Modeling and kinetics investigation of adsorptive properties and regeneration of modified clay on a

Abstract

Keywords: Adaptive Neuro-Fuzzy Inference Systems (ANFIS); Adsorption; Artificial Neural Network (ANN); Congo Red; Desorption; Optimization; Response Surface Methodology (RSM)

Introduction

Industrial wastewater discharge has increased as a result of industrialization. Industrial wastewater comprises numerous dissolved dyestuffs as well as other pollutants like salts, emulsifiers, leveling agents, dispersing agents, dye bath carriers, and heavy metals [1]. Among all industrial wastewater contaminants, dyes are most stable in aqueous solutions and difficult to deal with due to their artificial origin and their complex molecular structure, which make them not easily biodegradable via normal biological wastewater purification methods. Hence, dye effluent is often purified via chemical and physical techniques like degradation (sonochemical, photochemical, biological, electrochemical) methods [2-6], coagulation and flocculation [7], membrane separation [8], activated carbon adsorption [9], electrochemical removal [10], a fento-biological treatment scheme [8], and oxidation and zonation [11]. However, none of these methods, with the exception of the adsorption process, has efficiently and effectively removed color from waste water [12, 13].Congo red is a synthetic anionic secondary diazo dye or the sodium salt of 3,3’-([1,1’-biphenyl]4,4’-diyl)bis(4-aminonaphthalene-1-sulfonic acid). CR belongs to the azo dye group derived from benzidine. It’s a water-soluble dye that yields a red colloidal solution; its solubility is better in organic solvents such as ethanol. CR has a strong affinity for cellulose fibers, which is why it was first found in the textile industry. However, the use of Congo red in the cellulose industry and other dyeing industries has long been abandoned, primarily because of its carcinogenic, mutagenic, and other associated toxicities and tendency to run and change color when touched by sweaty fingers. CR still finds application in histology to stain tissues for microscopic examination and to serve as an acidbase indicator. The artificial origin, and complex molecular structure of CR, which made it resistant to microbial or bio degradation, together with its other harmful characteristics, necessitate the quest for its total removal from effluent before disposal to the environment. Adsorption is preferred to other methods because it is an effective, low-cost technique with high treatment efficiency [14, 15]. Of all the tested adsorbents, activated carbon has been the most widely accepted due to its great adsorption capacity. Other low-cost techniques utilizing non-conventional adsorbents are being investigated due to the high running cost of activated carbon adsorption [16, 17]. Some previously explored non-conventional, cost-effective adsorbents were modified biomass [18], Eggplant biomass [19], agriculture waste [20] neem leaf powder [21], black stone cherries [22], hazelnut shell [23], rice hull [24], rattan sawdust [25], garlic peel [26], wheat bran and rice bran [27], jackfruit peel [28], spent brewery grains [29], broad bean peels [30], fly ash [31], and rice husk [32], clay mineral [33, 34, 35], red Clay [36], zeolite-based bio membrane [37], modified clays [38], among others.

Clay’s exhibition of high adsorption prowess is a result of its microstructure, which provides highly specific surface areas [39] and the prospect to adsorb polar organic particles and ions on its external surface areas. High adsorption strengths of clay can be further improved through modification [40, 41]. Adsorption and desorption of organic particles on sorbents are controlled primarily by the properties of both the sorbents and the molecules [42]. Natural clays evince negatively charged surfaces and structures, which allow them to adsorb cationic dyes and charged particles at a higher rate than anionic dyes and particles [43]. Most research on the adsorption of CR has focused on studying the influence of factors using the OFAT (one factor at a time method), which is clumpy, dawdling, and cannot appropriately predict the optimum adsorption process parameters. The limits of the OFAT can be solved by using intelligent models in evaluating the adsorption process. Response surface methodology (RSM) aids in the simultaneous analysis of process parameters that affect a process, even in the presence of complex interactions. It is used to generate a mathematical model capable of determining the optimal operating parameters for a process. It requires a few experimental runs to establish an optimum condition [44, 45]. The artificial neural network (ANN), on the other hand, makes use of biological neurons in processing information. It is used for solving complicated non-linear problems with large data sets. It adequately analyzes complex data sets that are complicated by simple mathematical methods [46]. Adaptive neuro-fuzzy inference systems (ANFIS) are a synergic artificial intelligence model that incorporates the advantages of neural networks and fuzzy logic to create a hybrid model. It can be applied to both non-linear and linear systems with a high degree of accuracy. It is based on the Takagi-Sugeno fuzzy inference system [47, 48]. The objective of this work was to study the adsorptive properties of acid-modified Ihiala clay on the removal of azo dyes (Congo red (CR)) from industrial effluent. Secondly to model an equation for predicting the optimum removal of azo dye (Congo red (CR)) dyes from an aqueous solution using artificial intelligent optimization models (artificial neural network (ANN), and adaptive neuro-fuzzy inference systems (ANFIS)) and response surface methodology (RSM) as well as presenting a comparative study of the studied models. The novelty of the research was the ability to present a comparative study of artificial intelligent models (ANFIS and ANN) and RSM for predicting the optimum point of Congo red dye adsorption on modify Ihiala clay adsorbent.

Materials And Methods

The primary raw material used is pure white clay obtained from Uzoakwa village in Ihiala Local Government Area of Anambra State, Nigeria. The secondary raw materials used were HCl (35% purity), H3PO4 (85% purity),

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

NaOH (96% purity), etc. Distilled water was used to prepare the solutions. Acid Activation of Clay Sample The clay was dried under the sun for three days to prepare the material for activation. The dried clay sample was crushed to microscale size in a laboratory mortar and pestle, and 75 µm particles were sieved out using dry test sieves. The micro-sized clay sample was contacted with H3PO4 in a beaker at a weight-to-volume ratio of 1:5 for 24 hours. The clay slurry was filtered using a separation funnel with Whatman filter paper into residue and filtrate. The obtained residue was washed until a neutral pH of 7 was obtained using distilled water. The washed clay slurry was then oven-dried for four hours at 80 degrees Celsius. The dried activated sample was crushed and sieved again to obtain particle sizes ranging from 75 to 300 µm. Characterization The functional group of the activated and raw clay samples was determined by Fourier transform infrared (FTIR) analysis, which was carried out using a Shimadzu Spectrophotometer Model S8400, with samples prepared using the conventional KBr disc method. The adsorbent surface morphologies were determined using a scanning electron microscope (SEM), specifically the Joel Scanning Electron Microscope model JSM 6400.

