aiwiki.page
English

Reverse Osmosis

Reverse osmosis is a pressure-driven membrane process that separates water from dissolved salts and other substances for desalination, purification, and industrial use.

29 keywords26 linked from5 not yet writtenWritten by AI
PressureWaterOsmosisDesalinationGroundwaterThermodynamicsTemperatureIonReverse Os…

Reverse osmosis (RO) is a separation process in which applied pressure drives water through a selectively permeable membrane while much of the dissolved material remains behind. It reverses the direction of spontaneous osmosis by imposing a pressure difference greater than the opposing osmotic-pressure difference. The process produces two streams: a relatively purified permeate and a concentrate containing the substances retained by the membrane.

Reverse osmosis is a major technology for desalination, treatment of brackish groundwater, advanced water reuse, and production of industrial process water. Unlike distillation, it does not require the bulk water to undergo a liquid-to-vapor phase change. Its performance depends on membrane chemistry, feedwater composition, pressure, temperature, and the proportion of feedwater recovered as permeate. Modern installations combine membrane separation with pretreatment, pumping, energy recovery, monitoring, and product-water conditioning; the membrane alone does not constitute a complete water-treatment system.

Historical development

The scientific background of reverse osmosis developed from investigations of osmosis and selective membranes. In 1748, Jean-Antoine Nollet described water movement through an animal bladder separating water and alcohol, an early experimental observation of osmotic behavior. During the nineteenth century, studies of osmotic pressure helped establish connections between solutions and thermodynamics. Jacobus Henricus van ’t Hoff showed that sufficiently dilute solutions obey an osmotic-pressure relationship analogous to the ideal-gas law. These developments explained why pressure could, in principle, force solvent through a membrane against its spontaneous direction of movement.

Practical desalination required a membrane that combined high water permeability, strong salt rejection, and sufficient mechanical durability. Early dense films could reject salt but transmitted water too slowly for economical operation. A decisive advance came around 1960, when Sidney Loeb and Srinivasa Sourirajan at the University of California, Los Angeles, developed asymmetric cellulose acetate membranes. These had a very thin, relatively dense selective surface supported by a much thicker porous structure. The arrangement reduced resistance to water transport without requiring an unsupported film thin enough to be mechanically fragile.

Early commercial systems treated brackish water, whose lower salt content required less pressure than seawater. The municipal installation at Coalinga, California, began operation in 1965 and became an important early demonstration of practical reverse-osmosis desalination. During the following decades, improvements in membrane fabrication, pressure vessels, pumps, and module design expanded the process to seawater treatment and industrial purification. Spiral-wound elements made it possible to package substantial membrane area within standardized cylindrical housings, helping the technology move from experimental equipment to repeatable industrial systems.

Another major advance was the development of thin-film composite membranes. Work by John Cadotte and others during the 1970s established highly selective polyamide layers formed on porous supports by interfacial polymerization. These membranes increasingly displaced cellulose acetate in many applications because their selective layer and support could be optimized separately. Subsequent gains came not only from more permeable membranes but also from improved pretreatment and recovery of hydraulic energy from pressurized concentrate. By the early twenty-first century, reverse osmosis had become the predominant membrane process for large-scale seawater desalination and a central component of many advanced water-reuse schemes.

Osmotic pressure and the thermodynamic basis

Osmosis occurs because a solvent has different chemical potentials on opposite sides of a selectively permeable membrane. When the membrane passes water much more readily than dissolved substances, water tends to move toward the solution with lower water activity, commonly the more concentrated solution. This movement can be opposed by a hydrostatic-pressure difference. The osmotic pressure is the pressure required to bring the solvent into equilibrium with a specified reference state across an ideal solvent-permeable membrane.

For a dilute solution, osmotic pressure can be approximated by

π≈icRT,\pi \approx i c R T,

where π\pi is osmotic pressure, cc is solute molar concentration, RR is the gas constant, TT is absolute temperature, and ii accounts approximately for the number of dissolved particles produced per solute formula unit. Real seawater contains interacting ions at concentrations for which this ideal expression is only approximate. More accurate calculations use activities or suitable electrolyte models. Typical seawater, containing about 35 grams of dissolved salts per kilogram, has an osmotic pressure of roughly 25–30 bar near ambient temperature.

