Transactions of the Korean Nuclear Society Autumn Meeting
Gyeongju, Korea, November 2-3, 2006
Fatigue Characteristics of STS 304 Stainless Steel
for LNG Storage Tank at Low Temperature
Dosik Kim, Sang-Bok Ahn, Byung-Ok Yoo, Ki-Ha Kim, Yong-Sun Choo and Kwon-Pyo Hong
Irradiated Materials Examination Facility, Korea Atomic Energy Research Institute,
150 Duck-Jin Dong Yuseong-Gu., Daejeon, 305-353, KOREA, kimds@kaeri.re.kr
1. Introduction testing incorporates a personal computer to enable
digital data acquisition and plot the fatigue crack growth
In the engineering applications, many metallic rate (da/dN) versus the stress intensity factor range (∆K)
structures operate below the room temperature, such as graph during the tests.
liquefied gas tank, space shuttles, nuclear fusion
reactors and superconducting machinery [1]. In order to Table 1. Chemical compositions (wt%)
ensure the safety of these structures and to prevent C Mn P S Si Cr Ni
fatigue failure under service conditions, it is of practical 0.05 1.2 0.021 0.008 0.41 18.02 8.6
importance to predict accurately the fatigue life of
components applied at low temperatures. Table 2. Mechanical properties
Among the structural materials having high toughness Temp. 0.2% YS UTS Elong.
and high fatigue strength at room and cryogenic (K) (MPa) (MPa) (%)
temperatures, Type 304 stainless steel is widely used in 293 306.8 720.5 63.2
cryogenic structures due to favorable mechanical 193 496.9 1162.5 32.7
properties, service history and availability. In the 111 550.7 1495.0 28.9
cryogenic applications such as a liquefied natural gas
(LNG) storage tanks, the material will experience 2.2 Specimens and fatigue tests
thermal cycling over a range of room temperature to
111K (the service temperature of LNG tanks). And The compact tension (CT) specimens were 2 mm
cracks already exist, or are assumed to exist, due to thick and 40 mm wide, and three identical specimens
manufacturing and fabrication defects. In such cases, the were prepared for each test conditions. The notch was
life of the structure is determined solely by the crack machined in the L-T and T-L orientations, as defined in
growth rate, and a specific knowledge of fatigue crack ASTM E399-90. The constant amplitude fatigue tests
growth rates (da/dN) is essential for accurate fatigue life were performed with two stress ratios (R= σmin/σmax) of
predictions. 0.1 and 0.5, and a 6Hz sine waveform. Crack length was
In this study, the constant amplitude fatigue crack measured by a compliance method using a clip-on gage
growth tests of the cold-worked STS 304 stainless steel, mounted at the specimen edge. The fatigue crack growth
developed as a material for the membrane of LNG rates were estimated by a seven point incremental
storage tank, were conducted at a temperature range polynomial method [2].
from 293K to 111K. The effects of a stress ratio and a
crack orientation on the fatigue crack growth rate were 3. Results and discussion
estimated experimentally. Fractographic examinations
were also performed to reveal the differences of the 3.1 Fatigue crack growth behavior
fatigue crack growth characteristics at room and low
temperatures. As demonstrated by Paris et al. [3], a fatigue crack
growth rates (da/dN) can usually be described as power-
law function of the stress intensity factor range (∆K)
2. Experimental procedure
2.1 Material and apparatus
da = C (∆K )m (1)
dN
The material used in this study was a 2 mm thick where C and m are material constants that depend on
plate of STS 304 stainless steel produced by POSCO. environment and test variables.
Its chemical composition and mechanical properties are The fatigue crack growth rates at room (293 K) and
shown in Table 1 and 2. Fatigue tests at room and low 110 K are plotted against ∆K on logarithmic coordinates
temperature were conducted on MTS, 100 kN capacity, in Fig. 1. At equivalent ∆K values, da/dN increases with
with a cryostat. The specimens are installed into the an increase in a stress ratio (R) over the testing
clevises, and liquid nitrogen is poured into the chamber temperature. This trend is the same as the test results
to cool the specimen. All the tests began after the reported by Tshegg et al. [4] and Liaw et al. [5]. The
temperature has stabilized at 111, 193 and 273 K. The effect of stress ratio on fatigue crack growth rate is more
1/2
explicit at low temperatures than at room temperature.
There is little or no difference in da/dN in the L-T and
T-L orientations.
-2 -1
1x10 1x10
L-T orientation
L-T orientation
293K (R.T)
111K
Crack growth rate, da/dN (mm/cycle)
-3 -2
Crack growth rate, da/dN (mm/cycle)
1x10 1x10
(a) 293K (b) 111K
1x10
-4
1x10
-3
Paris' law
(※Arrow indicates a macrocrack growth direction.)
