Liquid-Liquid Coalescer Design Manual - AMACS

SourceStabilityDroplet SizeRangeFlash Drum Emulsionswith 5 % Dispersed Phase,Static MixersWeak100-1000 micronsFlash Drum Emulsions with 5 % Dispersed Phase,Impellor Mixers, ExtractionColumnsModerateStep 1 – Droplet CaptureThe first step of coalescing is to collect entraineddroplets primarily either by Intra-Media Stokes Settlingor Direct Interception. Figure 4 gives the useful zones ofseparation for various mechanisms. Elements that50-400 micronsCentrifugal Pump Discharges,Caustic Wash Drums, LowInterfacial Tension EmulsionsStrongHaze from Condensing in BulkLiquid Phases, SurfactantsVery StrongGiving Emulsions With VeryLow Interfacial Tensions10-200 microns0.1-25 micronsTarget Size, Microns1000FIGURE 4ZONES WHERE DIFFERENTCOALESCING MECHANISMS APPLY1001010.10.1Table 1Basic Design ConceptsOperating Principles of a Coalescer1101001000Droplet Size, Micronsdepend on Intra-Media Stokes Settling confine the disLiquid-Liquid Coalescers are used to accelerate the tance a droplet can rise or fall between parallel platesmerging of many droplets to form a lesser number of or crimps of packing sheets (Figure 5). This is comdroplets, but with a greater diameter. This increases pared to simple gravity separators in which the travelthe buoyant forces in the Stokes Law equation. SettlingFIGURE 5of the larger droplets downstream of the coalescerelement then requires considerably less residenceOIL DROPLETS RISING TO A COLLECTION SURFACEtime. Coalescers exhibit a three-step method of operaLtion as depicted in Figure 3.hFIGURE 3THREE STEPS IN COALESCINGSubmicron dropletsflow around target Several captured dropletscoalesce, forming largerdrops.Droplets striketarget and adhere.which trickle downand fall, becomingseparated1) Collection of Individual Droplets2) Combining of Several Small Droplets into Larger Ones3) Rise/Fall of the Enlarged Droplets by Gravitying distance is equal to the entire height of the pool ofliquid present in the separator. This effect is also seenin knitted wire mesh, but their high void fractions meanthe surface is very discontinuous.Meshes, co-knits of wire and yarns; and wire and glasswools all depend primarily on Direct Interceptionwhere a multiplicity of fine wires or filaments collectfine droplets as they travel in the laminar flow streamlines around them (Figure 6). As can be see in Figure4, in general they can capture smaller droplets thanthose that depend on enhanced Stokes Settling. Ageneral rule with Direct Interception is that the size ofthe target should be close to the average sized dropletin the dispersion. Finer coalescing media allow for theLIQUID-LIQUIDCOALESCER DESIGN MANUALTEL: 800-231-0077 FAX: 713-433-6201 WEB: www.amacs.com EMAIL: amacs@amacs.com3

