Corrosion studies of Ni-Mn-Ga alloys for shape memory applications

Investigations were performed to characterize the corrosion behaviour of polycrystalline Ni-Mn-Ga alloys for shape memory applications. The Ni₄₈Mn₃₀Ga₂₂ alloy stabilizes at room temperature a cubic austenite state, whereas the Ni₅₀Mn₃₀Ga₂₀ alloy is in a tetragonal NM martensite state. Both alloys exhibit low corrosion rates and a spontaneous anodic passivation in alkaline solutions (pH= 8.4; 11). In acidic media the alloys tend towards a more active state corresponding to a general shift of the corrosion potential to more negative values and an increase of the corrosion current density. While at pH= 5 the alloys are still spontaneously passive upon anodization, in sulphuric acid solution with pH= 0.5 a pronounced active-passive transition occurs before a passive state is attained by the formation of marginally protective surface layers (see figure below). Current transient measurements conducted at anodic passive potentials in electrolytes with pH= 5 and 8.4 revealed a diffusion-controlled anodic passive layer growth. XPS studies showed that while air-formed films comprise NiOOH, Ga₂O₃ and MnO₂, passive films formed in near neutral media (pH=5 to pH=8.4) are composed of Ni(OH)₂, NiOOH and Ga₂O₃ in the outer region and of NiO, MnO₂ and MnO in the metal-near region.
Under all conditions studied, but mostly pronounced in acidic environments, the martensite Ni₅₀Mn₃₀Ga₂₀ alloy is significantly more reactive than the austenite Ni₄₈Mn₃₀Ga₂₂ alloy. This can be attributed to a much higher density of surface defects in terms of a large number of twin boundaries, which can act as energetically favoured sites for corrosion attack. SEM analysis of the martensite Ni₅₀Mn₃₀Ga₂₀ surface revealed a preferential corrosion following the pattern of the twin boundaries. But also the effect of a slightly lower content of Ga, which is the most stable alloy component at low pH values, must be considered.

Potentiodynamic polarization curves in sulphuric acid solution (pH=0.5)

[1] A. Gebert, S. Roth, S. Oswald, L. Schultz, Corrosion Science 51 (2009) 1163-1171