Techno-economic and life cycle analysis of circular phosphorus systems in agriculture
Graphical abstract
Introduction
Food security and water scarcity are two of the most pressing problems that humanity is currently facing (Rosegrant, 2003; Mekonnen and Hoekstra, 2016). Populations of specific socio-economic and geographical backgrounds are affected unequally by these issues (D'Odorico et al., 2019). The deficiency and excess of a single nutrient i.e. Phosphorus (P) can be traced as a major contributor to both these problems. Consumption and replenishment of P occur at vastly different time scales, which leads to phosphate rock reserves diminishing in volume (Cordell et al., 2011). Thus, the unequal geographic availability of P reserves, their finite stocks in nature, and our inability to substitute P in food-chains ties P directly to the issue of food security. Despite the diminishing mineral reserves, agricultural systems abound in phosphate (Cordell et al., 2009). About 46 % of all loss of mined P, occurs through agricultural runoff (Rittmann et al., 2011). This P then becomes a limiting nutrient in algal blooms and causes eutrophication and anoxia in watersheds downstream (Elser et al., 2007). Eutrophication leads to hypoxia, causing the formation of dead zones in water bodies and the loss of fish habitat (Duce et al., 2008). Annually, 2.2 billion US$ are spent to purify eutrophication affected freshwater, and recover threatened species (Dodds et al., 2009).
A dynamic balance between N and P in watersheds determines which of these two will limit algal growth. The bloom of cyanobacteria is the principle cause behind eutrophication in lakes (Conley et al., 2009). Cyanobacteria recycle N by fixing it, thereby not letting the demand of N hinder their growth (Howarth and Paerl, 2008; Ryther and Dunstan, 1971). This leads to the availability of P, limiting eutrophication in freshwater bodies such as lakes. Higher salinities however, hinder N fixation by cyanobacteria (Howarth and Marino, 2006). Additionally, rapid P recycling by sediments and other hydrological phenomena, ensure the abundance of P in coastal regions and estuaries. The combination of abundance of P and absence of N, causes N to limit algal growth in saline waters.
In addition to accumulation in soil and water from farming practices, waste P is released to the environment by losses throughout the anthropogenic P cycle. Only 16 % of the P applied as fertilizer to crops, remains in the final food product, due to losses at all stages in the food system (Rittmann et al., 2011). In the face of dwindling reserves, only reducing demand and increasing use efficiency are not enough to meet projected demands. Thus, transforming anthropogenic P use from a linear to circular economy assumes paramount importance. Specifically, recycling P from agricultural systems seems to be the obvious solution in tackling the deficit and the excess, as well as the imbalance in P bioavailability in agricultural systems worldwide (Rowe et al., 2016). P recycling lowers the eutrophication potential of watersheds downstream to farms, while simultaneously providing an alternative to mined P. In addition, the opportunity of co-recovery of valuable products or co-generation of energy during the recycling process is an additional incentive to recycle P.
The Clean Water Act tasks the U.S. Environmental Protection Agency (USEPA) with monitoring and ensuring surface water quality in the United States. In 2011, EPA committed to partnering with states and stakeholders to reduce nutrient pollution (United States Environmental Protection Agency, 2011). This was accompanied by a framework recommending how to determine goal nutrient loads, target watersheds, and report loads. The agency has since maintained a database of the water quality of surface waters across the country. These monitoring data are available on the Assessment and Total Maximum Daily Load Tracking and Implementation System (ATTAINS) (US EPA, 2022). The methods of target load calculation have however, evolved utilizing a combination of both mechanistic models and stressor response models (Huo et al., 2018). The ability of watersheds to recover from nutrient overload can also be evaluated with EPA's Recovery Potential Screening (RPS) tool. Despite these efforts, USA has made little progress in transitioning to a nutrient-circular agricultural system (Peterson et al., 2022). This has led EPA to partner with USDA leadership, states, tribes, territories, and stakeholders to tighten nutrient pollution control (United States Environmental Protection Agency, 2022). The importance of market driven approaches such as water quality trading, third-party credit aggregation and banking in driving investments in clean water technologies has been stressed in the memorandum thus issued. Scientific literature has also made policy recommendations to the EPA such as the establishment of a Federal Advisory Committee on a circular economy, increasing funding and deregulating possible P circular products, to close the nutrient cycle (Goswami and Rouff, 2022).
