PWR AND VVER SECONDARY SYSTEM WATER CHEMISTRY – STAND ... · PWR and VVER Secondary System Water...

P W R A N D V V E R S E C O N D A R Y S Y S T E M W A T E R C H E M I S T R Y – S T A N D A L O N E R E P O R T – 2 0 1 0

© January 2010

Advanced Nuclear Technology International

Krongjutarvägen 2C, SE-730 50 Skultuna

Sweden

[email protected]

www.antinternational.com

PWR and VVER Secondary System Water Chemistry

Authors

Suat Odar Erlangen, Germany

Francis Nordmann Beauchamp, France


Disclaimer

The information presented in this report has been compiled and analysed by

Advanced Nuclear Technology International Europe AB (ANT International®)

and its subcontractors. ANT International has exercised due diligence in this work,

but does not warrant the accuracy or completeness of the information.

ANT International does not assume any responsibility for any consequences

as a result of the use of the information for any party, except a warranty

for reasonable technical skill, which is limited to the amount paid for this Report.

Copyright © Advanced Nuclear Technology International Europe AB, ANT International, 2010.

I(VI)


Acknowledgements

Each author reviewed the manuscript of the other author. These reviews were performed by Francis Nordmann and Suat Odar, consultants.


II(VI)


Contents

Acknowledgements II 1 1-1 Introduction and background information

1.1 1-2 Steam generator degradation problems1.1.1 1-4 Steam generator corrosion problems1.1.2 1-8 Steam generator thermal degradation problems1.1.3 1-12 Steam generator thermo hydraulic problems1.2 1-14 Corrosion of the secondary side systems and components1.2.1 1-14 General corrosion1.3 1-16 Economical and environmental aspects

2 2-1 Description of the behaviour of the impurities and chemical species in the secondary side

2.1 2-1 Impurities inside the steam generator2.1.1 2-1 Steam generator crevice and top of tube sheet deposits2.1.2 2-4 Concentration process of various species hide out phenomenon2.1.2.1 2-5 Adsorption on tube oxide surfaces2.1.2.2 2-5 Concentration in flow restricted geometries2.1.2.3 2-10 Concentration in tube surface deposit crevices2.1.2.4 2-12 Hide-out tests2.1.3 2-13 Hide out return process2.2 2-19 Impurities and chemical species in the entire secondary side systems2.2.1 2-19 Decomposition of amines in the secondary system2.2.2 2-24 Origin and behaviour of organic species in the system

3 3-1 Historical evolution of secondary water chemistry in the past during power operation

3.1 3-1 Rationale for chemistry evolution3.2 3-3 Chemistry evolution for various countries3.2.1 3-3 Chemistry evolution in the USA3.2.2 3-5 Chemistry evolution in Japan3.2.3 3-6 Chemistry evolution in France3.2.4 3-7 Chemistry evolution in Germany3.2.5 3-9 Chemistry evolution in other western European countries3.2.6 3-9 Chemistry evolution in eastern European countries3.3 3-12 Early secondary side water chemistry programs3.3.1 3-12 Phosphate chemistry3.3.2 3-15 Low pH ammonia AVT chemistry3.3.3 3-17 Boric Acid Treatment (BAT)3.3.4 3-23 Other inhibitors (titanium, lithium hydroxide)3.3.4.1 3-23 Titanium and cerium compounds3.3.4.2 3-24 Lithium oxide3.3.5 3-24 Molar ratio control (US, Japan)

4 4-1 Design and materials used in the secondary side

4.1 4-1 PWR steam generators4.1.1 4-1 Steam generator design and materials4.1.1.1 4-8 SG tubing material4.1.1.2 4-12 Tube support arrangements4.1.1.3 4-14 Tube sheet expansion4.1.1.4 4-17 Flow distribution baffles4.1.1.5 4-18 Preheater SGs and feedring SGs


III(VI)


4.1.2 4-18 Chemistry related steam generator tubing experience4.1.2.1 4-18 SG tubing material Alloy 600 MA (Inconel 600 MA)4.1.2.2 4-24 SG tubing material Alloy 600 TT4.1.2.3 4-24 SG tubing material Alloy 690 TT4.1.2.4 4-25 SG tubing material Alloy 800 NG4.2 4-25 VVER steam generators4.3 4-31 Entire steam water cycle4.3.1 4-31 PWR plants4.3.1.1 4-31 General design concept4.3.1.2 4-33 Materials used4.3.1.3 4-33 Condensers4.3.1.4 4-35 SG blow-down recovery4.3.2 4-36 VVER plants4.3.2.1 4-36 General design concept4.3.2.2 4-37 VVER Steam Generator Blowdown (SVO-5)

5 5-1 Encountered SG degradation and influence of various impurities on SG performance

5.1 5-2 SG degradation problems5.1.1 5-3 Wastage5.1.2 5-5 Denting5.1.3 5-8 Pitting5.1.4 5-11 IGA / SCC5.1.5 5-23 Flow induced vibration5.2 5-25 Influence of various impurities and parameters on SG performance5.2.1 5-28 Sulphate and sulphur compounds5.2.2 5-30 Lead

6 6-1 Adequate water chemistry for improved SG performance

6.1 6-1 Flow Accelerated Corrosion (FAC)6.1.1 6-5 Thermo-hydrodynamic parameters6.1.1.1 6-5 Temperature6.1.1.2 6-6 Flow rate6.1.1.3 6-7 Flow geometry6.1.2 6-8 Alloy composition6.1.3 6-11 Water chemistry6.1.3.1 6-11 pH values6.1.3.2 6-13 Oxygen concentration6.1.3.3 6-15 Hydrazine concentration6.2 6-18 Corrosion product control6.2.1 6-20 High ammonia concentrations6.2.2 6-22 Alternative (advanced) amines6.2.3 6-25 Support by oxygen6.3 6-25 Impurity control6.3.1 6-26 Make-up water6.3.2 6-31 Condenser leaks6.3.3 6-32 Pollution from other systems6.3.4 6-32 Ion Exchange Resins6.3.5 6-35 Regeneration6.3.6 6-35 Consumables6.3.7 6-37 Manufacturing and maintenance activities6.4 6-38 Oxygen control


IV(VI)


7 7-1 Different water chemistry strategies (state of art)

7.1 7-1 High pH operation (High-AVT)7.2 7-12 Alternative amines7.2.1 7-12 Various additives and objectives7.2.1.1 7-12 Alkaline reagent7.2.1.2 7-13 Hydrazine7.2.2 7-15 Amines properties7.3 7-23 Oxygen injections / oxygenated water chemistry7.3.1 7-23 German experience7.3.2 7-28 Japanese experience7.4 7-32 Dispersants7.5 7-33 Various options and rationale for selection7.5.1 7-34 Materials compatibility7.5.2 7-37 Relation with low and high temperature pH7.5.3 7-39 Relation purification means, liquid wastes and costs7.5.4 7-41 Rationale for treatment selections

8 8-1 Different strategies of using purification systems

8.1 8-3 Condensate polishing system: Various options and rationale for selection8.2 8-11 SG blow-down demineralizer: Various options and rationale for selection8.3 8-15 Possible recovery of condenser effluents

9 9-1 Different maintenance strategies in relation with chemistry (state of art), costs control

9.1 9-1 Chemical cleaning processes9.1.1 9-2 Hard chemical cleaning processes9.1.1.1 9-3 EPRI SGOG process (low temperature cleaning process)9.1.1.2 9-5 High temperature chemical cleaning process (AREVA process)9.1.2 9-7 Maintenance chemical cleaning processes9.1.2.1 9-7 Scale Conditioning Agents (SCA)9.1.2.2 9-7 Advanced Scale Conditioning Agents (ASCA)9.1.2.3 9-9 Ultrasonic Energy Cleaning (UEC)9.1.2.4 9-9 Deposit Minimization Treatment (DMT)9.1.2.5 9-10 Deposit Accumulation Reduction Treatment (DART)9.2 9-10 Mechanical cleaning processes9.2.1 9-10 Tube sheet lancing (hydraulic cleaning)9.2.2 9-12 Upper bundle hydraulic cleaning9.2.3 9-13 SG Tube scale removal by thermo-hydraulic effects9.3 9-14 Rationale for selection of cleaning technologies

10 10-1 Chemistry control and monitoring

10.1 10-1 Purpose of on line monitoring and grab sampling10.2 10-2 Criteria for representative sampling10.3 10-9 Rationale for selection of each type, number of control points10.4 10-13 Relation between pH, conductivity versus reagent content and some ions

11 11-1 Application of water chemistry control programs

11.1 11-1 Power operation11.1.1 11-1 EPRI guidelines11.1.2 11-3 EdF guidelines11.1.3 11-4 VGB guidelines11.1.4 11-5 VVER guidelines11.1.5 11-6 Discussion on various parameters11.1.6 11-10 Comparison of various guidelines11.1.7 11-12 Possible evolution of guidelines11.2 11-14 Plant start-up and shutdown11.3 11-18 Plant lay-up


V(VI)



VI(VI)

12 12-1 Conclusive summary and recommendations

12.1 12-1 Overall objectives12.2 12-3 Impurities behaviour12.3 12-3 SG degradation evolution12.4 12-5 Impurities mitigation and impact12.5 12-8 Corrosion product and FAC mitigation12.6 12-9 Treatment options and selection12.7 12-11 Purification options12.8 12-12 SG cleaning options12.9 12-14 Chemistry monitoring and guidelines12.10 12-15 Shutdown and lay-up

13 13-1 References

Acronyms and expressions Unit conversion


1 Introduction and background information

This report describes the secondary side water chemistry applied in Pressurized Water Reactor (PWR) and Voda Voda Energo Reactor (VVER) plants. It covers the range from basic information to current knowledge. In the following a brief introduction to the content of this report is given.

The objective of secondary side water chemistry control is to minimize corrosion damage and performance losses for all secondary system components and thereby to maximize the reliability and economic performance of the secondary system. To achieve this objective, the water chemistry has to be compatible with all parts of the secondary system including Steam Generators (SGs), turbines, condensers, Feed Water heaters, Moisture Separator Reheaters (MSRs), and finally piping. Special emphasis has to be put on SGs, because they are one of the key components of the PWR plants. Their degradation or performance loss greatly affects the overall plant performance. Most of the SG degradation problems are related to corrosion caused by inappropriate material and design selection or poor secondary side chemistry. For achieving a better understanding of the importance of water chemistry, the field experience regarding SG degradation problems and corrosion problems experienced in the secondary side components are briefly explained in Sections 1.1 and 1.2 respectively; more detailed description is given in Sections 3 and 5. For the selection of the secondary side chemistry, economical and environmental aspects need also to be considered; these aspects will be discussed in Sections 1.3 and 7.

The main causes for the SG corrosion problems are the concentrated impurities beneath the deposits on the Tube Sheet (TS) or in the Tube to Tube Support Plate (TSP) crevices. The behaviour of the different impurities and chemical species in the entire secondary side, as well as their transport into SGs and the mechanism of impurity concentration in the SGs are described in Section 2. Different impurities can influence the SG corrosion performance in a different degree; this will be described in Section 5. The most crucial SG degradation problems were experienced from the early 1970s until the end of 1990s. During this period of almost 30 years of PWR operation a lot of modifications and/or improvements of water chemistry and/or SG materials and design were performed to counteract these SG degradation problems. The world wide PWR secondary side water chemistry modifications and their historical evolution is described in Section 3. The improvements in SG design and materials are discussed in Section 4. In addition to these, this section also includes description of the design and the variety of materials typically used in secondary systems.

Based on thirty years of PWR operating field experience gained with counteracting the SG degradation problems. Common sense exists now worldwide, how to control the chemistry conditions to mitigate SG degradation problems. But this is unfortunately not always properly applied. Secondary side water chemistry selected to protect SGs is usually also satisfactory for entire secondary side components. The key control parameters of adequate secondary side water chemistry are described in Section 6. Section 7 describes the recent different water chemistry strategies applied worldwide to fulfill the water chemistry objectives.

The worldwide common sense to avoid SG corrosion problems is “keeping the SGs in clean conditions”. This can be maintained by minimizing the ingress of impurities and corrosion products into SGs, as well as by cleaning the SG deposits containing concentrated impurities. The secondary side purification systems, i.e. Condensate Polishing System (CPS) and Blow-Down Demineralizers, and their different operating strategies are described in Section 8. Section 9 explains the different SG mechanical and chemical cleaning technologies.

Adequate secondary water chemistry application relies on water chemistry monitoring results. The concept of water chemistry sampling and the injection of chemicals to control the specified chemistry goals are described in Section 10. Section 11 explains how the selected water chemistry control programs are applied in different plant operating modes, like in power operations, start-up or shutdown operations as well as lay-up chemistry program during annual outages. Finally, Section 12 concludes the information given in the previous sections and contains recommendations for a good secondary side water chemistry practice.


1-1(1-18)


1.1 Steam generator degradation problems

The secondary side of the early PWRs, which were designed and build in the 1960s and 1970s, were based on the industrial experience gained with fossil fired conventional power plants:

As structural materials, plain Carbon Steels (CS) were used for pipe work and building components overall on the secondary side. Because of the high thermal conductivity of copper, copper bearing materials were selected for the tubes of Feed Water Heat Exchangers (FW-HX), MSR, and Condensers. Ferritic Stainless Steel (SS) was selected for the turbine blades. After a short use of austenitic SS (18% Cr – 10% Ni) in few PWR plants, Alloy 600 MA1 (I 600 MA), a nickel based alloy, was used as SG tubing material. In Russian designed VVER plants, austenitic SS is still used for SG tubes. In PWR Recirculating Steam Generators (RSG), CS drilled hole TSP were used to support the SG tubes against vibration (see Section 4.1). These types of TSPs were experienced to be very sensitive against impurity concentrations and denting corrosion, as described later. For impurity control, CPS was designed especially for the PWRs with seawater cooling and also for the PWRs with Once-Through Steam Generators (OTSG), which do not have Steam Generator Blow-Down (SGBD) capability. The water chemistry selection was again based on the well-experienced fossil power plant chemistry; i.e., “Phosphate Chemistry” (PO4 Chemistry) was used for the RSGs and “All Volatile Treatment” (AVT Chemistry) was applied for the entire secondary side. For OTSGs, AVT chemistry was the respective SG chemistry. Feed Water (FW) pH values, sufficiently high to adequately control the Flow Accelerated Corrosion (FAC) of the CSs could not be applied, because copper bearing materials was used almost for all secondary side heat exchanger tubes. This resulted in high FW iron concentrations and accordingly high corrosion product deposits in the SGs. So called “combined chemistry”, which is also successfully applied to control the FAC in fossil power plant industry, using ammonia together with oxygen injection, could not be selected, due to corrosion sensibility of SG tubing material Alloy 600 MA under oxidizing conditions. This type of secondary side plant concept (selected design and structural materials) does not allow the selection of adequate water chemistry in order to avoid high FW iron transportation (corrosion product transport) into SGs. This poor corrosion product control in the secondary side was together with a sensitive SG tubing material and inadequate design of the SG tube support construction the main root cause of the SG degradation problems, as it is described in the following.

Since the early 1970s, lot of SG corrosion problems was experienced in the nuclear industry. Figure 1-1 shows the causes of SG tube degradation since 1973 up to 2006 [EPRI2, 2006In the period of early 1970s until mid 1990s, the majority of the tubes were repaired by pluggisleeving due to corrosion problems (see

]. ng or

worldwide.

