group » Research » theory

» Landau theory and mean field approximation

We have recently shown [1] how the Landau theory of phase transitions can be used to model magnetocaloric properties of ferromagnets, such as the dependence of ∆S_M in ∆H and T, and studied how the Landau coefficients, related to magnetoelastic coupling and their temperature dependence, can be used to interpret the change from 1st- to 2nd-order phase transitions (PT) [2]. A  similar application of this method was used in the study of the itinerant electron metamagnetic La(Fe1-xSix)13 system by Fujita et al. [3], where the dependence of the Landau parameters with temperature reflects the thermal variation of amplitude of spin fluctuations [4]. The insight given by the Landau theory is therefore of extreme value when one wants to understand how the couplings of the studied systems influence a 1st- or 2nd-order PT and the magnetic entropy changes, guiding materials modifications and conditions for optimization. However, the Landau theory is not designed to account for the magnetic behavior up to high fields/ magnetic saturation, since it depends on a power expansion of the order parameter. This limitation can be overcome by using a generalized mean-field analysis of magnetization data, which is also well suited to estimate magnetocaloric properties of materials. By using the general properties of the mean-field model, the mean field exchange field, usually described from H_exch=λM, is estimated directly by a scaling method from magnetization data. The generalization proposed does not suppose a constant λ, and instead allows extracting the dependence of λ in M and/or T to account for additional couplings of the magnetization to other degrees of freedom. We study results of this analysis for manganite systems with 1st- and 2nd order PT, and Tc ranging from 150 to 340 K. This method allows one to analyze the dependence of λ in M and/or T, which would respectively indicate adiabatic  magnetoelastic couplings or vibronic coupling proposed for manganites [5]. The analysis of the data in a large temperature and field range allows a suitable interpolation scheme and a consistent study of the magnetic entropy.

[1] A. Campos et al., Nature Materials 5 (2006) 802. ::: PDF

[2] P.J. von Ranke et al., Phys. Rev. B. (2007) in press.

[1] J.S. Amaral and V.S. Amaral, J. Magn. Magn. Mater. 272–276 2104–2105 (2004).

[2] J.S.Amaral, M.S. Reis, V.S. Amaral, T.M. Mendonça, J.P. Araújo, M.A. Sá, P.B. 

Tavares, J.M. Vieira, J. Magn. Magn. Mater. 290–291 686–689 (2005).

[3] A.Fujita and K.Fukamichi, IEEE Transct. on Magn., 41, nr 10, 3490-3492 (2005).

[4] H.Yamada and T.Goto, Phys. Rev. B, vol. 68, pp. 184417:1–184417:7 (2003).

[5] F. Rivadulla, M. Otero-Leal, A. Espinosa, A. de Andrés, C. Ramos, J. Rivas, and J. B.

Publications:

1. Journal of Magnetism and Magnetic Materials 272 (2004) 2104 ::: PDF

2. Second IIF-IIR International Conference on Magnetic Refrigeration

at Room Temperature (2007) 161 ::: PDF

» Anisotropic magnetocaloric effect

We study a new way of obtaining the magnetocaloric effect due to the crystal electrical field  quenching of the total angular momentum in magnetic system where a strong spin reorientation is present. The theoretical model is applied to DyAl₂  and the results predict a considerable magnetic entropy change by rotating a single-crystal in fixed magnetic field. The obtained temperature and magnetic field dependencies of the magnetization component along the <111>-crystallographic direction are in good agreement with the recently reported experimental data.

This work was proposed and developed by the brazilian groups. Our contribution was limited to discussion and comments of the results.

Publications:

1. Physical Review B (2007) in press

2. Journal of Magnetism and Magnetic Materials (2007) submitted