Competitive Fixed-Bed Adsorption of Pb(II), Cu(II), and Ni(II) from Aqueous Solution Using Chitosan-Coated Bentonite

1Department of Medical Laboratory Science and Biotechnology, Kaohsiung Medical University, Kaohsiung 80708, Taiwan 2Department of Chemical Engineering, University of the Philippines, Diliman, 1101 Quezon City, Philippines 3Environmental Engineering Program, National Graduate School of Engineering, University of Philippines, Diliman, 1101 Quezon City, Philippines 4Environment Business Line, Aecom Philippines Consultants Corporation, 1634 Taguig, Philippines 5Department of Environmental Resources Management, Chia Nan University of Pharmacy and Science, Tainan 71710, Taiwan


Introduction
The presence of heavy metals in groundwater and surface waters is considered a serious threat to the ecosystem and human health due to properties such as nonbiodegradability, bioaccumulation, and changing oxidation states [1].Globally, the Philippines is ranked to be the fourth and fifth largest producer of Cu(II) and Ni(II), respectively.According to Republic Act 9275 (Philippine Clean Water Act of 2004), Pb(II) is generated during the ore mining and processing of Cu(II) and Ni(II) [2].Based on the toxicological criteria, excessive intake of Pb(II) is considered to cause kidney damage, disruption of the nervous system, and delays in physical and mental development.Meanwhile, common metals such as Cu(II) and Ni(II) are considered to be essential minerals.However, ingestion of Cu(II) at high concentration may lead to central nervous system irritation, kidney failure, mucosal irritation, and liver damage while an increase in Ni(II) intake can cause health problems like lung, nose, and bone cancer as well as chronic bronchitis, respiratory distress, birth defects, and embolism [3][4][5].Due to the deleterious effects of heavy metals on human health, the US Environmental Protection Agency has set a maximum contaminant level for Pb(II) and Cu(II) in drinking water to be at 0 mg/L and 1.3 mg/L, respectively.
Adsorption is a simple, highly effective, economically feasible, and environmentally benign technology utilized in the removal of heavy metals from industry effluents [6,7].

International Journal of Polymer Science
In purification processes, liquid phase adsorption is typically carried out under varying configurations such as fixed-bed, fluidized bed, and batch.Fixed-bed adsorption has been proven to be an effective process under continuous flow conditions due to its operational simplicity, possibility of in situ generation, and ease of operation and handling [8,9].Moreover, only fixed-bed studies would provide readily available support data for the direct application of wastewater treatment in the industrial scale.
Chitosan-clay composites are considered to be a promising class of adsorbents due to their low production cost, improved materials, and mechanical stability [10].Several studies using chitosan-coated montmorillonite composites were performed in the removal of tungsten [11], tannic acid [12], basic dyes [13,14], humic acid [13], and reactive dyes [13].L. Wang and A. Wang [10] investigated the adsorption of Congo red using N,O-carboxymethyl-chitosan/montmorillonite nanocomposites.
Chitosan-bentonite beads have been utilized in several studies for heavy metal removal, where the preparative method was similar to the process utilized by Wan et al. [15].Basic comparative studies investigated the efficiency of removing Cu(II) from aqueous solution using chitosancoated bentonite (CCB) cross-linked with epichlorohydrin, glutaraldehyde, and ethylene glycol diglycidyl ether [16,17].Other articles on CCB have been published, including removal of oxidized sulfur compounds [18], As(V) [19], In(III) [20], Pb(II) [2], Cu(II) [2], and Ni(II) [2].Most of the previous studies on heavy metal removal using CCB were carried out in the batch mode.There were no reported fixedbed studies on the competitive adsorption of Pb(II), Cu(II), and Ni(II) from multiple aqueous solutions.Moreover, it is essential to evaluate the simultaneous adsorption of two or more metals since single metal species rarely exist in wastewater effluents and natural waters [21][22][23].In continuation of the previous work, this study aims to investigate the dynamics of the competitive adsorption of a multimetal system (Pb(II), Cu(II), and Ni(II)) using CCB under fixedbed conditions.The effects of operational conditions such as bed height, solution flow rate, and initial ternary metal ion concentration on the shape of experimental breakthrough curves and corresponding equilibrium adsorption capacities were examined.The fixed-bed adsorption data were fitted to common dynamic models such as Adams-Bohart, Thomas, and Yoon-Nelson models and were evaluated using linear regression analysis.The validity of the kinetic models, in terms of accuracy and adequacy, was measured using error analysis.

