Identifying novel wastewater treatment options through optimal technology integration

Wastewater management approaches are currently undergoing a paradigm shift from a public health and environmental protection focus, to becoming part of an overall integrated urban/industrial water cycle with maximal resource recovery objectives. The key resources to be considered in this context are the water itself (the most valuable component), the energy content (in the form of organics) and the nutrients (particularly nitrogen and phosphorus). This paper presents the outcomes of the first of four phases of a research project being conducted by the Advanced Water Management Centre at the University of Queensland, GHD Pty Ltd, Melbourne Water and Wide Bay Water, and sponsored by the Australian Water Recycling Centre of Excellence under the Australian Government's AU$12.9 billion Water for the Future Initiative. The research project has a particular focus on achieving maximal resource recovery outcomes with minimised environmental and economic footprint, through the optimal integration of novel, yet largely already demonstrated technologies, as part of a larger water recycling process train. Therefore, the aim of the project is to develop and demonstrate an integrated treatment train that can achieve water quality fit for recycling at a lower energy and chemical input and reduced capital and operating cost compared to current approaches.

The proposed treatment trains consist of three main elements; carbon removal, nitrogen removal, and a polishing stage. The research project will evaluate the first two of these three stages. Water of different qualities will be produced after each stage, each suitable for different reuse purposes. Water produced after Stage 1 could be used for agricultural, forestry and possibly limited horticultural irrigation applications, as the water contains substantial amounts of nitrogen and phosphorus. Product water from Stage 2 would be low in nitrogen (and possibly phosphorus) and would meet most environmental discharge requirements. As such, this water could be utilised to provide environmental flows, or used for a range of irrigation applications and certain industrial water recycling applications (though most likely requiring some further polishing). Stage 3 would then provide further water quality improvements through removal of solids and/or disinfection to allow more extensive recycling opportunities.

The research project consists of four phases: (1) a desk-top study; (2) lab-scale studies examining the proposed core treatment processes; (3) pilot-scale implementation of Stage 1 and 2 optimal process trains determined in Phases 1 and 2; (4) evaluation of environmental and economic benefits of the optimised process trains. Phase 1 of the project is an engineering and design study, which is the focus of this paper.

Figure 1 shows the overall process flow diagram of the proposed treatment processes investigated, incorporating several options for each of the three treatment stages.

Stage 1: Carbon removal processes. Two process technologies for carbon reduction are presented in Figure 1(A)) using a high rate anaerobic mainstream process, and (B) using a high rate aerobic mainstream process. Both technologies utilise anaerobic processes for carbon removal, generating methane for subsequent energy recovery. Process technology A incorporates the anaerobic process directly into the treatment train, whereas process B has the anaerobic process as a side-stream.

For process A, the mainstream anaerobic treatment could be achieved in an anaerobic membrane bioreactor (AnMBR) (Skouteris et al. 2012; Liao et al. 2012; Smith et al. 2012), a baffled anaerobic reactor (up-flow anaerobic sludge blanket (UASB) for example) (Seghezzo et al., 1998), or even an anaerobic lagoon. Biomass retention will be achieved through membrane separation, by granulation or flocular settling in the baffled reactor, or by lagoon settling. A key issue that has been identified in previous installations of such processes is the soluble methane present in the effluent from these stages, which needs to be stripped and harvested before the effluent is treated in the downstream process unit (Hartley & Lant 2006; Hatamoto et al. 2010; McCarty et al. 2011; Bandara et al. 2011; Cookney et al. 2012).

Process B utilises a very high rate aerobic (solids contact activated sludge) process (HRAS), whose function is to convert most of the soluble organics to biomass (Vesprille et al. 1984; Gray 2004). The biomass also captures most of the particulate pollutants from the wastewater. Excess biomass is then separated from the bulk liquid flow (typically via settling) and is passed into an anaerobic digester for methane generation. Novel variations are available for the side-stream biomass anaerobic digestion, with temperature phased anaerobic digestion (TPAD) selected for this study (Ge et al. 2011a, 2011b; Ho et al. 2013). The product-stream from the anaerobic digester (after solids separation and dewatering) will be high in nitrogen and phosphorus, and can be treated separately to both recover nutrients (as struvite) and to remove excess nitrogen (Sliekers et al. 2002; Le Corre et al. 2009).

Stage 2: Nitrogen removal processes.Figure 1 presents four proposed process technology options for nitrogen reduction, incorporating novel treatment processes. Process technology A utilises the recently discovered denitrifying anaerobic methane oxidation (DAMO) process (Hu 2010). In this process, the oxidized nitrogen species, nitrate and nitrite, are reduced to nitrogen gas using methane as the carbon source. The methane is supplied from the Stage 1 anaerobic process. Process B utilises a mainstream Anammox process, with the biomass growing as biofilm in a moving bed biofilm reactor (MBBR) arrangement (Kartal et al. 2010). Both the nitrite production and utilisation (by Anammox) occurs in the one tank concurrently, at various depths within the attached biofilm. Process C represents a more traditional nitrification/denitrification process (as either an Sequencing Batch Reactor (SBR) or continuous process), but with minimised carbon and energy usage by employing nitritation/denitritation via the ‘nitrite pathway’. Process D treats the dewatering liquor after the anaerobic digester for the Stage 1 B process. A nitritation/Anammox SBR was chosen to treat this high strength side-stream (Kuenen 2008), as this process is now widely achieving efficient nitrogen reduction in full-scale applications worldwide.

