Comprehensive nuclear materials 5 05 corrosion and stress corrosion cracking of austenitic stainless steels Comprehensive nuclear materials 5 05 corrosion and stress corrosion cracking of austenitic stainless steels Comprehensive nuclear materials 5 05 corrosion and stress corrosion cracking of austenitic stainless steels Comprehensive nuclear materials 5 05 corrosion and stress corrosion cracking of austenitic stainless steels Comprehensive nuclear materials 5 05 corrosion and stress corrosion cracking of austenitic stainless steels
Trang 15.05 Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels
U Ehrnste´n
VTT Technical Research Centre of Finland, Espoo, Finland
ß 2012 Elsevier Ltd All rights reserved.
Abbreviations
BWR Boiling water reactor
CGR Crack growth rate
ECP Electrochemical corrosion potential
EPR Electrochemical (potentiokinetic)
reactivation
HAZ Heat-affected zone
HWC Hydrogen water chemistry
IGSCC Intergranular stress corrosion cracking
K Stress intensity factor
K ISCC Threshold stress intensity for SCC
LWR Light water reactor
MIC Microbiologically influenced corrosion
NG Nuclear grade
NMC Nobel metal chemistry
NMCA Noble metals chemistry addition
NWC Normal water chemistry
PLEDGE Plant life extension and diagnosis by
GE (General Electric)
PWR Pressurized water reactor
RBMK Channel type graphite moderated
reactor
SCC Stress corrosion cracking
TGSCC Transgranular stress corrosion cracking
VVER Water-water energetic reactor
5.05.1 Introduction to Austenitic Stainless Steels
Austenitic stainless steels have rendered their vast use because of their good performance in corrosive environments, in addition to their excellent ductility, formability, toughness, and weldability The good corrosion resistance of austenitic stainless steels is mainly due to chromium alloying, resulting in a pro-tective, chromium-rich passive film on the material
in many environments Molybdenum, used as an alloying element in Type 316 stainless steels, further increases the corrosion resistance Chromium and molybdenum are, however, both ferrite-forming ele-ments, and to maintain a fully austenitic structure,
a balance between austenite-stabilizing elements (C, N, Ni, Mn, and Co) and ferrite-stabilizing ele-ments (Cr, Mo, Si, Ti, Nb, Al, V, and W) in solution must be established.1To compensate for the molyb-denum addition in Type 316 stainless steels, the amount of nickel is increased In materials for nuclear environments, the cobalt content is kept as low as possible (<0.02%), because of its strong influence
on radioactivity buildup
Stress corrosion cracking (SCC) is taken into account
in the design codes for light water reactors (LWRs,
93
Trang 2i.e., boiling and pressurized water reactors (BWRs
and PWRs)) through a statement that SCC should not
occur.2Intergranular stress corrosion cracking (IGSCC)
is still by far the largest damage mechanism for
austen-itic stainless steels in oxidizing BWR conditions, and
work on avoiding IGSCC is still going on.3,4 Several
factors affect IGSCC, one of them being sensitization
Sensitization is a result of nucleation and growth of
chromium-rich carbides on grain boundaries, causing
grain boundary chromium depletion
Chromium-depleted grain boundaries are prone to corrosion
and, in combination with a large enough stress and
suitable environment, to IGSCC All measures to
prevent sensitization are therefore taken in all steps
of component manufacturing and plant operation
Concerning the chemical composition, this can be
done by reducing the carbon content to levels below
0.03%, as is done in Types 316L, 316LN, and 316NG
stainless steels, where the carbon content is
typi-cally in the order of 0.02% Since carbon is a
strengthening element, nitrogen is added to these
steels to still achieve good mechanical properties
Nitrogen also reduces chromium-rich carbide
forma-tion, a concept that is utilized in the French RCCM
norms, which allow a carbon content of 0.035% in
their nitrogen-strengthened Type 316 stainless steel
with0.08% N The other approach to avoiding
sen-sitization is to tie up the carbon into precipitates
This is utilized in stabilized stainless steels, which
are of two main categories, titanium- and
niobium-stabilized stainless steels, Type 321 and 347,
respec-tively To ensure that the carbon is tied into Ti(C,N)
or Nb(C,N) precipitates, a high enough stabilization
ratio, that is, Ti/C or Nb/C above 5 or 10,
respec-tively, is specified
Good corrosion resistance is ensured by
restrict-ing the amount of harmful elements, especially sulfur
and phosphorus, which may cause intergranular
corrosion when segregated to grain boundaries
Sev-eral standardized grain boundary corrosion tests,
such as the Strauss test5 and the electrochemical
potential reactivation (EPR) test,6are employed
rou-tinely as part of acceptance tests for materials and
components
5.05.1.1 Types, Mechanical Properties, and
Microstructures
The chemical compositions and mechanical
proper-ties of the most common stainless steels are presented
summarized inTable 2
