Why Forward Osmosis for Brine Concentration Is Gaining Ground in 2026
Forward osmosis for brine concentration uses an osmotic pressure gradient — not hydraulic pressure — to pull water out of saline or hypersaline brine into a concentrated draw solution, reducing brine volume for downstream ZLD or salt recovery. Industrial reject brine typically runs 1.6–2× seawater salinity, or roughly 56–70 g/L TDS (per S4, Nature 2025), and at that range high-recovery RO and mechanical vapor recompression concentrators consume disproportionate thermal or electrical energy. Forward osmosis, as an osmotically driven membrane process, sidesteps hydraulic pressure: water diffuses from feed to draw because the draw's osmotic pressure exceeds the feed's, so membrane fouling is materially lighter than in RO (S2, Membranes 2026). This is the single most cited reason operators are revisiting FO after a decade of pilot-stage work. A 2026 Water Resources and Industry modeling study (S5) reinforces the case quantitatively: with an aquaporin hollow-fiber FO membrane and 50% isopropanol draw, the system reached 60.7% water recovery on highly saline feed (<35 g/L TDS) and 52.4% recovery on hypersaline feed (>70 g/L TDS) in a single pass. For a plant engineer weighing where FO slots in a brine management train, those recovery numbers are the first data point to compare against an existing RO recovery curve.
How Forward Osmosis Concentrates Brine: Mechanism and Driving Force
An FO cell has two compartments separated by a semipermeable membrane: the brine feed on one side and a high-osmotic-pressure draw solution on the other. Water moves from feed to draw because the chemical potential of water is higher in the feed; no hydraulic pressure is applied across the membrane. The driving force is the net osmotic pressure differential, and the working equation is:
Jw = A (πdraw − πfeed − πCP)
where A is the membrane's pure-water permeability coefficient, πdraw and πfeed are the bulk osmotic pressures of draw and feed, and πCP is the concentration-polarization loss that reduces the effective driving force. That πCP term is the single biggest reason FO flux is lower than the membrane's A-value would predict, and it has two forms. In AL-FS orientation (active layer facing feed) the dominant loss is dilutive internal concentration polarization inside the porous support, which becomes severe once feed TDS climbs past 35 g/L. In AL-DS orientation (active layer facing draw) the loss is concentrative external concentration polarization on the feed side — easier to model and easier to manage, but it cuts draw solute back-diffusion. For hypersaline brine volume reduction, most lab work has shifted to AL-DS for that reason. The other selectivity metric an engineer must monitor is reverse solute flux (RSF), measured as specific reverse solute flux (SRSF, g/L) — draw solute leaking back into the feed. The 2026 IPA/isopropanol study (S5) reported SRSF of 4.1 g/L on highly saline feed and 3.8 g/L on hypersaline feed, which is competitive with thermolytic draw systems when normalized against the osmotic pressure delivered.
FO Membrane Types Used for Brine Concentration
Three membrane families dominate the FO brine literature, trading off differently on flux, chemistry, and cost. Cellulose acetate (CA) membranes are the lowest-cost option, tolerate free chlorine up to ~1 ppm, but are limited to roughly pH 3–8 and temperatures below 35–40 °C — a narrow envelope that rules them out for hot mining or chemical brines. Thin-film composite (TFC) polyamide membranes, made by interfacial polymerization, are the dominant commercial choice because they deliver higher water permeability and salt rejection, but they are sensitive to chlorine (typically <0.1 ppm continuous exposure) and to surfactants. Aquaporin-based biomimetic hollow-fiber FO (HFFO) membranes embed aquaporin water channels in a TFC skin, and the 2026 modeling study (S5) reported 16.24 LMH on highly saline feed and 12.41 LMH on hypersaline feed with this membrane class — a clear step above first-generation CTA fibers. A fourth data point: a modified cellulosic FO membrane with a 1:1 NH4OH/NH4HCO3 draw reached 113 LMH in the Nature 2025 study (S4), but that draw chemistry must be compatible with the downstream regeneration step and with ammonia discharge limits, so the headline flux is not directly translatable to every plant.
