Characterization and applicability of the Naima–Tiaret–Algeria illite-montmorillonite clay for methylene blue dye removal: adsorption kinetic, isotherm and thermodynamic studies

Water is necessary for life on Earth. Nonetheless, different human activities, whether industrial, urban, or agricultural, make pollution worse. Dyes are used in a variety of industries, including textiles, paper, leather dyes, and the food and cosmetics industries (Batzias & Sidiras 2007). The global production of colorants is approximately 700,000 t per year (Taylor et al. 2021). Synthetic dyes, for example, utilized in the textile industry are released directly into the aquatic environment without any prior treatment. A dye is a substance with two distinct features that are unrelated to one another: color and the ability to be fixed on a support, such as a textile. Dyes with low concentrations in water (even less than 1 kgm⁻³ for some dyes) are noticeable, undesirable and persistent in the environment. The first medical dye was created by William Henry Perkin (quinine). This has benefited the industrial dye synthesis industry (Delgado-Vargas & Octavioparedes-Lopez 2003; Robert et al. 2004). The toxicity of various dyes has been studied in several studies on aquatic organisms (fish, algae, bacteria, etc.) and mammals (mutagenic, mortality and carcinogenic impacts). Furthermore, studies on the effect of dyes on the activity of both anaerobic and aerobic bacteria in wastewater treatment systems have been conducted. According to studies on various commercial dyes (Clarke & Anliker 1980; Thai & Ruey-Shin 2012), basic dyes are the most toxic to algae. Methylene blue (MB), one of the most common cationic dyes, is a heterocyclic aromatic chemical compound, which heavily applied in the food, rubber products, cotton and wool industries, wood and silk dyeing sectors (Vargas et al. 2011).

The aforementioned industries' effluent contains significant levels of MB (Beltrán-Heredia et al. 2011). A salt used as a color and medication is MB, also known as methylthioninium chloride (Modarai et al. 2002). As a medication, it is primarily utilized to treat methemoglobinemia (British National Form et al. 2015). It is used in particular to treat symptoms that do not go away after receiving oxygen treatment or methemoglobin levels exceeding 30%. Before this, it was recommended against using it to treat urinary tract infections and cyanide poisoning. It is usually administered via vein injection. Due to the constant structure of MB and its limited capacity to biodegrade, utilizing this dye will result in several significant environmental issues. Permanent exposure to MB will cause increasing in heart beats, cyanosis, shock, jaundice and irritation to the skin in humans (Xiao et al. 2015). Many water pollution control projects have been completed in recent years as a result of this major threat to aquatic life and the health of human populations because they are highly charged with highly toxic, harmful substances, responsible for the bad odor and abnormal coloring of the water.

Indeed, several conventional techniques, such as coagulation and flocculation, reverse osmosis, advanced oxidation filtration, electro-dialysis technology, adsorption, etc., have been employed in the removal of dyes from aqueous systems (Koulouchi 2007; Guezzen et al. 2023). However, most of these processes face various challenges and limitations, such as high cost, high energy consumption, complicated equipment, ineffective at low metal concentrations and long reaction times. Nowadays, the adsorption process is one of the effective, low investment and environmentally friendly techniques to remove recalcitrant organic and inorganic compounds from wastewater and landfill leachate (Benhadria et al. 2020; Bennama et al. 2022). Adsorption techniques for wastewater treatment have become more popular in recent years owing to their efficiency in the removal of pollutants too stable for biological methods.

Right now, various materials have been proposed to remove organic and inorganic pollutants form wastewater including activated carbon (Akar et al. 2006; Azharul et al. 2017; Cheng et al. 2018), clay minerals (Elaziouti et al. 2011a, 2011b; Gu et al. 2019; Guezzen et al. 2023; Ssouni et al. 2023), diatomite (Mohamed et al. 2019; Ebrahimi & Kumar 2021), agricultural residues (Robinson et al. 2002; Elaziouti et al. 2011a, 2011b; Bennama et al. 2022), layered double hydroxides (HDLs) (Tarmizi et al. 2019; Bouteiba et al. 2020a, 2020b) and other materials (Elaziouti et al. 2015; Dai et al. 2018; Bouhadjar et al. 2019; Benhadria et al. 2020).

