CC BY-NC-ND
0Available online at www.sciencedirect.com
SciVerse ScienceDirect Procedia
Engineering
Procedia Engineering 55 (2013) 438 - 442 =
www. el sevi er. com/1 ocate/procedi a
6th International Conference on Creep, Fatigue and Creep-Fatigue Interaction [CF-6]
Role of Microstructure on Creep Rupture Behaviour of Similar and Dissimilar Joints of Modified 9Cr-1Mo Steel
P. Parameswaran*, K. Laha
Metallurgy and Materials Group, Indira Gandhi Centre for Atomic Research, Kalpakkam - 603 102, India
Abstract
Creep tests carried out on the joints of ferritic steels at 823 K over a stress range of 100 - 380 MPa indicate that the similar joints possessed lower creep rupture strength than its ferritic base metal. On the other hand, failure of the dissimilar joint occurred at relatively lower stresses at the ferritic / austenitic weld interface. Microhardness measurements on the two different joints reveal that a hardness peak at the ferritic / austenitic weld interface and a hardness trough at the intercritical heat affected zone (HAZ) in ferritic base metal, which suggest the role of distinct micro structural consequence. Un-tempered martensite was found at the ferritic / austenitic weld interface imparting high hardness, resulting in inhomogeneity in the strength at the interface. On the other hand, annealing of martensitic structure of modified 9Cr-1Mo steel by intercritical heating during welding thermal cycle resulted in hardness trough in the intercritical HAZ. On creep exposure, while secondary phase, Z-phase that formed in the HAZ regions of similar joints resulted in extensive cavity nucleation leading to premature failure of the joint, the inhomogeneity in strength that resulted from the formation of martensite in the dissimilar joint is found to play a major role in the failure.
© 2013 The AuAors.Published by Elsevier Ltd.
Selection and peer-review under responsibility of the Indira Gandhi Centre for Atomic Research. Keywords: Ferritic steel; heat affected zone; dissimilar joint; type IV cracking
1. Introduction
For sustained long-term service at elevated temperatures of operation in power plants, the creep behavior of the ferritic steel weld joints is one of the most important aspects. During welding, the thermal cycle generates inhomogeneous microstructure in the HAZ of the modified 9Cr-1Mo steel, resulting in marked variation in mechanical strength across the weld joint [1]. The Cr-Mo ferritic steel weld joint under creep condition fails prematurely at the outer edge of HAZ, commonly referred as type IV cracking, following the classification scheme developed by Shuller et al. [2]. Creep tests conducted on modified 9Cr-1Mo steel joints indicated that at relatively lower stresses and higher test temperatures, the weld joint possessed lower creep rupture strength than the base metal, and the difference in creep rupture strength increased with decrease in stress and increase in temperature. Preferential accumulation of creep deformation coupled with extensive creep cavitation in the
^Corresponding author:
E-mail address: param@igcar.gov.in
ELSEVIER
1877-7058 © 2013 The Authors. Published by Elsevier Ltd.
Selection and peer-review under responsibility of the Indira Gandhi Centre for Atomic Research. doi: 10. 1016/j .proeng .2013.03.277
intercritical region of HAZ led to the premature failure of the weld joint in the intercritical region of HAZ, commonly known as type IV cracking. This suggests that the microstructures across the HAZ of the weld joint play an important role in promoting type IV cracking. In addition, dissimilar weld joints between Cr-Mo ferritic steels and austenitic stainless steels are employed extensively in conventional as well as in nuclear power generating plants and petrochemical industries. In the steam generator circuit of sodium cooled fast breeder reactors (SFRs), 316L(N) austenitic stainless steel pipes from the intermediate heat exchangers are required to join with the modified 9Cr-1Mo ferritic steel (9Cr-1MoVNb) pipes of steam generators. In such a joint, a ferritic / austenitic transition bond is formed across which the chemical composition, microstructure, stress state, physical properties such as thermal expansion coefficient and thermal conductivity, and mechanical properties vary quite appreciably. The mismatch in thermal expansion coefficient across the joint is reduced with the insertion of an Alloy 800 piece and adopting Inconel 182 welding electrode having thermal expansion coefficients intermediate between austenitic and ferritic steels [3]. However, premature creep failure was observed in dissimilar weld joint [4-9]. Therefore, an understanding of the microstructural changes across the weld bond between ferritic and austenitic alloys and their effects on high temperature creep deformation and fracture behaviour are of primary concern for a realistic life prediction of the dissimilar weld joints. Present paper discusses the detailed microstructural studies were carried out to understand the micromechanisms responsible for the choice of location of failure during creep exposure.
