Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Electromagnetic breathing dromion-like structures in an anisotropic ferromagnetic medium

 Sathishkumar Perumala,, J. Sivapragasama,b and M. Lakshmananc
a Department of Physics, Sri Paramakalyani College, Alwarkurichi - 627 412, Tenkasi District, Tamil Nadu, India.
b Department of Physics, K.S.R. College of Engineering, Tiruchengode-637 215, Tamil Nadu, India.
c Department of Nonlinear Dynamics, School of Physics, Bharathidasan University, Tiruchirapalli-620 024, Tamil Nadu, India
Corresponding author and E-mail address: perumal__\__sathish@yahoo.co.in, Tel:+91-4288-274213, Fax:+91-91-4288-274757
Abstract

      The influence of Gilbert damping on the propagation of electromagnetic waves (EMWs) in an anisotropic ferromagnetic medium is investigated theoretically. The interaction of the magnetic field component of the electromagnetic wave with the magnetization of a ferromagnetic medium has been studied by solving the associated Maxwell’s equations coupled with a Landau-Lifshitz-Gilbert (LLG) equation. When small perturbations are made on the magnetization of the ferromagnetic medium and magnetic field along the direction of propagation of electromagnetic wave by using the reductive perturbation method, the associated nonlinear dynamics is governed by a time-dependent damped derivative nonlinear Schrödinger (TDDNLS) equation. The Lagrangian density function is constructed by using the variational method to solve the TDDNLS equation to understand the dynamics of the system under consideration. The propagation of EMW in a ferromagnetic medium with inherent Gilbert damping admits very interesting nonlinear dynamical structures. These structures include Gilbert damping-managing symmetrically breathing solitons, localized erupting electromagnetic breathing dromion-like modes of excitations, breathing dromion-like soliton, decaying dromion-like modes and an unexpected creation-annihilation mode of excitations in the form of growing-decaying dromion-like modes.
PACS: 71.70.Gm, 76.50.+g, 75.60.Jk.
Keywords: Breathing dromion-like soliton, Breathing soliton, Nonlinear Schrödinger equation, Landau-Lifshitz-Gilbert equation, the variational method.
 

 

1 Introduction

The magnetization dynamics associated with the propagation of intense electromagnetic (EM) waves in nonlinear ferromagnetic media have attracted much attention from the perspectives of fundamental physics as well as technological applications like magnetic recording, fast data retrieval, and high density data storage devices [1-3]. When an EM wave propagates through a nonlinear ferro, ferri or anti-ferromagnetic medium, the interplay between the magnetic field components of EM wave and the magnetization of magnetic medium results in several interesting phenomena such as harmonic generation, self-focusing, domain wall propagation, EM soliton propagation and breather-like solitary excitations [4-8]. The nonlinear modes of excitations are described by the families of nonlinear dynamical equations that include nonlinear Schrödinger (NLS) equations for nonlinear spin excitations in the form of solitons, derivative nonlinear Schrödinger (DNLS) equation for breather-like solitary excitations, modified Korteweg-de-Vries (mKdV) equations for domain wall-type excitations and Korteweg-de-Vries (KdV) equations for polarisation in the long-wave approximation [9-14]. Further, it has been observed that the plane EM wave and the magnetization of the medium are modulated in the form of an EM soliton when it propagates through a charge-free, isotropic, and anisotropic ferromagnetic medium [15-17]. When free charges are present in the medium, however, the EM soliton slows down and the amplitude diminishes and becomes damped [18]. Moreover, the propagation of EM waves in an antiferromagnetic medium with Dzyaloshinskii-Moriya interaction has been studied theoretically and it was found that the propagation of EM waves is governed by breather-like solitary excitations [19].

The study of the breathing nature of the soliton excitations in the various physical systems has received a lot of attention over the past few years. Breathers are recognized as spatially localized nonlinear excitations in the nonlinear systems, i.e., solitary waves can breathe periodically with virtually undamped long-lived oscillations of amplitude and width. The interesting features of breathing soliton are that it does not damp down and no radiation is emitted [12, 20]. The breathing mode of soliton excitations has been experimentally observed in a quasi one-dimensional antiferromagnetic spin chain with magnetic field [21]. It has been shown that a classical anisotropic Heisenberg discrete spin chain with and without an external magnetic field can have a variety of classes of exact soliton solutions expressed in terms of Jacobian elliptic functions [22]. The stability and explicit analytical solutions of one, two and three spins excitations for the anisotropic Heisenberg spin chain with onsite anisotropy and constant external magnetic field have been studied in [23]. However, further work is needed in this direction to identify such interesting class of soliton solutions in a classical anisotropic Heisenberg spin chain.

Soliton explosions are the most striking and fascinating nonlinear dissipative phenomena studied in fiber laser, Frenkel-Kontorova model with anharmonic interatomic interactions and physical systems with nonlinear damping [24, 25]. Soliton explosions are unstable solutions caused by instabilities and their mechanisms have been studied in generalized Klein-Gordon equations with nonlinear damping and spatio-temporal perturbations [26]. The space-dependent perturbations and nonlinear damping that periodically create instability in the location where the soliton is located at that instant result in a highly non-stationary spatio-temporal state in which the soliton is not allowed to recover its original shape, leading to soliton explosion. In practice, it is realized that such mechanisms can account for some of the events that have lately been reported to happen in heart tissue and a transmission line at a Josephson junction [27, 28].