An ideal potential adsorbent should have an adequate capacity with a large surface area, both thermally and chemically stable, abundantly available, selective, low cost, sustainable and easily regeneratable (Manyangadze et al. 2020; Ebrahimi & Kumar 2021).

To date, few studies dealing with the removal of inorganic and synthetic pollutants by the interstratified clays are reported in the literature (Hajjaji et al. 2006; Missana et al. 2008; Da Silva & Guerra 2013; Ahrouch et al. 2019a, 2019b; Taibi et al. 2020).

In this work, Interstratified illite-montmorillonite (IMN) clays abundant in the Naima area (Tiaret-Algeria) were used as an alternative adsorbent to remove MB dyes from aqueous solutions. To understand the adsorption mechanism and to know the possible interactions between different dye functions and interstratified clay, the pristine IMN and its physically and chemically treated interstratified illite-montmorillonite form (TIMN) clays were characterized by XRD, FTIR, SEM-EDS, XRF, DSC-TG and DC methods. To optimize the adsorption parameters of the MB dye by interstratified clays, the adsorption kinetics of MB were examined under the impact of three various parameters including contact time, initial dye concentration and temperature. To fully explain the adsorption mechanism, kinetic data were correlated to the pseudo-first-order, pseudo-second-order and Elovich models. The adsorption isotherms for MB dye removal by clay materials were also adjusted using Langmuir, Freundlich and Dubinin–Radushkevich (D–R) isotherm models. Besides, the thermodynamic parameters, such as enthalpy, entropy and the free energy of adsorption, were determined in detail. To analyze the sorption behavior of MB dye on the interstratified clays, a mechanistic model for interpreting the sorption data in the illite-montmorillonite system was developed. Eventually, a comparison study between the adsorption capability values of our clays to that of other clay materials in previous relevant studies for MB removal was performed.

Chemicals used in this study, such as NaCl (CAS. 7647-14-5), KOH (CAS. 1310-58-3), NaOH (CAS. 1310-73-2) and AgNO₃ (CAS. 7761-88-8) with the highest analytical purity were obtained from Aldrich chemical company ltd. MB was purchased from Biochem. All the chemicals used in this study were of analytical grade without further purification. MB dye has been used to identify the surface area of clay minerals for decades. MB is a cationic dye in water that is adsorbed by negatively charged clay surfaces. The molecular structure and chemical properties of MB dye are illustrated in Table 1. Natural illite-montmorillonite of Naima region (IMN) was obtained from the region of Naima–Tiaret–Algeria.

Table 1

3D Molecular structure and physical properties of methylene blue dye (MB) (3D molecular structure was created by CrystalMaker software)

Molecular structure in 3D presentation .	Chemicals properties .
	Chemical name (IUPAC) Molecular formula λmax (m) Molecular weight (kg mol−1) Solubility in water at 20 °C (kg m−3) Melting point (°C) CAS number	chlorure de 3,7-bis (diméthylamino) phénothiazin − 5-ium C16H18CIN3S 0.665 × 10−6 0.319 0.040 190 61-73-4

Molecular structure in 3D presentation .	Chemicals properties .
	Chemical name (IUPAC) Molecular formula λmax (m) Molecular weight (kg mol−1) Solubility in water at 20 °C (kg m−3) Melting point (°C) CAS number	chlorure de 3,7-bis (diméthylamino) phénothiazin − 5-ium C16H18CIN3S 0.665 × 10−6 0.319 0.040 190 61-73-4

View Large

The IMN material was first dried in sunlight, disintegrated into small pieces and then passed through a 2 mm sieving. A fraction of 0.020–0.1 kg was treated with sodium hydroxide solution (NaOH; 0.1 M) and heated at 70 °C for 3.6 × 10⁺² s under continuous stirring. The obtained solid material was filtered, rinsed repeatedly with distilled water until free Cl⁻ was not detected in the suspension (AgNO₃ test) and was dried at 100 °C overnight to constant weight. It was called TIMN.

