Urea and ammonia plants can discharge nitrogen at staggering levels — in the tens of grams per liter. A stepwise characterization and treatability plan shows how to identify every contaminant and select processes that actually work on this mix.
The numbers are blunt. Studies report untreated urea‑plant wastewater with ammonia on the order of 10–30 g/L (10,000–30,000 mg/L) at about pH 10–11 (researchgate.net) (researchgate.net). Even after mixing and dilution, field surveys still find 50–55 mg/L NH₃‑N at urea plant discharge, tapering to 3–13 mg/L downstream (researchgate.net).
The primary load is nitrogen — inorganic and organic — with ammonia (NH₃/NH₄⁺), urea and other organic‑N (as total Kjeldahl nitrogen, TKN), plus nitrate/nitrite, and a supporting cast that can include phosphorus, chloride, sulfate, trace metals, organics, and oil/grease (nepis.epa.gov) (researchgate.net). Gas from the carbamate reaction (CO₂/H₂O) can elevate pH.
Baseline nitrogen loads and co‑contaminants
Initial analysis from one urea‑plant effluent showed high NH₃ and BOD₅ (five‑day biochemical oxygen demand) that made the discharge unsuitable, while metals like Fe, Cu, Cr, and Mn were measurable but low (<0.3 mg/L) (researchgate.net) (researchgate.net). Trace constituents (P, Cl⁻, SO₄²⁻, heavy metals, organics, oil/grease) depend on plant specifics and are usually ≪1–100 mg/L, but they must be tested (nepis.epa.gov) (researchgate.net). For field sampling, removing debris via primary screening (e.g., screens and primary separation) helps protect bench equipment.
APHA/EPA analytical methods and QC
The recommended suite follows APHA/AWWA/EPA methods: field basics (pH, temperature, conductivity); nutrients (NH₄⁺/NH₃, NO₂⁻, NO₃⁻, PO₄³⁻, TKN, urea); organics (COD, BOD₅, total organic carbon/TOC); solids (TSS, turbidity, oil & grease); inorganics (major cations/anions by ion chromatography); and metals (ICP‑AES) (researchgate.net). Standard Methods include 4500‑NH₃ (spectrophotometry, ion‑selective, or phenate), Kjeldahl for TKN, 5220 for COD (dichromate), 5210 for BOD₅, 2540 for TSS/TDS, 4500‑NO₃ (Griess) for nitrite/nitrate, and molybdate‑blue for phosphate. Quality controls and lab replicates are part of the plan. Baseline pH is often alkaline (about 8–11). Laghari et al. measured pH, TDS, TSS, BOD₅, COD, heavy metals (Fe, Cu, Cr, Mn) and NH₃ in urea‑plant effluent (researchgate.net).
Analysis should also compute mass loads (mg/day) using flow, not only concentrations. EPA’s 1979 data set puts urea‑plant effluent at about 120 g/m³ NH₄‑N (~120 mg/L) and 90 g/m³ organic N (nepis.epa.gov), while ammonium nitrate plants saw ~482 mg/L NH₄‑N and 121 mg/L NO₃‑N (nepis.epa.gov). For settling solids at bench scale, samples can be clarified before analysis with equipment analogous to a plant clarifier when scaling up.
Measurable outcomes and decision metrics
Key outputs include percent removal of NH₄‑N, COD/BOD, and the distribution of nitrogen species (NH₄, NO₃, NO₂). For process comparisons, track kinetics (e.g., nitrification rate in mg N/L·h), mg NH₄‑N removed per g VSS (volatile suspended solids, a proxy for biomass), mg NH₄‑N per g adsorbent, or per kWh for stripping. Record influent vs effluent values and removal efficiencies for each bench test. Final effluent concentrations and mass loads support compliance and sizing. (Appendix tables, if any, should report influent flows and concentrations.)
Bench nitrification and denitrification
Biological routes are tested in sequencing‑batch or continuous reactors. Inoculation uses activated sludge or nitrifying biofilm; acclimation to high NH₄‑N proceeds stepwise. Aerated reactors run near ~30°C and ~2 mg/L dissolved oxygen. One lab program increased influent NH₄‑N from 100 to 1500 mg/L, achieving ~94% conversion at 1500 mg/L with effluent around 90 mg/L (pmc.ncbi.nlm.nih.gov). An anoxic denitrification phase with added methanol/acetate reduces NOₓ to N₂; lab‑aged biomass helps tolerate high nitrogen. Partial nitritation is explored by controlling DO/pH to suppress nitrite‑oxidizing bacteria (e.g., DO ~0.3–0.5 mg/L “phantom bypass”) (pmc.ncbi.nlm.nih.gov) (pmc.ncbi.nlm.nih.gov). Bench reactors mirror full‑scale options such as an SBR or a biofilm system like a moving‑bed bioreactor.
Ion exchange and adsorption tests
Batch flasks with clinoptilolite or synthetic zeolite measure capacity against ammonia. Typical trials contact wastewater (e.g., 500–1000 mg/L NH₄‑N) with 1–10 g/L adsorbent for ~1–2 hours, then quantify residual NH₄‑N to calculate mg/g uptake. One study using zeolite‑A (from fly ash) reported ~32–41 mg NH₄‑N per g, with roughly 32–41% removal at 100–500 mg/L initial concentrations (pmc.ncbi.nlm.nih.gov). Regeneration is assessed using, for example, 2 M NaCl. Alongside zeolites, piloting can include resin media such as an ion‑exchange resin for ammonium. Powdered activated carbon (PAC) is also tested, though with lower NH₄ capacity; it remains useful for organics polishing with media like activated carbon.