(1) Percentage removal was calculated via expression 2. Percentage removal (% Adsorption) =

Where 𝐶𝑒(mg/L) and 𝐶𝑜 were equilibrium and initial dye concentrations in solution, V is volume of the solution in liter (L), and m is mass of dry sorbent in gram (g). Adsorption isotherms are of paramount importance in any adsorption system because they describe the relationship between the adsorbent and the adsorbate. In this research work, four isotherm models were employed: the Langmuir [49], the Freundlich [50], the Vieth–Sladek [51], and the Dubinin–Redushkevich [52], with their expressions in non-linear forms. The parameters of each model provide vital information on the adsorption mechanisms, surface properties, and adsorbent affinity. Langmuir Adsorption Isotherm The Langmuir adsorption isotherm is a two-parameter isotherm developed to describe solid-gas-phase adsorption. This empirical model is formulated under the assumption of monolayer adsorption, where the adsorbed layer has a thickness of one molecule. Moreover, it also assumes that adsorption is limited to a restricted number of indistinguishable localized sites and that there is no correlation between the adsorbed molecules, even on adjacent sites. Based on derivation, the Langmuir isotherm relates to homogeneous adsorption as a process with uniform energies of adsorption onto the surface without migration of adsorbate in the plane of the surface (all active sites have identical affection for the adsorbate). The non-linearized form of the Langmuir model can be expressed as follows:

Batch Adsorption Studies Batch adsorption was used to investigate the impact of temperature, initial dye concentration, adsorption time, solution pH, adsorbent dosage, and adsorbent particle size on CR uptake on PIC. The stock dye solution was prepared by dissolving a specific amount of dye crystal into a specific volume of distilled water. The solution pH was adjusted using 0.1 N hydrochloric acid (HCl) or sodium hydroxide (NaOH). The dye absorbance was determined with the aid of a UV-visible spectrophotometer. The one factor at a time method (varying one variable while other variables are held constant) was employed in studying adsorbent dosage, particle size, temperature, and pH, while the two factors at a time method was employed for initial dye concentration and adsorption time.

Where qm (mg/g) is Langmuir adsorption capacity constants, KL (L/mg) is Langmuir energy of adsorption constants, Ce (mg/L) is the equilibrium concentration, qe (mg g-1) is the amount of dye adsorbed at equilibrium

Equilibrium/isotherm and Kinetics Studies Equilibrium experiments were conducted by contacting specific amount of an adsorbent into glass beakers (250ml) containing a definite volume of different concentrations of CR dye solution of the same pH level. These beakers containing the mixture were placed on shakers (magnetic stirrers) regulated at the fixed temperature level for a period of 150 minutes to ensure equilibrium was attained. Equilibrium dye concentration was determined via a UV-visible spectrophotometer. Equilibrium dye uptake “𝑞𝑒 (mg/g)” was calculated via expression 1

Freundlich Adsorption Isotherm The Freundlich adsorption isotherm is a two-parameter isotherm and the foremost known relationship that describes reversible and non-ideal adsorption. Presently, this model is applied to multilayer adsorption, with an uneven distribution of heat from adsorption over the heterogeneous surface. The Freundlich model, in the past designed for adsorption, demonstrated that the adsorbate’s ratio to a specific mass of adsorbent to the solute was not constant at varying solution concentrations, and this signifies that the quantity adsorbed is the sum total of adsorption on all sites, with sites of strong binding affinity first

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

occupied prior to equilibrium or a decrease in adsorption energy. The non-linearized form of the Freundlich model can be expressed as follows: (4) Where Ce (mg/L) is the equilibrium concentration, qe (mg g-1) is the amount of dye adsorbed at equilibrium, Kf is Freundlich adsorption capacity constants and n is the Freundlich adsorption intensity constants, 1⁄𝑛 is a measure of the surface heterogeneity Dubinin-Radushkevich Adsorption Isotherm The Dubinin-Radushkevich model is a temperature-dependent model, and it predicts the adsorption mechanism. This model was first developed for the sorption of subcritical vapors onto micropore materials by a pore filling mechanism. Typically, it is used to represent the adsorption mechanism on a heterogeneous surface using a Gaussian energy distribution. Despite having unsatisfactory asymptotic properties and failing to predict Henry’s law at low pressure, the model has frequently been successful in fitting data on high solute activity and the intermediate range of concentrations. This model was primarily used to differentiate between the chemical and physical sorption of ions through the mean free energy, E per molecule of adsorbate. The non-linearized form of the Dubinin-Radushkevich model can be expressed as follows: (5) Where qm (mg/g) is Langmuir adsorption capacity constants, Ce (mg/L) is the equilibrium concentration, qe (mg g-1) is the amount of dye adsorbed at equilibrium. Vieth-Sladek Adsorption Isotherm The Vieth-Sladek isotherm is an empirical model developed to describe, at a constant temperature, the quantity of solute adsorbed onto a solid surface as a function of the concentration of solute in solution. This model was an upgrade to the Langmuir isotherm model, which had limitations in describing multilayer adsorption processes. The Vieth-Sladek isotherm presumes that the sorption process involves a multilayer of solute molecules on the adsorbent surface instead of a monolayer, as in the Langmuir isotherm model. These characteristics make it a valuable instrument for improving and understanding the adsorption process. The model accounts for both the solute concentration in solution and the extent to which the adsorbed molecules cover the solid surface. The non-linearized form of the Vieth-Sladek model can be expressed as follows: (6)

Where qm (mg/g) is Langmuir adsorption capacity constants, Ce (mg/L) is the equilibrium concentration, qe (mg g-1) is the amount of dye adsorbed at equilibrium, 𝑘𝑣𝑠 and 𝐵𝑣𝑠 are Vieth– Sladek constants. The same procedures as in an equilibrium experiment were employed in kinetic experiments; the only difference was that the aqueous samples were taken at different time intervals in a kinetic study to determine the uptake of dye at any preset time (t). Dye uptake at time (t) “𝑞𝑡 (mg/g)” was calculated as follows. (7) Where 𝐶𝑡 is dye concentration at any time (t). Adsorption kinetics is a very vital aspect of the adsorption system, which sheds light on adsorption mechanisms and dynamics with time. In this research, seven kinetic models were employed: the Elovich [53], pseudo-first order [54], pseudo-nth-order model [55], pseudo-second order [56, 57], intra-particle diffusion [58], liquid film [59], and Boyd models [60]. The liquid film, intra-particle diffusion, and Boyd model were proposed to identify the adsorption mechanism. Pseudo-first-order Model The pseudo-first-order model, or Lagergren equation, assumes that the model is applicable over the very initial stage of the sorption process. It also assumes that the degree of change in solute uptake with respect to time is directly proportional to the difference in equilibrium concentration and the quantity of solid uptake with time. Adsorption kinetics follows a pseudo-first-order model when adsorption happens by diffusion via the interface. The non-linearized form of the pseudo-first-order model can be expressed as follows: (8) Where qe (mg/g) is the amount of adsorbate adsorbed at equilibrium, qt (mg/g) is the amount of adsorbate adsorbed at time t. K1 (min-1) is the adsorption rate constants for pseudo-first-order. Pseudo-second-order Model The pseudo-second-order kinetic model, an empirical model developed by Ho et al., assumed that the rate-determining step of the adsorption process is purely chemisorption, and this predicts all the adsorption behavior. This aforementioned assumption signifies that the adsorption rate depends more on the adsorption capacity than the adsorbate concentration. This empirical model also forecasts that the rate-determining step of the sorption process should be linked to the covalent forces due to the sharing of electrons between the metal ions and functional groups. The major edge of this empirical model over pseudo-first-order

