Condensate Polishing for Nuclear & Fossil Fuel Power Plant

8/7/2019 Condensate Polishing for Nuclear & Fossil Fuel Power Plant

http://slidepdf.com/reader/full/condensate-polishing-for-nuclear-fossil-fuel-power-plant 1/10

Condensate Polishing For Nuclear AndSuper Critical Power Plants

For The 21 ST Century

AL TAVARES, Graver Technologies, LLC, Glasgow, Delawareand ROBERT A. APPLEGATE, Graver Water Systems, LLC,

Cranford, New Jersey

IWC-08-02

KEYWORDS : condensate polishing, mixed bed, deep bed, precoat filter powdered resin, hollow fiberfiltration, ultra low chloride resin

ABSTRACT : This paper provides a brief history and timeline of condensate polishing from the 1950’sto the present. It then discusses the equipment, designs, process strategies, and operating techniques thatare being employed and developed to address the increasingly stringent requirements of plants in the21st century.

INTRODUCTION AND HISTORICALPERSPECTIVE

Condensate polishing (CP) has a long history incritical, high-pressure power generationapplications beginning with mixed- bed ionexchange demineralizer in the 1950’s andfollowed by the introduction of powdered ionexchange precoat filter demineralizers in the1960’s. The following decades saw variousoperability improvements driven largely byunforeseen operating challenges and improved,more sensitive high purity water analysiscapabilities. In addition, higher-pressure systemdesigns are requiring more stringent CP waterquality that is resulting in innovativetechnologies in order to meet thoserequirements. This, in turn, has also resulted innew and more stringent EPRI and INPOguidelines. Suppliers to the power generationCP market have responded with advances in ionexchange resins, equipment design, andmonitoring systems.

In the framework of a renewed and globalpower generation growth cycle, our paper willpresent, briefly, the historical advances in CPsystems, but more importantly advancedequipment design and operating practices thatare currently available to the market. These CPsystem choices allow a technology to match auser’s specific needs. These technologiesinclude ion exchange technologies to maximizeCP performance and minimize operating costs,combined technology systems utilizing CP filterpolishing and deep bed ion exchange, advancedprecoat systems, and more recently developedhollow fiber filtration. Specific CP technologiescan be chosen based on a plant’s individualrequirements and often go hand in hand withrecently developed operating strategies such asoxygenated treatment (OT) and reducedchemical costs (particularly ETA and regenerantchemicals).

Table 1.0 provides a timeline showing some of the major developments in CP technology since1950.

- 1 -



DEEP BED ION EXCHANGECONDENSATE POLISHERS

Deep bed polishers typically employ a 3 to 4foot bed of mixed ion exchange resins. The ionexchange beds can be designed with variousratios of cation exchange resin and anionexchange resin, the most typical being 2:1 or 1:1cation to anion by volume. The systems aresized hydraulically with the accepted industrystandard rating of 50-gpm/sq. ft and can havedesign pressures in excess of 600 psig. Figure1.0 depicts a typical high-pressure sphericalservice vessel. These vessels are designed towithstand full system differential, and becauseof the high operational flow rates have elaboratedistribution and collection systems to assureuniform flow distribution and to prevent resinmounding or bed gouging. Historically thesepolishers have been regenerated on-site usingregeneration equipment external to the servicevessels. Figure 2.0 is a simplified version of atwo- (2) vessel regeneration system. The systememploys a state-of-the-art bottom resin transfersystem that provides an extremely high degreeof resin separation and is capable of operatingwith variable cation to anion resin ratios withminimal mechanical modifications.

Fossil and Nuclear PWRStandard Gaussian distributed cation and anionexchange resins were used in condensatepolishing deep beds for many years.Condensate grades were introduced in the early1970’s with reduced fine particle content.These cation and anion exchange resins were

somewhat easier to separate, resulting in lesscross contamination. Cation fines that remainwith the anion during separation are convertedto the sodium form during sodium hydroxideregeneration and can lead to poor performance.Inert resins, with clearly differentiable color anda density between the cation and anioncomponents, were sometimes added to themixed beds to act as a buffer layer and facilitateeven better separation, further reducing crosscontamination. However the inert resins did notalways perform consistently and in some casesbecame fouled, negating any further benefits.Uniform particle size resins were introduced inthe 1980’s as the best separating condensatepolishing resins. The particle size and densitydifferences enhance separation and subsequentregeneration quality. The uniform size allowsfor good flow characteristics and the lack of fineparticles can mean lower differential pressure.These resins have a much narrower particle sizedistribution than prior grades used forcondensate polishing and have become thevirtual standard for condensate polishingsystems.

