The Degradation Evaluation of Reactor Pressure Vessel Materials due to Neutron Irradiation by Ultrasonic Characteristic Analysis

SamLai Lee, KeeOk Chang, ByoungChul Kim, KwunJae Choi, UnSik Gong
Korea Atomic Energy Research Institute, 150Dukjindong, Yusonggu, Taejon, Korea.
Contact

ABSTRACT

Ultrasonic signals from Charpy impact test specimen have been analyzed. Base and weld metal specimens extracted from a reactor vessel according to the schedule of the surveillance test required by the related regulations during plant routine overhaul have been used. Ultrasonic test parameters including velocities were analyzed for different frequencies and were correlated with the quantity of material degradation such as the amount of neutron fluence. Test results showed that the ultrasonic parameters had close correlations with the amount of neutron irradiation for specimen and certain possibilities where a nondestructive evaluation method could be used to predict the fluence of the neutron irradiation in nuclear reactor vessel materials.

Keywords: ultrasonic analysis, Charpy impact test, neutron irradiation, radiation embrittlement, reactor pressure vessel

INTRODUCTION

The core beltline region in a reactor vessel is subject to neutron irradiation produced from nuclear fuel and the neutrons collide with the material lattice structure, causing changes in strength, hardness, ductility and toughness. Although irradiation hardening itself may show a positive effect from a metallurgical point, the integrity of vessel can be influenced by the accompanying decrease of ductility due to irradiation(Griesbach & Server 1993). The low alloy steel used for vessel material shows a ductile-brittle transition temperature shift in Charpy impact testing and the phenomenon occurs due to various defects residing in the material. Embrittlement by neutron irradiation is affected by various factors such as chemical composition, microstructure, etc. Various chemical compositions are added in order to improve the strength and stiffness while some remains after manufacturing. One of the elements, Cu, is known to deepen the irradiation embrittlement phenomenon (Astafev 1977). Many researchers have tried to measure the degree of embrittlement by various methods.

The life prediction of an operating reactor vessel has been performed through mechanical testing of the specimens such as Charpy and compact tension specimen and the safety of the operating reactor vessel has been convinced by correlating test parameters with the amount of neutron fluence. Although a sufficient number of specimens have to be available in order to maintain the integrity of the reactor vessel and make appropriate decision for life management, the numbers inserted into the reactor vessel during construction are limited due to the space allocation, thus causing difficulty in accurate evaluation due to neutron irradiation. Thus, USNRC recommended the development of various nondestructive evaluation methods to compensate the current integrity evaluation practice(Jackson 1996). One of the methods is ultrasonic testing widely used in detecting minor cracks within the material and measuring mechanical properties. Some research has been performed using ultrasonic characteristics such as Rayleigh wave criticality, electromagnetic acoustic effects(Alers & Ray 1999), magnetostrictive effects(Igarashi & Alers 1999) and ultrasonic nonlinearity(Hurley, Balzar & Purtscher 1999). Research has been also performed using magnetic methods such as Barkhausen noise measurement and magneto-hysteresis curve. However, up to now, much work using nondestructive method has not been appreciably noticed since such a research activity involves radiation contamination as well as specimen availability problem (Hiser & Green 1999).

THEORETICAL BACKGROUND

Since the ultrasonic beam is a stress wave, evaluation of the material properties is done using the fact that velocity and energy may vary when ultrasonic energy passes through the material and also depend on the density, grain size, dislocation and precipitates which may be affected by neutron irradiation and thus it becomes possible to correlate the interrelationship among attenuation, velocity and neutron fluence.

Experimental example showed that more impurities within the material by any reason hampered the propagation of velocity, thus causing slowness of velocity. Our previous result also showed that a certain correlation existed between phase velocity and the amount of neutron fluence in samples irradiated by neutrons in a nuclear reactor vessel. Generally, phase and group velocities are considered to be the same in ordinary polycrystalline material that is not scattered so much ultrasonically and these two velocities are interrelated mathematically. While our previous experiment was mainly focused on the measurement of phase velocity in the samples in order to see how the impurities caused by neutron fluence affect the phase information, the group velocity so called energy velocity is calculated at this time in order to see any changes compared to the previous measurement (Lee et al. 2000).