Figure 1-2). These SG problems often forced the plants to perform unscheduled or extended outages for preventive and/or corrective maintenance measures. In addition many SG replacements were necessary, which were costly in terms of repair work, personnel radiation exposure and loss of power [EPRI, 2000] and [EPRI, 2006]. Especially in US, where the majority of problems were reported on the secondary side of SG tubes, the industry made a lot of efforts to improve the design and materials of the SGs and the secondary side water chemistry to minimize the SG degradation problems during the period of 1970s to end of 1990s (see Sections 3, 4 and 5). Although tremendous progress has been made in controlling the SG degradation problems, as it can be seen in the improvement of the US PWR capacity factors (see Figure 1-3), minimizing its impact on plant operation will remain a continuing challenge

1 Mill Annealed 2 Electric Power Research Institute


1-2(1-18)


Figure 1-1: Worldwide causes of SG tube degradation [EPRI, 2006].

Figure 1-2: Worldwide percentage of SG tubes plugged [EPRI, 2006].


1-3(1-18)


Figure 1-3: US PWR capacity loss due to SG problems [EPRI, 2006].

In the following sections the SG degradation problems will be described in detail.

1.1.1 Steam generator corrosion problems

Majority of these SG tube corrosion problems were experienced in the period of early 1970s to end of 1990s. In many countries with some exceptions like in France, where most of the Stress Corrosion Cracking (SCC) was experienced on the primary side of the SG tubes, most of these corrosion problems were on the secondary side of the SG tubing. Figure 1-4 summarizes the worldwide SG degradation problems. It contains more information with respect to type of tube corrosion mechanism as compared to Figure 1-1. During this period of early 1970s to end of 1990s, the majority of the SGs worldwide used Alloy 600 MA as tubing material.

Figure 1-4: Worldwide SG corrosion experience [EPRI, 2000].


1-4(1-18)


Figure 1-5 illustrates the reported Alloy 600 MA tube degradation per each mechanism worldwide. Similar information is given in the Figure 1-6 for Alloy 800 SG tubing material, which is used in Siemens- KraftWerk Union (KWU) designed PWRs and in some Canadian Deuterium Uranium (CANDU) plants.

Figure 1-5: Worldwide causes of I600 MA SG tube repair by degradation mechanism [EPRI, 2006].

Figure 1-6: Worldwide causes of I800 SG tube repair by degradation mechanism [EPRI, 2006].


1-5(1-18)


As described already, based on good experience gained in fossil-fired conventional power plants, almost all early PWR plants started with PO4) chemistry in their SGs (see also Section 3). The applied SG PO4 concentration was depended on the FW purity. In low-purity plants, i.e. in plants with frequent condenser leaks, high PO4 concentrations of 40-50 mg/kg were applied, whereas in high-purity plants PO4 concentrations were in the range of 2 – 6 mg/kg (e.g. Siemens designed PWRs). In the early seventies phosphate wastage corrosion was experienced in many PWR plants, which were operating with high PO4 concentrations in their SGs, resulting in primary to secondary side SG leaks. This type of degradation is described in Section 5.

However, most of these plants performed the chemistry switch without improving their condenser leak tightness. PO4 was used to buffer the pH excursions in the medium beneath deposit, which was caused by concentrated salt impurities of condenser leaks. Hence, in many PWR plants, where PO4 chemistry was stopped without improvements in the condenser leak tightness, wastage corrosion was followed by denting corrosion. Actually denting is not SG tube corrosion but a mechanical deformation of the tubes. Denting occurs when the SG tubes are squeezed by corroded CS (drilled hole TSP) at the tube to TSP intersection as indicated in Figure 1-7. For the occurrence of denting at field, acidic and oxidizing conditions are required. The acidic conditions are usually caused by chlorides, which enters into secondary side by condenser leaks (mainly seawater) or by sulfates (for example in make-up water as residual of resin regeneration). Oxidizing conditions can be created by air ingress or also by copper ions in the SG deposits. The denting phenomenon will be described in Section 5 more in detail.

The occurrence of denting required that the SGs were replaced at two PWRs (Surry and Turkey Point) in the US. The problem has led to extensive tube plugging due to primary side cracking at supports and U-Bends and secondary side cracking at tube supports (see Section 5).

Figure 1-7: Schematic picture SG tube denting mechanism.

In Siemens designed PWRs, operated with lower PO4 concentrations, wastage corrosion was experienced much later in the end of 1970s. In these plants, chemistry was converted to “High AVT Chemistry” (AVT chemistry with high FW pH values), after replacing all copper bearing materials from the secondary side systems. Denting was not experienced after switching from PO4 chemistry.


1-6(1-18)


The occurrence of denting demonstrated the need for significant upgrading of the secondary system design and secondary side water chemistry practice. In the mid to late 1970s and early 1980s, the secondary side water chemistry was significantly improved by remedial actions in the secondary systems, like with the reduction of impurity and oxygen ingress by improvement of condenser integrity and the elimination of air leakages into the secondary system. The tube supports in new SGs or in replacement SGs were also changed from CS to SSs to reduce the corrosion rate, which caused the denting phenomenon. These SG modifications by various manufacturers are described in Section 4 for both TSP materials and design.

In many plants with old SG design and materials (CS drilled hole TSPs), where the secondary side modification was not or could not be performed, denting was counteracted and minimized by adding boric acid into SGs, which inhibits the CS TSP corrosion. This so called Boric Acid Treatment (BAT) was very successful so that no more severe denting problems were experienced like it had been in the 1970s in older PWRs, which still had CS drilled hole TSPs. This BAT is described in details in Section 3.

As a result of these improvements, denting has largely been controlled although some minor denting still occurs, mainly at Top of Tube Sheet (TTS), probably due to the abnormal presence of metallic particles coming from insufficient cleanliness after manufacturing and construction. Despite these improvements in water chemistry, other SG secondary side corrosion problems have occurred. These include pitting and SCC as already mentioned.

Pitting was experienced mainly in the early 1980s up to beginning of 1990s. It occurs in sludge pile regions under acidic and oxidizing conditions. The chlorides, which concentrate beneath deposits, are the main cause for the producing acidic conditions, damaging the protective oxide layers on the tube surface and initiating pits (see Section 5). Pitting can mainly be avoided by eliminating the ingress of chlorides. Although the improvements made in the area of condenser leak tightness and impurity monitoring has significantly reduced the SG tube pitting corrosion, some single events still occur.

The SCC has followed denting at the end of 1970s. Secondary side SCC occurs under stress conditions in corrosive media, either extreme caustic or extreme acidic (pHT

3 : > 9 or < 4). Stress is responsible for damaging the protective oxide layers on the tube surface, which protects the tubing material against corrosion. Initiation of SCC can also be caused by pitting, which also damages the protective oxide layer. SCC was experienced not only on the secondary side of the SG tubes (Outer Diameter Stress Corrosion Cracking: (ODSCC)), but also on the primary side (Inner Diameter Stress Corrosion Cracking: (IDSCC) or Primary Water Stress Corrosion Cracking: (PWSCC)). IDSCC is caused mainly due to corrosion sensitivity of the selected tubing material Alloy 600 MA in presence of stresses. As Coriou had found, this material is not stable even in pure water at high temperatures [Coriou et al, 1959]. The SCC will be explained in detail in Section 5.

As mentioned above, for the secondary side cracking, besides the sensitivity of Alloy 600 MA, corrosive conditions are responsible, which is produced by concentrated impurities within the tube to TSP crevices or beneath deposits at TTS. In the 1980s, it was considered that Alloy 600 MA ODSCC was mainly occurring under caustic crevice conditions. Therefore, many utilities started again to inject boric acid into the secondary side of their SGs with the objective to neutralize the caustic crevices to counteract the ODSCC. This chemistry measure was unfortunately not as successful as it was in the case of denting. The reason was probably the insufficient penetration of boron to the tube surface within the TSP crevice as explained in Section 3.

3 pH at temperature (operating temperature)


1-7(1-18)


This historical variation of the SG tube corrosion mechanism depends on the change of the environmental chemistry conditions of the SG tubing, caused by the change of the secondary side water chemistry (see Section 3). But in most cases, secondary side SG tube corrosion is caused by the presence of corrosion product deposits in contact with the SG tubes and by the presence of concentrated impurities within these deposits. Non-volatile impurities, which are always present even in trace quantities in the FW, enter into the SGs with the FW and concentrate in the deposits at the TTS, or in the tube support crevices and on the tube surfaces (see Section 2).

In addition to BAT, many utilities, developed and applied the so called “Molar Ration Control” (MRC) in their SGs, especially in USA and Japan [EPRI, 1980-1990] and [Millett, 1995]. This chemistry measure is balancing the concentrated anions and cations in the crevices to neutral or slightly acidic pH conditions by adding the missing content of acidic chloride into the SGs. Despite some promising laboratory data and the hope that these modifications be helpful, they were unfortunately not sufficient enough to stop or to drastically mitigate the corrosion of sensitive SG tubes. In the presence, the most experienced corrosion for the SGs with Alloy 600 MA tubing material remains the secondary side stress corrosion cracking (ODSCC) (see Figure 1-4). Consequently, numerous SG tubes had and still will have to be repaired by plugging and/or sleeving, or even SG will be replaced, which already has been initiated at most of the PWR units (see Figure 1-5).

1.1.2 Steam generator thermal degradation problems

Another SG problem caused by corrosion products is the primary to secondary heat transfer degradation, the so-called SG tube fouling. Majority of the corrosion products transported by FW into SGs are absorbed on the heat transfer surfaces of the tubes, building oxide scales.

According to the work, performed by Atomic Energy of Canada Limited (AECL) in collaboration with EPRI, the rate of tube fouling is strongly dependent upon the surface chemistry of the corrosion products and the amine used for pH control [Turner et al, 2002]. The fouling rates of fully oxidized iron oxides, such as hematite and lepidocrocite, are significantly greater than the fouling rate of magnetite under identical operating conditions (see Figure 1-8). The difference is related to the sign of the surface charge on the corrosion products at temperature. At operating temperature, the Alloy 600 MA heat surface has a negative charge. At this temperature, magnetite particles have a negative surface charge, whereas hematite and lepidocrocite are both positively charged. This surface charges accounts for the difference between the deposition rates of magnetite and hematite. In case of hematite, no repulsion exists between heat transfer surface and corrosion product particles resulting in high deposition. In contrary, the deposition of magnetite on heat transfer surfaces is less due to repulsive forces. This different surface chemistry behaviour of magnetite and hematite is important to understand the field experience regarding SG tube fouling: Some plants have reported increased SG tube fouling just after plant start-up operation, which was correlated to increased oxygen concentrations during start-up operation according to the utilities.


1-8(1-18)


Figure 1-8: Normalized fouling rate of various iron-based corrosion products under flow-boiling conditions [Turner et al, 2002].

At the beginning of plant lifetime, the deposition of corrosion products, magnetite, creates porous oxide scales on the tube surfaces, which increase the internal tube surfaces, resulting in an increase of nucleated boiling, accordingly enhancing the primary to secondary heat transfer. But these oxide scales grow and consolidate with the plant operating time and after a certain time the heat transfer rate begins to decrease, due to the isolation effect of the consolidated tube scales. Consolidation (or ageing) is a process whereby particles become chemically bonded to either the heat-transfer surface or to the pre-existing deposits. Deposit that has aged, or become consolidated, is strongly bound to the surface and therefore cannot be removed by the fluid. The driving force for the consolidation of deposit is postulated by the precipitation of dissolved species within the pores of the deposit (so called Ostwald ripening). It is enhanced by the temperature gradient at heat transfer surfaces. The process is thermodynamically favored because it is accompanied by a reduction in surface area and, therefore, by reduction of surface energy. Also boiling induced precipitation is a part of the deposit consolidation mechanism.

Figure 1-9 illustrates a typical SG tube fouling behaviour as a function of plant operating time. The fouling growth rate depends on the FW iron concentration as shown in the example of one PWR in Figure 1-10. In all plants where FW iron concentrations are drastically reduced, this tube foulinphenomena is either completely stopped or has become insignificant. Some examples together with the influence of pH agent (amine type and ammonia) on fouling will be discussed in Section

g

7.


1-9(1-18)


Figure 1-9: Typical SG tube fouling experience as a function of plant operation.

Figure 1-10: SG tube fouling and integrated iron ingress into SG at one German PWR.


1-10(1-18)


SG tube fouling caused by corrosion product deposits was also experienced in VVER SGs. In the past, in most of the VVER plants, due to their operating secondary side water chemistry with neutral or very low pHT values (ammonia treatment at low pH25 °C), FW iron transport values were very high. This resulted in heavy corrosion product deposits on the SG tubes (see Figure 1-11). After reaching a certain thickness, due to different thermal expansion coefficientof corrosion products and tube material, these deposits de-scale and fall down on to the bottomparts of SG during plant start-up and shut-down operations. An example of this phenomenon can be seen in

Figure 1-12. These heavy tube deposits were the reason for power loss in many VVER plants. The VVER manufacturer have specified a limit for oxide deposits on SG tubes for operating plants and new plants with 150 g/m2 and 100 g/m2 respectively. When these limits are exceeded removal of deposits is required by for example chemical cleaning [Mamet et al, 2002].

Figure 1-11: Heavy SG tube deposits (~ 2mm thick) in one VVER 400 plant [Odar, 2006].

Figure 1-12: SG deposits as flat scales in the lower rows of a VVER plant operating at low FW pH values [Mamet et al, 2002].


1-11(1-18)


In summary, it can be stated that the fouling behaviour of corrosion products under SG operating conditions is strongly dependent upon surface chemistry and on the amine used for pH control. Oxidizing conditions in the FW train, for example during plant start-ups, that favor formation of hematite, can result in higher fouling rates than reducing conditions, that favor the formation of magnetite. The amine used for pH control affects the SG tube fouling by influencing both the rate of particle deposition and the rate of deposit consolidation on the SG tubes. The fouling rate is the lowest when both the rate of deposition and deposit consolidation is low. The result of this tube fouling is the SG pressure decrease. Forty years of field experience with PWR and VVER plant operation confirmed that nearly all PWR and VVER plants are affected by this problem. For example, as of 2006, a decrease of SG pressure was reported in ten Electric de France (EdF) PWR units with a pressure drop rate of greater than 100 mbar/year, with some of them not even being able anymore to operate at 100% power [Corredera et al, 2008].

1.1.3 Steam generator thermo hydraulic problems

Since several years, after introducing broached hole TSP design, blockages of the broach holes by corrosion product deposits was reported in the industry. Based on performed visual inspections, it was found that the gaps between the tube and TSP at the lower part of the broached holes were getting blocked [Nordmann & Pitner, 2006a] (see Figure 1-13) [Nordmann & Pitner, 2006b] and [Corredera et al, 2008]. Depending on the operating water chemistry and the FW iron transport rates, the TSP blockage rates was found to be plant specific in the range of 2-6% per year.

Figure 1-13: Visual inspections: Blockage of broached hole TSPs by corrosion products [Nordmann & Pitner, 2006b].

In some PWR plants, this TSP blockage has caused SG water level oscillations, as shown as an example in the case of Maanshan PWR units (see Figure 1-14) [Chen, 2007]. Usually this TSP blockage is experienced in the upper TSPs. For safe operation, in most affected PWRs power had to be reduced, in case of Maanshan Units the power reduction was 5%.


1-12(1-18)


Figure 1-14: SG wide range water level oscillation in SG C at Maanshan Unit 2.

