Selective recovery of ammonium from complex aqueous environments is critical for circular nutrient economy and mitigating eutrophication. Highly selective nanoscale sorbents like Prussian Blue analogs offer an alternative to zeolites, but in continuous-flow systems, maximizing active-material loading often compromises structural stability and sorption kinetics. Here, we present a novel coagulation-assisted stabilization strategy to engineer a mixed-matrix membrane featuring a 50 wt% loading of zinc hexacyanoferrate nanoparticles, a record high loading for water filtration membranes. By doping the polyethersulfone casting solution with Fe³⁺, we induced a reduction in nanoparticle surface charge, triggering controlled intramembrane aggregation. These aggregates fill the evolving membrane voids during phase inversion, prevent nanoparticle leaching, and enable exceptional ammonium removal capacity of 2.3 ± 0.3 g·m⁻² in short contact times (10-26 sec). Crucially, high permeate fluxes may improve convective/dispersive access of NH₄⁺ to exchange sites within the ZnHCF membrane structure, overcoming the diffusion limitations of conventional packed beds. Pilot-scale evaluations (0.613 m²) under dynamic, low-concentration conditions successfully validated the system’s robust reusability and practical viability. This coagulation-based stabilization establishes a highly versatile platform for mixed-matrix membrane engineering, readily adaptable to immobilize diverse functional nanoparticles for a broad spectrum of separation applications, thus bridging the gap between ultra-selective nanomaterials and scalable membrane engineering.

This study conducted a cradle-to-gate life cycle assessment (LCA) of urine-based recovery of nitrogen (N), phosphorus (P), and potassium (K), which are essential nutrients and critical minerals vulnerable to supply risks. The main contribution of this work is the systematic evaluation of individual unit operations, integrated nutrient recovery processes, and alternative chemical input scenarios under consistent assumptions across ten midpoint impact categories. Individual recovery routes were compared, including hollow fiber membrane contactor (HFMC) versus air stripping for N, precipitation versus anion exchange (AEX) for P, and evaporation versus zeolite ion exchange (ZIX) for K, alongside two integrated NPK configurations: Process 1 (HFMC–ZIX–Precipitation) and Process 2 (Air Stripping–Precipitation–Evaporation). For separate nutrient recovery, air stripping, precipitation, and ZIX, lower impacts were observed in most of the ten impact categories for N, P, and K recovery, respectively. For integrated NPK recovery, Process 2 showed lower mean global warming potential (53.5 vs. 56.9 kg CO₂ eq) and ecotoxicity (172 vs. 233 CTUe); however, a Monte Carlo uncertainty analysis indicated no statistically significant differences across the ten impact categories. Sensitivity analysis further revealed that chemical inputs, particularly alkaline reagents used for pH adjustment in ammonia recovery, dominate variability in results in both processes, with electricity demand emerging as an additional key driver for the Process 2 configuration. Overall, ammonia recovery was identified as the primary environmental hotspot, underscoring the importance of reducing chemical demand; therefore, future optimization efforts should prioritize onsite chemical recovery, alternative pH adjustment strategies, and the integration of renewable energy to improve the sustainability of urine-based nutrient recovery processes and support circular economy objectives.

Desalination of brackish water is crucial for addressing the growing global water shortage. Although reverse osmosis is widely used, its indiscriminate solute rejection necessitates additional post-treatment processes such as remineralization. Therefore, ion-selective desalination using electrodialysis has emerged as a promising technology. However, the mechanisms behind ion selectivity in electrodialysis are not fully understood, particularly in the desalination of natural water sources. In this study, we desalinated natural brackish groundwater using an electrodialysis system, focusing on ion selectivity under varying applied currents. We find that the system becomes more monovalent selective with increasing current densities, aligning with previous studies. To further investigate this phenomenon, we developed a theoretical model that incorporates explicit mass transport calculations in the flow channels and the ion exchange membranes. Our model reveals that the commonly assumed formation of thin stagnant diffusion layers near the membrane surface does not hold in our system. Furthermore, our findings indicate that boundary layers alone cannot fully explain the observed ion selectivity, suggesting the need for a more comprehensive understanding of additional local mechanisms. We analyzed two potential mechanisms, non-equilibrium effects on ion partitioning and ion transport within the membranes, suggesting that non-equilibrium effects on ion partitioning better explain the experimental measurements. This study underscores the importance of explicitly accounting for mass transport within the electrodialysis flow channels and highlights the need for future research on the local processes within the membranes and at the channel/membrane interfaces.

