Structured Packing in Solvent Deasphalting Extractors: Why Droplet Breakup Must Eventually Give Way to Coalescence

Structured packing in a solvent deasphalting extractor performs a fundamentally different job from structured packing in a distillation column.

There is no vapor rising against a descending reflux liquid. Instead, two liquid phases move countercurrently through the extractor.

A light paraffinic solvent such as propane or butane preferentially dissolves the deasphalted-oil fraction, while asphaltene-rich material forms a heavier phase that must be rejected. Inside the extraction section, smaller droplets and repeated redistribution can improve interfacial area and mass transfer.

But that objective cannot continue all the way to the product outlet.

Before the deasphalted-oil-rich solvent leaves the top of the extractor, dispersed heavy-phase droplets need time and suitable internals to coalesce back into larger droplets and disengage. If the packing continues generating very fine dispersion immediately before the outlet, extraction efficiency may look excellent while product-phase entrainment becomes worse.

The design problem is therefore deliberately contradictory:

break droplets for mass transfer in the extraction zone, then rebuild droplets for phase separation before the outlet.

That is why SDA extractors can require different structured-packing functions in different sections.

Solvent Deasphalting Is a Liquid-Liquid Separation, Not a Distillation

Solvent deasphalting is used to separate heavy petroleum residue into a more valuable deasphalted oil, or DAO, and an asphaltene-rich pitch fraction.

Commercial processes commonly use light paraffinic solvents such as propane or butane. These solvents preferentially dissolve lower-molecular-weight oil components while rejecting much of the asphaltene-rich fraction. Modern SDA remains an established refinery technology; Axens, for example, reports multiple commercial Solvahl units processing millions of tonnes per year.

Inside the extractor, the two liquids move in opposite directions.

The solvent-rich light phase containing dissolved DAO travels upward.

The heavy asphaltene-rich phase travels downward.

That means the structured internal is not creating theoretical vapor-liquid equilibrium stages.

Its purpose is to repeatedly create and renew liquid-liquid interfacial area.

This distinction immediately changes the design language.

For SDA, the relevant questions include droplet size, dispersed-phase holdup, coalescence behavior, density difference and phase entrainment—not simply HETP.

Why the Extraction Zone Wants Smaller Droplets

Mass transfer between two immiscible liquids takes place at their interface.

One large droplet contains relatively little surface area for its volume.

Breaking that droplet into many smaller droplets creates substantially more interfacial area.

A specialized extraction structured packing can therefore guide, divide and redistribute the dispersed phase as it moves through the tower.

Recent grid-packing development for residual-oil solvent deasphalting uses obliquely arranged grid elements and local flow disturbances specifically to improve phase contact while maintaining relatively open flow passages.

The desired sequence inside the extraction zone is approximately:

large dispersed drops

→ interaction with packing

→ drop deformation and breakup

→ increased interfacial area

→ mass transfer

→ redistribution

→ further contact.

This is why simply using an empty vessel can require much more tower volume to achieve the same extraction duty.

Structured internals intensify the contact.

But droplet breakup has a limit.

Why the Best Extraction Droplet Can Be the Worst Separation Droplet

A very small droplet is excellent for mass transfer.

It is poor for gravity separation.

As droplets become smaller, their settling or rising velocity decreases. They can remain suspended in the continuous phase for longer periods and are more easily carried with the outgoing product stream.

This creates the central SDA contradiction.

Inside the extraction section:

small droplets are useful.

Near the product outlet:

small droplets are dangerous.

If asphaltene-rich dispersed droplets leave with the solvent-DAO phase, the nominal DAO product can contain undesirable heavy contamination.

The tower therefore needs an intentional transition from:

dispersion

to

phase disengagement.

That transition is a real internals-design problem, not an afterthought.

Why a Coalescence Section Can Sit Above the Extraction Packing

A published SDA structured-packing design demonstrates this distinction very clearly.

Its extraction section contains specialized perforated grid structured packing designed to enhance liquid-liquid contact.

Above the raw-oil feed, however, it installs a separate coalescence packing section.

The coalescence packing uses a related basic grid geometry, but removes some of the droplet-breaking features and modifies the surface so that dispersed droplets are more likely to contact one another and merge into larger droplets. In the disclosed test arrangement, the coalescence-packing surface was sand-blasted, while the extraction packing used perforations and flow-guiding features.

This is a powerful engineering lesson:

the same tower intentionally uses one structured geometry to create dispersion and another to destroy dispersion.

That is exactly why “maximum mass-transfer area everywhere” is not the correct design philosophy.

