Closing the textile dyeing loop with forward osmosis via disperse dye-bath reconcentration and water recovery

Forward osmosis (FO) has shown high dye rejection in textile applications, but many studies still treat it as an end-of-pipe step and rely on synthetic feeds, leaving limited evidence on its practical integration into dyeing operations using real industrial baths. In this study, FO was evaluated as an in-process circular strategy to reconcentrate real industrial dyeing baths and enable reuse of both concentrated baths and recovered water. By combining factorial optimization, long-term reconcentration of three industrial dyeing baths, membrane-fingerprint analysis, and re-dyeing validation, this work advances beyond previous studies by assessing not only dye rejection or water recovery, but also the operational feasibility and reuse potential of FO-retained dye baths. A factorial design assessed the effects of draw-solution (DS) concentration (NaCl, 1–5 mol/L) and recirculation velocities (20–60 cm/s). Flux ranged from 4.15 to 5.58 L·m −2 ·h −1 , with a strong improvement when DS increased from 1 to 3 mol/L, while gains were limited at 5 mol/L due to polarization and reverse salt diffusion. Response-surface optimization targeting concentration factor (CF) selected 3 mol/L NaCl, FS velocity 20 cm/s, and DS velocity 40 cm/s for long-term tests with three real dyeing baths (black, navy, royal). Average fluxes were 7.09, 8.01, and 10.40 L·m −2 ·h −1 , with normalized flux declines of 6%, 6%, and 19% over 2880 min. Overall water recovery in the long-term reconcentration runs reached up to 88% and the performance losses were governed predominantly by osmotic effects (reverse NaCl diffusion and feed salinity build-up), rather than by severe fouling-driven resistance. UV–Vis monitoring showed bath-dependent reconcentration (CF = 2.13 black, 1.20 navy, 2.28 royal). Complementary analysis using EEM suggested stronger interaction/deposition for the black bath, whereas royal and navy membranes remained close to pristine fingerprints. Re -dyeing tests confirmed that FO retentates remained usable for dyeing, supporting their use as value streams through blending and recipe adjustment. Long-term tests with DS replenishment showed incomplete flux recovery due to salinity build-up, with Js/Jw of 6.32–12.58 g·L −1 , highlighting the need to balance reuse benefits and salt carryover in NaCl-driven FO.