetang and W. Fan, Spinel CoMn₂O₄ nanosheet arrays grown on nickel foam for high-performance supercapacitor electrode, Appl. Surf. Sci., 2015, 357, 2013–2021.
27. X. Zhai, W. Yang, M. Li, G. Lv, J. Liu and X. Zhang, Noncovalent hybrid of CoMn₂O₄ spinel nanocrystals and poly (diallyldimethylammonium chloride) functionalized carbon nanotubes as efficient electrocatalysts for oxygen reduction reaction, Carbon, 2013, 65, 277–286.
28. M. H. Kim, Y. J. Hong and Y. C. Kang, Electrochemical properties of yolk–shell and hollow CoMn₂O₄ powders directly prepared by continuous spray pyrolysis as negative electrode materials for lithium ion batteries, RSC Adv., 2013, 3, 13110–13114.
29. Y. Jin, L. Wang, Q. Jiang, X. Du, C. Ji and X. He, Morphology controllable synthesis of CoMn₂O₄ structures by adjusting the urea concentration: From microflowers to microspheres, Mater. Lett., 2016, 168, 166–170.
30. L. Ren, J. Chen, X. Wang, M. Zhi, J. Wu and X. Zhang, Facile synthesis of flower-like CoMn₂O₄ microspheres for electrochemical supercapacitors, RSC Adv., 2015, 5, 30963–30969.
31. Y. Xu, X. Wang, C. An, Y. Wang, L. Jiao and H. Yuan, Facile synthesis route of porous MnCo₂O₄ and CoMn₂O₄ nanowires and their excellent electrochemical properties in supercapacitors, J. Mater. Chem. A, 2014, 2, 16480–16488.
32. M. Imran, D. H. Kim, W. A. Al-Masry, A. Mahmood, A. Hassan, S. Haider and S. M. Ramay, Manganese-, cobalt-, and zinc-based mixed-oxide spinels as novel catalysts for the chemical recycling of poly(ethylene terephthalate) via glycolysis, Polym. Degrad. Stab., 2013, 98, 904–915.
33. M. Y. Nassar, I. S. Ahmed, T. Y. Mohamed and M. Khatab, A controlled, template-free, and hydrothermal synthesis route to sphere-like [small alpha]-Fe₂O₃ nanostructures for textile dye removal, RSC Adv., 2016, 6, 20001–20013.
34. M. Y. Nassar and I. S. Ahmed, Template-free hydrothermal derived cobalt oxide nanopowders: Synthesis, characterization, and removal of organic dyes, Mater. Res. Bull., 2012, 47, 2638–2645.
35. M. Y. Nassar and I. S. Ahmed, Hydrothermal synthesis of cobalt carbonates using different counter ions: An efficient precursor to nano-sized cobalt oxide (Co₃O₄), Polyhedron, 2011, 30, 2431–2437.
36. M. Y. Nassar, Size-controlled synthesis of CoCO₃ and Co₃O₄ nanoparticles by free-surfactant hydrothermal method, Mater. Lett., 2013, 94, 112–115.
37. H.-P. Cong and S.-H. Yu, Shape Control of Cobalt Carbonate Particles by a Hydrothermal Process in a Mixed Solvent: An Efficient Precursor to Nanoporous Cobalt Oxide Architectures and Their Sensing Property, Cryst. Growth Des., 2009, 9, 210–217.
38. H. B. Wu, J. S. Chen, H. H. Hng and X. Wen Lou, Nanostructured metal oxide-based materials as advanced anodes for lithium-ion batteries, Nanoscale, 2012, 4, 2526–2542.
39. R. Jenkins and R. L. Snyder, Introduction to X-ray powder diffractometry, John Wiley & Sons, Inc., New York, 1996.
40. K. Cheng, F. Yang, G. Wang, J. Yin and D. Cao, Facile synthesis of porous (Co, Mn)₃O₄ nanowires free-standing on a Ni foam and their catalytic performance for H₂O₂ electroreduction, J. Mater. Chem. A, 2013, 1, 1669–1676.
41. Y. Yao, Y. Cai, G. Wu, F. Wei, X. Li, H. Chen and S. Wang, Sulfate radicals induced from peroxymonosulfate by cobalt manganese oxides (Co_xMn₃ − xO₄) for Fenton-Like reaction in water, J. Hazard. Mater., 2015, 296, 128–137.
