ROADMAP FOR IMPLEMENTATION OF OPTIMAL NUTRIENT RECOVERY TREATMENT TRAINS AT WRRFS

Transcription

1 ROADMAP FOR IMPLEMENTATION OF OPTIMAL NUTRIENT RECOVERY TREATMENT TRAINS AT WRRFS Vaneeckhaute, C. 1, Belia, E. 2, Meers, E 3, Tack, FMG 3, Vanrolleghem, PA 1 1 Université Laval, Canada, 2 Primodal Inc., Canada, 3 Ghent University, Belgium Abstract This paper develops a generic roadmap for implementation of optimal nutrient recovery teatment trains at water resource recovery facilities (WRRFs). First, guidelines are presented for setting up an optimal bio-based fertilization strategy as function of region-specific fertilizer regulations. Next, instructions are provided to evaluate the feasibility of bio-based fertilizer production based on the residuals characteristics. Finally, a conceptual algorithm is developed in order to facilitate the design of an optimal treatment train. As such, this paper provides a useful and comprehensive decision-support system for WRRFs aiming to implement nutrient recovery strategies. Keywords Anaerobic digestion, decision-support system, mathematical modelling, process optimization, renewable fertilizers, residuals management, resource recovery Roadmap availability The roadmap (excell or graphical version) can be obtained for free by contacting Celine Vaneeckhaute : Research team on green process engineering and biorefineries (BioEngine), Chemical Engineering Department, Universite Laval, 1065 avenue de la Medecine, Pavillon Adrien-Pouliot, Quebec, QC, Canada, G1V 0A6, Tel: , celine.vaneeckhaute@gch.ulaval.ca Introduction A recent review of nutrient recovery technologies (Vaneeckhaute et al., 2017b) has highlighted the potential for nitrogen (N) recovery as ammonium sulfate (AmS) fertilizer (Bonmati and Flotats, 2003) and for phosphorus (P) recovery as struvite, MgNH4PO4:6H2O, and/or calcium (Ca) / magnesium (Mg)- P precipitates (Rahman et al., 2014) at water resource recovery facilities (WRRFs), both from an economic and environmental perspective. Nevertheless, implementation of nutrient recovery technologies is still limited due to regulatory constraints, operational problems associated with the (variability of the) quality and quantity of the bio-based fertilizers produced, as well as the persisting uncertainty of fertilizer sales and inconsistency of marketing prices in regions where commercialization is possible (Seymour, 2009). Indeed, WRRFs must deliver high-value nutrient products that can compete with synthetically produced fertilizers already on the market. Finding the appropriate combination and sequence of technologies to treat a particular residuals stream and the optimal operating conditions for the overall treatment train are key concerns for producing high-quality marketable end-products (Carey et al., 2016; Guest, 2015). This paper aims to provide a roadmap for setting up optimal nutrient recovery trains at WRRFs, taking into account residuals characteristics, regional fertilizer regulations and markets. The scope of the study includes anaerobic digestion and the selected best available technologies applied at full-scale for the recovery of nutrients as marketable fertilizer commodities (Vaneeckhaute et al., 2017b), i.e. P precipitation/crystallization (end-products: struvite, Ca/Mg-P precipitates), NH3 stripping and absorption (end-product: AmS fertilizer), and acidic air scrubbing (end-product: AmS fertilizer). The selection of these technologies was made based on the stage of implementation, the technical

