FRO301238

NEW ION EXCHANGE RESIN DESIGNSAND REGENERATION PROCEDURES YIELD IMPROVED PERFORMANCE

FOR VARIOUS CONDENSATE POLISHING APPLICATIONS

STEPHEN W NAJMY, The Dow Chemical Company, Midland, MI

INTRODUCTIONCondensate polishing is an application that has seen a variety of developments over the past five years withrespect to new ion exchange resin products and alternate procedures for installing, regenerating and re-mixingthe mixed resin products. This paper describes experiences at several power stations that continue to benefitfrom the use of either a new resin product design or a modified resin handling procedure. These successstories come from1) Two BWR nuclear power stations seeking to achieve improved feedwater iron and reactor water sulfate

control, and2) One PWR nuclear power station looking to extend service cycle run length and better manage steam

generator chloride excursions due to salt water condenser inleakage

NEW LOW CROSS-LINKED CATION EXCHANGE RESIN FOR ENHANCED CRUD REMOVAL IN A BOILING WATER REACTOR

(BWR) NUCLEAR POWER PLANT

In a BWR nuclear power plant, excessive feedwater iron can cause two major concerns:• Heat transfer reduction - Iron buildup on the cladding of the nuclear fuel reduces heat transfer efficiency

and can lead to nuclear fuel cladding failures.• Shutdown dose rates - Non-optimal feedwater iron concentrations affect the transport of activated corrosion

products, causing an increase in radiation dose rate on the reactor piping and increasing worker exposure.

While some BWR stations have the benefit of an upstream pre-filter system, other stations rely on their deep-bedcondensatepolishersforcrudfiltration. SuchisthecaseatNiagaraMohawk'sNineMileUnitTwoNuclearPowerStation-aBWRplantincentralNewYork. Inthemid-1990'sitbecameevidentthatnewtechnologywas necessary for this plant to meet stringent industry guidelines for feedwater iron. However, the problem wasfurther compounded when this plant experienced an increase in condensate iron due to the loss of one of its'moisture separator/reheaters (MSRs) at the end of one of its operating cycles in 1971.

An opportunity for low cross-linked cation exchange resinsIn the early 1990's several manufacturers were producing commercial quantities of cation exchange resinsusing styrene-divinylbenzene (DVB) copolymers with 6% cross-linking. Previous work2in Japan suggested thepotential for enhanced crud removal with these products. This spawned interest in the U.S. and soonafterwards a pilot-scale study began at the Public Service Electric & Gas Hope Creek Station along with full bedtrials at several other BWR stationS3. It was during these full bed trials that stations reported increased reactorwater sulfate levels due to the installation of these low cross-linked cation resin products. By 1994, all BWRplants, with the exception of Commonwealth Edison's Dresden Station4, removed the so-called 'crud' resin fromtheir condensate polisher vessels. The explanation for the increased sulfate was simply the lower cross-linkedcation resin having inadequate thermal stability compared to the conventional 10% cross-linked styrene-DVBcation resin. So, although feed water iron reduction was an important goal, it could not come at the expense ofincreased reactor water sulfate. An increased level of sulfate will accelerate corrosion in nuclear fuel rods andother internal components, which can greatly reduce power generation economics.

Challenge: Meeting feedwater iron guidelines without spiking sulfatesTo meet guidelines set by the Institute of Nuclear Power Operations (INPO) and the Electric Power ResearchInstitute (EPRI), the Nine Mile plant needed to reduce feedwater iron from 2-3 ppb with MSRs in service, or 5ppb with MSRs out of service, to + 0.5 ppb. After researching the data5from a previous EPRI collaborationproject evaluating low cross-linked resin, Nine Mile concluded that a cation resin developed by researchers from

The Dow Chemical Company was a product worthy of study. The essence of the Company's invention 6,7 was aproprietary new technology for stabilizing low cross-linked resin.

