The chemistry of the aqueous solution, surface chemistry of zirconium oxide (ZrO2), and the physical structure of the corrosion film have to be considered for an understanding of the mechanism of corrosion of zirconium alloys in aqueous solutions. Based on information available in all these areas, we are proposing a model for oxide growth on Zircaloy-4 fuel cladding and Zr-2.5Nb pressure tube material, in lithium hydroxide (LiOH) solutions with and without added boric acid (H3BO3).
Corrosion exposures were at 360°C and were short term of four-day duration. Concentration of lithium covered the range 0.7 to 3500 ppm and boron was added at 300, 600 and 1200 ppm. Weight gain, Fourier Transform InfraRed (FTIR) spectroscopy, and Secondary Ion Mass Spectrometry (SIMS) were used to characterize the oxide films. Potentiodynamic polarization measurements were made in separate tests at 315°C. The chemistry of LiOH-H3BO3 system at 300 to 360°C was evaluated from the ionization constants of water, LiOH, and H3BO3.
There is no simple relationship between pH and corrosion. In the absence of boron acceleration in corrosion, in these short-term tests, is observed at concentrations >350 ppm Li for Zircaloy 4 and at >60 ppm Li for Zr-2.5Nb. This is attributed to the modification of normal oxide growth by surface OLi groups formed by reaction with undissociated LiOH in the solutions. In the case of Zircaloy-4, the inhibition of accelerated corrosion by H3BO3 is greater than what can be accounted for by its neutralizing action on LiOH. The mechanism suggested is the removal of surface OLi groups by reaction with non-ionized H3BO3. Boric acid at high concentrations has a detrimental effect on the accelerated corrosion of Zr-2.5Nb alloy.
According to the model proposed for oxide growth, the corrosion behavior can be classified into two categories: (1) growth of post-transition type films under acceleration conditions where solution had access into the oxide and (2) growth of pre-transition films under non-acceleration conditions where the solution had not gained access into the oxide. For Case 1, the pores have to be >2 nm in diameter so that interfacial double layer requirements are met for solution incursion in the oxide and the unleachable lithium and boron on the surfaces of oxide grains are from ion exchange and reactions with LiOH and H3BO3. In Case 2, the outer surfaces of oxide grains, which were exposed to the solution, have lithium and boron similar to that described for Case 1. The lithium and boron in the bulk of these pre-transition films are ascribed to LiOH and H3BO3 hydrogen-bonded to water and adsorbed on the surfaces of oxide grains.