Landau–Lifshitz–Gilbert equation
Template:Short description In physics, the Landau–Lifshitz–Gilbert equation (usually abbreviated as LLG equation), named for Lev Landau, Evgeny Lifshitz, and Thomas L. Gilbert, is a name used for a differential equation describing the dynamics (typically the precessional motion) of magnetization Template:Math in a solid. It is a modified version by Gilbert of the original equation of Landau and Lifshitz.[1] The LLG equation is similar to the Bloch equation, but they differ in the form of the damping term. The LLG equation describes a more general scenario of magnetization dynamics beyond the simple Larmor precession. In particular, the effective field driving the precessional motion of Template:Math is not restricted to real magnetic fields; it incorporates a wide range of mechanisms including magnetic anisotropy, exchange interaction, and so on.
The various forms of the LLG equation are commonly used in micromagnetics to model the effects of a magnetic field and other magnetic interactions on ferromagnetic materials. It provides a practical way to model the time-domain behavior of magnetic elements. Recent developments generalizes the LLG equation to include the influence of spin-polarized currents in the form of spin-transfer torque.[2]
Landau–Lifshitz equation

In a ferromagnet, the magnitude of the magnetization Template:Math at each spacetime point is approximated by the saturation magnetization Template:Math (although it can be smaller when averaged over a chunk of volume). The LLG equation describes the rotation of the magnetization in response to the effective field Template:Math and accounts for not only a real magnetic field but also internal magnetic interactions such as exchange and anisotropy. An earlier, but equivalent, equation (the Landau–Lifshitz equation) was introduced by Template:Harvtxt:[1] Template:NumBlk where Template:Math is the electron gyromagnetic ratio and Template:Math is a phenomenological damping parameter, often replaced by
where Template:Math is a dimensionless constant called the damping factor. The effective field Template:Math is a combination of the external magnetic field, the demagnetizing field, and various internal magnetic interactions involving quantum mechanical effects, which is typically defined as the functional derivative of the magnetic free energy with respect to the local magnetization Template:Math. To solve this equation, additional conditions for the demagnetizing field must be included to accommodate the geometry of the material.
Landau–Lifshitz–Gilbert equation
In 1955 Gilbert replaced the damping term in the Landau–Lifshitz (LL) equation by one that depends on the time derivative of the magnetization: Template:NumBlk This is the Landau–Lifshitz–Gilbert (LLG) equation, where Template:Math is the damping parameter, which is characteristic of the material. It can be transformed into the Landau–Lifshitz equation:[3] Template:NumBlk where
In this form of the LL equation, the precessional term Template:Math depends on the damping term. This better represents the behavior of real ferromagnets when the damping is large.[4][5]
Landau–Lifshitz–Gilbert–Slonczewski equation
In 1996 John Slonczewski expanded the model to account for the spin-transfer torque, i.e. the torque induced upon the magnetization by spin-polarized current flowing through the ferromagnet. This is commonly written in terms of the unit moment defined by :
where is the dimensionless damping parameter, and are driving torques, and Template:Math is the unit vector along the polarization of the current.[6][7]
Application to magnetic resonance
The response of a magnetic material to an oscillating magnetic field (e.g. the magnetic component of an electromagnetic wave) can be found using the LLG equation. Without loss of generality, the magnetization equilibrium magnetization vector , which is aligned with the static magnetic field can be taken to be in the z-direction. One can then consider a small oscillating magnetic field , which causes a small change in the magnetization . Thus, the total magnetization reads , and the total magnetic field is .
Filling this in to the LLG equation (neglecting the damping terms) gives
- .
The equilibrium magnetization does not vary over time. Additionally, the term is zero because they are parallel. Finally, the equation is considered in leading order, so the term is neglected because it is the product of two small terms. Thus, the equation becomes
The z-component of this equation is zero, meaning that the magnetization does not respond to an magnetic field oscillating in the z-direction. Assuming that oscillates together with , the time derivative turns into . The x- and y-components can then be written in matrix form as
Inverting this matrix gives the permeability tensor
Thus, there is a resonance at the ferromagnetic resonance frequency.[8]
Microscopic origin of the damping
In the Landau–Lifshitz–Gilbert equation the damping is described by the phenomenological, dimensionless parameter Template:Mvar, which does not by itself specify the physical mechanism responsible for the dissipation of energy and angular momentum from the precessing magnetization.[9] In conducting ferromagnets the dominant contribution to the intrinsic damping is generally attributed to spin–orbit coupling, which couples the magnetization to the itinerant electron system and allows energy to be transferred to the electronic degrees of freedom, where it is ultimately dissipated through electron scattering.[10]
Kamberský mechanism (torque-correlation model)
The most widely used microscopic description of intrinsic Gilbert damping is the torque-correlation model (TCM), often referred to as the Kamberský mechanism after the Czech physicist Vladimír Kamberský, who developed the theory in a series of papers between 1970 and 1976.[11][12] In this model the damping is expressed through a correlation function of the spin–orbit torque operator evaluated near the Fermi surface. When the magnetization is tilted slightly from equilibrium, spin–orbit coupling generates a non-equilibrium electron population that relaxes back towards equilibrium over a characteristic scattering time Template:Mvar, dissipating energy in the process.[13]