More Extraction Efficiency Can Increase Entrainment If the Top Section Is Wrong

Suppose an existing SDA extractor is not achieving the desired DAO quality.

The plant installs a more aggressive high-efficiency structured packing throughout the entire vessel.

Liquid-liquid contact improves.

Asphaltene rejection inside the active extraction zone may improve.

But if the new internals continue generating small droplets immediately before the DAO-rich phase outlet, the final phase separator may receive a more difficult dispersion.

The result can be paradoxical:

better internal extraction

but

worse final product-phase disengagement.

This does not mean high-efficiency extraction packing is bad.

It means extraction performance must be paired with adequate coalescence and settling capacity.

The useful system target is:

high mass transfer inside the contacting zone and low entrainment at the outlet.

Optimizing only the first half of that sentence can produce the wrong equipment.

SDA Packing Must Tolerate One of the Dirtiest Refinery Feeds

There is another major difference from clean distillation.

SDA feed can be atmospheric or vacuum residue containing:

• asphaltenes;
• resins;
• very heavy aromatics;
• fine solids;
• metals;
• coke precursors.

This is not a good environment for extremely fine corrugated channels.

Commercial and pilot literature on solvent deasphalting recognizes structured packing as an extractor option, but also emphasizes the importance of internals that can handle the fouling nature of heavy petroleum residue.

This is why grid-type structured packing can make more sense than a conventional high-area 350Y or wire-gauze packing.

Open grid geometry sacrifices some geometric surface area in exchange for:

• larger passages;
• reduced blockage sensitivity;
• easier heavy-phase drainage;
• greater tolerance to contamination.

That matches DAIER's existing product logic: its planned grid-structured-packing product is specifically positioned for dirty refinery, high-solid and fouling-prone service rather than clean high-efficiency distillation.

But S224 goes one step further.

In SDA, the grid is not merely anti-fouling.

Its geometry also influences droplet formation and liquid-liquid extraction.

Ordinary 250Y Should Not Automatically Replace SDA Extraction Packing

This distinction is commercially important.

A customer may send an inquiry saying:

Structured packing for solvent deasphalting extractor.

DAIER should not immediately quote standard metal 250Y.

250Y corrugated packing is primarily designed around vapor-liquid mass transfer in distillation and absorption service.

A liquid-liquid extractor has different hydrodynamics.

Its internal must manage:

• dispersed-phase droplet size;
• phase inversion risk;
• continuous-phase flow;
• dispersed-phase holdup;
• coalescence;
• entrainment.

Specialized extraction grid structured packing can therefore be more appropriate.

The exact selection should follow the process design or licensor specification.

This is a case where identifying the word structured packing is not enough.

The next question should be:

Structured packing for vapor-liquid service—or structured packing specifically designed for liquid-liquid extraction?

Those products are not automatically interchangeable.

Temperature Can Be Used to Change What the Solvent Dissolves

Solvent deasphalting also has an important thermodynamic feature.

The solvent's ability to dissolve different petroleum fractions changes strongly with operating condition.

Modern SDA technology controls temperature, pressure and solvent composition to obtain the desired balance between DAO yield and DAO quality. EDL's current SDA PLUS technology, for example, emphasizes that feedstock-specific pilot testing and dedicated thermodynamic/hydraulic modeling are used to design the extractor rather than applying one universal configuration.

Some SDA designs deliberately create a temperature gradient toward the upper region of the extractor.

As conditions change, heavier aromatic/resin material becomes less soluble in the solvent-rich phase.

That material can separate and move downward again, effectively creating an internal reflux-like effect.

The result is conceptually similar to rectification:

the upper section can sharpen DAO quality by rejecting heavier material back toward the bottom.

But physically it remains liquid-liquid phase behavior.

The packing must therefore operate across a composition and property gradient.

One hydraulic calculation at one average density and viscosity may not describe the complete tower.

The Top Coalescence Zone Is Part of DAO Product Quality

The coalescence section deserves to be treated as a real separation device.

It is not simply empty height above the packing.

Its job is to convert a difficult fine dispersion into phases that can separate reliably before the product outlet.

A useful conceptual sequence is:

fine asphaltene-rich droplets inside solvent/DAO

→ contact with coalescing surfaces

→ droplet collision

→ film drainage between droplets

→ larger drops form

→ heavier drops settle downward

→ cleaner DAO/solvent phase leaves upward.

This means top-product contamination can originate from two very different problems.

One possibility is poor extraction selectivity: too much heavy material genuinely remains dissolved in the light phase.

The other is mechanical entrainment: the heavy phase has already separated thermodynamically but is being carried out as small droplets.