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

is its ability to compute directly from the model the equilibrium adsorption capacity, which also puts to an end the theoretical need to determine the adsorption equilibrium capacity by experimentation. The non-linearized form of the pseudo-first-order model can be expressed as follows: (9) Where qe (mg/g) is the amount of adsorbate adsorbed at equilibrium, qt (mg/g) is the amount of adsorbate adsorbed at time t, K2 gmg-1min were the adsorption rate constants pseudo-second-order. Elovich Model The Elovich kinetic model assumes that the adsorbent surfaces are energetically heterogeneous and that neither interactions between the adsorbed ions nor desorption could affect the adsorption kinetics significantly at low surface coverage. The major effect of energetic heterogeneity surfaces on sorption equilibrium in gas-solid systems has been proven, while that of liquid-solid systems is yet to be established. The non-linearized form of the pseudo-first-order model can be expressed as follows: (10) Where qt (mg/g) is the amount of adsorbate adsorbed at time t, α is the Elovich constant for the initial adsorption rate, and β is the Elovich desorption constant. Pseudo-nth Order The pseudo-nth-order kinetic model is an empirical model that seeks to explain the kinetic analysis of order 1 and 2 kinetic parameters. This model will have a substantial effect on the computation of the rate constants since the rate constant is dependent on the reaction order. The non-linearized form of the pseudo-first-order model can be expressed as follows: (11) Where qe (mg/g) is the amount of adsorbate adsorbed at equilibrium, qt (mg/g) is the amount of adsorbate adsorbed at time t. Kn (gmg-1min) were the adsorption rate constants for pseudo-nth-order; n is the order of reaction. Adsorption Thermodynamics Thermodynamics studies the energy changes involved in the adsorption process. Thermodynamic parameters were calculated using the following equations. (12) (13) Where

(14) (15) Where ∆H is the change in enthalpy (heat of reaction), ∆Go is the standard change in Gibbs free energy and ∆S is the change in entropy. 𝐾𝑐 is the equilibrium constant, 𝐶𝑠 is the concentration of CR on adsorbent at equilibrium (mg/L), R is general gas constant (8.314J/mol/K), and T is the solution temperature in Kelvin scale. Activation Energy The size of activation energy gives insight on the nature of the adsorption i.e., whether the process is chemical or physical. To evaluate adsorption activation energy, Arrhenius rate expression was applied using the equations (16) and (17): (16) (17) Where 𝐸𝑎 is the activation energy (KJ/mol) 𝐾𝐴 is rate constant of adsorption (g/mg min), R is gas constant, and T is solution temperature in Kelvin (K) [40, 61]. 𝐸𝑎 is be obtained from the plot of 𝐼𝑛𝐾𝐴 against 1/T.

RSM Modeling RSM uses data obtained from the design of experiments to simultaneously solve multi-variant problems using statistical modeling techniques [44]. A Central composite design (CCD) with five factor levels was used to design the experiment. The factor levels can be assessed in Table S1 of the Supplementary Material. The process parameters used as independent variables were temperature, pH, concentration, and contact time, while the percentage adsorbed was the response (dependent variable). The number of experimental data sets in RSM-CCD can be obtained using equation (18) [62]. 𝑁 = 2n + 2𝑛 + 𝑛c

Where n stands for the number of input factors, 2n is the factorial points, 2n is the axial points, and nc is the center points. The total axial and factorial points were, however, tripled in order to enhance the design’s reliability. Ten center points were used. Hence, a total of eighty-two experimental data sets were evaluated. Systematic error was eliminated by carrying out the experiment at random. A second-order polynomial was used to estimate the affinity between the process variables and the response in the equation (19). (19)

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

where Y is the calculated response; βi, βii, and βij are the coefficients calculated from the second-order polynomial equation for the linear, pure quadratic, and cross products of Xi, Xi2 and Xi, Xj, respectively, while βo is the model constant. Design expert software version 13 was used in the design and analysis of the RSM-CCD. ANN Modeling The neural network toolbox of MATLAB R2015 (The Mathworks Inc.) was utilized in modeling and analyzing the prediction of the percentage of CR dye adsorbed using an artificial neural network (ANN). ANN is inspired by biological neurons, where the weighted inputs that arrive at each neuron are passed through a nonlinear activation function to produce an output signal [63]. For adequate modeling of the ANN process, the eightytwo data sets used in the RSM-CCD were tripled and used in the ANN. Hence, two hundred and forty-six data sets were used in the ANN. The data sets were divided into three categories: training, testing, and validation, with 70%, 15%, and 15%, respectively, allocated to each randomly. The ANN architecture consists of input nodes (four input parameters), hidden neuron layers, and an output node (the response). The ANN architecture is given in the Supplementary Material. The optimum number of hidden layers was obtained by iteration to avoid over-fitting and a reduction in the convergence rate [64]. Minimum mean square error (MSE) and maximum correlation coefficient (R2) were the parameters involved. Multi-Layer Perceptron (MPL), which made use of back propagation, was the learning algorithm used for the modeling. To regularize the bias value, trainlm was used as the training function. To reduce network error, the process parameters and response were normalized between 0 and 1 [65]. The process parameters and the response were normalized between 0 and 1 to reduce network error [65]. ANFIS Modeling The fuzzy logic toolbox of MATLAB R2015 (The Mathworks Inc.) was used in ANFIS modeling and prediction of optimal CR dye adsorbed. Using the fuzzy inference system (FIS) principle, the ANFIS is modeled as a five-layered network, as shown in Fig. S2 of the Supplementary Material. The first layer accepts the process variables (temperature, pH, concentration, and contact time) as inputs to the fuzzy system network. The second layer node decides the fuzzy rules, which it relays to the third layer for normalization of the activity rules The fourth layer adopts the nodes and converges the parameters, which it sends to the fifth layer as a single output layer (percentage of CR adsorbed) [66]. The same two hundred and forty-six data sets used in ANN modeling were utilized in ANFIS modeling. A hybrid method of optimization was used with zero error tolerance.

