PALE:ClassicArticles/GMR/Article17

From NSDLWiki

Spin-dependent Tunneling Conductance of Fe|MgO|Fe Sandwiches and Theory of Tunneling Magnetoresistance of an Epitaxial Fe/MgO/Fe(001) Junction

Butler, et al. Article Link

Marthon & Umerski Article Link

Butler, W.H., Zhang, X.G., Schulthess, T.C. & MacLaren, J.M., 2001. Spin-dependent Tunneling Conductance of Fe|MgO|Fe Sandwiches. Phys. Rev. B 63, 054416 and Mathon, J. & A. Umerski, A., 2001. Theory of Tunneling Magnetoresistance of an Epitaxial Fe/MgO/Fe(001) Junction. Phys. Rev. 63, 220403(R)

Essay about these articles

This is an interesting story of how a well-honed theoretical prediction of a very large MR ratio was indeed on target, in spite of the fact that the prevailing TMR were considerably smaller. Although it took its final form in 2001 in the papers cited above, among others, the idea that there were propitious conduction channels across certain oxide barriers, provided they were crystalline, was enunciated in the mid- to late-nineties by the groups of investigators in London, New Orleans, and Tennessee. Against the backdrop of extant TMR ratios of under 100%, their predictions of 1000%, while based on firm theoretical grounds, sounded slightly hubristic. The key observation was that barriers used in earlier experiments were made of alumina-- aluminum that has been oxidized--which produces an amorphous barrier.

For conduction across a barrier, the difference between an amorphous and a crystalline composition is that for amorphous barriers the momentum in the plane of the barrier is scrambled, i.e., lost, while for crystalline barriers it is conserved. This allows one to use symmetry arguments to match up, or join together, evanescent waves in the barrier to wavefunctions in the electrodes. As an example, for a manganese oxide, MgO, barrier, one particular state with DELTA1 symmetry, which is largely due to an s-like atomic wavefunction, decays far less than the other evanescent states in the barrier. The reason for this robustness is these states have a small transverse momentum, so that for a given total momentum, i.e., the Fermi momentum [momentum at the Fermi level], most of the momentum is in the longitudinal [forward] direction. Therefore, if s-like wavefunctions exist in the electrodes this barrier state is able to effectively couple the electrodes; in fact, far more than the other evanescent states. Iron, Fe, has an s-like wavefunction at the Fermi level majority spin, but the minority channel does not have a state with this character. I quote from Butler, “In the majority channel there is a state….The minority channel has four states with the same symmetries as the states of the majority channel with the crucial exception that the majority state is replaced by a minority state.”

The above cited article, along with a paper by Mathon and Umerski, found these symmetry considerations dictate that when the magnetization of the electrodes were parallel, there is a highly conducting path for majority, spin-up electrons between the Fe electrodes through the barrier’s state; but that the minority electrons in Fe do not have the appropriate symmetry to couple with this state. In the two-current model of conduction, this means that most of the current is carried by the majority state. When the magnetizations are antiparallel, while one spin state can couple to the state in one of the Fe electrodes, the other electrode does not have the correct symmetry [as long as spin-flip is not possible, which is usually the case]. Therefore, the current is considerably smaller for the antiparallel alignment of the electrode than for the parallel.

The predicted TMR ratios were of the order of 1000%. It took experimentalists three years to validate this prediction [1] . At first researchers were finding about 350%, but with further refinements scientist have recently achieved over 600% at room temperature for CoFeB/MgO/CoFeB tunnel junctions [2].

Other than this symmetry-based consideration there was another way of achieving high TMR ratios; there has been considerable activity using half-metallic electrodes, i.e., electrodes where only one spin state exists at the Fermi level. For example, lanthanum strontium titanate, (LSMO) [La2/3Sr1/3MnO3] was predicted by band structure calculations to be nearly half-metallic. In 2002 a group lead by Albert Fert found a TMR ratio of more than 1800 % at 4K, for a magnetic tunnel junction (MTJ) by using LSMO for magnetic electrodes and strontium titanate, STO [SrTiO3] as the insulating barrier, and they inferred an electrode spin polarization of at least 95 % [3]. Half-metallic electrodes have also been useful as spin analyzers inasmuch as that if one knows their spin polarization, e.g., if the density of states at the Fermi level is almost completely spin-up or spin-down, then one can determine the spin polarization of a counter electrode of an MTJ [4].

References

[1] S. Yuasa, A. Fukushima, T. Nagahama, K. Ando and Y. Suzuki, Jpn. J. Appl. Phys. Part 2 43, L588 (2004); S.S.P. Parkin, C. Kaiser, A. Panchula, P.M. Rice, B. Hughes, M. Samant and S.-H. Yang, Nat. Mater. 3, 862 (2004); S. Yuasa, T. Nagahama, A. Fukushima, Y. Suzuki, and K. Ando, ibid 3, 868 (2004).

[2] S. Ikeda, J. Hayakawa, Y. Ashizawa, Y. M. Lee, K. Miura, H. Hasegawa, M. Tsunoda, F. Matsukura, and H. Ohno, Appl. Phys. Lett. 93, 082508 (2008).

[3] M. Bowen, M. Bibes, A. Barthélémy, J.-P. Contour, A. Anane, Y. Lemaître, and A. Fert, Appl. Phys. Lett. 82, 233 (2003).

