Why Denim Washing Wastewater Needs More Than DAF and Biological Treatment
A conventional DAF and biological train reliably drops suspended solids and biodegradable COD from denim wash-house effluent, but it leaves true color, refractory COD, and dissolved process salts in the stream — which is exactly the gap a nanofiltration system for denim washing wastewater is engineered to close. The documented 2026 case study for a denim washing project (S3, 2025-07) confirms the standard train runs DAF → AAO → filtration, and the post-commissioning report explicitly notes the treated water meets discharge standards — not reuse standards.
Denim washing effluent carries indigo oligomers, hydrolyzed reactive azo dyes, sizing agents (PVA, starch), pumice fines, and large amounts of glauber's salt and sodium chloride used to drive dye exhaustion. A typical 2026 wash-house profile lands at COD 800–2,500 mg/L, TSS 200–800 mg/L, sulfate 500–3,000 mg/L, true color in the thousands of Pt-Co units, and pH 9–11 after alkaline desizing. DAF removes 60–85% of TSS; AAO removes 50–75% of soluble COD. Residual color after biology is typically 200–500 Pt-Co — visually still deep blue or yellow-brown.
ZDHC wastewater guidelines and buyer-brand water-stewardship questionnaires in 2026 push the KPI past discharge compliance toward freshwater reduction per kilogram of denim, often targeting 30–50% reuse of wash water. A DAF+biological train alone cannot reach that bar because the remaining color, salt, and recalcitrant COD are dissolved, not suspended. The next step is a membrane polish — and nanofiltration is the pressure-driven process that fits between ultrafiltration and reverse osmosis on this duty, rejecting dye chromophores and divalent salts while letting monovalent ions and water through at tractable feed pressure.
How Nanofiltration Works on Indigo and Azo Dye Effluent
Nanofiltration is a pressure-driven membrane process with a nominal molecular weight cut-off of 200–1,000 Da, sitting in the gap between tight ultrafiltration (1,000–10,000 Da) and reverse osmosis. That cut-off range is the engineering sweet spot for denim effluent: indigo oligomers and hydrolyzed reactive azo dye chromophores fall in the 600–1,500 Da window, so size exclusion alone retains most color bodies. Divalent salts — sulfate, calcium, magnesium — are rejected at 90–98%, while monovalent ions (Na⁺, Cl⁻) pass at 10–50%, which is what makes the process economically viable on textile feed (HydropureWater field data, 2026).
Beyond pore size, commercial polyamide NF membranes carry a slight negative surface charge at neutral pH, so Donnan exclusion adds electrostatic repulsion of anionic dye chromophores. On reactive azo dyes this typically pushes color rejection 5–15 percentage points above what size exclusion alone would predict, and the effect is strongest at pH 6.5–7.5 — which is also the band where downstream biology and spiral-wound elements perform best.
The trade-off is recovery versus osmotic pressure and scale risk. As recovery climbs toward 85%, the concentrate stream becomes saturated with calcium sulfate, and a 2026 scaling model for denim feed predicts calcium sulfate saturation is reached at 70–78% recovery for a feed containing 1,500 mg/L SO₄²⁻ and 400 mg/L Ca²⁺ at 30 °C (HydropureWater field data, 2026). That is why the realistic operating envelope for spiral-wound NF on textile feed is 60–85% recovery per stage, with multi-stage arrays used to keep concentrate volume manageable rather than pushing a single stage to its limit.
Process Flow: DAF, Biological Treatment, and NF in Series
The 2026 process train places nanofiltration as the final polish on a stabilized biological secondary, not as a standalone treatment. Each upstream step is justified by what it removes and what it protects.
- Bar screening and grit removal. Target: 2–6 mm aperture; protects downstream pumps, DAF scrapers, and membrane elements from pumice fragments, broken pocket liners, and fiber lint common in stone-wash and enzyme-wash operations.
- Equalization with pH adjustment. Two-basin EQ with pH correction to 6.5–7.5 using CO₂ or sulfuric acid. Enzymatic desizing swings pH into the 9–11 range; without EQ, AAO nitrification collapses and NF flux becomes erratic.
- Dissolved air flotation. A standard DAF system for denim effluent pre-treatment removes 60–85% of TSS, oils, and floating lint before biology. The S3 case study (2025-07) confirms DAF as the first clarifier in a documented denim train.
- Biological treatment (AAO or MBBR). Targets COD and partial color reduction; biology alone rarely achieves <100 Pt-Co true color, which is why the membrane step is needed.
- Sand or multi-media filtration. Brings SDI below ~5 and turbidity below 1 NTU. A multi-media filter for NF feed protection is the last particulate barrier before the membrane.
