Engineering Failure Analysis 33 (2013) 439448

Contents lists available at SciVerse ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

On a leak-before-break assessment methodology for piping systems of fast breeder reactor

Yeon-Sik Yoo a, Nam-Su Huh b,
Korea Atomic Energy Research Institute, 989-111 Daedeok-daero, Daejeon 305-353, Republic of Korea Department of Mechanical System Design Engineering, Seoul National University of Science and Technology, 172 Gongreung 2-dong, Nowon-gu, Seoul 139-743, Republic of Korea
b a

a r t i c l e

i n f o

a b s t r a c t
The Leak-Before-Break (LBB) concept has an effect on the safety design of Fast Breeder Reactors (FBRs), and thus its assessment has been one of the most signicant issues. In the case of a commercial-scale FBR, since the main loads are the thermal expansion and thermal transient stresses, ferrite steel with a low thermal expansion rate has been a candidate material. Moreover, thin-walled and large-diametric pipes have been used to reduce the number of loops, which might also result in an economical advantage. A conventional LBB assessment method is insufcient to consider these characteristics, thus an advanced method is required. In this context, in the present paper, the following points were proposed to apply the LBB assessment method to ferrite steel pipes with thin walls and large diameters: (1) The surface resistance correction factor against a ow through penetrated cracks was improved for a reasonable leakage assessment under low-pressure. (2) The R6 method was applied to an unstable fracture assessment for postulated cracks. (3) A buckling assessment was introduced in determining the critical crack length for elbows. The applicability of this proposed method has been veried through an LBB assessment on typical ferrite steel pipes. 2013 Elsevier Ltd. All rights reserved.

Article history: Received 2 June 2013 Accepted 20 June 2013 Available online 30 June 2013 Keywords: Buckling Creep Elbow Fast breeder reactor Leak-Before-Break

1. Introduction Application of the Leak-Before-Break (LBB) concept has an important role in realizing a reasonable design of Fast Breeder Reactor (FBR) plants with the assurance of safety and economy. In the case of commercial-scale FBRs, a low-pressure conditions and the use of ductile materials have an advantage in the sense that the risk of unstable fracture in key structures can be virtually eliminated [1]. For example, double-ended guillotine breaks in pipes can be prevented by detecting a leakage that occurs at the through-wall stage of a crack opening under these conditions. A consideration of these LBB characteristics can enlarge the degree of design freedom with respect to the materials, piping layouts, and safety measures against a leakage, thus leading to improved plant reliability and economy. To apply the LBB concept to FBR structures effectively, it is essential to notice their design characteristics. The representative inuence factors of the FBR design characteristics on the LBB assessment are shown in Fig. 1. The FBR plants have realized a comparatively good generating efciency under normal operating conditions of high temperature and low-pressure by virtue of coolant with a high boiling point [1]. Under high temperature, creep is not a negligible aspect for a defect assessment, and thus creep-fatigue crack growth evaluations considering the stress relaxation owing to creep and cyclic plant operating conditions were performed in this paper by reviewing the present defect assessment guidance [2]. Generally, a low-pressure condition makes leakage detection uneasy, and thus it may result in an unsafe leakage evaluation. For this
Corresponding author. Tel.: +82 2 970 6317; fax: +82 2 974 8270.
E-mail address: nam-su.huh@seoultech.ac.kr (N.-S. Huh). 1350-6307/$ - see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.engfailanal.2013.06.017

440

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

FBR Design Characteristics

Considerable Factors for LBB Assessment

R&D Issues

Due to Sodium Coolant Low Pressure Low Leakage Precise Considerations in Leakage Evaluation

High Temperature

Creep-fatigue

Creep-fatigue Crack Growth Evaluation

Due to Cost Reduction

Pipe Length Minimization

Ferrite Steel against Thermal Stress

Unstable Fracture Evaluation

Loop Number Minimization

Thin Walled and Large Diametric Structures

Bucking Evaluation in Elbow

Fig. 1. Inuence factors of FBR design characteristics on LBB assessment.