* Trademark of The Dow Chemical Company

That resin, Dowex* Guardian* CR-1 cation resin, has a low cross-linked copolymer backbone that offersenhanced particulate iron and copper removal compared to conventional cation exchange resins. However, ithas differentiated performance versus other low cross-linked resins because:

• the resin is produced using a unique stabilization technology to reduce the release rate of organicsulfonates"

• the resin combines a uniform particle size distribution with a small average bead diameter (approx. 550microns) and a lower true hydrated density to minimize the separation of the cation/anion resin duringtransfer back to the service vessel

To first qualify the new cation resin product engineers from Nine Mile devised a series of laboratory tests toassess the risk of sulfate leakage in the presence and absence of iron and copper impurities. This stud Y5

included an assessment for the impact on the surface kinetic properties of the anion resin. The Company'sconventional 1 0% cross-linked cation resin was used as the control system in these lab tests. Encouraged bythe lab results and the evaluation of the previous testing of low cross-linked resin, Nine Mile engineers incollaboration with EPRI proceeded to a full-scale plant trial.

In January 1998 the plant converted one of its 1 0 conventional deep-bed condensate polishing resin beds toGuardian CR-1 resin, rounding out the mixed-bed system with Dowex Monosphere* 550A Anion ExchangeResin. In July 1998 a second resin bed was converted to the Dowex Guardian CR-1/Monosphere 550Amixture.

Plant trial shows iron level reduction with stable sulfate levels

The plant's Chemistry Department sampled and analyzed the condensate demineralizer influent (CDI), polisherbed effluent, and final feedwater streams. The reactor water coolant was also monitored for sulfate levels. Theplant completed six, 30- to 40-day service cycles (January 1998 to April 1999) with the low cross-linked beds,comparing their results to the performance of the control bed (a new charge of Dowex HGR-W2/Dowex SBR-C).

During the first two cycles (with MSRs out of service), effluent iron averaged about 4 ppb (an 85 percentremoval rate) from the new resin bed versus about 8 ppb (a 60 percent removal rate) from the control bed. Afterthe MSRs were returned to service (cycles 36), the effluent iron averaged about ppb (a 97 percent removalrate) from the new resin bed, versus about 4.5 ppb (an 80 percent removal rate) from the control bed. Thesedata are depicted in Figure .

The impact on reactor water sulfate is shown in Figure 2. This plot includes the seasonal variation in the hotwellcondensate temperature. As shown in Figure 2, through the hot summer months sulfate levels in the reactorwater stayed near 1.5 ppb. In addition, after ultrasonic cleaning, sulfate spikes were not observed when thecleaned new resin beds were placed back into service. The plant's full-scale trial confirmed that the new resinprovided improved iron removal without increasing reactor water sulfates.

Successful trial leads to optimized-ratio of new resin and conventional cation resins

Based on the favorable iron and sulfate results, the Nine Mile plant converted two more condensate polishingbeds to this new resin pair during the remainder of Operating Cycle #7, leaving the other six beds operating withthe conventional 0%-XL gel cation resin. This allowed the plant to further reduce its iron level to about 1.0 ppb- well within the industry guidelines - and still maintain plant economics. To finalize the optimization, theoriginal two low cross-linked resin beds were replaced (typically after about 13,000 hrs of service) with the newresin. Plus, an additional bed was converted during a refueling outage in the spring season of 2000. The plantalso reduced its resin beds from ten to nine so that each bed has a dedicated vessel to prevent cross-

contamination.

The plant reports9 that the early results (after about 30 days) of Operating Cycle #8 with five low cross-linked

resin beds and four conventional resin beds have been exceptional with an integrated weekly iron average of

0.6 ppb and sulfate concentration less than 1.5 ppb.

2

18

16 m Control Resin (charge 2) 0

o New Resin (charge 9)

14 x New Resin (charge )

12

CL

a loLi_ AP0 ML

8

06 - 0 0

00 CD X %4 _E % 11PRO

OWD 0 % VW 0 X

00 X X

2 0 X X A

CP 00 0 0 X 0 X X0 got Qn� 0 X )N-Axv,0 .94.0 %��

lb 16 10,Fb cb �p ocb

Nq N:6 Nc� q Cb CZ;

1.,5", 1114r"

Figure 1. Effluent Iron from New Resin Beds Vs Control Resin Bed Nine Mile Station Unit 2

m ReactorWaterSulfate Hotwell Temperature eg-F) - % Reactor Power

40 - 160icate when DOWEX GUARDIAN resin beds

back into service after backwashing. 140

30 A; 120

KIND.: .100

virrrv0 20 - 80

L)

60

10 40

2006 M

ILL 00 lb (b

\, Nn�p §P C,

Figure 2. Effect on Reactor Water Sulfate during Trials with New Low Cross-linked Cation Resin Nine MileStation Unit 2