2. Experimental
Optical, scanning electron microcopy (SEM) and transmission electron microscopy (TEM) studies and microhardness testing of the weld joints were carried out in the as-welded, post weld heat treated and creep tested conditions. A two-stage etching procedure was used to reveal the microstructures of the different zones of dissimilar weld joint. An electrolytic etching in 10% chromic acid with 6 V was carried out to reveal the microstructures of the weld interface, Inconel weld metal and Alloy 800 base metal; whereas an immersion etching in Villela's reagent was found to etch the ferritic part of the joints. Energy Dispersive X-ray spectrometer (EDXS) attached with SEM was employed to analyze the elemental distribution across the ferritic / austenitic weld interface of the dissimilar weld joints. Transmission electron microscopy (TEM) was carried out on samples extracted from the different location of the joints. Discs of 3 mm in diameter were machined from the joint, reduced to 100 ^m thickness by mechanical polishing and finally thinned by jet electropolishing using an electrolytic solution of 20 % perchloric acid and 80 % ethanol at 260 - 270 K, 20 V and 0.1 A.
3. Results and discussion
Typical creep curves for the base metal and weld joints of dissimilar and similar joints are shown in Fig.1(a & b). The joints exhibited higher rate of creep than the base metal. The similar and dissimilar weld joint of 9Cr-1MoVNb steel had higher creep rupture than 9Cr-1Mo steel [10,11]. Systematic microstructural studies were carried out on the samples drawn near the failed regions.
3.1. Role of secondary carbides in similar joints
Carbide density across the HAZ of the post-weld heat-treated and creep tested weld joint specimens were measured employing SEM. A series of SEM photographs at an interval of 0.1 mm was taken from around the outer edge of HAZ. The presence of fine grain, devoid of typical martensitic microstructure having packets of lath as in the base metal, could clearly identify the intercritical region at the outer edge of HAZ. The very fine MX type of carbonitrides in the matrix (less than 100 nm) could not be resolved by SEM and it was considered that the particles in the SEM images were of M23C6 type of carbides. A lesser number of relatively coarser particles in the intercritical region of the HAZ compared to other constituents of the joint as is evident from the Fig. 2(a &b).
Fig.1 (a) Creep curves for base metal and dissimilar joints of ferritic steel.
Fig.1 (b) Variation of creep rupture life of modified 9Cr-1Mo base metal and weld joint with applied stress at different test temperatures.
The microstructures developed across the HAZ of the ferritic steels can be understood on the basis of the peak temperatures experienced by the base metal during weld thermal cycle and the phase transformation experienced by the steel. The region of HAZ closest to the weld-fusion boundary experienced peak temperatures well above the (a+y) / у (Ac3) phase transformation boundary. At these temperatures, the dissolution of carbides that impede the austenitic grains would occur resulting in the formation of coarse grain austenite.
Fig. 2. Intercritical region in similar weld joint (a) bulky carbides on boundaries (b) creep cavities associated with particles
Austenite has a high solubility for both carbon and nitrogen. This enables the dissolution of the chromium and iron rich M23C6 carbides in the austenite even in the short period of weld thermal cycle.
Partial dissolution of M23C6 carbides and loss of coherency of the prior existing MX type of carbonitrides in the austenite transformed phase result in the change of shape of the intragranular MX type of carbonitrides from needle shape to spheroidal with the consequent reduction in size and increase in inter-particle distance. Further on long term creep exposure, the formation of Z-phase occurs [12]. These microstructural changes in the intercritical region of HAZ result in the release of dislocations to form coarser sub-grain structure in the steel. The type IV cavity nucleation occurs at the coarser precipitate / matrix interface.
3.2. Role of microstructure in dissimilar joints
In the case of dissimilar joints, creep cavity was found to nucleate at the weld interface (Fig.3a) and the cavitation was associated with coarse particles at the weld interface (Fig.3(b)). Significantly less cavitation was observed in the diffuse ferritic / austenitic interface resulted from the weld bead overlap from subsequent weld passes and cracking at the weld interface had deviated to the coarse grain HAZ region in ferritic steel at the diffuse interface, leaving a ferritic ligament attached with the Inconel weld metal. This also raises the
possibility of increasing the creep rupture life of ferritic / austenitic dissimilar weld joint by increasing number of weld passes to increase the proportion of diffuse interface in the ferritic / austenitic interface.
