Multi-Metal Adsorption of Lead (II), Cadmium (II), and Manganese (II) Ions from Simulated Solution onto HDTMA-Br Modified Dijah-Monkin Bentonite Clay

.


INTRODUCTION
A healthy and rising human population necessitates the availability of drinking water.However, the widespread use of chemicals and increased industrialisation steadily increase the burden of toxins in drinking water [1].Heavy metals are among the most common pollutants found in wastewater, and even at low concentrations, they can be hazardous to humans and animals [2].Managing pollution caused by heavy metals has recently become a significant issue, particularly in developing countries.
Heavy metals are toxic or poisonous metals and metalloid elements with a relatively high density of 3.5 to 7 g/cm 3 .Many such metals are present extensively in the earth's crust and are non-biodegradable [3].Air, food and water are absorbed into the human body [2].Heavy metals are among the most prevalent pollutants identified in wastewater.
However, given the challenge of preventing the drainage of these pollutants into water supplies, implementing effective purification technologies is a relatively efficient way to maintain safer/cleaner bodies of water.Multiple treatment methods have been studied for water purification, such as using clay/modified clay, biological or inert content (bio-sorbent) for sorption, flocculation-coagulation by various chemical agents, etc. [4].Adsorption is preferred among many techniques because it is cost-effective, highly accurate, energy-efficient, versatile, and can quickly recycle spent adsorbent [5].Heavy metals differ widely in their chemical properties and are used extensively in electronics, machines and the artefacts of everyday life, as well as in high-tech applications.As a result, they can enter into the aquatic and food chains of humans and animals from various anthropogenic sources and the natural geochemical weathering of soil and rocks [2].Heavy metals in wastewater include lead, cadmium, arsenic, mercury, zinc, chromium, nickel, copper, vanadium, platinum, silver, gold, selenium and uranium [6].Nearly all of these metals have no known biological function in living organisms.At the same time, some, like copper, zinc, and nickel, are considered essential below 0.005 mg/l but are toxic when increased beyond the permissible unit [7].
Cadmium is one of the most toxic elements in the food chain and is the cause of kidney disorders and bone disease.Thus, cadmium is not essential for biological systems.It is used to manufacture nickel-cadmium batteries, plastics and pigments.Phosphate fertilisers and waste dumping are both routes for cadmium transference into the environment.Concern regarding the role and toxicity of cadmium in the environment is increasing because it can be highly toxic to humans and animals.Smoking cigarettes is one of the sources of cadmium poisoning in humans [2].The permissible limit for cadmium in drinking water is 0.003 mg/l [8].
Lead is another toxic heavy metal widely used in industrial applications such as batteries, printing, pigments, fuels, photographic materials, and explosives.However, lead can replace calcium in the skeletal system and accumulate.Exposure to lead (II) results in various neurodevelopmental effects, cardiovascular diseases, impaired renal function and fertility, hypertension, and adverse pregnancy outcomes [9].The admissible limit for lead in drinking water is 0.01 mg/l [8].
Furthermore, Manganese (II) ion is the twelfth most abundant element and fifth most abundant metal on the earth.This metal has a silver-grey colour and is very easy to oxidise.Thus, Manganese (II) is not a free element; it usually exists as oxides, carbonates, and silicates.It is absorbed by ingestion, inhalation and dermal permeation and administered intravenously.It is rapidly absorbed in the gastrointestinal tract and lungs and then distributed into different tissues through blood circulation.The liver, pancreas, bone, kidney and brain are the organs containing the highest manganese levels in the human body [10].The permissible limit for manganese (II) ions is 0.4 mg/l [8].Adsorption using bentonite clay has been proven over the years to be a preferred method in water and wastewater treatment because of the simplicity of the process, coupled with the abundantly cheap supply of adsorbent, as well as the excellent properties which translate to outstanding and efficient removal of target pollutants [11].Bentonites can adsorb a wide range of water pollutants due to their micron-size particles' large external and internal surface area [10].The adsorption capacities of clay minerals have experienced significant improvement due to modification with quaternary ammonium compounds (QACs) [12].Organoclays have progressively been used as adsorbents for heavy metal removal since they are cheaper than commercially available materials like zeolites and activated carbon [13].The modification method involves altering the surface polarity of the clay minerals by intercalation with cationic surfactants [14,15].
Bentonite clay offers an attractive and inexpensive option for removing inorganic and organic pollutants.The use of natural and modified bentonite for the adsorption of several heavy metals has drawn attention over the years [16].The adsorption capacities of bentonite clay activated using different activation techniques are shown in Table 1.Adsorption, like many other water and wastewater treatment methods, is a multi-factor dependent system whose efficiency can only be achieved when the independent variables such as pollutant load (usually in terms of initial concentration of the adsorbate (s)), adsorption time, temperature, dosage, and adsorbent particle size are chosen under favourable conditions.While many studies have been published on process optimisation using most of the listed criteria, the effect of particle size on the efficiency of surface processes has received less attention.
Response surface methodology is an optimisation tool that is very effective in determining the best conditions for a variety of processes, including adsorption [17,18,19,20], chemical reaction integrated separation processes [21,22,23] and many others.
Therefore, this experiment was conducted not only to modify Dijah-Monkin bentonite clay with HDTMA-Br but also to determine the optimal particle size of the adsorbent, the initial concentration of each Pb, Cd, and Mn coexisting in the same solution, the adsorption time for maximum heavy metals removal efficiency, and the adsorption capacity of the synthesised adsorbent.The model solution for this study was synthesised using a known concentration of the considered metals.