The gaps in knowledge with regards to closing the P cycle have thus naturally been studied widely. Multiple studies point to the need of a circular approach to P recovery (Cordell et al., 2011; Sarvajayakesavalu et al., 2018; Roy, 2017). The current literature has also identified how actors along the value chain can adopt themselves to design a circular system which drives itself to resilience, efficiency, and sustainability (Withers et al., 2018). Even though none have been implemented in the field, various technological and ecological options for P removal and recovery exist. These have been described in detail in the SI.
The degree of nutrient removal, the economic feasibility, and environmental impact of deployment of a technology, are three criteria which dictate the choice of nutrient removal option to be included in WWTPs or agricultural systems. Material flow analyses are particularly helpful to assess the degree of nutrient removal. We review various mass flow analysis (MFA) studies that analyze linear and circular P systems at the local, and country scales (Egle et al., 2015; Liu et al., 2004; Mehr et al., 2018; Pham et al., 2017). These are described in detail in the background section of the SI. The economic prospect of P recycling has been looked into by Sharpley et al. (2001) They tally the cost-effectiveness of best management practices against their efficiency. Vollaro evaluates the economic feasibility of using recycled P from technologies retrofit to wastewater treatment plants as an alternative to mineral fertilizer (Vollaro et al., 2016). Robles et al. and Sena et al. review the economics of P precipitated from wastewater (Robles et al., 2020; Sena et al., 2020). These are summarized in Table S1 in the SI. Similarly, many authors have performed life-cycle assessment (LCA) of technologies to recover P from urban wastewater. Table S1 in SI enlists some of these works. These studies focus on the reclamation of P from point sources which have a high concentration of the nutrient. P recycling from non-point or diffuse sources like agricultural systems however, is much more challenging due to its high dilution in runoff. The only demonstrated solution that reclaims nutrients from runoff is the Enhanced Biological Phosphorus Removal and Recovery system (EBP2R), that couples an anaerobic and aerobic reactor to produce growth medium for the cultivation of green microalgae (Valverde-Pérez et al., 2015a; Valverde-Pérez et al., 2015b; Valverde-Pérez et al., 2016). The concerns of heavy metal accumulation and NOx emissions hinder the direct agricultural application of EBP2R sludge (Kaljunen et al., 2022). Additionally, besides the life cycle impact of EBP2R systems, the economic and environmental impact analysis of P recycling in agricultural context has not been studied (Fang et al., 2016). These studies are crucial to determine the bottlenecks to recycling agricultural P, and to compare candidate solutions.
The application of technologies for P reclamation is marred by severe hurdles. Therefore, preliminary investigations into candidate technologies are critical, before they can be subjected to field deployment. For example, P removal by adsorption is particularly challenging since runoff cannot be filtered or pre-treated before it reaches the adsorbent bed. This shortens life of P sorption media (PSM) by premature clogging. In addition, most sorption media suffer from an imbalance of area requirement and hydraulic conductivity. Hydraulic conductivity, and hence nutrient removal rate are both high for highly porous media. However, water retention suffers due to this high porosity. Therefore, the use of porous media necessitates a large amount of land area for treatment of high flow events, which deliver majority of the nutrient loss (Penn et al., 2017a). On the other hand, materials with lower porosity and smaller land area requirement offer higher retention time at the cost of removal efficiency. These constraints, along with the lack of guidance and heuristics, make field-scale deployment of nutrient removal structures difficult. Phosphorus Transport Reduction App (P-TRAP) software is one of the few tools available, to design P removal structures for a removal target within a specified life-time, given a particular PSM (U.S. Department of Agriculture, 2021). Flow conditions, nutrient concentration and the design of the drainage system (surface, subsurface or tile) dictate the choice of PSM. Even though research in the area of flow-through testing and scale-up is ongoing, field testing is expensive and challenging (Shedekar et al., 2020; Penn et al., 2020; Gonzalez et al., 2020). The economics of P removal from runoff is the major constraint to deployment. The intervention of government policies and incentives is required unless the cost of recovered P can compete with that of mined P. Additionally, the setup of removal structures and disposal of novel materials to be used as sorption media may cause environmental burden. Therefore, the choice of technology, economic constraints and life-cycle impact of deployment must be evaluated thoroughly before field testing can commence.