The microstructure at the weld interface of dissimilar weld joint has been investigated further in detail. Fig.4 depicts the precipitation of intermetallic precipitates in addition to formation of martensite at the weld interface. It can be understood based on the redistribution of nickel, chromium and iron across the 9Cr-1MoVNb / Inconel 182 weld interface. It has been observed that the interface region had particularly higher nickel and iron contents. On subjecting to post weld heat treatment at temperatures between 973 - 1033 K, it is expected that a portion of the weld interface bond having relatively high nickel, iron and chromium contents would have transformed to austenite during heating and transformed to martensitic on cooling. The freshly formed martensite imparted high hardness to the interface. This resulted in higher creep strength gradient across the weld interface of dissimilar joint than that across the soft intercritical HAZ.
Fig.3. (a) Series of creep cavities at the weld interface; (b) Cavities associated with interface particles in 9Cr-1Mo-VNb dissimilar weld
joint creep tested at 180 MPa
Fig. 4. Ferritic/austenitic weld interface exhibit formation of martensite, Nb rich intermetallic precipitate.
4. Conclusions
• Dissimilar weld joints of ferritic and austenitic steels possessed lower creep rupture strength than the ferritic steel base metals.
• 9Cr-1Mo steel joints exhibited type IV failure which was associated with cavitation localization in the soft intercritical HAZ. In the 9Cr-1MoVNb steel weld joint due to (i) the replacement of martensite laths with high dislocation density by large sub-grains having low dislocation density, (ii) coarsening of M23C6 type of
carbides at grain and sub-grain boundaries, and (iii) the change in shape of the V/Nb carbonitrides from needle to spheroidal and the reduction of their misfit with matrix (iv) and formation of Z phase. • On the other hand, at lower stresses the failure in dissimilar weld joint occurred in the ferritic-austenitic weld interface which was associated with the creep cavitation at the interface particles under higher stress concentration resulting from higher creep strength gradient across the interface.
References
[1] C. D. Lundin, J. A. Henning, R. Menon, and K. K. Khan: 'Transformation, Metallurgical Response and Behaviour of the Weld Fusion Zone and Heat Affected Zone in Cr-Mo Steels for Fossil Energy applications', ORNL Report, 0RNL/Sub/81-07685/02 & 77,1987.
[2] Shuller, H. J., Haigh, L. and Woitscheck, A., 'Cracking in the weld region of shaped components in hot lines', Der Machinenschaden, 1974 (47) 1-13.
[3] A.K. Bhaduri, S. Venkadesan and P. Rodriguez: Intl. J. Pres.Ves.Piping, 1994 (58) 251-65.
[4] J. D. Parker: Mater. High Temp, 1994 (11) 25-33.
[5] J.D. Parker and G. C. Stratford: J. Mater. Sci, 2000 (35) 4099-4107.
[6] Yi Gong et al.: Fatigue Fract. Engg. Mater. Struct., 2010 , online.
[7] K. Laha, K. S. Chandravathi, K. Bhanu Sankara Rao, S. L. Mannan: Met. and Mat. Trans., 2001(32A) 115-124.
[8] M.Yamazaki, T. Watanabe, H. Hongo and M. Tabuchi: J. Power and Energy Systems, 2008 (2) 1140-49.
[9] Junchao An, Hingyang Jing, Guangchun Xiao, Lei Zhao and Lianyong Xu: J. Mater. Engg. Performance.; 06 November 2010, online.
[10] K. Laha, K. S. Chandravathi, P. Parameswaran, K. Bhanu Sankara Rao and S. L. Mannan: Met. and Mater. Trans., 2007 (38A) 58-68.
[11] K. Laha, K. S. Chandravathi, P. Parameswaran and K. Bhanu Sankara Rao: Met. and Mater. Trans, 2009 ( 40A) 386-97.
[12] P.Parameswaran, K.Laha, K.S.Chandravathi, M.D.Mathew and K.Bhanu Sankara Rao, Trans. IIM, 2010 (63) 479-83.