Comparative Analysis of RSM, ANN, and ANFIS Models The predicted percentage removal of CR by RSM, ANN, and ANFIS was compared to the experimental values and subjected to statistical error indices. Five statistical error functions were used. They were used to investigate the accuracy of the models in predicting the removal of CR from wastewater by the adsorption process. The significance of the model was based on smaller values of the root mean square error (RMSE), average relative error (ARE), sum of squared errors (SSE), hybrid fractional error function (HYBRID), and a higher value of the correlation coefficient (R2). Adsorbent Recycling/ Regeneration Recycling is the method of processing materials that would have otherwise been thrown away as trash into new products, which minimizes production costs as well. This phenomenon in the adsorption process is termed desorption. Desorption is simply using an eluent to create sorption equilibrium or destabilize the bond between the pollutant (bulk phase) and the adsorbing surface. Different desorption methods, including chemical regeneration, thermal regeneration, thermochemical regeneration, steam regeneration, bio-regeneration, vacuum regeneration, electrochemical regeneration, pressure swing regeneration, ozone regeneration, microwave regeneration, oxidative regeneration, and ultrasound regeneration, have been used in the regeneration or recycling of adsorbents. In this research, a chemical regeneration method was employed, and the eluents used were HCl and propanone (acetone). The use of HCl was because the pollutant to be desorbing is cationic, and therefore HCl destabilized or broke the intermolecular bond (van der Waals forces) between the pollutant (bulk phase) and the adsorbing surface, leading to adsorption, while Propanone was used because of its property to dissolve other substances without destroying their structure. Batch Desorption Exhausted adsorbent was mixed with the eluent and stirred at a specific temperature for 30 minutes per cycle until equilibrium was reached. It was then filtered into filtrate and residue adsorbent at the end of each cycle. The filtrate was for dye analysis using a UV spectrophoto2wmeter for desorption efficacy, while the residue sorbent was washed with water to remove the eluting agent, which was then oven dried and kept for reuse in a fresh process. The amount (quantity) in mg/g and efficacy in % of desorbed adsorbate (pollutant) were calculated using equations (20) and (21), respectively. Amount /quantity desorbed (mg/g)=

Where, 𝐶𝑠 is concentration of adsorbate desorbed in mg/g, 𝐶𝑎𝑑 is concentration of adsorbate adsorbed in mg/g,

M is weight of exhausted adsorbent in gram and V is volume of the eluent in liter.

Results And Discussion

Adsorbent Characterization FTIR analysis The FTIR spectra of the raw Ihiala clay and the acid-modified Ihiala clay are given in Figures 1a and 1b, respectively. Comparing the figures, a higher number of

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

peaks was observed in Fig. 1b, and this is as a result of the acid modification that removed impurities contained in the raw clay material. The peaks observed in Fig. 1b are composed of various functional groups that are responsible for the binding of CR dye, and they are amide, alkane, alcohol, and alkyne halide. SEM micrographs The scanning electron micrographs (SEM) of the raw Ihiala clay and the acid-modified Ihiala clay are shown in Figures 2a and 2b. Figure 2b displayed a greater number of heterogeneous layers of pores and the internal surface of the

Figure 1. FTIR spectra of (a) raw Ihiala clay and (b) of acid activated Ihiala clay.

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

Figure 2. SEM Micrograph of the clay (2a) before activation and (2b) after activation. clay material than Figure 2a. This observed development is a result of the modification that exposed the interior and inner surface of the clay sample to the acid particles. This increased the pores of the adsorbent, thereby making it a good adsorbent. Effects of Process Variables on CR Adsorption Effect of adsorbent particle size The effect of sorbent particle size was studied at particle size ranges of 75, 150, 300, 650, and 850 µm at a constant temperature of 303 k, an adsorbent dosage of 1 g, initial dye concentration of 100 mg/l, a time of 60 min, and a pH of 2. The result of the study, as reported graphically in Fig. 3a, shows that the percentage dye uptake decreased with sorbent increases in particle size, from 72.3% at 75 µm to 44.6% at 850 µm. This outcome proved that large particles produce a less heterogeneous layer of pores and surface area. Moreover, the disintegration of large particles opened channels and tiny cracks on the adsorbent surface and thereby increased the availability of pores for better adsorption and diffusion. The result of this study concurs with the previous work, which reported that an increase in adsorbent particle size distinctly decreased the amount of dye adsorption as well as the percentage of dye removed [67]. El-Halwany (2010) also reported that dye removal increased with the decrease in particle size [24]. Effect of adsorbent dosage Adsorbent dosage effect was investigated for the dosage range of 0.2, 0.4, 0.6, 0.8, and 1 g/100 ml of CR at a constant temperature of 303 k, an adsorbent particle size of 150 µm, an initial dye concentration of 100 mg/l, a time of 60 min, and a pH of 2. The result of the study, as reported graphically in Fig. 3b, shows that the percentage of adsorption increased with an increase in adsorbent dosage, from 45.4% at 0.2 g to 83.5% at 1 g. This significant increase in dye adsorption per increase in adsorbent dosage is due to

an increase in the surface area of active functional groups, which also increased the availability of adsorption sites [68]. The result also showed a notable decrease in quantity adsorbed per unit mass of adsorbent with an increase in dosage. This decrease in unit adsorption as dosage increases is due to adsorbent sites remaining unsaturated during the adsorption reaction. The result of this study is in agreement with the previous works, which reported that an increase in Fe-Bent dosage distinctly decreased the amount of arsenic in solution [69, 70]. Effect of concentration and contact time The effect of initial dye concentration was studied at concentration ranges of 100, 200, 300, 400, and 500 mg L-1 per 1 g adsorbent dosage at a contact time range of 2–150 minutes at a constant temperature of 323 k, an adsorbent particle size of 75 µm, and pH 2. The result as reported in Fig. 3c shows that as the rate of CR adsorbed per unit mass increased with an increase in initial ion concentration, the adsorption percentage decreased. This may be due to the low ratio of the initial number of dye molecules to the available surface area at a lower concentration. Furthermore, the result also revealed an increase in quantity of adsorption with increasing contact time at all concentrations until equilibrium was reached after 90 minutes. This is a result of the driving force supplied by the initial dye concentration to conquer the resistance to the mass transfer of dye between the solid phase and liquid phase. Finally, the result also showed that the process followed three-step processes: rapid initial adsorption (bulk diffusion), a period of slower uptake (pore diffusion and intra-particle diffusion), and a period of no significant uptake (equilibrium stage). The findings are consistent with previous work, which reported that at higher concentrations of As(V), adsorption efficiency decreased due to a decrease in available active sites [70]. Mahmoud et al. (2019) also reported that the equilibrium of the PNP adsorption on the montmorillonite clay was reached after 120 min [71].