The key design characteristics and operating parameters of these prospective treatment technologies were reviewed and summarised, with the main purpose of the review to provide supporting literature information for the key process parameters to be used in the engineering and economic desk-top assessment.

To shortlist the potential processes for the new treatment trains, the novel treatment technologies reviewed were evaluated against established current technologies by a multi-criteria analysis (MCA) approach. A ‘Base Case’ of established technology was included to provide relativity in the MCA assessment, with anaerobic lagoons selected as the Base Case for the carbon removal stage, and extended aeration activated sludge selected for the nitrogen removal stage. The criteria considered included financial (capital and operational costs), operational (complexity and robustness), water quality, environmental impacts/benefits, and technology risks and uncertainty.

The short-listed technologies selected by the MCA for the carbon stage were a HRAS process and an AnMBR process, and those selected for the nitrogen reduction stage were an Anammox MBBR and a nitrogen removal SBR process (refer Tables 1 and 2). In all cases, these processes ranked higher than the ‘Base Case’ established technologies.

The two highest ranking technologies selected from the MCA for both the carbon and nitrogen removal stages, were incorporated into two alternative treatment train options. Option 1 (shown in Figure 2) includes an AnMBR for carbon removal, dissolved methane stripping and recovery from the AnMBR effluent, followed by nitrogen removal in a mainstream combined nitritation/Anammox MBBR.

Option 2 (shown in Figure 3) includes a high rate aerobic activated sludge (HRAS) process for carbon removal, with further mainstream nitrogen removal treatment using an SBR process. The biosolids generated from the high rate aerobic process are digested by a TPAD system. The effluent from the TPAD system passes through a struvite crystalliser process for nutrient recovery, before being treated with a side-stream Anammox process for nitrogen removal.

For comparison with these two new designs, a current treatment train ‘Base Case’ was selected (as shown in Figure 4). The ‘Base Case’ option included a single-stage extended aeration activated sludge main-stream process for both the carbon and nitrogen removal requirements. Side-stream aerobic digestion was selected for the excess biosolids stabilisation. This type of process train has been widely adopted in Australia (NWQMS 2000) to meet low effluent nutrient levels (of less than 5 mg/L nitrogen and 2 mg/L phosphorus).

Two typical scales of implementation for Australian conditions were selected for the assessment; 10 ML/d and 100 ML/d average flow, in order to determine the potential scale effects on treatment plant economics.

The engineering analysis carried out for each alternative process train used the key design characteristics and operating parameters estimated from the literature for these prospective treatment technologies. High-level concept design and sizing, estimates of performance, assessment of environmental footprint, and whole-of-life cost estimates (including capital and operating expenses) were developed for each process train option. The operational and capital costs for the two treatment train options were estimated and compared with typical current costs of the ‘Base Case’ option, for both the 10 ML/d and 100 ML/d plant scales. Some items (such as land, labour and maintenance costs) were excluded from the capital and operating cost calculations.

The results of the engineering and economic evaluation are summarised in Table 3. These show that the new treatment trains have the potential to substantially decrease the economic cost of wastewater treatment. The potential economic advantage over the Base Case is predicted for both alternative treatment train options, and for both plant scales considered.

For the 10 ML/d case, the estimated capital costs of the new treatment trains are slightly higher than for the Base Case. However, the estimated operational costs are significantly lower, especially for the Option 1 mainstream AnMBR/Anammox. The lower operational costs of the new treatment trains are typically due to lower treatment process energy consumption and higher biogas production for energy generation. For Option 1, the value of the power produced from the biogas production is estimated to be higher than the power required for the treatment process, leading to a potential negative operational cost for this option.

The economic evaluation results for the 100 ML/d case are similar to the 10 ML/d case, where it was also estimated that both treatment train options have advantages over the Base Case. The estimated capital costs of the novel treatment trains are more or less the same as the Base Case. However, the predicted operational costs are significantly lower, especially for the Option 1 mainstream AnMBR/Anammox train.

Of interest for both plant sizes evaluated, is that the estimated capital costs for all technology options are quite comparable. Hence, the key differences are generated by the significantly lower operating costs (and mainly the differences in energy consumption/production) of the novel treatment trains compared to current technologies.

Both shortlisted processes have the potential to substantially decrease the cost of wastewater treatment and recycling (by between 10 and 46% based on Net Present Value estimates) based on high-level engineering and economic analysis. The new treatment trains proposed are predicted to be not only economically beneficial, but also environmentally advantageous, having the potential to reduce energy and chemical usage, while also improving nutrient and energy recovery. Compared to the Base Case of sewage treatment, the new designed treatment trains may also deliver equal or better quality of water for recycling at a much lower cost.

The performance of the sequential combined processes and their design and operational parameters is currently being evaluated in the Phase 2 and 3, lab-scale and pilot studies. These studies will further evaluate the performance treatment trains and unit processes, and confirm the design/operating parameters of the core processes. The further studies will also attempt to demonstrate the reliability of the process trains and the optimal integration of the treatment elements, as well as confirm the energy and other inputs required to achieve satisfactory performance.