Resumen de Assessment and optimization of the operation of integrated membrane systems for wastewater reclamation
The combination of two membrane technologies coupled together in series has become a standard technology when it comes to producing reclaimed water of high quality for potable reclamation or industrial applications. This combination of two membrane processes is referred to as integrated membrane systems (IMS). Despite the widespread experience gained utilizing such a process technology around the world, there are a number of aspects of the process technology which require further investigation including the fate of compounds of emerging concern (CEC), the control of N-Nitrosodimethylamine (NDMA) formation, the use of energy associated with the process and the total cost of producing the reclaimed water, and monitoring membrane integrity in RO treatment processes. The objective of this work was to further the knowledge in one aspect related to each of these four challenges and then bring each of these areas together in the discussion to understand whether proposing a decision support system for the online monitoring and operation of integrated systems would allow improvements to the current state-of-the-art.
The first results chapter deals with the fate of pharmaceuticals and their transformation products through the MBR-RO/NF process. The majority of published work focus on the removal of parent compounds of CEC through treatment processes. Often these parent compounds are excreted from the human body with a number of human metabolites which could be found at concentrations much greater than the corresponding parent chemical and can themselves be pharmacologically active (Petrie et al., 2015). For this reason, the focus of this work was to advance the knowledge in terms of understanding the fate of a number of pharmaceuticals and particularly their main human metabolites through the IMS process. The results showed that the two consecutive membrane processes, when seen as a whole, become a highly efficient process to remove all the studied compounds. When comparing the removal efficiencies of the RO and NF membranes, as expected, the RO membrane showed near complete removals (>99%) of all the compounds over various process conditions, whereas the NF membrane resulted also in high removal efficiencies (> 90%).
It has been shown that nitrifying WWTPs are better able to remove nitrosamines precursors when compared to non-nitrifying facilities (Sgroi et al., 2016a). Due to the poor removal of nitrosamines by RO and NF membranes and the high energy cost of removing them using UV processes as a last step in a treatment train, the focus of the second chapter of work was the removal of NDMA formation potential and individual precursors under nitrifying and non-nitrifying conditions (achieved by changing the aeration conditions of the bioreactor). The work also looked at the removal of NDMA formation potential and individual precursors by using an NF membrane to understand whether an NF membrane would provide a high enough rejection of NDMA to achieve potable reclamation water quality targets. The results showed that during normal aerobic operation, implying a fully nitrifying system, the MBR pilot plant was able to reduce NDMA formation potential above 94%, however this removal percentage was reduced to values as low as 72% when changing the conditions to avoid nitrification. These results suggest that a fully nitrifying MBR system will support better removal of NDMA precursors during wastewater reclamation.
Unlike seawater RO systems, were systems operate with fairly constant process conditions due to the relatively constant feedwater salinity and temperature, wastewater shows diurnal variations which are catchment dependant. The objective of the next chapter was to explore whether these diurnal variations in wastewater quality, in terms of inorganic constituents and temperature, would justify the modification of RO process conditions (in terms of system recovery and pre-treatment dosing) to minimise the operational cost while considering the control of membrane fouling. The results showed that although there are limitations to the use of electrical conductivity (EC) as a main parameter to deduce the individual ionic constituents in a wastewater, given the right assumptions, the work has shown that it is possible to obtain a useful profile for a particular EC value which could be used in an online / real-time optimisation system. Through the presentation of a case considering the cost of energy and pre-treatment chemicals, the work showed that there is scope for online optimisation tool in terms of associating a cost to the current operating process conditions and then questioning whether there is a more cost-effective way to ‘set’ the system.
An EC sensor is a standard component on any RO/NF system. The way it is currently utilised in a system doesn’t provide much information to the operator in terms of membrane integrity monitoring. The final results chapter explores the limits of detection of membrane integrity methods using EC sensors and provides a number of strategies to obtain a better characterisation of membrane integrity.
The results are brought together in the discussion of the work to provide a framework together with an initial rule-base for a knowledge-based decision support system for the real-time control of integrated membrane systems for wastewater reclamation.