Preparation and Modification of Dijah-Monkin
Bentonite Clay.This work used raw bentonite clay collected from the deposit site at Dijah-Monkin town in Zing LGA, Taraba State of Nigeria, subjected to preliminary treatment, beneficiated, modified with cationic surfactant HDTMA-Br, and characterised [24].
Adsorption Studies.All chemicals used are analytical-grade reagents.The multi-metal solution was prepared according to a procedure given by [16] to obtain 0.0048, 0.0089 and 0.0223 mole of lead (II), cadmium (II) and manganese (II) ion, respectively.The stock solutions were further diluted to desired concentrations of 10, 30 and 50 mg/l to mimic heavy metals concentrations in industrial effluents.The multi-metal solution (lead, cadmium and manganese) was obtained by mixing equal volumes of the known concentrations of the individual stock solutions where the particular concentration of the metals in the multi-metal solution retained their concentration as con-firmed by atomic absorption spectrophotometer (Buck scientific model: VGB 210).
The sorption process involves mixing 50 mg of HDTMA-Br modified bentonite clay of selected particle size ranges in 50 ml of the multi-metal solution according to the design matrix.The supernatant was filtered using 110 mm diameter Whatman filter paper and analysed for metal ion residual concentration using the atomic absorption spectrometer (Buck Scientific model: VGB 210).
where C 0 (mg/l) and C e (mg/l) are the initial and final (equilibrium) concentrations, respectively, V (l) is the volume of the solution treated, and m (g) is the mass of the adsorbent.

Design of Experiment with Response Surface Methodology (RSM). Design of experiment (DOE)
is a well-accepted statistical technique that can design and optimise the experimental process.It involves choosing the optimal experimental design and estimating the effect of several variables independently and simultaneously interacting.
Response surface methodology (RSM) is a statistical method used for experimental modelling and analysing the relationship between the input and response variables [26,27,28].This study selected three process variables, particle size, initial metal ion concentration and contact time, to study the effect on adsorption capacity and percentage removal.Adsorption experiments were performed at selected level factors: particle size ranges of 50-150 μm, contact time of 30-180 minutes, and initial multi-metal concentration of 10-50 mg/l, as shown in Table 2.
Analysis of variance (ANOVA) was used to analyse regression coefficients, prediction equations, and case statistics.Diagnostics Plots and model graphs were obtained using Design Expert v.6.0.8 (Trial) to examine the effects of variables individually and their interactions to determine their optimum level.The point prediction method was used to optimise the levels of each variable for maximum response.

RESULTS AND DISCUSSION
Characterisation and Structural Formula of Modified Bentonite Clay.Oloyede et al. (2021) reported the XRD, FTIR, SEM, and XRF of the modified Dijah-Monkin bentonite clay.The XRF data were used to calculate the structural formula for the natural bentonite clay using the method presented by [24].The calculation is presented in Table 3.The structural formulas were in close agreement with the general molecular formula for the montmorillonite group, with silicon occupying all the tetrahedral sites.In contrast, aluminium occupied two-thirds of the octahedral sites.Potassium (K) and calcium (Ca) are exchangeable, and superscript "iv" is used to indicate tetrahedrally coordinated ions, while superscript "vi" is for octahedrally coordinated ions [12].The general molecular formula of the montmorillonite group is (Ca, Na, H)(Al, Mg, Fe, Zn)2 (Si, Al)4O10(OH)2-xH2O.
The structural formula of the modified clay revealed the reduction of aluminium ions in the tetrahedral group.Furthermore, the exchangeable interlayer cations, such as calcium and potassium ions, also experienced a decrease, indicating more available sites for adsorption.The one-layer structural formula for modified clay in this study is [(Al 3+ 2.99 Ti 4+ 0.07 Fe 3+ 0.61 Mn 2+ 0.02) -0.88 (Si 4+ 8.12 Al 3+ 0.12) -0.12 O20(OH)4] -1.00 (Ca 2+ 0.14 K + 0.47) +0.61   Box-Behnken Design (BBD).Table 4 presents the results of the responses (adsorption capacities and removal percentages) under different conditions (particle size, initial concentration, and contact time) from the design matrix using BBD.
The lead adsorption capacity (LAC) experimental result in Table 4 showed that the highest adsorption for lead removal was 49.98 mg/g at serial run 13.The lowest was 9.81 mg/g at serial run 10, while the lead percentage removal (LPR) was highest at 100% at serial run four and lowest at 98.07% at serial run 10.
The cadmium adsorption capacity (CAC) experimental result showed that the highest adsorption for cadmium removal was 49.41 mg/g at serial run 16, and the lowest was 9.77 mg/g at serial run 7.In contrast, the cadmium percentage removal (CPR) was highest 99.96% at serial run 13 and lowest 98.70% at serial run 10.The manganese adsorption capacity (MAC) experimental result showed that the highest adsorption for manganese removal was 49.96 mg/g at serial run 5.The lowest was 9.86 mg/g at serial run 10, while the manganese percentage removal (MPR) was highest 99.93% at serial runs 3 and 5; and lowest 98.60% at serial run 10.Many authors have also used CCD to analyse the adsorption data critically.Obtaining various values of responses at different levels of the factor combination in the design matrix indicated that the chosen 3-factor BBD adequately represented the adsorption system.In a recent investigation, CCD findings were obtained for pH, adsorbent dose, and initial concentration, with the maximum experimental removal of Cd 2+ and Pb 2+ being 98.90% and 99.99%, respectively [29].