Hollow fiber membrane contactors (HFMC) can recover high-purity ammonium (liquid) fertilizer from wastewater with low energy and area footprint. Previous studies examined factors such as pH, initial ammonia concentration, flow rate, stream configuration, and acid-stripping solution. However, water flux through the membrane, impacting %N in acid stripping, remains a key barrier to producing commercial-grade liquid fertilizer. This study tested %N enhancement by increasing feed-side salinity to reduce vapor pressure, thereby reducing undesired water flux to the acid side. Two salinity levels (2 M and 5 M NaCl) representing zeolite regeneration solutions were tested. With 5 M NaCl, water flux was nullified, yielding 12 %N (NH4+) over two cycles, considered very high for this technology. In contrast, 2 M NaCl allowed water flux, achieving only 9 %N. In a third cycle, 5 M NaCl further increased %N to ∼14 %, an unprecedented result for HFMC. Water flux was negative (-0.031 L/m²·h) with 5 M and positive (0.015 L/m²·h) with 2 M NaCl, indicating reverse or forward flow. Ammonia removal efficiency and transfer coefficient (K) remained stable. Furthermore, the membrane prevented ion cross-contamination, producing high-purity liquid fertilizer. Operating at higher salinity, as in ammonia-laden regeneration solutions, may be economically feasible due to solution reusability. This approach optimizes feed properties for highly concentrated liquid fertilizer production.

Wastewater reuse is essential to water supply and conservation efforts. Anaerobic membrane bioreactors (AnMBR) are emerging as an efficient treatment, enabling high effluent quality and low energy consumption. Subsequent reverse osmosis (RO) treatment can achieve potable water quality. Although mineral fouling in RO has been extensively studied, research on iron-based minerals, particularly under anaerobic conditions, remains notably scarce. Specifically, the potential fouling caused by Vivianite (Fe3(PO4)28H2O) - a mineral prone to precipitate and clog pipes of anaerobic wastewater streams - during high recovery RO applied to anaerobic effluents (e.g., following AnMBR) had not been systematically studied. This study used geochemical modeling and filtration experiments to test the potential mineral fouling of ferrous ion-bearing minerals in RO of synthetic and real AnMBR effluents. The geochemical model calculated mineral precipitation potential at pH 5-10 and an 85-95 % recovery ratio. Vivianite and amorphous calcium phosphate (ACP) were the dominant precipitants. Filtration experiments at varying feed Fe2+ concentrations (0-10 mg/L) and pH (6.0-7.1) revealed the significant effect of these parameters on mineral fouling type and extent, agreeing with the geochemical model. Vivianite precipitated at all pH values, significantly reducing (>90 %) at pH 6 compared to pH 6.75 & 7.1. Higher Fe2+ feed concentrations were correlated with a higher extent of vivianite mineral fouling. ACP was observed only at pH 7.1 since it competed with vivianite over P utilization. This research presents the first systematic investigation of vivianite mineral fouling in RO processes, informing process design of anaerobic effluent water reuse processes.

Applying nanofiltration (NF) membranes to treat tertiary wastewater effluent can significantly upgrade water quality with minimal energy and chemical requirements. However, membrane fouling limits water recovery, typically under 85 %, and generates a large, costly-to-treat retentate. Minimizing fouling is key to significantly increasing recovery and reducing environmental impact. Here, we studied the application of physical hydraulic backwash to achieve ultra-high recovery effluent filtration by controlling fouling, minimizing concentrate volume, and improving water quality. We conducted lab-scale semi-batch effluent NF with/without backwash, targeting over 95 % recovery using backwash-compatible hollow-fiber polyelectrolyte multilayer membranes. Two backwash protocols were tested: 10 % TMP increase and Time (2 h) + 10 % TMP increase. The latter showed better performance and was selected for comparison vs filtration without backwash. Without backwashing, we reached 94 % water recovery before the trans-membrane pressure increased by 40 %. When applying automatic backwash with the permeate, we achieved up to similar to 98 % water recovery, reducing the concentrate volume by similar to 300 % compared to filtration without backwash, with a low energy demand increase (5-7 %). Backwashed filtration consistently produced better water quality that meets the Israeli effluent reuse regulations for unrestricted irrigation and river discharge. Furthermore, we found that the membrane performances can be fully restored and maintained in multiple filtration cycles by applying a standard chemical cleaning-in-place protocol. Our results indicate that backwashing nanofiltration membranes is a practical strategy to control fouling in ultrahigh recovery nanofiltration, significantly reducing concentrate volume and increasing the production of high-quality permeate.