In addition to compositional, mechanical, and cor-rosion resistance requirements, several other require-ments are put on austenitic stainless steel materials for nuclear components These include, for example, requirements on grain size A small enough grain size is needed to enable reliable nondestructive inspec-tion requirements, using ultrasonic techniques A com-mon requirement is that the grain size must not exceed ASTM number 4.0, which corresponds to an average grain size of 90mm The grain size of stabilized stainless steel components is typically smaller than this
Stainless steels are generally welded with a slightly over-alloyed filler metal to ensure good corrosion resistance of the final joint The weld shall contain
a small amount (>3% but <10%) of d-ferrite to avoid solidification and liquation cracking.7Welding induces residual stresses, which together with the operational stresses enhance crack initiation and growth The aim is, naturally, always to minimize the residual stresses by a proper choice of welding parameters, by securing a good fit between the parts
to be welded, etc The use of a narrow-gap welding technique has increased remarkably during the last few decades The narrow-gap welding method has many advances as it results, for example, in a lower level of residual stresses, a reduced weld volume, a narrower heat-affected zone (HAZ) with lower risk of sensitization, and less grain growth.8
Steels in BWRs and PWRs Austenitic stainless steel is the main construction material in nuclear power plants (NPPs) owing to its good corrosion resistance, ease to manufacture dif-ferent shapes, and good weldability In BWRs, stainless steels are used for piping and reactor pressure vessel cladding and structures inside the pressure vessel, including the core shroud (which separates the pri-mary water upward flow through the core from the downward flow in the annulus), the core plate (which supports the bottom of the fuel), the top guide (which aligns the top of the fuel bundles), the shroud dome, the steam separators, etc Austenitic stainless steel is also largely used for other components such
as pumps, valves, shafts, sleeves, and in auxiliary sys-tems such as water tanks
The material choices for PWRs are essentially the same The steam generator (SG) tubes are made of
Trang 3Table 1 Chemical composition of common stainless steel alloys
alloys a
DIN 1.4301
1.4306
1.4311
DIN 1.4436/DIN 1.4401
DIN 1.4404/DIN 1.4435
1.4429
min Ti
SS 2337 DIN 1.4541
min Nb
SS 2338 DIN 1.4550 Source: Wegst, C E Stahlschluessel Key to Steel; Stahlschluessel Wegst GmbH: Marbach, 1995.
a The corresponding alloys according to the Swedish SS and the German DIN standards are also given.
Trang 4Ti-stabilized stainless steel in the Russian designed
Water–water energetic reactors (VVERs), which are
PWRs with slightly different water chemistry and
horizontal instead of vertical SGs, as in Western
PWRs The SG tubes in western PWRs are made
of nickel-based materials (Alloy 600, 690, or 800)
Further, both the SG vessel and the pressurizer are
clad with austenitic stainless steel
5.05.2 Stress Corrosion Cracking
SCC is a failure mode caused by a combination of a
susceptible material, stresses, and an aggressive
envi-ronment (Figure 1) There are two modes of SCC in
austenitic stainless steel, namely intergranular and
transgranular stress corrosion cracking (IGSCC and
TGSCC) IGSCC in austenitic stainless steel is the
major failure mode in BWRs, while it has not been
considered as a plausible failure mode in PWR
pri-mary water under normal operation conditions
However, the number of IGSCC cases in PWRs has
increased by time, showing that PWRs are not totally
immune to IGSCC
Most TGSCC cases are due to chloride-induced
SCC TGSCC is rare in the primary system, and
the failure cases are typically observed in auxiliary
systems
In this section, the main factors affecting IGSCC in
BWR environment are reviewed, that is, degree of
sensitization, deformation, electrochemical corrosion
potential (ECP), water purity, and stress Also several
other parameters affect IGSCC susceptibility, but a comprehensive description of these is out of the scope
of this chapter Among these parameters are tempera-ture, hydrogen, mechanisms related to localization
of deformation and to corrosion deformation interac-tions, such as effect of strain rate, subtle differences in chemical composition, dynamic strain aging, dynamic recovery, vacancy injection, selective dissolution, grain boundary segregation, and relaxation.9–21
5.05.2.1.1 Degree of sensitization Sensitization is the result of nucleation and growth of chromium-rich carbides M23C6at the grain bound-aries, which results in a depletion of chromium at the grain boundaries because of faster diffusion rate
susceptibility
Stress
Region of potential stress corrosion cracking
Figure 1 The classic presentation of stress corrosion cracking includes the three circles: material, environment, and stress Reproduced from General Electric Company Alternative Alloys for BWR Pipe Application; NP-2671-LD, Final Report; San Jose, CA, 1982, with permission from BWR Owners Group.