42. H. M. Aly, M. E. Moustafa, M. Y. Nassar and E. A. Abdelrahman, Synthesis and characterization of novel Cu(II) complexes with 3-substituted-4-amino-5-mercapto-1,2,4-triazole Schiff bases: A new route to CuO nanoparticles, J. Mol. Struct., 2015, 1086, 223–231.
43. K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Applications in Coordination, Organometallic, and Bioinorganic Chemistry, Wiley, 2009.
44. T. Xiaowei and Y. Hong, Synthesis of magnetic nanocomposites and alloys from platinum–iron oxide core–shell nanoparticles, Nanotechnology, 2005, 16, S554.
45. E. Westsson and G. Koper, How to Determine the Core–Shell Nature in Bimetallic Catalyst Particles?, Catalysts, 2014, 4, 375.
46. S. A. Hosseini, A. Niaei, D. Salari and S. R. Nabavi, Nanocrystalline AMn₂O₄ (A = Co, Ni, Cu) spinels for remediation of volatile organic compounds—synthesis, characterization and catalytic performance, Ceram. Int., 2012, 38, 1655–1661.
47. A. Goyal, S. Bansal, V. Kumar, J. Singh and S. Singhal, Mn substituted cobalt ferrites (CoMn_xFe₂ − xO₄ (x = 0.0, 0.2, 0.4, 0.6, 0.8, 1.0)): As magnetically separable heterogeneous nanocatalyst for the reduction of nitrophenols, Appl. Surf. Sci., 2015, 324, 877–889.
48. A. V. Salker and S. M. Gurav, Electronic and catalytic studies on Co₁ − xCu_xMn₂O₄ for CO oxidation, J. Mater. Sci., 2000, 35, 4713–4719.
49. M. Y. Nassar and M. Khatab, Cobalt ferrite nanoparticles via a template-free hydrothermal route as an efficient nano-adsorbent for potential textile dye removal, RSC Adv., 2016, 6, 79688–79705.
50. P. Pascariu, A. Airinei, M. Grigoras, N. Fifere, L. Sacarescu, N. Lupu and L. Stoleriu, Structural, optical and magnetic properties of Ni doped SnO₂ nanoparticles, J. Alloys Compd., 2016, 668, 65–72.
51. C. Sánchez, E. Flores, M. Barawi, J. M. Clamagirand, J. R. Ares and I. J. Ferrer, Marcasite revisited: Optical absorption gap at room temperature, Solid State Commun., 2016, 230, 20–24.
52. M. Thommes, K. Kaneko, V. Neimark Alexander, P. Olivier James, F. Rodriguez-Reinoso, J. Rouquerol and S. W. Sing Kenneth, Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution (IUPAC Technical Report), in Pure and Applied Chemistry, 2015, p. 1051.
53. M. Ahmaruzzaman and S. Laxmi Gayatri, Batch adsorption of 4-nitrophenol by acid activated jute stick char: Equilibrium, kinetic and thermodynamic studies, Chem. Eng. J., 2010, 158, 173–180.
54. D. H. K. Reddy and Y.-S. Yun, Spinel ferrite magnetic adsorbents: Alternative future materials for water purification?, Coord. Chem. Rev., 2016, 315, 90–111.
55. G. Newcombe and M. Drikas, Adsorption of NOM onto activated carbon: Electrostatic and non-electrostatic effects, Carbon, 1997, 35, 1239–1250.
56. S. Liu, Y. Ding, P. Li, K. Diao, X. Tan, F. Lei, Y. Zhan, Q. Li, B. Huang and Z. Huang, Adsorption of the anionic dye Congo red from aqueous solution onto natural zeolites modified with N,N-dimethyl dehydroabietylamine oxide, Chem. Eng. J., 2014, 248, 135–144.
57. O. Gulnaz, A. Kaya and S. Dincer, The reuse of dried activated sludge for adsorption of reactive dye, J. Hazard. Mater., 2006, 134, 190–196.
58. K. Chinoune, K. Bentaleb, Z. Bouberka, A. Nadim and U. Maschke, Adsorption of reactive dyes from aqueous solution by dirty bentonite, Appl. Clay Sci., 2016, 123, 64–75.
59. W. J. Wber and J. C. Morris, Proceedings of the International Conference on Water Pollution Symposium, Pergamon Press, Oxford, New York, 1962.
60. F. A. Batzias and D. K. Sidiras, Dye adsorption by prehydrolysed beech sawdust in batch and fixed-bed systems, Bioresour. Technol., 2007, 98, 1208–1217.
61. A.-N. M. Salem, M. A. Ahmed and M. F. El-Shahat, Selective adsorption of amaranth dye on Fe₃O₄/MgO nanoparticles, J. Mol. Liq., 2016, 219, 780–788.