been retained. Whether a coalescer medium ishydrophilic (likes water) or oleophilic (likes oil) dependsDROPLET INTERCEPTIONon the solid/liquid interfacial tension between it and theLiquid FlowStreamlinesDroplet Trajectorydispersed phase. In general an organic disperseddDROPLETphase ‘wets’ organic (that is plastic or polymeric)Filamentd/2Area for efficientmedia, as there is a relatively strong attraction betweendroplet collectiond/2Dthe two, while an aqueous dispersed phase preferably‘wets’ inorganic media, such as metals or glass. ThisDROPLETDropletTrajectorydaids in the coalescence step as the droplets adhere tothe media longer. Also assisting coalescing is the denseparation of finer or more stable emulsions (Table 2). sity of media: lower porosities yield more sites availableNote that fine media will also capture or filter fine solid for coalescing. In the case of yarns and wools, capillaryparticulates from the process stream. Therefore, unless forces are also important for retaining droplets.the emulsion is very clean, an upstream duplex strainerOnce several droplets are collected on a plate, wire, oror filter is needed to protect a high efficiency coalescer.fiber, they will tend to combine in order to minimize theirinterfacial energy. Predicting how rapidly thisMax Droplet Flow Rangewill occur without pilot testing is very difficult toMediaSource2Diameter, µgpm/ftdo. Judgments of the proper volume, andtherefore residence time, in the coalescersSeparators with15-75Corrugated40-1000Coarse EmulsionsSheets(35-180 m3/hr/m2) are guided by experience and the following& Static Mixersproperties:Overhead Drums,7.5-45Wire Mesh, Extraction Columns,20-300(20-110 m3/hr/m2) Coalescing Media:Wire Wool Distillation Tower Feeds,Impeller Mixers Media/Dispersed PhaseSteamStripperBottoms,Interfacial TensionCo-Knits of7.5-45Caustic Wash Drums,10-200Wire &3/hr/m2) Porosity(20-110mHigh Pressure DropPolymerMixing Valves CapillarityHaze from Cooling inGlass Mat,Bulk Liquid Phase,Co-Knits of7.5-45Liquid Phases:1-25Surfactants GivingWire &(20-110 m3/hr/m2)Emulsions with Very Continuous/DispersedFiberglassLow Interfacial TensionInterfacial Tension Continuous/DispersedMediaHydro/Oleophilic Porosity Target Size Fouling/CostDensity DifferenceMetal/PlasticH/O98-99%3/8" - 1"Low/LowSpacing/CrimpsCorrugated Sheets Continuous Phase Viscosity Superficial Velocity.002" - .011"Wire/Plastic MeshH/O95-99%FIGURE 6Wire WoolHCoalescers work better in laminar flow for several reasons. First, as mentioned above,8 - 10 micronHigh/HighWire/FG Co-Knits,H92-96%droplets will stay in the streamlines around aGlass Matwire or fiber target. Second, high fluid velocitiesTable 2overcome surface tension forces and stripdroplets out of the coalescer medium. This results in reStep 2 – Droplet CoalescenceThe second step is to combine, aggregate, or coa- entrainment in co-current flow and prevents dropletslesce captured droplets. Increasing the tendency for from rising/sinking in counter-current flow. Lastly, slowdroplets to adhere to a medium, increases the proba- er velocities result in greater residence time in thebility that subsequent droplets will have the opportuni- media and therefore more time for droplet-to-targetty to strike and coalesce with those that already have impact, droplet-to-droplet collisions, and Intra-MediaStokes Settling.Wire/PolymerCo-KnitsO94-98%21-35 micronLIQUID-LIQUIDCOALESCER DESIGN MANUALTEL: 800-231-0077 FAX: 713-433-6201 WEB: www.amacs.com EMAIL: amacs@amacs.com4

The guidelines in Table 2 are used for selecting theproper coalescer for a given source based on themedia’s Droplet Collection ability. Also given are typicalflow ranges for each type of coalescer media.Step 3 – Stokes Settling With Coalesced DropletsThe third step is the Stokes Settling of the coalesceddroplets downstream of the medium. The degree ofseparation primarily depends upon the geometry of thevessel and its ability to take advantage of the largecoalesced droplets that were created through stepsone and two as described above.taking into account the effects of any particulates orsurfactants present. AMACS has several of these available,both as hand-held batch testers and continuous unitsºgle, double, or triple coalescer stages (Figure 7). Thisallows a coalescer system to be developed that isoptimized for its removal efficiency, on-stream time,and cost effectiveness.FIGURE 7PILOT FILTER AND COALESCERBasis for Sizing and SelectionA preliminary procedure for determining how difficult itis to separate two immiscible liquids involves the performance of a simple field test. A representative sample of the emulsion is taken from a process pipeline orvessel. It is either put it in a graduated cylinder in thelab or, if it is under pressure, in a clear flow-throughsample tube with isolation valves. The time required toobserve a clean break between phases is noted. If thecontinuous phase has a viscosity less than 3 centipoise, then Stokes Law says the following:SeparationTime 1 minute 10 minutesHoursDaysWeeksEmulsionStabilityVery WeakWeakModerateStrongVery StrongDroplet Size,Microns 500100-50040-1001-40 1 (Colloidal)Fortunately, the experienced designer with knowledgeof the application, equipment, and physical propertiescan often estimate the strength of the emulsion anddetermine which medium will be successful. A moredefinitive approach, and one that is often needed toprovide a process warranty, is the use of an on-sitepilot unit.Liquid-liquid coalescer performance is often rated inparts per million of dispersed phase allowable in thecontinuous phase effluent. Even trace amounts of contaminants such as emulsifiers and chemical stabilizerscan have dramatic effects on results at these levels. Ina pilot program, several alternate media are providedto the customer so that their performance can bedocumented on the actual process stream, therebyFor liquid-liquid coalescers, as with any process equipment, successful sizing and selection is always acombination of empirical observation/experience andanalytical modeling. Of the three steps in coalescing –droplet capture, combining of the collected droplets,and gravity separation of the enlarged droplets – thefirst and the last can be modeled with good accuracyand repeatability. The modeling of the middle and theactual coalescing step is a complex function of surfacetension and viscous effects, droplet momentum, andthe dynamics of the sizes of the droplets in the dispersion. This has been done successfully in porousmedia, but is beyond the scope of this brochure.Droplet capture, the first step in liquid-liquid coalescing, is the most important. The next two sectionsdescribe the formulas used for the collection mechanisms of Intra-Media Stokes Settling and DirectInterception.LIQUID-LIQUIDCOALESCER DESIGN MANUALTEL: 800-231-0077 FAX: 713-433-6201 WEB: www.amacs.com EMAIL: amacs@amacs.com5

