Tank Farm Drained Water: Two Separation Mechanisms in Series at 100,000 mg/L

Two properties that defeat every single unit

The difficulty of treating tank farm drained water comes down to two numbers and one physical fact. The oil content is extremely high, exceeding 100,000 mg/L under some operating conditions, and the emulsion is deep, with fine droplets held inside a stable interfacial film. Neither property is unusual on its own. Together they rule out any single separation technology, because a unit sized for the oil load is never fine enough for the droplet size, and a unit fine enough for the droplets is blocked by the load long before it performs.

The SINOKLE answer is not a better single unit but a sequence: two separation units with different mechanisms, connected in series, forming a two-stage purely physical process. Understanding why it works is mostly a matter of understanding what each stage is asked to do.

Why the stages cannot be merged

The choice of separation technology follows from the physical form of the pollutant, not from its concentration. Free oil and dispersed oil form relatively large droplets that can be removed by density difference or by bubble attachment. Emulsified oil is different: the droplets are tiny and wrapped in an interfacial film, so they are difficult both to give sufficient rising velocity and to capture effectively with bubbles.

A single technology usually covers only part of the droplet-size range. Whatever falls outside that range passes through and becomes the source of out-of-spec effluent, and in a tank farm drain the out-of-range fraction is large. Hence the logic of two stages. The first addresses droplet size, using coalescence to make fine droplets grow before separation is completed. The second addresses floatability, taking out the easily floated pollutants first and lowering the load on everything downstream. Load staging plus a division of mechanisms is the whole framework.

Stage one: droplet growth in a cyclonic field

The HCC high-efficiency cyclonic coalescer works at two levels that reinforce each other.

At the cyclonic level, the oily drained water forms a high-speed rotating flow field inside the unit. Oil and water differ in density, so they experience different centrifugal forces: the oil phase gathers toward the low-pressure core while the water phase moves outward, and a preliminary phase separation is achieved. Centrifugal action amplifies the effect of the density difference far beyond what gravity settling can deliver in the same footprint. That amplification is the reason a cyclone can handle a load that would blind a gravity separator.

At the coalescing level, fine oil droplets repeatedly collide, adhere and merge on the surface of the coalescing media and in the shear field of the flow. Once small droplets merge into larger ones, their buoyancy and their centrifugal separation velocity rise markedly, and they acquire the conditions for effective removal. The engineering difficulty sits here rather than in the cyclone: the coalescing process depends on providing sufficient collision opportunities and suitable flow-field conditions, which is exactly what the structural design of the coalescer controls. Too gentle a shear and droplets never meet; too violent a shear and the merged droplets are torn apart again.

Stage two: giving the residual droplets buoyancy

CDFU cyclonic dissolved air flotation is built around the combination of dissolved gas and flotation. The system dissolves gas into water under pressure to form dissolved-air water; when that water enters the atmospheric separation zone, the sudden pressure drop releases the dissolved gas as microbubbles. Those bubbles are small, numerous and high in specific surface area, which gives them ample contact opportunity with the oil droplets and suspended solids still in the drained water.

Once bubbles attach to a droplet, a combined body forms whose density is markedly lower than that of water. Fine droplets that would otherwise rise extremely slowly gain enough buoyancy from the attached bubbles to rise quickly to the surface, where they form a scum layer that the skimming device collects and discharges. This is the strengthening effect of microbubble flotation on emulsified oil, and it is worth stating precisely what it does and does not do: it does not change the chemical nature of the oil, only the forces acting on the droplets. That distinction is why the separated oil can go back into the crude slate without further treatment.

What the combination delivers

The results show up on two sides. On the water side, emulsified oil and suspended solids are removed efficiently and the effluent meets discharge standards. On the oil side, the oil separated in both stages is collected and returned directly to the crude oil tanks, which converts a disposal cost into recovered product.

No chemical agent is dosed at any point, so no oily chemical sludge is generated and there is no secondary pollution to manage. That single design decision removes an entire downstream problem: no sludge thickening, no dewatering, no hazardous waste classification, and no disposal contract for a stream that is mostly water and oil.

Three differences from gravity plus flocculation

Compared with the conventional route of gravity separation followed by flocculation, the differences sit in three dimensions.

• Separation driving force. The conventional route relies on gravity and chemical destabilisation; the two-stage process relies on microbubble buoyancy and a centrifugal field, both of which act more strongly on fine droplets than gravity ever will.
• Secondary pollution. The conventional route produces chemical sludge, which has to be dewatered, transported and disposed of. The two-stage process doses no chemicals, so the sludge problem is eliminated at its source rather than managed downstream.
• Resource flow. The conventional route treats oil as a pollutant to be destroyed or landfilled. The two-stage process recovers it and returns it to the tanks.

Where the design needs care

Two-stage physical separation is not maintenance-free. The coalescing media has a finite void volume and will eventually collect waxes, asphaltenes and particulate matter, so the pressure drop across the coalescer is the parameter to trend rather than the effluent quality. Once the drop starts climbing at constant flow, the bed is loading and cleaning is due; waiting for the effluent to fail means the second stage has been carrying the first stage's work.

The flotation stage has its own limits. Microbubble attachment depends on the oil being present as discrete droplets rather than as a true dissolved phase, so a drain that has been sitting with a strong surfactant load may not respond. In that case the first stage is doing the work and the second is only polishing, which is a different operating regime with a different chemical demand.

The water quality characteristics of tank farm drained water mean no single technology solves the problem in one step. SINOKLE combines CDFU cyclonic dissolved air flotation with the HCC high-efficiency cyclonic coalescer into a two-stage purely physical process: flotation solves floatability, coalescing solves droplet size, and compliant discharge and oil recovery are achieved together. The key to understanding the process is the division of mechanisms between the two stages, not the performance of either unit alone.