Recently, in two EdF PWRs, Cruas Unit 1 and 4, this TSP blockage has even caused severe problems due to circumferential cracking of three tubes (one at Cruas-1 and two at Cruas-4), [Nordmann & Pitner, 2006a and b]. TSP blockage in certain areas resulted in increased flows in other areas, which caused this fatigue cracking of some tubes by vibration. Due to these cracks, SG leaks occurred (at Cruas-1: 9 l/h; at Cruas-4: 500 l/h). The phenomenon and its influencing parameters are described in Section 5.

Summarizing conclusions of Section 1.1 SG problems

All these SG degradation problems, not only the corrosion but also the thermal efficiency and the thermo hydraulic issues, can be counteracted by keeping the SGs clean from corrosion products. As it is confirmed by field experience: All plants, where adequate secondary side corrosion product control was performed, were free of these SG degradation problems, or at least less affected in the case of highly sensitive Alloy 600 MA tubing. In the past when large amount of acidic or alkaline impurities entered the SGs, corrosion easily occurred. Then with more satisfactory water chemistry in the plants, where insufficient attention was paid for the corrosion product control, the SG problems still continued. This confirms the importance of adequate selection of the secondary side water chemistry in addition to avoiding the presence of large impurities.


1-13(1-18)


1.2 Corrosion of the secondary side systems and components

1.2.1 General corrosion

Bare CS (non-alloyed) is not stable in presence of oxygen in water. If it is exposed to water, bare iron metal will be dissolved and iron ions will be released into water, according to the following chemical equation, which is called anodic reaction:

Fe0 Fe2+ + 2e-

The remaining electrons accumulate at the metal surface. This excess of electrons at the metal surface and the ion-layer in the water just close to metal surface build an electrical field with Electro-potential difference between those electrical layers (see Figure 1-15). Corrosion is the dissolution of metal (Me0 Me+) at anodic locations and it takes place, when the electrons at the metal surface (cathodic location) are removed by cathodic reactions.

Figure 1-15: Metal dissolution in water and corrosion mechanism.

Typical cathodic reactions in aerated water are the following:

2H+ + 2e- H2 and

O2 + 4e- + H2O 4OH-

If anodic and cathodic locations are homogeneously distributed at the metal surface, the corrosion occurs also homogeneously, and is called general corrosion. If the anodes are concentrated at one location, due to electro neutrality rule (number of anodes = number of cathodes), at this location anodic current density will be increased resulting in an accelerated corrosion, causing pits (pitting corrosion).

Because H+ and/or OH- are involved in the above given cathodic reactions their electro potential depends on the pH value of the water according to the Nernst’s equation (see Figure 1-16):

E = E°H – RT / nlog COx/CRed (Nernst’s equation)

Prerequisite for a corrosion reaction is:

Electro Potential of cathodic reaction > Electro Potential anodic reaction


1-14(1-18)


Iron metal (CSs) in water has an anodic potential of – 0.44 VH. In the pH range of 4 to 12, oxygen electrode potential is ranging between + 0.5 and +1.0 VH, i.e. always higher than the Fe potential (see Figure 1-16). Accordingly, in presence of oxygen in water, non-alloyed bare iron metal will always corrode. In absence of oxygen this corrosion can be avoided by increasing the pH value (e.g. at pH = 9.0) Fe potential will be above the potential of H2 electrode.

As a consequence, in order to avoid CS corrosion, it is required to be oxygen free and to increase the water alkalinity. Under alkaline conditions the Fe ions, which are released from CS surfaces into water react with OH- ions of the alkaline agents building passive oxide layers, In return, this protects the CS from further corrosion. Therefore, under adequate water chemistry conditions, CSs have been widely used as economical contractual materials on the secondary side of power plants.

Figure 1-16: Equilibrium potential for H2 and O2 electrodes in water.

One of the main objectives of the secondary side water chemistry is building and maintaining these protective layers on CS surfaces. If this passive protective layer is locally destroyed (e.g. by chlorides enriched beneath deposits) it can result in a pitting corrosion or even in a SCC when stresses are simultaneously present on steel (for brief explanation of these selective type of corrosion please refer also to Sections 1.1 and 5).

In presence of protective oxide layers CSs corrode in very small extent, due to the dissolution of oxide layers. Under reducing and alkaline conditions magnetite generally is built as a protective layer in the secondary side. If this protective layer is properly built and in absence of FAC, this magnetite dissolution results in FW iron concentrations in the range of 1-2 μg/kg. This is the basic contribution of general CS corrosion, in the FW iron transportation during the power operation. However, increased general CS corrosion can also occur during other plant operation modes like plant start-up and shut-down, due to injection of aerated make-up water for maintaining a constant SG water level; or during annual outages if proper lay-up of systems is not implemented. Therefore oxygen control during plant start-up and shut-down operations and proper lay-up of systems during annual outages are important for minimizing the corrosion product ingress into SGs.

If the secondary side chemistry is not adequately applied to maintain the protective oxide layers on the CS surfaces, this results not only in increased general CS corrosion, but also in many cases in so called FAC (also named Erosion Corrosion(EC)) in systems with high medium velocities. This is the main source of corrosion product transport into SGs, and will be explained in Section 6.


1-15(1-18)


1.3 Economical and environmental aspects

At the beginning of the design and operation of Nuclear Power Plants (NPP), Environmental consideration was not a priority, in the same way as for many other industrial activities or way of living. But with the increasing influence of political issues, the sustainable environment concern and moreover the impact of a growing number of plant in operation, it appeared necessary to better take into account these environmental issues.

There is no need to explain why the economical aspects are of always growing importance since this is the case in any other activity in a more open and competitive world. On this aspect, economical aspects were not so crucial at the beginning of the nuclear program in various countries in the seventies for various reasons:

• The weight of administrative and regulatory constraints were much lower in the past than now.

• Some countries decided to build NPP under the government responsibility and long term governmental financial involvement, meaning the absence of need of return from investment on the short or medium term.

• Some of the above countries decided to build a nuclear program for a strategy of electricity generation independence, due to insufficient national resources of fuel, coal or gas (like France) and nuclear energy was the solution, independently of the economical aspect; some other countries, although having fossil resources decided to have nuclear power plants for long term strategy or various other reasons.

• Some countries or private utilities decided to build NPP, purely for economical reason when it was more competitive than producing electricity from fossil resources, particularly when the regulatory constraints were not as high as now and fuel prices was high.

• Nowadays, global warming is another reason to have NPP.

But whatever the original reason for operating a NPP, with the present worldwide economical world and interconnected market in several countries (particularly within Western Europe), it become very important to build and operate NPP in an economical way, without nevertheless jeopardizing safety.

Consequently, some advantage of components reliability improvement should be taken into account to adjust NPP design in a more economical way, either related to investment or to operating costs. A typical example is the presence of condensate polishing plants, which were very common in the past due to the use of condenser tubing with copper alloys. They were not tight enough on the long term. With the current use of tight and reliable integrity condenser technology in new NPP or in old NPP where condenser are progressively replaced, the need for condensate polishers is not at all the same as in the past. Thus, for new NPP, it is extremely important to reconsider old practices of some utilities to systematically decide the condensate polisher’s implementation in NPP with RSGs having a BlowDown (BD) capability. Avoiding the installation of condensate polishers, as fully explained in Section 8, allows significant investment savings. When they already exist, keeping them in by-pass mode is an excellent way, whenever possible with tight condensers and RSGs, to minimize operating costs.

The advantage of avoiding the use of condensate polishers is not only economical but also environmental, drastically decreasing liquid wastes releases from regeneration and higher content of chemical injection. Besides these advantages, the absence of condensate polishers in permanent operation is an excellent way for better optimizing the secondary system chemical treatment and associated beneficial aspects (pH, CS corrosion and associated consequences) that are described in other sections. Exceptionally, the absence of condensate polishers has a positive impact both for some materials behaviour, for economical and for environmental aspects while it is frequent to see that economical options are in contradiction with a more reliable behaviour of components.


1-16(1-18)


Economical and environmental approaches must also be taken into account for SGBD design and operating mode, which are not independent from the condensate polishers’ presence and operating strategies.

This is why a great emphasis is given to these options, fully detailed in Section 8.

For economical aspects, optimization of chemistry should be considered, with for example the SGBD limits for impurities during normal power operation or start up. With the use of more resistant SG tubing, such as Alloy 690 TT or Alloy 800 NG, of a behaviour completely different from Alloy 600 MA, it is extremely interesting to take advantage of this improvement to eliminate unnecessary limits that had been progressively set for SG with Alloy 600 MA, sensitive to corrosion. Such an approach has been proposed for example to EPRI in the last Water Chemistry Guidelines Revision, without of course jeopardizing the long term integrity of SG tubing. Siemens-KWU designed PWRs with SGs having Alloy 800 NG tubing material are using since mid of 1980s water chemistry guidelines without such unnecessary limitations. Such chemistry specification or guidelines allows a greater flexibility of operation and thus less unavailability.

Such evolution of chemistry specification, in addition to economical benefits, are also friendly environmentally positive by lowering the quantity of solid and liquid wastes into the environment. Example also detailed in Section 8 is related to recycling of non-condensable parts to the condenser instead of releasing them to the environment. Another example is the longer use of resin without replacement or regeneration with environmental benefit superseding the economical aspect.

As far as environmental consideration is concerned, the strategy is more complicated when defining the secondary water treatment, since there are several contradictory aspects and potential consequences on material behaviour. The very best example of what is discussed in Section 7 is concerning the reagent selection and associated pH and concentration. Selecting the amine depends on many parameters while the concentration increase, and thus the pH has:

• A positive impact for FAC of CS components, corrosion product transport and consequently SG behaviour and associated maintenance.

• A negative impact on operating costs, not particularly due to reagent costs but mainly to regeneration or replacement frequencies of Ion Exchange Resins (IER) from purification systems.

• A negative impact on liquid effluents associated with the reagent direct release into environment or these induced by regeneration of solid wasted from IER replacements.

These most crucial examples demonstrate the need to better take into account the various aspects, including economical and environmental ones and not only the material behaviour. A specific emphasis is given in this report to these problems since more and more NPP have new design or upgraded components. Such NPP do not require the same options as in the past when only material behaviour was considered.

The environmental impact is also related to the global warming and increased industrial activities with consequences such as always higher temperatures in the river water. When regulatory limits are achieved, it may be necessary, in extreme situation, to decrease for example the BD flow rate (if the BD coolers are cooled by river water) so as to minimize the river water temperature increase that may become unacceptable, even in presence of cooling towers. A compromise between material behaviour and environmental consequences has to be adequately balanced which was rarely the case in the past.

Another interesting way of minimizing liquid effluents release into the environment is to select physical methods for make-up water preparation. This is described in Section 6 and includes methods such as Reverse Osmosis and Electro Deionization, already implemented in Sweden. In addition, this has an economical advantage.


1-17(1-18)



1-18(1-18)

In conclusion of theses environmental and economical aspects, serious consideration should be given to take into account the use of more resistant materials (tight condensers with titanium tubing, SGs with Alloy 690 TT or Alloy 800 NG tubing resistant to corrosion under reasonable chemistry conditions, larger use of SSs in the secondary systems) when adjusting the secondary water chemistry options. Thus, the secondary water treatment as well as the control of impurities and operating options should be decided keeping in mind the material integrity but without forgetting economical and environmental aspects.


2 Description of the behaviour of the impurities and chemical species in the secondary side

2.1 Impurities inside the steam generator

Chemical species (pH control agents, corrosion products) salt impurities (e.g. condenser leaks) and organics, which are transported by FW into SGs, concentrate there more or less depending on their volatility due to steaming or they may deposit / be adsorbed beneath deposits at different locations, like at TTS and TSP crevices. The majority of the FW corrosion products (usually dissolved iron) deposit and build magnetite scales on the heat transfer tube surfaces causing tube fouling, as explained in Section 1.2. Besides building tube scales, they also accumulate within the heat transfer crevices (tube to TSP crevices and at TTS) creating blocked crevices. Approximately 5-10% of the FW corrosion products accumulate at TTS building hard caked deposits. These blocked heat transfer crevices and TTS hard deposits are the special locations where the FW salt impurities preferentially concentrate and may cause tube corrosion. The building of these crucial SG deposits and the concentration mechanism of the salt impurities will be explained in the next sections.

2.1.1 Steam generator crevice and top of tube sheet deposits

At high SG operating temperatures solubility of dissolved FW corrosion products (iron) decreases with increasing temperatures (retrograde solubility or reverse solubility). Therefore they precipitate and build deposits at heat transfer surfaces especially in flow-restricted areas like at TTS or in Tube to TSP crevices. Also suspended corrosion products (magnetite) deposit on such heat transfer surfaces due to their surface charge, as described in Section 1.2. Evolution of such deposits at TTS in a RSG is schematically shown in the Figure 2-1. At initial conditions, just after starting the commercial operation of the RSGs, the TTS area and the tube to TS crevices at the top of TS are free of deposits. But very soon in early operating period FW iron transported into SGs start to build thin deposit layers covering the TS crevices. Such a deposit at the TTS is called sludge. If these deposits are not removed e.g. by annual Tube Sheet Lancing (TSL) they grow quickly during later operation and build thick hard caked sludge usually at the center of tube bundle within several years (based on experience within 3-4 years). Beneath these deposits, circulating bulk water cannot properly cool the heat transfer tubes, which results in temperature increase of the tube surface within the deposit. This temperature increase over the secondary side bulk water temperature is called “super heating”. In such super heated deposits salt impurities concentrated preferentially (Section 2.1.2). These hard deposits with concentrated impurities, which cannot be removed with recirculating flows and which therefore remain at all times on tube surfaces, are of potential risk for tube corrosion.

Figure 2-1: Accumulation of deposits on TTS with time [Staehle & Gorman, 2002].


2-1(2-28)


Similar corrosion product deposition occurs also within the tube to TSP crevices. As an example, progressive accumulation of deposits in a drilled hole TSP is schematically given in Figure 2-2.

Figure 2-2: Progressive accumulation of deposits in an eccentric drilled hole TSP [Staehle & Gorman, 2002].

TSP crevices with circular drilled holes have an average gap size of ~ 0 - 200μm between tubes and TSPs, where usually tubes are eccentrically located. Deposition of the crevices starts near the contact location of the tubes, where the highest superheating is expected as shown in Figure 2-2 (b) and progresses very rapidly (c) filling the whole crevices. Within a short operation time of few years these crevices can get fully blocked if they are not cleaned, as usually. These blocked heat transfer crevices cannot be cooled by bulk water; therefore the superheating at the tube secondary side surface can increase up to primary temperature. Again such heat transfer crevices are the crucial locations for the preferential impurity concentrations causing SG tube corrosion. A photograph of such a drilled hole TSP crevice, pulled from a retired SG is given in Figure 2-3 as an example (the tube was very tightly stuck with hard deposits in this TSP sample, so that it could even not be removed by pulling mechanically).

Figure 2-3: Example of drilled circular hole TSP crevice, pulled from a retired Westinghouse RSG [Bjornkvist, 1994].


2-2(2-28)


The design of CS drilled hole TSPs were modified to SS line contact supports, like “Broached hole TSPs” or “egg crate supports”, in order to minimize the accumulation of deposits in the support constructions and provide proper tube cooling, which prevents superheating as accelerating or driving force for impurity concentration causing tube corrosion (Section 4). Even with these improved design of tube supports corrosion product deposits occur at tube line contacts as shown in the Figure 2-4.