This study explores a novel multi-stage process for recovering valuable nutrients-nitrogen, phosphorus, and potassium-from real hydrolyzed urine as value-added products. The approach utilizes a combination of membrane contactor, zeolite ion exchange, and mineral precipitation techniques. A closed-loop system was established by reusing the acid regeneration solution from ion exchange as the acid-stripping solution in the hollow fiber membrane contactor (HFMC), thereby minimizing chemical usage. Ammonia recovery using the HFMC achieved over 90 % removal across three cycles from hydrolyzed urine. Zeolite columns of chabazite and clinoptilolite demonstrated consistent potassium recovery from HFMC-treated urine, with slightly higher uptake by chabazite compared with clinoptilolite. This suggests zeolite selection can be based on cost and availability. The regeneration of the potassium-saturated zeolite columns using sulfuric acid exhibited rapid and substantial amounts of potassium desorption. Potassium regeneration remained stable over two cycles, with potassium concentrations reaching up to 14 g/L. The release of other ions, such as sodium, was minor compared with potassium, highlighting the minimal impact of sodium interference. The combined ammonia-potassium liquid fertilizer exhibited a favorable N:K mass ratio (3.6 % N and 0.7 % K), with negligible amounts of other ions, making it suitable for facilitating plant growth. Iron phosphate precipitation, a promising alternative resource for fertilizer or lithium iron phosphate batteries, was successfully achieved. Iron doses were more effective in precipitating phosphate at neutral pH than basic pH, reaching over 90 % phosphate removal. This study provides a promising approach for recovering valuable resources from human urine, promoting a more sustainable approach to wastewater management and nutrient recycling.

High water-recovery reverse osmosis (RO) desalination of wastewater effluent is one promising solution to combat the global water scarcity. However, its implementation is often constrained by membrane scaling, with calcium phosphate scaling (Ca-P precipitates) being a primary challenge. This study explores the application of granulated ferric hydroxide (GFH) adsorption as a pretreatment to recover phosphorus (P), a key precursor to CaP precipitates related scaling, from P-rich wastewater effluent thereby facilitating subsequent effluent RO desalination at high water recovery ratios. The results highlight the GFH's strong affinity for phosphate and its partial adsorption of calcium, magnesium, and organic matter from the effluent, which are all advantageous in mitigating fouling and scaling. Optimized acid-base regeneration retained similar to 85 % of P adsorption capacity after five cycles, indicating strong reusability. Moreover, the GFH treatment increased the water recovery ratio from 4 % to 75 % during subsequent RO desalination. Adjusting the GFH-treated effluent pH to 6.5 further improved the water recovery ratio to > 90 %. These results can be attributed to scaling mitigation, as confirmed by surface analyses and saturation index calculations. Overall, these findings highlight P adsorption as an effective and sustainable pretreatment approach for wastewater effluent RO desalination, enabling desalination at a high water recovery ratio while simultaneously achieving efficient P recovery.

Removing iron, aluminum, and magnesium ions from the primary wet-process phosphoric acid solution (WPA) is a promising strategy for mitigating scaling in downstream processing. Here, Donnan dialysis using Nafion membranes, with HNO3 as the stripping solution, was proposed and evaluated. The species in WPA as a function of temperature were modeled. The influence of HNO(3 )molarity (1-5 M), temperature (25-65(degrees) C), and Nafion membrane type on the metal ion mass transport was systematically studied. The effect of H(3)PO(4 )molarity on the metal ion mass transport was analyzed with synthetic WPA solutions, considering the form (free cation or complexed with anions) in which these metal ions transport through the membrane by analyzing the ionic composition of membranes in equilibrium with WPA. The coupled cation diffusion coefficients in membranes are modeled via the Nernst - Planck flux equation. It is found that the cation flux always follows this order: JMg > JAl > JFe for WPA. These factors can be attributed to: 1) in WPA Mg appears mostly as Mg2+, while almost all Fe ions and around half the Al ions are in a paired state, giving the bulky Fe(H2PO4)(2+), Al(SO4)(+ )and Al(SO4)Z, respectively; 2) only solution viscosity influences the Mg2+ transport, while both the solution viscosity and the complexation with anions influence Fe and Al transport; 3) Mg2+ ions have the largest diffusion coefficient in the membrane phase. These metal ions transport in the form of free cations through the membranes. It is concluded that Donnan dialysis is suitable for removing the fast-diffusing Mg2+ ions from diluted WPA, and increasing the solution temperature could enhance the Mg2+ transport rate.