NEW LESS-SEPARABLE MIXED RESIN HELPS TO REDUCE REACTOR WATER SULFATE IN A BOILING WATER REACTOR (BWR)NUCLEAR POWER PLANT THAT DOES NOT HAVE THE BENEFIT OF OPERATING UNDISTURBED CONDENSATE POLISHER BEDS

The use of a deep-bed condensate polisher in a BWR primary cycle is to help maintain reactor water chemistrycontrol. In this application the mixed resin bed functions to remove both ionic and particulate species from thecondensate return stream. This minimizes the accumulation of insoluble iron (crud) deposits and contaminant

3

ions, such as sodium, chloride and sulfate - which together create conditions conducive to inter-granular stresscorrosion cracking (IGSCC) of BWR pressure vessels and reactor internals.

Identifying the major source of reactor water sulfate

In spite of the best efforts to optimize the performance of the deep-bed condensate polishers, meeting the EPRIguideline for final feed water iron and the operational target for reactor water sulfate does not come easy.Reactor water sulfate control and minimization continues to challenge many BWR stations. To this day, stationsstill report's contributions to reactor water sulfate from one or more of the following sources:• ionic sulfate not removed by the condensate polisher resin during conditions of condenser inlealkage• ionic and organically-bound sulfate (post-UV) in the effluent streams of the condensate polisher and reactor

water clean-up vessels• cation exchange resin fines that are sufficiently small to pass through the screened under-drain laterals of

the condensate polisher vessels

So a key part of determining the most cost-effective solution is to understand which of these sulfate sources iscontrolling reactor water sulfate levels. For most situations the answer is the presence of organically boundsulfur in the polisher effluent stream. Clearly the cation resin can be a source of sulfur due to the resins'compositional structure and its thermal stability properties. However, the anion exchange resin readily adsorbssulfate and low molecular weight organic sulfonates resulting from the cation resins' natural decompositionmechanisms. But since cation resins are denser than anion resins, it seems logical that during the transfer ofthe mixed resins back to the service vessel a cation resin 'heel' can form at the bottom of the vessel.Consequently, sulfate species that leach from the cation resin will carry down stream and concentrate in thereactor water system. Having an underlay of anion resin is one approach to reduce this source of sulfate, aswas demonstrated" at Pennsylvania Power & Light's Susquehanna Station.

Operating with undisturbed resin beds

Many plants have opted to retrofit their condensate polishers systems with non-precoat filters. These filtersremove insoluble iron and are installed upstream of the mixed resin beds. This eliminates the need to transferthe mixed resin beds to an ultrasonic resin cleaning (URC) system since crud accumulation on the resin beds isreduced to negligible quantities. In this fashion the station can choose to operate the resin beds in anundisturbed mode. Today, many BWR stations are operating with undisturbed resin beds that were initiallyloaded with an underlay of anion resin.

An opportunity for a new, less-separable mixed resin combination

For BWR stations that cannot justify the expense (estimated between $4 and $1 0 million dollars) to retrofit theircondensate polisher system with a non-precoat filter system, the task of controlling feed water iron and reactorwatersulfateseemsevenmoredifficult. SuchisthecaseattheAmergenEnergyOysterCreekStationinsouthern New Jersey. During Cycle 16 (November 1996 - September 1998), the station experienced 12

abnormally high levels of elevated reactor water sulfate and the root cause was linked to residual cation resinremaining on the under-drain laterals following the resin transfer to the JRC (Japanese Resin Cleaner) system.At the time, all but one of the condensate polisher vessels contained a 11 equivalent ratio of Dowex HGR-W2and SBR-C resins. The other vessel contained a new, less-separable combination of Dowex resin in theprocess of a field trial evaluation.

In the next refueling outage (17R) the under-drain laterals were replaced with a new higher strength design toalleviate pressure drop issues and improve resin transfer capability. In addition, the new Dowex less-separableresins were loaded in all vessels. This new resin mixture was selected because of performance advantagesdemonstrated in a full-bed field trial in the previous operating cycle. Specifically,1. Reducedseparabilitycomparedtostandardresinmixtures(lesstendencyforresinstoclassifytranslatingto

better sulfate control).2. Smaller cation resin bead diameter (approx. 500 microns) with uniform size distribution (results in a lower

void fraction of the resin mixture and provides better crud filtration).

The differential pressure across the condensate polisher system is higher with this new resin pair (see Figure 3)but this was accommodated with the newly installed under-drain laterals.

