N,N-Dimethylformamide (DMF, CAS 68-12-2) functions in polyurethane synthetic leather manufacture as a dipolar aprotic solvent with a normal boiling point of 153 °C, a vapour pressure of 0.37 kPa at 20 °C, a closed-cup flash point of 58 °C, and flammability limits of 2.2 % v/v to 15.2 % v/v in air. DMF is fully miscible with water, and this property is exploited in wet coagulation: a cast polyurethane film is immersed in a water/DMF bath, the solvent exchanges with water, and phase inversion generates a microporous matrix. Under the CLP Regulation, DMF carries the hazard statement H360D, which is the toxicological basis for the article restriction that shapes production practice. The European Union restriction under Commission Regulation (EU) 2021/2030, Annex XVII entry 76, applies a finished-article limit of 0.3 % w/w from 12 December 2023. DMF is not removed efficiently by drying because its boiling point is far above typical coated-fabric dryer temperatures of 120–140 °C; therefore residual DMF must be managed before drying.

Industrial lines recover DMF from three main sources: coagulation bath overflow, wash-water spent streams, and dryer exhaust. The first two are liquid streams and are treated by distillation or evaporative preconcentration; the third is an air stream that is captured by carbon adsorption or thermal oxidation. The liquid recovery train is normally coupled to the washing line because the residual DMF in the finished article is set by the final wash bath purity and the washing intensity.

What Limits Residual DMF During Wet Coagulation and Washing?

The coagulation bath is the first residual-control point. The bath is a DMF–water mixture, and its DMF mass fraction governs precipitation rate. At bath DMF fractions above 30 % w/w, the solvent gradient across the cast film collapses; a dense skin forms, water ingress slows, and the inner layer retains high residual DMF. At bath DMF fractions below 15 % w/w, precipitation is rapid enough to generate macrovoids, reducing tear resistance measured by ISO 4674-1 and flex resistance measured by ISO 32100:2018. For fast-coagulating soft-segment polyurethane grades, the process window around 22 % w/w can be as narrow as ±2 % w/w; outside this window, split-tear values decline or surface pinholes appear. Bath temperature is held between 25 °C and 40 °C. Higher bath temperature shortens phase-inversion time but increases DMF vapour emission; lower temperature slows solvent exchange and increases downstream washing load. On production-scale lines, plate heat exchangers with ±2 °C control are used because bath temperature and DMF fraction act as coupled variables.

Wash water temperature and residence time then control desorption. The post-coagulation polyurethane gel retains DMF in the polymer-rich phase; because DMF is fully miscible with water and strongly solvates the PU hard segments, diffusion is grade-specific. Published diffusion data for DMF in wet PU gel vary with hard-segment content and are not transferable across resin suppliers. Wash intensity is therefore established by factorial pilot trials at 40 °C, 50 °C, and 60 °C wash water, with final DMF measured on the dried coated fabric.

The following table lists representative DMF mass fractions across line streams. The values are operating envelopes reported in wet-process PU leather plants; local set points must be validated for the specific polyurethane grade, fabric weight, and line speed.

From the coagulation bath and first wash tanks, high-strength DMF liquor at 20–30 % w/w is segregated from low-strength wash water. The high-strength stream is filtered through a 10–25 µm cartridge filter to remove fibre debris before entering a vacuum rectification column. Low-strength wash water at 0.5–3 % w/w is preconcentrated by multi-effect evaporation or vacuum stripping; the preconcentrated stream at 8–12 % w/w is then blended into the column feed. DMF and water do not form an azeotrope, but DMF hydrolyses under acid or base catalysis at elevated reboiler temperatures, producing dimethylamine and formic acid. Recovery columns therefore operate under vacuum with base temperatures below 130 °C to limit hydrolysis. Overhead water is recycled to the washing train; recovered DMF is drawn as a side stream or bottom product. Free acidity as formic acid is controlled below 0.005 % w/w, because acidic recovered DMF accelerates hydrolytic degradation and yellowing of PU resin in the next coating batch.

