Corrosion is a major service life-limiting mechanism for both pressurized water reactors (PWR) and boiling water reactors (BWR). While most of the corrosion research emphasis in the nuclear corrosion community has been focused on environmentally assisted cracking (EAC) of austenitic stainless steels and nickel-based alloys and weld metals, in particular stress corrosion cracking (SCC), there are other corrosion phenomena that seriously affect plant life extension that cannot be ignored. This paper presents three other important corrosion areas, i.e., general corrosion of the light water reactor (LWR) containments, flow-accelerated corrosion of carbon steel piping systems, and the corrosion of buried piping.

Corrosion is a major service life-limiting mechanism for both pressurized water reactors (PWR) and boiling water reactors (BWR). While most of the research effort has been focused on intergranular stress corrosion cracking (IGSCC) in BWR, primary water stress corrosion cracking (PWSCC) in PWR and irradiation-assisted stress corrosion cracking (IASCC) in both systems, less attention has been paid to other plant life-limiting phenomena such as general corrosion of LWR containments, flow-accelerated corrosion of carbon steel systems in both designs, and the corrosion degradation of buried piping.

Nuclear reactor containment is an airtight steel structure enclosing the reactor normally sealed off from the outside atmosphere. The steel is either free-standing or attached to concrete. It is designed, in any emergency, to contain the escape of radiation. The containment is the fourth and final barrier to radioactive release (part of a nuclear reactor's defense in depth strategy), the first being the fuel ceramic itself, the second being the zirconium alloy fuel cladding tubes, and the third being the reactor vessel and coolant system.

Oyster Creek

The Oyster Creek BWR, which entered service in 1969, utilizes a Mark I carbon steel containment as shown in Figure 1. During the 1980 refueling outage, water was noted around various containment penetrations and floors.1 The presence of water indicated that an intrusion of water into the annular space between the carbon steel drywell shell and concrete shield wall had occurred. Radiological analysis of the water samples indicated an activity level similar to primary water. This suggested that the source of the water was the reactor cavity located immediately above the dry-well as indicated by the arrow in Figure 1.

FIGURE 1

Oyster Creek Mark I containment with sand cushion.

FIGURE 1

Oyster Creek Mark I containment with sand cushion.

Close modal

While initial analyses suggested that the water was from leaks at the bellows drain line gasket, it was eventually determined that the leaks were due to numerous through-wall corrosion fatigue cracks located approximately 2 m (~6 ft) above the cavity seal floor, Figure 2.1 This cavity seal leakage plus any condensation in the gap between the insulation and the drywell shell, the initial installation of moist sand in the sand cushion, and the seepage of water from the initial sprayed-on insulation slurry (magnesium oxide [MgO], magnesium chloride [MgCl2], and H2O) could provide an electrolyte for the corrosion of the drywell shell.

FIGURE 2

Close up of the Oyster Creek Mark I reactor cavity with stainless steel liner.

FIGURE 2

Close up of the Oyster Creek Mark I reactor cavity with stainless steel liner.

Close modal

As illustrated in Figure 3, the following series of factors/events most likely affected the corrosion of the Oyster Creek drywell:1 

  1. Backfilling of moist sand into the transition zone creates an initial electrolyte. Sand is contaminated by open exposure to marine environment during storage and installation. Backfilling also affects porosity of the sand, which affects moisture retention quality and creates random air pockets.

  2. Expansion of drywell during pressure testing “squeezes” water out of the insulation slurry that flows down into the sand bed. This water initially contains high quantities of chloride and sulfate.

  3. Corrosion of the steel drywell initiates. While the red lead primer provides some initial protection, carbon dioxide (CO2) from the air and sulfate from the sand or insulation accelerate the breakdown of the limiting inhibitive qualities of the red lead primer.

  4. Areas with more ready access to oxygen such as the insulation gap and drain become local cathodes.

  5. Areas adjacent to concrete are provided some corrosion protection because of local alkalinity. A macro-galvanic cell is established between the steel adjacent to the concrete and the steel adjacent to the sand cushion.

  6. Condensation cycles and leaks from the fuel pool cavity liner when filled during outages contribute air-saturated water to maintain moist sand cushion.