reason, careful considerations are required to determine the appropriate leakage evaluation techniques for an LBB assessment of FBR structures. Meanwhile, an innovative FBR design is ongoing to minimize the piping length and number of loops for cost reduction. The minimized piping length is considered to increase the thermal expansion stress under high temperature. As a solution to the thermal stress problem, ferrite steel with a small thermal expansion has been recommended for FBR structures. Actually, thermal stress can be classied into the secondary stress of a displacement-controlled type, which is well known to reduce the risk of unstable fractures in the piping structures [3]. However, it is thought that the fracture assessment as a critical failure mode is necessary to conrm the LBB concept of FBR structures. As shown in Fig. 1, a minimized number of loops and low internal pressure lead FBR structures for thin-walled and largediametric piping systems. Another comprehended critical failure mode in these structures is considered to be buckling rather than an unstable fracture, and the representative object is an elbow with an axial crack. Based on these considerations, the present paper proposes a prospective LBB assessment methodology applicable to commercial-scale FBR piping structures with an emphasis on leakage evaluation under low-pressure condition, unstable fracture assessment taking into account material resistance to crack growth, and the possibility of buckling as a LBB assessment category in thin-walled and large-diametric elbow structures. 2. Proposed LBB assessment methodology for FBR piping structure 2.1. LBB assessment ow The LBB assessment ow in a FBR structure with a postulated crack can be constructed as in Fig. 2 based on the LBB definition. Fig. 2 indicates that the LBB behavior arises and the fracture mode corresponds to a partial penetration when the detectable crack length due to leakage evaluation is smaller than the crack length at a critical state. The crack length caused by creep and fatigue at penetration is naturally deduced to be smaller than the detectable crack length, considering the crack growth under plant operating conditions. The critical crack length can be obtained from the fracture assessment considering the material and structural characteristics. 2.2. Creep-fatigue crack growth evaluation Creep is a meaningful damage factor, together with fatigue, for crack growth in FBR structures subjected to high temperature. Generally, creep-fatigue crack growth may be evaluated from the following equations, which neglect the interaction between creep- and fatigue-induced damages:

da=dN da=dN c da=dNf da=dN c C c DJ c mc da=dN f C f DJ f mf 1

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

441

Penetrated Crack Length Due to Creep-fatigue Crack Growth Evaluation : 2Cp

Critical Crack Length Due to Failure Mode Assessment : 2Ccr

Detectable Crack Length Due to Coolant Leakage and Crack Growth Evaluation : 2Cd

Cd < Ccr

No

Guillotine Break

Y es

LBB

Leak Rate Assessment

Fig. 2. Current LBB assessment ow.

where da and dN denote the increments of crack growth and number of cycles, respectively, and Cf and mf denote the coefcient and exponent of Paris law to fatigue crack growth, and Cc and mc denote those of creep crack growth. DJ denotes the J-integral range in a cracked structure, and various studies have been performed for evaluating parameter J in the elasticplastic region [2,4]. The initial crack size for a creep-fatigue crack growth evaluation may be determined by adding a safety margin of 2 to the detectable limit of ultrasonic testing (UT) at pre-service inspection (PSI). From the safety margin and capability of UT in FBR structural components, an initial semi-elliptical crack whose depth (a0) and length (2C0) are 0.2 and 1.0 times the wall thickness (t) is assumed to be located circumferentially on the inner surface in the case of a straight pipe. To obtain the crack length at a penetration (2Cp) from crack growth evaluation, the cyclic stresses throughout the lifetime should be considered. These stresses are induced by the following loading conditions in FBR plants: Internal pressure, Thermal expansion, and Thermal transients. Under these loading conditions, internal pressure and thermal expansion have their maximum values under a normal operating state, and thermal transients occur as transient events. 2.3. Detectable crack evaluation at leakage Considering that coolant leaks out at the outer crack surface after penetration by the initial inner crack, it is necessary to evaluate the outer crack length corresponding to a threshold leakage. In the case of FBR structures, the threshold leakage is known to be 200 g/h, which results from the detection capability of 100 g/h and a safety factor of 2. Fig. 3 indicates that a detectable crack length (2Cd) at the inner surface can be evaluated from the crack growth behavior after penetration and the outer crack length corresponding to a leakage of 200 g/h. As it is natural for the detectable crack length to be evaluated under the normal operating conditions with a certain period, the corresponding loadings are considered to be as follows: Internal pressure, Thermal expansion, and Dead weight. Based on the above evaluation procedure, this study deals with certain aspects of the present leakage evaluation method, as follows:

442

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

Fig. 3. Denition and evaluation method of detectable crack at leak.