An overwhelming majority of research on nutrient-circular agricultural systems is devoted to the reclamation of nutrients from point sources (Vollaro et al., 2016; Robles et al., 2020; Bradford-Hartke et al., 2012; Linderholm et al., 2012; Bradford-Hartke et al., 2015; Daneshgar et al., 2019; Liu et al., 2011; Altamira-Algarra et al., 2022). These sources include livestock waste, manure, treated sludge etc. P, recovered as struvite from WWTPs has been found to be economically feasible and attractive, which has been implemented in practice (Achilleos et al., 2022; Yetilmezsoy et al., 2017). This study however, operates in the context of diffuse sources of P, the total P content of which are in the range 0.01–3.5 mg/L (Rattan et al., 2021; Lang et al., 2013), compared to 5–20 mg/L in municipal wastewaters (Kõiv et al., 2010). Thus, the amount of runoff that must be processed for recovery of equivalent P is about an order of magnitude greater for agricultural systems than urban wastewater treatment systems. Thus, diffuse P treatment facilities would produce a smaller throughput of recovered P through their lifespan at the same processing power, compared to urban or municipal WWTPs. The conclusions from WWTP centric articles thus are not applicable for our study. The economic infeasibility of diffuse P recovery, on the other hand, has been pinpointed already. Rosemarin et al. summarize the literature in this area and advocate for policy that internalizes the environmental impacts of fertilizer use, if recycled nutrients are ever to compete economically with artificial supplements (Rosemarin et al., 2020). However, the loss of nutrients through agricultural runoff and the application of circular strategies to tackle it has not been studied conclusively. Nutrient-circular systems utilizing point source of nutrients have been shown to benefit human health and aid emission abatement (Tonini et al., 2019a). A similar effect for nutrient circularization from non-point sources remains to be investigated. The only class of solutions investigated in detail in this regard are EBP2R systems (Valverde-Pérez et al., 2015a; Valverde-Pérez et al., 2015b; Valverde-Pérez et al., 2016; Fang et al., 2016). We add to this literature by investigating the economics of hitherto unstudied P recovery systems such as pyrolysis of wetland biomass and wetlaculture™.
The recovery of P from sewage sludge in WWTPs is practiced widely and the efficiencies of various methods of recovery have been compared in literature. Egle et al. compare nineteen technologies in the context of wastewater and sludge treatment systems using a combination of methods including MFA and cost analysis (Egle et al., 2016). Chemical precipitation, adsorption, and ion exchange methods that recycle P from aqueous phase, sewage sludge as well as ash are explored in the comparison. The P adsorption efficiency of both waste derived and engineered materials have also been compared at laboratory scale for surface waters, rich in organic matter (Boyer et al., 2011). More specialized techniques such as wet leaching, struvite precipitation, magnetic vivianite separation, sludge melt gasification, the thermochemical sodium sulfate process, and white phosphorus recovery from sewage sludge have been compared on the basis of technological maturity, recovery efficiency and cost (Kaljunen et al., 2022). The landscape of similar studies for non-point sources however, is much more limited. At the laboratory scale, bacterial populations and side stream enhancements across EBPRs have been compared for recovery efficiencies (Onnis-Hayden et al., 2020; Schuler and Jenkins, 2003). Of on-field nutrient removal structures, removal efficiencies for different PSMs have been modelled and compared (Penn et al., 2017b). Nutrient removal efficiencies of wetlands have been compared for varying hydrological parameters such as retention capacities and substrates (Pietro and Ivanoff, 2015; Tang et al., 2009). Studies also compare chemical (precipitation based) and microbial (anaerobic) P removal efficiencies (Leng and Soares, 2022). The reference document by US EPA summarizes nutrient removal technologies to enable decision and policy-making (United States Environmental Protection Agency, 2008).
