Materials Science and Engineering A319321 (2001) 316320

A study on the mechanical strength change of 2.25Cr1Mo steelby thermal aging

Hyuntae Yang *, Sangtae KimDepartment of Mechanical Engineering, Yeungnam Uniersity, Oyongsan, 712-749, South Korea

Abstract

The purpose of this study is to investigate the thermal embrittlement and the mechanical properties of 2.25Cr1Mo steel agedat high temperature for extended periods. Original and aged materials were tested to obtam the tensile strength, hardness andimpact-absorbed energy. The tensile strength, hardness and impact-absorbed energy decreased as aging time was increased. X-raydirtraction was used to study changes in carbide structure. These changes lead to thermal embrittlement. 2001 Elsevier ScienceB.V. All rights reserved.

Keywords: Carbide extract; Energy transition temperature; Fatigue test; High temperature tensile test; Thermal embrittlement

www.elsevier.com/locate/msea

1. Introduction

Structural steel components should have good tough-ness even after being subjected to elevated temperaturesfor extended periods, a process that can cause temperembrifflement. However, material degradation such asthermal embrittlement occurs during prolonged serviceat the working temperature range of steel power plantsteam pipes. Previous research on thermal embrittle-ment has addressed the following subjects:1. the embrittlement mechanism and fracture morphol-

ogy [1,2],2. heat treatment for preventing brittle fractures,3. the effect of microelements on degradation [35],4. the correlation between impact energy and fracture

toughness [69].The embrifflement mechanism has not been clearly

examined until recently. Now the embrittlement due tograin-boundary carbide precipitation and the equiva-lent segregation can be exammed with the advent ofbetter analytical equipment.

This study was conducted to examine changes in thestrength of 2.25Cr1Mo steel caused by high-tempera-ture aging and to determine the characteristics of the

change with different aging time and temperature. Thisalloy is used extensively in the steam pipes and pressurevessels of power plants.

2. Experimental procedure

The steel used in the experiments was 2.25Cr1Mosteel. The mechanical properties and chemical composi-tion of 2.25Cr1Mo steel are listed elsewhere [1,2]. Thetest specimens for investigating the degree of thermalembrittlement were made from artificially aged andin-service-aged materials. The artificially aged speci-mens were annealed for 500, 1000, and 5000 h at530 C. The in-service-aged materials came from theelbow tube of a steel pipe, which had been used forabout 10 000 h at 530 C. These specimens were cut asshown in Fig. 1.

All specimens were tested to obtain the impact-ab-sorbed energy, high-temperature tensile strength, andhardness. The effect of fatigue crack growth with in-creasing aging time was also investigated. After eachtest, the specimen was examined to compare the mor-phology of the carbides. For this, the carbide particleswere extracted by electrolytic dissolution.

Concerning the tensile test at the elevated tempera-ture, tests were periormed up to 600 C at intervals of100 C and the tensile strength was measured at each

* Corresponding author. Tel.: +82-53-8102456; fax: +82-53-8133703.

E-mail address: stkim@yu.ac.kr (H. Yang).

0921-5093/01/$ - see front matter 2001 Elsevier Science B.V. All rights reserved.PII: S0 9 21 -5093 (01 )01008 -5

H. Yang, S. Kim /Materials Science and Engineering A319321 (2001) 316320 317

temperature. We also performed Rockwell hardnesstests, micro-Vickers hardness tests, and impact tests.Chevron V-notched (CVN) specimens were used for theimpact tests. The impact test data were assumed to berepresented by the following equation [10]

E=A+B tanhTT0

C

(1)

where A, B, and C are parameters, T is the temperatureof the impact test, E is the CVN impact-absorbedenergy, and T0 is the energy transition temperature.

Fig. 4. Temperature dependence of the CVN impact-absorbed energyof 2.25Cr1Mo steel.

Fig. 1. Schematic for the location of the tensile and impact testspecimens.