2.3.1. Coolant leakage velocity Coolant leakage from a through-wall crack, Q, can be calculated as follows:

Q qV S

where q and S denote the coolant density and crack opening area, respectively. The coolant leakage velocity, V, can be evaluated from the following equation [2]:

v u 2p u o V t n fF 1rkC 1r kB t q 1:5 D
h

where p is the internal pressure, fF is the friction coefcient, and Dh is the equivalent diameter of the crack opening area. The parameter, r, is known as the stress ratio, which is the ratio of membrane stress to the combined membrane and throughwall bending stresses. However, the effect of through-wall bending stress on the crack opening is certainly different from that of the membrane stress [4,5], and thus the following parameter, r0 , which is called the crack opening ratio, is proposed as an alternative in this study:

r0

dm dm db

where dm is the crack opening displacement (COD) owing to membrane stress, and db is the COD from the through-wall bending stress. The correction factors kB and kC of Eq. (3) represent the degree of resistance to a leakage ow of a crack surface subjected to a particular stress history [2]. They are called surface resistance correction factors. 2.3.2. Resistance correction on crack surface A leakage threshold value of 200 g/h [1] is actually a very low quantity which may merely represent seepage. Under lowpressure conditions, the surface resistance correction factors in Eq. (3) are considered to be signicant for an evaluation of a low level leakage, and are therefore examined in this study.

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

443

The effect of the surface resistance correction factor on the coolant leakage velocity was investigated simply in Fig. 4, using an example where only kC was considered. Fig. 4 indicates that kC should be treated as a function of the COD, taking the stress histories into account. In what follows, kC and kB are studied in cases where a structure is subjected to fatigue loadings with a stress ratio (Min. stress/Max. stress) of 1.0. (1) Stress history and crack opening behavior in which only kC is considered. The factor kC is affected by the surface roughness owing to crack growth controlled by membrane stress [2]. It is thought that most normal operating conditions of nuclear plants are represented by this case. The fact that kC may be considered as the sole representative correction factor for surface irregularity means that the characteristic of crack opening from throughwall bending stress is not apparent in the overall stress histories. (2) Stress history and crack opening behavior in which only kB is considered. The factor kB is a shape correction factor owing to crack surface generation and related variables, and is controlled by through-wall bending stress [2]. In general, such a through-wall bending stress-controlled case is expected to have an r0 of 0. (3) Stress history and crack opening behavior in which both kC and kB are considered. This mainly deals with the case in which a crack opening is due more to through-wall bending stress than to membrane stress, but in which the characteristics of the crack surface due to both stresses are signicant. It is clear that the resistance correction term is inuenced by the variable r0 in this case, regardless of the COD of the outer crack surface at the point when a leakage is detected. From the consideration above, it is judged that the surface resistance correction factors are signicant parameters for a low leakage evaluation, and thus systematic research on them should also be performed experimentally. 2.4. Structural failure mode assessment In LBB assessment, it is clear that one of the most important concerns is to evaluate the crack length at the structural critical point (2Ccr) in a pipe. This study deals with the ductile unstable fracture of a circumferentially cracked straight pipe and the buckling of an axially cracked elbow as critical points to be considered as part of the loading and structural characteristics of an FBR. 2.4.1. Ductile unstable fracture Generally, the failure mode in an LBB assessment is DEGB (double-ended guillotine breakage) owing to a plastic collapse or ductile unstable fracture. In the case of comparatively small-diametric piping structures, consideration of the net section stress owing to a plastic collapse concept has often been adopted as an effective fracture assessment method. However, the plastic collapse concept is considered to produce an unreasonable crack length for thin-walled and large-diametric structures such as FBR piping systems, in which a ductile unstable fracture may precede a plastic collapse. Therefore, it is thought that the two-parameter method of R6 [5] or the tearing modulus concept [6], which considers the fracture behavior of a
3.0E+06

2.5E+06

kc = f( at leakage detection) Previous Study [Guideline, 2001] kc = f( due to stress history) - Present Study -

2.0E+06

V (mm/hr)

1.5E+06

kc = 0

1 5

1.0E+06

10
5.0E+05

0.0E+00 0

0.01

0.02

0.03

0.04

out at leakage detection (mm)

Fig. 4. Effect of surface resistance correction factor on coolant leakage velocity (p = 0.182 MPa, Q = 200 g/h, D0 = 1270 mm, t = 15.9 mm).

444

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

crack tip, is suitable for evaluating a critical crack in the FBR structure. This approach is especially important for evaluating the fracture behavior of ferrite steel, with its low ductility. For these reasons, the two-parameter method of R6 was adopted as a representative fracture assessment method for FBR piping structures in this study. Fig. 5 concisely shows the assessment procedure by the R6 method of the relation between Kr, which is a parameter relating to fracture toughness, and Lr, which is related to the hardening characteristics of the material. The following loading conditions contributing to an unstable fracture are considered: DF1 DF2 DF1 DF1 times times times times internal pressure. thermal expansion. dead weight and seismic load.