  7. Some regions of the sand cushion see alternate wetting and drying during startup/shutdown cycles. This results in a concentration of chloride at the carbon steel/sand interface.

  8. Sand maintains corrosion products close to metal surface and can physically reduce the corrosion rate, Figure 4.

FIGURE 3

Schematic of corrosion mechanism of Oyster Creek drywell.

FIGURE 3

Schematic of corrosion mechanism of Oyster Creek drywell.

Close modal

Subsequently, the sand cushion was removed and the drywell was triple-coated with an epoxy coating and the corrosion has been mitigated.2 A strippable coating is applied to the reactor cavity during each outage to prevent any additional water leaking into the drywell region. Figure 5 presents an example of the general corrosion history of sand bed bay no. 11 where ultrasonic testing (UT) measurements of the drywell thickness have shown that the corrosion rate is statistically nil.

FIGURE 4

Corrosion product on Oyster Creek drywell.

FIGURE 4

Corrosion product on Oyster Creek drywell.

Close modal
FIGURE 5

Corrosion rate reduction of carbon steel drywell at Oyster Creek.

FIGURE 5

Corrosion rate reduction of carbon steel drywell at Oyster Creek.

Close modal

Beaver Valley Power Station, Unit 1

On April 23, 2009, during a refueling outage at Beaver Valley Power Station, Unit 1, which initiated service in 1987, the PWR licensee performed a visual examination of the interior reactor containment building steel liner.3 At a containment elevation of 227 m (746 ft), the licensee identified an area approximately 76 mm (3 in) in diameter that exhibited blistered paint, Figure 6.4 The paint blister was intact at the time of discovery. Collapse of the blister during further inspection revealed a protruding corrosion product underneath. The licensee then cleaned this area to allow further evaluation. The cleaning activity uncovered a rectangular area of approximately 25 mm (1 in) (horizontal) by 10 mm (0.375 in) (vertical) that penetrated through the entire liner plate thickness. UT of the surrounding area showed liner thinning within an area of approximately 65 cm2 (10 in2). The licensee removed the corroded section of the liner and discovered a partially decomposed piece of wood approximately 50 by 100 by 150 mm (2 by 4 by 6 in) embedded in the concrete behind the section of the liner, Figure 7.4 The wood, which was left behind as a result of inadequate housekeeping and quality assurance practices during the original construction of the containment wall in the early 1970s, was approximately 37 years old, had a moisture content of 13%, and a pH of 3.5. This was a classic corrosion case of a small anode surrounded by a large cathode.

FIGURE 6

Paint blister identified at Beaver Valley Unit 1 containment.4 

FIGURE 6

Paint blister identified at Beaver Valley Unit 1 containment.4 

Close modal
FIGURE 7

Portion of carbon steel liner plate removed to investigate the debris behind the liner plate at Beaver Valley Unit 1.4 

FIGURE 7

Portion of carbon steel liner plate removed to investigate the debris behind the liner plate at Beaver Valley Unit 1.4 

Close modal

Corrosion mitigation was achieved by welding in a new portion of the liner, pressure testing the area, performing a volumetric examination of welds, and restoring the paint.4 

Containment Corrosion Summary

Nuclear plant operating experience indicates that containment liner corrosion is often the result of liner plates being in contact with objects and materials (e.g., wooden pieces, workers' gloves, wire brush handles, felt) that are lodged between or embedded in the containment concrete.3 Liner locations that are in contact with objects made of an organic material are susceptible to accelerated corrosion because organic materials can trap oxygenated water that will provide the cathodic reactant for carbon steel corrosion. Organic materials also can cause a localized low-pH area when they decompose. Corrosion that originates between the liner plate and concrete is a greater concern because visual examinations typically identify the corrosion only after it has significantly degraded the liner. In some cases, licensees identified such corroded areas by performing an ultrasonic examination of suspect areas (e.g., areas of obvious bulging, hollow sound).