X-ray diffraction (XRD) was conducted utilizing a Siemens D 5000 automatic diffractometer with Cu-K radiation (=0.154178 × 10⁻⁹ m) and a rear monochromator to remove iron fluorescence. A 0.2 × 10⁻³ m rear slit was inserted, while the front and back windows were both set at 2 × 10⁻³ m. Low rotational speed (0.01°s⁻¹) was chosen to produce distinct spokes over 10–80° for both the IMN and TIMN samples. A Shimatzu 8400 spectrometer with a resolution of 0.02 m⁻¹ and a range of 4–45 × 10⁴ m⁻¹ was used to gather FTIR spectra. An FEI Quanta 650 Scanning Electron Microscopy (SEM) from Bruker Nano GmbH in Berlin, Germany, connected to BRUKER XFlash 6/10 energy dispersive X-ray detector (EDS) was used to assess the elemental composition and powder morphology. The elements in crude clay purification were analyzed utilizing a Bruker S1 Titan XRF X-ray fluorescence (XRF) Spectrometer. An Instek 821 LCR meter with a temperature range of 27–167 °C was employed to measure conductivity. An Agilent spectrophotometer agile model was utilized for the optical density analysis, which was controlled by an 8543 computer. The peak wavelengths are obtained via automatic scanning between 2 × 10⁻⁷ and 8 × 10⁻⁷ m. To prevent interference over time, adsorbents are developed. We monitored the pyrolysis of these clays using DSC-ATG. SETARAM-Labs evo TGA was used. The samples (m = 0.026 × 10⁻³ kg) were put in a boat and heated at a rate of 10 °C per minute with an airflow ranging from 25 to 1,000 °C.

The adsorption of MB dye by IMN and TIMN adsorbents was investigated using batch experiments. 8 × 10⁻⁵ m³ of dye solution was added into an Erlenmeyer flask with 10⁻⁴ kg of the absorbent. The solution was then shaken in a water bath shaker at room temperature until equilibrium was attained. The supernatant solution was separated by filtration, and the residual dye concentration was analyzed using the spectrophotometry method.

K_F (m³ kg⁻¹) and 1/n are Freundlich constants. According to predictions, the adsorption process will either be linear (n = 1), chemical (n < 1) or a beneficial physical procedure (n > 1). The intercept and slop of the linear plot of ln q_e vs ln C_e were used to determine K_F and 1/n.

The mechanism of withdrawal is mainly physical connection when E < 8 kJ/mol and ion exchange if E 8 × 10⁻³ J mol⁻¹ ≤ E ≤ 16 × 10⁻³ J mol⁻¹ (Felhi et al. 2008).

The relative ability of soils to store one particular group of nutrients, the cations, is referred to as cation exchange capacity or CEC. It has two origins. One origin is isomorphic substitution in the tetrahedral- and/or octahedral sheet of the clay mineral layer. Substitution of aluminum by magnesium or silicon by aluminum leads to a negative net charge. This part of the CEC is considered to be constant since it is almost not sensitive to the pH of the system. The second origin is the dissociation of aluminol groups on the edges. Since the acidity of these groups is weak, the edge charges are pH dependent and the CEC depends on the pH. Table 2 presents the estimated CEC values for IMN and TIMN samples along with these reported for different materials in the literature (Eloussaief et al. 2009). The CEC of the IMN and TIMN, estimated by the MB method (Pal & Ghoshal 1977) are 35–50 cmol (+) kg⁻¹ and 51 cmol (+) kg⁻¹ of IMN, respectively.

Commonly, the surface active sites of aluminosilicates are characterized in terms of Broensted and Lewis acidity/basicity comprising exchangeable cations, coordinatively unsaturated ions Al³⁺, Mg²⁺, Fe³⁺, acid/basic hydroxyl groups and oxygen anions of aluminosilicates (Novikova et al. 2014). Surface Broensted acid sites (donate protons) and Lewis acid sites (accept electrons) are represented by different surface groups and exchangeable ions (Alda et al. 2017). In the aspect of mineralogy, lattice defects such as isomorphous substitution present in the clay or nanosheet surface are known to significantly affect clay properties. The resulting structure surfaces can be either electrically neutral or negatively charged (Zhang et al. 2017). The nature of the surface sites of the IMN sample is determined by using the potentiometric titration method. As reported in Table 3, the surface of IMN clay is positively charged in the whole range of pH acid. The weak isomorphic substitution of low valence cations for higher valence cations and the influence of the contaminant minerals such as kaolinite and illite are expected to be the primary reasons for the elevated yield of acidic products (Šucha et al. 2009). It is wealthy to know that kaolinite and illite, which are alumino-silicate clays, have higher Lewis acid components. Besides, Brönsted acid sites are characterized by M^VI–OH (M = Si, Al, Mg, Fe), in agreement with FTIR results which revealed an intense band at 16–14 × 10⁴ m⁻¹ region.