The torque-correlation model can be regarded as a generalization of Kamberský's earlier breathing Fermi surface model, and it makes fewer approximations; in particular it describes damping over a wide range of temperatures and in the presence of chemical (alloy) disorder.[14] A characteristic feature of the resulting expression is that it separates into two contributions with different dependence on the electronic scattering rate: an intraband (conductivity-like) term that grows as the scattering rate decreases, and an interband (resistivity-like) term that grows as the scattering rate increases. This partitioning has been used to interpret the non-monotonic temperature dependence of the damping observed in ferromagnetic metals such as iron, cobalt and nickel.[10]
Kamberský later re-derived the damping formula using exact linear-response theory, expressing it in a form convenient for numerical estimates in terms of the Fermi golden rule.[15]
Divergence problem and later developments
Calculations based on Kamberský's formula predict that the intraband contribution, which is of third order in the spin–orbit coupling parameter, diverges for a perfect crystal in the limit of zero temperature (vanishing scattering rate).[10] Later analytical work argued that this divergence is an artifact of applying the formula beyond its regime of validity, showing that when the derivation is carried out consistently and restricted to second order in the spin–orbit parameter, no divergent intraband terms appear.[16] In real materials the presence of defects, reflected in a finite residual resistivity, ensures that the damping parameter remains finite even as the temperature approaches zero.[16]
The torque-correlation model has since been implemented in first-principles electronic structure calculations and extended in several directions, including tensorial and non-local formulations of the damping.[17] It has been complemented by an equivalent formulation based on scattering theory, which relates the damping to the scattering matrix of the ferromagnet and can avoid explicit reliance on an external relaxation-time parameter.[18]
References and footnotes
- ↑ 1.0 1.1 Template:Cite book
- ↑ Template:Cite journal
- ↑ Template:Cite book
- ↑ For details of Kelly's non-resonant experiment, and of Gilbert's analysis (which led to Gilbert's modifying the damping term), see Gilbert, T. L. and Kelly, J. M. "Anomalous rotational damping in ferromagnetic sheets", Conf. Magnetism and Magnetic Materials, Pittsburgh, PA, June 14–16, 1955 (New York: American Institute of Electrical Engineers, Oct. 1955, pp. 253–263 Template:Webarchive). Text references to Figures 5 and 6 should have been to Tables 1 and 2. Gilbert could not fit Kelly's experiments with fixed usual gyromagnetic ratio Template:Math and a frequency-dependent Template:Math, but could fit that data for a fixed Gilbert gyromagnetic ratio Template:Math and a frequency-dependent Template:Math. Values of Template:Math as large as 9 were required, indicating very broad absorption, and thus a relatively low-quality sample. Modern samples, when analyzed from resonance absorption, give Template:Math's on the order of 0.05 or less. J. R. Mayfield, in J. Appl. Phys. Supplement to Vol. 30, 256S-257S (1959), at the top left of p.257S, writes "As was first pointed out by J. C. Slonczewski, the observed torque peak can be interpreted in terms of rotational switching effects (abrupt reorientations of M) which must occur when K/M ≤ H ≤ 2K/M." Therefore the interpretation given by Gilbert was not universally accepted.
- ↑ J. Mallinson, "On damped gyromagnetic precession," in IEEE Transactions on Magnetics, vol. 23, no. 4, pp. 2003-2004, July 1987, doi: 10.1109/TMAG.1987.1065181.
- ↑ Template:Cite journal
- ↑ Template:Cite journal
- ↑ Template:Cite book
- ↑ Template:Cite journal
- ↑ 10.0 10.1 10.2 Template:Cite journal
- ↑ Template:Cite journal
- ↑ Template:Cite journal
- ↑ Template:Cite journal
- ↑ Template:Cite journal
- ↑ Template:Cite journal
- ↑ 16.0 16.1 Template:Cite journal
- ↑ Template:Cite journal
- ↑ Template:Cite journal
Further reading
- Template:Cite journal This is only an abstract; the full report is "Armor Research Foundation Project No. A059, Supplementary Report, May 1, 1956", but was never published. A description of the work is given in Template:Cite journal The original abstract reads: T. L. Gilbert, Armour Research Foundation of Illinois Institute of Technology.–The gyromagnetic equation, , for the motion of the magnetization field , in a ferromagnetic material can be derived from a variation principle, as first shown by Doering (W. Doering, Z. Naturforsch. 3a, 374 (1948)). Here is the effective internal field, including the magnetic field and contributions from exchange, anisotropy, and magnetoelastic effects. Using the variational principle, the equations of motion can be recast into a Lagrangian form. This makes possible a consistent derivation of the equations of motion of the magnetization field and other fields to which it may be coupled (e.g., the displacement field of the lattice and the electromagnetic field). It also permits the introduction of viscous damping effects in a consistent manner using the Rayleigh dissipation function. It is shown that viscous damping of the magnetization fields leads to an equation of motion which reduces to the Landau-Lifshitz equation only when the damping is small. It is also shown that this Lagrangian formalism permits the introduction of damping due to disaccomodation in a consistent and very general way.
- Template:Cite journal
- Template:Cite journal
- Template:Cite book
- Template:Cite journal
- Template:Cite journal