Figure 2-4: Schematic view of deposit accumulation at line contact crevices associated with broached holes (a) and egg crates (b), [Staehle & Gorman, 2002].

Deposits were experienced at line contact of broached hole TSPs (see Section 1.1.3 and 5) and of egg crates, as indicated in Figure 2-4. However, up to date, despite the occurrence of accumulations as indicated in Figure 2-4a and b, there is no evidence of any serious tube corrosion at these line contacts, except in the case of intermediate design with CS broached hole or egg crate supports and Alloy 600 MA tubing (see Section 4). But, in some egg crate support design, where the egg crate bars are located at the one and same level (i.e. the tube has two line contacts at obtuse angle area), in that case, at the small volume region associated with the obtuse angle of egg crate intersection (Figure 2-4b) deposits conditions can be expected at limited area similar to drilled hole TSP crevices.

Also at the deposit filled contact area of broached hole TSP crevices superheating in some extent can be expected as indicated in Figure 2-5.


2-3(2-28)


Figure 2-5: Different temperatures at open and at deposit filled contact area of broached hole TSP [Staehle & Gorman, 2002].

2.1.2 Concentration process of various species hide out phenomenon

Soluble species, like salt impurities organics etc., which are transported by FW into RSGs concentrate in the bulk water due to steaming. The concentration factor of these species depends on their volatility, i.e. their distribution coefficient between steam and water. In order to limit the concentration mechanism in the bulk water SGBD system is designed, which is in service with a typical BD rate of 0.5 to 1% of the FW flow rate. Many of the impurities, like NaCl, NaOH etc as main SG contaminants are more or less soluble in bulk water but have very small solubility in steam. Their steam to water distribution coefficient is in the range of 10-6 to 10-5 at the RSG operating temperatures (see Figure 2-6). For these non-volatile species, the expected concentration factor is ~ 200 and 100 in the RSG bulk water for the SGBD rate of 0.5% and 1% respectively. But based on experience, the observed SG bulk water impurity concentrations are usually remarkably less than the expected ones according to these concentration factors. This is because the impurities disappear from the bulk water by adsorption and/or by concentration mechanism beneath the deposits. This phenomenon is called “Hide Out” (HO). HO phenomenon is observed mainly during plant start-up operation, especially when power escalation exceeds 30% power. During normal power operation, HO phenomenon cannot be recognized easily due to very low impurity concentrations in the FW and in the RSG bulk water.

Figure 2-6: Steam to water distribution coefficient of salts as a function of temperature and pressure [Jonas, 1977].


2-4(2-28)


Soluble species in the RSG bulk water can hideout by following competing mechanisms (see Figure 2-7):

• Adsorption on tube surfaces or surface oxide deposits (Section 2.1.2.1).

• Concentration in locally flow restricted geometries like TSP crevices or TTS deposits (Section 2.1.2.2).

• Concentration in tube surface deposit crevices (Section 2.1.2.3).

Adsorption on tube oxide surfaces 2.1.2.1

Some soluble species, particularly polyvalent anions like sulfate and phosphate have affinity to adsorb on metal oxide (magnetite) surfaces. Boiling as driving force is not required. This means adsorption and desorption will occur during heat-up as well as during power operation. Once the oxide surface at a given condition is saturated, additional adsorption does not occur, unless the bulk water concentration increases.

Figure 2-7: Competing hideout mechanism [EPRI et al, 1995].

Concentration in flow restricted geometries 2.1.2.2

Soluble impurities in bulk water concentrate preferentially in flow restricted areas, where water flow in/out is restricted by densely packed deposits, such as in tube to TS and tube to TSP crevices. The driving force for the impurity concentration in such flow restricted geometries is the heat transfer with super heating and the boiling. Local super heat in crevices is defined as temperature difference between the crevice and the bulk water:

ΔT = TCr - TBW where TCr = Local temperature in the crevice TBW = Secondary side SG bulk water temperature


2-5(2-28)


3 Historical evolution of secondary water chemistry in the past during power operation

3.1 Rationale for chemistry evolution

Historically, based on fossil fired unit experience, most of the PWR units were using the Phosphate (PO4)treatment, since such a treatment is able to buffer contaminants entering the system and concentrating in the SG. The reason for this is that condenser tubes were historically made of copper alloys, for their high thermal conductivity, but were not always tight.

This PO4 treatment, described in Section 3.3.1 was thus selected to buffer (neutralize) any acidity or alkalinity coming from contaminations, mainly cooling water in the condenser.

Besides this treatment, some plants were operated on AVT (ammonia treatment) which was the case of the old French PWR of Chooz A, 300 MWe, in the 1960’s, which has SS SG tubing. But most of the PWR up to 1974 were operated with PO4treatment.

In the early 1970’s, under PO4 treatment, Alloy 600 MA started to suffer from corrosion. It has been first considered that locally, the chemistry was either too alkaline or too acidic respectively inducing SCC or wastage (thinning) of the SG tubing. Several Na/PO4 ratios have been successively tried to avoid both type of corrosion but without success. In fact even both environments were able to coexist in different local parts of the SG and it became obvious that it was impossible to avoid corrosion of Alloy 600 MA tubes, whatever the Na/PO4 molar ratio that was in the range of 2.0 to 2.6.

It must be mentioned that even under PO4 treatment, some volatile reagent (hydrazine that thermally decomposed to ammonia) is also added to get the desired pH in the whole secondary system, PO4 being added only to keep the expected buffered chemistry within the SG.

In 1974, Westinghouse decided to move to AVT i.e. without any solid additive that would not be volatile.

Most of the PWR with Alloy 600 MA tubing effectively moved to AVT chemistry, with the exception of a few old units of limited power that did not have encountered major problems and did not want to take any unnecessarily risk and have the burden of changing their chemistry (this was for example the case of Zorita in Spain and H.B. Robinson in USA).

But at the time of PO4 treatment, utilities were used to operate with rather high impurity levels, much higher than what they are now, particularly for seawater cooled plants. This has been the cause for the “denting” phenomenon (Section 5). But the AVT chemistry, without any buffering effect, was unable to neutralize the acidity or alkalinity of cooling water ingresses or other pollutions, when they concentrate in the SGs. This denting phenomenon induced extremely quick and severe degradation, requiring urgent Steam Generator Replacements (SGR) in several PWR units in the USA, particularly high for seawater cooled units.

Consequently, much more stringent specifications have been progressively applied in various countries having SG of the Westinghouse and/or Combustion Engineering (CE) design with drilled circular holes in CS TSP. This together with BAT (explained in the following) has been able to mitigate denting in remaining American SGs in service and to avoid it in Japan. In some other countries with similar weak design, the improved chemistry allowed to avoid denting (France, Switzerland, and Sweden).

Despite the application of such chemistry specifications, SCC appeared under AVT on the secondary side of Alloy 600 MA SG tubing in many countries, starting in the early 1980’s (Japan 1981 for the first evidence of such cracking).


3-1(3-27)


Since beginning of 1970s the PWR secondary side chemistry was modified worldwide significantly due to experienced SG corrosion problems (see Section 1). The reasons for these SG degradation problems were the inadequate SG design and selected tubing material, as well as the applied secondary side chemistry. Based on lessons learned the SG design and materials as well as secondary side chemistry were modified continuously in order to prevent or mitigate these SG problems.

In the late 1980’s and early 1990’s, Outer Diameter-Inter Granular Attack / Stress Corrosion Cracking (OD-IGA/SCC) was still growing in many units where it had already started and was appearing in many others. Only a few units were remaining free of such corrosion at TSP or sludge pile level on top of TS. Additional remedies in chemistry, besides lowering the specification limits in impurities that reached the lowest feasible values, have been decided on units with Alloy 600 MA

After the 2000s, when most of SG with Alloy 600 MA were replaced or scheduled to be replaced on the medium term, the main focus of chemistry evolution was related to mitigation of FAC, corrosion transport and deposition within the SG. This means that evolution were more to optimize the use of alkaline reagent that continuing to permanently decrease the level of impurities entering the SG, that was already very low in most countries.

It could have been considered, with the progressive replacement of SG and the use of much more resistant alloy, Alloy 690 TT in many countries that there could be relaxation of the chemistry specification. This hope was based on what was applied at units with the German design SGs (Alloy 800 NG tubing) or the VVER Russian plants with a less resistant alloy but horizontal SGs without TS and lower operating temperatures, where much less stringent chemistry specification is used and better corrosion results than SG with Alloy 600 MA experienced.

Nevertheless, many people, used to be frighten by inadequate chemistries and poor performances of Alloy 600 MA, did not think to apply less severe specification on units with Alloy 690 TT although this alloy was at least as resistant as Alloy 800 NG or more resistance than Russian Alloy where no major degradation had been observed with much less severe chemistry specifications.

The reasons are:

• The bad experience feedback with Alloy 600 MA making the utilities very cautious.

• The interpretation by some authors of IGA/SCC occurring due to lead instead of Alloy sensitivity and thus considering that Alloy 690 TT being also sensitive to lead corrosion, no specification relaxation should be applied.

• Taking the benefit of applying a good chemistry to increase the safety margin with Alloy 690 TT.

On the contrary, some authors [Nordmann, 2004] considered that keeping for Alloy 690 TT the high severity of Chemistry specification progressively developed for Alloy 600 MA has no rationale and that the application of more realistic specification, comparable to what is applied in PWR of German design was stringent enough and would allow several benefits, such as:

• Higher operating flexibility.

• Higher availability (in case of very small impurity ingress, if thermal release restriction in the river requiring a lower BD rate, shorter start up, etc).

• Lower operating and investment costs (less use of condensate polishers and less frequent IER regeneration or replacement).

• Lower environmental consequences by decreasing the quantity of liquid and solid wastes with less IER regeneration or replacement and more recycling of some systems (like incondensable parts of condenser).


3-2(3-27)


Finally, EPRI decided in 2008 to apply slightly less severe guidelines for Alloys 600 TT and even less severe for 690 TT as compared to Alloy 600 MA. Among other utilities with these various types of nickel base alloy SG Tubing, several are more or less applying EPRI guidelines. All the US utilities have to apply them. French ones, less restrictive than those of EPRI are considering the situation to avoid any risk of increasing by harmful deposits, the last problems of Flow Induced Vibration (FIV) that occurred in some units.

The approach for the water chemistry modification and the selected strategies was different in various countries and based on plant specific situations. In the following, the historical evolution of the water chemistry strategies will be explained in different countries, focusing mainly on those where these chemistry modifications were developed.

3.2 Chemistry evolution for various countries

3.2.1 Chemistry evolution in the USA

USA is the country with the earliest PWR operating experience. The first commercial PWR, Haddam Neck went into service in early 1968 and more than half of the now operating 70 PWRs started with their commercial operations in 1970s. Accordingly, the first SG degradation problems caused by selected inadequate SG TSP design and tubing material Alloy 600 MA were experienced in USA PWRs. In parallel to improvements made in SG design and tubing material for new replacement SGs or for new plants a big effort was performed in 1970s and 1980s to continuously modify the SG water chemistry with the objective to mitigate the degradation problems of already operating old design SGs. These modifications of secondary side chemistry were a continuous process because in some cases the change of water chemistry conditions brought another SG degradation phenomenon. In other countries, water chemistry modifications were also necessary to mitigate the SG degradation problems, as it is explained in the following. However, they were not so significant or drastically because they could benefit from the lessons learned from USA PWR operating experience.

The secondary side water chemistry evolution in USA is shown in Figure 3-1. Based on the experience gained with fossil power plants, the US PWR operators started with PO4 chemistry in the SGs and low AVT (All Volatile Chemistry with ammonia + hydrazine) on the entire secondary side.

Figure 3-1: History of secondary side water chemistry in US PWRs [Fruzzetti & Wood, 2006].


3-3(3-27)


4 Design and materials used in the secondary side

In this section it will be described how the selected design and materials in SGs and in the entire secondary side can demand the selection of secondary water chemistry and influence the plant performance. Special emphasis will be given to the SG design and especially to its tubing material. In the following sections, these issues will be described for the PWR and VVER plants.

4.1 PWR steam generators

4.1.1 Steam generator design and materials

In PWRs, two types of SGs are used: The most use one is the Recirculating SGs (RSG), designed by Westinghouse and its licensees (MHI, Framatome), CE (now part of Westinghouse) and Siemens –KWU (now AREVA NP GmbH). The other type is the Once Through SGs (OTSG) designed by Babcock & Wilcox (B&W)(now part of AREVA NP). Schematic view of a typical RSG designed by Westinghouse (Model D3) together with OTSG is shown in Figure 4-1. The other RSGs designed by CE and Siemens - KWU are shown in Figure 4-2 and Figure 4-3 respectively.

Figure 4-1: Schematic view of (a) RSG (Westinghouse Model D3) and (b) OTSG (B & W) [Staehle & Gorman, 2001].


4-1(4-37)


Figure 4-2: Schematic view of recirculation SG designed by CE.

Figure 4-3: Schematic view of the RSGs (a) with and (b) without economizer designed by Siemens-KWU [Schwarz, 2009].


4-2(4-37)


All these PWR SGs have vertical tube bundles (vertical SGs), on contrary to VVER plants, where horizontal recirculating SGs are used (see Section 4.2). Several early PWRs SGs, e.g. Shippingport, the N-Reactor, and Indian Point Unit 1, have used horizontal tube bundles. The advantage of the horizontal RSG is the absence of TS, where deposits accumulate and corrosive media may occur. But this arrangement needs more floor space and thus requires larger reactor buildings with higher costs. Therefore, all modern RSGs have vertically oriented tube bundles.

Most of the PWR SGs are recirculating SGs. The initial reason to selecting a recirculating design was to avoid boiling of water until complete evaporation in the SGs, where rapid accumulation of deposits and high local impurity concentrations can be produced resulting in accelerated tube corrosion. However, in the 1960s satisfactorily performance of the fossil fired once through boilers of conventional power plants was experienced, when water chemistry is stringently controlled. In these boilers, boiling occurs on the inner surface of the tubes and they do not have any heat transfer crevices. Based on this experience, B&W developed OTSGs for PWRs, which with minimized crevices and strictly controlled water chemistry, very similar to that used in fossil power plants. OTSGs have two TSs, one at the top the other at the bottom of the SG (see Figure 4-1b). FW enters the annulus above the 9th TSP level and is preheated by primary coolant to saturation on the way down to bottom TS, and enters then into the tube bundle almas steam. This steam is then superheated within the tube bundle above 8

ost th TSP. At OTSG steam

outlet it is superheated more than 30 °C (see Figure 2-13). In another words, OTSG has no water inside of the SG. Accordingly, a BD system is also not considered to limit the impurity concentration. Therefore, this type of SGs requires strictly controlled water chemistry, and an absence of impurities. Experience has shown that OTSG have improved tube corrosion performance when compared to Westinghouse or CE designed RSGs of the same vintage; but that they are not immune to tube corrosion problems. The tracer of FW salt impurities was concentrated on the tube surfaces in the superheated regions, causing Intergranular corrosion year’s later [Sherburne, 1998] (see Figure 4-4).

Figure 4-4: OTSG experience with respect to tube corrosion at Oconee unit 1: (a): tubes plugged vs time; (b): Comparison of total plugged tubes with those plugged for SCC in the free span [Sherburne, 1998].


4-3(4-37)


5 Encountered SG degradation and influence of various impurities on SG performance

As explained in Section 3, the various types of degradation encountered in SG are directly related to their design, particularly the tubing material and the TSP. Obviously, the type of corrosion depends on the chemical environment (acidic, neutral or alkaline, oxidizing or reducing) the chemical treatment (PO4 or AVT) and clearly on the type and concentration of the various impurities that are described in this section.