Modelling and analysis of variance of adsorption capacity.
To establish a mathematical relationship between each of the responses and the factors, the Fit Summary of the Design Expert 6.0.8 was used.The lead adsorption capacity ANOVA presented in Table 5 shows that the Model Fvalue is 1697439.16,which implies the statistical significance of the model with chances of error as low as 0.01%.
Also, based on a 95% confidence level, which gives an allowable error value to be equal to or less than 5%, the significant terms for the linear model developed are A and B. Values for the standard deviation (std.), mean, and correlation value (CV) were obtained to be 0.025, 29.92 and 0.084 respectively.Another coefficient of determination (R 2 ), adjusted R 2 , predicted R 2 and adequate precision were also obtained to be 1.000, 1.000, 1.000 and 3293.114,respectively.The values of these parameters indicate the accuracy of the model.
Similarly, ANOVA results showed that the model developed for cadmium adsorption capacity was statistically significant with an F-value of 194975.95 and a p-value accounting for more than 99% confidence level.Moreover, these terms significantly affected the model because the p-values obtained for B, C, and B 2 were far less than the allowable probability of error (5%).
Values for the std., mean, and CV were obtained to be 0.042, 29.64 and 0.14, respectively.In addition, the R 2 , adjusted R 2 , predicted R 2 and adequate precision were also obtained to be 1.000, 1.000, 1.000 and 1224.602,respectively.Accordingly, Pb, Cd, and Mn adsorption capacity data were respectively fitted to linear, quadratic, and linear models.After that, the obtained models were analysed for variance.
The final empirical model in terms of coded factors for LAC, CAC and MAC is given in Eqs. ( 3)-( 5) respectively.A positive sign in front of the terms indicates a synergistic effect, whereas a negative sign indicates an antagonistic effect.The quality of the model developed was evaluated based on the correlation coefficient value.The R 2 value for these equations is 1.0000.7 shows that the Model Fvalue was 27.18 for lead, which implied the model was significant with 0.01 % probability of error.A, B, B2, C2 and AB were the only considerable model terms in this model.Values for the std, mean, and CV were obtained as 0.081, 99.60 and 0.082, respectively.The R 2 , adjusted R 2 , predicted R 2 and adequate precision were obtained as 0.9722, 0.9364, 0.8145 and 19.324, respectively.
The cadmium percentage removal ANOVA suggested a Model F-value of 5.07, which implies a significant model.In this case, C and B 2 are the only considerable model terms.Values for the std., mean, and CV were obtained to be 0.21, 98.77 and 0.21, respectively.The R 2 , adjusted R 2 , predicted R 2 and adequate precision were obtained as 0.8669, 0.6958, 0.7438 and 1224.602,respectively.The standard deviation reflects the degree of deviation of the experimental from the actual value.
The software suggested the quadratic model for LPR and CPR, while a two-factor interaction was suggested for MPR.The heavy metals percentage removal data were modelled based on the Fit Summary suggestions.Subsequently, analyses of variance for each of the models were carried out.
The final empirical model in terms of coded factors for LPR, CPR, and MPR is given in Eqs. ( 6)-( 8), respectively.
Furthermore, the 3D view for the LPR depicts the relationship between initial concentration and particle size at a constant time of 105 minutes, shown in Figure 10.The model graph for the LPR shows that the increase in the initial concentration resulted in a higher percentage removal, while a higher particle size also increases the percentage removal.A slight increase in the per-centage removal was observed, with the percentage removal approaching equilibrium as the concentration increased.In contrast, the increase in the particle size linearly enhanced the percentage removal.
The 3D view for the CPR, presented in Figure 11, shows that the percentage removal increased at lower concentrations and sharply declined as the concentration increased.Furthermore, the per-centage removal reduced at shallow particle size and later increased as the particle size crossed 100 µm.
However, the 3D view of the model graph for the MPR, presented in Figure 12, shows that an increase in the initial concentration of manganese results in a higher percentage removal and increases steadily with an increase in the particle size.
Notes: *LAC -Lead adsorption capacity, *CAC -Cadmium adsorption capacity, *MAC -Manganese adsorption capacity Summary of Optimised Adsorption Capacity and Percentage Removal.The numerical optimisation method involved setting goals for each response to generate the optimal conditions.The optimisation goal was to maximise the adsorption capacities and percentage removal of the adsorbates, while the ranges of the factors were set as presented in Table 3. Table 9 shows the optimal adsorption limits for the adsorption capacity and percentage removal of the metals in the solution, while Table 10 shows the optimum condition suggested by the software.
Validation of the optimum condition.Adsorption experiments were carried out at the following optimum conditions (particle size 150 µm; initial concentration 50 mg/l and contact time 171 minutes) to validate the predicted optimisation results.The responses and their residuals are presented in Table 11.Comparison of optimal operating conditions.The suggested optimal values for lead, cadmium and manganese were almost similar to those proposed in other reported studies.Despite this, the values obtained differed, which shows that conditions were specific for each system according to the adsorbent characteristics, adsorbate type, and operating factors.A study conducted by [20] on the adsorption capacity of bentonite-kaolinzeolite pellet in a multi-metal solution revealed 99.84% and 61.93% removal for lead and cadmium, respectively, which confirms the preferential removal of lead over cadmium.Another study on the adsorption capacity of zeolite by [30] revealed a 95.08% and 80.77% removal for lead and cadmium, respectively, while a study on the adsorption capacity of kaolin conducted by [31] for a multi-metal solution revealed a slightly different trend with 99.2% cadmium removal and 96.3% lead removal.The result of this study, however, agrees with the superior uptake of lead over cadmium in a multi-metal system.However, manganese experienced a higher uptake compared to both lead and cadmium, although the cation size indicated by the ionic radii increases from lead to manganese, i.e. 0.401, 0.426, 0.438 nm for Pb (II), Cd (II) and Mn (II) respectively [32] indicating that the smaller the cations, the faster their adsorption and the quantity adsorbed because they passed through the pores and channels within the bentonite structure with ease [33].This deviation is probably due to the variation in operating factor time.Lead uptake is spontaneous and attains equilibrium early (about 30 mins).Still, though not sporadic, manganese uptake continues as the contact time increases, accounting for more uptake at 170 mins optimal time.