[4] J. M. De Teresa, A. Barthélémy, A. Fert, J. P. Contour, R. Lyonnet, F. Montaigne, P. Seneor, and A. Vaurès, Phys. Rev. Lett. 82, 4288 (1999); Jose Maria De Teresa, Agnès Barthélémy, Albert Fert, Jean Pierre Contour, François Montaigne and Pierre Seneor, Science 208, 507 (1999).

Discussion Question

What characteristics must be present in tunnel junctions for one to observe the large TMR ratios predicted by the groups of Butler and Mathon?

The above article is reprinted with permission from the author(s) of Butler, W.H., Zhang, X.G., Schulthess, T.C. & MacLaren, J.M., 2001. Spin-dependent Tunneling Conductance of Fe|MgO|Fe Sandwiches. Phys. Rev. B 63, 054416 and Mathon, J. & A. Umerski, A., 2001. Theory of Tunneling Magnetoresistance of an Epitaxial Fe/MgO/Fe(001) Junction. Phys. Rev. 63, 220403(R). Copyright (2001) by the American Physical Society. Readers may view, browse, and/or download material for temporary copying purposes only, provided these uses are for noncommercial personal purposes. Except as provided by law, this material may not be further reproduced, distributed, transmitted, modified, adapted, performed, displayed, published, or sold in whole or part, without prior written permission from the American Physical Society.

Select articles citing these papers

Butler

Wang, S. G., R. C. C. Ward, et al. (2008). "Temperature dependence of giant tunnel magnetoresistance in epitaxial Fe/MgO/Fe magnetic tunnel junctions." Physical Review B 78(18).

Karpan, V. M., P. A. Khomyakov, et al. (2008). "Switching on magnetism in Ni-doped graphene: Density functional calculations." Physical Review B 78(19).

Isogami, S., M. Tsunoda, et al. (2008). "In situ heat treatment of ultrathin MgO layer for giant magnetoresistance ratio with low resistance area product in CoFeB/MgO/CoFeB magnetic tunnel junctions." Applied Physics Letters 93(19).

Jiang, X., R. Wang, et al. (2005). "Highly spin-polarized room-temperature tunnel injector for semiconductor spintronics using MgO(100)." Physical Review Letters 94(5).

Zhang, X. G. and W. H. Butler (2004). "Large magnetoresistance in bcc Co/MgO/Co and FeCo/MgO/FeCo tunnel junctions." Physical Review B 70(17).

Zhang, X. G., W. H. Butler, et al. (2003). "Effects of the iron-oxide layer in Fe-FeO-MgO-Fe tunneling junctions." Physical Review B 68(9).

Tsymbal, E. Y., O. N. Mryasov, et al. (2003). "Spin-dependent tunnelling in magnetic tunnel junctions." Journal of Physics-Condensed Matter 15(4): R109-R142.

Wunnicke, O., N. Papanikolaou, et al. (2002). "Effects of resonant interface states on tunneling magnetoresistance." Physical Review B 65(6).

Meyerheim, H. L., R. Popescu, et al. (2001). "Geometrical and compositional structure at metal-oxide interfaces: MgO on Fe(001)." Physical Review Letters 8707(7).

Klaua, M., D. Ullmann, et al. (2001). "Growth, structure, electronic, and magnetic properties of MgO/Fe(001) bilayers and Fe/MgO/Fe(001) trilayers." Physical Review B 6413(13).

Mathon

Wang, S. G., R. C. C. Ward, et al. (2008). "Temperature dependence of giant tunnel magnetoresistance in epitaxial Fe/MgO/Fe magnetic tunnel junctions." Physical Review B 78(18).

Kuzmenko, I. and V. Fal'ko (2008). "Canted magnetization texture in ferromagnetic tunnel junctions." Physical Review B 78(18).

Karpan, V. M., P. A. Khomyakov, et al. (2008). "Switching on magnetism in Ni-doped graphene: Density functional calculations." Physical Review B 78(19).

Jiang, X., R. Wang, et al. (2005). "Highly spin-polarized room-temperature tunnel injector for semiconductor spintronics using MgO(100)." Physical Review Letters 94(5).

Djayaprawira, D. D., K. Tsunekawa, et al. (2005). "230% room-temperature magnetoresistance in CoFeB/MgO/CoFeB magnetic tunnel junctions." Applied Physics Letters 86(9).

Zutic, I., J. Fabian, et al. (2004). "Spintronics: Fundamentals and applications." Reviews of Modern Physics 76(2): 323-410.

Yuasa, S., T. Nagahama, et al. (2004). "Giant room-temperature magnetoresistance in single-crystal Fe/MgO/Fe magnetic tunnel junctions." Nature Materials 3(12): 868-871.

Yuasa, S., A. Fukushima, et al. (2004). "High tunnel magnetoresistance at room temperature in fully epitaxial Fe/MgO/Fe tunnel junctions due to coherent spin-polarized Tunneling." Japanese Journal of Applied Physics Part 2-Letters & Express Letters 43(4B): L588-L590.

Parkin, S. S. P., C. Kaiser, et al. (2004). "Giant tunnelling magnetoresistance at room temperature with MgO (100) tunnel barriers." Nature Materials 3(12): 862-867.

Tsymbal, E. Y., O. N. Mryasov, et al. (2003). "Spin-dependent tunnelling in magnetic tunnel junctions." Journal of Physics-Condensed Matter 15(4): R109-R142.