Intra-Media Stokes SettlingVC (C1) Q h µ( S.G.) d2(2)In a horizontal 3-phase separator, in order for efficient separation to take place, droplets of some min- Whereimum size which exist in both the gas and the liquidVC Coalescer volume, cubic feetphases must be captured within the equipment.When coalescing media is installed in the lower segmentC1 164 for Plate-Pak w/horizontal sheetsof the vessel, the furthest a droplet has to travel is219 for STOKES-PAK w/horizontal sheetsfrom plate to plate or sheet to sheet, rather than312 for STOKES-PAK w/vertical sheetsdown from the liquid level to interface level and/or upQ Liquid/liquid emulsion flow, US GPMfrom the vessel wall to the interface level (dependingwhether the dispersed phase is heavier or lighterh Corrugated plate spacing or structuredthan the continuous phase).packing crimp height, inchesAMACS offers a number of Corrugated Plate Interceptorsd Minimum droplet diameter, microns(CPI) to enhance coalescence, such as Plate-Pak andµ Continuous phase viscosity, centipoiseSTOKES-PAK crimped sheet packing (Figure 8).FIGURE 8COALESCING MEDIA THATDEPENDS ON STOKES SETTLINGPlate-Pak is the most efficient CPI and thus has thesmallest C1. The reason for this is that the height, h, adroplet must traverse before hitting a solid surface isminimized in this construction (see Figure 9 a-c).FIGURE 9Operating by enhanced gravitysettling, Plate-Pak vanesare especially effectivefor removing largerdroplets.DISTANCE BETWEEN PLATES INVARIOUS STOKES-PAK COALESCERS9a Plate-Pak corrugations perpendicular to the flowOil Dropletsh EmulsionClear liquidPlate-Pak Axis of CorrugationAxis of Corrugation9bStokes-Pak withHorizontal Sheetsh1/2"Axis of CorrugationStokes-Pak 9cThey make more efficient use of a vessel volume thana straight PPI (Parallel Plate Interceptor) since moremetal is used and the specific surface area is greater.It can be shown from Equation 1 for Vt that the volumeof media necessary to remove virtually all dropletsequal to a minimum, typically 30-60 microns, is givenby:Stokes-Pak withVertical Sheetsh1/2"LIQUID-LIQUIDCOALESCER DESIGN MANUALTEL: 800-231-0077 FAX: 713-433-6201 WEB: www.amacs.com EMAIL: amacs@amacs.com6

In order to decrease solid retention the axis of the cor- can be found by trial-and-error substitution of the terminalrugations of Plate-Pak should be parallel to the flow. settling velocity from Equation 1 into Equation 3 belowHowever, vessel geometry often necessitates that thes (vt/h)/ (vs/L) .999corrugations be perpendicular to the flow, especially in(3)round vessels. Due to its light, self-supporting struc- wheres Fractional Collection Efficiencyture and ease of installation, the overall project cost isby Stokes Settlingnormally less for STOKES-PAK than Plate-Pak whenηηvs Superficial VelocityL Element Lengthvt/h Droplet Rise Timevs/L Droplet Residence TimeFIGURE 10GAS OUTOIL OUTADJUSTABLEOIL EWASTEWATERINLETIn horizontal flow when this length is over four elements, 32" (813 mm), the coalescer is usually split intwo or more beds with intermediate spacers or spacerrings. Also, cross-flow designs are often used in thissituation to allow for more frequent removal of thecollected dispersed phase.Direct InterceptionSOLIDSDRAINSOLIDSDRAINthey both have sheets in the horizontal. STOKES-PAKwith vertical sheets, on the other hand, retains fewersolids than the horizontal sheet version and so is oftenrequired in fouling situations. In this case, there issome loss in coalescer efficiency due to the longer distance a droplet could travel (see Figure 9 b and c). Theentire CPI unit can also be put on a 45 to 60 angle inorder to retard fouling. However, this requires muchmore support structure and an additional 40 to 100%of coalescer volume since droplet trajectory is lengthened (Figure 10).Direct Interception occurs when a droplet follows astreamline around a target but collides with it becausethe approach distance is less than half its diameter,d/2 (Figure 6). The formulas for Direct Interception inmesh, co-knits, wire and glass wools are given below.Given first is a formula for the collection of a droplet ona single target. Following that is a formula which,based on this factor, calculates the depth of the coalescer element necessary to achieve a desired overallcollection efficiency at a selected minimum dropletsize.(4)Equation 2 incorporates empirical factors that increasethe coalescer design volume over the theoretical inD Collection Efficiency of a Singleorder to compensate for the effects of bypass andTarget by Direct Interceptionback mixing. With knowledge of the cross-sectionalarea of a fully flooded coalescer vessel or the lowersegment available in a horizontal 3-phase separator, E Effective Length Multiplierthe required depth can easily be calculated from Vc.AMACS Plate-Pak and Stokes-Pak both come in units α Volume Fraction of Fibers or Wireswhich are 8" (203 mm) deep as a standard, but customdepths are also available.d Droplet Diameter, inchesηOnce the final coalescer length is selected the minimumdroplet size that can be collected at 99.9% efficiency K Kuwabara’s Hydrodynamic Factor-0.5 ln α -0.25 α 2 α -0.75LIQUID-LIQUIDCOALESCER DESIGN MANUALTEL: 800-231-0077 FAX: 713-433-6201 WEB: www.amacs.com EMAIL: amacs@amacs.com7