Table 2 Minimum room temperature mechanical properties of stainless steels (hot finished and/or annealed forging)
(min MPa)
0.2% yield strength (min MPa)
Elongation (min.%)
Reduction
in area (min.%)
Source: Davis, J ASM Specialty Handbook; ASM International: Materials Park, OH, 1994; ASM Standard A 240.
Trang 5along the grain boundaries compared to that within
the grain interior Chromium-rich carbides form
within the temperature range of 500–750C, but
continue to grow down to much lower temperatures
Sensitization can therefore occur as thermal
sensiti-zation during heat treatment and welding or as
low-temperature sensitization during long-time exposure
to LWR temperatures, below the chromium-rich
car-bide precipitation temperature.22 In the latter case,
the nucleation of carbides must have occurred
previ-ously, and the nucleated carbides grow during the
long-time exposure and deplete the grain boundaries
of chromium The degree of sensitization is typically
measured using the EPR test, which is sensitive to the
area where the grain boundary chromium content is
below 15%, and is, thus, not a true measure of the grain boundary chromium content
The time for carbide precipitation increases as the carbon level decreases as seen inFigure 2 Nitrogen alloying delays carbide precipitation (Figure 3), while deformation accelerates diffusion and precipi-tation The obvious remedy to avoid sensitization
is, thus, to decrease the amount of free carbon, as explained earlier, by reducing the carbon content, or
by tying carbon to Ti- or Nb-carbides and by restricting the degree of deformation
IGSCC in sensitized stainless steels occurs typi-cally in the weld HAZ at a distance of 4–8 mm from the fusion line, at the location where a high degree of sensitization combined with high residual stresses
Time (min)
304 (0.053% C)
1000
900
800
700
600
500
1800
1600
1400
1200
1000
Figure 2 Time–temperature–precipitation diagram for stainless steels with different carbon contents Reproduced from Shah, V N.; MacDonald, P E Aging and Life Extension of Major Light Water Reactor Components; Elsevier:
Amsterdam, 1993.
Time (h)
1000 900 800 700 600 500
1832 1652 1472 1292 1112 932
0.039% N
0.069% N 0.145% N 0.247% N
Figure 3 Effect of nitrogen on precipitation of M 23 C 6 in a 0.05C–17Cr–13Ni–5Mo stainless steel Reproduced from Peckner, D.; Bernstein, I M Handbook of Stainless Steels; McGraw-Hill: New York, 1977; pp 4-35–4-53, pp 751–757.
Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels 97
Trang 6results in most severe conditions for IGSCC The
typical location of IGSCC is different in
nonsensi-tized stainless steels, where IGSCC occurs very close
to the fusion line, within the first few grains
5.05.2.1.2 Deformation
Deformation increases the susceptibility of stainless
steels to IGSCC, in sensitized as well as in
nonsensi-tized stainless steels, where the role of strain is
deci-sive Deformation occurs as bulk cold work from
rolling, bending, grinding, etc., as surface cold work
from machining, grinding, etc and as weld shrinkage
from welding Weld shrinkage can lead to up to 25%
equivalent room temperature strain in the weld HAZ
very close to the fusion boundary, and this is also the
location of observed IGSCC cracking in
nonsensi-tized stainless steels pipes.23–27 The importance of
cold deformation in IGSCC is shown in Figure 5
Deformation is estimated to be the main affecting
parameter in 50% of all IGSCC cases covered in
the survey (including sensitized and nonsensitized
stainless steels) The effect of deformation has been
studied using bulk-deformed materials,28–35and the
results show a correlation between IGSCC crack
growth rate (CGR) and yield strength (Figure 6)
Much effort is nowadays put on deformation in terms of restrictions on bulk and surface deformation and on the development of sophisticated surface treat-ment procedures to remove surface cold work at criti-cal locations.36,37Application of narrow-gap welding results both in a decrease in the degree of deformation
in the HAZ and in lower residual stresses It should be pointed out that some components, such as bolts, can
be made of intentionally cold-worked stainless steel to increase the material strength
5.05.2.1.3 Environment
As mentioned earlier, one of the main reasons for the good behavior of austenitic stainless steels in LWR conditions is the formation of a protective passive film in high-temperature water (around 300C) The oxide film formed in high-temperature water has a double-layered structure The inner layer grown on the metal surface consists of a chromium spinel or magnetite and is covered by an outer layer
of magnetite or Fe–Ni spinel precipitated from the aqueous phase.38,39The double-layered oxide struc-ture forms so that faster diffusing elements pass through the inner layer to the outer layer while the slower diffusing elements, such as chromium, remain
in the inner layer and therefore the outer layer con-tains mainly of iron and the inner layer is enriched with chromium Although consensus is not yet reached
on the mechanistic details for corrosion and SCC in LWR environments, breakage of the passive film is generally considered to be of major importance because of the fact that if the oxide film breaks, the corrosion rate is high until passivation occurs.28,40,41 The CGR of IGSCC is highly dependent on the oxidizing power of the environment, that is, the ECP
oxygen content increases in the high-temperature
0
5
10
15
20
Distance from weld fusion line (mm)
Various BWR stainless steel
weld HAZs
Figure 4 Deformation versus distance from the weld
fusion line in various stainless steel weld HAZs Deformation
is expressed in terms of equivalent tensile strain at room
temperature, and results from weld shrinkage strains
during welding Reproduced from Andresen, P L.; et al.
In Corrosion 2000, NACE 55th Annual Conference,
Orlando, FL, Mar 26–31, 2000; p 12, Paper No 00203,
with permission from BWR Owners Group.
Cold work Chemistry Material Nuclear grade Ni-base alloy Weld repair Residual stresses Sensitization
Figure 5 Cause of IGSCC in Swedish nuclear power plants Cold work is the biggest singular parameter affecting IGSCC Gott, K Personal communication, April 2010.
Trang 7water, but it is not a linear relationship, and small
changes in oxygen concentration can result in large
changes of ECP and CGR Important to notice is
that the correlation between CGR and ECP is
differ-ent for sensitized stainless steels and deformed,
non-sensitized stainless steels, which show much higher
CGRs compared to sensitized materials at low
poten-tials (although lower than at high ECP)
Two solutions to lower the ECP in BWRs have
been developed, that is, hydrogen water chemistry
(HWC) developed by General Electric and Asea
Brown Boveri, and noble metal chemistry addition
(NMCA, NobleChem™), developed by General
Elec-tric The ECP in a BWR recirculation circuit during
normal operation and using normal water chemistry
(NWC), that is,200 ppb oxygen, is above 100 mVSHE
The ECP is still higher in the core because of
radio-lytic decomposition of water forming hydrogen
per-oxide, H2O2 The ECP is remarkably lower in plants
using either HWC, where 40–250 ppb hydrogen is
added to the feed water, or noble metal chemistry
(NMC™), where a small amount of platinum is
added to the reactor water either at about 130C
during startup or during full power operation
(OnLine NobleChem™), creating an electrocatalytic
surface layer.42,43 The ECP of the buffered PWR
environment is in the lower range of the ECP curve,
that is, about600 mVSHE The trend, especially for
the US BWR fleet is toward HWC and NMCA None
of the US BWR plants operates on NWC, and 75% apply NMCA.44The majority of the European BWRs operate on NWC