Effect of pH The effect of pH is essential when the adsorbing molecules can ionize in response to pH. CR is a diazo/anionic dye and maintains its red color at pH 5, while below pH 2, the solution changes from red to dark blue, at pH 3, it changes to blue-violet, and the original red color is different above pH 10. The effect of pH is studied between the pH ranges of 2 and 12, at a constant temperature of 303 k, an adsorbent particle size of 150 µm, an initial dye concentration of 100 mg/l, a time of 60 min, and an adsorbent dosage of

Anfis

power when compared with the experimental data. It was closely followed by ANN. Therefore, it was concluded that ANFIS has the best generalization capacity regarding CR dye removal from wastewater. Onu et al. (2021), reported that ANFIS gave better prediction than ANN and RSM in the removal of eriochrome black-T dye [95].

Comparative Performance of the Adsorbent A comparison of the adsorption of Congo red dye with other adsorbents was made to ascertain the efficiency and relevancy of the modified clay used in this study. Maximum adsorption capacity (qmax) and removal efficiency were the parameters utilized in the comparison, as shown in Table 7.

Table 7. Performance evaluation of different clay adsorbents based on removal efficiency and adsorption capacity of the clays in removal of organic pollutants from sewage Adsorbent

Composite clay Natural clay Organo clay (H-Bt) Organo clay (B-Bt) Treated clay Cu-pillar clay Montomorillonite Bentonite clay Fe-bent Bentonite composite Pillared bentonite Bentonite/GO Modified clay (PCH) Clay Pillared clay (CrPC)

Pillared clay (AlPC) Modified red clay (MRC) formulated clay-lime

(F13 and F23) Modified zeolite Natural clay Moroccan Clay Modified clay

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

Figure 10. Effect of concentration of eluent and temperature on desorption of CR on PIC. It is obvious from the table that the adsorption capacity of the modified Ihiala clay was higher than that of some other adsorbents. Besides, Ihiala clay is readily available as a large clay deposit in Ihiala town, Nigeria. This indicates that the abundant PIC, which is readily available with a relatively cost-effective method of activation, could be utilized as an effective and alternative adsorbent in the removal of Congo red dye from contaminated water. Regeneration of Adsorbent The effects of three different process variables were examined: concentration, pH, and temperature of the eluent. From the results of Figure 10, it is observed that the highest desorption percentages of 88.1, 91.5, and 97.5 % were achieved at an eluent (HCl) concentration of 0.09 mol, pH 5, and temperature of 303 k for a specific time of 150 minutes, while the highest desorption percentage of

Conclusion

The present study established PIC as a reliable and quality adsorbent for dye uptake from an aqueous solution due to its high percentage adsorption and desorption capacities. The uptake of CR onto PIC was found to depend on the following factors: temperature, initial dye concentration, pH of solution, adsorbent particle size, contact time, and adsorbent dosage. The equilibrium adsorption data was best fitted by the Langmuir isotherm model, while the adsorption kinetics are best represented by pseudo-second-order kinetic models. The Langmuir model gave the highest maximum adsorption capacity (qm) value of 127.24 mg g-1. The ANN, ANFIS, and RSM models were found adequate in predictive modeling of the adsorption process, and the statistical analysis indicated that the ANFIS was marginally better than the ANN and RSM.

Nomenclature

Adsorption capacity, mg g-1. Adsorption capacity at equilibrium, mg g-1. Adsorption capacity at time, mg g-1. Langmuir constants, L mg-1. Dimensional separation factor Freundlich constants, L g-1. Freundlich constants Effective diffusion coefficient, m2s-1. Universal gas constants, J mol-1 K Phosphoric acid Ihiala activated clay Temperature, oC, K Dubinin-Radushkevich constant, mg g-1. Constant related to sorption energy, mol2 Mean free energy, KJ mol-1.

Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025

Co K2 Kn Cs Ce Kd ∆H ∆S ∆G Ea qm M W ∆qe HCl CR A nth t Ct K1 Kc

Initial concentration, mg L-1. Pseudo second-order kinetic constant Pseudo nth order kinetic constant Equilibrium concentration of dye on adsorbent, mg L-1 Equilibrium concentration of dye in solution, mg L-1 Intra-particle diffusion rate constant, mg g-1min1/2 Free enthalpy change, KJ mol-1 Free entropy change, J mol-1 K Free energy change, KJ mol-1. Activation energy, KJ mol-1. Maximum adsorption capacity Total mass of the adsorbent, g Weight of adsorbent Normalized standard deviation Hydrochloric acid Congo red Elovich constants Order of reaction Time (min) Concentration at time t, mg L-1. Pseudo first order kinetic constant Equilibrium constant

Acknowledgement

The authors wish to appreciate Mr. Emmanuel Okafor of English department of St. John School Awka for proof-reading the manuscript.

Data Availability Statement

The authors confirm that the data that supports the findings of this study are available within the article. Raw data that support the finding of this study are available from the corresponding author, upon reasonable request.

Conflict Of Interest

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics

There are no ethical issues with the publication of this manuscript.

