Corrigendum to Building an operational framework for nitrogen recovery via electrochemical stripping [Water Research, 169 (2020), 115226]

Electrified Ion Exchange Enabled by Water Dissociation in Bipolar Membranes for Nitrogen Recovery from Source-Separated Urine

“…We continued the recovery process without electricity (but with electrolyte circulation) to achieve higher recovery efficiency (80%) during the second and third regeneration steps. The slower recovery (80% in 48 h) than regeneration (100% migration in 6 h) indicated that ammonia diffusion across the GPM was rate-limiting for the AMCT configuration, consistent with previous electrochemical stripping (ECS) studies without resin. , During recovery, ammonia diffusion into the trap solution will increase the solution pH (initially sulfuric acid) due to proton consumption that converts ammonia (NH 3 ) to ammonium (NH 4 + ). However, the trap solution pH stayed below 1.5 during cyclic nitrogen recovery tests (Figure S16) because protons in the trap solution (20 mmol proton in 200 mL pH 1 sulfuric acid) were more abundant than the recovered ammonium (8 mmol; 200 mL of 800 mg/L ammonium in the trap solution as indicated in Figure c).…”

“…To evaluate EXS robustness with varying urine compositions, we also tested synthetic urine with twice as much ammonium (∼8000 mg/L NH 4 + ) as the real urine collected in this study (∼4000 mg/L NH 4 + ). Our previous nitrogen recovery studies using ECS indicated that the recovery efficiency varied with the influent nitrogen concentration. , We expected EXS to be more compatible with fluctuating nitrogen concentrations because it separates nitrogen from urine onto resins; indeed, three nearly identical ammonium breakthrough curves were again observed with concentrated synthetic urine, albeit with half as many bed volumes (Figure S18a). The mass of ammonium removed by the resin was identical for synthetic and real urine (around 300 mg NH 4 + for the 12 mL resin used, Figure S18b), which demonstrated that resin equalized nitrogen loading from different streams.…”

“…The middle chamber electrolyte was maintained at 50 mM because we observed low transference of NH 4 + with increasing middle electrolyte concentration (decreasing NH 4 + /K + concentration ratio). Thus, a trade-off exists between chemical inputs (electrolyte conductivity) and energy consumption, which has been reported for conventional ECS . However, EXS enables electrolyte engineering to optimize energy consumption without using unamended wastewater as the electrolyte, which is distinct from both ECS and other BPM-assisted electrochemical processes. , To identify strategies for minimizing both energy consumption and chemical inputs (overcoming the trade-off), we performed preliminary feasibility tests for reusing EXS electrolytes (details in Section S4).…”

“…Compared to BPM-assisted electrochemical processes (without IX resin), BPM-assisted IX processes are advantageous for nitrogen recovery from urine because they can avoid using unamended wastewater as an electrolyte. BPM-assisted electrochemical processes have enhanced ammonia stripping from urine with electrochemical pH adjustment, such as BPM electrodialysis with membrane contactor and bipolar electrodialysis. , However, these BPM-assisted electrochemical processes typically use unamended wastewater (e.g., urine) as an electrolyte, which inevitably influences system energy efficiency when wastewater composition changes (particularly due to conductivity). , Urine typically exhibits fluctuating compositions when collected from different people, locations, time periods, and flush water dilution ratios. ,, Thus, using urine as an electrolyte could cause highly variable energy efficiency . IX resins can be leveraged to separate and equalize urine nitrogen flows, while facilitating integration with electrochemical nitrogen recovery using a more stable electrolyte composition.…”