Water purity has a profound effect on both crack initiation and CGR in oxidizing environments The main concerns are chlorides and sulfates for SCC and additionally copper for pitting corrosion (although Cu also has a synergistic effect on SCC) Sulfate and/or chloride levels already in the ppb-range increase the IGSCC susceptibility Power plants monitor online the conductivity, which is a mirror for water purity, and analyze the amounts of impurities on regular basis from grab samples The conductivity of BWR primary water of today has been reduced from a typical range
of about 0.4mS cm1in the 1970s to 0.1–0.2mS cm1 (the conductivity of theoretically pure water is 0.056mS cm1)
Dissolved oxygen is consumed inside cracks and crevices, and the local ECP is reduced to low levels, creating a potential gradient between the outer sur-face and the crack tip This results in migration of anions into the lower potential area, which results in very high anion levels in the crack despite low levels
in the surrounding environment.45Further, the local environment in a crack can remain aggressive for a long time after, for example, short periods of higher impurity levels in the bulk environment
It is not only the environment during steady-state operation that needs attention, but also the
1E – 08 1E – 07 1E – 06
Yield strength (MPa)
Two sensitized points for comparison
Annealed cold worked
Unsensitized 304, 304L and 316L SS and A600
288 ⬚C high purity water, 2000 ppb O 2
CT tests at 27.5–30 MPa m½ Circles = High martensite SS Triangles = Alloy 600
Very high martensite
Very low or no martensite
Predicted response
Figure 6 Effect of yield strength (and martensite content) on stress corrosion crack growth rate of unsensitized stainless steels in oxygenated, high purity water The predicted response is based on the PLEDGE model (plant life extension and diagnosis by GE) Reproduced from Andresen, P L.; et al In Proceedings of 11th International Symposium on Environmental Degradation of Materials in Nuclear Power Systems – Water Reactors, Stevenson, WA, Aug 11–14, 2003; American Nuclear Society: La Grange Park, IL, 2003; p 435, with permission from BWR Owners Group.
Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels 99
Trang 8environment during shutdown, downtime periods,
and during startup The possible role of these will
increase with plant age and amount of shutdowns and
startups
5.05.2.1.4 Stress
The stresses causing IGSCC are a combination of
residual and operational stresses, although the first is
considered more decisive in IGSCC failures This is
because operational stresses are kept low by design
and components are usually designed to operate
below 80% of their yield strength The CGR of
intergranular stress corrosion cracks increases with
increasing stress intensity factor (K ) (Figure 8) The
effect of stress intensity on CGR varies depending on
the material and environment Knowledge of the
dependency betweenK and CGR is very important
for structural integrity calculations, which are made
to show that flaws, either postulated or detected using
nondestructive inspections, are tolerable and do not
pose a safety risk Huge efforts have been put on the
production of high-quality laboratory CGR data and
efforts are still going on Approved relationships (i.e., agreement reached between national safety authority and plant operators) are called disposition lines, and examples of published lines are shown inFigure 8 Several methods to mitigate IGSCC have been applied over the years, such as last pass heat sink welding, mechanical stress improvement, and weld overlay cladding.46All these aim at producing a com-pressive stress state in the HAZ However, these methods are usually applied as temporary remedies Measurement of residual stresses is an area of increased focus nowadays, and lack of knowledge can result in excessive under- or over-conservatism in design and in structural integrity calculations Also other stress-related factors affect IGSCC, such as vibratory loading, thermal loads from, for example, stratification, as well as load cycles during shutdowns and startups Much effort was earlier put on defining theKISCC; that is, the stress intensity, below which SCC would not occur With improved laboratory testing techniques, lower and lower KISCC values have been measured and a true threshold value may not exist
1E -09
1E-08
1E-07
1E-06
1E-05
–0.6 –0.5 –0.4 –0.3 –0.2 –0.1
Corrosion potential (VSHE)
–1 )
0.0 0.1 0.2 0.3 0.4
Sensitized 304 stainless steel
30 MPa m ½ , 288 ⬚C water
0.06–0.4 μS cm –1 , 0–25 ppb SO4
SKI Round Robin Data
Filled triangle = Constant load
Open squares = Gentle cyclic
42.5 28.3 14.2 μmh –1
GE PLEDGE
predictions
30 MPa m½
sens SS
0.5
0.1 0.25
0.06 μS cm –1
0.1μS cm –1
Means from analysis of
120 L sens SS data 0.06 μS cm –1
2000 ppb O2
200 ppb O2 Ann 304SS
316L (A14128, square)
304L (Grand Gulf, circle)
non-sensitized SS
50%RA 140C (black)
10%RA 140C (grey)
20% CW
A600
20% CW A600
4 dpa 304SS
GE PLEDGE predictions for unsens SS (upper curve for 20% CW)
(a)
10−1 (b)
10−6
10−7
10−8
10−9
Solution conductivity (μS cm –1 )
Predicted curves from PLEDGE code for typical range in ECP
316L stainless steel
25 mm CT specimen constant load
288 ⬚C water Test conditions:
0 ⬚C cm-2 EPR
ª27.5 MPa m½
200 ppb O2
Figure 7 Summary of crack growth rates of sensitized stainless steels versus corrosion potential, ECP (a) (reproduced from Andresen, P L.; Morra, M M J Nucl Mater 2008, 383(1–2), 97–111) and for nonsensitized stainless steels versus solution conductivity (b) (reproduced from Andresen, P L Corrosion 1988, 44(7), 450) The prediction curves for different water conductivity levels are according to the PLEDGE model, with permission from BWR Owners Group RA ¼ reduction
in area; CW ¼ cold work.
