The Title Goes Here

INTRODUCTION

Perspective accelerator driven subcritical systems (ADS) employing liquid Pb and PbBi as a coolant and / or target [1] require investigation of corrosion effects of steels in such environment at relevant temperatures. The paper describes corrosion experiments with AISI 316 L, 1.4970 and MANET steel in original state and with an Al alloyed surface (austenites only).

EXPERIMENTAL

The experiments which were planed up to an exposure time of 7200 h were interrupted for specimen investigations at 2000 and 4000 h. They were carried out in two different LBE loops, one at IPPE, Obninsk, and the other at PROMETEY, St. Petersburg, with liquid metal velocities of 1.3 and 0.5 m/s, respectively. The concentration of oxygen in the liquid alloy was controlled at 10^-6 wt%. Specimens consisted of tube and rod sections which were stacked together to a test rod [2].

RESULTS

Thick protective oxide layers develop after 2000 and 4000 h on the surface of the martensitic MANET steel at 420 and 550 °C. These oxide layers however, spall off afterwards and after 7200 h a new oxide layer is observed which, however, contains inclusions of PbBi, but with no sign of a progressive dissolution attack.

As opposed to the martensitic steel, very thin layers on austenitic steel develop at 420 °C which consist of chromia only. However, the austenites produce thick layers of magnetite and spinel also, if a temperature of 550 °C is applied. After 4000 h the 1.4970 austenitic steel shows inclusions of PbBi in the spinel layer, not however, in the magnetite scale upper part. These inclusions indicate a dissolution attack before or during the formation of the new magnetite scale after spalling off between 2000 and 4000 h. Indeed, there is no further penetration of PbBi into the spinel zone during extension of exposure to 7200 h. The magnetite scale on top of the surface is dense enough to seal it against the PbBi bath. It does, however, not hinder the oxygen to migrate to the spinel zone and to diffuse into the metallic substrate.

The other austenite, AISI 316L, shows a different behavior. Its oxide layers spall also off between 4000 and 7200 h, but a new oxide scale develops with no dissolution attack in its transient state. Therefore, no inclusions of LBE are observed in the spinel zone. The different behavior is explained by the different structures of both austenites. The AISI 316L rod specimen is coarse grained, while the 1.4970 tube specimen has a fine grained, plate like structure (< 1 µm thick)

The martensitic steel is not tested at 600 °C. But, both austenitic steels show catastrophic dissolution attack at 600 °C already after 4000 h of exposure.

Surface Al-alloyed austenitic steels do not have any dissolution attack at all temperatures and exposure times. They keep their thin protective alumina scales unchanged.

The table I shows the suitability of the steels, derived from the experiments conducted so far.

TABLE I. Suitability of steels for use in PbBi.
	4000 h			7200 h
steel	420°C	550°C	600°C	550°C
AISI 316L	+	+	-	+
1.4970	+	(+)	-	(+)
MANET	+	+		(+)
austenite Al-alloyed	+	+	+	+
(+) PbBi inclusions in the renewed oxide scale

REFERENCES

1. C. Rubbia, J.A.Rubio, S. Buono, F. Carminati, Conceptual Design of a Fast Neutron Operated High Power Energy Amplifier, CERN/AT/95-44 (ET), September 29, 1995.

2. G. Müller, A. Heinzel, J. Konys, G. Schumacher, A. Weisenburger, F. Zimmermann, V. Engelko, A. Rusanov, V. Markov, J. Nucl. Mater., 301 (2002) 40-46