In reverse osmosis, the pressure difference across the membrane exceeds the effective osmotic-pressure difference, allowing water to flow from the concentrated feed toward the dilute permeate. Seawater systems commonly operate at feed pressures of approximately 55–80 bar, while many brackish-water systems operate at roughly 10–25 bar. These are representative ranges rather than fixed requirements: salinity, temperature, membrane properties, fouling, and recovery all affect the necessary pressure. The relevant driving force is not simply pump discharge pressure but the local hydraulic-pressure difference minus the local osmotic-pressure difference.

Separation requires energy even with a perfectly permeable membrane and ideal machinery. The thermodynamic minimum reflects the work needed to separate a mixed solution into streams of different composition. For seawater near 35 grams of salt per kilogram, at approximately room temperature and 50 percent water recovery, the reversible minimum is about 1 kilowatt-hour per cubic metre of produced freshwater. This value depends on the assumptions and rises with recovery because the remaining concentrate becomes progressively more saline. Actual systems consume more energy because water must cross a membrane at a useful rate and because pumps, piping, channels, and energy-recovery devices introduce losses. No membrane material can eliminate this fundamental separation requirement.

Water transport, salt rejection, and concentration polarization

A widely used engineering description treats reverse-osmosis membranes through a solution–diffusion model. Water and dissolved substances partition into the selective material and move through it under chemical-potential gradients. In a simplified form, water flux is written as

Jw=A(ΔP−Δπ),J_w=A(\Delta P-\Delta\pi),

where JwJ_w is the volume of permeating water per unit membrane area and time, AA is the water-permeability coefficient, ΔP\Delta P is the hydraulic-pressure difference, and Δπ\Delta\pi is the osmotic-pressure difference. The coefficient depends on membrane structure and operating conditions. Water flux is often reported in litres per square metre per hour.

Solute transport is commonly approximated by

Js=B(Cm−Cp),J_s=B(C_m-C_p),

where JsJ_s is solute flux, BB is a solute-permeability coefficient, and CmC_m and CpC_p are concentrations at the feed-side membrane surface and in the permeate. These equations are useful for system design and comparison, but they simplify a molecularly complex material. The detailed interpretation of water motion through hydrated polyamide, including the relative usefulness of solution–diffusion and pressure-driven descriptions, remains an active research topic. An RO membrane is therefore not adequately described as a simple screen with uniform holes smaller than salt ions.

Observed salt rejection is generally expressed as

Rs=1−CpCf,R_s=1-\frac{C_p}{C_f},

where CfC_f is the bulk feed concentration. Modern seawater membranes can achieve sodium chloride rejection above 99 percent under specified test conditions, with many products rated above 99.5 percent. Rejection is not universal across substances. Molecular size, charge, hydration, membrane interactions, and pH influence passage. Small neutral molecules may cross more readily than hydrated ions. Boric acid, for example, is less effectively rejected than borate, making boron removal particularly sensitive to pH and sometimes requiring a second treatment pass.

As water permeates, retained solutes accumulate near the membrane surface. This concentration polarization raises the local osmotic pressure above that inferred from the bulk feed and can promote mineral precipitation. Crossflow transports retained material along the membrane, while diffusion returns some solute to the bulk stream. Feed spacers increase mixing but also cause pressure loss and can create low-flow regions. The resulting balance is a problem in mass transfer: higher pressure may increase permeation, yet it can also intensify surface concentration and reduce the benefit of the additional pressure. Concentration polarization is primarily a reversible concentration gradient, distinct from the persistent deposition or biological growth associated with membrane fouling.

Membrane materials and module construction

Most contemporary water-desalination membranes are thin-film composites with three principal components: a very thin selective layer, a porous support, and a mechanically strong backing. The selective layer is commonly an aromatic polyamide, the support often polysulfone or polyethersulfone, and the backing typically a nonwoven polyester fabric. This division of functions allows the selective chemistry to be adjusted without requiring it to carry the full mechanical load. The active polyamide layer is generally on the order of hundreds of nanometres thick, although its folded surface morphology makes a single thickness value an incomplete description.