Introduction

Flow-accelerated corrosion (FAC, also called flow-assisted corrosion and, misleadingly, erosion/corrosion) causes wall thinning of carbon steel piping, tubing, vessels, and components. The wall thinning is caused by an increased rate of dissolution of the normally protective oxide layer (e.g., magnetite, Fe3O4) that forms on the surface of carbon steel and low-alloy steels when exposed to high-velocity water or wet steam, Figure 8.5 The oxide layer eventually reforms and the process subsequently continues. The problem is widespread in all types of conventional (e.g., fossil and combined cycle) and nuclear power plants. Wall thinning rates as high as 3 mm/y (~120 mpy) have been observed. If the thinning is not detected in time, the reduced wall cannot withstand the internal pressure and other applied loads. The result can be either a leak or complete rupture.

FIGURE 8

Sketch of the FAC mechanism in carbon steel.5 

FIGURE 8

Sketch of the FAC mechanism in carbon steel.5 

Close modal

The rate of wall loss (unfortunately aka “wear rate”) of a given component is affected by temperature, local flow velocity, component geometry on local hydrodynamics (e.g., turbulence), the pH at temperature, the liquid phase dissolved oxygen concentration, and the alloy composition (e.g., especially Cr, Mo, and Cu).

Surry 2 and Mihama 3 Fatal Flow-Accelerated Corrosion Incidents

Even though FAC control programs have been implemented in most nuclear and fossil plants, major damage still occurs. For example, the first FAC in a single-phase system in an LWR occurred on December 9, 1986, in a 46 cm (18 in) carbon steel condensate system elbow rupture in the secondary side of the Surry 2 PWR after approximately 13 years of operation, Figure 9.6 This event killed four workers and injured an additional four workers. Unfortunately, almost 18 years later on August 9, 2004, a carbon steel steam line ruptured at Mihama 3 PWR killing five workers and severely injuring an additional six workers after 28 years of operation, Figure 10.7–8 The Mihama 3 event occurred during the process of assembling scaffolding to support workers to inspect the carbon steel steam line for FAC. Since the Mihama 3 incident occurred more recently than Surry 2, it is considered prudent to provide some additional details on this unfortunate and unwarranted event.

FIGURE 9

Photo of the 46 cm (18 in) carbon steel condensate system elbow rupture in the secondary side of the surry.2 

FIGURE 9

Photo of the 46 cm (18 in) carbon steel condensate system elbow rupture in the secondary side of the surry.2 

Close modal
FIGURE 10

Photo of the 56 cm (22 in) carbon steel steam line rupture at Mihama.3 

FIGURE 10

Photo of the 56 cm (22 in) carbon steel steam line rupture at Mihama.3 

Close modal

The Mihama piping was fabricated from carbon steel (JIS SB42) with a nominal outside diameter of 558.8 mm (22 in) and wall thickness of 10 mm (0.39 in).7–8 The system was designed for a maximum service temperature of 195°C (383°F). The actual service temperature was approximately 140°C (284°F) with the flow rate of approximately 1,700 m3/h (450,000 gal/h). The feedwater was under all-volatile treatment (AVT) with hydrazine as the oxygen scavenger and ammonia (ethanolamine in later service years) for pH control.

The rupture was located in the A-loop feedwater line from the 4th low pressure (LP) heater to the de-aerator, just downstream of an orifice for measuring flow rate in the piping, Figure 11.7–8 The fish-scale-like (scalloped) surface pattern, which is characteristic of FAC damage, was observed on a large portion of the straight pipe. A critical observation to note here is that at the same location no such damage was found on the (inside) bottom of the pipe where the wall thickness was still close to the nominal and it was covered with a surface film (oxide) almost 0.5 mm (20 mils) thick.8 

FIGURE 11

Mihama carbon steel pipe rupture (schematic).7–8 

FIGURE 11

Mihama carbon steel pipe rupture (schematic).7–8 

Close modal

The same portion of piping from the B-loop was similarly investigated that also showed some wall thinning, but with less severity. However, it is important to note that the piping surface upstream of the orifice did not show any significant thinning.8 That is, despite the nominally same water chemistry and piping material, the FAC damage was confined only where the flow characteristics were affected downstream of an orifice, i.e., a location of locally high flow rate, turbulence, and flow reattachment.