Nuclear BWRAlthough early BWR deep bed plants includedregeneration systems, none are operable todayand only a few were ever operated. Some plantsused ultrasonic resin cleaners to remove theparticulate iron oxides and then return the resinsto the polisher vessels. It was preferred to usemixed bed condensate polishing resins that werenot easily separable. Gaussian distributed resins

- 2 -



worked well for years. To improve the ironfiltration, a new dual-morphology cation wasintroduced in the 1990’s. This “CRUDremoval” resin was installed in the condensatepolishing systems at a majority of the US BWRplants. Iron removal results were excellent inall cases but elevated sulfate levels wereobserved in most of the systems after about nineto twelve months of operation. Newopportunities were presented as iron removalremained a priority. Uniform size resin was thenext step although there was some reluctance toaccept the resins because of their ease of separation. In the BWR non-regenerable

systems, it is desirable to keep the mixed bedsintact. Smaller uniform particle size cationswere developed to make mixed beds with lessseparable tendencies.

The process selection conditions that favordeep beds include :• High TDS cooling water such as sea water

or high TDS cooling towers• The requirement to operate long term with

small condenser leaks• The inability to control excessive air in-

leakage

Figure 1.0

B E D

D E P T H

RESIN BAFFLE

OUTLET

RESIN VOLUME

RESIN TRANSFER INLETINLET

INERT FILL

BAFFLE

INLET DISTRIBUTORHIGH RATE

DIFFUSION RING

OUTLETSRESIN TRANSFER

UNDERDRAINHEADER LATERAL

WITH HIGH PRESSURE UNDERDRAINTYPICAL SERVICE VESSEL

- 3 -



Figure 2.0:SepraEight™ Two Vessel Bottom Transfer Regeneration System

CATION TRANSFER LINE

ANIONTRANSFER

LINE

XE

XT

DRAIN RESINOUTLET

RESININLET

BACKWASHINLET

DRAIN

Separation/Anion Regen. Cation Regen/Mix & Hold

21 st CENTURY DEEP BEDIMPROVEMENTS

• High capacity strong acid cation (SAC)exchange resins with higher than typicalcrosslinking have been evaluated in bothBWR and PWR nuclear plants. Thesehigher capacity resins can provide longerrun lengths to the amine break in PWRand fossil condensate service. However,results were economically neutral whenconsidering run length vs. resin costs.Other PWR plants evaluated the cationsto reduce the phenomenon characterizedas anion kinetic impairment. Onethought is that the solvent properties of ethanolamine (ETA) accelerate cationleachables, which in turn begin to foulthe anion resin, causing the poor kineticperformance. Many have returned to thetraditional UPS cations for condensatepolishing.

The high crosslinked cations are alsobeing used in BWR condensates inattempts to reduce the system sulfates.Small pieces of polystyrene sulfonateoligomers diffuse out of cation exchangeresins over time, and form sulfates in the

reactor. The higher crosslinked cationsare better able to tie up these oligomers.However, all cation resins undergo someamount of polymer degradation ordesulfonation over time. The highcapacity cations have a greater numberof sulfonated sites providing morepotential for eventual degradation.Improved sulfate levels have been

experienced when the cation resins areprocessed by the supplier prior toshipment.

More recent specifications for TOC andsulfate extractables have been driven tosome degree by improved analyticalcapabilities and the ability to correlatethese to operating issues. Specificationsmay apply to both cation and anionexchange resins. As-manufactured base

resins are not made specifically to meetthese requirements, nor are thoseproperties routinely evaluated. Somespecifications approach the extent of electronics grade resins regardingleachable TOC and other contaminantlevels. To achieve these new morestringent quality requirements oftenrequires post manufacturing processing.

- 4 -



Some ion exchange resin suppliers arewell qualified to perform the significantlevel of processing required. Multiplesteps can include special cleaningtechniques and UPW rinsing.Coordination of processing, shipping,and use between the customer andsupplier can be key to achieving the bestperformance.