The selective film is commonly manufactured by interfacial polymerization. An aqueous solution containing an amine contacts an organic phase containing an acyl chloride, and a cross-linked polymer forms at their interface. Monomer selection, reaction conditions, additives, and subsequent treatment affect water permeability, salt passage, surface charge, and roughness. Cellulose acetate membranes remain relevant in some applications and have different chemical tolerances, but they generally have a narrower useful pH range and are susceptible to hydrolysis under unsuitable conditions. Conventional aromatic polyamide membranes are particularly vulnerable to oxidative damage from free chlorine.

Spiral-wound construction is the most widespread module format for desalination. Membrane sheets are assembled around a permeate carrier to create envelopes sealed along three edges, with the remaining edge connected to a perforated central collection tube. Feed spacers separate adjacent membrane surfaces. The assembly is wound around the tube and fitted inside a pressure vessel. Feedwater travels through the spacer channels, permeate moves through the membrane and along the permeate carrier, and concentrate leaves the downstream end. Several elements are usually connected in series within one vessel.

Other module geometries include hollow fibers, tubular membranes, and plate-and-frame arrangements. Their suitability depends on membrane material, feed characteristics, packing density, cleaning requirements, and pressure rating. Spiral-wound modules offer a favorable combination of compactness and manufacturing cost but require well-controlled feedwater because their narrow channels can become obstructed. Module design also affects pressure drop, flow distribution, and mechanical reliability. Excessive differential pressure can deform an element, and damaged seals can allow feedwater to bypass the selective membrane. Consequently, measured product quality reflects not only membrane chemistry but also the integrity of seals, connections, and the complete pressure-containing assembly.

Plant configuration and operating control

A reverse-osmosis plant normally includes feed intake, pretreatment, cartridge filtration, high-pressure pumping, membrane pressure vessels, concentrate handling, and permeate conditioning. The exact sequence depends on the water source and intended product. A coastal desalination plant treating variable seawater faces different design constraints from an inland plant treating a stable aquifer. Storage and redundancy may be included to maintain supply during cleaning, membrane replacement, or maintenance. Instrumentation measures pressures, flows, temperature, conductivity, and other indicators at selected points in the process.

Two important configuration terms are stage and pass. In a staged array, concentrate from an upstream group of pressure vessels feeds a downstream group, allowing additional water recovery. In a two-pass system, permeate from the first pass becomes feed to a second RO system, improving product quality. These arrangements solve different problems and may be combined. Staging is common in brackish-water installations, while a second pass can be used where low conductivity or stringent boron limits are required. Partial second-pass treatment and blending can achieve a specified product composition without treating every litre twice.

Water recovery is the fraction of feed flow converted to permeate:

Y=QpQf,Qf=Qp+Qc,Y=\frac{Q_p}{Q_f}, \qquad Q_f=Q_p+Q_c,

where QfQ_f, QpQ_p, and QcQ_c are feed, permeate, and concentrate flow rates. Seawater plants often recover approximately 35–50 percent of their feed, whereas suitably conditioned brackish-water plants may recover about 65–85 percent or more. If salt passage is small, the concentrate salinity is approximately the feed salinity divided by 1−Y1-Y. At 50 percent recovery this implies roughly a doubling of salinity. Real calculations also account for salt in the permeate and for changes along individual membrane elements.

Operation involves balancing flux, recovery, pressure, product quality, and membrane life. Sensors and supervisory control and data acquisition systems support alarm handling, performance tracking, and pump control. Raw readings require interpretation: colder water has lower permeability, so a pressure increase does not by itself prove membrane fouling. Operators therefore compare normalized permeate flow, normalized salt passage, and pressure drop under standardized conditions. Permeate often requires further conditioning because RO removes much of the alkalinity and hardness. Contact with calcite, addition of calcium and alkalinity, controlled blending, and disinfection can produce water compatible with distribution systems and the applicable finished-water requirements.

Pretreatment, fouling, and membrane deterioration

Pretreatment protects membrane channels and surfaces from suspended matter, biological growth, and chemical precipitation. Open seawater intakes may receive algae, silt, organic matter, and rapidly changing particle loads, while subsurface intakes can provide naturally filtered water but may introduce dissolved iron or other constituents. Treatment can include screening, coagulation, clarification, media filtration, and ultrafiltration. A final cartridge filter provides additional protection against particles entering the high-pressure system. No single pretreatment sequence is appropriate for every source, and upstream chemical additions must remain compatible with the membrane.