- Nanofiltration membrane stage. 2–3 stages of 4-inch or 8-inch spiral-wound NF elements, permeate 60–85% recovery, concentrate 15–40% sent to RO, evaporator, or brine handling.
- Optional RO polish. For mills targeting boiler-feed reuse; otherwise NF permeate returns to the wash line for rinsing and stone-washing.
| Stage | Equipment | Target Contaminant | Design Output |
|---|---|---|---|
| 1 | Bar screen | Lint, pumice, fiber | >2 mm capture |
| 2 | EQ basin + pH correction | pH swing, flow variation | pH 6.5–7.5 |
| 3 | DAF | TSS, oils, floatables | <100 mg/L TSS |
| 4 | AAO / MBBR | Soluble COD, partial color | 50–75% COD removal |
| 5 | Multi-media filter | Particulates, SDI | SDI <5, turbidity <1 NTU |
| 6 | NF membrane array | Color, sulfate, hardness | 60–85% permeate recovery |
| 7 | RO polish (optional) | Monovalent salts | Reuse-grade TDS |
NF Module Design Parameters for Denim Duty in 2026
The spec below is the operating envelope a process engineer can put on a 2026 purchase order for a 50–200 m³/h denim wash house. Membrane type is polyamide thin-film composite, NF270 or equivalent, chosen for high divalent rejection (≥97% MgSO₄), low monovalent rejection, and documented cleanability with both alkaline and acid CIP. Operating pressure sits at 0.5–1.5 MPa (5–15 bar), with transmembrane pressure 0.4–1.2 MPa; flux at design conditions is 10–25 L/m²·h, and field experience shows flux drops 15–30% over 12–24 months before a CIP cycle restores performance (HydropureWater field data, 2026).
Recovery is set at 60–85% per stage, with two-stage arrays used to keep the concentrate at 15–40% of feed and to bring single-element recovery into the 30–50% band where calcium sulfate scaling risk is controlled. Feed limits are non-negotiable: SDI <5, turbidity <1 NTU, Fe <0.05 mg/L, free chlorine <0.1 mg/L. Chloramine residual from biological clarification must be quenched with activated carbon or sodium bisulfite dosing — exceeding 0.1 mg/L free chlorine permanently damages polyamide NF in days.
Cleaning frequency is typically 1–4 weeks depending on feed variability. Alkaline CIP at pH 11–12 removes organic fouling (size, dye oligomers, biomass carryover); acid CIP at pH 2–3 removes calcium carbonate and metal oxide scale. Antiscalant dosing on the NF feed is standard, and the engineering controls — antiscalant pump, CIP skid, pH and ORP probes — are integrated through an automatic chemical dosing for NF CIP and antiscalant package that ties to the NF PLC.
| Parameter | 2026 Design Value | Notes |
|---|---|---|
| Membrane type | Polyamide TFC, NF270-class | 400 ft² 8-inch elements standard |
| MWCO | 200–1,000 Da | Indigo/azo dyes 600–1,500 Da |
| Feed pressure | 0.5–1.5 MPa | TMP 0.4–1.2 MPa |
| Design flux | 10–25 L/m²·h | Drops 15–30% over 12–24 months |
| Recovery (per stage) | 60–85% | Multi-stage arrays preferred |
| Feed SDI | <5 | Verified at <3 for warranty |
| Free chlorine limit | <0.1 mg/L | Quench with SBS or activated carbon |
| Alkaline CIP | pH 11–12, 35 °C | Organic fouling |
| Acid CIP | pH 2–3, 35 °C | CaCO₃, metal oxide scale |
| CIP frequency | 1–4 weeks | Trend on normalized flux |
NF vs. UF, MBR, and RO for Denim Wash Water: Which Membrane Fits?
The membrane choice is a trade-off between color and salt rejection, pressure, and concentrate volume. Ultrafiltration removes colloids and suspended solids but lets dissolved dye and salt pass, so it functions as a clarifier upstream of NF, not as a stand-alone color/salt barrier. A submerged PVDF MBR, such as an integrated MBR system, replaces the secondary clarifier and produces near-reuse-quality effluent for suspended solids but does not reject dissolved color, sulfate, or hardness. Reverse osmosis, such as an RO polish downstream of NF for boiler-grade reuse, delivers 99%+ rejection of everything including monovalent salt, but at 1.0–2.0 MPa higher feed pressure than NF and a concentrate stream 30–50% larger because monovalent salt drives osmotic pressure.