DF1 and DF2 are the design factors for the primary and secondary stresses in an unstable fracture. 2.4.2. Buckling assessment Since FBR structures have thin-walled and large-diametric piping systems, another critical failure mode may be buckling. Although buckling does not induce a double-ended guillotine break, it is certain that the buckling mode impairs the durability of the piping structures. A representative object for the buckling assessment is elbow structures. When an elbow has a circumferential crack on its intrados or extrados surface, buckling is generally expected to precede an unstable fracture. In this case, the buckling mode is considered to complement the structural critical state assessment from the viewpoint of safety design. Meanwhile, although an axially cracked structure is considered not to undergo an unstable fracture, it must be an object of the buckling assessment. Considering that a great stress acts on its crown surface of an elbow under normal operating conditions, an elbow with an axial crack on its crown surface can be qualied as a main object for a buckling assessment. Generally, a buckling assessment can be performed using a limit load concept. The conditions that bring about buckling are as follows [7]:

Mcr M0 hc
For an axial crack,

M cr > M 1 hc 1 0:15C =D
For a circumferential crack,

M cr > M 1 M 2 hc 1 0:2137C =D 0:0485C =D2 1:0559C =D3

where M0 denotes the plastic collapse moment and M1 and M2 are the global bending moment due respectively to the primary and secondary stresses. The parameters C and D denote a half crack length and pipe mean diameter. Although the critical crack length can be obtained from Eqs. (5)(7), it is considered that the signicance of this assessment is to compare the critical moment at a detectable crack with the moment components contributing buckling. The loading conditions for the buckling assessment are as follows: DF2 times thermal expansion, DF1 times dead weight and DF1 times seismic load DF1 and DF2 are the design factors for the primary and the secondary stresses in the buckling assessment.
Kr = KI /KIC FAC P Y P : Crack Growth Initiation XY : Stable Crack Growth YZ : Supposed Crack Growth CO : Path to Failure X Unstable Fracture Occurrence C

O Z

Lr = / y
Fig. 5. Ductile unstable fracture assessment method by two-parameter method.

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

445

3. LBB assessment using proposed methodology for FBR piping In this work, an LBB assessment based on the proposed method was performed for a ferrite steel piping structure (O.D. = 1270 mm, thickness = 15.9 mm) with the two-looped primary hot-leg system using the FBR design data [1]. From the results of stress analysis using a nite element method (FEM) with design operating conditions, the assessment objects were identied as the Y-piece and the elbow crown of the piping system. Based on the present methodology, the respective crack lengths were obtained for the LBB assessment. 3.1. Geometry, loading conditions and material Fig. 6 shows a schematic drawing of the LBB assessment object and the normal operating condition, with a holding period of about 700 h under 550 centigrade [1]. As a thermal transient state, a manual trip was selected in this study. Fig. 7 shows the variation of coolant temperature and heat transfer coefcient (h) with time at the thermal transient. The ow rate was assumed to be 9.08 103 kg/s at the steady state. The method for calculating the heat transfer coefcient is shown in the following equation [8]:

h Nu j=Di Nu 5 0:025Pe0:8 Pe Pr Re Re V Di =l
where Nu is the Nusselt number, Pe is the Peclet number, Di is the inner pipe diameter, and V is the velocity of coolant. SebanShimazakis equation was used to calculate Nu in this study. The material is Mod. 9Cr1Mo ferrite steel, which is recognized to be eligible for use in large-scale FBR piping structures, and the coolant is liquid sodium. The material properties of Mod. 9Cr1Mo ferrite steel and sodium under normal operating conditions are shown in Table 1 [1]. The complete material data for creep-fatigue crack growth and a fracture assessment were not available, and thus Cf, mf, Cc, mc shown in Eq. (1), and JIC (fracture toughness) and JR curve (material resistance to crack growth) of Mod. 9Cr1Mo ferrite steel were used in this study [1]. 3.2. Stress analyses by FEM Finite element analyses (FEA) were performed elastically at a steady state under a normal operating condition and at the thermal transient. An FEA model for the steady state was constructed by a shell element, and the other model for thermal
Elbow : p = 0.105 MPa
IHX (550 oC )

I H

G Temperature
RV (550 oC )

- E G : 550 oC -G -H H : 300 oC I : 50 oC

F Y-piece : p = 0.182 MPa

Fig. 6. Schematic drawing of LBB assessment object and normal operating condition of the primary hot-leg piping structure of large-scale FBR [1].