Figure 5-1 shows the various types of degradation that may occur on Recirculating SG, both on primary and secondary side of SG tubing, in the case of Alloy 600 with AVT chemistry.

Figure 5-1: Degradation on Alloy 600 in recirculating SG under AVT treatment.

Wastage was particularly present with the PO4 treatment. When switching to AVT treatment and in presence of a specific design of TSP, denting occurred, mainly in acidic environment (seawater cooled plants). Pitting occasionally occurred in a few plants under specific polluting environment.

The most severe type of corrosion has been IGA/SCC of Alloy 600 MA that will remain the major issue on the secondary side of units up to complete elimination of this alloy by SG replacement. Although this degradation is hardly avoidable, even with the most severe chemistry specification, it can at least be slowed down and is mainly affecting the units with an alkaline environment (river cooling water, condensate polishers) or with high level of some pollution such as lead.

In this section, as detailed in Sections 2 and 6, it will be explained the influence of deposits with corrosion products on the various types of degradation. This is even more important for insuring a long and safe duration of the SGs with sufficiently resistant tubing (Alloy 800, 18Cr-10Ni, Alloy 690 TT, and hopefully Alloy 600 TT on the secondary side).


5-1(5-40)


5.1 SG degradation problems

SG tubing materials Alloy 600 MA, Alloy 600 TT and Alloy 690 TT are nickel based alloys whereas Alloy 800 NG and 18%Cr-10%Ni are iron based alloys. However they all have nickel, chromium and iron as main constituents. Under operating alkaline and reducing conditions these metals are more stable as oxides than as metals. In Figure 5-2 stability of nickel, chromium and iron is shown under SG operating conditions. On free span tube surfaces, the pH is about 6 – 6.5 depending on the alkalizing agent used and the potential is slightly above hydrogen line for fully deaerated AVT conditions without any other oxidants such as copper. The stable form of the constituents of the tubing alloy is the nickel oxide, chromium oxide and magnetite. This means that a thin oxide layer (nickel-chromium ferrites) is built on the surface of the tubing material, which protects the tubing alloy against corrosion. If somehow these protective oxide layers are destroyed, alloy losses its protection and can corrode. The following main water chemistry related factors can destroy the protective oxide layers and cause SG tubing corrosion:

• pH: Extremely high or low local pH values can dissolve the oxide layers and accelerate different types of corrosion.

• Electro Chemical Potential (ECP): ECP is a measure of the strength of the oxidizing or reducing conditions. Local ECP affects strongly the occurrence and rate of the corrosion.

• Specific species: Certain specific impurities, like sulphur, chlorides, lead, etc, accelerate tubing corrosion either indirectly by changing the local pH and ECP in the crevices or beneath deposits or they directly react with the protective oxide layers on the tubing surface. This will be explained later in Section 5.2.

Figure 5-2: Potential-pH (Pourbaix) Diagram for nickel-, chromium-, and iron-water at 288 °C.

Such extreme pH excursions are usually caused by concentrated impurities in flow restricted heat transfer crevices or beneath deposits (see Section 2). Depending on existing local environmental chemistry conditions (pH and ECP), different type of SG tube corrosion can occur in those areas (see Figure 5-3). If instability of the oxide layer occurs in a large area, it results in general corrosion or wastage. If localized breakthrough of the oxide layer happens, pitting can occur at that location. In the presence of stresses, SCC is usually occurring.


5-2(5-40)


Figure 5-3: Different type of crevice corrosion [EPRI, 1980-1990] and [Millet, 1995].

In the following sections such different type of SG tube corrosion degradations are explained.

5.1.1 Wastage

The wastage corrosion is the tube thinning beneath the corrosion product deposits. The pulled tube examinations confirmed that wastage is caused by concentrated PO4s within these deposits (see Section 3.3.1). Wastage corrosion is experienced in the past at PWR SGs where PO4 chemistry was applied. Because of the retrograde solubility behaviour of PO4s, they precipitSG operating temperature beneath deposits due to concentration mechanism, creating acidconcentrated PO

ate at ic

2 re 5-5).

VT.

4 medium. This concentrated PO4 damages the protective oxide layers on the tube surface, which results in tube thinning by dissolution of the tubing material (see Figure 5-4). Field experience confirmed that wastage corrosion rate increases with temperature. At older Siemens designed PWRs, which were operating with PO4 chemistry in their SGs, wastage was experienced in coldest SGs after 7 to 8 years of operation, whereas in hottest SGs it took only years (see Figu

In some PWR, the Na/PO4 ratio was adjusted to avoid the presence of acidic environment, but this has been at the origin of corrosion under alkaline environment (IGA/SCC of Alloy 600 MA, see Section 5.1.4 since it was impossible to perfectly control the environment beneath deposits in all the parts of the SG tubing. As a consequence of wastage and/or IGA/SCC, in mid seventies many PWR plants stopped using PO4 Chemistry in their SGs, and switched to A

Nowadays, wastage is not any more a problem under AVT chemistry.


5-3(5-40)


6 Adequate water chemistry for improved SG performance

As discussed in Section 1 and 2, the root cause of all SG degradation problems is the concentrated impurities and the corrosion product deposits. The oxidizing conditions in the SGs accelerate the degradation problems. Accordingly, to improve the SG performance, this chemistry related parameters, corrosion products, impurities and the oxidizing conditions need to be controlled adequately. From these three chemistry parameters, probably the corrosion products is the most important one, because they create in SGs conditions for impurity concentration, responsible for corrosion, as well as tube deposits, which cause thermal degradations. The main source of these corrosion products is the FAC of the CSs in the secondary side. Therefore it is appropriate to discuss the FAC mechanism first before discussing how it can be controlled by chemistry measures.

6.1 Flow Accelerated Corrosion (FAC)

Based on field experience, the main corrosion product source is the EC, also called “Flow Accelerated Corrosion” (FAC) of CSs during the power operation. FAC is a degradation process that affects CSs and in a less extent low-alloy CSs under specific plant conditions of temperatures, flow rates, and water chemistry. Unfortunately these specific conditions often exist in PWR and VVER plants, and CSs are used in various components almost in the entire secondary systems. FAC occurs under high flow rates, if the protective oxide layers on the surface of CS components and pipes cannot be built. The protective oxide layers are built on the CS surface by the reaction of iron ions, which are dissolved from the metal, with water at high temperatures (see left hand side of Figure 6-1). If these protective layers dissolve in an iron unsaturated fluid medium at the metal-fluid interface, or if iron ions released from the CS surfaces are immediately removed by a high flow (see right hand side of Figure 6-1), the protective layers cannot be built, and this results in the FAC degradation of CSs.

Figure 6-1: Schematic description of protective layer built-up on metal surfaces and FAC mechanism at high flow conditions.

FAC can occur in both, one-phase flows, like in condensate and FW systems, and also in two-phase (wet steam) flows. Visual pattern of FAC is different depending on under which flow condition it was caused. FAC caused by one-phase flow shows “horse shoe pits” whereas it has “tiger stripping” pattern if it is produced under two-phase flow conditions in wet steam (see Figure 6-2) [Kastner & Heitmann, 1982].


6-1(6-44)


Figure 6-2: FAC degradation pattern under two-phase and one-phase flow conditions.

Single phase FACs are the published well known events because they caused in some plants severe accidents, which resulted in severe injuries and even death of plant personnel. The first published big FAC accident in the nuclear industry, a break of condensate line, occurred in Surry-2 in 1986 (see Figure 6-3).

Figure 6-3: FAC accident at Surry-2 PWR in 1986.


6-2(6-44)


The second severe FAC accident happened at Mihama 3 in August 2004; again due to a rupture of condensate pipe downstream of an orifice, causing the death of five plant personnel again. It was a pity that this accident occurred only a few days before the shutdown for outage where the inspection of the pipe should have taken place and when a large staff was present in the turbine hall ready for the outage. The ruptured pipe was located behind an orifice plate. The flow-accelerated-corrosion attack caused a local reduction of the original wall thickness from 10 mm down to 1.4 mm within the 20 years of operation including 18 years at low pH with ammonia treatment (see Figure 6-4).

Figure 6-4: FAC accident at Mihama-3 PWR in 2004.

These two severe FAC accidents are not the only FAC damages experienced in the PWR plants. Some other examples, including the CS SG TSP damage, are given in Figure 6-5.

Similar severe FAC events were also experienced in VVER plants: In Loviisa unit 1 and in Balakovo unit 2 FW line rupture happened in 1990 and 2004 respectively (see Figure 6-6).

Figure 6-5: Some examples of FAC damage at PWR plants [Turner, 2007].


6-3(6-44)


7 Different water chemistry strategies (state of art)

Field experience with SGs in the 1970s and 1980s and their root cause confirmed that the concentrated impurities beneath the corrosion product deposits on the SG tube surfaces together with SG tubing sensitivity is the main reason for the SG tube degradation problems (See previous sections, especially Section 2). Accordingly, industry started to develop chemistry strategies to counteract these degradation problems (see Section 3). Many of these strategies were unfortunately not good enough to stop or avoid the SG degradation problems (see Section 3); because they were trying to improve only the chemistry environmental conditions in the SGs without sufficiently considering the necessary improvement in FW corrosion product control. As of today, the industry clearly understood that adequate corrosion product control is indispensable for improved SG performances. The first chemistry strategy, addressing this issue, was introduced in mid of 1980s by Siemens-KWU; this is the “High pH Ammonia AVT” chemistry, called High-AVT. This was followed by introducing the “Advanced Amine Chemistry” developed by EPRI and various utilities in the world. At the beginning of 2000s, partial oxygen injections were introduced in several PWRs, to counteract local FAC problems occurring under extremely high flow conditions. Since recently, use of dispersants developed by EPRI, is introduced as a trial in several US PWRs, to increase the removal of SG corrosion products by the BD system. In the following sections of this section, these state of art chemistry strategies and their effectiveness will be described. Also, for the selection of the chemistry regimes (treatment), their relation with Investment and operating costs, as well as the issues of associated release of liquid effluents to environment and solid wastes will be discussed.

7.1 High pH operation (High-AVT)

High AVT is high pH ammonia chemistry, introduced by Siemens-KWU at the beginning of the 1980s. The objective of pH increasing was to stop the FAC in two phase flow steam systems like the area of Cross-under Line, MSR, MSR-Drains, and Steam Extraction Lines for an improved FW corrosion product control (see Figure 6-7, Section 6.1.3.1 and Section 6.2). This chemistry is using ammonia for adjusting the necessary FW pH sufficiently high to stop the FAC in the steam systems. The source of the ammonia is the hydrazine, which starts to thermally decompose at higher temperatures above 180 °C producing ammonia (see Figure 3-9). Hydrazine is injected continuously in the condensate system for oxygen control; and the excess of hydrazine decomposes thermally to produce the ammonia. Plant specific FW chemistry concentrations are in the range of:

• Ammonia: 5 – 15 mg/kg (ppm9).

• Hydrazine: 20 – 150 μg/kg (ppb).

• pH25 values: > 9.8 up to 10.1.

• Specific conductivity values: 15 – 35 μS/cm.

To achieve these chemistry conditions, only hydrazine is injected. Additional ammonia injection is not necessary in Siemens-KWU designed plants, because the secondary side is very leak tight; accordingly the capacity of the Condenser Air Removal is not designed too big. These results in relatively small ammonia sink in the secondary side. Therefore, the continuous ammonia removal from the Secondary Side by SGBD and Condenser Air Removal systems can be balanced only by thermal decomposition of the hydrazine excess to achieve a FW pH25 value of at least 9.8. For the control of the chemistry program, on-line monitoring facilities are considered in SGBD, FW. In addition, for impurity control, cation conductivity cells are designed almost in all systems and the condenser leak monitoring system consists of sodium and cation conductivity measurements. The concept of High AVT chemistry in the Siemens-KWU designed plants is given in Figure 7-1.

9 parts per million (mg/kg)


7-1(7-46)


High AVT chemistry is not applied only in Siemens-KWU designed plants, but also in many other plants with Siemens-KWU designed Replacement SGs in Europe and also in several plants in Japan and Korea and even in some VVER units. At some of these plants, it was not possible to adjust the necessary FW pH25 only with hydrazine injection, due to very high ammonia sink in the condenser caused by high capacity air removal system. At these plants, it was necessary to inject ammonia in addition to hydrazine.

Figure 7-1: Concept of High-AVT Chemistry application in Siemens-KWU designed plants.

Water chemistry specifications for the High-AVT chemistry used in the Siemens-KWU designed plants and/or replacement SGs is called VGB10 Guidelines and that is explained in Section 11. This VGB water chemistry guideline is established based on the field experience of Siemens-KWU designed plants and reflects the High-AVT chemistry operating practice of more than 25 years. FW pH25 of > 9.8 is specified for the corrosion product control, because this FW pH value is sufficient to remarkably suppress the FAC in two phase flow steam systems, if Low Chromium Alloyed Steels with 2.5% Cr content are used in the FAC endangered areas of the steam systems like in Siemens-KWU designed plants. Figure 7-2 shows decreasing FW iron concentration with increasing FW pH25. Remarkable reduction of FW iron concentration starts above pH25 of 9.8. The Siemens-KWU designed plants are operating with FW pH25 of 9.9 – 10.1; and the FW iron concentrations of these plants are in the range of 1 to 0.5 μg/kg (see Figure 7-3 as an example).

10 Technische Vereinigung der Großkraftwerkbetreiber e.V. (Large Power Plant German Owners Group).


7-2(7-46)


Figure 7-2: FW iron concentration as a function of pH increase at Neckarwestheim Unit 1 [Odar, 2005].

Figure 7-3: FW iron concentration as a function of pH increase at Gösgen NPP [Odar, 2005].


7-3(7-46)


8 Different strategies of using purification systems

With the evolution of corrosion and other issues in the secondary system, the purification options have largely evolved.

In the past, the key requirements were the elimination of:

• Impurities coming from raw water condenser leaks, by the use of CPSs.

• Corrosion products by condensate polishing plants (powder resin or deep beds).

• SG bulk water concentration decrease by SGBD.

But many units were operating at low pH, which was easily compatible with the Ion Exchange Resin (IER) regeneration frequency. This is important for SGBD but even more for CPS due to the higher treated flow rate.

It must be mentioned that there are new possibilities with physical methods of purifying the water without using IER, thus with less environmental constraints related to regeneration. This will be also discussed, based on the Swedish experience at Ringhals plant.

Impurities are efficiently eliminated by SGBD since it is the location, where most impurities are present at the highest concentration. Thus, it is more efficient, for a defined flow rate, limited for thermal efficiency reasons, to purify SGBD rather than other parts of the system such as condensate water. Furthermore, it is technically easier to eliminate most of the impurities (e.g. 80%) of the water when they are more concentrated in SGBD rather than in a point very close to ultra pure water like in condensate water. This difference has a considerable impact on cost, efficiency, friendly environmental impact.

On the opposite, purification through the condensate polishing plant allows to eliminate potential impurities before they enter into the SG, where they may concentrate and induce corrosion phenomena.

The decision for purification systems selection and operating mode depends on various criteria:

• Coping with the different materials.

• Efficiency of purification.

• Liquid and solid effluents releases into the environment.

• Pollution risks.

• Reactivity in case of urgent need.

• Investment and operating costs.