Nanofiltration is the middle path. It removes color and divalent salts at 5–15 bar, lets monovalents pass so osmotic pressure stays manageable, and is the cost-effective choice when the goal is recycling permeate to the wash line, not producing boiler feed. The decision rule used in 2026 design reviews: if the KPI is freshwater reduction per kilogram of denim (target 30–50% reuse), NF typically wins on energy and concentrate volume; if the KPI is zero liquid discharge or boiler-feed reuse, NF + RO + evaporator is the correct train.
| Criterion | UF | MBR | NF | RO |
|---|---|---|---|---|
| Pore size / MWCO | 0.01–0.1 µm | 0.04 µm PVDF | 200–1,000 Da | <200 Da |
| Color rejection | <20% | 20–40% | 85–98% | >99% |
| Salt rejection | <5% | <5% | SO₄²⁻ 90–98%, Na⁺ 10–50% | 99%+ |
| Feed pressure | 0.1–0.3 MPa | 0.05–0.2 MPa | 0.5–1.5 MPa | 1.0–3.0 MPa |
| Energy (kWh/m³ permeate) | 0.1–0.3 | 0.3–0.6 | 0.5–1.2 | 1.5–3.0 |
| Best fit on denim duty | Pre-clarifier | Secondary clarifier | Color/salt split + wash-line reuse | Boiler-grade reuse, ZLD |
Water Reuse Mass Balance and Operating Cost Considerations
For a 50 m³/h denim wash house sending secondary effluent through a two-stage NF array at 70% recovery, the mass balance lands at approximately 35 m³/h of permeate and 15 m³/h of concentrate. At 6,000 operating hours per year, that is roughly 210,000 m³/year of freshwater displaced from municipal or borehole supply — a defensible number for a ZDHC water-stewardship submission.
Energy use is 0.5–1.2 kWh/m³ of NF permeate, which is 3–5× lower than RO at 1.5–3.0 kWh/m³ and the most defensible cost lever when presenting to a CFO (HydropureWater field data, 2026). The concentrate stream still requires handling: 15 m³/h can be sent to a downstream RO for further recovery, an evaporation pond (where climate and land permit), or a dedicated textile brine line. Antiscalant and CIP chemicals (NaOH, HCl, citric acid) typically run 0.05–0.15 kg per m³ of permeate and show up as a small but visible line item in the operating cost narrative; for context on the broader regulatory push driving this economics, see the 2026 water-reuse drivers and ZDHC pressure on textile mills.
For mills pushing toward irrigation reuse of the permeate, the 2026 reclaimed-water quality standards for alternative reuse set the target envelope, and the NF permeate comfortably meets the heavy-metal and pathogen criteria; only TDS and sodium may need a blend with freshwater before land application.
Frequently Asked Questions
When should a denim mill choose NF over RO for wash-water reuse?
Choose nanofiltration when the goal is to recycle permeate back into the wash line and the mill's KPI is freshwater reduction per kilogram of denim, typically targeting 30–50% reuse. NF runs at 0.5–1.5 MPa and rejects color plus divalent salts at 90–98% while letting monovalent salt pass, which keeps osmotic pressure and energy (0.5–1.2 kWh/m³) manageable. Choose reverse osmosis only when the mill needs boiler-grade reuse or zero liquid discharge, since RO requires 1.0–2.0 MPa higher feed pressure and a 30–50% larger concentrate stream.
What feed-water quality is required to protect NF membranes in denim duty?
Polyamide thin-film composite NF elements require SDI <5 (verified at <3 for warranty), turbidity <1 NTU, Fe <0.05 mg/L, and free chlorine <0.1 mg/L. The biological secondary effluent must pass through a multi-media filter to meet these limits, and any chloramine residual from biological clarification must be quenched with activated carbon or sodium bisulfite before the NF feed; exceeding 0.1 mg/L free chlorine permanently damages polyamide in days.
How much freshwater can a 50 m³/h denim wash house realistically save with NF?
A 50 m³/h wash house running DAF + biological + two-stage NF at 70% recovery produces approximately 35 m³/h of permeate and 15 m³/h of concentrate. At 6,000 operating hours per year, that displaces roughly 210,000 m³/year of municipal or borehole freshwater — a defensible number for ZDHC water-stewardship reporting and a clear freshwater reduction per kilogram of denim KPI.
What is the realistic recovery range for spiral-wound NF on denim effluent, and what limits it?
The realistic operating envelope is 60–85% recovery per stage, with multi-stage arrays used to keep single-element recovery at 30–50%. Recovery is limited by calcium sulfate scaling — on a feed with 1,500 mg/L SO₄²⁻ and 400 mg/L Ca²⁺ at 30 °C, saturation is reached at 70–78% recovery — and by osmotic pressure as dye and salt concentrate on the feed side. Antiscalant dosing and a 2–3 stage array are the standard 2026 controls.