Fig. 3 SEM micrographs of the fracture surfaces
m
da/dN = c ∆K
(∆K ≈21.5MPa m1/2)
Paris' law
-5 m -4
1x10 da/dN = c ∆K 1x10
-6 -5
1x10 1x10
R = 0.1, c=6.768e-10, m=3.708 R = 0.1, c=2.905e-15, m=6.474
R = 0.5, c=8.468e-10, m=3.862 R = 0.5, c=5.531e-14, m=6.310
1x10
-7
1x10
-6
4. Conclusion
6 10 100 200 6 10 100 200
1/2 1/2
Stress intensity factor range, ∆K (MPa m ) Stress intensity factor range, ∆K (MPa m )
The fatigue crack growth behavior of the cold-rolled
(a) 293 K (b) 111 K
STS 304 stainless steel developed for a membrane
Fig. 1 The effect of stress ratio on the fatigue crack growth
rate material of LNG storage tank was examined
experimentally at 293K, 193K and 111K. The following
Fatigue crack growth data at 293, 193 and 111 K are conclusions are made.
presented in Fig. 2. In Figure 2(a), the dot line is the test 1. The effect of stress ration (R) on fatigue crack growth
results of SUS 304 stainless steel tested at 77 K by Reed rate (da/dN) is more explicit at low temperatures than at
et al. [6]. The rates at low temperatures are lower than at room temperature. There is little or no difference in
room temperature. Previous researchers indicated that da/dN in the L-T and T-L orientations. The resistance of
this trend is attributed to the extent of strain-induced fatigue crack growth at low temperatures is higher than
martensitic transformation at the crack tip [4,7]. At at room temperature, which is attributed to the extent of
lower ∆K region, the fatigue crack growth resistance is strain-induced martensitic transformation at the crack
higher at low temperatures than at room temperature. tip.
The temperature dependence of fatigue crack growth 2. The temperature dependence of fatigue crack
resistance is gradually vanished with an increase in growth resistance is gradually vanished with an increase
stress intensity factor range (∆K), which correlates with in stress intensity factor range (∆K), which correlates
a decrease in fracture toughness with decreasing with a decrease in fracture toughness with decreasing
temperature. temperature.
3. Fractographic examinations reveal that the
-1
differences of the fatigue crack growth characteristics
1x10
-2
L-T orientation
R = 0.1
1x10
-2 L-T orientation
R= 0.5
between room and low temperatures result in the crack
1x10
Crack growth rate, da/dN (mm/cycle)
closure and the strengthening due to the martensitic
Crack growth rate, da/dN (mm/cycle)
-3
1x10
-3 1x10
Reed's data
transformation.
-4
1x10 -4
1x10
1x10
-5 Paris' law
da/dN = c ∆K
m
1x10
-5
Paris's law
da/dN = c ∆K
m REFERENCES
-6
1x10
293K, c=6.768e-10, m=3.708
1x10
-7 193K, c=3.966e-15, m=6.413
111K, c=2.905e-15, m=6.474
1x10
-6
293K, c=8.486e-10, m=3.862
193K, c=4.484e-13, m=5.788
[1] X.-L. Zheng and B.-T. Lo, Fatigue Crack Propagation in
77K(R=0.125), c=9.51e-12, m=4.34
1x10
-8
1x10
-7
111K, c=5.531e-14, m=6.310
Metals at Low Temperature, Handbook of Fatigue Crack
6 10
Stress intensity factor range, ∆K (MPa m )
100
1/2
200 6 10
Stress intensity factor range, ∆K (MPa m )
100 200
1/2 Propagation in Metallic Structures, Vol.2, p.1385, 1994.
[2] Standard Test Method for Tension Testing of Metallic
(a) R=0.1 (b) R=0.5 Materials, Annual Book of ASTM Standards, E647-95a, 1995.
Fig. 2 The effect of low temperature on the fatigue crack [3] P. C. Paris and F. Erdogan, A Critical Analysis of Crack
growth rate Propagation Laws, Journal of Basic Engineering, Vol.85, p.
257, 1960.
3.2 Fractography [4] E. Tshegg and S. Stanzl, Fatigue Crack Propagation and
Threshold in b.c.c. and f.c.c. Metals at 77 and 293K, Acta
Micrographs taken with a SEM are shown in Fig. 3, Metallurgica, Vol.29, p.33, 1981.
which are typical examples for the fatigue fracture [5] P. K. Liaw, W. A. Logsdon and M. H. Attaar, Cryogenic
appearance tested at 293 and 111 K. The fracture Temperature Near-Threshold Fatigue Crack Growth Rate Data
surface at l11 K is characterized by a very smooth for JBK-75 Stainless Steel, Cryogenics, October, p.523, 1983.
[6] D. T. Reed and R. P. Reed, Materials Studies for Magnetic
appearance. Therefore, it is found that the differences
Fusion Energy Applications at Low Temperatures-Ⅱ,
of the fatigue crack growth behavior at room and low NBSIR79-1609, National Bureau of Standards, Boulder, Co,
temperatures result in the crack closure and the p.81, 1979.
strengthening due to the martensitic transformation. [7] Z. Mei and J. W. Jr. Morris, Influence of Deformation-
Induced Martensite on Fatigue Crack Propagation in 304-
Type Steels, Metallurgical Transactions A, Vol.21A, No.12,
p.3137, 1990.
2/2