• A major advantage of deep bedcondensate polishers is the high ionexchange capacity. However, they dohave limited CRUD filtrationcapabilities. This is driving advances inpre filtration to the ion exchange deepbeds. When combining technologies,CRUD removal is done prior to the deepbeds using high flux rate filtration. Thisprocess step can consist of backwashablepleated elements, high flow ratedisposable pleated elements, orpowdered resin precoated elements.These technologies can be used eitherduring the entire cycle, at start-up, or asneeded at other times such as condenserleaks with high suspended solids.

These deep bed pre-filters were alsoconsidered for plants with and withoutprecoat filter demineralizers alreadyinstalled. Various types of pleated filtershave been designed and installed as newequipment, or in some cases in theexisting filter demineralizer vessels. Thepleated filter elements containsignificant surface areas of small micronsize media. Ten, five, and one micronsizes are the most frequently used.Some are specially designed as precoatsepta to gain the additional advantagesof powdered resin use. A more recentdesign employs a comparatively largediameter disposable, non-precoat,pleated filter used in a horizontalorientation.

• Highly regenerated ion exchange resinsreduce ionic leakage to lower levels,

extend the operating cycle, reduceregeneration periods, and extendoperating life. This includes thecomplementary use of amine form cationresin with an ultra-low-chloride anion toachieve operational improvements.Several benefits have been realized andsavings more than cover the cost of thespecialty, value-added premium anionexchange resin. One goal was to reduceiron transport to a PWR’s steamgenerators by increasing the aminecontent, which raises pH and driveschloride off even a typical nuclear gradelow chloride anion. A newly developedspecialty regeneration process has beenused to prepare the highly converted,ultra low chloride anion resin. This resinmaintained low steam generatorchlorides as the increased amine contentreduced iron transport. The amine formcation and ultra-low-chloride anionresins have been operating for more thanfour years without regeneration and areexpected to exceed five years.Economic benefits are realized due to asignificant reduction in use of ETA andregenerant chemical savings for fouryears worth of regenerations. Thisspecial anion may also be used in BWR,neutral pH condensate and along with acomplementary highly regenerated,very-low-sulfate specialty cation, nowunder development, may providesuperior performance in this non-regenerable application.

• Off site regeneration can eliminate onsite handling of regenerant chemicals.More recently the use of off-siteregeneration service providers hasbecome an option for plants operatingdeep bed condensate systems. Atpresent, experience has been limited, andcosts can be even higher thanregenerating on-site when all costs arecalculated. In addition to regenerantchemicals, these costs can includetransportation, facility leasing or

- 5 -



overhead, licensing site for nuclear plantresins, insurance, and others. Costs arelikely to be relatively high because thevolume of resin to be regeneratedannually may not be sufficient todevelop a regeneration facility unlessmultiple plants are serviced from thesame facility. However, the higher costmust be weighed against potentialbenefits derived by plant operationimprovement, time saved, and theelimination of on-site chemical handlingand storage. Perhaps plants interested indiscontinuing their on site regenerationsshould consider the non-regenerablespecialty resin option to determinewhether it could be their best option.

PRECOAT FILTER DEMINERALIZERSPrecoat filter demineralizers combine superiorfiltration, compared to deep bed polishers, withkinetically superior ion exchange by usingfinely ground, highly regenerated resinsprecoated onto septa in pressure vessels that arespecially designed to provide uniform precoatapplication and process flow distribution.Systems of this type have been in use since theearly 1960’s in a variety of applicationsincluding condensate polishing in BWR, PWR,Fossil, and industrial plants; reactor watercleanup, fuel pool treatment, and radwasteprocessing. The first generation of precoatsystems included the use of individual ionicform powdered ion exchange resins.Technology improvements have includedpremixed resins in various ratios of cation toanion (in several ionic forms for the cation), andsometimes combined with fibers that enhancedphysical filtration and produce extended runlengths.