446

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

600

120

Temp. at MT Q at MT
500 100

Temperature (C)

400

80

300

60

200

40

100

20

0 0 200 400 600 800

0 1000

Time (sec.)

(a) Variation of Coolant Temperature and Flow

25000

MT
20000

h (W/m2 K)

15000

10000

5000

200

400

600

800

1000

Time (sec.)

(b) Variation of Heat Transfer Coefficient

Fig. 7. Loading condition at thermal transient.

transient analysis was composed of an axi-symmetric element with an assumption that stress at the thermal transient was distributed axi-symmetrically. On the basis of the steady state analysis, a Y-piece with a circumferential crack and an elbow with a circumferential crack on the intrados surface and an axial crack on the crown surface were selected as the LBB assessment objects for this study. The stress components for evaluating each item are laid out in Table 2, where P denotes the primary stress and Q denotes the secondary stress. In Table 2, other stress components except for the stress analyses by the present FEA were taken from Ref. [1]. 3.3. LBB assessment results Table 3 shows the evaluation results of the cracks for the LBB assessment by the present methodology. Crack penetration was assumed to have occurred when the normalized crack depth (a/t) reached 0.8, considering the yielding of the outer surface during crack growth. In evaluating the detectable crack size, the surface resistance correction factor kC was assumed to be 10, referring to the result of Fig. 4 for a conservative evaluation, and the COD solutions considering that both membrane and through-wall bending stresses were used [9,10]. Alternatively, in assessing the unstable fracture of a circumferentially cracked structure, DF1 and DF2 adopt the values of 1.5 and 1.1, respectively, considering elastic follow-up in the elasticplastic region applied to the design for FBR structures. As the critical crack obtained from unstable fracture assessment was greater

Q (%)

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448 Table 1 Material property of Mod.9Cr1Mo ferrite steel and sodium at 550 C. E (GPa)

447

rY (MPa)
333

rU (MPa)
397

j (W/m K)
29.5

C (J/kg K) 744

a 106
12.0

q (kg/m3)
7860.0

Mod. 9Cr1Mo ferrite steel 179 0.302

j (W/m K)
Sodium 64.6 E: Youngs modulus m: Poissons ratio rY: Yield stress rU: Tensile strength j: Thermal conductivity C: Specic heat a: Thermal expansion rate q: Density l: Kinematic viscosity Pr: Prandtls number

l 107 (m2/s)
2.74

Pr 0.0044

q (kg/m3)
820.25

Table 2 Stress components used in LBB assessment. Y-piece (circumferential crack) Pm (MPa) Qbg (MPa) Qb (MPa) Pbg-W (MPa) Pbg-S (MPa) 3.6 210.1 29.7 0.8 11.0 Elbow (circumferential crack) 3.1 43.0 29.7 2.0 13.0 Elbow (axial crack) 6.2 0 282.6 0 13.0

Pm: Stress due to internal pressure Qbg: Stress due to thermal expansion Qb: Stress due to thermal transient Pbg-W: Stress due to dead weight Pbg-S: Stress due to seismic load

Table 3 Crack lengths and LBB assessment results. Y-Piece (circum. crack) Crack lengths Cp (mm) Cd (mm) Ccr (mm) LBB assessment Ccr/Cd Mcr/M1 or Mcr/(M1 + M2)
a

Elbow (circum. crack) 20.1 62.2 1217.8 19.6 12.2

Elbow (axial crack) 49.9 Crack closure 29.1a

17.4 21.4 800.5 37.4

Mcr in an axial cracked elbow was evaluated by Cd (68.1 mm) neglecting crack closure.

than the detectable crack from a leakage evaluation, it can be said that the LBB behavior was shown in these cases. The ratio of Ccr to Cd is generally called the safety margin for LBB behavior. The same judgment can be made on an elbow with a circumferential crack from the evaluation results in Table 3. Meanwhile, considering that an elbow with a thin wall and a large diameter is likely to have a buckling mode, the buckling assessment with regard to Cd was performed on the present elbow. This showed a smaller safety margin for the buckling mode than for an unstable fracture, and thus it is thought that the buckling mode may become a complementary LBB assessment category in this case. In the case of an elbow with an axial crack, although Cd can be obtained from the present leakage evaluation methodology, the evaluation results were omitted from Table 3. The main reason for this was a prominent through-wall bending stress that induced a crack closure on the inner surface. Moreover, when an axially cracked elbow structure is subjected to a displacement-controlled load, the stress near the crown surface is very low, and in general, the crack does not grow. After all, it is considered that in these circumstances an axially cracked elbow is unlikely to undergo an unstable fracture. Therefore, it can be judged that LBB behavior does not arise in this case. However, considering that most ferrite steel is a work-softening