Draw Solution Selection: Matching Chemistry to Brine Application
Draw solution regeneration is the variable that decides whether FO is economically viable for a given brine, not the raw flux number. The 2026 review (S2) classifies draw solutes into five families: gaseous (SO2, NH3/CO2), organic (isopropanol, glycerol, glucose), inorganic (NaCl, MgCl2, CaCl2), magnetic nanoparticle, and polymer gel. For industrial brine concentration, only two systems have credible regeneration economics today: thermolytic NH3/CO2 and alcohol-based isopropanol (IPA). The table below summarizes the operating trade-off.
| Parameter | Thermolytic NH3/CO2 | 50% Isopropanol (IPA) |
|---|---|---|
| Osmotic pressure (typical draw conc.) | ~25–30 MPa at 60 °C saturated | ~14–17 MPa at 50 wt% |
| Regeneration method | Heating to 60–80 °C; NH3 and CO2 strip overhead | Fractional distillation; permeate to 0.2 ppm IPA |
| Reverse solute flux (SRSF) | 0.5–2 g/L (system-dependent) | 3.8–4.1 g/L (S5, 2026) |
| Thermal energy for regeneration | Lower per m³ water; ~6–10 kWh/m³ thermal | Higher per m³ water; distillation column |
| Residual concern in permeate | Ammonia slip; requires polishing | Volatile organic; 0.2 ppm in distilled permeate (S5) |
| Best-fit brine | High-flux needs, ammonia-tolerant downstream | Hypersaline brine where SRSF is acceptable (S5) |
The IPA system (S5) is attractive for hypersaline brine because the draw reaches osmotic pressures high enough to drive 12.41 LMH even against a 70 g/L feed, and fractional distillation regenerates it cleanly. NH4HCO3 and 1:1 NH4OH/NH4HCO3 systems (S4) dominate high-flux lab demonstrations but require thermal decomposition above ~60 °C and can carry ammonia residuals into the product stream. The selection rule of thumb is simple: pick the draw with the lowest $/m³ of regenerated permeate, not the highest raw flux.
Real Performance Data: 2026 Flux and Recovery for Saline and Hypersaline Brine
The table below consolidates 2026 quantitative performance so an engineer can benchmark a planned FO step against the published envelope. The headline numbers come from the Water Resources and Industry 2026 modeling study (S5), which used an aquaporin hollow-fiber FO membrane and a 50% isopropanol draw modeled in MATLAB/Aspen Plus.
| Feed class | Feed TDS (g/L) | Water flux (LMH) | Water recovery (%) | SRSF (g/L) | Draw solution | Source |
|---|---|---|---|---|---|---|
| Highly saline | < 35 | 16.24 | 60.7 | 4.1 | 50% IPA | S5 (2026) |
| Hypersaline | > 70 | 12.41 | 52.4 | 3.8 | 50% IPA | S5 (2026) |
| Brackish (intermediate) | 5–15 | 20–30 (typical pilots) | — | — | NaCl or thermolytic | Industry pilots, 2024–2025 |
| Modified cellulosic, NH4 draw | ~35 (saline) | 113 | — | — | 1:1 NH4OH/NH4HCO3 | S4 (Nature 2025) |
The brackish intermediate range is where published FO pilots consistently run 20–30 LMH — the upper end of what the technology delivers today, but it is also the range where RO is already efficient. The 113 LMH result in S4 illustrates membrane capability but, again, draw regeneration and ammonia handling dominate the actual project economics.
Fouling, Pretreatment, and Operating Limits You Must Plan For
FO is not a "drop-in" membrane step. The 2026 Membranes review (S2) classifies FO fouling into four categories — inorganic scaling, organic, colloidal, and biofouling — and each maps to a different brine source. Mining reverse-osmosis reject typically drives calcium sulfate and barium sulfate scaling near the saturation limit; food and dairy brine brings heavy organic loading; landfill leachate concentrate carries both colloidal silica and active biology. The pretreatment that mitigates all four is consistent: a multi-media filter ahead of the multi-media filter ahead of the UF stage, followed by a UF pretreatment skid for FO membrane protection at ≤0.03 µm nominal pore size to drive SDI15 below 3 before the FO membrane. Operating limits to plan for: feed temperature typically 25–40 °C (TFC polyamide loses rejection above 45 °C); pH 2–11 for TFC; draw solution must remain thermally stable across the regeneration cycle; and recovery must be capped before the feed crosses the scaling onset for its dominant scalant. A practical sludge dewatering reference for the FO brine pretreatment line helps when sizing the downstream solids handling if lime softening is added ahead of UF.
