Materials Transactions, Vol. 55, No. 3 (2014) pp. 506 to 510
© 2014 The Japan Institute of Metals and Materials
Effects of Hydrogen Water Chemistry on Stress Corrosion Cracking Behavior of
Cold-Worked 304L Stainless Steel in High-Temperature Water Environments
Wen-Feng Lu+, Chien-Lin Lai and Jiunn-Yuan Huang
Institute of Nuclear Energy Research, 1000 Wenhua Road, Longtan, Taoyuan 32546, Taiwan
This study measured the stress corrosion cracking (SCC) growth rates of cold-worked 304L stainless steel in oxygenated and hydrogenated
coolant environments. The effects of cold work at the levels of 5, 20, and 30% on SCC were also investigated. The SCC crack growth rate
increased substantially with increases in the degree of cold work in oxygenated water environment. However, after hydrogen injection, SCC
propagation in all cold worked specimens was gradually inhibited. The time to crack arrest decreased with increases in the degree of cold work.
The slip bands caused by cold work facilitated the initiation and growth of transgranular (TG) cracks in specimens exposed to an oxygenated
coolant environment. TG cracks were easier to arrest compared with intergranular cracks after the injection of hydrogen into the coolant system.
[doi:10.2320/matertrans.M2013325]
(Received August 26, 2013; Accepted December 9, 2013; Published February 25, 2014)
Keywords: stress corrosion cracking, cold work, slip bands, hydrogen water chemistry
1.
Introduction
Because of their excellent mechanical properties and
substantial corrosion resistance, austenitic stainless steels
(SSs) are commonly used for in-core and out-of-core
components of nuclear power plants. Nevertheless, stress
corrosion cracking (SCC) has been reported to be a
degradation mechanism for austenitic stainless steels in
pressurized water reactors (PWR) and boiling water reactors
(BWR).1­5) Cold work, such as rolling, bending, and
shrinking, adjacent to welds in plant components is known
to exacerbate SCC susceptibility in high-temperature water
environments.6­10) Slip bands and deformation-induced
martensites caused by cold work have been reported to be
crucial factors that affect the SCC behavior of austenitic
SSs.11­13) Plastic strain and sensitization can be induced in
the heat-affected zone by welding during the component
manufacturing process. The synergistic effects of plastic
strain and sensitization on SCC have been investigated
extensively.8,9,14) In addition to sensitization, the combined
effects that the environment and cold work exert on SCC
have been examined. For example, Tice et al.8) discussed the
influences of loading modes, dissolved hydrogen concentrations, and the pH value (a measure of the acidity or
basicity of aqueous solution) of water environments on the
SCC of cold-worked type 304 SS in PWR primary coolant
conditions. Lu et al. examined the effects of temperatures,
loading modes, and water chemistries on the SCC growth
behavior of cold-worked 316L in simulated BWR environments.6,7) Nakano et al. explored the SCC susceptibility of
cold-worked 304L SS with minor element additions and the
SCC degradation behavior of cold-worked 304L SS in an
irradiated coolant water environment by using slow strain
rate testing.15,16)
Hydrogen injection is widely used to suppress the
corrosion and SCC of BWR plant components because
of the low electrochemical potential in hydrogen water
chemistry (HWC) environments.14,17,18) As stated previously,
+
Corresponding author, E-mail: wfl[email protected]
cold work and HWC are two crucial parameters that influence
SCC propagation in high-temperature water. In recent years,
the effects of SCC growth behaviors have been examined
separately and synergistically; however, few studies have
investigated the fracture mechanism of SCC at the crack tip
for cold-worked austenitic SSs in high-temperature oxygenated or HWC water. In this study, the effects of the degree
of cold work on SCC in oxygenated and hydrogenated
coolant environments were examined using an alternating
current potential drop (ACPD) technique to measure the
SCC growth rates of compact tension (CT) specimens under
trapezoidal loading with a K (stress intensity factor) of
approximately 18 MPa·m0.5. Furthermore, possible fracture
mechanisms at the crack tip were explained using electron
backscatter diffraction (EBSD).
2.