Instrumentation.
Infrared spectra of CCB before and after adsorption of Cu(II), Ni(II), and Pb(II) were obtained using a Nicolet 6700 Fourier Transform Infrared spectrometer using disc composed of 1 : 10 ratio of sample to KBr within the spectral range of 400-4000 cm −1 .The surface morphology and surface elemental analysis of CCB were observed by JEOL JSM-7001 scanning electron microscope (SEM) and energy dispersive X-ray spectroscopy (EDX) under a vacuum of 1.33 × 10 −6 mBar, running at 20.0 kV and using a tungsten filament.The quantitative analysis for the residual concentration of metal ions was analyzed using ICP-OES Perkin Elmer DV2000 series.

Synthesis of Chitosan-Coated Bentonite.
Based on the method by Wan et al. [15], CCB beads were synthesized by dissolving 5.0 g chitosan in 5% (v/v) HCl (300 mL) where the mixture was stirred for 2 h at 300 rpm.About 100 g of bentonite was slowly added to the solution and stirred for another 3 h.Sodium hydroxide (1 N) was added dropwise until neutral pH was obtained.The resulting beads were thoroughly washed with deionized water and dried in an oven (Chanel Precision Oven DV452, Memmert GmbH) for 24 h at 65 ∘ C.Then, the CCB beads were pulverized and sieved where particles with size range of 0.50 to 0.21 mm were used in the fixed-bed studies.

Fixed-Bed Experiments.
The schematic diagram of the fixed-bed adsorption set-up is shown in Figure 1.Fixedbed adsorption experiments were carried out using a 1.8 cm i.d., 76 cm length borosilicate glass column (7740, IWAKI).Near the column inlet, glass beads and sand were placed to prevent channeling and ensure even distribution of the solution.Glass wool and sand were placed at the bottom to prevent loss of the CCB and avoid outlet clogging.The fixedbed was continuously operated in a down flow mode using a peristaltic pump (Model 7518-00, Cole Palmer) at room temperature (25 ± 1 ∘ C).The pH of the influent was maintained at a constant pH 4.0 [17].Effluent samples were collected from the bottom of the fixed-bed at predetermined time intervals.Operating parameters such as initial concentration (50-200 mg/L), flow rate (0.6-1.0 mL/min), and bed height (1.0-3.0 cm) were varied.

Analysis of the Breakthrough Curves.
The column performance in the removal of Pb(II), Ni(II), and Cu(II) from a multimetal system is analyzed using breakthrough curves.The breakthrough time is described as the time when the effluent metal concentration is about 3-5% of the influent concentration [24].In this study, the breakthrough time (  , min) is set as the time when the effluent metal concentration reaches 1% of the influent concentration while exhaustion time ( total , min) is set as the time when the concentration of the effluent becomes constant and breakthrough curves become flat.The total adsorbed metal quantity (, mg) in the column is represented by the area under the plot of the adsorbed metal ion concentration, which is calculated through numerical integration, given as where  corresponds to the solution flow rate (mL/min),  0 is the influent concentration (mg/L),   is the effluent concentration (mg/L) at any time  (min), and  refers to the area under the breakthrough curve from  0 to   (mg⋅min/L) from time 0 to any time  [24].The total metal uptake at equilibrium,   (mg/g), is computed using where  (g) corresponds to the mass of adsorbent [24].The total equilibrium adsorption capacity,  total (mg/g), for  number of components is calculated using The length of the mass transfer zone, MTZ (cm), is dependent on the bed height,  (cm), and breakthrough and equilibrium point as computed in [25] MTZ = (  total −    total ) .
The volume of effluent treated,  eff (mL), is calculated using (5)

Error Analysis.
Error analysis is performed in order to validate the most applicable model that would describe the performance of the breakthrough curves in the fixed-bed adsorption of a multimetal system using CCB.In general, the fit between experimental data and linearized forms could be illustrated using the linear regression correlation coefficient ( 2 ).On the other hand, the average percentage error (%) describes the agreement between the theoretical and experimental values of   / 0 that was utilized in plotting breakthrough curves [26].Values of % that are <35% are considered to be acceptable [27]:

Effect of Flow Rate.
The effect of flow rate (0.6 to 1.0 mL/min) on the fixed-bed removal of Pb(II), Cu(II), and Ni(II) was evaluated while the bed depth and initial concentration were maintained at 2.0 cm and 100 mg/L, respectively.Based on Figure 5, the breakthrough curves Transmittance (%) were observed to shift from right to left with increasing flow rate, which indicates shorter service time of the bed.Unlike the steeper and more defined breakthrough curves of 0.8 and 1.0 mL/min, those of 0.6 mL/min are characterized by a gradual slope (lower value of /).In Table 2, an increase in flow rate resulted in lower breakthrough and exhaustion times as well as longer mass transfer zones.This is due to decrease in residence time and higher intraparticle diffusion effect of the aqueous solution in the fixed-bed at higher flow rates [30].In addition, the total volume of effluent treated ( eff ) and quantity of metal ions adsorbed at exhaustion (  ) were observed to also decrease at higher flow rates.The reduced contact time at greater flow rates indicates that there is less time for lateral diffusion to occur within the CCB bed due to weak distribution of the liquid inside the fixed-bed, which results in lower diffusivity of the solute among the adsorbent particles [31][32][33].