Fatigue crack growth rates in the Paris region weremeasured. The specimens with different aging condi-tions were tested under load control, using a sine waveload form, with a load ratio of 0.1 and a frequency of10 Hz. The crack length was monitored continuouslyusing a traveling microscope.

Carbide morphology following thermal embrittle-ment was studied on the extracted carbides. The elec-trolyte used was a mixture of 95% methanol and 5%muriatic acid. X-ray diffraction was used to identify themorphology of the carbides.

3. Experimental results

The tensile strengths of 2.25Cr1Mo steels in theoriginal and the in-service-aged conditions are shown inFig. 2. The tensile strength of the aged steel is lowerthan that of the original material. At 530 C, thisdifference is about 150 C. Generally the tensilestrength at high temperatures relates directly to thealloy composition. Typically, temperatures between 400and 500 C affect the tensile strength of steels contain-ing C, Cr and Mo. In case of C, this effect starts at200 C. The strengthening effect decreases with thechange of carbide morphology caused by a long expo-sure at high temperature.

In the case of the original material, secondary hard-ening is clearly seen near 400 C. However, in the caseof the used material, there is a weaker effect. Thisreduction of secondary hardening is due to the forma-tion of Cr-rich carbides during the long time in use.

The Rockwell and Vickers hardness tests results areshown in Fig. 3. A clear difference in hardness is seenbetween the original and used material. As the heatingtime increased, both hardnesses decreased almostlinearly.

Fig. 2. Temperature dependence of the tensile strength of virgin andused tube material.

Fig. 3. Rockwell and micro-Vickers hardness as a function of agingtime.

H. Yang, S. Kim /Materials Science and Engineering A319321 (2001) 316320318

Fig. 4 shows CVN impact test results and approxi-mate curves for materials artificially aged at the differ-ent aging times. Fig. 5 shows the approximate absorbedenergy transition temperature obtained from the data inFig. 4. The energy transition temperature can be repre-sented by the following equation, which was obtainedby curve fitting the data in Fig. 5.

T0=37+0.1027 t0.574 (2)

Here T0 is the energy transition temperature (C) andt is the thermal aging time (h). This equation can beused to estimate the degradation of structural steelsafter a long exposure to high temperatures.

Fig. 6 shows the result of fatigue crack growth rateversus the stress intensity factor during fatigue cyclictests. There is no clear difference between the growth

Fig. 7. X-ray diffraction patterns (CuK radiation) of the carbidesextracted from 2.25Cr1Mo steel in the original condition and fromsteel aged for various times.

Fig. 5. Energy transition temperature as a function of thermal agingtime.

rates of the original and aged specimens, but a some-what different growth rate is observed in the usedmaterial. In spite of the fact that the morphology anddistribution of the carbides changes with aging time,the behavior of the carbide does not show any changeduring the fatigue cycling. This means that the behaviorof the carbide is not relevant to fatigue cycling.

The carbide precipitation obtained by bulk elec-trolytic extraction was analyzed by X-ray diffraction, asshown in Fig. 7. M23C6, M2C and M3C were foundboth in the original and aged specimens. With increas-ing aging time, the amount of M23C6 increased. For thematerial used for a long time at high temperature, theX-ray analysis showed different types and quantities ofcarbides. In particular, the amount of M7C3 increasedwith increasing aging time. Generally the behavior ofcarbides for the 2.25Cr1Mo steel used at high temper-atures for a long period evolved with time as follows[11]:

e-Carbide M3C M7C3+ M3C +

M3C M2C M23C6M6C.

It is known that M6C stabilizes at temperaturesbetween 815 and 980 C, which is higher than thestabilizing temperature of M23C6, around 760 to850 C. Actually, the temperature of heat treatmentand the in-service temperature were about 530 C.Therefore, M23C6 and M6C carbides could not be iden-tified in specimens aged for a long period at a tempera-ture of 530 C. It thus seems that M23C6 carbides hadalready precipitated during the steel manufacturingprocess. The M7C3 carbides identified in Fig. 7 wouldoriginate from the transformation of M23C6, M2C, andM3C. Similar observations have been documented else-where [12].