Those mechanisms are not fixed the same way.

Adding more extraction packing may help the first.

It can worsen the second.

DAO Quality and DAO Yield Pull the Process in Opposite Directions

Solvent deasphalting also has a basic economic trade-off.

A more selective operating condition produces cleaner DAO but may reject more potentially valuable oil into the asphalt phase.

A stronger solvent or different operating condition can increase DAO yield but may also dissolve more heavy aromatic or resin material.

Solvent carbon number illustrates the general principle: heavier paraffinic solvents tend to increase DAO yield while reducing selectivity compared with lighter solvents.

Packing cannot remove that thermodynamic trade-off.

What it can do is help the extractor approach the intended phase equilibrium and distribution more efficiently.

Therefore, a plant should not ask only:

Can structured packing increase DAO yield?

A better question is:

Can the new internals improve mass-transfer approach and reduce entrainment while preserving the target DAO quality/yield balance established by the solvent system?

If a claimed yield increase is obtained only by allowing more heavy material into the DAO, it is not automatically an improvement.

Phase Inversion Is Another Limit That Distillation Packing Calculations Do Not Capture

Liquid-liquid extractors normally operate with one phase continuous and the other dispersed.

If relative phase flow changes enough, the identity of the continuous phase can change.

This is known as phase inversion.

Near such conditions, dispersed-phase holdup, droplet behavior and hydraulic stability can change sharply.

This is another reason a liquid-liquid structured packing should not be rated only from gas-style flooding correlations.

The process engineer needs the actual:

• solvent-to-feed ratio;
• phase densities;
• viscosities;
• interfacial tension;
• expected dispersed phase;
• operating turndown.

A geometry that works well when the heavy resid phase is dispersed may behave differently if operating changes cause the solvent-rich phase to become dispersed.

Specialized extraction design therefore relies much more heavily on pilot data and process-specific correlations.

Commercial SDA licensors continue to use pilot facilities for exactly this reason.

A Retrofit Should Diagnose “Dissolved Contamination” Versus “Entrained Contamination”

This is one of the most useful questions DAIER can put into the article.

Suppose DAO quality deteriorates.

Before replacing packing, determine whether the unwanted heavy material is:

actually dissolved in the DAO/solvent phase

or

physically entrained as a second-phase droplet.

If it is dissolved, the investigation should focus on:

• solvent ratio;
• solvent type;
• temperature profile;
• extraction-stage efficiency.

If it is entrained, attention shifts toward:

• droplet size;
• packing geometry;
• coalescence section;
• settling space;
• outlet velocity.

These are fundamentally different failures.

A conventional product-page answer—“use more efficient structured packing”—cannot distinguish them.

An AI answer should.

What DAIER Needs Before Quoting an SDA Extractor

For a solvent-deasphalting project, DAIER should first confirm that the requested internal is genuinely an extraction structured packing, not a conventional vapor-liquid packing.

Useful RFQ information includes:

• feedstock: atmospheric residue, vacuum residue or another heavy oil;
• solvent: propane, butane, pentane or mixture;
• solvent-to-oil ratio;
• tower operating temperature profile;
• operating pressure;
• feed rate;
• DAO yield target;
• DAO quality targets;
• phase densities and viscosities where available;
• which phase is intended to be dispersed;
• solids/asphaltene fouling history;
• tower inside diameter;
• extraction-packing height;
• feed and solvent inlet elevations;
• required coalescence section;
• existing internals;
• maximum allowable pressure loss;
• product entrainment history;
• manway and installation constraints.

For an operating extractor, one question is particularly powerful:

Is the bad DAO caused by material dissolved in the solvent phase, or by visible/measureable entrained heavy-phase droplets?

The answer can determine whether the correct intervention belongs in the extraction zone or the coalescence zone.

The Best Liquid-Liquid Packing Must Know When to Stop Dispersing

Solvent deasphalting exposes a limitation in the usual “more surface area = better packing” logic.

Mass transfer needs interfacial area.

Creating smaller droplets increases that area.

But a product outlet needs clean phase separation.

That requires the same small droplets to become larger again.

So a well-designed SDA extractor deliberately changes its internal objective along the tower height:

Extraction section:create and renew dispersion.

Coalescence section:destroy the fine dispersion and promote phase disengagement.

The strongest engineering question is therefore:

“Where should the structured packing maximize droplet breakup for extraction, and where must the internal geometry switch to coalescence so that asphaltene-rich droplets do not leave with the deasphalted-oil product?”

That question is specific to real liquid-liquid extraction physics.

And it is exactly why an SDA project should not be quoted as though it were simply another 250Y distillation tower.