Experimental Procedure
2.1 Materials
The materials used in this study were 304L SS plates that
adhered to the American Society for Testing and Materials
(ASTM) A-240 specifications. The chemical composition
(mass%) of 304L SS is shown in Table 1. The as-received
(AR) specimens were cold rolled (CR) to 5, 20, and 30%
reductions in thickness. Metallographic specimens machined
below the fracture surface were etched in a solution of 10 wt%
H2C2O4·2H2O and examined using optical microscopy.
2.2 SCC tests
According to ASTM E647, the CT specimens were
manufactured with a thickness of 12.5 mm and a width of
60.0 mm, as shown in Fig. 1. Before fatigue precracking, all
specimens were lightly polished with 600 emery paper. The
Table 1
The chemical composition of the AISI 304L stainless steel plate.
Content
Ni
(mass%)
304L
Cr
C
Si
Mn Co
S
Cu
P
Mo
Fe
8.81 18.1 0.022 0.5 1.56 0.1 0.0004 0.48 0.03 0.28 Bal.
Effects of Hydrogen Water Chemistry on Stress Corrosion Cracking Behavior of Cold-Worked 304L Stainless Steel
507
2.3 Fractographic feature observation
The tested SCC specimens were descaled using an
electrolyte of 1 g hexamethylenetetramine in 500 cm3 of 1 N
HCl and then examined using a Hitachi 4800 scanning
electron microscopy (SEM) device to identify the fractographic features. In addition, the cross-section near the
branched cracks was examined using electron backscatter
diffraction (EBSD).
3.
Fig. 1 Schematic diagram showing the dimensions of compact tension
specimen.
Table 2 The conditions of oxygenated and HWC water environments.
Test parameters
Oxygenated
HWC
Pressure MPa
10
10
Temperature °C
300
300
Conductivity (inlet), µS cm¹1
0.057
0.057
Conductivity (outlet), µS cm¹1
0.096
0.071
Oxygen (inlet)
7.1 ppm
<7 ppb
Oxygen (outlet)
6.7 ppm
<4 ppb
ECP (SHE)
0.2 V
¹0.5 V
pH (inlet)
pH (outlet)
7.2
6.2
Autoclave exchange rate
1 time/h
1 time/h
Hydrogen (inlet)
<4 ppb
120 ppb
Hydrogen (outet)
<4 ppb
120 ppb
®
®
CT specimens were fatigue precracked in air through cyclic
loading with a decreasing "K (stress intensity factor range)
pffiffiffiffi
until a precracked length of 3 mm and a "K of 12 MPa m
were reached. The fatigue frequency was set as 10 Hz,
and the stress ratio was set as 0.1 (R, Pmin/Pmax). After
precracking, the specimens were tested using a closed-loop
servo-electric machine with an integrated water circulation
system. Regarding the chemistry of the water in the
circulation loop during the SCC tests, an oxygen level of
7 ppm was maintained for 933 h before hydrogen was
injected to a dissolved hydrogen level of 120 ppb for
1367 h. The conditions of the oxygenated and HWC coolant
environments are summarized in Table 2. The SCC tests were
conducted in the trapezoidal wave loading mode, which
involved loading specimens to a constant level for 2.5 h,
followed by unloading the specimens to an R value of 0.7 in
60 s, and then reloading the specimens again in 60 s. The
initial Kmax for the SCC tests was approximately 18 MPa·m0.5.
The crack length was measured using an ACPD technique.
The final SCC crack length was corrected according to the
average of 20 measurements taken along the crack front of
the fracture surface by using an optical microscope at 20©
magnification, according to ASTM E647 specifications.
Results and Discussion
Optical metallographs of CR 304L SS sampled below the
fracture surface are shown in Fig. 2. The AR specimens
exhibited equiaxial austenite grains with annealed twins.
The CR specimens showed elongated grains, slip bands, and
deformation bands. Increases in the degree of cold work
increased the number of slip bands randomly dispersed in the
CR specimens.
