Water Water Everywhere Not A Drop To Drink

Getting Fresh Water From Saltwater: The Actual Process

Water Water Everywhere Not A Drop To Drink

The phrase comes from Coleridge, but the problem it describes is purely engineering now. Coastal regions with access to endless ocean water but zero freshwater sources face this daily. I spent three years on a desalination retrofit for a mid-sized treatment facility in the Gulf and learned more about what actually goes wrong than what the brochures say. Reverse osmosis is the dominant method. Seawater gets pushed through semi-permeable membranes at roughly 600 to 800 psi. Salt ions and contaminants can't pass through. Fresh water comes out the other side. The rejection rate on modern membranes sits around 99.7% for dissolved salts. It sounds efficient until you look at the energy cost. That pressure requires high-energy pumps. A typical seawater RO plant runs between 3 to 4 kilowatt-hours per cubic meter of product water. For a community of 100,000 people consuming about 300 liters per person per day, you are moving roughly 30,000 cubic meters daily. That is 90,000 to 120,000 kilowatt-hours just for the membrane stage before you factor in pretreatment and post-treatment.

What Happens Before the Membrane

This is where most plant failures start. Seawater pulled directly from the coast contains suspended solids, algae, bacteria, and organic matter. If any of that reaches the membrane surface, you get fouling. Biofouling is the worst kind because microorganisms form a slime layer that degrades flux rates within days rather than months. The standard pretreatment train looks like this: screening to remove debris, coagulation and flocculation with chemicals like ferric chloride, sand filtration, and then cartridge filtration down to 5 microns. Some plants add activated carbon to strip organic compounds that membranes can't reject. In my experience, the cartridge filtration stage is the most overlooked. People size the big equipment correctly and then cheap out on the 5-micron blocks. Those blocks clog fast in bloom seasons and the pressure differential spikes without anyone noticing until the RO skid alarms go off.

Membrane Maintenance and Replacement

Membranes don't last forever. A standard element in a seawater RO system has a design life of about 5 to 7 years with proper maintenance. The real killer is chlorine exposure. Polyamide membranes, which dominate the market, are extremely sensitive to oxidants. Even brief exposure to chlorinated water during a pretreatment upset can degrade the selective layer permanently. I once tracked down a gradual salt passage increase over six months across an entire membrane array. The root cause was a broken dosing pump that let raw chlorinated water bypass the dechlorination stage for about 40 minutes during a routine shutdown. The replacement cost for those elements was around $47,000. Chemical cleaning is routine. You cycle sodium metabisulfite for oxidative cleaning and citric acid or specialized detergents for organic and biofouling. The interval depends on feed water quality. Good pretreatment pushes cleaning intervals to every 3 to 6 months. Poor pretreatment means you are taking the skid offline weekly. There is a rule of thumb among operators: if you are cleaning more often than monthly, your pretreatment is wrong, not your membranes.

What Comes After the Membrane

The permeate from reverse osmosis is essentially pure water with one exception. It is aggressive. It has no minerals and a low pH, typically around 5.5 to 6.5. If you send that straight into a distribution pipe, it will leach metals from the plumbing. Copper, lead, zinc. You have to remineralize it. Most plants inject CO2 to adjust pH or blend with a mineral slurry containing calcium carbonate or magnesium oxide. The target is a stable pH between 7 and 8.2 with enough alkalinity to prevent corrosion in the downstream infrastructure. Skipping or shortening this step is a common mistake when plants are trying to reduce OPEX. I have seen it in two separate facilities and both ended up with customer complaints about metallic tasting water and corrosion damage to building plumbing within the first year of distribution.

Brine Disposal

The concentrate stream is where the environmental headaches live. For every liter of fresh water you produce, you are generating roughly 0.5 to 0.75 liters of brine that is twice as salty as seawater plus residual pretreatment chemicals. Dumping it back into the ocean requires careful diffuser design to ensure rapid mixing. A poorly designed outfall creates a localized hypersaline zone that kills benthic organisms within a few hundred meters of the discharge point. Solar evaporation ponds work in arid coastal areas but consume large land areas and still leave concentrated residue. Zero liquid discharge systems exist but run 40 to 60% more expensive than conventional brine management. There is no cheap solution here and nobody talks about it enough in project feasibility studies.

When Reverse Osmosis Is the Wrong Call

Not every salty water problem needs RO. Small-scale applications under 50 cubic meters per day sometimes make more sense with thermal distillation, especially where waste heat is available from industrial processes. Multi-effect distillation or mechanical vapor compression can be more energy-efficient at small scale because they recover latent heat across multiple stages. The capital cost is higher but the operating cost flips when you have free or cheap thermal energy sitting around. Nanofiltration is another option for brackish water where the salt content is lower than seawater. The pressure requirement drops to 150 to 300 psi and the energy consumption falls accordingly. If your source water is groundwater with moderate TDS rather than ocean water, you are almost certainly over-engineering the problem by going straight to full RO. The energy equation is improving slowly. Regenerative pressure exchangers like those from Energy Recovery Inc. have brought the specific energy consumption of large plants down significantly by recovering pressure from the brine stream. Newer projects in the Middle East are achieving 2.8 kWh/m3 at the membrane stage. That is better but still requires a serious power source. If your coastal community has access to cheap solar or wind, pairing desalination with renewable generation is the only way the economics stay sane without subsidies.

I keep coming back to pretreatment. It is the unglamorous part of the process that determines whether your plant runs for a decade or fails in eighteen months. Size it properly, monitor the differential pressures, and change those cartridge filters on schedule. Everything else depends on it.