EXPERIMENT

Standard Charpy impact test specimens made of Mn-Mo-Ni low alloy steel and extracted at 1/4T from the surface of the reactor vessel beltline region during the reactor vessel fabrication were used. The specimen consists of base and weld metal and their composition and heat treatment condition are given in tables 1 and 2. Figure 1 shows the optical metallographs for base and weld metal. The irradiated specimens inserted into the reactor vessel before the operation of the reactor are withdrawn after a certain period of irradiation by neutrons according to the plant surveillance program. Table 3 shows the irradiation period and the amount of neutron fluence for the specimens.

Element	C	Si	Mn	P	S	Mo	Ni	Cr	Cu	Al	Co	Fe
Base	0.16	0.17	1.28	0.006	0.005	0.48	0.71	0.17	0.06	0.011	0.012	Bal.
Weld	0.06	0.37	1.68	0.007	0.008	0.55	0.58	0.03	0.04	0.004	0.011	Bal.
Table 1: Chemical composion (wt%) of specimens.

Mat'l	Temperature	Time	Cooling method
Base metal	Austenizing (871 ^°C)	4 hr	Water-quenched
	Tempered (663 ^° C )	4 hr	Air-cooled
	Stress relief Heating : 595 ^° C Holding : 595 ^° C Cooling : 310 ^° C	15 ^° C/hr 6 hr 10 ^° C	Air-cooled
Weld metal	Post weld, Stress relief Heating : 595 ^° C Holding : 595 ^° C Cooling : 310 ^° C	15 ^° C/hr 6 hr 10 ^° C	Air-cooled
Table 2: Heat treatment of specimens

(a)	(b)
Fig 1: Optical micrographs of base(a) and weld metal(b).

Ultrasonic signal by pulse echo method was obtained in order to identify the change of ultrasonic characterization due to neutron irradiation. 5 MHz ultrasonic transducer of 6 mm diameter with a delay line in order to reduce the effect of the near zone due to the size of the specimen was used. Although the signal amplitude changes slightly depending on the contact pressure between transducer and each specimen surface, the amplitude information does not greatly affect the test result since we are looking at the signal itself in frequency domain. The measurement for the irradiated specimen was performed using a lead shielded facility as shown in Fig. 2, since the specimen is highly contaminated by radioactive materials.

Withdrawal schedule of specimens	Fuel cycle	Neutron fluence (10¹⁹ n/cm² )
0^th	0^th	0
2^nd(5.28EFPY)	6^th	2.2
3^rd(7.68EFPY)	8^th	3.8
Table 3: History of neutron irradiation.

Fig 2: UT data acquisition system.

RESULT AND DISCUSSION

Figure 3 shows a typical ultrasonic signal acquired from the specimen by the transducer. Signals from delay (1) and (2) represent the return echoes of the delay line used to reduce the near zone effect and steel back signals (1) and (2) are from the back surface of the specimen. Each signal in the time domain acquired from the specimen was transformed into the frequency domain by FFT and the phase velocity at each frequency was calculated from the phase information and the frequencies, and the changes were then observed for each specimen as well as the amount of neutron fluence (Jeong & Hsu 1995). Group velocity was also calculated by differentiating phase velocity numerically and adjusting using frequency information as well as phase velocity information. The velocity calculation strictly from phase information may face some errors including system time delay and diffraction effect and the result may show problems such as lack of accuracy as well as reliability microscopically compared to the analysis done by sophisticated ultrasonic apparatus (Fortunko et al. 1992). However, the result obtained from conventional ultrasonic equipment are considered to have shown enough information to sufficiently demonstrate macroscopic range since the errors from such problems does not greatly affect slope information between phase and frequency domain.

Fig 3: Typical UT signal from specimen.

The evaluation was performed for the irradiated as well as the unirradiated specimen and the relationships between the velocities versus frequencies are shown in Fig. 4 and 5. Fig. 4 shows phase velocity information while Fig. 5 presents group velocity value versus frequency. One may notice that both velocities decrease while the amount of neutron fluence increases in base metal as shown in Fig. 4(a) and 5(a). Such results may be considered to be consistent with previous work where ultrasonic wave propagation could be slow down by various defects present caused by various reasons(Cohen & Mal 1989). Systematic changes were not noticed in the weld metal shown in Fig. 4(b) and 5(b) compared to the case of the base metal although these two graphs showed similarity with each other which was predicted from the theoretical background. Such inconsistent behavior is considered to stem from the complicated metallurgy and compositional characteristics of weld metal although the area where the ultrasonic signals were acquired represented real weld. One may also notice that the velocity difference was not appreciably noticeable at the specimen extracted at 2^nd and 3^rd withdrawal schedule. This phenomenon may be considered due to the reason that once the reactor starts operation, vast amount of defects and clusters are generated at the beginning in the material and their amount of generation does not increase any more although the amount of neutron fluence increases. This phenomenon is very similar to the results of mechanical testing, called Charpy impact testing where ductile and brittle transition temperatures are measured ( Final Report 1996). One can see the big drop of the CVN energy in Fig. 6, when comparing the specimen extracted at 2^nd withdrawal schedule with the unirradiated specimen at approximately ambient temperature and also slight difference in the CVN energy at the specimen extracted at 2^nd and 3^rd withdrawal schedule. Consistent behavior is also observed in the base metal shown in Fig. 6(a), while one can see some inconsistency in the weld metal shown in Fig. 6(b).

