Vacancies in thermal equilibrium and ferromagnetism near the Curie temperature

Abstract

Abstract In several refractory body-centred cubic metals (α-Fe, V, Nb, Ta) the binding energy of positrons (e+) trapped in vacancies is too small to permit accurate determinations of the enthalpy of formation of monovacancies, H 1V F , $ H_{{\rm{1V}}}^{\rm{F}},$ by high-temperature positron annihilation. Owing to their larger mass, trapped positive muons (μ+) and π-mesons (π+) are much more firmly bound to vacancies. It is argued that the lattice steering (channelling or blocking) of their charged decay products (e+ or μ+) allows us to obtain accurate H 1V F $ H_{{\rm{1V}}}^{\rm{F}}$ values of the refractory bcc metals. In ferromagnets with high Curie temperatures T C, such as α-Fe, Co, and FeCo alloys, H 1V F $ H_{{\rm{1V}}}^{\rm{F}}$ may also be deduced from muon spin rotation (μ+SR) measurements. However, in Fe and Co this approach is limited by the strong sensitivity of the spontaneous magnetization against temperature fluctuations near T C. The reduction of this sensitivity in the so-called asymptotic critical regime by applying sufficiently strong external magnetic fields is investigated on the basis of the Arrott– Noakes equation. A method for determining the critical amplitudes occurring in this equation is proposed. In disordered Fe1–xCox alloys (0.2 ≤ x ≤ 0.75) the Curie temperatures are sufficiently high for the spontaneous magnetization in the bcc phase not to be critically affected by temperature fluctuations, hence these alloys are well suited for μ+SR investigations of thermal vacancies. From an analysis of the available positron-annihilation and self-diffusion data the vacancy migration enthalpy in disordered Fe0.5Co0.5 is found to be (1.1 ± 0.2) eV, in good agreement with quenching data and with the value established for α-Fe.