Trang 95.05.2.1.5 Components at risk
The earliest incidents of SCC in BWRs occurred in
stainless steel fuel cladding, before zirconium alloys
were used.47IGSCC plagued the BWRs in the 1970s
and caused a clear reduction in capacity factors
Crack-ing was first observed in the recirculation and water
cleanup systems in pipes with small diameter and later
also in larger diameter pipes The material was mainly
Type 304 with a high carbon content of0.6% Owing
to large efforts to solve the problem, including the
development of Type 316NG, narrow-gap welding
technique, as well as low-potential water chemistries,
the number of IGSCC incidents has remarkably
reduced.3In the late 1980s, cracking in Ti-stabilized
stainless steel piping was detected.48–50 Robust
miti-gation measures, including adoption of narrow-gap
welding, change of material to Nb-stabilized stainless
steel with higher stabilization ratio requirements, and
reduction of the amount of welds, were applied in
Germany to solve the problem Also from Russian
channel type graphite moderated reactors (RBMKs),
which operate under BWR-like conditions, numerous
IGSCC cases have been reported.51,52 In the 1990s,
the first cases with IGSCC in nonsensitized stainless
steels were reported in BWRs,23,25 first in pipings,
and later numerously in core shrouds.27Deformation
(weld shrinkage in piping and surface grinding in
the core shrouds) is considered to be of major impor-tance in these cases
PWRs operate at low corrosion potentials and very low oxygen levels, <30 ppb The risk of IGSCC
in austenitic stainless steels in nonoxidizing envi-ronment is, thus, much lower than in NWC BWR environment.31,53Incidences with IGSCC under nom-inal PWR conditions have not been reported Oxygen can, however, be enclosed in certain situations, such as startups, and can lead to IGSCC in austenitic stainless steels Although the number of IGSCC cases in PWRs
is still very low, the number seems to be increasing.54,55 IGSCC has been observed in pressurizer heater sleeves, canopy seals in the control rod drives, SG safe-ends, etc Laboratory tests on cold-worked stain-less steels show that IGSCC is possible also in normal PWR environment, indicating (although not gener-ally accepted) that more IGSCC failures may occur in the operating PWR plants with time
Environments Austenitic stainless steels are prone to TGSCC when exposed to aggressive oxidizing water, for example,
1E – 11 1E – 10 1E – 09 1E – 08 1E – 07 1E – 06 1E – 05 1E – 04
Stress intensity (MPa m ½ )
–1 )
NRC 0313, BWR NWC MD-01, BWR NWC JSME, BWR NWC JSME, BWR HWC
Figure 8 Crack growth rate versus stress intensity according to dispositions lines for sensitized stainless steels in normal water chemistry boiling water reactors environment Compiled by author from NRC Generic Letter GL880011 NRC Position on IGSCC in BWR Austenitic Stainless Steel Piping; January 25, 1988; http://www.nrc.org; Jansson, C.; Morin, U.
In Proceedings of 8th International Symposium on Environmental Degradation of Materials in Nuclear power Systems – Water Reactors, Amelia Island, FL, Aug 10–14; American Nuclear Society: La Grange Park, IL, 1997;
pp 667–674; Kobayashi, H.; Kashima, K Int J Press Vess Pip 2000, 77, 937–944 JSME is the Japan Society of Mechanical Engineers; NRC is the Nuclear Regulatory Commission.