Fig 4: Relationship between phase velocity and frequency for base(a) and weld metal(b)

Fig 5: Relationship between group velocity and frequency for base(a) and weld metal(b)

Fig 6: Relationship between CVN energy and temperature for base (a) and weld metal (b)

CONCLUSIONS

Ultrasonic evaluation for Charpy impact test specimen has been performed in order to evaluate the degree of the embrittlement of the reactor vessel and following results have been drawn;

The ultrasonic phase velocity decreases as the amount of neutron fluence increases in base metal, while such a consistent relationship was not shown in weld metal. A similar relationship has been observed in group velocity analysis, which was predicted from our theoretical background. Velocity saturation phenomena observed at the specimen extracted at 2^nd and 3^rd withdrawal schedule may be interpreted to stem from the generation of vast amount of defects and clusters at the beginning of reactor operation and slowing down later on although the amount of neutron fluence continuously increases with reactor operation. Such a similarity is also observed in the results of mechanical testing, called Charpy impact testing where ductile and brittle transition temperatures are measured. More parametric analysis by ultrasonic testing may be useful in identifying the material state related to neutron irradiation embrittlement and can be considered useful for supplementing the current technology for the evaluation and prediction of the degree of reactor vessel embrittlement.

REFERENCES

Alers, G. A., and Ray Santoyo 1999, 'Shear Wave Velocity Response to Copper Precipitate and Neutron Irradiation' in Workshop on Nondestructive Characterization of Embrittlement in Reactor Pressure Vessel Steels Chapter 19, National Institute of Standards and Technology, United States Department of Commerce.
Astafev, A. A., et. al. 1977, Atomnaya Energiya 42, pp.187-193.
Cohen, Y. B., and Mal, A. K. 1989, 'Ultrasonic Inspection' in Metals Handbook 17 , Nondestructive Evaluation and Quality Control, ASM International, Metals Park, pp.231-277.
Final Report for the 3^rd Surveillance Test on the Reactor Pressure Vessel Material 1996.
C. M. Fortunko, G. L. Peterson and B. B. Chick, M. C. Renken and A. L. Preis 1992, 'Absolute Measurements of Elastic-Wave Phase and Group Velocities in Lossy materials', Rev. Sci. Instrum., Vol.63, No.6, pp3477-3486.
Griesbach, T. J., and Server, W. L. 1993, 'Management of Neutron Radiation Embrittlement: A View from the United States' in Radiation Embrittlement of Nuclear Reactor Pressure Vessel Steels: An International Review (Fourth Volume)-1990, edited by L.E. Steele, ASTM STP 1170, Philadelphia, pp.30-39.
Hiser Jr, A. L., and Green Jr, R. E. 1999, 'Evaluation of Ultrasonic Attenuation Measurements for Estimating Irradiation Embrittlment of RPV Steels' in ibid, Chapter 18.
Hurley, D. C., Balzar, D., and Purtscher, P. T. 1999, 'Ultrasonic Nonlinearity Parameter in Precipitate-hardened Steels' in ibid, Chapter 16
Igarashi, B., and Alers, G. A. 1999, 'Ultrasonic Measurement of Dynamic Magneto-striction' in ibid, Chapter 23.
Jackson, S. 1996, 'The Direction and Trend in US Nuclear Regulatory Policy' in The First Annual KAIF/KNS Meeting, Seoul.
Jeong, H., and Hsu, D. K. 1995, 'Experimental Analysis of Porosity-induced Ultrasonic Attenuation and Velocity Change in Carbon Composites', Ultrasonics 33, pp195-203.
Lee, et al. 2000, 'Ultrasonic Evaluation of Radiation Embrittlement in Reactor Vessel Materials', Proc. of 2000 Rev. of Progress in QNDE, Ames, Iowa, July16-21.