XRD is the primary tool for identifying and quantifying crystalline compounds. Powder XRD diffractograms characterize minerals and elements present in solids. Clay powders have been distinguished into smectites and kaolinites based on XRD patterns. The XRD patterns of IMN and TIMN samples, as presented in Figure 2, exhibited mainly basal reflections of montmorillonite, and illite with interstratified illite-montmorillonite, quartz and kaolinite as the major phase, besides minor amounts of muscovite and chlorite minerals. In the XRD patterns of IMN, the peaks (001) located at 26.49 ° and (003) at 217.17° obviously corresponded to the montmorillonite fraction while the illite phase was evidenced by the main peak (002) at 28.87°. The basal spacing at 7.20 and 3.58 × 10⁻¹⁰ m matching to the (001) and (002) reflections, approving the presence of kaolinite. The interstratified illite-montmorillonite (I/M) was evidenced by the main peak located at 2θ = 3.42° with a calculated basal spacing of 25.83 × 10⁻¹⁰ m. Additionally, quartz was presented by the prominent peak (011) at 3.33 × 10⁻¹⁰ m. In the TIMN sample, no significant change in the XRD profile was observed during the purification process. However, the decrease in intensity of the quartz reflections at 2θ = 27.8° (3.2 × 10⁻¹⁰ m), indicates the effectiveness of the purification process (Bhattacharyya & Gupta 2008).

Clay mineralogy is useful in characterizing the nature of clay minerals and is critical to many geoscience projects and understanding of palaeoclimate. As briefly mentioned in Table 4, chemical analysis showed three primary ingredients of IMN and TIMN samples: silica (SiO₂), alumina (Al₂O₃) and ferric oxide (Fe₂O₃). Besides, the IMN clay contained Ca, K, and Mg Ti oxides, as minor amounts, with other metal cations, and trace fractions. The mass fraction of SiO₂ and Al₂O₃ increased from 42.2 to 46.45% and from 11.97 to 12.98%, respectively. Conversely, the CaO and MgO mass fractions decreased from 1.52 to 1.23% and from 2.05 to 1.97%, respectively. The SiO₂/Al₂O₃ molar ratio of 3.52 and 3.57 for IMN and TIMN, respectively. The prominent K₂O content may indicate the possible presence of large amounts of illite in both samples (Hamdaoui et al. 2008).

The evolution of maximum absorption bands of dye in aqueous solution at various pH. The location of the absorption bans maxima of dye are collected in Table 6.

The adsorption isotherm is a crucial fundamental for describing the mechanism of dye removal onto the adsorbent surface. As represented in Figure 12, the adsorption isotherms of MB onto IMN and TIMN materials are L type according to Gilles et al. classification (Giles et al. 1960). The maximum adsorption capacities, q_m,exp, were 4.327 × 10⁻² and 4.914 × 10⁻² kg kg⁻¹ for IMN and TIMN, respectively.

The usual way to validate a model is to consider the goodness-of-fit using the linear regression coefficients, R². Results from using the Langmuir, Freundlich and D–R isotherm models have low regression coefficient values (R² ≤ 0.91), suggesting that it is not appropriate to use these types of linearization for this study. Additionally, the theoretical equilibrium adsorption capacities, q_e,th, for the whole applied models were significantly similar to that of the experimental data, excluding that deduced from the linear regression of the Langmuir for the TIMN sample. Besides, the obtained free energy (E) values for adsorbing MB using IMN and TIMN ranged from 0.58 to 0.75 × 10⁺³ J mol⁻¹, respectively, which are less than 8 × 10⁺³ J mol⁻¹, signifying that the process is physisorption (Elaziouti et al. 2015).

The smallest Chi-square χ² and APE values designate the similarity between the experimental and the theoretical data and reflect the best-fit isotherm model for the adsorption process (Auta & Hameed 2012). Table 9 presents two error functions (Chi-square χ² and APE) extracted from the linear regression of the Langmuir, Freundlich and D–R isotherm models.