Both purification systems (CPS and SGBD) will be discussed in this section.

An important issue in IER is their potential regeneration:

• The option of operating the cation resin (either as a cation bed or within a mixed bed) after the amine breakthrough avoids IER regeneration with all the associated burden described below. When the water quality at resin outlet becomes insufficient, the IER is just discarded and replaced by a new one.

• Replacing the resin by a new NG resin instead of regenerating the used resin may potentially better quality at resin outlet since this IER be manufactured with the desired quality (low impurities and high level of regeneration and rinsing during the manufacturing final step).


8-1(8-16)


The regeneration burdens are:

• Use of highly concentrated regenerants (acid or caustic used for regeneration) with safety issues, risk of pollution.

• Production of liquid wastes, needed to be neutralized before release to the environment.

• Need of resin transfer inducing some deterioration and resin fines pollution that may be more or less eliminated by filtration.

• Possible cross contamination of resins during regeneration, which will cause some higher content of impurities at resin outlet.

• The regeneration should be perfectly controlled to avoid pollution by regeneration reagents (acid, base).

The regeneration should preferably be carried out in a location external to the resin vessels where the resins are transferred for this external regeneration, thus minimizing the risk of pollution; in case of internal regeneration, additional precaution should be taken.

When separating the cation and anion resins of the mixed bed, this should be carefully done to avoid the presence of cation resin in the anion and vice versa (cross contamination) during regeneration, otherwise the undesirable resin would be polluted by the reagent during regeneration and then bring a significant impurity level (throw). There are several ways for avoiding such a cross contamination:

• Use of a TriobedTM with an inert resin layer of intermediate density between cation and anion resins; only a small quantity of inert resin will be present in the each of the resin to be regenerated and will not bring any pollution.

• Remove the intermediate part of insufficiently separated resins, around the separation line between the anion and cation resin and keep this layer away from the regeneration process to avoid its contamination by the reagent. To limit the amount of solid waste, this portion of the resins can be added to the exhausted resins for separation before the next cycle regeneration.

Regenerating the IER may allow more options. The benefit and constraints of IER regeneration are different for CPS and SGBD and will be more discussed in each appropriate section.

For SGBD, regeneration permits operating at higher pH and with a wider range of amine selection while for CPS, regeneration is not compatible with a high pH or an amine requiring a high molar concentration. The difference between CPS and SGBD is due to the flow rate, which is much lower on the SGBD and thus compatible with a higher pH because the regeneration frequency and quantity of reagent and liquid wastes will be more acceptable.


8-2(8-16)


8.1 Condensate polishing system: Various options and rationale for selection

A Condensate Polishing System (CPS) is installed for purifying the water coming from the condenser, which, after the FW train, will be the FW for the SG. The CPS is not always necessary and the decision to install it or not and to use it or not depends on various criteria which will be explained here.

It consists in a series of Ion Exchange Resin (IER) beds and/or filters with powdered resins, to eliminate corrosion products and ions present in condensate water i.e. ~55% to 2/3 of the FW. The purpose is to avoid the impurity transportation toward the SG, where they may deposit and concentrate, induce corrosion of SG tubing components, (or even potentially the turbine) or decrease the heat transfer.

The greatest efficiency to purify the FW would require to install the CPS in the final FW, but this may not be considered since the IER cannot operate at high temperature and the thermal balance would be unacceptable.

Generally, there is a cation bed for conditioning reagent elimination followed by a mixed bed (polishing) to better eliminate in a neutral environment the final traces of impurities coming from the condenser in case of leak. Most of CPS includes traditional regenerable resins, called deep beds, but there are, in the USA, some stations with powdered resins (Powdex) to insure an efficient filtration.

Powdex systems, with powdered resins were of interest at the time where the target was mainly to eliminate corrosion products and are used in about ½ of BWR in the USA, but only in a few PWR units. The advantage of Powdex is an easy use and the absence of regeneration (no risk of chemicals handling or contamination). On the opposite, Powdex is unable, on the long term, to cope with cooling water ingress in the condenser for ions elimination, due to its low ion exchange capacity.

The CPS is followed by filters to avoid release of resins fines into the FW and the SG where there would be highly detrimental for SG tubing corrosion due to the presence of sulphur compounds (Section 5.1). Sulphates mainly come from cationic resin oxidation and mechanical deterioration.

There are several advantages and inconveniencies in using CPS. The benefits of having a CPS are:

• To continue operation with a leaking condenser when this is a river water of low or intermediate condenser leak rate or when this is a seawater with low leak rate below the threshold to be localisable (< 0.05 l/h).

• Allows to better deciding on the best moment for a shutdown with the dispatching in case of leak.

• To avoid the large pollution from an important condenser leak to reach the SG; however, this may also be avoided or minimized without a CPS with proper actions; this may be (i) action sheets in control room as they exist in French PWRs for an immediate shutdown in case of large seawater leaks confirmed by condensate water signal before the pollution reaches the SG or (ii) put the leaking part of condenser out of service if the design permits it, as in Siemens plants.

• To partially or totally replace the SGBD purification treatment.

• To allow a quicker start up after shutdown for pollution or for refueling.

• To partially eliminate suspended solids (upstream CPS), although it does not eliminate suspended solids downstream of the CPS which may be much more important than without CPS, due do a lower pH.


8-3(8-16)


9 Different maintenance strategies in relation with chemistry (state of art), costs control

There are several types of maintenance activities in relation with chemistry. The In Service Inspection is not covered here, being out of the scope of the document. This section is describing the various maintenance options that are potentially applied and dealing with corrosion products elimination from the SGs.

It has been seen in other sections that either corrosion products (iron or other oxides) or some compounds having a low solubility (such as calcium, sulphate) may deposit in the SG where they may have an impact on heat transfer performances or corrosion process (Section 6).

Soluble impurities like sodium and chloride that may induce SG tubing corrosion are eliminated:

• In case of significant pollution on a short term basis after power decrease by feed and bleed of the polluted systems (condensate, FW and SG).

• More generally by Hide-Out Return (HOR) also during power decrease either immediately after the pollution occurred or at the end of the fuel cycle (or in other cases of shutdown for other reasons).

The hide out return is not a maintenance activity and is covered by Sections 2 and 11. Preventive methods are described in other sections and are concerning corrosion products minimization by appropriate chemistry treatment (Sections 6 and 7) mitigation of deposits accumulation with dispersant addition (Section 7).

In this section, only insoluble compound elimination by maintenance cleaning activities is covered. These may be divided in two groups:

• Chemical Cleaning (CC) processes with solution aimed at dissolving the corrosion products and other species included in such deposits; this is generally used for the entire SG, and may be applied with the purpose either to remove the tube scales for heat transfer recovery (when the free span of the tubes are covered by thick deposits) or for elimination of deposits from specific areas where they may have adverse effects described in Section 5 (IGA/SCC or other types of corrosion at TSP level, FIV, presence of an pollution such as lead that has been inadvertently introduced in the system).

• Mechanical cleaning processes, with water at HP and velocity, particularly applied to the sludge pile at top of Tubesheet, where corrosion products are depositing by gravity but also at upper bundle area to clean the deposits on TSPs.

It is clear that any maintenance activity should be minimized as much as reasonably possible. This means that it must be applied if there is some need associated to safety issues or if the economical benefit associated to its implementation is higher than its cost. The cost evaluation includes the application itself, time loss on the critical path during the outage and wastes elimination.

In the following sections the SG cleaning activities are described.

9.1 Chemical cleaning processes

Presently, there exist two main application procedures with respect to temperatures for the SG chemical cleaning processes. These are applications at low and high temperatures. Both procedures may have advantages as well as disadvantages for the individual specific applications: The “High-Temperature Processes” need shorter application time as advantage due to fast reaction kinetic; however they are fully in the “critical path” of the plant outage, especially when the primary side plant heat is used. This may become a disadvantage for short annual outages.


9-1(9-18)


On the other side, the “Low Temperature Processes” have longer application duration, which may be a disadvantage. However, they can be applied every time during the annual outage so far it fits to the outage schedule. This may in return be an advantage. The utilities are always considering these aspects of the processes for their decision to select the cleaning process.

In addition to these “Hard or Full Scale Chemical Cleaning” processes which were developed in the past, several organizations started to developed also so called “Soft or Maintenance Chemical Cleaning” (MCC) processes in the 1990s and 2000s. Whereas the ”Hard Cleanings” are applied usually as active cleanings to remove huge amount of deposits and/or hard TTS and TSP crevice deposits, the purpose of “Soft Cleanings” is usually to remove periodically portions of the SG deposits for preventive reasons. Some examples of these soft processes are ASCA (Advanced Scale Conditioning Agents), ASCA & UEC (ASCA supported by Ultrasonic Energy Cleaning), DMT (Deposit Minimization Treatment) and DART (Deposit Accumulation Reduction Treatment).

In the following sections only a short description of the SG chemical cleaning processes will be given. For more detailed information the readers should refer to the LCC5 Special Topic Report “Decontamination and SG chemical cleaning”.

9.1.1 Hard chemical cleaning processes

All Hard Chemical Cleaning (HCC) processes are usually using two different solvents for removing iron and copper deposits. This is because, under the reducing operating conditions of PWR and VVER plants, the SG deposits mainly consist of magnetite and certain amount of metallic copper, if copper bearing materials are used in the secondary side heat exchanger components. The chemical solvents used for SG cleanings are usually organic complexing agents, mainly EDTA, which dissolve preferentially metal oxides, but are less or non active for dissolving the metals. Accordingly, for the chemical removal of these deposits, consisting in magnetite and metallic copper, two different chemical environmental conditions are necessary:

• Magnetite removal process under reducing conditions.

• Copper removal process under oxidizing conditions, to convert the metallic copper to copper oxides for dissolution. If lead is present, this process is also used, since lead behaves similarly to copper.

Although different SG Chemical Cleaning vendors use different formulation of the cleaning solvents, the chemistry of these processes is very similar. The main difference is the use of inhibitors to prevent the CS corrosion. All low temperature magnetite solvents have to use inhibitors to inhibit CS corrosion. Also some high temperature solvents are using inhibitors, which have temperatures below 150 °C. They are applied mainly in US plants. Use of inhibitors is limited at about 140-150 °C due to their thermal decomposition. The only inhibitor-free high temperature magnetite solvent is the one that is used by AREVA High Temperature SG Chemical Cleaning process. This process is applied at temperatures between 160 and 175 °C and the CS corrosion is inhibited by stringent reducing conditions. The main source for the CS corrosion during the chemical cleaning is the iron-III, which is released by magnetite dissolution. Iron-III is a very strong oxidation agent, and can very fast corrode CS (metallic iron) into iron-II, if the CS surfaces are not protected by corrosion inhibitors. Alternative way of protecting the CS is the immediate scavenging of the iron-III by strong reducing agents. This is the way how the high temperature cleaning processes can inhibit the CS corrosion under strong reducing conditions. Hydrazine is used as reducing agent for this purpose, which starts to react fast as a strong reducing agent above 150 °C. Below this temperature the hydrazine reaction is not fast enough to protect CSs explaining why the low temperature process needs an inhibitor.


9-2(9-18)


HCC processes are usually applied as active cleanings to counteract corrosion problems or to remove huge amount of deposits or tenacious hard deposits. For this purpose they use high concentrations of cleaning chemicals with long exposure time. Accordingly, they may produce higher corrosion rates than the maintenance cleaning processes do, which are using dilute concentrations (see Section 9.1.2). Hard cleaning processes can be applied several times during the service life time of the SGs (some SGs were cleaned up to 8 times). However, it is not economical and therefore not recommended to apply them annually. For frequent cleaning, MCC processes are available.

In the following sub-sections two most applied HCC processes are briefly described.

EPRI SGOG process (low temperature cleaning process) 9.1.1.1

The EPRI SGOG process was developed in response to SG corrosion problems in the time frame of 1978 to 1983 [EPRI, 1994b]. The major focus of this program was the removal of corrosion products from CS drilled hole TSP crevices that caused denting corrosion. The EPRI SGOG process consists of three distinct solvent steps. These steps are referred to as “Iron Solvent”, “Copper Solvent” and “Crevice Solvent” steps. All these steps are based on the di-ammonium salt of EDTA. The process is applied with appropriate rinsing sequences and generally ends with a passivation step. The formulation and the application parameters of the generic EPRI SGOG process are given in Table 9-1. The individual steps and times required for typical cleaning without crevice cleaning is shown in Table 9-2. Actual cleaning process steps and the time required for each cleanings are plant specific and depend on the specific conditions of the SG to be cleaned and various utility constrains. If crevice cleaning step is to be applied, additional time is required which is again SG specific and has to be determined by plant specific qualification program.

Table 9-1: Generic EPRI SGOG chemical cleaning process parameters [EPRI, 1994b].

Process parameters Magnetite solvent Copper solvent Crevice solvent Passivation

EDTA 10% wt 5% wt 20% wt -

Hydrazine 1% wt - 0.03%

Hydrogen Peroxide - 2-3% wt - -

pH25 7 (NH4OH)

9.5 (EDA)

6 (NH4OH)

10 (NH4OH)

CCI-801 Inhibitor 0.5 – 1% vol. - 1% wt -

Temperature 190-205 °F (88-96 °C)

90-110 °F (32-43 °C)

245-255 °F (118-124 °C)

195-205 °F (91-96 °C)

The EPRI SGOG process was originally designed to use the reactor coolant system as a heat source for the SG solvent. This design was replaced by external recirculation and heating of the solvent due to utility concerns for cleaning “on critical path”. A typical layout of the EPRI SGOG process equipment for the external heat approach is shown in Figure 9-1. This method is usually used for applications in North America.


9-3(9-18)


10 Chemistry control and monitoring

For an adequate chemistry control reliable chemistry monitoring is indispensable. All decisions regarding the control of chemistry, like injection rate of pH agents or even plant shut-down in case of big impurity incidents due to condenser leaks, relies on the results of chemistry monitoring. A wrong decision based on biased or wrong monitoring results can be detrimental for the plant operation. Hence, representative sampling, selecting adequate sampling locations use of adequate monitoring equipments and finally right evaluation of the monitoring results for correct decisions are of highest importance for protecting the key components and for the reliable plant operation. In the following sections, important issues with respect to representative sampling and adequate monitoring are discussed.

10.1 Purpose of on line monitoring and grab sampling

Each type of chemistry control is associated with chemistry parameters that are describing the expected conditions for undisturbed plant operation or the chemistry boundaries, within which the chemistry should be controlled in case of pollution incident. These chemistry parameters, which needs to be monitored are called, depending on different chemistry guidelines, Expected, Diagnostic, Control and/or Limit values. This means that no monitoring is requested for research purposes if it is not associated with an action in case of deviation. Research or tests are of course important for specific needs, but are not included in the normal chemistry specifications and monitoring program. Moreover, the number of monitored parameters is limited to what is necessary so that the plant chemistry staff may concentrate on reliable measurements and spend more time on the various ways to achieve a good chemistry rather than spending time to redundancy of monitoring. Of course, in case of abnormal or specific situations, additional monitoring for diagnosis purpose is requested.

Monitoring limitation is one of the ways of cost saving as well as focusing on the most important parameters, able to guarantee a correct plant operation. In some cases, such a limitation contributes to avoid unnecessary chemical handling and wastes.

Another example of minimizing the monitoring program may be associated to the use of more resistant SG tubing Alloys with different chemistry specifications. The benefit would be:

• Less monitored parameters (e.g. no need for on-line or frequent measurements of anions at SGBD when cation conductivity values are low and stable).