CONCLUSIONS
This research focuses on the optimisation of process parameters such as clay particle size (50-150 µm), metal ions initial concentration (10-50 mg/l) and contact time (30-180 minutes) for the maximum multi-metal adsorption capacity and percentage removal.The results of statistical analysis of the adsorption experiments carried out according to the Box-Behnken design showed that only the particle size and initial concentration had significant individual effects on the responses.However, the impact of contact time was substantial for cadmium removal, suggesting that the adsorption of cadmium ion on the adsorbent increases with increased contact time; this is due to high cadmium cation hydration energy of -1807 kJ/mol -1 compared to lead and manganese with -1481 kJmol -1 , -1760 kJmol -1 respectively.Furthermore, the model equations for the responses were developed, and the optimum adsorption condition for the multi-metal adsorption that maximised the adsorption capacity and the percentage removal was obtained to be 150 µm particle size, 50 mg/l initial concentration over 171 minutes.

Figures 1 ,
Figures 1, 2 and 3 represent the standard probability plot of the studentised form of residuals, which indicates whether the residuals follow a normal distribution, in which case the points will follow a straight line.In this study, the residues for LAC, CAC and MAC have a normal distribution

Table 2 -
Factors Level Selected for Adsorption Experiment

Table 3 -
Calculation of Structural Formula for Modified Dijah-Monkin Bentonite Clay

Table 4 -
Selected Factors and Responses for the Adsorption Experiment of Multi-Metal Solution

Table 5 -
Analysis of Variance for Adsorption Capacity

Table 6 -
Model Summary Statistics of Adsorption Capacity

Table 7 -
Analysis of Variance for Percentage Removal

Table 8 -
Model Summary Statistics of Lead Percentage Removal

Table 10 -
Optimal Point and Predictions

Table 11 -
Residuals of Validation of the Optimum Condition