4

30

25 IN_ -- NOMMENE.O.NMENEENEL

20

1 5 -- E)MEMEMEco ......... (:IEEM .............. I.E.0C 1 0 ME an MW IN ME IN ME ME 0 ME AN W ME M I

5

0 ------- -

75 85 95 105 115 125

Temperature (deg F)

NewResin----0---HGR-W2/SBR-C-N ON IncreaseinTotalBedDPI

Figure 3 Clean Resin Bed Differential Pressure Increase at Oyster Creek 13

New resin provides improved reactor water sulfate and feedwater iron control 13

During the next operating cycle (Cycle 17) the reactor water chemistry was dramatically improved. Figure 4shows a plot of the reactor water sulfate and the reactor water clean-up (RWCU) flow rate following Cycle 17startup.

50 - 800

MEMO 0 700

40 -600

500 E30 - CL

MEN NEWS_Effi ME-00 No 'won 0 moon 400

LL

CO 20 300

200

10 AL

+ 100

AAI,AAAJL 2A AA A'k A0 A A A AAAA AAAA �kAk" AA, A AAAAA AAAjAA'AA 0

11/21/1998 12/05/1998 12/19/1998 01/0211999 01/16/1999 01/30/1999

1--A--Sulfate III CUS Flowl

Figure 4. Reactor water sulfate following Cycle 17 startup - Oyster Creek Station 13

The startup plan called for two-pump RWCU operation at approximately 700 gpm until control could bemaintained below EPRI Action Level (5 ppb) with one pump. An outage of the RWCU system from 12/09/98 -12/11/98 caused a sulfate increase, but the system was restored prior to exceeding reactor water sulfate of 20ppb. The plant went to one-pump RWCU operation on 12/13/98, and sulfate continued to trend down to <1 ppb.

5

Based on these results, the Oyster Creek goal of controlling reactor water sulfate with one-pump IRWCUoperation was met.

Iron control results following the Cycle 17 startup are shown in Figure 5. After some initial results in the 4 -ppb ranges, iron was down to 25 ppb on 01/20/99 prior to the loss of the moisture separator reheater. Theimpact of reheat loss can be seen in the 26 - 30 ppb CDI iron concentrations measured in the latter part ofJanuary and early February 1999. The reheater problems were repaired in February.

The process of cleaning all of the resin beds takes time and by late-February feedwater iron resumed adownward trend. This downward trend with operating time was expected at Oyster Creek where iron removalwith standard resins has been observed to improve with resin age. After mid-May 1999, feedwater iron wascontrolled in the 07 - 2 ppb range.

100

10

0S

0.112/31/1998 01/30/1999 03/01/1999 03131/1999 04/30/1999 05/30/1999 06/2911999

I- �> feed Water -&-influent Fe Ins --<�--Effluent Fe Ins

Figure 5. Iron control following Cycle 17 startup - Oyster Creek Station 13

Successful trial leads to installation of a new, high-capacity, less-separable resin13

Based on the favorable results for reactor water sulfate and feed water iron control, the Oyster Creek plantcontinued to operate with one RWCU pump in service at 400 gpm. However, in late-Spring 1999, the plantbegan a battle with a small, continuous condenser leak. And in November 1999 a leak rate occurred that wasestimated to be approximately 3.5 gpm. The resin beds maintained reactor water chemistry under good controlduring and immediately following the large leak; however, calculations suggested that the condensate polisherresin had over 20% of the original capacity exhausted due to all the exposure to salt water inleakage. So it wasdecided to begin replacing all the resin beds in December 1999.

All subsequent replacements were with a new, high-capacity, less-separable resin mixture with a cationexchange resin that had recently been developed by research scientists at Dow Chemical. Oyster Creekdecided to install this new cation resin. The main reasons for this decision include:• the new resin has a bead diameter that is 50 microns larger than the previous cation resin but still much

smaller than the conventionally sized 1 0% cross-linked resin• the new resin has 12% cross-linkage compared to 1 0% cross-linkage for the previous cation resin

6

The performance expectations for the new 12% cross-linked cation resin were improved resin stability, reducingorganic sulfur release, along with somewhat improved selectivity for sodium, in addition to enhanced removal ofinsoluble iron.