• Less grab sampling with associated spent time and effort.

• No sophisticated application of MRC with potentially detrimental addition of ammonium chloride.

• More flexibility on shutdown schedule when some moderate and non detrimental contamination (e.g. river water condenser leak) occurs.

As proposed by Nordmann in the DAWAC13 project of IAEA, it is advisable to use on-line monitoring whenever it is feasible, with 4 main advantages over grab sampling [IAEA, 2006].

• There is an immediate indication in control room in case of large deviation for important parameters requiring an immediate remedial action, which may include shutdown of the unit if there is no other solution. Thanks to redundancy of the on-line monitors, the plant operators do not wait for validation of the large pollution by grab sampling before deciding for example to shut down the unit. This avoids material degradation and pollution of the system, which would require more cleaning time before restart-up the unit. Moreover, this may be even more important for plants without chemist on shift during normal situation.

13 Data processing technologies and diagnostics for water chemistry and corrosion in Nuclear Power Plants.


10-1(10-22)


• The trending of small pollution is possible when it is close to the detection limit or to the blank variation with grab sampling. This makes possible detection and fixing of small pollution.

• The spent time and money are lower.

• The plant chemists will have time to concentrate their effort on studying the results and will be more inclined to take the appropriate remedial actions.

Grab sampling is carried out in the following cases:

• When no on-line monitor is available for a chemical parameter (e.g. ammonia, amine).

• When the required frequency of analysis would induce more time for on-line monitor maintenance than for grab sampling and analysis (e.g. once a month, diagnosis such as calcium at SGBD on river water cooled units).

10.2 Criteria for representative sampling

An adequate control of the water chemistry program relays on a proper monitoring program and reliable sampling, which allows an efficient follow-up of the chemical conditions in the secondary side. In any case, the selected chemistry control parameters must be related to the key parameters (corrosion product, oxygen and dissolved impurities) and they, in return, must be associated to the key components of the system, Table 10-1 gives a general overview for chemistry control concept in the secondary side.

Table 10-1: General concept for chemistry control of secondary side.

Secondary side key component

Chemistry key parameter

Control parameter Monitoring parameter Sampling location

SGs

Corrosion products pH Specific conductivity Final FW

Impurities

Strong anions Sodium

Cation conductivity Condenser SGBD

Sodium SGBD Condenser hot-well

Chloride Sulphate

Chloride Sulphate

SGBD

Oxygen/oxidizing conditions

Oxygen

Oxygen

Redox potential Final FW

Hydrazine

Copper Copper Final FW

Turbine Impurities Organics CO2

Cation conductivity Moisture separator drain

The design of the sampling technique and location should consider the selected control parameters of the water chemistry. For example, it is inadequate to design a grab sampling with daily sampling frequency for a control parameter, which has Action Level(AL) that requires immediate plant shut-down within several hours. For such parameters, online monitoring techniques are required.


10-2(10-22)


Adequate chemistry control relies on reliable and representative sampling results. Many parameters are influencing the representativeness of the sampling, such as sampling nozzle, sampling transportation in the sampling line (i.e. sampling line material, length, isometrics, and temperatures), sampling flow rates, etc. As a general rule, the sampling lines should be designed with SS material, as short as feasible, with fewer elbows. They should also have an adequate diameter (usually 6 or 8 mm) that provides turbulent flow rates. The main purpose of selecting such a design is to avoid that sampling may change its composition along the sampling line by precipitation or chemical reactions with the sampling line surface, which may result in non-representative biased samples. In the following the most important issues are discussed that may affect the representativeness of the sampling.

In the secondary side, representative samples can be collected for most of the dissolved chemical species without any problems. However, special emphasis should be given for the sampling of FW oxygen, hydrazine and corrosion products. The sampling of these three parameters can be significantly biased by the sampling line during their transportation to monitoring equipment. The most crucial issue is the thermo-hydraulic conditions within the sampling line, which affects the representativeness of the sampling:

The sampling lines are usually small diameter pipes (6 – 8 mm), which provides huge amount of surface if compared to the volume of the sampling medium that flows in the line. This surface to volume ratio in the sampling line is significantly bigger than, for example, in the FW line. The plant sampling lines are usually long and their sampling coolers are installed far away from the sampling locations. Accordingly, they are not only long, but they are also hot lines. In the hot lines oxide scales built-up on the inner surface of the sampling lines. These oxide scales interact with the sampling medium. Especially FW oxygen and hydrazine sampling is biased by the sampling line surfaces, because the reaction between oxygen and hydrazine is catalyzed by corrosion products at higher temperatures (see Figure 10-1b). Also the inner surface of the FW line has oxide layers. However the surface to volume ratio in the FW line is negligible if compared to the sampling line. Therefore the decrease of oxygen and/or hydrazine concentration in the FW line is not so remarkable (see Figure 10-1a). These oxide scales in the sampling line influence the corrosion product (iron) measurements also by absorption and/or desorption mechanism. Therefore, especially for the FW sampling, it is recommended to design short lines with coolers that are installed as close as possible to the sampling location to shorten the hot line portion.

Figure 10-1: Model calculations for oxygen decay and ECP profile at 180 °C pH= 9.2 and N2H4: 100 µg/kg in a FW line (a) and in the sampling line (b) [Takiguchi, 2007].


10-3(10-22)


11 Application of water chemistry control programs

11.1 Power operation

This section explains some chemistry control programs that are applied by various utilities worldwide during different plant operation modes such as power operation, start-up and shut-down operation or lay up during outages. Each plant is applying its water chemistry program according to its chemistry specifications. In different countries, water chemistry guidelines are used, depending on the OEM of the plant, especially of the Key Components SGs and turbine. Utility experience feedback may also be taken into account for using the guidelines when preparing the chemistry specifications officially applied in each plant. The most used and published water chemistry guidelines are the EPRI (USA), EdF (France) and VGB (Germany) Guidelines. In Japan, MHI Water Chemistry Guidelines are used for PWR plants, however these guidelines are not publicly known. In VVER units, initially, Russian guidelines were applied in all the cases, and then some countries applied slightly different chemistry specifications. Therefore, in the following sections the different practices are explained for the power operation and for the plant start-up and shut-down operations that are based on only EPRI, EdF, VGB and some VVER Guidelines.

All Guidelines specify chemistry parameters according which the chemistry program is applied and in addition they have rules for corrective measures. In case of EPRI and VGB Guidelines these chemistry parameters are called Control and Diagnostic Parameters and the rules for corrective measures are the so called ALs. In the case of EdF Guidelines, the chemistry parameters are specified as Expected and Limit Values and also include control and diagnosis parameters; when there are several levels of values or when the limits correspond to a set of 2 values (e.g. cation conductivity + sodium at SGBD or pH + Li in the primary coolant), the corrective rules are defined as Zones, which correspond to ALs. The ALs or zones are prescribing values of parameters above which long-term system reliability may be jeopardized. Operating below lowest level value of the ALs or Zones provides a great degree of assurance that corrosion risk is minimized. However, field experience has shown that operation with impurity concentrations below these limits has resulted in significant corrosion in several plants, particularly for Alloy 600 MA. Thus it is recommended to operate according to ALARA principle with respect to impurities, instead of using intentionally the full allowance of the guidelines if there is no other important constraint.

In the following, the power operation chemistry practices according to these different guidelines with respect to their control parameters for FW and SGBD chemistry are explained.

11.1.1 EPRI guidelines

The EPRI Guidelines define the chemistry parameters as follows [EPRI, 2004a]:

• Control Parameters: Parameters that have a demonstrated relationship to SG degradation.

• Diagnostic Parameters: Important parameters to monitor program effectiveness, identify programmatic problems, or assist in problem diagnosis.

• Action Level 1: In the case of violation, corrective actions to be implemented to return to below AL 1 within one week. If this does not succeed, go to AL 2 for those parameters having AL 2 values.

• Action Level 2: Reduce power immediately to plant-specific level (usually to 30%); and return parameter to below AL 1 within 100 hours. If this does not succeed, go to AL 3 for those parameters having AL 3 values. If AL 2 is entered as a result of being in AL 1 for more than one week, and parameter concentration remains below the AL 2 values, operation at ~ 30% may continue. Escalation to full power can be resumed once below AL 1 values.

• Action Level 3: Shut down as quickly as safe plant operation permits and SG clean-up actions to be initiated.


11-1(11-21)


Guidelines for control parameters at ≥ 30% reactor power for FW, SGBD and condensate water are given in Table 11-1, Table 11-2 and Table 11-3, respectively.

Table 11-1: EPRI FW control parameters for > 30% power operation [EPRI, 2004a].

Action Level

Parameter Frequency 1 (1 week)

2 (100 hours)

3 (immediate)

pH Agent daily (a) - -

Hydrazine, [µg/kg] continuous <8 X CPD [O2] or < 20 µg/kg

(b) (b)

Total Iron, [µg/kg] weekly > 5 - -

Total Copper, [µg/kg] weekly > 1 - -

Oxygen, [µg/kg] continuous > 5 > 10 (a): Specific conductivity can be used in lieu of inline pH when correlation with pH additive concentrations can be established. (b): In event of loss of hydrazine feed that is not restored within 8 hours, commence shut-down as quickly as safe plant operation permits. If

hydrazine feed is restored, the plant may be returned to full power.

Table 11-2: EPRI SGBD control parameters for > 30% power operation [EPRI, 2004a].

Action Level


2 (100 hours)

3 (immediate)

Cation conductivity a [µS/cm]25 continuous - > 1 > 4

Sodium, [µg/kg] continuous > 5 > 50 > 250

Chloride, [µg/kg] daily > 10 > 50 > 250b

Sulfate, [µg/kg] weekly > 10 > 50 > 250b

(a): At plants operating with Advanced Amines, cation conductivity will be elevated as a result of organic acids. Therefore, it should be confirmed that the increase is a result of the presence of chloride or sulfate before initiating AL 2 and AL 3 requirements.

(b): If these parameters are not monitored continuously, cation conductivity may be substituted in lieu of imposing AL 3 values for chloride and sulfate.

Table 11-3: EPRI Condensate control parameter for > 5 % power operation [EPRI, 2004a].

Action Level


2 (100 hours)

3 (immediate)

Oxygen a [µg/kg] continuous > 10a > 30a,b

(a): If an appropriate FW oxygen monitoring is in use, and if plant is copper free, then condensate oxygen can be used as diagnostic parameter rather than as control parameter.

(b): Reduction to 30% power may not be a proper response to this AL 2 situation.


11-2(11-21)


11.1.2 EdF guidelines

EdF document specifies limit values, according to the applied chemistry control. These limit values corresponds to Control Values of other Guidelines. ALs for SGBD chemistry are illustrated as allowable zones for certain time duration, where sodium and cation conductivity limit values are shown in relation to each other. The FW chemistry limit values and the SGBD zones are given in Table 11-4 and Figure 11-1 respectively.

Table 11-4: EdF FW pH limit values for > 25% power operation and pH agent used [Staudt et al., 2002].

Parameter Expected value Limit value Materials - Treatment

pH25 value 9.1 to 9.3 9.0 to 9.4 Presence of Copper Morpholine treatment

Morpholine, [mg/kg] 4 to 6 4 to 8

Hydrazine, [µg/kg] > 10 > 5

pH25 value 9.5 to 9.6 9.2 to 9.8 No copper

Morpholine treatment Morpholine, [mg/kg] 6 4 to 8

Hydrazine, [µg/kg] 100 > 50

pH25 value 9.6 to 9.8 9.5 to 10.0 No copper

Ammonia treatment

Ammonia, [mg/kg] 2 to 5 -

Hydrazine, [µg/kg] 100 > 50

Figure 11-1: EdF SGBD water chemistry limit values for > 25% power operation, for (a): River cooled plants and (b): Seawater water cooled plants, [Staudt et al., 2002].


11-3(11-21)


12 Conclusive summary and recommendations

12.1 Overall objectives

This report describes the secondary side water chemistry applied in PWR and VVER plants. It covers the range from basic information to current knowledge. In this section, conclusion are more focusing on recommendations and the best solution for each specific plant, rather than on the explanation on the phenomena, on the Research and Development (R&D) results, on the description of the problems that are fully described in the corresponding sections. This is why some of the tables that have been proposed as a rationale for selection of the best options in some sections, are duplicated in this section.

The objective of secondary side water chemistry control is to minimize corrosion damage and performance losses for all secondary system components and thereby to maximize the reliability and economic performance of the secondary system. To achieve this objective, the water chemistry has to be compatible with all parts of the secondary system including SGs, turbines, condensers, FW heaters, MSRs, and finally piping. Special emphasis has to be put on SGs, because they are one of the key components of the PWR plants. Their degradation or performance loss greatly affects the overall plant performance. Most of the SG degradation problems have been related to corrosion caused by inappropriate material and design selection or poor secondary side chemistry. Then, in the past there has also been important corrosion of CS components, which induced severe failures and accidents.

The main causes for the SG corrosion problems on the secondary side have been observed on Alloy 600 MA, a highly sensitive material due to its composition (high nickel and low chromium content). The IGA/SCC occurred on the tubing of many plants and are due to the presence of this inadequate material in regions where the stress level is high or the chemistry is not appropriate. This is encountered where there are concentrated impurities beneath the deposits on the TS or in the Tubes to TSPs crevices and is of concern for any type of SG. The difference between various SG tubing materials is that even in presence of extremely small amount of pollution, IGA/SCC may develop on Alloy 600 MA while the other tubing materials (Alloy 800 NG, Alloy 690 TT, or 18%Cr-10%Ni-Ti) will only suffer from corrosion:

• In presence of important deposits where the impurities will concentrate.

• Or in presence of high level of impurities.

• Or in case of highly detrimental pollution such as significant lead content or Ion Exchange Resin (IER) fines.

But according to the feedback, although it is clear that lead is a highly detrimental pollutant, it is not the main cause of all the degradation that occurred in the secondary side of operating units.

With the progressive replacement of SG with Alloy 600 MA and the predominant presence of sufficiently resistant SG tubing in existing units, the new challenges of secondary water chemistry are keeping the SG in clean condition, selecting options so as to operate the unit in the most economical safe and friendly environmental way.


12-1(12-16)


The key points for keeping the SG in clean condition are the following:

• Selecting an adequate secondary system treatment (ammonia at high pH or amine) depending on several parameters and conditions in order to minimize the generation, transport and deposition of corrosion products in which the impurities will concentrate. This also means avoiding the use of permanent full flow condensate polishing stations, which are not compatible with the operation at a high pH. The absence of copper alloys will be also an important choice for allowing this pH selection.

• Keeping a small amount of oxygen in the condensate water (but keeping a low value and a reducing environment in the final FW and inside the SG), or add some dispersant in the SG.

• Performing regular cleaning of SG based on the visual inspection results performed during the previous outage.

• Avoiding the presence of high level of pollution, which is now easily achieved with tight condensers and good quality make-up water purification system.

• Avoiding the inadvertent pollution of the system by highly detrimental pollution such as lead and reduced sulphur compounds coming from resin fines.

• Using the proper purification systems for eliminating the impurity that are unavoidable.

There are other phenomena or constraints to take into account in the secondary system:

• Mitigating FAC of CS, which is achieved by selecting a chemistry that roughly fulfills the same requirements as corrosion transport mitigation.

• Avoiding the heat transfer loss associated with deposits on the SG tubes.

• Avoiding TSP blockage that may induce FIV.

