Effect of cold-work on self-welding susceptibility of austenitic stainless steel (alloy D9) in high temperature flowing sodium C. Meikandamurthy a,⇑ , Hemant Kumar b , Gopa Chakraborty b , S.K. Albert b , V. Ramakrishnan a , K.K. Rajan a , A.K. Bhaduri b a Fast Reactor Technology Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603 102, India b Materials Technology Division, Indira Gandhi Centre for Atomic Research, Kalpakkam 603 102, India article info Article history: Received 14 June 2010 Accepted 5 October 2010 abstract Self-welding susceptibility of alloy D9 (15Cr–15Ni–2Mo titanium–modified austenitic stainless steel), used as wrapper in the fuel subassemblies of sodium cooled fast reactor, was studied in flowing sodium. Specimens were tested at 823 K in annealed and in 20% cold-worked condition up to a maximum contact stress of 24.5 MPa and maximum duration of 9 months. The results showed that the annealed alloy D9 showed good resistance to self-welding in all the tests. But 20% cold-worked alloy D9 got self-welded in all the tests except in the test carried out for 3 months duration indicating that tests conducted at high contact stresses and long duration reduce the resistance of the steel to self-weld. Microstructural changes observed in the cold-worked alloy D9 at the location of contact between the mating surfaces indicate dynamic recovery resulting from high contact stress and temperature facilitating self-weld. Ó 2010 Elsevier B.V. All rights reserved. 1. Introduction In sodium cooled fast reactor, core contains fuel subassemblies where the fuel pins are packed in a hexagonal wrapper. Minimum contact area between the fuel subassemblies is set by providing wrapper pads on each of the six faces of the hexagonal wrapper. For India’s Prototype Fast Breeder Reactor (PFBR), which is under construction, 20% cold-worked alloy D9 is the material chosen for hexagonal wrapper of fuel subassemblies and wrapper pads. Cold-work is given to impart improved resistance to void swelling for the alloy D9. During the operation of the reactor, the maximum contact stress to which some of the wrapper pads are subjected to is estimated as 3.3 MPa at 823 K. It is important that high operating temperature, contact stress and fairly long resident time (24 months) of the subassemblies inside the reactor do not lead to self-welding of the subassemblies at the wrapper pads. In the event of self-welding of the wrapper pads, additional force has to be applied to separate the self-welded subassemblies during fuel handling operation. Self-welding is essentially a diffusion bonding phenomenon that occurs when two virgin metallic surfaces are pressed against each other for a particular duration at high temperature. During this process, deformation of the surfaces in contact takes place due to loading and during the subsequent recovery and recrystal- lisation that follow, diffusion of atoms takes place across the con- tact interface resulting in self-welding of the mating surfaces [1]. Liquid sodium in inert environment removes the oxide films that would normally be present on the metallic surfaces, and thus facil- itates self-welding. Limited literature is only available on self-welding studies on stainless steel in sodium. Few sodium experiments were reported in 304, 316 and 321 stainless steel. Relations between the break- away shear force with contact pressure, contact temperature and surface finish were established [2,3]. The self-welding coefficient (W), defined as the shear stress for breakaway per unit contact stress, is related to square root of duration (t) of the self-welding test [4]. Agastini has studied self-welding in 20% cold-worked AISI 316 stainless steel [5]. Recently self-welding studies on annealed and cold-worked alloy D9, and 316LN austenitic stainless steel [6,7] were reported by our team. As 20% cold-worked alloy D9 exhibited susceptibility to self- welding, it is important to understand the mechanism of self-weld- ing of this alloy in the cold-worked condition and also to ensure that self-welding does not take place in the hexagonal wrapper pads of PFBR during the residence time of the reactor. Hence, the present study was conducted. 2. Test facility for self-welding The test set-up for evaluating susceptibility to self-welding is shown in Figs. 1 and 2. It is provided with a rod, outer tube, disc spring, nut, load cell and provision to hold the specimens at bot- tom. The specimens for studying self-welding susceptibility are hollow cylinders of 21.4 mm OD, 15.8 mm ID and 15 mm height. 0022-3115/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jnucmat.2010.10.002 ⇑ Corresponding author. E-mail address: cmm@igcar.gov.in (C. Meikandamurthy). Journal of Nuclear Materials 407 (2010) 165–170 Contents lists available at ScienceDirect Journal of Nuclear Materials journal homepage: www.elsevier.com/locate/jnucmat