Fig. 6. Crack elongation rate, da/dN, as a function of the stressintensity factor for virgin and aged 2.25Cr1Mo steel.

H. Yang, S. Kim /Materials Science and Engineering A319321 (2001) 316320 319

2.25Cr1Mo steel has ferrite and pearlite phases. Asshown in Fig. 8, the ferrite phase transformed topearlite by exposure at high temperature. It is due tothe fact that Fe3C in the ferrite was combined withother components during the heat treatment. Fig. 9shows the microstructures of different sections of thein-service-aged steel tube. Micrographs of the lateralside, inside, and outside section show long grains hav-ing shapes like rugby balls. However, the front sectionhas spherical grains. During operation at high tempera-tures for a long period, the elongated grains becomemore spherical and carbides grow at the grainboundaries. These became very sensitive to crack initia-tion and the steel becomes brittle.

4. Conclusions

The purpose of this study was to investigate thethermal embrittlement and the mechanical properties of2.25Cr1Mo steel after different aging conditions. Theresults are summarized as follows.

1. The difference between the tensile strength of theoriginal and in-service-aged steel is larger at the testtemperature of 530 C than at room temperature.The tensile strength of used material is about 150MPa less than the original material.

2. Energy transition temperatures and hardness can bedescribed as a function of aging time. However,there was no apparent dependence of fatigue crackgrowth rate on aging time.

3. Several carbides were observed in the original andthe aged steel and some carbides were transformedduring high-temperature aging.

4. The 2.25Cr1Mo steel contains both ferrite andpearlite phases, and the pearlite phase increasedwith increasing aging time. The microstructure ofthe used material shows crystalline elongated grainalong the longitudinal direction of the tube. Duringoperation at high temperatures for long period, theelongated grains become more spherical and car-bides grow at the grain boundaries. These becamevery sensitive to crack initiation and the steel be-comes brittle.

Fig. 8. Microstructure of 2.25Cr1Mo steel, (a) original (b) used material.

Fig. 9. Microstructures of used 2.25Cr1Mo steel tube, (a) axial cross section, (b) diagonal cross section, (c) outside surface, and (d) inside surface.

H. Yang, S. Kim /Materials Science and Engineering A319321 (2001) 316320320

References

[1] K.D. Challenger, A.K. Miller, Trans. ASME 103 (1981) 714.[2] Sung-Kang Hur, Kyung-Tae Hong, Jung-Man Doh, Bull. Ko-

rean Inst. Met. Mater. 7 (1994) 356371.[3] Yutaka Watanabe, Tetsuo Shoji, Metal. Trans. 22A (1991)

20972106.[4] Y. Nishizaka, Y. Hara, A. Hori, H. Ysukahara, K. Miyano, T.

Wada, T.B. Cox, Trans. ASME J. Pressure Vessel Technol. 107(1985) 285294.

[5] Lee Seung-Gun, Chang Yoon-Suk, Jin Tae-Eun, Tran. 15th Int.Conf. Struc. Mech. Reactor Tech., (1999) III-167-III-174.

[6] J.K. Kim, J.Y. Yoon, K.W. Song, J.J. Lee, S.H. Chung, Trans.KSME 15 (1991) 898906.

[7] T. Iwadate, Trans ASME J. Pressure Vessel Technol. 107 (1985)230.

[8] T. Iwadate, T. Karaushi, J. Watanabe, ASTM STP 631 (1977)493506.

[9] R. Roberts, G.V. Krishna, J. Nishanian, ASTM STP 700 (1980)552577.

[10] K.E. Stahlkopf, R.E. Smith, W.L. Server, R.A. Wullaert, ASTMSTP 601 (1975) 291307.

[11] A. Arrouz, M.J. Collins, R. Pilkington, Metals Technol. 10(1983) 461463.

[12] R.L. Klueh, P.J. Maziasz, Met. Trans. 20A (1989) 373382.

A study on the mechanical strength change of 2.25Cr1Mo steel by thermal agingIntroductionExperimental procedureExperimental resultsConclusionsReferences