Modelling and optimization of competitive bio-sorption of copper and lead ions using fruit peels

Abstract

The application of various agricultural-based materials as adsorbents for the removal of heavy metal ions from aqueous solutions has attracted the interest of many researchers. Many studies have been conducted on the removal of heavy metals from wastewater using the bio-sorption process with a focus on wastewater containing single solutes. In addition, the existing column adsorption models were developed to describe the dynamic behaviour of single solute biosorption processes. However, the application of a linear driving force model which makes use of batch experimental results to describe the bio-sorption process in a fixed-bed has not been reported for binary solute systems. In this study, the performance of orange and banana peels was investigated for the removal of copper and lead ions from wastewater in both single and binary systems. These bio-sorbents were used in their natural form. The characterization of the bio-sorbents before and after adsorption was achieved using analytical techniques. Fourier Transform Infrared Spectroscopy (FTIR) was used to determine the functional groups present on the surface of the bio-sorbents. The results showed that the bio-sorbents contain various functional groups such as carboxyl, hydroxyl, carbonyl, aminyl, and alkyl groups that enhanced the adsorption process. After adsorption, there were significant shifts in the peaks representing hydroxyl and carboxyl groups which were common to both biosorbents. Therefore, it was concluded that ion exchange is the sorption mechanism responsible for the adsorption of metal ions. The Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy (SEM/EDS) was used to determine the morphological structure and the elemental composition of the bio-sorbents. The surfaces of the bio-sorbents showed an irregular, and microporous structure while the elemental composition revealed the presence of carbon, oxygen, hydrogen, and potassium. The surfaces of the bio-sorbents became uniform, smooth, and covered after the adsorption and the EDS showed the presence of adsorbed metal ions. The X-ray diffractometer (XRD) which explained the crystallinity of the bio-sorbents showed that both bio-sorbents are amorphous. The values of the point of zero charge (pHpzc) for orange and banana peels obtained were 3.85 and 4.83 respectively. This revealed that the surfaces of both bio-sorbents were acidic and therefore suitable for the adsorption of cations. The design of experiments (DOE) was employed in the batch study to investigate the interactive effect of the operating parameters, and a 24 full factorial design was used to generate the experimental runs. The factors studied and their ranges are initial concentration (10 – 100 mg/L), solution pH (2 - 6), adsorbent dosage (0.1 – 1 g), and particle size (75 – 455 µm), for the single solute system. The interaction of the factors was studied using response surface methodology (RSM) with the central composite design (CCD). The highest removal of lead and copper for orange peels was 99.75% and 98.53% while 99.32% and 98.12% were obtained for lead and copper using banana peels. The results showed that both bio-sorbents have a high affinity for lead and copper while the order of influence of the factors gave adsorbent dosage > pH > initial concentration > particle size for the bio-sorption of copper and lead using both bio-sorbents. The optimum pH of 5.5 was obtained for both metals hence three factors (initial concentration, adsorbent dosage, and particle size) with the same parameter range specified for a single solute were considered for the binary solute interactive study using the same method. The initial concentration of the metal ions in the binary system was in the ratio of 1:1. In most of the experimental runs, the percentage yield of lead was higher than copper. The highest removal for lead and copper was 98.85 % and 87.32 % for orange peels, while for banana peels, lead was 97.85 % and copper was 97.6 %. The isotherm and kinetic studies of single and binary systems of copper and lead were carried out using both bio-sorbents. The Langmuir isotherm fitted the adsorption data signifying a monolayer adsorption mechanism while the pseudo-second-order model fitted the kinetic data which suggested the chemisorption process for both bio-sorbents in single and binary systems. The adsorption of lead was higher than copper in both single and binary systems and the adsorption of copper was sensitive to the co-existence of lead in binary systems. Orange peels bio-sorbent performed slightly better than banana peels hence, it was chosen for the dynamic column studies The fixed bed experiments were conducted to investigate the effect of column parameters such as flow rate (1 and 3 mL/min), initial concentration (10, 50, and 100 mg/L), and bed height (1 and 3 cm). The results showed that the performance of the bed was improved with an increase in the bed height while the volume of solution treated at breakthrough decreased with an increase in flow rate for both metals in a single solute system. The Thomas, Yoon Nelson, and Bohart Adams models were applied to the experimental data. The Thomas and Yoon Nelson models performed well with a high coefficient of correlation (R2 > 0.9) and the lowest mean absolute error value of less than 0.1. The breakthrough curves for the binary solution of copper and lead showed slightly different shapes than the single solute system. This can be ascribed to the influence of the co-existence of metal ions which led to competition for the limited binding sites on the bio-sorbent. The breakthrough time decreased with an increased initial concentration for both metal ions in the binary system. However, the breakthrough curves representing copper bio-sorption, reached the breakthrough point faster than lead suggesting a lower affinity of copper to bind to the active site. The bio-sorption capacity of lead was consistently higher than copper for all the initial concentrations considered. A mathematical model was developed for the binary solute system of copper and lead. The model was developed from the mass balance equation of the solid and liquid phases of an elemental section of the column. An assumption of axially dispersed plug flow was made, and a linear driving force (LDF) was used to describe the intraparticle mass transfer. The partial differential equation obtained from the mass conservation equation was discretized to form an ordinary differential equation (ODE) using the finite difference method. The resulting ODEs were solved using the ode15s solver in MATLAB. The mathematical model results followed a similar trend with the experimental results, such that the breakthrough curve of copper reached the breakthrough point faster than lead for all the initial concentrations considered. The model results showed that the mathematical model based on the linear driving force can be used to describe the dynamic behaviour of a bi-solute fixed-bed adsorption column. The mathematical model performed well at high initial concentrations. The Thomas model gave the lowest mean absolute error (MAE) value of 0.08 while the mathematical model gave an MAE value of 0.9 which explains the deviation of the models from the experimental results. In conclusion, the equilibrium isotherm studies carried out in the batch experiments were used to assess the adsorption capacity of the bio-sorbent which was used in the LDF expression of the model. Hence, this study has demonstrated that the mathematical model developed for the binary system is suitable for predicting the breakthrough curves using batch experimental results