Struvite precipitation jar tests
Chemical precipitation targets magnesium ammonium phosphate hexahydrate (struvite, MgNH₄PO₄·6H₂O). Jar tests add Mg²⁺ (e.g., MgCl₂ or bittern) and PO₄³⁻ to approximate Mg:N:P ~1:1:1, then adjust pH to ~9–9.5. One study used bittern plus KH₂PO₄ at around pH 9.6 to recover ammonia as struvite (researchgate.net). Supernatant NH₄‑N removal is typically on the order of 50–90%, subject to conditions; solids can be verified by XRD/FTIR. Accurate reagent dosing is aided by a bench‑scale dosing pump.
In scale‑up, precipitated crystals are settled and separated, with compact settlers such as a lamella settler offering footprint advantages. Jar tests should log struvite yield (e.g., g struvite per g Mg or per g NH₄ removed) as a potential fertilizer recovery metric.
Ammonia stripping bench trials
Bench columns or vessels strip ammonia by bubbling air or steam through raw or pH‑adjusted wastewater. Operation at elevated pH (~11) shifts NH₄⁺ toward gaseous NH₃. EPA fact sheets describe removal exceeding 90% under optimized conditions (nepis.epa.gov). Trials should record energy use and off‑gas ammonia concentration; capture via acid scrubbing can produce ammonium sulfate. Supporting equipment (e.g., blowers, columns, and packings) falls under typical treatment ancillaries.
Advanced oxidation for COD polishing
For organics, jar tests can apply Fenton’s reagent (H₂O₂ + Fe²⁺) or UV/H₂O₂. Ammonia is not directly oxidized by Fenton under normal conditions; severe oxidative regimes (e.g., catalytic ozonation) can partially oxidize NH₃ to NO₃. If process‑derived organics are present, advanced oxidation can reduce COD/BOD; report COD/BOD removal. UV delivery hardware, as used in ultraviolet systems, is a common bench and plant platform for UV‑based AOPs.
Sequential and hybrid treatment trains
After single‑unit trials, multi‑stage schemes are tested. Examples include biological nitrification with downstream chemical precipitation, or adsorption followed by oxidation. Lab simulations can, for instance, biologically convert a portion of NH₄‑N and then polish the effluent through an adsorption column; the reverse sequence is also evaluated. Overall performance is judged on total nitrogen and COD removal, reuse water quality, and by‑product recovery potential.
Where membrane polishing is in scope, options span integrated reactors such as a membrane bioreactor to standalone steps like ultrafiltration. Reuse‑driven projects also pilot desalting or polishing via brackish‑water RO, depending on final water quality targets.
From bench metrics to design and compliance
Each unit yields quantitative metrics. Biological tests report NH₄‑N removal (%) and rates (e.g., kg N/m³·d or mg N/g VSS·h). Chen et al. achieved 94.1% NH₄‑N removal with effluent ~90 mg/L at a 1500 mg/L feed (pmc.ncbi.nlm.nih.gov). Adsorption needs capacity and breakthrough curves (e.g., ~41 mg NH₄‑N/g zeolite at pH 6–8) (pmc.ncbi.nlm.nih.gov), struvite trials quantify percent nitrogen recovered and yield, and stripping reports percent volatilized and off‑gas NH₃. Results are compared to effluent targets (for example, Indonesian Permen standards cite Baku Mutu NH₄‑N ~10 mg/L for similar industries). If bench nitrification removes 95% of a 1000 mg/L NH₄‑N load (down to 50 mg/L) but the limit is 10 mg/L, additional polishing — e.g., adsorption or stripping — is justified. These data then inform sizing, reagents, and costs, including whether final polishing should leverage integrated membrane systems.
Sources and methods references
This characterization and treatability plan draws on fertilizer‑industry wastewater studies and standard environmental practice: major nitrogen components (ammonia, nitrate, organic N) are typical of the sector (nepis.epa.gov); ammonia loads range from ~50–55 mg/L at discharge to tens of g/L in concentrated plant streams (researchgate.net) (researchgate.net). Effluent analyses commonly include pH, TSS, TDS, BOD₅/COD, NH₃, and metals (researchgate.net) (link.springer.com); bench nitrification studies show ~94% NH₄ removal at 1500 mg/L load (pmc.ncbi.nlm.nih.gov); zeolite adsorption has reported ~32–41 mg NH₄/g capacities with 30–41% removal at 100–500 mg/L (pmc.ncbi.nlm.nih.gov); struvite crystallization at pH ~9.5–10 efficiently recovers NH₄ as MgNH₄PO₄·6H₂O (researchgate.net). Measurement methods follow accredited Standard Methods, including Nessler or ion‑chromatography for NH₃ (researchgate.net) and PDAB colorimetry for urea. Taken together, the literature and bench protocols map a data‑driven route to treating ammonia‑dominated effluent and, where desired, recovering value.