Precoat filter demineralizer technology has alsoadvanced with a number of equipment andoperating strategies that have improved thesesystems. These include advanced precoattechniques, high energy air surge backwashingin order to effectively clean septa of spentprecoats and crud, internal distribution tubeadditions to ensure more uniform flow patterns

and even distribution of precoat material alongthe entire surface of each septum, as well asstate of the art process controls to optimizesystem performance. In many cases oldersystems have been improved by incorporatingthese upgrades along with evolved septa andhardware improvements that enhance filtrationperformance and system maintenance.

Figure 3.0 is a schematic diagram of a precoatsystem with an enhanced flow distribution tube,air surge type backwash, and advanced precoatsystem. During the service mode condensateenters the service vessel at the center of thebottom head and is directed to the chamberabove the tube sheet. At this point, some of thecondensate enters the filter chamber below thedistribution tube while the balance enters abovethe distribution tube. This split assures that thecondensate is distributed equally to the top andbottom of the filter elements and avoidslocalized high velocities that could disruptproper distribution of the precoat material.Treated condensate passes through the filtersepta and proceeds through the tube sheet to thevessel outlet.

A major advantage of precoat filterdemineralizers is their highly effective filtrationcapability. Over the years there has beenconsiderable advancement in septa technologyincluding improved yarn types in wound septa,absolute rated multi-media pleated filters and acombined technology of pleated filters with aprecoatable surface. Other specialty filtrationmedia exist including stainless steel wedge wireelements, carbon fiber wound septa for hightemperature applications, and spun bond filtersthat appeal to the cost conscious user but haveshown limited life when compared to othertypes.

Precoat filter demineralizers use a relativelysmall amount of ion exchange resin in theprecoat, and therefore have less total ionexchange capacity than deep beds. Howevermany of the disadvantages of bead resins,including regenerations, are avoided. In processconditions that favor precoat demineralizers

- 6 -



they have proven to be highly flexible andeffective condensate polishing systems.

• Need for frequent start ups and restarts(precoat filters provide superior crudremoval and offer the most cost efficientmeans for controlling corrosion transport)The process selection conditions that favor

precoat filter demineralizers include : • Need for rapid start up• Need to minimize pressure drop

requirements for condensate polishing• Low TDS cooling water • Operating policy of orderly shut down andrepair in the event of a condenser leak

• High temperature condensate as typicallyseen in air cooled condenser plants

• Need to minimize costs includingequipment, installation, and operating costs.

• Titanium condensers, with welded tubesheets.

• Limited operator availability and training(regeneration of deep beds is a complexregeneration sequence and involves the useof strong acid and base)

• Desire to avoid handling and neutralizinglarge quantities of acid and base

• Limited space availability. The footprint of aprecoat dimineralizer can be as much as50% of a deep bed installation.

Figure 3.0: Precoat Filter Demineralizer System with Advanced Precoat Feature

FEED IN

SUMP

DRAINPRECOAT INLET

HOLD PUMP

VENT FEED OUT

AIRSURGESUPPLY

SUPPLYWATER

SLURRYTANK

SERVICE VESSEL

ELEMENTS

DISTRIBUTIONTUBE

TANKAUXILIARY

PUMP

RECYCLEPRECOAT

PRECOATINJECTION

PUMP

SUMP

SUMP

WATERSUPPLY

SUMP

FEED IN

SUMP

DRAINPRECOAT INLET

HOLD PUMP

VENT FEED OUT

AIRSURGESUPPLY

SUPPLYWATER

SLURRYTANK

SERVICE VESSEL

ELEMENTS

DISTRIBUTIONTUBE

TANKAUXILIARY

PUMP

RECYCLEPRECOAT

PRECOATINJECTION

PUMP

SUMP

SUMP

WATERSUPPLY

SUMP

- 7 -



21 st Century Precoat Filter DemineralizerImprovements

• Increased operating run lengths from oneor more new equipment designupgrades: advanced precoat systems,distribution tubes, and state-of-the-artprogrammable logic control systems

• Precoat formulations designed to extendrun lengths improving direct operatingcosts and reducing associated costsincluding the disposal of radwastes.Longer runlengths reduce the totalquantity of powdered precoat usagebenefitting any precoat filterdemineralizer operation by reducingoperating costs. For BWR plants, thevolume of radwaste is reduced. This issignificant because in most cases theradwaste treatment cost can exceed thepurchase price.

