Sour Water From Delayed Coking: Three Mechanisms, Three Stages, One Train

Where sour water comes from

Refinery sour water is a by-product of several units at once: atmospheric and vacuum distillation, catalytic cracking, delayed coking and catalytic hydrogenation. Sulfur and nitrogen compounds in the feedstock are cracked at high temperature or converted through catalytic cracking and hydrocracking, forming H2S and NH3-N, and those become sulfur- and ammonia-bearing wastewater once the product stream is condensed, dehydrated or water-washed.

Coking sour water is the difficult member of that family. It carries coke fines at small particle size, oil at high concentration and a well-developed emulsion, plus elevated sulfur, ammonia nitrogen and phenols, and the composition swings widely between campaign periods. A plant that designs for the average of that stream will be out of specification half the time.

Three problems, not one

The instinct when a stream fails is to look for a single better unit. Sour water does not cooperate with that approach, because it contains three distinct problems that need three distinct mechanisms. Solids have to be intercepted before they reach anything that can blind off. Emulsified oil has to be grown into droplets large enough to separate. And the variability of the feed has to be absorbed somewhere upstream, or every unit downstream spends its life chasing a moving target.

Stage one: fine filtration as the interceptor

Fine filtration sits in the middle of the train and traps fine coke particles there, so they cannot enter the downstream units. The placement is deliberate. Put filtration at the very front and it blinds within days on free oil; put it at the end and the coalescing media is already loaded with fines it was never designed to hold. In the middle it protects the equipment that cannot tolerate solids while seeing a feed the upstream stage has already partly cleaned.

Stage two: the KHC coalescing separator at the tail

The KHC high-efficiency coalescing separator works at the tail end, where its job is the fraction nothing else could touch. Fine emulsified oil droplets coalesce on the media, grow into droplets large enough to rise, and separate. Because the unit sits last, it sets the final effluent number, which is why its performance is the one the permit depends on.

Stage three: CDFU at the front, absorbing the shocks

CDFU, the high-efficiency compact cyclonic flotation unit, responds quickly at the front end and offsets influent shocks with stable flotation separation. The short hydraulic retention time that makes the unit compact is also what makes it responsive: there is very little inventory to turn over before a change at the inlet shows up as a change at the outlet, so the control response arrives while the excursion is still happening.

Why three stages instead of one good unit

Flotation, filtration and coalescence each specialise in one mechanism, and they connect in a sequence where each one hands the next a feed it can handle. Flotation takes the bulk oil and rides the swings. Filtration takes the solids. Coalescence takes the fine emulsified remainder. Stack the three and the effluent oil content stays firmly at a low level without any single unit being asked to do something it cannot.

The numbers that define the claim

Emulsified oil removal precision reaches 0.1 um, and effluent oil content stays below 10 ppm. Those two figures are the core indicators of the combined process, and they are worth reading together: the first says what size of droplet the train can capture, the second says what the water actually carries out. A process that quotes one without the other is describing half a result.

What this buys the refinery

For a refining and chemical enterprise the value goes past compliance. A 0.1 um removal precision leaves nowhere for micro-emulsified oil to hide, and effluent oil content below 10 ppm creates stable conditions for downstream stripping and biological treatment. That matters because sour water is only one link in the refinery wastewater system and it often affects the whole chain: incomplete front-end oil removal sends emulsified oil travelling through the process, aggravating foaming and scaling in the stripper, pushing chemical dosing up and eventually disturbing biological stability. SINOKLE addresses the problem at the source with 0.1 um precision, which is the cheapest place to solve it.

The shock-tolerance argument

Coking sour water, with its complex composition and wide content fluctuation, imposes hard requirements on shock resistance, and the value of a combined process is that buffers form naturally between stages. CDFU cyclonic flotation absorbs the influent shocks, fine filtration stabilises mid-stage water quality, and the KHC coalescing separator secures the final output. Three stages share the load, so the system keeps deterministic effluent performance while the influent wanders.

Reading the leadership claim correctly

Claims of a leading position are usually about a single machine, and that reading misses the point here. What is being claimed is the leadership of combined-process logic: for the specific form in which coke fines, emulsified oil and high-concentration pollutants coexist, three mechanisms are assigned precisely and applied in progressive stages, which is what makes the result both feasible and reliable. Any plant comparing options should test that logic against its own analysis rather than against the equipment list.

The direction the industry is moving

Sour water management is shifting from end-of-pipe compliance towards front-end purification: solve the emulsified oil and coke fines before they enter stripping and biological units, and the whole flow relaxes. A train benchmarked at 0.1 um removal precision and effluent oil content below 10 ppm fits that shift, and it gives refining and chemical enterprises a quantifiable basis for comparison rather than a claim to take on trust.