Regression model, artificial neural network, and cost estimation for biosorption of Ni(II)-ions from aqueous solutions by Potamogeton pectinatus

ABSTRACT

This study investigated the application of Potamogeton pectinatus for Ni(II)-ions biosorption from aqueous solutions. FTIR spectra showed that the functional groups of –OH, C–H, –C = O, and –COO– could form an organometallic complex with Ni(II)-ions on the biomaterial surface. SEM/EDX analysis indicated that the voids on the biosorbent surface were blocked due to Ni(II)-ions uptake via an ion exchange mechanism. For Ni(II)-ions of 50 mg/L, the adsorption efficiency recorded 63.4% at pH: 5, biosorbent dosage: 10 g/L, and particle-diameter: 0.125–0.25 mm within 180 minutes. A quadratic model depicted that the plot of removal efficiency against pH or contact time caused quadratic-linear concave up curves, whereas the curve of initial Ni(II)-ions was quadratic-linear convex down. Artificial neural network with a structure of 5 – 6 – 1 was able to predict the adsorption efficiency (R²: 0.967). The relative importance of inputs was: initial Ni(II)-ions > pH > contact time > biosorbent dosage > particle-size. Freundlich isotherm described well the adsorption mechanism (R²: 0.974), which indicated a multilayer adsorption onto energetically heterogeneous surfaces. The net cost of using P. pectinatus for the removal of Ni(II)-ions (4.25 ± 1.26 mg/L) from real industrial effluents within 30 minutes was 3.4 $USD/m³.

KEYWORDS:

• Agro-based sorbent
• Ni(II)-ions biosorption
• Statistical and intelligence models

Funding

This work was supported by Environmental Sciences Department at Faculty of Science, and Environmental Studies Department at Institute of Graduate Studies and Research, at Alexandria University, Egypt.