Application of chitosan-carbon nanotube hydrogel beads composite in the removal of antibiotic compounds and perfluoroalkyl substances from aqueous solution

Abstract

The environmental occurrence of antibiotics and perfluoroalkyl acids (PFAAs), particularly in portable water sources, is very evident that current wastewater treatment plants cannot completely eradicate these emerging contaminants of environmental concern. On the other hand, long term exposure to these contaminants poses serious health risks to human and aquatic life. Available literature suggests that antibiotics and PFAAs can be eradicated by solid-liquid adsorption. Therefore, there is an urgent need to develop environmentally green and cost-effective adsorbents for the remediation of antibiotics and PFAAs. As such, the present study focuses on investigating the treatment efficiency of chitosan-carbon nanotube (CCNT) hydrogel beads for the removal of antibiotics viz. amoxicillin (AMX), ciprofloxacin (CIP), and sulfamethoxazole (SMX) as well as PFAAs i.e., perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS) from synthetic aqueous solutions. The scope of the present work includes elucidating the adsorption kinetics, isotherms, thermodynamic parameters as well as breakthrough curves for the uptake of AMX, CIP, SMX, PFOS and PFOS on CCNT hydrogel beads. Moreover, Monte Carlo simulations were performed to elucidate the interaction of PFOA, PFOS and sulfamethazine (SMT) with polyethylene (PE) and polypropylene (PP) relative to water. Herein, CCNT hydrogel beads were synthesised using a two-step precipitation approach and analysed using Fourier transformation infrared (FTIR) technology. From the FTIR results it was evident that the synthesised model adsorbent was characterised with strong peaks of carboxylic and hydroxyl functional groups which were essential for the uptake of the model adsorbates i.e., AMX, CIP, SMX, PFOA and PFOS. Post the adsorption process, band stretches were observed cementing the uptake of the model adsorbates on CCNT hydrogel beads. Single adsorption kinetics experimental data for the uptake of AMX, CIP, SMX, PFOA and PFOS were well fitted by the nonlinear pseudo-first order (PFO) kinetic model recording R2 values of more than 0.9 for all model adsorbates. It is worth noting that the competing PFO and pseudo-second order (PSO) kinetic models were validated by applying the Bayesian Information Criterion (BIC) as a statistical analysis parameter. Furthermore, the findings of the present study from the Weber-Morris kinetic model suggest that multiple processes were limiting the overall adsorption rate of AMX, CIP, SMX, PFOA and PFOS on CCNT hydrogel beads. Adsorption isotherm studies were conducted for a temperature range of 283 K, 293 K and 303 K for a contact time of 24 hours for AMX, CIP, and SMX, and 168 hours for PFOA and PFOS. The findings of the present study suggest that AMX, CIP, SMX, PFOA and PFOS single adsorption experimental data were well fitted by the nonlinear Freundlich isotherm model suggesting the heterogeneity of the surface as well as the exponential distribution of the active sites of the model adsorbent. On the other hand, the binary and ternary AMX, CIP, and SMX adsorption experimental data were well fitted by the nonlinear competitive extended Sips adsorption isotherm model. Furthermore, results for the binary and ternary adsorption systems explicitly demonstrated that the multicomponent adsorption systems exhibited both antagonistic and synergistic effects on the uptake of AMX, CIP and SMX. Interestingly, binary adsorption experimental data for PFOA and PFOS were well fitted by the extended-Langmuir isotherm model (R2=0.996 for PFOA and R2=0.995 for PFOS) and extended-Sips isotherm model (R2=0.996 for PFOA and R2=0.997 for PFOS), with the system strictly exhibiting antagonistic effects for the uptake of one adsorbate in the presence of another adsorbate. From thermodynamic studies, it is evident that the uptake of AMX, CIP, SMX, PFOA and PFOS is an endothermic process, and it cannot be explicitly classified as a chemical nor physical adsorption process but as a physicochemical adsorption process. The presence of sodium chloride (NaCl) and humic acid (HA) demonstrated antagonistic effects on the uptake of AMX, CIP and SMX by CCNT hydrogel beads due to formation of aggregates with an increase in ionic strength. Single factor analysis of variance results recorded p-values of less than 0.05 for the uptake of AMX, CIP and SMX indicating that there was a statistical difference between the means of the independent and depended variables, thus cementing the negative effect of increasing ionic strength on the uptake of the mode adsorbates. However, NaCl ions exhibited minimal competitive effects with adsorbate molecules for active adsorption sites on CCNT hydrogel beads compared to HA. On the other hand, the presence of NaCl as a competing ion exhibited synergistic effects in the uptake of PFOA and PFOS from aqueous solutions. Furthermore, for the present work, breakthrough curves from the experimental data were well fitted by the Thomas model recording R2 and adjusted-R 2 values of greater than 0.9 for all adsorbates investigated indicating that the breakthrough curves for the present work can be described by a symmetrical function. Additionally, the breakthrough points time predicted by the Thomas model was aligned with the experimentally determined breakthrough points time cementing its practical utility and superiority over the log-Gompertz and Bohart-Adams models. On the other hand, from the Monte Carlo simulation results, it is evident that, in an aqueous environment, both PFOA and PFOS may be taken up preferentially by PP and PE, although less strongly by PE. The degree of polymerisation of PE and PP did not significantly influence the observed behaviour. In terms of sorption affinity, the observed affinity was PFOA>PFOS>SMT which was consistent for both PE and PP. Based on the results obtained, it was concluded that CCNT hydrogel beads composites have the potential to be applied as adsorbents for the removal of antibiotics and PFAAs from aqueous solutions. Furthermore, the simulation results obtained suggest that Monte Carlo simulations in Material Studio can be used as an effective tool in elucidating the interaction between antibiotics and PFAAs with microplastics relative to water as co-existing contaminants. Therefore, the findings of the present work have successfully addressed the research questions for the current study