References

1. Noroozi B, Soria GA, Bahrami IH, Arami M. Congo Red dye from simulated wastewater using Equilibrium and kinetic adsorption study of a cat- activated carbon derived from corn cobs; kinetics ionic dye by natural adsorbent-silkworm pupa. J and equilibrium studies. Glob J Pure Appl Chem Hazard Mater 2007;139:167–174. [CrossRef] Res 2023;11:1–19. [CrossRef] 336 Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025
2. Tabak A, Eren E, Afsin B, Caglar B. Determination of adsorptive properties of a Turkish sepiolite for removal and isotherm studies of a cationic dye using agri- of Reactive Blue 15 anionic dye from aqueous solu- cultural waste: Broad bean peels. J Hazard Mater tions. J Hazard Mater 2009;161:1087–1094. [CrossRef] 2008;154:639–648. [CrossRef]
3. Akar ST, Recep U. Untreated clay with high adsorp- tion capacity for effective removal of C.I. Acid Red Kinetic study and equilibrium isotherm analysis of 88 from aqueous solutions: Batch and dynamic Congo Red adsorption by clay materials. J Chem flow mode studies. J Chem Eng 2010;162:591–598. Eng 2009;148:354–364. [CrossRef] [CrossRef] [32] Onu CE, Oguanobi NC, Okonkwo CO, Nnamdi-
4. Mohammed GH, Magdy AW, Hosni AG, Ahmed Bejie J. Application of modified agricultural waste in SE. Adsorption of Rose Bengal dye from wastewa- the adsorption of Bromocresol Green dye. Asian J ter onto modified biomass. Sci Rep 2023;13:14776. Chem Sci 2020;7:15–24. [CrossRef] [CrossRef] [33] Juraj B. Controversial issues related to dye adsorp-
5. Zhi L, Xiaohai H, Yi M, Bing G, Yiling S, Jianghui tion on clay minerals: A critical review. Molecules Z, Soon HT. Eggplant biomass-based porous carbon 2023;28:6951. [CrossRef] for fast and efficient dye adsorption from wastewa- [34] El Kerdoudi Z, Bensalah J, Ferraa N, El Mekkaoui A, ter. Ind Crops Prod 2022;187:115510. [CrossRef] Berisha A, Safi Z, et al. Physicochemical characteri-
6. Dal SK, Meenal G. Treatment of textile dyeing efflu- zation of clay and study of cationic Methylene Blue ent using agriculture waste-based adsorbents: A dye adsorption. ACS Omega 2023;8:40848–40863. review. Chem Sci Int J 2022;32. [CrossRef]
7. Bhattacharyya KG, Sharma A. Kinetics and thermo- dynamics of Methylene Blue adsorption on Neem dye using natural clay from water. J Environ Eng Sci (Azadirachta indica) leaf powder. Dyes Pigments 2022;17:175–183. [CrossRef] 2005;65:51–59. [CrossRef] [36] Muhammed O, Çiğdem SÖ. Equilibrium studies for
8. Arana JMRR, Mazzoco RR. Adsorption studies of dye adsorption onto red clay. MTU J Eng Nat Sci Methylene Blue and phenol onto black stone cher- 2022;3:36–45. ries prepared by chemical activation. J Hazard Mater [37] Sabarish R, Jasila K, Aswathy J, Jyotishkumar P, 2010;180:656–661. [CrossRef] Jaewoo L, Jyothi Mannekote S, et al. Adsorption of
9. Doğan M, Abak H, Alkan M. Adsorption of anionic dye onto ZSM-5 zeolite-based bio mem- Methylene Blue onto hazelnut shell: Kinetics, mech- brane: Characterizations, kinetics and adsorption anism and activation parameters. J Hazard Mater isotherm. J Polym Environ 2022. Preprint. Doi: 2009;164:172–181. [CrossRef] 10.21203/rs.3.rs-1191111/v1 [CrossRef]
10. El-Halwany MM. Study of adsorption isotherms and kinetic models for Methylene Blue adsorption on H. Adsorption behaviors of modified clays prepared activated carbon developed from Egyptian rice hull with structurally different surfactants for anionic (Part II). Desalination 2010;250:208–213. [CrossRef] dyes removal. Environ Eng Res 2023;28:210076.
11. Hameed BH, Ahmad AL, Latiff KNA. Adsorption of basic dye (Methylene Blue) onto activated car- ous solutions using activated carbon: External mass bon prepared from rattan sawdust. Dyes Pigments transfer processes. J Chem Technol Biotechnol 2007;75:143–149. [CrossRef] 1983;33A:205–218. [CrossRef]
12. Hameed BH, Ahmad AL. Batch adsorption of Methylene Blue from aqueous solution by garlic of a dye (Crystal Violet) on acid-modified non-con- peel, an agricultural waste biomass. J Hazard Mater ventional adsorbent. J Chem Technol Metall 2009;164:870–875. [CrossRef] 2019;54:95–110.
13. Xue SW, Jing PC. Biosorption of Congo Red from aqueous solution using wheat bran and rice bran: of Methylene Blue dye from aqueous solution Batch studies. Sep Sci Technol 2009;44:1452–1466. using modified Ngbo clay. J Mater Sci Res Rev [CrossRef] 2020;5:33–46.
14. Hameed BH. Removal of cationic dye from aqueous solution using Jackfruit peel as non-conventional Hsieh MF. Adsorption of acid dye onto activated low-cost adsorbent. J Hazard Mater 2009;162:344– carbons prepared from agricultural waste bagasse
15. [CrossRef] by ZnCl₂ activation. Chemosphere 2001;45:51–58.
16. Jaikumar V, Shathish Kumar K, GnanaPrakash D. [CrossRef] Biosorption of acid dyes using spent brewery grains: [43] Mahammedi F, Benguella B. Adsorption of Characterization and modeling. Int J Appl Sci Eng Methylene Blue from aqueous solutions using nat- 2009;7:115–125. ural clay. J Mater Environ Sci 2016;7:285–292. Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025 337
17. Venkatesh PM, Karthikeyan R. Comparative stud- ies on modeling and optimization of hydrodynamic bon from solutions. J Sanit Eng Div 1963;89:31–39. parameters on inverse fluidized bed reactor using [CrossRef] ANN-GA and RSM. Alexandria Eng J 2018;57:3019– [59] Spahn H, Schlunder U. The scale-up of activated car- 3032. [CrossRef] bon columns for water purification based on results
18. Onu CE, Igbokwe PK, Nwabanne JT, Nwanjinka OC, from batch tests. I. Theoretical and experimental Ohale PE. Evaluation of optimization techniques in determination of adsorption rates of single organic predicting optimum moisture content reduction in solutes in batch tests. Chem Eng Sci 1975;30:529– drying potato slices. Artif Intell Agric 2020;4:39–47. 537. [CrossRef] [CrossRef] [60] Boyd GE, Adamson AW, Meyers LS Jr. The exchange