• Selecting operating options and treatment that are producing reasonable quantities of liquid and solid wastes and are associated with reasonable costs.

The chemistry specifications or guidelines that are applied by various utilities have largely evolved with time, from poor chemistry conditions at the time where the condenser where leaking, to more and more restrictive guidelines, particularly for units with Alloy 600. It is now time to adjust these specifications to the new challenges that are encountered and that are slightly different from what they were in the 1990s. As mentioned, above, they should focus with greater priority to corrosion product mitigation, selecting the type of impurities that are detrimental to each type of material, and include economical and environmental constraints that are much more predominant than in the past.

Finally, the main objectives of the secondary water chemistry are:

• Keeping the optimum and sufficient pH by reagent addition.

• Keeping a reducing environment by hydrazine addition.

• Mitigating Intergranular corrosion on OD of SG tubes, by impurities limitation, particularly those that are alkaline or acidic and by sufficient care to avoid detrimental pollution during maintenance activities (such as lead).

• Mitigating FAC of CS and of transport of corrosion products thanks to a sufficiently high pH and a low deposition rate inside SG; this will avoid components failures, keep the thermal heat flux and hydraulic flow within the SG and avoid the concentration of impurities beneath deposits.

• Optimising operating costs and applying a good chemistry (specifications, operating practices, design options, conditioning selection).

• Keeping an acceptable level of waste releases into the environment through specification, operation practices, design options, conditioning selection.


12-2(12-16)


12.2 Impurities behaviour

Chemical species (pH control agents, corrosion products) salt impurities (e.g. condenser leaks) and organics which are transported by FW into SGs concentrate there more or less depending on their volatility due to steaming or they may deposit / be adsorbed beneath deposits at different locations, like at TTS and TSP crevices. The majority of the FW corrosion products (usually dissolved iron) deposit and build magnetite scales on the heat transfer tube surfaces causing tube fouling.

The impurities entering the SG may either concentrate or be eliminated by steam if they are volatile. Most of them will concentrate in the local areas of the SG were they may either reach high concentration on the tube surface and even more in crevices, or precipitate, depending on their solubility. The most detrimental are the compounds that may highly concentrate like sodium hydroxide while calcium or metallic compounds may easily precipitate. When impurities are precipitating or concentrating instead of being eliminated by the SGBD, this is called hide out phenomenon, that is occurring when the heat flux is sufficient, typically for power operation >25% of nominal power. Reversely, when the power is decreased, a part of the compounds that were hidden out will return to the solution, and this is called Hide-Out Return (HOR).

Besides building tube scales, they also accumulate within the heat transfer crevices (tube to TSP crevices and at TTS) creating blocked crevices. Approximately 5-10% of the FW corrosion products accumulate at TTS building hard-caked deposits. These blocked heat transfer crevices and TTS hard deposits are the special locations where the FW salt impurities preferentially concentrate and may cause tube corrosion.

12.3 SG degradation evolution

In the past, the tubing for SG was the typical SS tubing, nearly the 18% Cr – 10 % Ni austenitic SS, comparable to what is used in many nuclear systems, such as the primary coolant. The main weak point of this alloy is its sensitivity to SCC at high temperature in presence of oxygen and chloride. This is why the nickel base alloy, Inconel 600 MA (Alloy 600) has been selected by the American industry with Ni> 72%. However, a French scientist [Coriou et al, 1959] published results at a very early stage, before the start up of most of the international nuclear programs of PWR units, showing that Alloy 600 MA is sensitive to SCC in pure water conditions. This was meaning the use of this alloy, selected to avoid SCC in presence of chloride and cooling water ingress, may crack even in absence of any impurity. Nevertheless, several American scientists claimed that this Alloy was resistant and that degradation of French tests were due to the presence of lead impurity [Copson & Dean, 1965]. Insufficient attention has been given to French results for various reasons different from country to country. German people paid attention to these results and only built 1 unit with Alloy 600. MA. When it was evidenced that Alloy 600 MA was cracking, Siemens immediately moved, thus at an early stage, to Alloy 800 NG (Incoloy 800 NG), a iron base Alloy with lower Ni and higher Cr content. VVER of Russian design kept their 08X18H10T (or 08Cr18Ni10Ti) with 18% Cr 10 % Ni austenitic SS stabilized with titanium.

As far as the other components of the secondary systems are concerned, the main evolution has been the progressive elimination in many plants of copper alloys allowing:

• Increasing the pH of the secondary system and thus decreasing the corrosion of CS with all the associated consequences.

• Using condenser with other alloys for the tubing (titanium or SS) and an overall design permitting the plant operation with very few leaks.

SS has been used for several other components (heaters, MSR) in some existing plants, in replacement of CS or copper alloys components or in new plants. This is avoiding FAC and drastically decreasing the corrosion product transport. Although for FAC mitigation, the use of low alloy steel with a content of 2.5% Cr is sufficient, in several situations, SS has been used.


12-3(12-16)


13 References

Alves-Vieira M., Nordmann F., Molé D. and Fourcroy H. "Investigations on the Secondary Water Treatment - pH Control and Compounds behavior". EPRI, SGMP- TAG meeting. Santa Fe, NM. 5 - 7 December 2006.

Andersson, private communication, 2009.

Aspden J.D., Dedekind I., Gericke G. and Wilshire S., Proc., VGB Conference Chemistry in Power Plants (Essen, Germany), VGB PowerTech e.V., TB 437, P1, Essen, Germany, 2001.

ASTM D 1066, Standard practice for sampling steam, 2001.

Balakrishnan P., “High-Temperature Properties of some Alternative Amines”, EPRI Workshop on Use of Amines in Conditioning Steam Water Circuits, Tampa, FL, USA, 25-27 September 1990.

Battaglia P. J. and Prentice K., “Advancements in steam generator deposit removal”, Nuclear Plant Journal, May-June 2003.

Battaglia P. J. and Rogosky D., “The use of advanced scale conditioning agents for maintenance of the secondary side of nuclear plant stean generators”, 5th CNS International Steam Generator Conference, Toronto, Canada, Nov. 26-29, 2006.

Baum A. J., “Restricted geometries and deposits”, Chapter 6 of the ASME Handbook on Water Technology for Thermal Power Systems. Paul Cohen, Editor-in-Chief. New York: The Americal Socity of Mechanical Engineers, pp. 279-338, 1989.

Baum A. J. and Rootham W. M., “Operating strategies for minimizing the occurrence of steam generator ODSCC”, Water Chemistry of Nuclear Reactor Systems 7, British Nuclear Energy Society, BNES, London, pp. 625, 1996.

Bengtsson B., Svanberg P., Bothin U., Svärd G. and Velin A., “Experiences of Water treatment APPLICATIONs at Ringhals”, NPC’08” International Conference on Water Chemistry of Nuclear Reactor Systems – Berlin 14-18 September 2008.

Berge Ph. and. Donati J. R. Nuclear technology, vol.55 n°1, pp.88 to 104, October 1981.

Berge Ph., “The Use, Need, and Desirable Properties of Amines for Use in Steam-Water Circuits”, CEGB - Amine Workshop Proceedings SWR/SSD 0791/N/96. Bristol, England, 29 April - 1 May 1986.

Berger M., Brun C., Clinard M-H. and Albanese A., “The Flamanville EPR: The Role of Chemistry in the Design and Development of a Third Generation PWR”, Proceedings of the International Conference on Water Chemistry in NPP, Berlin, Germany, October 2008.

Bjornkvist L., Siemens-KWU received from Ringhals for test purposes, 1994

Bojinov et al., “Corrosion Behaviour of A Steam Generator Tube Material In Simulated Steam Generator Feedwater Containing Chlorides and Sulphates”, International Conference on Water Chemistry in Nuclear Reactor Systems – SFEN (French Nuclear Energy Society). Avignon, France, 22 - 26 April 2002.

Bouecke R., Odar S. and Stellwag B., “Operating experience with steam generators”, Siemens Service Report Power Plants, Dec. 1989

Bretelle J-L., Stutzmann A., Nordmann F., “Design and Operating Conditions Consequences for PWR NPP’s chemistry”, NPC’08” International Conference on Water Chemistry of Nuclear Reactor Systems – Berlin 14-18 September 2008.


13-1(13-13)


Brozova et al, “Nature of Tube degradation in Horizontal Steam Generators”, International Symposium of Fontevraud V, France, SFEN (French Nuclear Energy Society), September 2002.

Castaño Marín Ma, et al, “Steam Generator Replacement: Inconel 690TT and Incoloy 800 Mod. as an Alternative to Inconel 600”, Progress in the Understanding and Prevention of Corrosion, Vol. 1, Barcelona, Spain, July 1993, p447-457, The Institute of Materials, London, 1993.

Cattant F., “Chemical analyses of OD deposits removed from SG tubes”, EPRI IGA/IGSCC workshop. Minneapolis, MN, USA, 1993.

Cattant F., Dupin M., Sala B. and. Gelpi A., “Analysis of deposits and underlaying surfaces on the secondary side of pulled out tubes from a French plant”, Fontevraud, 12-16 September, 1994.

Cattant F., “Chemical Analysis of OD Deposits Removed from SG Tubes”, Proceedings: 1993 EPRI Workshop on Secondary-Side IGA/SCC, pA7-1 to A7-6, EPRI TR-107883, March 1998.

Chen M-J., “The application of steam generator scale cleaning process at Maanshan NPP”, Symposium on Water Chemistry and Corrosion of Nuclear Power Plants in Asia, Taipei, Taiwan, Sept. 26-28, 2007.

Choi S. and Lumsden J., “Investigation of Lead Equilibrium Lines on Pourbaix Diagram Using Electrochemical Measurements at 280°C”, NPC’08 International Conference on Water Chemistry of Nuclear Reactor Systems – Berlin 14-18 September 2008.

Choi S., Haas C. Sawochka S. and.Marks C., “PWR Secondary Chemistry Benchmarking”, International Conference on Water Chemistry of Nuclear Reactor systems, Berlin, Germany, 2008.

Cohen P. “Water Coolant Technology” Monograph published by American Nuclear Society, 555 North Kensington Avenue, La Grange Park, Illinois 60525. ISBN: 0-89448-020-0. Copyright 1980.

Combrade P. et al., “Effect of Sulfur on the protective Layers on Alloys 600 and 690 in Low and High Temperature Environments”, Fourth International Symposium on Environmental Degradation of Materials in Nuclear Power Systems. Jekyll Island (GA), Ed. TMS p 5 (1990), USA, August 1989.

Copson H. R. and Berry W. E., “Corrosionof Inconel Nickel-Chromium Alloy in Primary Coolant of Pressurized Water Reactors”, Corrosion, vol 18, N°1, pp 21t-26t, 1962.

Copson H. R. and Dean S. W., “Effect of contaminants on Reistance To Stress Corrosion Cracking of Ni-Cr Alloy 600 in Pressurized Water”, Corrosion, vol 21, N°1. 1965.

Cordovi M.A., “Corrosion considerations in light water-cooled nuclear power plants”, paper No. 7, International Nickel Power Conference, International Nickel, New York, p. 35, 1972.

Coriou H et al, “Corrosion sous contrainte de l’Inconel dans l’eau à haute température”, 3° Colloque de Métallurgie sur la Corrosion. Saclay, 1959 p. 161-169. CEN Saclay (France) et North Holland Publishing Amsterdam, 1959.

Corredera G., Alves-Vieira M. and de Bouvier O., “Fouling and TSP blockage of steam generators on EDF fleet: Identified correlations with secondary water chemistry and planned remedies”, International Conference on Water Chemistry of Nuclear Reactor systems, Berlin, Germany, 2008.

Cullen W., Partrige M., Paine, J. P. N., “Corrosion of Alloys 600 and 690 in acidic Sulphate and Chloride Solutions”, Sixth International Symposium on Environmental Degradation of Materials in Nuclear Power Systems. San Diego, Ca, USA, August 1993.

Cullen W., EPRI, SGMP meeting, 2006.


13-2(13-13)


Daret J. et al, “Secondary side Degradation of SG tubing. Which inhibitor for which causes”, 8th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors, Amelia Island, August 1997.

Daret J. et al, “Evidence fo rthe reduction of Sulfates under Reporesentative SG Secondary Side Conditions, and for the Role of Reduced Sulphates on Alloy 600 Tubing Degradation”, 9th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors, Water ANS and NACE, Newport Beach, USA, p 567, 1999.

Dauvois V., Lambert J., Desmoulins D. and Nordmann F., “Laboratory and plant investigations on decomposition products of morpholine in the secondary system of French PWRs”, Water chemistry of nuclear reactor systems 4, BNES, London p.369-374,1986.

Davis J. and Battaglia P. J., “Advanced Scale Conditioning Agents”, Nuclear Plant Journal, May-June 2004.

de Bouvier O., Demasse P., Bouchacout M. and Fruzzetti K., “Redox conditions effect on flow accelerated corrosion: Influence of hydrazine and oxygen”, International Conference on Water Chemistry in Nuclear Reactor System, Avignon, France, April 22-26, 2002.

De Bouvier O., Lemaire P. and Nordmann F., “How to manage SG life duration and degradation with 600 MA tubing”, SGMP / TAG presentation of EDF – Annual results. Washington DC, 9-11 August 2005.

Debray W. and Stieding L., “Materials in the primary circuit of water cooled power reactors”, Paper No 3, International Nickel Power Conference, Lausanne, Switzerland, p. 3/1, 1972.

Denk J. and Svoboda R., “Stress corrosion cracking due to carbon dioxide and organic impurities in the steam/Water cycle”, Power Plant Chemistry, 8(7), 2006.

Dijoux M., Le-Genti P., Nordmann F., Odar S., Roumiguiere F. and Hollwedel U., “Chinon B1 SG Chemical Cleaning - Application of Framatome ANP Inhibitor-Free High Temperature Chemical Cleaning Process”, Proceeding of International Conference on Water Chemistry of Nuclear Reactor Sytems, San Francisco, US, 2004.

Dijoux M., Bouvier O., Mercier S., Pages D., Bretelle J-L., Leclercq P., Tempier F. and Roumiguiere, F., “Application of AREVA inhibitor-free high temperature chemical cleaning process against blockages on SG tube support plates”, International Conference on Water Chemistry of Nuclear Reactor systems, Berlin, Germany, 2008.

DIN 25477, “Probenahmesysteme in Kernkraftwerken mit Leichtwasserreaktoren”, 1989.

Dinov K. et al., “Lead Assisted Corrosion of Alloy Components in PGV-1000 Steam generator of NPP Kozloduy”, Second International Seminar on primary and secondary water chemistry at NPP, Balatonfüred, Hungary, 19-23 September 1995.

Dobrovolska I., Arkhypenko A. and Nordmann F., “Morpholine Secondary Water Chemistry in Ukrainian and French Units”, International Conference on Nuclear water Chemistry in Reactor Systems, Jeju, Korea, October 2006.

Dow B.L., “Steam generator progress report”, Rev. 10, EPRI, 1994.

Eaker R., FryeC., Rochester D., “Evaluation of Alternate Oxygen Scavengers for Steam Generator Wet Layup”. Proceeding of International Conference on Water Chemistry of Nuclear Reactor Sytems, San Francisco, US, 2004 2004.


13-3(13-13)

Date post:	02-Nov-2019
Category:	Documents
Upload:	others
View:	40 times
Download:	1 times

PWR AND VVER SECONDARY SYSTEM WATER CHEMISTRY – STAND ... · PWR and VVER Secondary System Water...

Documents