2 performance, and financial aspects, on top of the current fertilizer marketing potential. Besides the information acquired in Vaneeckhaute et al. (2017a,b), additional data were obtained from extensive communications with technology providers. Hence, the roadmap is partially based on full-scale operational experience. As such, this paper provides a valuable decision-support system for residuals and wastewater processing utilities considering the implementation of anaerobic digestion and subsequent recovery and recycling of nutrients as marketable agricultural commodities. Methods In a first phase, field and greenhouse trials were performed in order to evaluate the effectiveness of various bio-based fertilization scenarios using the selected products as compared to traditional fertilization strategies using manure and synthetically produced mineral fertilizers (Vaneeckhaute et al., 2014, 2016). Besides the crop yield, nutrient use efficiencies as well as soil organic carbon effects were determined. The economic and ecological benefits of using bio-based fertilizers in agriculture were also defined (Vaneeckhaute et al., 2013). Based on the results, a decision-tree was produced for setting up bio-based fertilization strategies as function of regional fertilizer regulations. In a second phase, an inquiry was performed among various nutrient recovery technology providers in order to define default operational conditions and design parameters for the selected processes. Moreover, information on the feasibility of nutrient recovery processes based on operational experience was compiled. Based on the obtained insights, guidelines were provided for evaluating the feasibility of bio-based fertilizer production as function of the residuals characteristics. In a third phase, a generic nutrient recovery model (NRM) library was developed and validated (Vaneeckhaute et al., 2017a). It includes dynamic mathematical process models for the best nutrient recovery systems currently available as identified in Vaneeckhaute et al. (2017b). In addition, a compatible biological-physicochemical anaerobic digester model was built. The models are based on detailed chemical solution speciation, as well as physicochemical and biochemical reaction kinetics. Model simulations and global sensitivity analyses were performed in order determine the main factors impacting a wide range of 25 performance indicators of a nutrient recovery treatment train, such as methane and biogas production, digestate composition and ph, ammonium sulfate recovery, struvite production, product particle size and density, air and chemical requirements (acid, base), etc. Finally, model-based optimization experiments were performed in order to optimize the design and operating conditions of nutrient recovery treatment trains in view of maximized revenue at minimal environmental impact. The information obtained from the model simulations, global sensitivity analyses and optimization experiments was incorporated in a conceptual algorithm for setting up an optimal treatment train design. Based on all acquired knowledge, a generic three-phase roadmap for implementation of optimal nutrient recovery treatment trains was developed as descibed below. Results and Discussion Based on the experimental and simulations results, the two most important factors determining the optimal treatment train configuration for nutrient recovery are i) regional fertilizer regulations and markets, and ii) residuals characteristics. A three-phase roadmap for setting up nutrient recovery treatment trains as function of these influencing factors is presented and discussed below.

3 PHASE 1: Selecting a bio-based fertilization strategy as function of fertilizer regulations Depending on regional fertilizer regulations, N or P can be the limiting factor for digestate application as organic/organo-mineral fertilizer. This limiting factor determines for which fertilizers the market demand is the highest in the particular region. Overall, based on the analyses performed in this study, in P-poor regions, the agricultural demand for digestate as organic fertilizer, and recovered AmS, struvite and Ca/Mg-P precipitates as mineral fertilizers is expected to be high. In P-saturated regions, among the considered recycled products, the most interesting fertilizers for agriculture are the liquid fraction of digestate as organo-mineral fertilizer and AmS as mineral fertilizer. Following bio-based fertilizer selection, the feasibility of producing these products has to be determined (PHASE 2). PHASE 2: Evaluating the feasibility of bio-based fertilizer production as function of residuals characteristics An important point to consider when designing nutrient recovery treatment trains is the physicochemical characterization of the residuals. Obviously, the macronutrients, especially N and P, of the waste stream have to be quantified in order to verify if there is an interest for N and P recovery. Technology providers confirmed that, based on operational experience, P recovery is only of interest if the P-load is higher than 80 kg d -1, whereas N stripping and absorption only becomes economically feasible at concentrations in the range of or higher than mg N L -1. Moreover, struvite production is only of interest if the residuals stream has an N:P molar ratio higher than 1. If, based on initial N and P measurements, recovery of these nutrients seems feasible, additional physicochemical analyses must be performed in order to design the optimal treatment train (PHASE 3). PHASE 3: Using the conceptual algorithm for designing the optimal treatment train A conceptual algorithm was developed based on simulation results and contact with technology providers. It provides an overview of guidelines for designing nutrient recovery treatment trains, taking into account residuals characteristics (PHASE II) and fertilizer market demands (PHASE I). The following treatment train configurations for nutrient recovery at WRRFs were identified: 1. N and P recovery = not feasible: No action should be taken; 2. P recovery = not feasible, N recovery = feasible: The recommended treatment train configuration includes anaerobic digestion, NH3 stripping and absorption, with optional pre- and post-treatments, depending on the nature of the residuals; 3. P recovery = feasible, N recovery = feasible or not feasible, and a market for struvite is available at acceptable transportation cost: The recommended treatment train configuration includes struvite precipitation, followed by AmS recovery (if N recovery = feasible); 4. P recovery = feasible, N recovery = feasible or not feasible, and the N:P-ratio and/or fertilizer markets are not favourable for struvite recovery: The recommended treatment train configuration depends on the operational conditions of the anaerobic digester : a. Thermophilic digestion, followed by AmS recovery (if N recovery = feasible) and subsequent Ca/Mg-P precipitation; b. Mesophilic or psychrophilic digestion, followed by Ca/Mg-P precipitation and subsequent AmS recovery (if N recovery = feasible).