Ammonia-rich saline wastewater is a common byproduct of the pharmaceutical industry. Hollow fiber membrane contactors (HFMCs) show potential for NH3 3 recovery, enabling valuable product recycling. Our previous work indicated that excessive NaOH usage and increased salinity of the treated effluent are major setbacks when employing HFMCs for NH3 3 recovery from pharma wastewater. Therefore, finding alternative approaches to reduce chemical use and salinity is vital. This study demonstrates the integration of bipolar membrane electrodialysis (BMED) with HFMC for NH3 3 recovery, chemical recycling, and salinity reduction. This novel, highly circular concept was tested using real pharmaceutical wastewater. We varied the volume ratios between compartments to assess the BMED flexibility in attaining different treatment goals. BMED achieved over 90% salinity reduction and produced concentrated base and acid in all volume ratios tested, saving 70% of the NaOH required for pH increase before the HFMC step. The HFMC achieved over 98% NH3 3 recovery in two sets of five consecutive cycles, using either a BMED-generated base or NaOH. A preliminary economic analysis revealed a 47% reduction and 86% increase in expenses on chemicals and energy, respectively, compared to HFMC without BMED. Nevertheless, since the largest operating expense was the purchase of chemicals, integrating the BMED step reduced the overall operating expenses of the treatment by 33% (from $3.76 to $2.54 per kg N). The treated effluent's quality allows discharge to conventional wastewater treatment plants, resulting in economic and environmental benefits. This work highlights the importance of innovative separation processes in advancing toward cleaner drug manufacturing.

Predicting ion uptake and selectivity in ion-exchange membranes is desired for many applications, yet a suitable physical description defining the most appropriate ion-specific parameters is still challenging. Here, we systematically develop an ion-association-based approach to modeling ion uptake in ion-exchange membranes from solutions of symmetric and non-symmetric salts. The model treats association in an ion-specific manner, self-consistently accounting for equilibria between free ions in solution and within the membrane phase (salt injection) and between free and associated species within the membrane (association equilibria), subjects to overall membrane electroneutrality. The resulting models, including different possible association equilibria, were employed to fit the reported data for Nafion 117 and CR61 cation-exchange membranes in equilibrium with NaCl, MgCl2, CaCl2, and Na2SO4 single-salt solutions. The results are compared with the previously reported fits to the Manning condensation model, which shows that both models produce similarly good fits for NaCl, MgCl2, and CaCl2 solutions in the 0.01 to 1 M range. However, the greater flexibility and specificity of the association model allow addressing deviations observed for Na2SO4 solutions and for CaCl2 above 1 M as free-ion paring and possible formation of charged NaSO4- and CaCl+ pairs, respectively. The results demonstrate the present model may be a sound non-mean-field alternative to the Manning condensation model, capable of addressing ion-specificity and multiple modes of association.

This work describes the preparation of a monovalent selective cation-exchange membrane (MSCEM) with high selectivity, stability, and low resistance compared to state-of-the-art membranes. These key properties were successfully improved by structural tuning of cross-linking chemistry between the selected polyelectrolyte, polyethyleneimine (PEI), and cation exchange membrane (CEM) in a controlled manner. A facile three-step single-side modification approach was implemented, where the bio-inspired polymer polydopamine (PDA) strongly binds PEI to the CEM, followed by Glutaraldehyde (GA) cross-linking via Schiff base reaction. The change in surface morphology, physico-chemical, and electrochemical properties of the modified membranes was investigated. Selectivity of the prepared membranes was tested at varying current densities (i.e., 30 %, 50 %, and 70 % limiting current density) with two different synthetic solutions mimicking the cation composition of brackish groundwater and reverse osmosis concentrate. The observed permselectivities at different current densities and compositions ranged between 2-3.5 (Na+ over Ca2+) and 4-16 (Na+ over Mg2+) - comparable to best-in-class commercial MSCEM. The electric resistance was towards the lower range of the commercial MSCEM. The modified membranes were stable after 8 operation cycles with almost constant selectivity.

