Magnetic phase transitions and the magnetothermal properties of gadolinium

TLDR

Interstitial impurities lower the paramagnetic–ferromagnetic transition temperature and can bias magnetocaloric measurements in pulsed fields. The study measured the magnetocaloric effect in Gd samples using direct adiabatic temperature rise, magnetization, and heat‑capacity techniques. All three techniques consistently determined a zero‑field Curie temperature of 294 K; the transition shifts linearly (~6 K/T) above 2 T, the spin‑reorientation transition at 227 K is suppressed by fields >2–2.5 T, and Gd remains ferromagnetic down to ~4 K, with an additional anomaly near 132 K.

Abstract

A study of four Gd samples of different purities using ac susceptibility, magnetization, heat capacity, and direct measurements of the magnetocaloric effect in quasistatic and pulse magnetic fields revealed that all techniques yield the same value of the zero-field Curie temperature of 294(1) K. The Curie temperature determined from inflection points of the experimental magnetic susceptibility and heat capacity is in excellent agreement with those obtained from the magnetocaloric effect and Arrot plots. Above 2 T the temperature of this transition increases almost linearly with the magnetic field at a rate of $\ensuremath{\sim}6\mathrm{K}/\mathrm{T}$ in fields up to 7.5 T. The spin reorientation transition, which occurs at 227(2) K in the absence of a magnetic field, has been confirmed by susceptibility, magnetization, and heat-capacity measurements. Magnetic fields higher than 2--2.5 T apparently quench the spin reorientation transition and Gd retains its simple ferromagnetic structure from the ${T}_{C}(H)$ down to $\ensuremath{\sim}4\mathrm{K}.$ The nature of anomaly at $T\ensuremath{\cong}132\mathrm{K},$ which is apparent from ac susceptibility measurements along the $c$ axis, is discussed. The presence of large amounts of interstitial impurities lowers the second-order $\mathrm{paramagnetic}\ensuremath{\leftrightarrow}\mathrm{ferromagnetic}$ transition temperature, and can cause some erroneous results in the magnetocaloric effect determined in pulsed magnetic fields. The magnetocaloric effect was studied utilizing the same samples by three experimental techniques: direct measurements of the adiabatic temperature rise, magnetization, and heat capacity. All three techniques, with one exception, yield the same results within the limits of experimental error.