448

Y.-S. Yoo, N.-S. Huh / Engineering Failure Analysis 33 (2013) 439448

material, if the cyclic stress of an elbow is close to the yield stress, a crack closure can be lost owing to the residual deformation during crack growth. In this limited case, it is considered that the leakage can be detected, and therefore the buckling mode as a critical aspect is valid as an LBB assessment category. 4. Discussion Though an LBB assessment is generally performed on objects that are highly-stressed under normal operating conditions, another object where potential crack growth may be anticipated might be a welded part that contains residual stress. It is difcult to claim that the residual stress is considered in evaluating Cp since it is not a uctuating stress component if the stress relaxation effect under normal operating conditions can be neglected. In evaluating Cd of a welded part, however, if a crack penetrates the pipe wall, the effect of the residual stress may be considered for evaluating the corresponding COD. Meanwhile, in assessing a ductile unstable fracture, if the residual stress distributes the thickness as a self-equilibrated stress component in a circumferential cracked structure, its effect is considered to be negligible. Further consideration is given to the design factors for the primary and secondary stresses. As the structural design for FBR permits an elasticplastic region, precise studies on the design factors should be performed considering the structural characteristics and stress classications. The design factors also need to be obtained from a statistical analysis of the scatterings of the experimental data. 5. Conclusions This paper dealt with an advanced LBB assessment methodology for ferrite steel piping structures with thin walls and large diameters. The applicability of the proposed method has been veried by an LBB assessment of ferrite steel pipes in a commercial-scale FBR design, and thus the following conclusions were obtained: (1) The crack length detectable by a leakage is sensitive to the surface resistance to the leak ow under low-pressure conditions. (2) The unstable fracture mode in FBR piping structures can be properly assessed by the R6 method, taking the material resistance to crack growth into account. (3) Buckling occurs prior to an unstable fracture in the case of thin-walled and large-diametric elbow structures. The present study has demonstrated the appropriate methodology for an LBB assessment of commercial-scale FBR piping structures. The present methodology investigated in this paper is expected to drive advances in structural design technologies, leading to a reliable and reasonable plant design for FBRs. Acknowledgments The present author would like to express his sincere thanks to Prof. N. Kasahara of Tokyo University and Dr. T. Asayama of Japan Atomic Energy Agency for providing signicant advice on FBR design characteristics and an LBB assessment for FBR structures. References
[1] Ichimiya M, Mizuno T, Konomura M, A promising sodium-cooled fast reactor concept and its R&D plan, Global 2003, New Orleans, 435, 2003. [2] Guideline for structural integrity evaluation for high-temperature components. Central research institute of electric power industry; 2001 (in Japanese). [3] Yoo YS, Ando K. Leak-before-break behavior of statically indeterminate piping system. Nucl Eng Des 2003;224:25363. [4] Wakai T, Aoto K. A study on the guideline of defect assessment procedures for large scale fast breeder reactor components (1) Development of creepfatigue crack growth assessment procedure. JNC TN9400 2001078; 2001 (in Japanese). [5] R/H/R6. Assessment of the integrity of structures containing defects. Nuclear Electrics Limited Rev. 3; 1995. [6] Paris PC, Tada H, et al., The theory of instability of the tearing mode of elastic-plastic crack growth. vol. 5. ASTM STP, No. 668; 1975. [7] Zahoor A. Ductile fracture handbook, EPRI NP-6301; 1989. [8] Seban RA, Shimazaki TT. Heat transfer to a uid owing turbulently in a smooth pipe with walls at constant temperature. Trans Am Soc Mech Eng 1951;73:8039. [9] Machida H, Yoo YS. Crack opening displacement of a through-wall crack in a plate subjected to bending load. Nucl Eng Des 2003;226:193204. [10] Yoo YS. Crack opening displacement of circumferential through-wall cracked cylinders subjected to tensile and through-wall bending loads. Nucl Eng Des 2004;229:12338.