The Local Atomic Joule Magnetostriction of Fe81Ga19

Abstract

Using X-ray Absorption Spectroscopy (XAS) in differential mode (DiffXAS), the magnetostriction of a Fe81Ga19 splat cooled ribbon has been measured and the strain coefficients quantified. Due to the local atomic nature of XAS, this represents the first microscopic analysis of such a system, and was made possible only by recent advances in synchrotron radiation based techniques, capable of detecting atomic strains on the scale of femtometres [1]. Previously, magnetostriction measurements have relied on macroscopic techniques, commonly via strain gauges. However, such measurements on thin films and ribbons, which tend to be of the greatest importance from a technical perspective, are notoriously difficult. This is in part due to the measured strains being extremely small over a sample thickness of a few tens of microns, but also due to the practicalities of coupling such a sample to a sensor. Consequently, published magnetostriction coefficients vary immensely. In some cases, giant magnetostrictions have been reported for Fe(1-x)Gax ribbons [2][3] although doubt has recently been cast upon their validity [4]. This serves to assert a need for a more fundamental approach to measuring magnetostrictive strains. A need which is satisfied by DiffXAS. Being based upon XAS, DiffXAS probes changes in local atomic structures and is just dependent upon the short-range order of the first one or two atomic shells surrounding an absorbing atom. However, whilst even giant magnetostrictive strains exhibited by a number of rare-earth based Fe alloys are on the very limits of detection by conventional XAS techniques, DiffXAS offers an increase in sensitivity of two orders of magnitude and so makes such strains easily measurable. Direct detection of strains on the scale of tens or hundreds of ppm then becomes possible. Furthermore, in principle, this is true for any type of strain that is reproducible upon the modulation of some external sample property [5]. Additionally, since x-ray absorption is chemically selective, these structural changes may be viewed from different positions within the crystal lattice, and so the underlying significance of different atomic species in the overall process elucidated. Contributions from different types of bonds within the structure may then be decoupled and analysed. Such information has immense value when trying to obtain fundamental knowledge of atomic strains, and particularly when attempting to verify theoretical models. Concerning Fe(1-x)Gax, a theory for the observed strain enhancement was first put forward by Wu [6] in 2002, which has more recently been developed by Cullen et al [7] after modelled the behaviour of the material’s magnetocrystalline anisotropy. Experimental verification of these proposals has yet to be presented, but is something that DiffXAS has the potential to provide. Using this technique, we have focused on the problem of enhanced magnetostriction observed in the Fe(1-x)Gax system. Such systems have attracted significant interest from a technological and device applications perspective since, although they do not possess truly giant magnetostrictions, they are both absent of expensive rare-earth components and have desirable mechanical properties. They also show appreciable low-field magnetostriction, saturating at fields of only several hundred Oersteds. Working within these saturation conditions, DiffXAS detects the changes in photoelectron scattering path length induced by structural distortions that occur when the sample’s magnetisation vector is modulated between two states – parallel and perpendicular to the x-ray polarisation vector. Subsequent data analysis uses a framework of Cartesian tensors to model the structural properties of the sample, and ab initio XAS theory to model the observed perturbations. From this, the atomic strain tensor may be derived and related to the sample magnetisation vector in order to find the coefficients of the Joule magnetostriction tensor [8]. These coefficients may then be reduced to the more familiar macroscopic 100 and 111 coefficients by exploiting the crystal symmetry elements [8]. Working on a splat cooled foil of Fe81Ga19, preliminary analyses performed with this technique have yielded a magnetostriction coefficient of (3/2)100 = 250±20ppm (111 coefficient is approximately zero for this composition), based upon a disordered A2 structure, determined from analysis of the sample’s conventional XAS signal. This analysis failed to detect the D03 structure reported by some authors for this composition. Further experiments are planned to examine the full range of compositions over which magnetostriction enhancement is observed in this system. [1] R.F. Pettifer, O. Mathon, S. Pascarelli, M.D. Cooke, M.R. J. Gibbs, Nature 435, 79 (2005) [2] M.C. Zhang, H.L. Jiang, X.X. Gao, J. Zhu, and S.Z. Zhou, J. Appl. Phys. 99, 023903 (2006) [3] G.D. Liu, L.B. Liu, Z.H. Liu, M. Zhang, J.L. Chen, J.Q. Li, G.H. Wu, Y.X. Li, J.P. Qu, T.S. Chin, J. Appl. Phys. 84, 2124 (2004) [4] R. Grössinger, R. Sato Turtelli, N. Mehmood, S. Heiss, H. Müller, C. Bormio-Nunes, J. Mag. Mag. Mater., in press [5] M. P. Ruffoni, R.F. Pettifer, S. Pascarelli, A. Trapananti, O. Mathon in X-Ray Absorption Fine Structure XAFS13, edited by B. Hedman and P. Pianetta, AIP Conf. Proc. No. 882 (AIP, 2007), p. 838. [6] R. Wu, J. Appl. Phys. 91, 7358 (2002) [7] J. Cullen, P. Zhao, M. Wuttig, J. Appl. Phys. 101, 123922 (2007) [8] E du T, de Lacheisserie in Magnetostriction: Theory and Applications of Magnetoelasticity (CRC Press, 1993), p133 and 161-162