Why Cartridge Coalescers Fail on Heavy Tar: Four Mechanisms Behind Rapid Plugging and Separation Loss

The pattern seen on site

In petrochemical coking, coal-tar deep processing and heavy-oil pretreatment, cartridge-type coalescing oil separators are a habitual first choice. They work beautifully on gasoline, diesel and light lubricating oil. Put heavy tar through them and they fail fast: pressure drop climbs, separation collapses and the unit stops meeting production requirements.

Field symptoms are consistent enough to be diagnostic. Cartridges clog within half a day, effluent oil exceeds limits, alarms fire repeatedly and operating conditions never settle. The root cause is single and structural: the design principle of a cartridge coalescer is mismatched with the medium and impurity characteristics of heavy tar.

First, what the cartridge is designed to do

A cartridge coalescing oil separator is bespoke equipment for low-viscosity, clean light oils. Its core is coalescence growth followed by gravity settling, and the whole mechanism depends on stable fluid properties.

Demulsification coalescence happens when the porous fibrous medium captures micron-scale emulsified water droplets, which collide, adsorb and merge into millimetre-scale droplets. Hydrophobic separation then uses hydrophilic-oleophobic or oleophilic-hydrophobic media and interfacial-tension differences to detach the grown droplets from the oil phase. Gravity settling finishes the job as large droplets sink under their own weight and the layers separate.

The prerequisites are strict: low oil viscosity, no large amount of sticky resins and asphaltenes, few solid impurities, good fluidity and droplets free to settle. That describes gasoline, diesel and light white oil. It does not describe heavy tar.

Reason one: high viscosity and high resin content disable coalescence

Heavy tar, coal tar and coking heavy fractions are defined by ultra-high viscosity, high resin content and high asphaltene content, and each of those breaks a link in the chain above.

Light oil passes through cartridge pores with little resistance, so tiny droplets dwell, collide and coalesce inside the medium. Heavy tar is extremely viscous even at operating temperature and flows poorly, which produces two fatal effects. Coalescence efficiency falls to zero because the viscous oil wraps the droplets and suppresses their diffusion and collision, so droplets that should clump and grow stay as a micron-scale emulsion and never form anything settleable. Worse, the few large droplets that do form are torn apart again by the drag of the viscous fluid and the shear of the cartridge pores, producing a futile cycle of coalescing as much as it breaks up.

Meanwhile resins and asphaltenes are strongly adhesive. They lay down a viscous coating over the fiber surface, cover the coalescence active sites and destroy the interfacial-tension characteristics of the medium, so the cartridge loses all coalescence and demulsification ability.

Reason two: coke fines plug the medium irreversibly

Heavy-tar fractions inevitably carry fine coke powder, carbon powder and solid particles. These are inherent to coking and tar operations and they are the nemesis of a cartridge coalescer, whose pores must be dense and regular to capture fine droplets.

Resins first form a mucous film on the surface, then coke fines adhere and accumulate into a dense filter cake, triggering severe bridging pore-plugging. This is not ordinary uniform clogging but a dual failure of viscous coating plus solid embedding. Plugging is extremely fast: in most field cases the inlet-outlet differential passes its limit within hours of a new cartridge entering service, so continuous operation becomes impossible.

It is also irreversible. Ordinary dust clogging can be blown or washed out, but tar resins penetrate and stick inside the fibers and then harden and scale under high-temperature, high-viscosity conditions. The cartridge is finished, cannot be regenerated, and must be replaced, which makes O&M cost extremely high.

Reason three: the density-difference logic fails

The final step of cartridge coalescence depends on an oil-water density difference plus gravity settling, and heavy tar breaks that assumption as well. Light oil and water differ enough in density for coalesced droplets to sink quickly. Heavy tar is very dense, with some fractions approaching or nearly reaching water density, so the difference is compressed. At the same time the viscous oil generates strong fluid entrainment.

Grown water droplets cannot escape the oil-phase entrainment, so they stay suspended in the tar instead of settling. Some heavy oil droplets sink and mix into the water phase instead, creating mutual mixing and two-way entrainment. The effluent ends up severely over its oil limit, dewatering targets are missed and the separation becomes meaningless.

Reason four: the design envelope does not overlap

Industry standards state plainly that this equipment suits only low-viscosity light oils, and that once medium viscosity exceeds 220 mm²/s, coalescence separation essentially fails. Heavy tar operating viscosity is far past that threshold, which makes it a prohibited condition rather than a difficult one.

Heavy-tar service is additionally high-temperature, high-impurity and highly emulsified-stable, while the fibrous medium has limited temperature, viscosity and pollution resistance. Long-term running deforms, ages and delaminates the medium, which not only fails to separate but adds secondary impurity pollution from the medium itself.

What to use instead

Heavy-tar oil-water separation requires abandoning the fine-filtration logic of cartridge coalescence in favour of filter-free, high-throughput, viscosity-tolerant, anti-clogging macro-stratification processes. Three mature options exist.

• Gravity settling plus demulsification separation: dedicated demulsifiers with large-volume chambers for slow stratification, suited to high-viscosity, high-impurity tar systems.
• Horizontal three-phase separator: density difference, flow rectification and baffle coalescence with no precision pore structure, so there is no membrane to plug and nothing to fail, well suited to complex heavy-tar conditions.
• Heated demulsification plus centrifugal separation: moderate heating lowers tar viscosity and weakens entrainment while centrifugal force strengthens separation, suited to highly stable emulsified tar.

The governing principle is straightforward: heavy tar shows high viscosity, high resins, high coke fines, small density difference and strong emulsification, and a cartridge coalescer fails on every one of them. Matching the process to the medium is the whole job.