Molten lead is a potential electrodynamic fluid for magneto-hydrodynamics (MHD) liquid metal (LM) generators in a project developed in the Center for MHD studies at Be n-Gurion University [1]. Obviously, the structural materials of the generator have to be corrosion-proof in liquid lead turbulent flow, and the material of the wall close to the electrode zone should not be ferromagnetic. Austenitic type 316 stainless steel (SS316) is a promising structural material which meets these requirements. To our knowledge, no data are yet available about the behaviour of SS316 in liquid lead turbulent flow. There is very little information about the behaviour of SS316 in contact with static or low speed flow molten lead in the literature [2]. The interaction of SS316 with Pb-17Li liquid alloy is described in a larger number of works, perhaps because this melt is a potential breeder in fusion reactors [3-5]. As a rule, selective nickel dissolution (leaching) from steel is mentioned. The selective process is related to higher nickel solubility in the liquid eutectic Pb-Li as compared with iron and chromium. In this letter the results of an investigation of the corrosion of SS316 samples in turbulent liquid lead flow are described. The experimental conditions determining either selective or uniform types of SS316 corrosion are discussed and the role of oxygen impurities is discussed in greater detail. An analysis of the equilibrium states in the P b P b O F e C r Ni-Mn system was also performed to estimate the characteristics of steel-liquid interaction under our experimental conditions. The samples under investigation were in the form of discs 40 mm in diameter and 8 m thick. The samples were polished and degreased with acetone. Tests were carried out by the rotating disc method described previously [6]. The test conditions were: temperature 500 °C, sample rotation rate 2400 rpm. The test container for the molten lead (99.97%) was made of carbon steel and the melt volume was approximately 1 I. Nitrogen (99.7%) was padded through the container above the melt surface at a flow rate of 4 1 min -1. The duration of the experiment was 800h. Every 200h the sample was removed from the container for X-ray diffraction (XRD) and scanning electron microscopy (SEM)/electron probe X-ray microanalysis (EPMA) [6]. According to EPMA, the steel composition was (wt %): Cr 17.9, Ni 10.5, Mo 2.5, Mn 1.8, Si 2.0, Fe remainder. The steel in the initial state proved to be two-phase: the main phase was 7 austenite (a = 0.3602 nm) and the second phase was oL-ferrite (a = 0.2870 nm). This corresponds to the equilibrium phase diagram of Fe -Ni -Cr [7]. It should be noted that the steel retained its phase composition during the entire test. The y phase cell parameter did not change during the test, but the parameter of the o~ phase increased to 0.2890 nm after 800 h. The concentrations of iron, nickel and chromium near the sample surface, according to EPMA data, did not change during the test period (Fig. 1). The presence of 7 and ol phases, complex oxide phase FeCr204 and small quantities of Fe2SiO4 (faylite) and o~-MnSiO3 was determined by XRD analysis of the sample surfaces. As follows from the above data, the composition of the subsurface zone remained unchanged for a long time. The dimensions of the samples also remained constant to within + 10 #m. An obvious question arises as to why there is no selective dissolution of nickel and, as a result, no build-up of o~-phase on the surface of the sample tested, despite the large solubility difference of Ni, on the one hand, and iron and chromium, on the other, in liquid lead (this difference amounts to four or five orders of magnitude at 800 K [8-10]). Firstly, the rate of nickel leaching from SS316 by liquid lead turbulent flow under the rotating disc regime was estimated. The rate of pure nickel dissolution, Jr~i, under our experimental conditions is expressed by [11]:
Read full abstract
Sep 19, 2014
Boris B Bokhonov + 6