Meeting the standards and goals set by Exelon 17

Amergen ownership includes Exelon and British Energy, so consequently, Exelon's standards and goals areapplied to Oyster Creek. The chemistry goals are tied to the Chemistry Performance Indicator (CPI) which iscalculated from an equation with terms for reactor water quality and feed water iron concentrations. OysterCreek was challenged to lower its' CPI value to less than 1. 1 and to achieve this goal Oyster Creek needed tocontrol feed water iron below 21 ppb and reactor water sulfate and chloride below 20 ppb.Oyster Creek has met the reactor water sulfate and chloride goals each month so far in 2001. And feed wateriron has been less than 21 ppb for most months in 2001.

Today, Oyster Creek operates with five beds of the new, less-separable mixed resin and 2 beds of theconventional mixed resin. As a result, year-to-date average CPI through September 2001 is 107.

TECHNIQUES TO ENHANCE MIXED RESIN PERFORMANCE INA PRESSURIZED WATER REACTOR (PWR) NUCLEAR POWER

PLANT

The use of mixed resin in a deep-bed condensate polisher in the secondary cycle of a PWR nuclear power plantserves to filter insoluble iron oxides and protect the downstream steam generation equipment in the event ofcondenser inleakage. Minimization of insoluble iron and dissolved contaminant ions in the feedwater is criticalto protect the steam generator tubes from a variety of corrosion related mechanisms. These mechanisms aretied to such problems as tube denting and pitting, inter-granular attack (IGA) and inter-granular stress corrosioncracking GSCC). Major changes and improvements to the operation and monitoring of the secondary cycle toeffect these corrosion mechanisms include, but not limited to:• replacing ammonia with ethanolamine as the controlling agent for secondary cycle pH to reduce iron

transport rates• installing on-line ion chromatography equipment on the condensate polisher effluent lines to better

understand the mechanisms that control contaminant ion leakage (sodium, chloride and sulfate) from thecondensate polisher resin

PWR plants continue to look for ways to reduce operational costs and improve effluent water quality from theircondensate polishing systems. Interestingly, some recent discoveries have emerged with regards to increasingservice cycle run lengths without compromising effluent water quality, and maintaining steam generatorchemistry under high TDS condensate conditions (i.e., condenser inleakage). The source of these new findingscomes from the Dominion Engineering Millstone Station - a PWR nuclear power plant in Connecticut that usesDowex Monosphere 650C and 550A resins. The first published report 14 was last year and since that time atleast one other PWR plant15 has benefited from this work.

Meeting the secondary cycle steam generator chemistry targets

Table provides a summary of the EPRI guidelines of the major contaminant ions in a secondary cycle steamgenerator. Recognize that plant chemists are striving for much lower concentration levels of all contaminant

ions compared to the maximum allowable levels to satisfy Action Level One status.

Table 1. Contaminant Ion Criteria in Secondary Cycle PWR Steam Generator Systems

I Parameter I I Action Leve I I I rational Targets

Sodium 5 ppb < 0.5Chloride 10 ppb <

Sulfate 10 ppb <

In order to achieve the target for contaminant sodium, PWR stations need to maintain sodium leakage at levelsnear 5 parts-per-trillion (ppt) or less. In addition, the service cycle of the polisher resin cannot exceed the onsetof amine breakthrough. Amine breakthrough occurs when the influent ethanolamine and/or ammonia haveexhausted the cation exchange resin to the point at which these amine species appear in the polisher effluentas indicated by specific conductivity. Under these conditions the service cycle run length of the polisher bed isessentially the operating capacity the cation resin affords for amine pick-up. The term "percent utilization", in

this particular application, refers to the amine pick-up capacity relative to the total exchange capacity of the

cation resin.

Means to increase utilization of the cation exchange resin

7

Engineers at Millstone Station explored two changes to the resin handling and regeneration procedures: 1) in-vessel mixing of the freshly regenerated mixed resin, and 2 extending the regeneration of the cation exchangeresin. The in-vessel mixing uses compressed air to re-mix the condensate polisher resin in the service vesselimmediately following the transfer from the mix and hold tank. This redistributes the cation resin equallythroughout the vessel and consequently more of the cation resin settles in the useable portion (upper section) ofthe service vessel. Unit 2 first implemented this procedure in February 1996 and Unit 3 put into practice inAugust 1999. Extending the regeneration is a simple approach that involves regenerating the cation resin until95% or more of the ethanolamine elutes from the cation resin bed. Sampling and analyzing the regeneranteffluent makes possible an elution curve (see Figure 6 that can be used to determine the time required forcompleting the regeneration step. Figure 7 shows that the system at Millstone reaches the desired endpointafter approximately 90 minutes.