Effect of Bed Height.
As seen in Figure 6, the breakthrough curves of Pb(II), Cu(II), and Ni(II) become less steep as bed depth was increased from 1.0 to 3.0 cm.Moreover, as the bed height was increased, breakthrough curves were observed to shift from left to right that resulted in longer times to reach breakthrough and exhaustion (Table 3).It can also be seen that the increasing adsorbent bed height prolonged the breakthrough time (  ) and exhaustion time ( total ) as well as a significant increase in volume of treated effluent ( eff ) of the fixed-bed.The increase in the removal of Pb(II), Cu(II), and Ni(II) can be attributed to greater amount of CCB present in the fixed-bed at 2.0 and 3.0 cm, which implies that there are more binding sites available that would result in higher breakthrough adsorption capacity of the column [34].The good performance of metal removal at bed height of 3.0 cm could also be attributed to more contact opportunities between the metal and CCB particles.However, the mass transfer zone broadened since an increase in bed height causes greater resistance to mass transfer and slower kinetics of adsorption [35].The increased mass transfer resistance is caused by the repulsive forces between the adsorbed metals on the CCB's surface and metals in the aqueous film.

Effect of Influent Concentration.
The breakthrough curves of Pb(II), Cu(II), and Ni(II) under varying influent concentration are shown in Figure 7.When the ternary influent concentration was increased from 50 to 200 mg/L, steeper and sharper breakthrough curves were obtained with higher values of / at 150 and 200 mg/L.In Table 4, results show that the fixed-bed was saturated more quickly, where earlier breakthrough time and exhaustion time were achieved at high metal concentration (200 mg/L).As a result, the total volume of effluent treated ( eff ) decreased.However, adsorption capacity at equilibrium (  ) was observed to increase at high influent concentration.In adsorption, the main driving force is the concentration gradient, which is the difference between the concentration of solute on the sorbent and concentration of solute in the solution [36].At higher metal concentration, the driving force of adsorption is greater due to the high concentration difference facilitated by high mass transfer coefficient values.Thus, higher adsorption capacities are achieved at higher metal concentration.The values of MTZ were observed to decrease with increase in influent metal concentration.

Application of Mathematical Kinetic Model.
Fixed-bed adsorption data was fitted with established mathematical models in order to predict breakthrough curves and to subsequently determine the model parameters.The modeling of the breakthrough curves utilized experimental data with   / 0 values higher than 0.01 to less than 0.99 using Adams-Bohart, Thomas, and Yoon-Nelson kinetic models.The Thomas model evaluates the maximum solid phase concentration of solute on the adsorbent and the rate constant in a fixed-bed [27].The model is based on the following  assumptions: negligible axial and radial dispersion in the fixed-bed; adsorption that follows Langmuir isotherm of adsorption-desorption; and the rate driving force obeying second-order reversible reaction kinetics [37].The linearized form of the Thomas model is expressed as where  TH is the Thomas rate constant (mL/min⋅mg);   is the adsorption capacity at equilibrium (mg/g);  is the amount of adsorbent in the column (g); and  is the solution flow rate (mL/min) [38].
The Adams-Bohart model is an analytical expression based on the assumption of a rectangular isotherm, where the adsorption rate is proportional to both the residual capacity of the sorbent and adsorbate concentration [39,40].The linearized Adams-Bohart equation is provided as  where  AB is the Adams-Bohart kinetic rate constant (L/mg⋅min);   is the saturation concentration of the adsorbent (mg/L);  is the bed height of the adsorbent (cm);  is the linear flow rate (cm/min).
The Yoon-Nelson model assumes that the rate of decrease in the probability of the adsorbate molecule being adsorbed is proportional to the probability of the adsorbate adsorption and adsorbate breakthrough on the adsorbent [31,41].It is expressed by where  is the time required for 50% adsorbate breakthrough (min) and  YN is the Yoon-Nelson rate constant (L/min).All three kinetic models are considered to be mathematically equivalent, which can be represented by a similar fitting equation: The values of the model parameters derived from Adams-Bohart, Thomas, and Yoon-Nelson are presented in Table 5.Based on the high  2 values (0.85 <  2 < 0.99) and low values of %, all three kinetic models (Adams-Bohart, Thomas, and Yoon-Nelson) best describe the simultaneous fixed-bed adsorption of Pb(II), Cu(II), and Ni(II) from a multimetal system.
As the flow rate was increased, all kinetic coefficients of the three models ( AB ,  TH , and  YN ) were observed to increase while   ,   , and  decreased.This implies that external mass transfer dominates the fixed-bed kinetics that occurs at the initial part of the adsorption [31].Moreover, the values of the rate constant are determined by the mass transfer in the fluid that is dependent on the flow rate [42].
Based on Table 5, increasing the bed height from 1.0 to 3.0 cm causes the values of kinetic rate constants ( AB ,  TH ,  YN ) and   ,   , and  to increase.A longer bed height translates to longer contact time of the solution through the fixed-bed.Moreover, a lower rate constant indicates there is sufficient time for the contaminants to migrate from bulk solution to be adsorbed onto CCB surface.
At higher influent concentration, the adsorption bed capacity (  ) and adsorption capacity at equilibrium (  ) were observed to increase while 50% time to reach breakthrough () decreased.However, there was no apparent trend observed for  AB and  TH while values of  YN significantly increased with increasing influent concentration.The increase in  YN was due to the increase in driving force of mass transfer in liquid film which leads to easy saturation of the fixed-bed and resulted in a decrease in  50 [43].Based on Figures 5-7, the predicted breakthrough curves are in good agreement with the experimental data, which implies that models such as Adams-Bohart, Yoon-Nelson, and Thomas are adequate in predicting the fixed-bed adsorption behavior of Pb(II), Cu(II), and Ni(II) in a multimetal system.