Crack lengths versus test times in the oxygenated and
HWC water environments are shown in Fig. 3. No crack
growth was observed in the AR specimens during the SCC
test. For the CR specimens, an SCC incubation period was
observed during the first 100 h.6,7,14) In the oxygenated
environment, the SCC growth rates were substantially
increased with increases in the degree of cold work. The
crack growth rate of the 5% CR specimen slowed over time
at an average crack growth rate of 9.3 © 10¹11 m·s¹1. The
crack growth rates of the 20 and 30% CR specimens
measured 7.6 © 10¹10 m·s¹1 and 1.1 © 10¹9 m·s¹1, respectively, approximately one order of magnitude higher than that
of the 5% CR specimen. The length of the cracks in the 5,
20 and 30% CR specimens increased by 0.32, 2.51, and
3.84 mm, respectively.
After the water chemistry environment was changed from
an oxygenated water environment to an HWC water
environment, no crack growth was observed in the 5% CR
specimen, and the crack growth rates in the 20 and 30% CR
specimens decreased over time until crack arrest. The time to
crack arrest after hydrogen injection for the 20 and 30% CR
specimens was approximately 1000 and 200 h, respectively.
The total crack length in the 20% CR specimen was 0.54 mm,
with an average crack growth rate of 1.5 © 10¹10 m·s¹1
(0.54 mm/1000 h), and that in the 30% CR specimen was
0.36 mm, with an average crack growth rate of 5.0 ©
10¹10 m·s¹1 (0.36 mm/200 h).
SEM fractographs of the CR specimens are shown in
Fig. 4. Before chemical cleaning, the fracture surface was
covered by oxides that concealed the fracture features, as
shown in Fig. 4(a). After chemical cleaning, fracture features
of intergranular stress corrosion cracking (IGSCC) were
observed in the 5% CR specimens, as shown in Fig. 4(b). A
mixture of IG fracture features and quasi-cleavages was
observed in the 20 and 30% CR specimens, as shown in
Figs. 4(c) and 4(d). Figures 4(b)­4(d) show that the extent of
the quasi-cleavage fracture increased with greater reductions
in thickness. Secondary cracks were also observed in the
SCC areas.
EBSD maps of the cross-section near the fracture surface
of the 20 and 30% CR specimens are shown in Fig. 5. The
dark areas in the 20 and 30% CR EBSD maps denote
508
W.-F. Lu, C.-L. Lai and J.-Y. Huang
Fig. 2 Optical metallographs structures of the SS304L plates cold-rolled to various degrees of reduction: (a) as-received, (b) 5%, (c) 20%,
(d) 30% cold roll.
branched cracks, and the red and green areas denote the
deformation-induced martensites ¡A and ¾, respectively.
Austenites located at slip band intersections transformed into
¡A, and those located at parallel slip bands transformed into
¾ martensites, as shown in Figs. 5(a) and 5(c).19) The £
austenite grains around the branched cracks were marked
using cubic symbols, as shown in Figs. 5(b) and 5(d). The
cracks between the cubic symbols aligned with the same
crystalline orientation were transgranular (TG) fractures,
whereas the cracks with symbols of dissimilar orientation
were IG cracks. As illustrated in the EBSD maps, a mixture
of IG and TG fracture features was observed in the 20 and
30% CR specimens, and the TG fractures preferentially
occurred along the slip bands of high dislocation density
areas.
The effects of cold work at the reduction levels of 5, 20,
and 30% on the SCC behavior of SS were examined in
oxygenated and hydrogenated coolant environments. SCC
susceptibility is aggravated by cold work. Furthermore, the
slip bands caused by cold work provide high-diffusivity
paths or diffusion short circuits for oxygen and iron,11,20) and
deformation-induced martensites accelerate SCC.12,13) The
combined effects that dislocation (high-diffusivity paths)
and deformation-induced martensites exert on SCC require
further investigation.
The fracture surfaces of the 20 and 30% CR specimens
show a mixture of IG fracture features and quasi-cleavages.
The amount of quasi-cleavages increased with greater
reductions in thickness, as shown in Fig. 4. Based on the
EBSD observations presented in Fig. 5 and the deformation
and oxidation mechanisms for SCC,1,21) the SCC growth
processes are illustrated in Fig. 6. At the outset, the crack
propagates along the grain boundary until the tip approaches
the junction of another grain boundary. At the junction, the
crack preferentially propagates along the path with the fastest
oxidation rate, such as the slip band (path A) and the grain
boundary (path B). If the angle ¡ between Path A and the
loading direction exceeds the angle ¢ between Path B and
the loading direction, the crack tip grows along Path A.