The lower values of Chi-square χ² of the applied linearized models (Table 9) approve the better correlation between the experimental and calculated data. However, the higher APE values (APE was more than 10%) established that the Langmuir (APE = 44.2669) and D–R (APE = 11.7102) isotherm models did not present a suitable fit to the experimental results of MB by TIMN sample compared to the Freundlich model (APE = 3.5307). In contrast, the lower values of APE values obtained from the linear form of Langmuir (APE = 7.9838) and D–R (APE = 0.7510) isotherm models, in comparison to that of the Freundlich model (APE = 10.4910), confirm the better correlation between the experimental and calculated data of the adsorption of MB by IMN sample. In spite of lower standard error values like Chi-square χ² with all linearized models, they do not describe the equilibrium data adequately, because of the poor linear regression and high error function APE.

The values of the thermodynamic parameters for the sorption of dyes onto the TIMN clay sample are given in Table 10. The negative value of ΔH° (−381.281 J mol⁻¹) indicated the exothermic nature of the adsorption of MB onto TIMN for all tested temperatures. The positive value of ΔS° (18.518 J mol⁻¹ K⁻¹) showed that the sorption process was irreversible and random at the solid–liquid interface during the sorption of MB onto TIMN clay adsorbent.

The negative values of free energy ΔG°, as mentioned in Table 10, revealed that the adsorption is highly favorable and spontaneous (Benguella & Yacouta-Nour 2009). The ΔG° values obtained in this investigation are < −12,000 J mol⁻¹, indicating that physical adsorption is the predominant mechanism in the sorption process.

As indicated in Table 11, RM (raw material of clay) samples considerably reduced pollution compared to the other samples. This finding demonstrates that the physico-chemical features of the clay and the pollutant affect the removal efficiency. The estimated monolayer coverage capacities, q_e,th, were 4.673 × 10⁻² and 2.739 × 10⁻² kg kg⁻¹ for the adsorption of MB onto IMN and TIMN, respectively. The removal efficiency of IMN and TIMN samples was found to be great (Kannan & Sundaram 2001; Gücek et al. 2005; Rafatullah et al. 2010; Pimolpun & Pitt 2014; Allam et al. 2016), as a result of their affinity for MB removal.

According to the available functional groups (coordinatively unsaturated ions Al³⁺, Mg²⁺, Fe³⁺, exchangeable cations, acid hydroxyl groups and oxygen anions of aluminosilicates) on the surface of the TIMN, as buttressed by the XRD, FTIR and surface acidity results, there are possible:

• (1)

  Hydrogen bonding. The highly electronegative nitrogen atoms on the adsorbate MB molecule could generate hydrogen bonding with the hydrogen atoms covalently bonded to the surface of the TIMN clay (Liu et al. 2021)

• (2)

  Electrostatic attractions (ionic bonding). The edges of the TIMN clay surface contain hydroxyl surfaces (Si–OH, Al–OH, and Mg–OH groups), which could engender hydrophilic adsorption and can be attributed as a result of electrostatic attractions between the positively charged species of the MB (the positive charge is located on the N or S hetero–atoms of the MB molecule) and the negatively charged functional groups (variable charge at broken edges Si–O, Mg–O or hydroxyl surfaces (Si–OH, Al–OH and Mg–OH groups). In acidic environments, the release of H⁺ ions from the edge (more active) of the TIMN structure causes dye adsorption, from aqueous matrices, onto the edges of TIMN sample for cations like MB⁺ (Haounati et al. 2021).

• (3)

  Additionally, cationic exchange can take place due to the interlayer nature of the smectite (montmorillonite) fundamental particle in TIMN adsorbent with acceptable host interlayer space, thus cationic exchange is considered a potential mechanism. Based on the XRD results and assuming dimensions of 17 × 10⁻¹⁰ m × 7.6 × 10⁻¹⁰ m × 3.25 × 10⁻¹⁰ m for the MB molecule, we can assume that the exchangeable interlayer sodium cations could easily exchange with the MB⁺ cations through cation exchange mechanism and the intercalation of MB molecules in the interlayer space of TIMN has occurred. In the present study, the basal spacing for TIMN sample was 25.83 × 10⁻¹⁰ m, and by subtracting 10 × 10⁻¹⁰ m (d₀₀₁-spacing of dehydrated illite) and 10.1 × 10⁻¹⁰ m (basal space of hydrated montmorillonite), resulted in an interlayer space of 5.73 × 10⁻¹⁰ m, which is much larger than the thickness of one layer of MB orientated in the interlayer space in a nearly horizontal configuration (Li et al. 2010).

