Refinery Overhead Sour Water: Breaking a Surfactant-Locked Emulsion in Minutes
Crude processed worldwide is getting heavier and more variable, and the units at the front of the refinery feel it first. The atmospheric and vacuum distillation column is the forerunner of the whole plant, and its overhead sour water now carries high oil content that readily forms a stable emulsion. Removing that oil efficiently and cheaply has become one of the practical problems standing between a refiner and long-cycle, low-carbon operation.
Why overhead water has to be treated at all
Crude is first split into straight-run fractions by boiling point, then the off-spec components are removed or converted until the products meet specification. Distillation remains the most common and most economical separation step in refining, and it is always the first one.
During pre-distillation and atmospheric/vacuum distillation, the overhead vapour is condensed and separated into oil, water and gas in the overhead separator. Because the condensed water carries a high concentration of sulfides, the industry calls it sour water. In principle it goes to the stripper and then onward as purified water for reuse.
In practice, heavier and more diverse crude slates mean the overhead condensate emulsifies and carries oil. That costs the refinery high-value light product and sends a shockwave downstream in three specific ways. Oil is a surfactant, so it foams under the intense vapour-liquid contact inside the stripper; foam destroys mass transfer, raises steam consumption and cuts tower capacity. Emulsified oil deposits on trays and reboilers, blocking them, degrading mass-transfer efficiency and creating local hot spots that shorten equipment life. And hydrocarbons in the feed produce black sulfur during stripping, while oil carried into liquid ammonia damages by-product purity.
What holds the emulsion together
Gravity settling cannot break this emulsion, and the reason is interfacial chemistry rather than droplet size. Heavy inferior crude brings natural surfactants, asphaltenes, resins and petroleum acids. On top of that, extraction additives such as alkylbenzene sulfonic acid and fatty alcohol polyoxyethylene ether, plus the alkaline corrosion inhibitor dosed into the atmospheric/vacuum unit, together build a large population of anionic surfactants in the water.
These large molecules hydrate weakly, so they penetrate the water film, reach the inner Helmholtz plane, displace the water dipole layer and adsorb strongly onto the oil-droplet surface. Interfacial tension collapses. The negative charge on the droplets and the anionic surfactants then form a dense Stern electric double layer with a high x potential (Zeta potential). At refinery temperature, thermal motion and diffusion thicken that double layer and push the x potential higher, so electrostatic repulsion between droplets becomes very large. Finally, the surfactants build an oriented adsorption layer at the interface, a protective interfacial film that physically prevents droplets from fusing when they collide. Gravity has no answer to any of this.
What the industry does today
To keep the stripper stable, oil in the sour water feeding it is generally required to stay below 50 mg/L. Most refineries still pursue that with large-tank gravity settling, usually two tanks in series, one settling and one buffering, with a retention time of 50-70 hours. The approach consumes enormous tank volume and plot area, still cannot guarantee removal of a surfactant-stabilised emulsion, and makes the floating oil that does separate awkward to collect.
Two physical lines of defence
More tank capacity and more chemical will not solve an interfacial-film problem. SINOKLE has accordingly developed a purely physical combination: a silicon-carbide filter followed by a KHC high-efficiency coalescing oil remover.
SiC filter: interception and pre-demulsification
Overhead sour water carries coke powder and ferrous sulfide (FeS) particles, and these fines adsorb at the oil-water interface to form a solid-stabilised Pickering emulsion that is far more rigid than a normal one. The first stage is therefore an SiC tubular or membrane filter. Silicon-carbide ceramic tolerates high temperature and strong acid and alkali, and its surface is strongly hydrophilic and oleophobic. At high flux it removes micron-level suspended solids, destroying the basis of the solid-stabilised emulsion, while pore shear and the oleophobic surface begin coalescing and breaking the emulsified droplets.
KHC coalescing oil remover: collision and phase separation
The pre-filtered water then enters the KHC unit, packed with a polymer oleophilic modified-fibre coalescing cartridge developed and patented in-house. Water carrying micro- and nano-scale droplets passes through the dense bed, and under combined hydrodynamic and surface-chemical action the droplets go through four steps on the fibre surface: interception, wetting, attachment and fusion. Captured across a very large interfacial contact area, they are guided past the Stern double-layer repulsion and merge into droplets of tens of microns or even millimetres. Those rise quickly to the oil-collection zone at the top of the vessel and are drawn off.
Industrial data from SINOKLE at Luoyang Petrochemical, on a 15 m3/h sulfur-wastewater system, is instructive: with average influent oil as high as 15%, the effluent oil concentration stayed below 50 mg/L and the single-system removal efficiency reached 99.96%.
The economics are the point
Against 50-70 hours of gravity settling, this combination needs about 5 minutes of residence time. Because no demulsifier is dosed, the recovered light emulsified oil is clean and can go straight back into the unit for re-refining, avoiding the processing loss that poor recovered-oil quality imposes under the old route.
Take a 30 m3/h sour-water stream at 2% oil running 8,000 hours a year. If the combined process recovers 300 mg/L more light emulsified oil than gravity settling would, the light-oil recovery alone is worth roughly 430,000 RMB a year at 6,000 RMB per tonne. Add the avoided light-oil processing loss of 2.4 million RMB and the annual comprehensive benefit against gravity settling exceeds 2.84 million RMB.
There is a second, larger effect. Removing oil ahead of the stripper reduces the oil load on the tower. Temperature and pH strongly influence stripping, and under normal temperature, pH 7.5 and a well-controlled gas-liquid ratio, effective front-end de-oiling combined with other desulfurisation measures can partly substitute for high-energy steam stripping. At the 30 m3/h scale, partial substitution plus the saving on stripper blockage maintenance, about 120,000 RMB a year, can bring total energy and direct benefits up to 10.56 million RMB per year.
Refining margins are decided in details like this one. Turning a stream that fouls equipment and loses product into re-refinable oil and compliant purified water, in five minutes rather than three days, is a process change that pays for itself several times over.