Figures and 9 show the cation resin utilization performance for Millstone Station Units 2 and 3 After theinitial five or six service cycles, both units show fairly consistent cation resin utilization over the next ten servicecycles, near 75%. This is a significant improvement compared to a utilization range of 45% to 65%; reported 16

in 1996 from a survey of 8 different PWR units in the United States.

Condenser inleakage causes high chloride in the steam generators

High TIDS conditions existed in the hotwell during re-start preparations for Unit 2 due to ingress of seawaterfrom the coolant side of the condenser. Although the condensate polishers provided ion exchange capacity toremove the majority of the contaminant ions, the steam generators were showing unacceptably high chloridelevels. Efforts to correct this problem proved futile even after applying twice as much regenerant NaOH for theanion exchange resin, carefully examining the resin separation and regeneration process, and analyzing thebulk regenerant NaOH for high chloride impurity. The root cause of the problem was finally discovered whenengineers at Millstone measured the elution of contaminant ions from the anion resin.

35 -----------------------------------------------------------------

30 ------- -----------------------------------------------

Mco=' 25 ------- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -0

a- 20 ------ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

EL-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

0

10 ---- ---------------------- -------------------------

0O

5-- ----------------------- --------------

0

0 20 40 60 80 100 120 140

Time (minutes)

Figure 6. Ethanolamine Elution Curve during Regeneration of Charge #6 Cation Resin Millstone Station, Unit 2

8

120 ------------------------------------------------------------------

100 ---------------------------------------------------------

80 ------------------------ -------------------------------

60 ------------------ ------------------------------------------

FEw

40 ------------- -----------------------------------------------

20--- -----------------------------------------------------

0 1 i i i i0 20 40 60 80 100 120

Time (minutes)

Figure 7 Regeneration Efficiency of Cation Resin -- charge 6 at the Millstone Station, Unit 2

9

100

90

80----------- ------------------ --- --- --- --- --- --- --- --- --- --- ---

70

60

50

40

30

20

10

01 2 3 4 5 7 8 9 10 11 12 13 14 15 16

No. of Service Runs

Figure 8. Cation Resin Utilization for Charge 3 at Millstone Station Unit 2

100

90

80-------------------------- ----------- -- --- --- --- --- --- --- --- --- ---

70

60

so

40

30

20

io

01 2 3 4 5 6 7 8 9 10 11 12 13 U 15

No. of Service Runs

Figure 9 Cation Resin Utilization for Charge #1 at Millstone Station Unit 3

Extending the NaOH regeneration to correct the problem

Figure 10 is an example elution curve for one of the four condensate polisher charges that was exposed toabnormally high levels of free mineral acidity (chloride and sulfate) due to the condenser inleakage situation.As can be seen in this plot, about 150 minutes of NaOH regeneration is required to remove most of thecontaminant sulfate and about 300 minutes is required to achieve efficient removal of the contaminant chloride.This amount of NaOH is much higher than standard recommendations that this author has seen reported in theliterature. The results speak for themselves. Figure 1 1 shows the impact on steam generator chloride levelsonce the extended NaOH regeneration was applied to all four condensate polisher charges.

10

350 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

8% Caustic Used Throughout300 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

250 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

or de0-

200 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Sulfate ------C0coC 150 --------- ---------------------------------W

00 100 ------------------- ----- -----------

50 -- ------------------------ ---------------------------

0 i i i a __f___ m __*

0 50 100 150 200 250 300 350

Time (minutes)

FigurelO. ElutionofAnionsduringRegenerationofCharge#5Anion--MillstoneStationUnit2

14 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

12 -- - - - - - - - - - -SG-1--Non-Volatile Anions -----------------------------1 00% Power Operations

10 ------------------------------------------------- Chloride ---

CL Sulfate

8-1 -----------------------------------------------------------------0

--------- --- -- ----------------------------------------W

04 - - - - - - - - - - - - - - - - - - - - - - - - - -

2 ---- --- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

06/7 6/14 6/21 6/28 715 7/12 7119 7/26 8/2 8/9 8/16 8/23 8/30 9/6 9/13 9/20

Date (1 999)