The life-cycle impact of P recovery as struvite is well studied. Existing studies compare sludge derived P fertilizers to mineral fertilizers, as well as to sludge and sludge ash derived fertilizers (Linderholm et al., 2012; Bradford-Hartke et al., 2015; Pradel and Aissani, 2019; Amann et al., 2018; Nakakubo et al., 2012; Sena et al., 2021; Behjat et al., 2022). Process improvements like hydrothermal treatment, nitrification, struvite precipitation and biological removal improve emissions (Chen et al., 2022; Rashid et al., 2020). Moreover, when externalized societal costs are included in the analysis, nutrient recycling is favored across most impact categories (Tonini et al., 2019b). Such LCA studies do not currently include comparison with pyrolysis coupled systems, even though P recovery has been shown to increase in such cases (Fan et al., 2022; Cui et al., 2019; Gbouri et al., 2022). Finally, even though the merits of wetlaculture™ system have been demonstrated on field, there have been no life-cycle studies investigating its emissions (Altamira-Algarra et al., 2022; Jiang and Mitsch, 2020; Jiang et al., 2021; Boutin et al., 2021; Mitsch, 2017). The focus of this article is therefore, to study novel technologies to remove and reclaim nutrients from diffuse sources i.e. farm runoff. More importantly, we aim to identify emission hotspots and economic bottlenecks in said technologies so that we may guide technological innovation and the expedite the circular revolution.
In this work, we review MFA, LCA and TEA studies conducted for different P circular systems at different spatial scales. We identify specific ecological and technological strategies which can recycle nutrients from diffuse sources in agricultural systems. We calculate the degree of circularity for these strategies using MFA. We evaluate the economic feasibility of these recycle pathways. We also calculate their life-cycle emissions. We identify the processes within these modules that contribute the most to their environmental impact. Thus, scope for reduction in emissions before field scale implementation is recognized. Since implementation issues are difficult to account for in quantitative assessments, we assume perfect scalability of lab-scale results. This work is the first to evaluate economic feasibility and environmental implications of P recovery from agricultural runoff, even though similar studies have been conducted on treatment of urban wastewater.
The rest of the article is organized as follows. We choose four methods for P recycling from agricultural systems. These are: P removal using ion-exchange resins followed by precipitation of phosphate from the resin dewatering solution, P interception by letting runoff pass through wetlands and recovery of the P in char left after wetland biomass pyrolysis (WBP), P removal by passing runoff through bioreactor and recovery of the P in char after pyrolysing the substrate of the Denitrifying Bioreactor (DNBR), and using legacy P left in a wetland by converting it to farmland (wetlaculture™) (Jiang and Mitsch, 2020; Jiang et al., 2021; Boutin et al., 2021). The technologies available for nutrient removal from agricultural runoff operate on one of four basic principles: chemical precipitation, crystallization, adsorption and biological removal (Egle et al., 2015; Karunanithi et al., 2015). Of these principles, adsorption lends itself directly to the possibility of nutrient recovery by regeneration of the adsorbent (Kumar et al., 2019). Precipitation and crystallization have not been directly applied to treat effluent water from non-point sources of P due to the lower concentrations of P in these waters. However, ion-exchange resins adsorb P and allow P recovery from the resin dewatering solution via chemical precipitation. Thus, the inclusion of ion-exchange resins within the analysis covers two removal principles: adsorption and chemical precipitation. Crystallization of phosphates from P rich solutions, yields similar products as precipitation, employing pH control and seeding to induce nucleation instead of chemical precipitants (Peng et al., 2018). The recovery rates of crystallization and precipitation are comparable as well. Thus the ion-exchange process represents crystallization as well. Finally, the biological recovery of P from Enhanced Biological Phosphorus Removal and Recovery (EBP2R) systems has been well demonstrated in current literature and supported with life-cycle studies (Guisasola et al., 2019; Xia et al., 2014; Geng et al., 2018). Moreover, despite the efficiency of EBP2Rs to reach effluent concentrations with ultra-low P concentrations, concerns of heavy metal accumulation and NOx emissions hinder the direct agricultural application of EBP2R sludge (Kaljunen et al., 2022). We have therefore excluded EBP2Rs from this study. Thus in summary, our choice of technologies is suited to represent both technological and ecological means of recovery and all feasible principles of nutrient recovery from diffuse sources.