• Fine filtration precoats for BWRparticulates. Specialty precoat productsare designed with improved filtrationcharacteristics in order to remove veryfine iron and copper particulates in orderto meet reactor contaminantconcentration guidelines.

• Layered precoat systems to targetspecific treatment goals such as solublecopper. One example is a specialtyprecoat product containing powderedweakly acidic cation that is precoatedover a fiber containing mixed bedprecoat to selectively remove copper infossil condensate during startups.

• Development of extremely highlyregenerated powdered ion exchangeresins that combine the kineticsuperiority of powdered resins with ultralow ionic contamination values thatachieve low ppt levels of ionicleachables. Powdered resin analogs of ultra-low-chloride anion and thecomplementary very low leachablesulfate cation bead resins describedearlier. High purity resins for powderedproducts are perhaps even more criticalthan for bead resins because the internalstructure is exposed during the grinding

process. Any contaminants that woulddiffuse out of the bead over time areimmediately released from a powder.

• Advanced generation, dual functioningprecoatable septa that combine absolutefilter ratings with excellent precoatablecharacteristics and backwash efficiencythat extend septa life.

HOLLOW FIBER FILTRATION SYSTEMSHollow fine fiber membrane filtration (HFF) isa filtration only technology. The use of thistechnology in condensate filtration wasdeveloped in Japan and it is used extensively inthe Japanese nuclear industry. It also haslimited application in Japanese fossil units.The first system of this type in North Americawas installed and has been operating for justabout one year.

The filtering fibers are made of polysulfonewith 1.0 mm cross-section and a 0.1 μ m poresize (Figure 4.0). The fibers are potted inmodules (Figures 5.0 and 6.0). The modules arethen installed in tube sheets (Figure 7.0) withinpressure vessels. Figure 8.0 shows aninstallation in Japan.

The condensate feed is introduced to the outsideof the fiber, flowing through the membrane intothe hollow center. Filtered condensate thenflows to collection through the fiber lumen.This product is very effective for the removal of corrosion products down to <1 microgram/literfor particles greater than 0.1 micron.

The cost for an HFF system is significantlymore than that of a pleated membrane filtrationsystem. However, the membrane life has beenshown to be significantly longer and can greatlyreduce element replacement and disposal costsover the life of the system. Elements have beenin service in Japan in excess of 10 years withoutneeding replacement.

SUMMARYFor more than 50 years, condensate polishingsystems have been used to treat the water in the

- 8 -



power generation cycle. While the principle hasremained the same, requirements have becomemore stringent and the condensate polishingsystems and products have evolved through the

joint efforts of the power industry and thededicated suppliers who serve the needs of thatindustry. Countless advancements in equipmentdesign, product development, and operationaltechniques have been introduced throughoutthese 50+ years of condensate polishing, manymore than could be described in this paper.

Today more than ever, continued advancementsand new technologies in condensate polishingare critical to improving plant operations inorder to help produce a viable supply of powerto meet the expanding needs of this country andthe world. We cannot rest after 50 years of improvement, but must guide the way for thenext 50 years of technology development.

At the beginning of the 21 st century we arealready viewing some results of continuingdevelopment through:

• Application of extensive resin cleaningto achieve the lowest possible TOCleachables.

• Coordination of the processing,shipping, loading, and use of condensatepolishing resins.

• Ability to run condensate polisherswithout regenerations by using ultra-low-chloride anion in conjunction withoperational changes, achieving bothregenerant and cycle chemical savings.

• Considering an off-site regenerationoption when it makes sense.

• Installation of dual function prefiltersthat operate effectively with precoatswhen necessary, and without whendesired.

• Installation of new systems containinghigh flow rate, disposable prefilters.

• Use of custom powdered products andprecoat combinations to achieve specifictargeted results.

• Evaluation of low leachable sulfateproducts as they become available.

• Installation of parts with designimprovements and control systemupgrades to maximize performance of condensate polishing systems.

These are a few of the advancements already inprogress, being considered, or underdevelopment for the continuous evolution of condensate polishing applications. What willthe next steps be?

Figure 4.0:

Figure 5.0:

Figure 6.0:

- 9 -



Figure 7.0: Figure 8.0:

- 10 -

Documents

Condensate Polishing for Nuclear & Fossil Fuel Power Plant