Fibre fragments that bypass filtration carbonise on reboiler tubes and lower the heat-transfer coefficient. Production lines report quarterly cleaning of the first-effect reboiler when cartridge filters are improperly seated. The same lines monitor overhead DMF concentration in the water distillate; keeping the overhead DMF content below 0.1 % w/w allows the water to return to low-strength washing without overwhelming biological treatment. DMF is biodegradable, but high DMF loads increase chemical oxygen demand and total nitrogen in wastewater.

When Recovered DMF Re-enters the Coating Batch

Recovered DMF quality is not solely a distillation issue; residual water, dimethylamine, and formic acid alter coating solution rheology and colour. A recovered DMF specification of water ≤0.05 % w/w and free acidity ≤0.005 % w/w as formic acid is normally required for reuse in one-component polyurethane systems. Water is measured by Karl Fischer titration per ASTM E203-16. If recovered DMF contains dimethylamine, the solution pH shifts and ester-based PU soft segments can undergo hydrolytic chain scission during storage, causing viscosity drop at 25 °C and pinhole defects after coagulation. Batch-to-batch variation is monitored by refractive index and Karl Fischer titration; Fourier transform infrared spectroscopy is used when dimethylamine contamination is suspected. Production lines that blend virgin and recovered DMF report stable coating solution viscosity when the recovered solvent meets the water and acidity limits; published data for specific resin grades remains limited.

Recovered solvent is added to the mixing vessel after cooling below 40 °C. If recovered DMF is added hot, local gel formation can occur at the vessel wall. The coating solution temperature is maintained at 25 °C during viscosity measurement with a rotational viscometer, and the viscosity acceptance band is set by the resin supplier.

Countercurrent washing trains are configured as three-to-five stages with pneumatic squeeze rollers between baths. Wash water flows opposite to fabric direction; the final stage receives fresh water at 40–60 °C, while overflow cascades to earlier stages. Squeeze pressure at each nip removes interstitial DMF-laden water before the next dilution stage. The final bath DMF concentration must remain below 0.2 % w/w to achieve finished-article DMF below 0.3 % w/w after drying at 120–140 °C. Drying exhaust is not a minor stream: forced-convection dryers remove water and residual DMF, and the exhaust air is routed to carbon adsorption or thermal oxidation. Production lines exhibit a transient residual DMF spike after grade changes; countercurrent wash baths retain DMF from the previous resin system, and the first 50–100 linear metres after a changeover can exceed the target. Operators divert this transition length until bath conductivity returns to baseline.

Thick compact-grain articles require an additional steam-assisted desorption step after washing. Saturated steam at 100–105 °C raises the temperature of the polymer matrix and increases the diffusion coefficient of DMF; plant trials show that steam treatment removes residual DMF that ordinary water washing leaves trapped in the dense skin. The steam chamber is followed by a dry-air purge before drying to prevent water spotting.

Residual DMF Analysis in Finished Synthetic Leather

For compliance testing, representative samples are cut from three positions across the web width and extracted with methanol or acetonitrile. Gas chromatography with flame ionisation detection or mass spectrometry is used; the method limit of quantification should be ≤10 mg/kg to support decisions against the 0.3 % w/w limit, which corresponds to 3,000 mg/kg. Calibration standards are prepared from DMF of purity ≥99.9 % in the same extraction solvent. Because DMF is hygroscopic, samples are sealed immediately after cutting and extracted within 24 h. The analytical result is not corrected for moisture regain. Non-compliance is declared when the measured value exceeds 3,000 mg/kg after taking the expanded measurement uncertainty into account. A finished article at 2,900 mg/kg is borderline: the difference from the limit is smaller than the typical expanded uncertainty of ±10 %, so replicate analysis or a confirmatory GC-MS method is required before release.

Control laboratories maintain a retained sample for 30 days after release. Residual DMF can migrate slowly within the polyurethane matrix during storage, so retained samples support dispute resolution when a released batch is re-tested after warehousing.