62. I. Langmuir, The Adsorption of Gases on Plane Surfaces of Glass, Mica and Platinum, J. Am. Chem. Soc., 1918, 40, 1361–1403.
63. G. D. Halsey, The Role of Surface Heterogeneity in Adsorption, in Advances in Catalysis, ed. V. I. K. W. G. Frankenburg and E. K. Rideal, Academic Press, 1952, pp. 259–269.
64. T. G. Venkatesha, R. Viswanatha, Y. Arthoba Nayaka and B. K. Chethana, Kinetics and thermodynamics of reactive and vat dyes adsorption on MgO nanoparticles, Chem. Eng. J., 2012, 198–199, 1–10.
65. B. von Oepen, W. Kördel and W. Klein, Sorption of nonpolar and polar compounds to soils: Processes, measurements and experience with the applicability of the modified OECD-Guideline 106, Chemosphere, 1991, 22, 285–304.
66. H. R. Mahmoud, S. M. Ibrahim and S. A. El-Molla, Textile dye removal from aqueous solutions using cheap MgO nanomaterials: Adsorption kinetics, isotherm studies and thermodynamics, Adv. Powder Technol., 2016, 27, 223–231.
67. M. Ghaedi, A. Hekmati Jah, S. Khodadoust, R. Sahraei, A. Daneshfar, A. Mihandoost and M. K. Purkait, Cadmium telluride nanoparticles loaded on activated carbon as adsorbent for removal of sunset yellow, Spectrochim. Acta, Part A, 2012, 90, 22–27.
68. J. Ma, Y. Jia, Y. Jing, Y. Yao and J. Sun, Kinetics and thermodynamics of methylene blue adsorption by cobalt–hectorite composite, Dyes Pigm., 2012, 93, 1441–1446.
69. G. J. Copello, A. M. Mebert, M. Raineri, M. P. Pesenti and L. E. Diaz, Removal of dyes from water using chitosan hydrogel/SiO₂ and chitin hydrogel/SiO₂ hybrid materials obtained by the sol–gel method, J. Hazard. Mater., 2011, 186, 932–939.
70. O. Ozdemir, B. Armagan, M. Turan and M. S. Çelik, Comparison of the adsorption characteristics of azo-reactive dyes on mezoporous minerals, Dyes Pigm., 2004, 62, 49–60.
71. N. A. Travlou, G. Z. Kyzas, N. K. Lazaridis and E. A. Deliyanni, Graphite oxide/chitosan composite for reactive dye removal, Chem. Eng. J., 2013, 217, 256–265.
72. K. Vijayaraghavan and Y.-S. Yun, Biosorption of C.I. Reactive Black 5 from aqueous solution using acid-treated biomass of brown seaweed Laminaria sp, Dyes Pigm., 2008, 76, 726–732.
73. B. Armaǧan, O. Özdemir, M. Turan and M. S. Çelik, The removal of reactive azo dyes by natural and modified zeolites, J. Chem. Technol. Biotechnol., 2003, 78, 725–732.
74. K. Y. Foo and B. H. Hameed, Insights into the modeling of adsorption isotherm systems, Chem. Eng. J., 2010, 156, 2–10.
75. Z. Eren and F. N. Acar, Equilibrium and kinetic mechanism for Reactive Black 5 sorption onto high lime Soma fly ash, J. Hazard. Mater., 2007, 143, 226–232.
76. M. Rachakornkij, S. Ruangchuay and S. Teachakulwiroj, Removal of reactive dyes from aqueous solution using bagasse fly ash, Songklanakarin, Journal of Science and Technology, 2004, 26, 13–24.
77. P. Pengthamkeerati, T. Satapanajaru and O. Singchan, Sorption of reactive dye from aqueous solution on biomass fly ash, J. Hazard. Mater., 2008, 153, 1149–1156.
78. J. F. Osma, V. Saravia, J. L. Toca-Herrera and S. R. Couto, Sunflower seed shells: A novel and effective low-cost adsorbent for the removal of the diazo dye Reactive Black 5 from aqueous solutions, J. Hazard. Mater., 2007, 147, 900–905.
79. M. M. El-Zawahry, F. Abdelghaffar, R. A. Abdelghaffar and A. G. Hassabo, Equilibrium and kinetic models on the adsorption of Reactive Black 5 from aqueous solution using Eichhornia crassipes/chitosan composite, Carbohydr. Polym., 2016, 136, 507–515.

---