Brackish water desalination is imperative for meeting water demands in arid regions far from the seashore. Reverse osmosis, the leading desalination technology, removes nearly all calcium and magnesium ions, which are essential in drinking and irrigation water. Multistep process schemes combining reverse osmosis with ion-selective membrane processes can maintain or reintroduce these minerals without external chemical addition. Previous efforts emphasized membrane processes that retain multivalent ions, focusing primarily on nanofiltration and monovalent-selective electrodialysis. The potential of processes where monovalent ions are retained and divalent preferentially transported through the membrane has not been studied systematically. Here, we explored applying divalent-selective electrodialysis to transfer calcium and magnesium from the influent into the brackish water reverse osmosis permeate. This novel concept enables chemical-free remineralization of the reverse osmosis permeate while reducing membrane scaling by sparingly soluble calcium salts. We tested this concept experimentally using commercial membranes and natural brackish water, evaluated the product water based on quality criteria for domestic and agricultural use, and assessed the techno-economic feasibility. We found that water suitable for potable use and irrigation can be attained at a reasonable cost, depending on current density. These findings highlight the need for more research on divalent-selective electrodialysis and offer future directions.

Monopolar membrane-assisted electrolyzers enable water electrolysis using acid-alkali asymmetric electrolytes. However, understanding how such an electrolyzer works remains a significant challenge. By assessing the concentration-polarization state in membranes, measuring the ion concentration change in electrolytes, and determining the corresponding transmembrane resistance, we reveal that this electrolyzer can prevent the negative effect of the water dissociation process. The electrolyzer functions by the chemical potential gradient between the asymmetric electrolytes. Briefly, the delta in pH between asymmetric electrolytes significantly modifies the reversible hydrogen electrode potential at both electrolyte compartments and electrodes, and therefore decreases the required external potential. Notably, the unavoidable ion diffusion slightly reduces this positive effect. The electrolyzer performance depends on the membrane property, working temperature, elec-trolyte compositions as well as electrocatalysts. When adopted with state-of-the-art electrocatalysts, this elec-trolyzer achieves an industrially relevant current density of 200 mA cm-2 at a cell voltage of only 1.39 V, outperforming most conventional water electrolyzers, and to the best of our knowledge also those fed by asymmetric electrolytes. Overall, this work highlights the promise of coupling chemical potential energy and electrical energy for hydrogen production, which provides a new strategy to lower the potential for driving water splitting.

Drinking water contamination by per- and polyfluorinated alkyl substances (PFAS) is a global concern. Nanofiltration is a promising PFAS removal technology due to its scalability and cost-effectiveness. However, nanofiltration cannot typically reduce PFAS concentrations below current drinking water recommendations. To enhance PFAS removal, we developed mixed-matrix-composite nanofiltration (MMCNF) membranes—an active nanofiltration layer on porous adsorptive support that synergetically combines filtration and adsorption. We synthesized MMCNF membranes comprising thin polyelectrolyte multilayer films deposited on thick (~400 µm) polyethersulfone supports incorporating β-cyclodextrin microparticles. These membranes achieved near complete removal (>99.9%) of model PFAS (PFOA: perfluorooctanoic acid) for significantly longer filtration times compared to a control membrane without β-cyclodextrin, but otherwise identical. The spent MMCNF membrane was regenerated using ethanol, and high PFOA removal performance was regained during three filtration cycles. Perfluorooctanoic acid was concentrated 38-fold in the ethanol eluent. Further concentration by evaporation is straightforward and can enable eluent recycling and effective PFAS removal.

While the detrimental effect of concentration polarization (CP) on water flux and solute rejection in pressure-driven membrane processes has been extensively explored, the impact of CP on the selectivity between solutes in these processes has been somewhat overlooked. Considering the growing interest in solute- solute selectivity, in this study, we explored the effect of CP on ion-ion selectivity in nanofiltration (NF) membranes. We first show and discuss the "reversed" observed rejection trend of monovalent cations in NF, which is opposite to the trend of the ions' hydrated size and mobility in solution. Next, we apply the film theory using three independent approaches to evaluate the extent of CP in the boundary layer adjacent to the membrane surface, from which the real rejection of the ions can be calculated. Our calculated real rejections of monovalent cations, which were in higher correspondence with the ions' hydration properties and mobility in solution, suggest that CP played a major role in the "reversed" selectivity observed. Last, we demonstrate how CP adversely affects the commonly pursued monovalent-divalent ion separation in NF. Overall, our results highlight the necessity to rigorously account for CP in future studies on NF and suggest minimizing CP as a primary step to improve the selectivity between solutes.