Integrating FO into a Brine Management Train: FO + RO + MD / Crystallizer
FO is rarely a terminal step, as its function is to lift feed concentration into the operating window of a downstream membrane or thermal process. Three realistic trains are deployed today:
| Train | Configuration | Typical recovery | Best-fit application |
|---|---|---|---|
| 1. Water reuse | UF → FO → high-recovery RO for the post-FO polishing step | 85–95% water reuse | Industrial reuse where permeate water quality is the priority |
| 2. Near-ZLD | UF → FO → membrane distillation (MD) | >95% liquid recovery | High-salinity brine where RO pressure penalty is too high |
| 3. Salt recovery | UF → FO → crystallizer | ~99% (with crystal purge) | Lithium, sodium, or potassium recovery; ZLD engineering blueprint for high-salinity industrial brine |
The draw-solution regeneration loop is co-sized with the FO skid and is the largest auxiliary consumer of thermal or electrical energy in any of these trains. For the water-reuse case, sizing the post-FO RO is a different exercise from greenfield RO design because the FO step has already reduced the feed volume; the RO sizing guide for the post-FO polishing step walks through the corrected flux and recovery assumptions. For long-term membrane replacement budgeting, plan on stocking replacement RO and UF membrane elements for the FO brine train on the same service interval.
Frequently Asked Questions
What flux and recovery
Frequently Asked Questions
What is forward osmosis for brine concentration and how does it differ from reverse osmosis?
Forward osmosis (FO) is a membrane separation process that utilizes the natural osmotic pressure gradient between a feed solution and a highly concentrated draw solution to drive water permeation across a semi-permeable membrane. Unlike reverse osmosis (RO), which relies on high-pressure mechanical pumps to overcome osmotic pressure, FO operates at low or near-zero hydraulic pressure.
Because FO does not require the extreme hydraulic pressures (often exceeding 80–100 bar) needed to process high-salinity brines, it is significantly less prone to membrane compaction and fouling. While RO is limited by the osmotic pressure of the feed, FO can process hypersaline feeds by utilizing a draw solution with an even higher osmotic pressure, allowing for greater concentration factors.
Which draw solution is best for concentrating hypersaline industrial brine?
The selection of a draw solution depends on the intended end-use and the required regeneration method, but thermally responsive solutes and high-solubility salts are current industry standards. For systems requiring easy regeneration, thermolytic salts like ammonium bicarbonate or proprietary switchable polarity solvents are preferred as they can be separated from water via low-grade waste heat (below 70°C).
For applications where the draw solution can be integrated into a larger process, concentrated brine streams—such as magnesium chloride (MgCl2) or calcium chloride (CaCl2)—are highly effective due to their high solubility and high osmotic potential. These solutions can achieve osmotic pressures exceeding 200 bar, facilitating the concentration of industrial brines that would otherwise reach the thermodynamic limit of traditional RO systems.
What water flux and recovery can a forward osmosis system realistically achieve on brine above 70 g/L TDS?
For feeds exceeding 70 g/L TDS, FO systems typically maintain a water flux ranging from 2 to 8 LMH (liters per square meter per hour), depending on the membrane type and the specific osmotic pressure gradient maintained. While these flux rates are lower than those seen in brackish water RO, they remain stable due to the low-pressure operating environment.
System recovery rates in FO for hypersaline brines are highly variable but can realistically reach 50% to 80% volume reduction in a single stage. When integrated into a multi-stage configuration, FO can push feed concentrations toward saturation points, significantly reducing the volume of brine sent to downstream thermal evaporators or crystallizers.
What pretreatment does an FO membrane need before handling industrial brine?
Pretreatment for FO is critical to mitigate the risk of membrane scaling and irreversible fouling, particularly when dealing with industrial streams rich in silica, calcium, and magnesium. Standard pretreatment typically includes multi-media filtration (MMF) or ultrafiltration (UF) to remove suspended solids and colloids that exceed 0.1 micrometers.
In addition to physical filtration, chemical conditioning is often necessary to prevent scaling. This includes pH adjustment to maintain solubility limits and the addition of antiscalants designed for high-salinity applications. If the brine contains high organic loads, advanced oxidation processes (AOP) or activated carbon adsorption may be required to protect the membrane's active layer from chemical degradation.
How does forward osmosis fit into a zero liquid discharge (ZLD) brine treatment train?
In a Zero Liquid Discharge (ZLD) configuration, forward osmosis acts as a high-efficiency volume reduction step positioned between initial RO stages and final thermal treatment. By concentrating the brine stream prior to entering a mechanical vapor recompression (MVR) evaporator or a crystallizer, FO significantly reduces the thermal energy demand of the entire plant.
By removing a large portion of the water content at low energy cost, FO allows the downstream thermal equipment to be downsized. This reduces both the capital expenditure for massive evaporators and the operational expenditure associated with high-energy thermal processes, effectively bridging the gap between standard RO desalination and final salt crystallization.