Important Factors Affecting FAC

The critical parameters to cause FAC in for carbon steel are the following:9 

  • Water chemistry

    • ○ pH (<9.3)

    • ○ Dissolved oxygen (<40 ppb)

    • ○ Temperature maximums at 135°C (275°F) for water and at 177°C (350°F) for steam

  • Material Chemistry

    • ○ Cr, Cu or Mo (<0.5%)

  • Hydrodynamics

    • ○ Velocity – (>4.6 m/s [>15 ft/s] water and >27 m/s [>90 ft/s] steam)

    • ○ Geometry – (turbulence in elbows, tees, etc.)

    • ○ Steam Quality – (0.1 to 0.9)

LWR single-phase FAC susceptible carbon steel components and systems are the following:9 

  • Condensate and feedwater

  • Auxiliary feedwater

  • Heater drains

  • Moisture separator drains

  • SG blow-down

  • Reheater drains

  • Lower head drain lines

  • Other drains

LWR two-phase FAC susceptible carbon steel components and systems are the following:9 

  • High- and low-pressure extraction steam lines

  • Flashing lines to the condenser (miscellaneous drains)

  • Feedwater heater vents

  • Moisture separator drains

  • Lines with leaking valves

  • Main steam lines

  • Reheat steam lines

The mitigation of FAC in BWR includes replacing carbon steel with low-alloy steel or stainless steel and controlling the dissolved oxygen content. PWR can also use the option of replacing carbon steel with low-alloy steel or stainless steel. However, since PWR water chemistry is deaerated, their water chemistry FAC mitigation option is increasing the pH or using different chemical additions to control the pH.

Virtually every nuclear site has kilometers (miles) of buried piping where often 100% of it has never been inspected or otherwise assessed relative to its structural integrity.10 For all sites, the continued structural integrity of buried piping will be a key item to achieving plant life, especially for license renewal. Degradation of buried piping is a significant issue facing nuclear power plant owners.

Unlike above-ground piping systems, buried pipes corrode from both the fluid side (ID) and corrode or experience mechanical damage from the soil side (OD).10 This continuing degradation is difficult to assess since the pipes are very difficult to reach for inspection. When buried pipes do leak, the source of leakage can be difficult to locate, evaluate and subsequently repair. As a result, several LWR have experienced very costly leaks and repairs of their buried piping systems.

Leaks or ruptures in safety-related lines can also pose challenges to plant safety and safe shutdown.10 Leaks in lines containing radioactive materials can affect the ground water or cause airborne emissions and have the potential to breach the plant boundary. Leaks of pipes containing other hazardous materials (e.g., diesel oil, fuel oil, and hydrazine) can also contaminate the surrounding soil and have the potential to enter the ground water.

To mitigate these potential problems, the nuclear industry has initiated several actions to provide tools and technology as well as provide a consistent and consensus approach to deal with buried piping.10 These activities are being supported by several industry organizations including:

  • —Electric Power Research Institute's (EPRI) Buried Pipe Integrity Group (BPIG) and Balance of Plant Corrosion (BOPC) Integration Committee. Among other activities, BPIG and BOPC have sponsored the development of EPRI report 1016456, “Recommendations for an Effective Program to Control the Degradation of Buried Pipe” that contains a recommended approach to deal with buried pipe issues.10–11 

  • —Nuclear Energy Institute's (NEI) Nuclear Strategic Issues Advisory Committee (NSIAC) has sponsored two related initiatives: Ground Water Protection Initiative NEI-07-0710,12 and the Guideline for the Management of Buried Piping Integrity, NEI-09-14.10,13 

  • —Institute of Nuclear Power Operations (INPO) has provided several related buried piping guidelines to their member organizations.

The integrity of buried pipes has received increased interest from the U.S. Nuclear Regulatory Commission (NRC). This includes guidance provided by Inspection Procedure 62002,10,14 requirements for inspection of buried pipe as identified in the GALL report10,15 and more recently, a directive to the staff to evaluate the adequacy of regulations for buried pipe.10,16 Recently, INPO has made integrity of buried pipe a focus area of plant assessments. Also, buried pipe integrity is of interest for new plants, to ensure that new designs build upon lessons learned from operation of current facilities.