Selectivity of CCB in the Removal of Pb(II), Cu(II), and Ni(II)
. In all experimental runs, Pb(II) was observed As such, the total amount of metal ions adsorbed by the column ( ad ) was in the order of Pb(II) > Cu(II) > Ni(II).The selectivity can be attributed to the difference in metal properties such as electronegativity and hydrolysis constant, with Pb(II) being the most electronegative and being easily hydrolyzed in comparison to Cu(II) and Ni(II) [44].Thus, more Pb(II) was adsorbed by the functional groups of CCB.Previous batch studies showed similar results, where Pb(II) has the greatest affinity for adsorption onto CCB over Cu(II) and Ni(II) [2].

Conclusion
The performance of CCB in removing Pb(II), Cu(II), and Ni(II) from multimetal solution under fixed-bed conditions was evaluated.Breakthrough curves were derived under varying flow rate, bed height, and initial metal concentration.Results indicate that the breakthrough time, exhaustion time, and adsorption capacity at breakthrough increase with decreasing flow rate and initial concentration and increasing bed height.The adsorption preference of CCB could be arranged in the order of Pb(II) > Cu(II) > Ni(II).Experimental breakthrough curves were observed to be in good agreement with theoretical curves generated by Adams-Bohart, Thomas, and Yoon-Nelson models, which were validated by high  2 ( 2 > 0.85) and low % values (% < 35%).FTIR analysis showed that hydroxyl (O-H) and amine (N-H) groups of chitosan and Si-O group of bentonite are involved in the adsorption process.This study demonstrates that CCB is an effective adsorbent in the removal of Pb(II), Cu(II), and Ni(II) from a multimetal solution.

Table 1 :
FTIR analysis of CCB (before and after adsorption).

Table 2 :
Column adsorption data and parameters of ternary metal solution of Pb(II), Cu(II), and Ni(II) at different solution flow rates with initial metal concentration of 100 mg/L and bed height of 2.0 cm.

Table 3 :
Column adsorption data and parameters of ternary metal solution of Pb(II), Cu(II), and Ni(II) at different bed heights with initial metal concentration of 100 mg/L and flow rate of 0.6 mL/min.

Table 4 :
Column adsorption data of ternary metal solution of Pb(II), Cu(II), and Ni(II) at different initial metal concentration with bed height of 2.0 cm and solution flow rate of 0.4 mL/min.

Table 5 :
Model parameters of Adams-Bohart, Thomas, and Yoon-Nelson equation for the fixed-bed adsorption of Pb(II), Cu(II), and Ni(II) onto CCB at 25 ∘ International Journal of Polymer Science to be preferentially adsorbed over Cu(II) and Ni(II) in a multimetal system.Breakthrough and exhaustion of the fixed-bed occurred earlier for Ni(II) over Cu(II) and Pb(II).