• (4)

  Methachromasy effect. The substitution of the main absorption band, which corresponds to the absorption of monomer, MB⁺ (MB is a metachromatic dye), by a new absorption band placed at shorter wavelengths. This metachromatic effect is observed in several clay/dye systems attributed to different processes.

• • (i) Self-association of dye molecules (dimers and H-aggregates) when they are adsorbed on the clay surface.

Surface charge density plays an important role in metachromatism. Thus, MB cations form predominantly larger H-aggregates (composed of three or more MB molecules) on the surface of smectites with a high charge density (high negative charge) (Bujdák et al. 2001; Cenens & Schoonheydt 1998; Elaziouti & Laouedj 2010; Li et al. 2010).

• (ii) n–π interactions. Intermolecular interaction between the electronic π-system of the dye with the electron lone- pairs of the oxygen atoms or at the clay surface. OH − π interactions between the OH groups on the surfaces of the clay and aromatic rings of MB dye may be one of the important interactions for the adhesion of MB dye including aromatic rings contributing to the adsorption of dye molecules (Nakamura et al. 2020).

The surface functional groups pre-demonstrated by FTIR analysis allow for interaction with the dye molecule through n–π interactions. The n–π interaction is proposed for clays with partial tetrahedral substitution of Si⁺⁴ by Al⁺³. Since TIMN Clay has this unique property, methachromatic effect has to be related to the n-π interactions between the organic functions of the MB dye molecule with the electron lone-pairs of the oxygen atoms in Si–O–Al, Si–O–Mg and Si–O–Fe groups of the TIMN clay surface through their different C = C double bonds of the aromatic rings (Wang et al. 2014; Jawad & Abdulhameed 2020; Ssouni et al. 2023).

A novel modified illite-montmorillonite clay was successfully elaborated via a chemical treatment process and characterized using XRD, FTIR, SEM-EDS, XRF, DSC-TG and DC methods. The adsorption capacity of the bare and treated IMN clays for MB dye was tested under conditions including contact period, initial dye concentration and temperature. It increased with the rise in the initial dye concentration, whereas it decreased significantly with increasing temperature. The adsorption process is well-fitted to the pseudo-second-order kinetic model. The maximum adsorption capacities, q_{e exp}, were 4.327 × 10⁻² and 4.914 × 10⁻² kg kg⁻¹ for IMN and TIMN, respectively. Results from using the Langmuir, Freundlich and Dubinin–Radushkevich (D–R) isotherm models have low regression coefficient values (R² ≤ 0.91), suggesting that it is not appropriate to use these types of linearization for this study. In spite of lower standard error values like Chi-square χ² with all linearized models, they do not describe the equilibrium data adequately, because of the poor linear regression and high error function APE. The obtained free energy (E) from D–R model values for adsorbing MB using IMN and TIMN ranged from 0.581 to 0.745 × 10⁻³J mol⁻¹, respectively, suggesting that the process is physisorption. According to the thermodynamic aspect, the negative value of ΔH° (−318.281 Jmol⁻¹) suggested that the sorption was exothermic in nature. The positive value of ΔS° (18.518 J mol⁻¹K⁻¹) showed that the sorption process was irreversible and random at the interface between adsorbent (IMN) and adsorbate solution (MB). The negative values of free energy ΔG° for all tested temperatures suggest that the adsorption was a highly favorable and spontaneous process. The main adsorption mechanisms of MB were hydrogen bonding, electrostatic attraction, cation exchange, n-π and OH − π interactions. These findings demonstrated the viability of an efficient TIMN adsorbent for MB dye removal in wastewater treatment systems.

The author gratefully acknowledges the material support from the Directorate-General for Scientific Research and Technological Development (DGSRTD) and the Ministry of Higher Education and Scientific Research (Algeria). We are also greatly indebted to the Laboratory of Agro-biotechnology and Nutrition in Semi-arid Zones, Ibn Khaldoun University, Tiaret, Algeria for their material support and the University of Science and Technology of Oran - Mohamed Boudiaf (USTO M.B.), Oran, Algeria.

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

The authors declare there is no conflict.