Some Common Types of Corrosion Degradation Failures in Buried Piping

General Corrosion — Aside from through-wall penetration, Figure 12, general corrosion can have a significant impact on structural integrity when the stress from the applied load (e.g., pressure, dead weight, etc.) on the system exceeds the strength of the component.10 The corrosion allowance built into the plant design was intended to accommodate general corrosion that was expected to occur over the life of the plant, such that the stresses in the piping at the end of life were sufficiently low that the piping would always retain its structural integrity. The key parameters related to general corrosion are the extent of metal loss and their effects on the load-carrying capability of the piping under normal and off-normal conditions.

FIGURE 12

General corrosion (exterior) failure of a carbon steel buried pipe.10 

FIGURE 12

General corrosion (exterior) failure of a carbon steel buried pipe.10 

Close modal

Pinhole Leaks — Pinhole leaks, which usually have extremely low leak rates, in buried piping will occur as the result of general or localized corrosion in carbon steels and low-alloy steels or from highly localized corrosion (e.g., the result of microbiologically influenced corrosion [MIC] or chloride pitting in stainless steel or copper-based alloys), Figure 13.10 Pinhole leaks generally have no effect on the function of the system, i.e., flow rates from the leak are typically measured in drops per minute as compared to system flows of hundreds or thousands of liters/gallons per minute. However, these leaks may have an effect on neighboring equipment or can have environmental effects (e.g., leaks of radionuclides, oil, or fuel). Both the potential collateral damage (e.g., flooding or drips or sprays on nearby electrical equipment) and the potential impact of the localized thinning on the structural integrity of the piping must be considered.

FIGURE 13

Pinhole failure of a buried carbon steel buried pipe.10 

FIGURE 13

Pinhole failure of a buried carbon steel buried pipe.10 

Close modal

Occlusion — The buildup of corrosion products or deposits can dramatically decrease the flow-carrying capability of a service water pipe, Figure 14. The overall occlusion will be affected by debris, macro-fouling, and corrosion products.10 In some cases, biological fouling can cause occlusion of the service water piping. Degradation as a result of occlusion is much more common in smaller bore piping, where the surface-to-volume ratio is higher so that the corrosion products, which are less dense than the underlying material from which they were formed, will “impinge” upon each other producing a restriction to flow. In large bore pipe, the same volume of corrosion product (or other deposit) will be a much smaller fraction of the cross-sectional area of the pipe. Corrosion products can be released from surfaces from any size pipe and deposit elsewhere, producing detrimental effects at those locations. Corrosion product buildup also can affect the flow-carrying capabilities of the pipe as a result of the increases in surface roughness, even before the cross-sectional area of the pipe is significantly degraded.10,17–18 

FIGURE 14

Occluded carbon steel pipe from pitting and tubercles.

FIGURE 14

Occluded carbon steel pipe from pitting and tubercles.

Close modal

Cavitation Corrosion — Cavitation, i.e., the repeated nucleation, growth, and violent collapse of cavities in the liquid at a solid/liquid interface, in service water piping, is most common immediately downstream of throttling valves and may be especially prone to occur downstream of valves that are not intended for throttling, such as butterfly valves. Albeit not from a buried pipe, per se, Figure 15 illustrates an example of cavitation corrosion in a carbon steel elbow immediately downstream of a butterfly valve in a service water system.10 Cavitation damage is avoided by reduction of either the flow velocity or temperature where reduction of flow velocity is preferable.10,19 Other cavitation mitigation measures could include system redesign with a valve or orifice less likely to produce cavitation conditions, selection of more cavitation-resistant material, or erosion-resistant welded overlays.

FIGURE 15

Surfaces of carbon steel service water elbow subjected to cavitation.10 

FIGURE 15

Surfaces of carbon steel service water elbow subjected to cavitation.10 

Close modal
  • ❖ While most of the corrosion emphasis in the nuclear corrosion community has been focused on EAC of austenitic stainless steels and nickel-based alloys and weld metals, there are other corrosion phenomena that affect plant life extension that cannot be ignored, such as the general corrosion of carbon steel containments, FAC of carbon steel piping, and the various corrosion mechanisms affecting buried piping. All organizations involved with LWR life extension need to have a global approach to all LWR corrosion concerns.

