Modelling of nanostructured membranes for wasterwater purification

Abstract

The removal of pollutants as well as the recovery of high added value molecules with low molecular weight is a current challenge in the wastewater treatment. Membrane processes can be considered as a viable option to solve these issues. In separations based membranes, high rejection of organic matter and water high permeability are two very important aspects that must be optimised. In general, the increase of the former comes at the expense of the latter because high rejection is cause of membrane fouling which in turns reduces dramatically the water permeability. Thus, membrane fouling constitutes one of the major limitations in membrane separation processes. The design of innovative materials which show high water permeability and at same time molecular rejection but with a low propensity to fouling is of fundamental relevance in this area. Mixed-matrix composite membranes where nanostructures such as Carbon Nano Tubes (CNTs) or Nano particles (NPs) are added, are receiving a huge focus since the properties of these nanostructures can confer an increase of the membrane efficiency in terms of permeability, selectivity, resistance and finally antifouling. The choice of the right nanomaterial can be highly accelerated by computational modelling. Thus, in this thesis an ab-initio modelling study in the frame of the Density Functional Theory (DFT) is carried out in order to investigate the structure-property relationships of nanostructures like CNTs and polyoxometalates NPs (POM). These nanostructures are used in the preparation of composite membranes in the frame of BioNexGen project [1] with the aim to propose novel membranes to be used in bioreactor for industrial waste water treatment. In particular, the rejection of CNTs towards organic solutes with low molecular weight coming from industrial waste water such as cosmetic, textile and oil olive was analysed in conjunction to the optimization of water permeability in CNTs. Also, the adsorption of POMs on polymeric membrane surface was analysed in order to understand if an efficient antifouling property can be imparted to the membranes by adding these NPs which possess excellent antibacterial and oxidants properties. While nanotube permeability has been extensively studied in both experimental and modelling works resulting in orders of magnitude water flow enhancements for tube diameters in the reverse osmosis and nanofiltration range [2], less work has been done on CNT selectivity of small organic solutes. Precise control of the CNTs synthesis allows for fine-tuning of the outer and inner diameters in the case of single- and multi-wall carbon nanotubes (SWNTs and MWNTs, respectively), offering the possibility of controlling their selectivity toward small solutes. In this thesis a modelling study was, therefore, addressed to define the optimal CNT internal diameter that should be used in order to achieve a total rejection of several target compounds very difficult to separate. Three novel algorithms [3, 4] were implemented during the thesis for reach the aforementioned objective. These algorithms do not make use of any adjustable parameters (i.e. fitting parameters) being based on geometry optimizations carried out in the frame of high level of quantum approach (DFT) and topological analysis of the considered systems. In addition, the CNTs geometrical characteristics should be optimized in order to get the best compromise between permeability and rejection. The functionalizing of the nanotubes tip with specific functional groups capable of hindering the passage of solutes while allowing water molecules to flow has been suggested as a solution to this problem [5, 6]. Thus, a theoretical study of CNT-composite membranes was carried out in this work with the aim of optimizing the CNTs tip, through their functionalization, to maximise both water permeability and solutes selectivity. This study is the basis of the third implemented algorithm [7]. Concerning the membrane fouling, the analysis of the adsorption of antibacterial anionic NPs (i.e. POM) on the surface of polymeric membranes was carried out in this thesis in order to achieve an optimal adsorption of these NPs. Noncovalent interactions between POM and commercial surfactants, used in polymer membrane preparations, were evaluated. Anionic POMs are in fact promising antibacterial agents [8], therefore, an efficient exchange with the bromide ions, used to counterbalance the positive charge of the commercial surfactants, would allow to increase the efficiency of the membrane by reducing the organic and bio fouling. In order to achieve this result, besides the aforementioned POM-surfactants noncovalent interactions the electronic hydration energy of POM and bromide anions was also evaluated in order to predict the probable exchange of these anions on the polymer surface. The conclusions of this thesis are achieved by using only ab-initio methods thus the results can be considered quite generals and homogeneous, free from empirical or fitting parameters. A validation of the theoretically prediction is provided concerning the POM-bromide exchange by means of an ad hoc designed experiment. The experimental results are in agreement with the theoretically predictions