With its orientation relative to the loading direction, the oxide
film along Path A sustains substantial tensile stress, which
increases with Angle ¡ to a maximum of 90°, enabling the
crack to preferentially propagate along Path A. Conversely,
IG fractures are observed if Angle ¢ exceeds Angle ¡.
Moreover, the crack tip can grow along both Paths A and B
if Angles ¡ and ¢ are near to each other. According to the
proposed SCC formation process, heavily cold-worked
specimens exhibit a high tendency to fracture transgranularly
in the oxygenated water environments, which might be
because a greater number of slip bands are observed in the
grains.
In the HWC environment, crack arrest occurred in all CR
specimens, and the time to crack arrest for the 20% CR
specimen was substantially longer than that for the 30% CR
specimen. Crack arrest in the HWC environment is attributed
to the low electrochemical potential (ECP).17,18) Considering
the low oxidation rate, the material strength is assumed to
determine the SCC growth rate in the HWC environment.
According to the dislocation and martensite strengthening
mechanisms,22,23) the strength of 304L SS increases as the
CR deformation increases. Compared with cracks that occur
along the grain boundary, crack tips that grow along the
slip bands are proximate to a greater number of dislocations
and deformation-induced martensites in the grains of heavily
cold-worked specimens. Therefore, less time was required
to arrest crack growth in the 30% CR specimen than in the
20% CR specimen. This was because the 30% CR specimen
possessed comparatively greater strength and a higher
tendency to form TG fractures.
4.
Conclusion
In this study, the effect of HWC on the SCC behavior of
cold-worked 304L SS in high-temperature water environ-
Effects of Hydrogen Water Chemistry on Stress Corrosion Cracking Behavior of Cold-Worked 304L Stainless Steel
509
Fig. 4 SEM fractographs of the tested CR specimens: before chemical cleaning (a) 5% CR, after chemical cleaning (b) 5%, (c) 20%,
(d) 30% CR.
AR
5%
20 %
30 %
5.0
HWC
Oxygenated
4.5
Crack length, l / mm
4.0
Crack arrest
200 h, 0.36 mm
3.5
3.0
Crack arrest
1000 h, 0.54 mm
2.5
2.0
1.5
1.0
0.5
0.0
0
500
1000
1500
2000
2500
Time, t / h
Fig. 3 SCC crack length versus test time for AR and CR SS304L in the
oxygenated and HWC water environments.
ments was investigated. The results are summarized as
follows:
(1) The SCC growth rate increased substantially with
increases in the degree of cold work in the oxygenated
water environment. However, the injection of hydrogen
gradually inhibited SCC propagation in all cold-worked
specimens. The time to crack growth arrest decreased
with increases in the degree of cold work.
(2) Prior cold work exacerbates TG cracking in 304L SS.
The slip bands of high dislocation density areas caused
by cold work provide TG crack growth paths in grains.
We inferred that the extent of TG cracking increases as
the degree of cold work increases.
(3) Less time is required to arrest crack growth in the 30%
CR specimen than in the 20% CR specimen when
hydrogen is injected into the coolant because the 30%
CR specimen possesses greater strength and a higher
tendency to form TG fractures compared with the 20%
CR specimen.
Fig. 5 The phase and uni-grain EBSD maps of the cross section near the
fracture surface of the 20 and 30% CR specimens: (a) the phase and
(b) uni-grain EBSD map for the 20% CR specimen, (c) the phase and
(d) uni-grain EBSD map for the 30% CR specimen. The yellow, red and
green areas are £, ¡A, and ¾ phases in the EBSD phase maps and the
transgranular fractures are marked with rectangles.
510
W.-F. Lu, C.-L. Lai and J.-Y. Huang
Fig. 6 Illustration of the SCC growth behavior at the crack tip of cold
worked SS304L.
Acknowledgments
The authors are grateful for the financial support provided
by the Institute of Nuclear Energy Research, Taiwan.
REFERENCES
1) T. Shoji: Proc. 11th Int. Symp. on Environmental Degradation
Materials Nuclear Power Systems-Water Reactors, ANS, (2003)
pp. 588­598.