The formulas for Direct Interception have no velocityterm in them, but to allow coalescence to take placedesigns are normally done for the middle of the flowranges given in Table 2. K, the KuwabaraHydrodynamic Factor, above is a correction to the collection efficiency term that assumes a laminar/viscousflow field. The effective length multiplier, E, is anempirical factor that takes into account the uneven distribution of curved and crinkled targets in a wool medium and/or the shielding effects of the loops of knittedmesh and twists of adjacent filaments in a strand ofyarn. The idealized layout of fiber targets where E 1 ina coalescer is shown in Figure 11, while what actuallyexists in a co-knit is shown in Figure 12. The finer the filament or wire the more the nesting/shielding effect andthe lower the value of E.FIGURE 11INTERCEPTOR LAYOUT IN AN IDEAL COALESCEROrderedTargetsSFLOWDFIGURE 12(5) Overall Collection Efficiency by Direct InterceptionL Element length required for removal of all droplets a minimum size at a .999, inchesAs can be seen in Figure 4, there are two broad categories of Interceptor-Pak Coalescers that depend inDirect Interception, those that are made with fine wiresand those that are made with fine fibers. The factors toDropletApplication rSheen4.50.037 .04Fiberglass MatFiberglass Co-Knit 8.9/0.00035 0.027 .02Interceptor-PakTMCausticWash Drums11.0Teflon21/0.00083 0.019 ster24/0.00095 0.021 .002WoolInterceptor-PakTM0.028 ptor-PakTM0.014 .60Table 3be used in the formulas above for these media, theappropriate minimum droplet size to use; and theCO-KNIT MESH COALESCER THATapplications where they have found success are givenDEPENDS ON DIRECT INTERCEPTIONin Table 3. In wire-yarn co-knits the wire occupies asmuch as a third of the volume fraction as the yarn, butAs with CPI coalescers, sizing of a liquid-liquid coalescer exhibits only a few percent of the surface area.that operates primarily on Direct Interception also corre- Therefore, for the sake of conservatism, the constantslates well to an Overall Collection Efficiency of 99.9% of given in the table do not take into account either factor.a minimum droplet size. Once this droplet size, empirically found to be approximately half the target diameter, The equations for droplet collection above can also beis substituted into Equation 4, the length, L, required for used to derive the dispersed phase’s concentration inthe effluent stream. First, a measured distribution ora clean break can be predicted as follows.the curve estimated with Mugele’s droplet size distri-LIQUID-LIQUIDCOALESCER DESIGN MANUALTEL: 800-231-0077 FAX: 713-433-6201 WEB: www.amacs.com EMAIL: amacs@amacs.com8

bution equations is broken up into a large number ofdiscrete diameter ranges. The fractional collectionefficiency is then calculated at the mid-point of therange using either equation 3 or 5 (rewritten to beexplicit in ) thereby deriving the volume of dispersedphase that penetrates at that diameter. The effluentcurve is then plotted. The area under both curves isfound with the influent normalized to 1 (Figure 13).With knowledge of the influent dispersed phase concentration, the effluent level is found by multiplying bythe ratio of these areas.Volume Fraction per MicronFIGURE 13Droplet Diameter, micronsGravity Separation Downstreamof a Coalescer ElementSuccessful gr

PHASE IN GAS OUT GAS OUT 20 Ft. Gravity Separator 12 Ft. Coalescer Vessel 16" INTERFACE LEVEL THREE PHASE IN LIQUID LEVEL LIQUID LEVEL 3 0 " 36 " I 60 D " I D LIGHT PHASE OUT LIGHT PHASE OUT HEAVY PHASE OUT Liquid-Liquid Coalescer Design Manual 800-231-0077 14211 Industry Road