Corrosion and Stress Corrosion Cracking of Austenitic Stainless Steels 101
Trang 10containing chlorides, under sufficiently high stresses.
As chloride levels in BWR and PWR primary systems
are kept low, TGSCC is rare under normal operating
conditions The exception is TGSCC from the
sec-ondary side in VVER SG tubes, manufactured of
Ti-stabilized stainless steels Condenser leakages,
fre-quently reported in VVERs, can result in chloride
contamination of the secondary water and eventually
in TGSCC in the SG tubes Copper release from brass
condensers and crevices formed by iron deposits
fur-ther enhance both TGSCC and pitting corrosion on
the secondary side of the SG tubes
A risk of TGSCC exists at locations where (a slow)
buildup of aggressive conditions can occur The risk
of TGSCC increases with plant age, as the buildup of
aggressive conditions can be very slow and can occur
at unknown (uninspected) locations Known chloride
sources are old insulation and sealing materials (e.g.,
asbestos), leakage from cables, polymers, paints,
con-crete, etc Wet insulation is the worst of these, as
the insulation provides crevice conditions in addition
to a chloride source Strict regulations for expendables
(grease, cleaning agents, sealing materials, etc.) allowed
in NPPs are applied to reduce the risk of buildup of
aggressive conditions All bare outer surfaces of
aus-tenitic stainless steel components, where humidity
may exist, can be at risk for TGSCC
TGSCC has been reported in valves,56for
exam-ple, where the source for the chlorides is assumed to
be asbestos sealing used early in time and in water
tanks, where chlorides probably stem from humidity
and concentration buildup at the waterline Stainless
steel bellows in the BWR reactor containment are, in
principle, at risk because of the high degree of cold
work in the bellows However, no SCC has been
reported in these New components can also be at
risk for TGSCC, if proper measures are not taken to
avoid contamination of components during
transpor-tation, storage, and installation
TGSCC can occur in oxidizing concentrated boric
acid solutions although laboratory results are not fully
conclusive whether chloride is also needed or not.56–58
5.05.3 Pitting Corrosion
Pitting corrosion occurrence has several similarities to
TGSCC, that is, it requires oxidizing conditions and
presence of water with harmful ions, such as chlorides,
fluorides, sulfates, and/or copper, but no stress is needed
The Type 304 stainless steel is more prone to pitting
corrosion than Type 316 stainless steel Pitting corrosion
is very often observed at same locations as TGSCC, but pitting corrosion can also occur without SCC and vice versa The risk of pitting corrosion under normal BWR conditions is extremely low However, pitting corrosion can occur in pressure boundary systems at locations where (slow) buildup of aggressive local conditions can occur Such locations are, for example, areas with low water flow, dead ends, and valves with sealing
As pitting occurs only in oxidizing conditions, it is not a plausible degradation mechanism in PWR pri-mary water under nominal environmental conditions However, the environment may be oxidizing both locally and/or temporarily because of startups, for example Different systems during shutdown may be filled with air, and this may cause air pockets during startup The oxygen from air will then dissolve into the primary water and local oxidizing conditions temporarily emerge until the oxygen is consumed by the oxidation of metal surfaces The risk of pitting corrosion (and TGSCC) is, however, highest in auxil-iary systems, for example, at outer surfaces, where the temperature is low enough for condensation to occur Thus, pitting corrosion can occur at nominally dry locations Accumulation of aggressive local conditions
is enhanced by crevices
The sources of chlorides were listed earlier Sulfate sources have been introduced earlier, for example, in molybdenum disulfide greases, but since the harmful influence of this material was iden-tified, it is not an allowed expendable material Again, copper can enter the system from copper-containing structural components
Pitting corrosion is seldom considered to pose a safety problem, as the wall thicknesses of pressure boundary components are usually large enough to sustain pitting corrosion for long times without leak-age However, pitting corrosion is always an indica-tion of a harmful environment existing at the locaindica-tion and is often associated with the risk of TGSCC, which can cause wall cracking in short time periods Pitting corrosion enhances the risk of SCC as the pits increase the local stress concentration and thus act as crack initiators Observation of pitting corrosion shall therefore not be omitted as insignificant
5.05.4 Microbiologically Induced Corrosion
A rather rare corrosion mode is microbiologically induced corrosion, or nowadays, microbiologi-cally influenced corrosion (MIC) MIC is normal