1.
B.M.
Gordon
,
“Corrosion Issues in the BWR and Their Mitigation for Plant Life Extension,”
paper presented at the ANS Topical Meeting on Nuclear Plant Life Extension, Snowbird, UT
,
July 1988
.
2.
B.L.
Lipford
,
J.C.
Flynn
,
“Drywell Corrosion Stopped at Oyster Creek,”
Power Engineering
,
November 1993
,
p
.
47
.
3.
NRC Information Notice 2010-12: Containment Liner Corrosion, June 18, 2010
.
4.
D.
Weakland
,
D.
Hecht
,
“Beaver Valley Unit 1 1R19 Containment Liner Inspection,”
presented at the APOG Meeting
,
June 2009
.
5.
R.B.
Dooley
,
PowerPlant Chemistry
10
,
2
(
2008
):
p
.
68
.
6.
U.S. Nuclear Regulatory Commission
,
“Feedwater Line Break,”
Information Notice 86-106, Supplement 1, U.S. Nuclear Regulatory Commission, Washington, DC
,
February 13, 1987
.
7.
NISA
,
“Secondary Piping Rupture Accident at Mihama Power Station, Unit 3, of the Kansai Electric Power Co., Inc. (Final Report),”
(Translated by JNES, Rev 1 as of May 14, 2005), The Nuclear and Industrial Safety Agency
,
March 30, 2005
.
8.
Y.S.
Garud
,
“Issues and Advances in the Assessment of Flow Accelerated Corrosion,”
paper presented at the 14th Int. Conf. on Environmental Degradation in Nuclear Power Systems—Water Reactors, Virginia Beach, VA. August 23-27, 2009. Published in proceedings of same, eds
.
T.R.
Allen
,
J.T.
Busby
,
G.
Ilevbare
,
American Nuclear Society
,
La Grange Park, IL
,
2010
.
9.
“Flow-Accelerated Corrosion in Power Plants,”
EPRI, Palo Alto, CA, 1998, TR-106611-R1
.
10.
“Balance of Plant Corrosion—The Buried Pipe Reference Guide,”
EPRI, Palo Alto, CA, 2010, 1021470
.
11.
Recommendations for an Effective Program to Control the Degradation of Buried Pipe. EPRI, Palo Alto, CA, 2008, 1016456
.
12.
Ground Water Protection Initiative, Nuclear Energy Institute, NEI 07-07, August 2007
.
13.
Guideline for the Management of Buried Piping Integrity, Nuclear Energy Institute, NEI-09-14, December 2009
.
14.
Inspection Procedure 62002, Inspection of Structures, Passive Components, and Civil Engineering Features at Nuclear Power Plants, Nuclear Regulatory Commission, December 21, 1996
.
15.
NUREG-1801, Generic Aging Lessons Learned (GALL) Report, Nuclear Regulatory Commission, July 2001
.
16.
Memorandum from G. Jaczko to R.W. Borchard, Nuclear Regulatory Commission
“Evaluation of Buried Pipe at Nuclear Reactor Facilities,”
dated September 3, 2009
.
17.
C.F.
Bowman
,
“Microbe Induced Corrosion of TVA Raw Water Cooling Piping,”
Proc. EPRI Workshop on Microbe Induced Corrosion (MIC)
,
May 1986
,
EPRI
,
Palo Alto, CA
.
18.
“Predictive Model for Non-Microbially Influenced Corrosion Tuberculation—Initial Model for Formation and Growth of Tubercles Caused by Electrochemical Corrosion,”
EPRI Palo Alto, CA
,
December 2003
,
1003442
.
19.
J.H.
Fitzgerald
, III
,
Heating/Piping/Air Conditioning
46
,
3
(
1974
):
p
.
83
.

Author notes

Reprinted from Proceedings of the CORROSION 2012 Research Topical Symposium “Corroson Degradation of Materials in Nuclear Power Reactors—Lessons Learned and Future Challenges” (Houston, TX: NACE International, 2012).

* Structural Integrity Associates, Inc. San Jose, CA 95138-1025. E-mail: bgordon@structint.com.