19. Betiku E, Osunleke AS, Odude VO, Bamimore A, adsorption of ions from aqueous solutions by Oladipo B, Okeleye AA, et al. Performance eval- organite-zeolites. II. Kinetics. J Am Chem Soc uation of adaptive neuro-fuzzy inference system, 1947;69:2836. [CrossRef] artificial neural network and response surface meth- [61] Nwabanne JT, Okpe EC, Asadu CO, Onu CE. odology in modeling biodiesel synthesis from palm Sorption studies of dyestuffs on low-cost adsorbent. kernel oil by transesterification. Biofuels 2018;1–17. Asian J Phys Chem Sci 2018;5:1–19. [CrossRef] [CrossRef] [62] Arulkumar M, Sathishkumar P, Palvannan T.
20. Gonzalez CRT, Subathra MSP, Manoj NKN. Optimization of Orange G dye adsorption by acti- Modeling the daily reference evapotranspiration in vated carbon of Thespesia populnea pods using semi-arid region of South India: A case study com- response surface methodology. J Hazard Mater paring ANFIS and empirical models. Inf Process 2011;186:827–834. [CrossRef] Agric 2020;2–12. [63] Manpreet SB, DhritiKapoor RK, Kalia ASR, Ashwani
21. Najafi B, Faizollahzadeh Ardabili S. Application of KT. RSM and ANN modeling for electrocoagula- ANFIS, ANN, and logistic methods in estimating tion of copper from simulated wastewater: Multi- biogas production from spent mushroom compost objective optimization using genetic algorithm (SMC). Resour Conserv Recycl 2018;133:169–178. approach. Desalination 2011;274:74–80. [CrossRef] [CrossRef] [64] Mingyi F, Tongjun L, Jiwei H, Rensheng C, Xionghui
22. Langmuir I. The constitution and fundamental W, Xuedan S, et al. Artificial neural network mod- properties of solids and liquids. J Am Chem Soc eling and genetic algorithm optimization for cad- 1916;38:2221–2295. [CrossRef] mium removal from aqueous solutions by reduced
23. Freundlich H. Over the adsorption in solution. J graphene oxide-supported nanoscale zero-valent Phys Chem 1906;57:385–470. [CrossRef] iron (nZVI/rGO) composites. Materials (Basel)
24. Vieth WR, Sladek KJ. A model for diffusion in a glassy 2017;10:1–22. [CrossRef] polymer. J Colloid Sci 1965;20:1014–1033. [CrossRef] [65] Mourabet ME, Rhilassi A, Bennani-Ziatni M, Taitai
25. Dubinin MM, Radushkevich LV. Equation of the A. Comparative study of artificial neural network characteristic curve of activated charcoal. Proc Acad and response surface methodology for modeling Sci Phys Chem Sect USSR 1947;55:331–333. and optimization of the adsorption capacity of fluo-
26. Elovich SY, Larinov OG. Theory of adsorption from ride onto apatitic tricalcium phosphate. Univ J Appl solutions of non-electrolytes on solid: (I) Equation Math 2014;2:84–91. [CrossRef] adsorption from solutions and the analysis of its [66] Bahman N, Sina FA. Application of ANFIS, ANN, simplest form, (II) Verification of the equation of and logistic methods in estimating biogas produc- adsorption isotherm from solutions. Izv Akad Nauk tion from spent mushroom compost (SMC). Resour SSSR Otd Khim Nauk 1962;2:209–216. [CrossRef] Conserv Recycl 2018;133:169–178. [CrossRef]
27. Ritchie AG. Alternative to the Elovich equation for bents. Chem Eng J 2012;180:91–98. [CrossRef] the kinetics of adsorption of gases on solids. J Chem [68] Ladhe UV, Wankhede SK, Patil VT, Patil PR. Soc Faraday Trans 1 Phys Chem Condens Phases Adsorption of Eriochrome Black-T from aqueous 1977;73:1650–1653. [CrossRef] solution on activated carbon prepared from Mosambi
28. Ho YS, Wase DAJ, Forster CF. Removal of lead ions peel. J Appl Sci Environ Sanit 2011;6:149–154. from aqueous solution using sphagnum moss peat [69] Ismat AE, Marufa K, Most Afroza K, Owaleur RM, as adsorbent. Water SA 1996;22:219–224. Anis-Ul-Haque KM, Jegede MM, et al. Sequestration
29. Ho YS, McKay G. Pseudo second-order model for of hazardous dyes from aqueous solution using sorption process. Process Biochem 1999;34:451– raw and modified agricultural waste. Adsorpt Sci
30. [CrossRef] Technol 2021;2021:6297451. [CrossRef] 338 Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025
31. Barakan S, Aghazadeh V, Samiee BA, Mohammadi S. Thermodynamic, kinetic, and equilibrium isotherm from aqueous solution and industrial effluent using studies of As(V) adsorption by Fe(III)-impregnated natural and agricultural waste. Colloids Surf B bentonite. Environ Dev Sustain 2020;22:5273–5295. 2013;107:77–106. [CrossRef] [CrossRef] [83] Kausar A, Bhatti H, Mackinnon G. Equilibrium
32. Mahmoud EO, Mohamed L, Hicham AO, Younes kinetic and thermodynamics studies on the removal B, Abdelhadi A, Abdelaziz E, et al. Efficient removal of U(VI) by low-cost agricultural waste. Colloids of p-nitrophenol from water using montmorillonite Surf B 2013;111:124–133. [CrossRef] clay: Insights into the adsorption mechanism, pro- [84] Malash GF, El-Khaiary MI. Methylene Blue adsorp- cess optimization, and regeneration. Environ Sci tion by the waste of Abu-Tartour phosphate rock. J Pollut Res 2019;26:19615–19631. [CrossRef] Colloid Interface Sci 2010;348:537–545. [CrossRef]
33. Imessaoudene A, Cheikh S, Hadadi A, Hamri N, Bollinger JC, Amrane A, et al. Adsorption perfor- modynamics and isosteric heat of adsorption of toxic mance of zeolite for the removal of Congo Red dye: metal ion onto bagasse fly ash (BFA) and rice husk Factorial design experiments, kinetic, and equilib- ash (RHA). Chem Eng J 2007;132:267–278. [CrossRef] rium studies. Separations. 2023;10:57. [CrossRef] [86] Cheung WH, Szeto YS, McKay G. Intraparticle dif-
34. Adebayo MA, Jabar JM, Amoko JS, Openiyi EO, fusion processes during acid dye adsorption onto Shodiya OO. Coconut husk-raw clay-Fe compos- chitosan. Bioresour Technol 2007;98:2897–2904. ite: Preparation, characteristics, and mechanisms [CrossRef] of Congo Red adsorption. Sci Rep 2022;12:14370. [87] Cáceres-Jensen L, Rodríguez-Becerra J, Parra- [CrossRef] Rivero J, Escudey M, Barrientos L, Castro-Castillo V.
35. Reddy SMC, Sivaramakrishna L, Reddy VA. The use Sorption kinetics of Diuron on volcanic ash-derived of an agricultural waste material, Jujba seeds, for the soils. J Hazard Mater 2013;261:602–613. [CrossRef] removal of anionic dye (Congo Red) from aqueous [88] Hu Z, Chen H, Ji F, Yuan S. Removal of Congo Red medium. J Hazard Mater 2012;203–204:118–127. from aqueous solution by cattail root. J Hazard [CrossRef] Mater 2010;173:292–297. [CrossRef]