Figure 1 1. Steam Generator Chemistry - Millstone Station Unit 2

The other important item to note from Figure 1 1 is the absence of a steam generator sulfate excursion prior tothe extended NaOH regenerations. This is explained by the concentration-dependent selectivity in the

*DE0192646�9-

exchange between divalent and monovalent ions. Although excess sulfate remained on the anion resinfollowing the regenerations up to mid-July 1999, the sulfate remained on the anion resin during the servicecycles due to its' extremely high affinity for strong base anion resin under conditions of low hydroxideconcentrations. Conversely, under conditions of high hydroxide concentration (i.e., during NaOH regeneration),the sulfate possesses much lower selectively and is more easily displaced when compared to chloride (seeFigure 10).

CONCLUSIONS

Condensate polishing is an application with many different design and operational aspects. The past decadehas brought new challenges for improved water quality with respect to both soluble and insoluble contaminants.Nonetheless, the endeavors to understand the compositional complexities of the ion exchange resin bead andthe convoluted dynamics of ion exchange chemistry and chemical engineering mechanisms occurring within themixed bed condensate polisher have brought new ideas and expectations for ion exchange resin in deep-bedcondensate polishers than ever before.

The new products and procedures presented here are a collaboration of a great deal of effort on the part ofresearchers, consultants, system engineers, station chemists, lab technicians and others. The studiesdiscussed in this paper unequivocally demonstrate the merits of:1. Aspeciallydesignedcationresintoachievegreaterthan95%insolubleironremovalefficiency2. Aless-separablemixedresinforimprovedcontrolofreactorwatersulfateinBWRprimarycycles3. Applying increased levels of regeneration chemicals and retrofitting the service vessels with re-mixing

capability to improve the operation of deep-bed condensate polishers in PWR secondary cycles

REFERENCES

1. Becker, M.W., Private Communication, September 1997.2. Izumi, T., et al, "Crud Removal Characteristics of Newly Developed Ion Exchange Resins", International

Water Conference, 1989.3. Wilson, J., et al, "Improved Corrosion Product Removal with Deep Bed Condensate Polishers", EPRI

Condensate Polishing Workshop, 1993.4. Malauskas, D., "Four Years of Operating with Iron Removal Resins at Dresden', EPRI Condensate

Polishing Workshop, 1997.5. Becker, M.W., "EPRI Collaboration Project TC6388 Re-qualification of Low Cross-Linked Resin for Iron

Control", EPRI Condensate Polishing Workshop, 1999.6. U.S. Patent No. 5302,623, Dhingra, Y.R. and VanTol, P.R. inventors, "Method of Stabilizing Cation

Exchange Resins Against Oxidative Degradation", assigned to The Dow Chemical Company, April 12,1994.

7. U.S. Patent No. 4973,607, Stahlbush, J.R., and Strom, R.M. inventors, "Cation Exchange ResinsIncorporating Antioxidants", assigned to The Dow Chemical Company, November 27, 1990.

8. Najmy, S.W., and Stahlbush, J.R., "TOC Leachables for Dow Stabilized GUARDIAN* and HGR-W2 Resins'EPRI Condensate Polishing Workshop, 1997.

9. Becker, M.W., "EPRI Collaboration Project TC6388 Re-qualification of Low Cross-Linked Resin for IronControl", EPRI Condensate Polishing Workshop, 2000.

10. Giannelli, J., et al, "Resin Improvements for Sulfate Reduction at LaSalle", EPRI Condensate PolishingWorkshop, 1996.

11. DAngelo, D.R., and Morgan, D.J., "Anion Resin Underlay Reduces Sulfur Throw from BWR Deep BedCondensate Polisher", International Water Conference, 1993.

12. Giannelli, J., Private Communication, April 1998.13. Giannelli, J., and Hillman, R., "Sulfate from A Salt Water Condenser Leak at Oyster Creek Nuclear

Generating Station", EPRI Condensate Polishing Workshop, 2000.14. Crone, L., "In-Vessel Mixing and Extended Regenerant Application Techniques to Enhance Resin

Performance", EPRI Condensate Polishing Workshop, 2000.15. Raught, D., Private Communication, August 2001.16. Najmy, S.W., "The Next Generation of MONOSPHERE Resin Condensate Polishing", EPRI Condensate

Polishing Workshop, 1997.17. Giannelli, J., Private Communication, October 2001.

12