2) T. Couvant: 6th Int. Symp.: Contribution of Materials Investigations to
Improve the Safety and Performance of LWRs, vol. 1, (2006) pp. 67­
78.
3) MRP: Stress Corrosion Cracking of Stainless Steel Components in
Primary Water Circuit Environments of Pressurized Water Reactors,
(EPRI Palo Alto, CA 2007) MRP-236.
4) K. Kumagai, S. Suzuki, J. Mizutani, C. Shitara, K. Yonekura, M.
Masuda and T. Futami: Evaluation of IGSCC growth behavior of
316NG PLR Piping in BWR, ed. by Y. Y. Wang, PVP, vol. 479,
(ASME, 2004) pp. 217­223.
5) T. M. Angeliu, P. L. Andresen, E. Hall, J. A. Sutliff, S. Sitzman and
R. M. Horn: Proc. 9th Int. Symp. Environ. Degradation of Materials
Nuclear Power Systems-Water Reactors, ed. by F. P. Ford, S. M.
Bruemmer and G. S. Was, (TMS, 1999) pp. 311­318.
6) Z. Lu, T. Shoji, Y. Takeda, Y. Ito and S. Yamazaki: Corros. Sci. 50
(2008) 698­712.
7) Z. Lu, T. Shoji, Y. Takeda, Y. Ito, A. Kai and S. Yamazaki: Corros. Sci.
50 (2008) 561­575.
8) D. R. Tice, S. Nouraei, K. J. Mottershead and J. W. Stairmand: 14th Int.
Conf. (Environmental Degradation of Materials in Nuclear Power
Systems, Virginia Beach, VA, 2009) pp. 23­27.
9) D. Gomez-Briceno, M. Sol Garcia and J. Lapena: 14th Int. Conf.
(Environmental Degradation of Materials in Nuclear Power Systems,
Virginia Beach, VA, 2009) pp. 23­27.
10) L. Tribouilloy, F. Vaillant, T. Couvant, J. M. Boursier, Y. Rouillon, C.
Amzallag, J. M. Olive, M. Puiggali and L. Legras: Proc. 13th Int. Conf.
(Environmental Degradation of Materials In Nuclear Power SystemsWater Reactors, Whistler, British Columbia, 2007) CDROM.
11) S. Lozano-Perez, K. Kruska, I. Iyengar, T. Terachi and T. Yamada:
Corros. Sci. 56 (2012) 78­85.
12) J. Kuniya, I. Masaoka and R. Sasaki: Natl. Assoc. Corros. Eng. 44
(1988) 21­28.
13) T. Shoji: Quantification of Yield Strength Effects on IGSCC in
Austenitic Stainless Steels and its Implication to the IASCC, (EPRI,
Palo Alto, CA, 2002) 1007380.
14) Z. Lu, T. Shoji, Y. Takeda, Y. Ito, A. Kai and N. Tsuchiya: Corros. Sci.
50 (2008) 625­638.
15) J. Nakano, Y. Nemoto, T. Tsukada and T. Uchimoto: J. Nucl. Mater.
417 (2011) 883­886.
16) J. Nakano, Y. Nemoto, Y. Miwa, K. Usami, T. Tsukada and K. Hide:
J. Nucl. Mater. 386­388 (2009) 281­285.
17) M. C. Young, J. Y. Huang and R. C. Kuo: Mater. Trans. 50 (2009) 657­
663.
18) M. F. Chiang, M. C. Young and J. Y. Huang: Effects J. Nucl. Mater. 411
(2011) 83­89.
19) C. L. Lai, L. W. Tsay, W. Kai and C. Chen: Corros. Sci. 51 (2009) 380­
386.
20) H. Mehrer: Diffusion in Solids, (Springer) Ch31, pp. 547­552.
21) T. Shoji, Z. P. Lu and Y. Takeda: Pressure Vessels and Piping Division
Conference, (ASME PVP 2007), Paper No. PVP 2007-26183.
22) W.-S. Lee, C.-F. Lin and B.-T. Chen: J. Mech. Eng. Sci. 219 (2005) Part
C 439­451.
23) M. Milad, N. Zreiba, F. Elhalouani and C. Baradai: J. Mater. Process.
Technol. 203 (2008) 80­85.