36. Popoola TJ, Okoronkwo AE, Oluwasina OO, Adebayo MA. Preparation, characterization, and Zurina ZA. Batch adsorption of basic dye using application of homemade graphene for the removal acid-treated kenaf fiber char: Equilibrium, kinetic, of Congo Red from aqueous solutions. Environ Sci and thermodynamic studies. Chem Eng J 2011;181– Pollut Res 2021;28:52174–52187. [CrossRef] 182:449–457. [CrossRef]
37. Lafi R, Montasser I, Hafiane A. Adsorption of Congo Red dye from aqueous solutions by prepared nanotubes: Equilibrium, kinetics, and thermody- activated carbon with oxygen-containing functional namics. J Hazard Mater 2007;144:93–100. [CrossRef] groups and its regeneration. Adsorpt Sci Technol [91] Amini A, Younesi H, Bahramifar N, Lorestani A, 2019;37:160–181. [CrossRef] Ghorbani F, Daneshi A, Sharifzadeh M. Application
38. Mahmoodi M, Hayati MB, Arami M, Lan C. of response surface methodology for optimiza- Adsorption of textile dyes on pine cone from col- tion of lead biosorption in an aqueous solution by ored wastewater: Kinetic, equilibrium, and thermo- Aspergillus niger. J Hazard Mater 2008;154:694– dynamic studies. Desalination 2011;268:117–125. 702. [CrossRef] [CrossRef] [92] Kim HK, Kim JG, Cho JD, Hong JW. Optimization
39. Salleh MAM, Mahmoud DK, Karim WA, Idris A. and characterization of UV-curable adhesive for Cationic and anionic dye adsorption by agricultural optical communication by response surface meth- solid waste: A comparative review. Desalination odology. Polym Test 2003;22:899–906. [CrossRef] 2011;280:1–13. [CrossRef] [93] Onu CE, Nwabanne JT. Application of response sur-
40. Lata H, Garg V, Gupta R. Removal of basic dye from face methodology in Malachite Green adsorption aqueous solution by adsorption using Parthenium using Nteje clay. Open J Chem Eng Sci 2014;1:19–33. hysterophorus: An agricultural waste. Dyes [94] Iheanacho CO, Nwabanne JT, Onu CE. Optimum Pigments 2007;74:653–658. [CrossRef] process parameters for activated carbon production
41. Ho YS, Porter JF, McKay G. Equilibrium isotherm from rice husk for phenol adsorption. Curr J Appl studies for the sorption of divalent metal ions onto Sci Technol 2019;36:1–11. [CrossRef] peat: Copper, nickel, and lead single-component [95] Onu CE, Nwabanne JT, Ohale PE, Asadu CO. system. Water Air Soil Pollut 2002;141:1–33. Application of ANN and RSM techniques in opti-
42. Ho YS, Wase DAJ, Forster CF. Removal of lead ions mal parameter evaluation for turbidity removal from aqueous solution using sphagnum moss peat from abattoir effluent using valorized chicken bone as adsorbent. Water SA 1996;22:219–224. coagulant. S Afr J Chem Eng 2021;36:24–42. Sigma J Eng Nat Sci, Vol. 43, No. 1, pp. 316−339, February, 2025 339
43. Foroutan R, Mohammadi R, Adeleye AS, Farjadfad S, Esvandi Z, Arfaeinia H, et al. Efficient arsenic(V) lared clays. Proceedings of the 9th World Congress removal from contaminated water using natural clay on New Technologies (NewTech’23). 2023;Paper No. and clay composite adsorbents. Environ Sci Pollut ICEPR 138. Res 2019;26:29748–29762. [CrossRef] [105] Carazeanu PI, Roscaa I, Dumbravaa A. Modified red
44. Paul ED, Mutsee T. Treated clay mineral as adsor- clays as adsorbents in the removal of cationic dyes bent for the removal of heavy metals from aqueous from aqueous solutions. Dig J Nanomater Biostruct solution. Appl Sci Eng Prog 2021;14.
45. Georgescu AM, Nardou F, Zichil V, Nistor ID. 2023;18:567–578. [CrossRef] Adsorption of lead(II) ions from aqueous solutions [106] Fumba G, Essomba JS, Ankoro NO, Kouotou D, onto Cr-pillared clays. Appl Clay Sci 2018;152:44– Bélibi BPD, Ndi JN, Ketcha MJ. CI Acid Orange 52
46. Ahmadi A, Foroutan R, Esmaeili H, Tamjidi S. The role materials: Isotherm, kinetic, and thermodynamic of bentonite clay and bentonite clay@MnFe₂O₄ com- studies. J Mater Sci Chem Eng 2023;11. [CrossRef] posite and their physico-chemical properties on the [107] Laura AP, Adrián B, María EG, Yenetzi V, Ana GC, removal of Cr(III) and Cr(VI) from aqueous media. María SB. Adsorption of arsenic, lead, cadmium, Environ Sci Pollut Res 2020;27:14044–14057. [CrossRef] and chromium ions from aqueous solution using
47. Chang YS, Au PI, Mubarak NM, Khalid M, Jagadish a protonated chabazite: Preparation, characteriza- P, Walvekar R, Abdullah EC. Adsorption of Cu(II) and Ni(II) ions from wastewater onto bentonite and tion, and removal mechanism. Adsorpt Sci Technol bentonite/GO composite. Environ Sci Pollut Res 2023;2023:2018121. [CrossRef] 2020;27:33270–33296. [CrossRef] [108] Fuad HSR, Dler MSS, Stephan K. Adsorption study
48. Barakan S, Aghazadeh V. Structural modification of and removal of Basic Fuchsin dye from medical lab- nano bentonite by aluminum, iron pillarization, and oratory wastewater using local natural clay. Adsorpt 3D growth of silica mesoporous framework for arse- Sci Technol 2023;2023:9398167. [CrossRef] nic removal from gold mine wastewater. J Hazard [109] Abdelfattah EM, Yassine E, Souad Z, Issam M. Mater 2019;378:120779. [CrossRef] Elaboration and characterization of Organo-
49. Darlyson TG, Liana MRM, Luiz BdSS, Edy SB, Ghassoul (Moroccan clay) as an adsorbent using Enrique V, Enrique R, et al. Purification of anthocya- cationic surfactant for anionic dye adsorption. Phys nins (Brassica oleracea var. Rubra) from purple cab- Chem Res 2023;11:913c928. bage using natural and modified clays as adsorbent. Adsorpt Sci Technol 2023;2023:2724122. [CrossRef] [110] Wang G, He D, Fengchun ZF, Hu J, Lee Y, Shi J, Xu J.
50. Metin A, Ahmet G, Kübra G, Elif Ş. Adsorption Extraction and purification of ustiloxin A from rice of textile dyes from aqueous solutions onto clay: false smut balls by a combination of macroporous Kinetic modeling and equilibrium isotherm analy- resin and high-speed countercurrent chromatogra- sis. Front Chem 2023;11. [CrossRef] phy. Food Prod Process Nutr 2020;2:29. [CrossRef]

Share and Cite