Refinery Sour Water: Breaking the Emulsion Before the Stripper Tower

Why sour water resists conventional separation

Sour water is the stream that comes off coking fractionators, catalytic cracking units and hydrocracking trains in a refinery. It is not simply oily: it carries high concentrations of hydrogen sulphide, ammoniacal nitrogen and volatile phenols, and the intense agitation plus high temperature and pressure of the production units turn it into a stable emulsion, either water-in-oil or oil-in-water. Oil droplets typically sit below 10 μm, a size band where conventional gravity separation gives up entirely.

Three operational headaches follow. Demulsification is difficult, so oil-water separation efficiency drops and the effluent carries excess oil into the downstream stripper tower, where it drives up scaling and corrosion. Compliance pressure is constant, because standards such as the Emission Standard of Pollutants for the Petroleum Refining Industry (GB 31570-2015) keep tightening the limits on oil content and suspended solids. And recovery is poor: incomplete separation means less waste oil is salvaged, which wastes a resource and inflates sludge disposal cost.

What the mainstream options actually deliver

Most refineries pre-treat sour water with one of three approaches, and each has a clear ceiling.

• Conventional gravity settling and oil separators rely on the density difference between oil and water. They remove floating oil, but emulsified oil removal sits below 30%, running cost is low, and adaptability is very poor — the process only works on lightly emulsified water. For a highly emulsified stream, effluent oil content misses the standard by a wide margin.
• Conventional dissolved-air flotation (DAF) attaches microbubbles to droplets and floats them out. Emulsified oil removal reaches 60% to 70%, but holding effluent oil content consistently inside the limit is difficult, running cost is high because of chemical dosing and energy demand, and resistance to water quality swings is only moderate. Demulsification capability is simply insufficient for this duty.
• Membrane separation intercepts droplets with microporous media and delivers high outlet precision, but fouling and failure come quickly. Membrane replacement and cleaning costs are extremely high, and the process cannot cope with high emulsification and high impurity loads.

Put simply, all three share the same weak points: incomplete demulsification, low separation efficiency and high operating cost. That is the gap SINOKLE set out to close with process integration rather than incremental tuning.

The three-stage SINOKLE train

The complete process runs CDFU cyclonic dissolved-air flotation first, then a precision filtration stage, then a KHC coalescing oil remover, forming a progressive separation chain in which each unit targets a smaller droplet population than the one before.

Stage one: CDFU cyclonic dissolved-air flotation

Sour water from the coking fractionator enters the CDFU unit, where cyclonic intensified air dissolution generates ultra-fine bubbles 5–30 μm across under high shear. Working with a dedicated demulsifier, the unit breaks the stable emulsion so that fine droplets coalesce and float, achieving a first three-phase separation of oil, water and solids. Separated waste oil is discharged to a receiving tank for recovery, while nitrogen and VOCs released at the top go to a gas recovery system.

Stage two: precision filtration

The CDFU effluent is pumped to precision filters arranged A/B, one duty and one standby. High-precision media intercept residual fine droplets and suspended solids, and an on-line backwash water and gas system keeps the filters running continuously. Backwash gas also returns to the gas recovery system rather than venting.

Stage three: KHC coalescing oil removal

Precision-filtered water then enters the KHC coalescing oil remover. The oleophilic and hydrophobic properties of the dedicated coalescing packing cause the remaining droplets to coalesce and grow, completing deep oil-water separation. Final effluent oil content stays consistently within limit and flows on to the stripper tower for deammoniation and desulphurisation, while separated waste oil returns to the coking fractionator for resource recovery.

Throughout, the design is fully enclosed, which keeps hydrogen sulphide and other toxic gases inside the system rather than in the working area. Automatic control across the whole flow keeps manual intervention low, which matters on a unit that runs around the clock.

What the configuration changes in practice

Against a stable, highly emulsified feed, CDFU flotation combines cyclonic intensification with ultra-fine bubbles to achieve demulsification well beyond conventional flotation, and the KHC stage deepens the separation further. The result is an emulsified oil removal rate above 95% with effluent oil content held steadily below 10 mg/L — and the specification the equipment is designed to hit is a limit of <10 mg/L, so there is margin rather than a knife-edge.

Running cost falls in several places at once. The air dissolution system and chemical regime are optimised, no chemical agents are added at all, energy consumption drops by 30% to 50%, and sludge production falls sharply. Waste oil recovery climbs above 90% and the recovered oil goes straight back to production units, turning a disposal cost into a modest revenue line.

Adaptability is the third gain. The system tolerates wide swings in influent oil content, degree of emulsification and temperature, so it holds compliance even under high sulphur, high emulsification and high suspended solids conditions. The complete unit is modular and skid-mounted, occupying about one third of the footprint of a conventional process, which means it can be installed into an existing refinery with no large-scale civil work and a short commissioning window.

The enclosed design also removes the leakage risk that makes operators nervous about sour water service, and full recovery of waste oil and gas supports the wider move toward energy saving and resource utilisation in refining. Deployments across several large domestic refining and petrochemical enterprises have handled cumulative sour water volumes above one million tonnes per year at consistent compliance, which is the kind of track record that matters when a retrofit has to justify itself on operating data rather than on a brochure.