journal article·2020-01-21Open access

Systematic derivation of realistic spin models for beyond-Heisenberg solids

Markus Hoffmann, Stefan Blügel

Physical Review B· Volume 101· Issue 2

Abstract

The magnetic properties of many ultrathin films on metal substrates cannot be explained by the Heisenberg exchange interaction. Other known interactions, such as the biquadratic or the four-spin interaction, are also insufficient. This suggests the existence of further multispin interactions. The authors provide a systematic derivation of those interactions based on a multiband Hubbard model combined with a downfolding technique and propose a three-spin interaction. They show that it is strong in the systems with Fe on Ru(0001) and Fe on Rh(111), and even comparable in size to the Heisenberg interaction.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Magnetic properties of thin films · Theoretical and Computational Physics · Physics of Superconductivity and Magnetism

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Peter Grünberg Institut and Institute for Advanced Simulation, Forschungszentrum Jülich and JARA, 52425 Jülich, Germany

Funding

Defense Advanced Research Projects Agency · Deutsche Forschungsgemeinschaft

References (64)

Show source-provided references
1. P. W. Anderson · 1963
2. H. J. Zeiger · 1973
3. C. de Graaf · 2015
4. 10.1038/natrevmats.2017.3110.1038/natrevmats.2017.31 ↗
5. 10.1103/PhysRevLett.87.11680110.1103/PhysRevLett.87.116801 ↗
6. 10.1038/nphys223110.1038/nphys2231 ↗
7. 10.1038/ncomms1361310.1038/ncomms13613 ↗
8. 10.1038/ncomms503010.1038/ncomms5030 ↗
9. 10.1103/PhysRevB.84.22441310.1103/PhysRevB.84.224413 ↗
10. 10.1103/PhysRevB.92.02040110.1103/PhysRevB.92.020401 ↗
11. 10.1088/0305-4608/15/6/01810.1088/0305-4608/15/6/018 ↗
12. 10.1103/PhysRevB.58.29310.1103/PhysRevB.58.293 ↗
13. 10.1103/PhysRevB.64.17440210.1103/PhysRevB.64.174402 ↗
14. M. Fähnle · 200710.1002/9780470022184.hmm122 ↗
15. 10.1103/PhysRevLett.107.01720410.1103/PhysRevLett.107.017204 ↗
16. 10.1103/PhysRevLett.107.11990110.1103/PhysRevLett.107.119901 ↗
17. 10.1007/BF0132860110.1007/BF01328601 ↗
18. 10.1038/nature0580210.1038/nature05802 ↗
19. 10.1080/1478643050050404810.1080/14786430500504048 ↗
20. 10.1103/PhysRevB.72.18441510.1103/PhysRevB.72.184415 ↗
21. In this paper the spin-orbit interaction is neglected and thus the Dzyaloshinskii-Moriya interaction or the magnetic anisotropies are not part of the discussion.
22. 10.1103/PhysRevB.14.309010.1103/PhysRevB.14.3090 ↗
23. 10.1103/PhysRevLett.101.02720110.1103/PhysRevLett.101.027201 ↗
24. 10.1103/PhysRevLett.86.110610.1103/PhysRevLett.86.1106 ↗
25. 10.1103/PhysRev.120.33510.1103/PhysRev.120.335 ↗
26. 10.1016/0038-1098(72)90755-710.1016/0038-1098(72)90755-7 ↗
27. 10.1016/0304-8853(93)91197-F10.1016/0304-8853(93)91197-F ↗
28. 10.1088/0305-4470/10/3/01110.1088/0305-4470/10/3/011 ↗
29. 10.1103/PhysRevB.18.507810.1103/PhysRevB.18.5078 ↗
30. 10.1103/PhysRev.140.A180310.1103/PhysRev.140.A1803 ↗
31. 10.1103/PhysRevLett.11.1010.1103/PhysRevLett.11.10 ↗
32. 10.1103/PhysRevB.33.628710.1103/PhysRevB.33.6287 ↗
33. 10.1103/PhysRevB.74.09240610.1103/PhysRevB.74.092406 ↗
34. 10.1063/1.59389810.1063/1.593898 ↗
35. 10.1103/PhysRevLett.85.238010.1103/PhysRevLett.85.2380 ↗
36. 10.1103/PhysRevB.11.472510.1103/PhysRevB.11.4725 ↗
37. 10.1088/0022-3719/10/8/03110.1088/0022-3719/10/8/031 ↗
38. 10.1103/PhysRevB.37.975310.1103/PhysRevB.37.9753 ↗
39. 10.1007/s00257005015310.1007/s002570050153 ↗
40. 10.1143/JPSJ.70.308910.1143/JPSJ.70.3089 ↗
41. 10.1063/1.36167810.1063/1.361678 ↗
42. 10.1103/PhysRevLett.108.08720510.1103/PhysRevLett.108.087205 ↗
43. 10.1038/nphys204510.1038/nphys2045 ↗
44. 10.1002/pssb.20114709010.1002/pssb.201147090 ↗
45. 10.1103/PhysRevLett.120.20720210.1103/PhysRevLett.120.207202 ↗
46. 10.1103/PhysRevLett.120.20720110.1103/PhysRevLett.120.207201 ↗
47. We would like to note that the nomenclature three-spin interactions has recently drawn criticism as the interaction indicates an interaction of three spins, which would lack time-inversion symmetry. In the context of this paper we use it synonymously to four-spin-three-site interaction.
48. 10.1143/JPSJ.36.4810.1143/JPSJ.36.48 ↗
49. 10.1143/JPSJ.47.78610.1143/JPSJ.47.786 ↗
50. 10.1016/0375-9601(80)90148-610.1016/0375-9601(80)90148-6 ↗
51. 10.1143/JPSJ.24.116910.1143/JPSJ.24.1169 ↗
52. 10.1103/PhysRevB.76.13241210.1103/PhysRevB.76.132412 ↗
53. 10.1016/S0304-8853(97)00166-210.1016/S0304-8853(97)00166-2 ↗
54. 10.1063/1.174806710.1063/1.1748067 ↗
55. R. Winkler · 200310.1007/b13586 ↗
56. 10.1103/PhysRev.149.49110.1103/PhysRev.149.491 ↗
57. 10.1016/j.aop.2011.06.00410.1016/j.aop.2011.06.004 ↗
58. M. Hoffmann, N. Ohs, and S. Blügel (2019), doi: 10.5281/zenodo.3609778.
59. 10.1098/rspa.1963.020410.1098/rspa.1963.0204 ↗
60. 10.1098/rspa.1964.001910.1098/rspa.1964.0019 ↗
61. 10.1098/rspa.1964.019010.1098/rspa.1964.0190 ↗
62. We use ti,α,j,α′′ instead of only ti,α,j,α′ to emphasize the two different types of hoppings. This notation will also be used later on in the paper.
63. This ensures that a ferromagnetic alignment (either all spins up or all spins down) gains no energy in higher-order. This agrees well with the underlying Hubbard model where none of the hopping terms have any effect on those ferromagnetic states as no hopping is possible.
64. 10.1103/PhysRevB.79.09441110.1103/PhysRevB.79.094411 ↗