We describe steady state models for each method and analyze their material flow. This is followed by a discussion of methodology for the techno-economic analysis, the life-cycle analysis and the basis of calculation of circularity metrics. Finally, we interpret results of the TEA and the LCA, comparing costs, life-cycle emissions and circularity metrics of the recycle routes. We also do sensitivity analyses to study the effects of parameters like P removal efficiency and cost factors on the results. We find that the economics and life-cycle impact of ion exchange deem it infeasible, despite its high efficiency. Wetlaculture™ emerges to be the most efficient and cost effective option. Although wetlaculture™ costs may be brought down further, they still remain twice as much as for fertilizer made from mined P.
Section snippets
Material flow analysis
A systems analysis is employed to compare economic and life-cycle environmental impacts across four P recycle pathways. The pathways are illustrated in Fig. 1. We begin with tracking the movement of P through the technology chain for each recycle pathway. This chain involves collection, removal, processing and recovery modules. We construct material flow analyses on these chains, using process descriptions of emerging recovery paths found in the literature. The recycle of an equal amount of P
Material flow analysis
Fig. 3 shows the flow of P through recycle schemes at steady state as Sankey diagrams. Each unit is represented as a node here. Reagents are shown to originate from an “input” node and waste streams go to the “Environment” node. All flows shown are in mass units. Circularity of the ion-exchange route is the greatest (0.68), followed by Wetlaculture™ (0.60), DNBR (0.37), and finally WBP (0.34). Therefore, the ion exchange method provides the greatest opportunity to close the P loop. A version of
Conclusions
We combined P recovery technologies with best management practices (BMPs) and used them to develop circular economy strategies from diffuse nutrient sources in agriculture. A mix of novel technological and ecological recycle options were analyzed. These included using ion exchange resins to capture phosphate and precipitating the phosphate as hydroxyapatite from the P rich solution using NaOH, pyrolyzing the biomass from wetlands to utilize biochar thus obtained as a nutrient rich soil
CRediT authorship contribution statement
Amrita Sen: Conceptualization, Data curation, Investigation, Methodology, Writing – original draft. Bhavik R. Bakshi: Conceptualization, Project administration, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (107)
- et al.
Struvite precipitation within wastewater treatment: a problem or a circular economy opportunity?
Heliyon
(2022) - et al.
A review of technologies for closing the p loop in agriculture runoff: Contributing to the transition towards a circular economy
Ecological Engineering
(2022) - et al.
Environmental impacts of phosphorus recovery from municipal wastewater
Resour. Conserv. Recycl.
(2018) - et al.
Advanced bio-fuels from pyrolysis oil: the impact of economies of scale and use of existing logistic and processing capabilities
Fuel Process. Technology
(2012) - et al.
A meta-analysis of lcas for environmental assessment of a conceptual system: phosphorus recovery from dairy wastewater
J. Clean. Prod.
(2022) - et al.
An evaluation of corn production within a wetlacultureTM system at Buckeye Lake, Ohio
Ecological Engineering
(2021) - et al.
Comparison of low-cost and engineered materials for phosphorus removal from organic-rich surface water
Water Res.
(2011) - et al.
Phosphorus recovery from centralised municipal water recycling plants
Chemical Engineering Research and Design
(2012) - et al.
Analysis of the effects of prepyrolysis hydrothermal treatment on phosphorus recovery from sewage sludge using a life cycle assessment
J. Clean. Prod.
(2022) - et al.
The story of phosphorus: global food security and food for thought
Global Environmental Change
(2009)