Separation of specific ions from water could enable recovery and reuse of essential metals and nutrients, but established membrane technologies lack the high-precision selectivity needed to facilitate a circular resource economy. In this work, we investigate whether the cation/cation selectivity of a composite cation-exchange membrane (CEM), or a thin polymer selective layer on top of a CEM, may be limited by the mass transfer resistance of the underlying CEM. In our analysis, we utilize a layer-by-layer technique to modify CEMs with a thin polymer selective layer (∼50 nm) that has previously shown high selectivity toward copper over similarly sized metals. While these composite membranes have a CuCl2/MgCl2 selectivity up to 33 times larger than unmodified CEMs in diffusion dialysis, our estimates suggest that eliminating resistance from the underlying CEM could further increase selectivity twofold. In contrast, the CEM base layer has a smaller effect on the selectivity of these composite membranes in electrodialysis, although these effects could become more pronounced for ultrathin or highly conductive selective layers. Our results highlight that base layer resistance prevents selectivity factors from being comparable across diffusion dialysis and electrodialysis, and CEMs with low resistance are necessary for providing highly precise separations with composite CEMs.

Production of certain pharmaceuticals generates wastewater with a high ammonia load, posing economic and environmental burdens. Recovering ammonia as a nitrogen fertilizer can alleviate these burdens and is vital from a circular economy perspective. For this purpose, this study, jointly performed by a research university and a large pharma company, evaluated the use of Hollow fiber membrane contactors (HFMC). The experiments were conducted using real pharmaceutical wastewater in two separate laboratories at two process scales (bench and pilot) and two operating modes (batch and continuous). The study evaluated the effect of feed concentration, flow rate, and process configuration and described the results using a single-parameter (mass transfer coefficient K) mathematical model. We then used the model to design scaled-up batch and continuous processes, based on which we performed a detailed economic assessment. A robust and consistent ammonia removal and recovery were obtained for all the experimental conditions tested. The model adequately described the empirical results, and K was similar across all the experiments in this study, which were conducted in practically relevant conditions. Both batch and continuous process designs were found feasible, increasing process flexibility. The economic evaluation revealed that acid and base consumption, rather than the initial investment, is the most significant expense. The cost of membranes, although relatively high, had little contribution to the overall expenses. Revenues from selling the recovered nitrogen as ammonium-sulfate fertilizer reduces the evaluated net cost of the novel treatment to a minimum of $3.76 per kgN - a potential cost savings of up to 29% relative to the current treatment ($5.28). Therefore, applying HFMC to recover nitrogen from pharmaceutical wastewater is promising from both economic and environmental viewpoints.

Reactive transport modeling refers to the coupling of chemical equilibrium reactions with rate-dependent phenomena such as transport and phase change. The concept originated and is widely used in geosciences but has been infiltrating other fields. Reactive transport can contribute to model prediction and fundamental understanding of transport phenomena by illuminating the interrelations between species distribution of solutes and their displacement. Despite many potential benefits, the Chemical Environmental Engineering field has been slow at adopting the reactive transport approach. Here we focus on reactive transport modeling of membrane separation processes in aqueous solutions for environmental applications. We briefly review the progress made in the last decades for selected technologies and discuss the potential in further developing reactive transport modeling for each case. We then discuss several common misunderstandings and knowledge gaps while highlighting the challenges and opportunities of implementing the reactive transport approach in membrane transport modeling and simulation.

Phosphoric acid production generates large volumes of industrial wastewater that cannot be treated efficiently by existing processes because of its low pH and high precipitation potential. At present, the wastewater is generally stored in evaporation ponds that are prone to breaches, leakage, and flooding. We developed an alternative three-step process for the treatment of phosphoric acid wastewater including selective electrodialysis, reverse osmosis, and neutralization. Testing the process with synthetic wastewater yielded promising results. An exceptional Na/Ca selectivity (up to 18.3 was observed in low-pH electrodialysis, enabling the separation of concentrated H2SO4 without gypsum scaling. Sulfate removal from the electrodialysis diluate prevented scaling in the subsequent high-recovery (>90%) reverse osmosis step, generating high-quality water. A final reaction between the reverse osmosis concentrate and natural phosphate rock enabled P recovery and neutralization of remaining acidity. The electric-power requirement of the process was estimated to be 4.4 kWh per m3 of wastewater, from which 0.78 m3 of clean water, ~3 kg H2SO4, and ~2.5 kg P were recovered. Overall, lab-scale results indicate that this process would be a sustainable and techno-economically viable solution for the treatment of hazardous wastewater byproducts of the phosphoric acid industry