4 Current work targets the integration of the developed roadmap in a more holistic spatiotemporal decision-support simulation tool, including transports, greenhouse gas emissions and social aspects, on top of the technical, economic and regulatory aspects discussed in this study. Conclusions A generic roadmap for implementation of optimal nutrient recovery teatment trains at WRRFs is presented. It includes: 1. A decision-tree of bio-based fertilization recommendations as function of fertilizer regulations; 2. Guidelines for determining the feasibility of nutrient recovery based on operational experience; 3. A conceptual algorithm for designing optimal nutrient recovery treatment trains as function of the residuals characteristics and fertilizer markets. As such, this paper provides a useful decision-support system for wastewater and residuals processing utilities considering the implementation of nutrient recovery practices. Current work targets the integration of the developed roadmap in a more holistic spatiotemporal decision-support simulation tool. Acknowledgements The first author is funded by the Natural Science and Engineering Research Council of Canada (NSERC) through the award of an NSERC Discovery Grant (RGPIN ). Peter Vanrolleghem holds the Canada Research Chair in Water Quality Modelling. References Bonmati, A., Flotats, X. (2003). Air stripping of ammonia from pig slurry: Characterisation and feasibility as a pre- or post-treatment to mesophilic anaerobic digestion. Waste Manage., 23(3), Carey, D.E., Yang, Y., McNamara, P.J., Mayer, B.K. (2016). Recovery of agricultural nutrients from biorefineries. Bioresource Technol., 215, Guest, J. (2015). Tailoring innovation to achieve financially viable energy and nutrient recovery from wastewater. In : Proceedings of the Government Affairs Conference IWEA, University of Illinois, IL, USA. Rahman, M.M., Salleh, M.A.M., Rashid, U., Ahsan, A., Hossain, M.M., Ra, C.S. (2014). Production of slow release crystal fertilizer from wastewaters through struvite crystallization: A review. Arab. J. Chem., 7(1), Seymour, D. (2009). Can nutrient recovery be a financially sustainable development objective? In : Proceedings of the 2009 Annual Conference of the Pacific North West Clean Water Association, Boise, ID, USA. Vaneeckhaute, C., Claeys F.H.A., Tack, F.M.G., Meers, E., Belia, E., Vanrolleghem, P.A. (2017a). Development, implementation, and validation of a generic nutrient recovery model (NRM) library. Environ. Mod. Softw., 98, 1-40.

5 Vaneeckhaute, C., Ghekiere, G., Michels, E., Vanrolleghem, P.A., Tack, F.M.G., Meers, E. (2014). Assessing nutrient use efficiency and environmental pressure of macro-nutrients in bio-based mineral fertilizers: A review of recent advances and best practices at field scale. Adv. Agron., 128, Vaneeckhaute, C., Janda, J., Vanrolleghem, P.A., Tack, F.M.G., Meers, E. (2016). Phosphorus use efficiency in bio-based fertilizers: Bio-availability and fractionation. Pedosphere, 26(3), Vaneeckhaute, C., Lebuf, V., Michels, E., Belia, E., Vanrolleghem, P.A., Tack, F.M.G., Meers, E. (2017b). Nutrient recovery from digestate: Systematic technology review and product classification. Waste Biomass Valor., 8(1), Vaneeckhaute, C., Meers, E., Michels, E., Buysse, J., Tack, F.M.G. (2013). Ecological and economic benefits of the application of biobased mineral fertilizers in modern agriculture. Biomass Bioenerg., 49,