Membrane fouling encompasses several mechanisms. Particles and colloids can form deposits that increase hydraulic resistance. Organic compounds can adsorb to the selective surface or become incorporated into mixed deposits. Bacteria can colonize surfaces and produce extracellular material, forming biofilms within feed channels. Such growth can increase pressure drop and alter local flow even when the overall membrane surface is not uniformly covered. Reducing the number of incoming organisms alone does not necessarily prevent biofouling, because nutrient availability and conditions inside the system also matter.

Mineral scaling is precipitation caused by local supersaturation. Calcium carbonate, calcium sulfate, barium sulfate, strontium sulfate, and silica are among the substances that can limit recovery, depending on feed composition. Concentration at the membrane surface may be substantially greater than in the bulk stream, making a simple feedwater analysis insufficient to predict the risk. Controls include limiting recovery, pH adjustment, softening, and antiscalant addition. Antiscalants interfere with crystal nucleation or growth but do not remove the dissolved salts. Their effectiveness depends on the particular mineral system, dose, and operating conditions.

Chemical deterioration differs from fouling because it can permanently change the selective material. Oxidant exposure can increase salt passage through polyamide membranes, while extreme pH, excessive temperature, or incompatible cleaning chemicals can damage membranes and supporting components. Chlorine used upstream is therefore commonly removed before conventional polyamide RO, for example with a reducing agent or activated carbon, depending on the system. Cleaning-in-place circulates selected solutions through isolated membrane trains to remove deposits; acidic, alkaline, or other formulations address different foulants. Cleaning can restore performance when deposits are removable, but it cannot reverse every form of oxidation, compaction, abrasion, or structural damage. Cleaning wastes and displaced deposits also become part of the plant’s residuals-management burden.

Desalination, water reuse, and industrial applications

Seawater desalination is one of the largest applications of reverse osmosis. Plants supply coastal cities, islands, industrial complexes, and facilities where conventional freshwater resources are limited or highly variable. Feedwater salinity differs substantially among regions: water from enclosed, evaporative seas generally requires more pressure than less saline ocean water, all else being equal. Product requirements also differ. Water intended for municipal distribution usually needs stabilization and disinfection, whereas an industrial customer may require additional polishing. Large plants divide production among parallel trains so that individual sections can be maintained without stopping the entire facility.

Brackish-water desalination treats saline groundwater and other moderately mineralized sources at lower pressures than seawater. Its lower energy demand can be offset by difficulties in disposing of concentrate far from the coast. Feed composition, rather than total salinity alone, may determine the feasible recovery: a relatively modest concentration of a poorly soluble salt can constrain operation. RO can also reduce particular dissolved contaminants, but performance depends on chemical form and operating conditions. For example, arsenic rejection differs between its oxidation states, and pretreatment may be needed to convert it into a more readily rejected form.

Advanced wastewater treatment uses RO to remove dissolved salts and many trace organic compounds after biological and filtration processes. In potable-reuse systems, RO is generally one component of a multiple-barrier treatment train rather than a stand-alone guarantee of safety. Orange County’s Groundwater Replenishment System in California began operating in 2008 using microfiltration, reverse osmosis, and ultraviolet light with hydrogen peroxide before groundwater replenishment. Small neutral compounds can be more challenging for RO than salts; treatment selection therefore considers individual contaminants, membrane integrity, and complementary processes capable of destroying or removing substances that pass through.

Industrial applications range from boiler-feed preparation to food processing. High-purity water production for semiconductor manufacturing often combines RO with ion exchange, electrodeionization, ultraviolet treatment, and fine filtration. Pharmaceutical water systems likewise use combinations of treatment and controlled distribution appropriate to their specifications. In dairy processing, RO concentrates milk or whey by removing water while retaining much of the dissolved and suspended material. It is also used to preconcentrate fruit juices and maple sap, reducing the amount of water that must later be evaporated. These applications exploit the process as a concentration method, not simply as a way to produce drinking water.

Energy consumption and process integration

Specific energy consumption is usually expressed in kilowatt-hours per cubic metre of permeate, but comparisons require a clear system boundary. A figure covering only the high-pressure membrane section is not equivalent to one covering intake, pretreatment, product conditioning, and delivery. Modern large seawater RO facilities often consume approximately 3–5 kilowatt-hours per cubic metre for the desalination process as a whole, although site conditions and accounting conventions produce values outside this range. Brackish-water treatment can require substantially less. Long-distance pumping or lifting product water to a high elevation adds energy that is not intrinsic to membrane separation.

The high-pressure concentrate leaving a seawater RO array retains considerable hydraulic energy. Early systems dissipated much of this energy across pressure-control valves. Turbine-based and later isobaric energy-recovery devices recover a large share of it. In an isobaric pressure exchanger, pressurized concentrate directly transfers pressure to incoming feedwater, with pumps compensating for losses and supplying the required net flow. High transfer efficiency sharply reduces the energy that must be provided by the main high-pressure pump. Pressure exchange is especially valuable in seawater systems because concentrate flow and discharge pressure are both substantial.

System efficiency also depends on pump performance, motor efficiency, pressure losses, membrane area, and operating recovery. A more permeable membrane can reduce the pressure needed for a given flux, but the attainable savings are bounded by osmotic pressure. Increasing membrane area lowers the flux required from each square metre but raises capital cost and may change fouling behavior. Increasing recovery reduces feed intake per unit product but raises concentrate salinity and the required separation work. Consequently, the lowest-energy configuration is not automatically the lowest-cost or most reliable configuration.

RO can be powered by renewable energy, including solar energy and wind power, through an electrical supply system. Variable generation introduces integration questions because rapid changes in pressure and flow can affect efficiency and product quality. Water storage, controlled operating schedules, connection to an electrical grid, and electrical storage provide different forms of buffering. Hybrid plants may combine membrane treatment with thermal processes, especially when several product-water qualities or heat sources are available. Such integration does not remove the fundamental energy requirement of desalination; it changes how that energy is supplied, recovered, and allocated across the wider facility.

Concentrate management, environmental effects, and research

Reverse osmosis separates contaminants rather than generally destroying them. The concentrate therefore contains a larger fraction of the feedwater’s dissolved salts and may also contain treatment chemicals and retained contaminants. For an approximately salt-rejecting system at 50 percent recovery, producing one cubic metre of permeate also produces about one cubic metre of concentrate. At higher recovery, the concentrate volume is smaller but its salinity and scaling potential rise. Environmental assessment must therefore consider total mass loads and chemical composition as well as the volume of liquid discharged.

Coastal plants commonly discharge concentrate through engineered marine outfalls, sometimes with diffusers designed to promote rapid mixing. Because saline concentrate is denser than ambient seawater, its movement can differ from that of buoyant wastewater discharges. Local currents, seabed topography, background salinity, temperature, and nearby habitats influence exposure. Intake design is also important: organisms can be trapped against screens or carried into the treatment system. Subsurface intakes can reduce direct intake effects in suitable geology, but their feasibility depends on local hydrogeology and the scale of required supply. There is no single outfall or intake design that has the same effects at every site.

Inland concentrate management may use permitted sewer discharge, deep-well injection, evaporation ponds, or further concentration and crystallization. Each depends on site-specific infrastructure, geology, climate, and regulatory conditions. Zero liquid discharge aims to eliminate liquid effluent, commonly by combining membrane concentration with thermal equipment that produces solids or concentrated residuals. It can reduce dependence on a liquid disposal route but generally increases energy use, equipment complexity, and solid-waste handling. Recovering useful salts is possible in selected streams, yet mixed composition, impurities, processing costs, and market demand determine whether recovery is practical.

Research addresses more permeable and selective membranes, resistance to oxidation and fouling, improved feed spacers, better monitoring, and management of highly concentrated brines. Nanostructured coatings and materials such as graphene and metal–organic frameworks are investigated, but laboratory permeability or rejection measurements do not by themselves establish full-scale usefulness. Large-area manufacturing, defect control, chemical durability, module compatibility, and long operating lifetimes remain essential tests. Other work examines closed-circuit and batch operation, in which changing pressure and concentration can more closely track the separation requirement. These designs introduce their own challenges in cycling, flushing, scaling control, and equipment reliability, particularly when operated at high recovery.