In this study, we demonstrate the existence of localized electromagnetic breathing dromion-like structures, dromion explosions and decaying breathing dromion-like modes of excitation in an anisotropic ferromagnetic medium with Gilbert damping when subjected to a varying EM wave. In order to put forward the TDDNLS equation for understanding the evolution of the magnetization of the medium, we apply the multiple-scale perturbation method to the system in which almost all explicit parameters appearing in the equations are redefined. The present work is organized as follows. Section 2 introduces the governing dynamical equations. Section 3 employs a reductive perturbation technique that yields the TDDNLS equation. The magnetization components of an anisotropic ferromagnetic medium are constructed by using the variational method with the Lagrangian density function in Section 4. Section 5 provides a summary of the results.

2 Model and dynamics

The magnetization dynamics of a classical continuum anisotropic ferromagnetic medium with Gilbert damping in the presence of EM field is governed by the Landau-Lifshitz-Gilbert equation [29],

𝐌t=γ[𝐌×𝐅𝐞𝐟𝐟]+Λ[𝐅𝐞𝐟𝐟(𝐌.𝐅𝐞𝐟𝐟)𝐌],𝐌2=1,\frac{\partial{\bf{M}}}{\partial t}=\gamma\bigl{[}{\bf{M}}\times{\bf{F}_{eff}}% \bigr{]}+\Lambda\bigl{[}{\bf{F}_{eff}}-({\bf{M}}.{\bf{F}_{eff}}){\bf{M}}\bigr{% ]},~{}~{}~{}~{}~{}{\bf{M}}^{2}=1,divide start_ARG ∂ bold_M end_ARG start_ARG ∂ italic_t end_ARG = italic_γ [ bold_M × bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT ] + roman_Λ [ bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT - ( bold_M . bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT ) bold_M ] , bold_M start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = 1 , (1)

where 𝐌(𝐫,𝐭)𝐌𝐫𝐭\bf{M(r,t)}bold_M ( bold_r , bold_t ) represents the magnetization 𝐌=(Mx,My,Mz)𝐌superscript𝑀𝑥superscript𝑀𝑦superscript𝑀𝑧{\bf{M}}=(M^{x},M^{y},M^{z})bold_M = ( italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT , italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT , italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) of the ferromagnetic medium. The effective field 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT in the presence of a varying magnetic field 𝐇𝐇\bf{H}bold_H as the magnetic field component of the EM wave propagating through the anisotropic ferromagnetic medium along the z-direction is written as 𝐅𝐞𝐟𝐟=J2𝐌2AMz𝐧^+2β𝐇subscript𝐅𝐞𝐟𝐟𝐽superscript2𝐌2𝐴superscript𝑀𝑧^𝐧2𝛽𝐇{\bf{F}_{eff}}=J\nabla^{2}{\bf{M}}-2AM^{z}{\bf{\hat{n}}}+2\beta\bf{H}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT = italic_J ∇ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT bold_M - 2 italic_A italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT over^ start_ARG bold_n end_ARG + 2 italic_β bold_H, where J𝐽Jitalic_J is exchange integral which arises due to spin-spin exchange interaction and 𝐧^=(0,0,1)^𝐧001{\bf{\hat{n}}}=(0,0,1)over^ start_ARG bold_n end_ARG = ( 0 , 0 , 1 ). The anisotropy parameter A𝐴Aitalic_A describes the strength of the crystal field anisotropy along the z-direction as the easy axis of magnetization, and the contribution due to interaction of the magnetization with the external magnetic field is represented by the term proportional to β=γμB𝛽𝛾subscript𝜇𝐵\beta=\gamma\mu_{B}italic_β = italic_γ italic_μ start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT in which γ𝛾\gammaitalic_γ and μBsubscript𝜇𝐵\mu_{B}italic_μ start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT represent the gyromagnetic ratio and the Bohr magneton, respectively. The parameter ΛΛ\Lambdaroman_Λ is designated as γλ𝛾𝜆\gamma\lambdaitalic_γ italic_λ, where λ𝜆\lambdaitalic_λ is the dimensionless Gilbert damping parameter which stands for nonlinear spin relaxation phenomenon. Eq. (1) is equivalent to the Landau-Lifshitz-Gilbert (LLG) equation in which the first term depicts the precession of the magnetization vector about the effective field 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT with angular frequency ω=γ|𝐅𝐞𝐟𝐟|𝜔𝛾subscript𝐅𝐞𝐟𝐟{\bf{\omega}}=-\gamma|{\bf{F}_{eff}}|italic_ω = - italic_γ | bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT |. The second term in Eq. (1) represents the Gilbert damping torque which describes the relaxation of magnetization towards the direction of the effective field 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT for a positive value of λ𝜆\lambdaitalic_λ i.e., it controls the rate at which the magnetization vector relaxes to equilibrium state. For negative value of λ𝜆\lambdaitalic_λ, the magnetization 𝐌𝐌{\bf{M}}bold_M spirals away from the direction of 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT into a direction opposite to 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT thereby gaining energy and corresponds to switching of magnetization. The Gilbert damping plays a key role in the spin dynamics of magnetic systems. It is commonly assumed that the origin of the Gilbert damping is the spin-orbit coupling and two-magnon scattering process [30, 31]. Materials with large spin-orbit coupling exhibit strong interactions between their magnetic moments and the surrounding environment, resulting in high damping of the magnetization dynamics. Therefore, the processes of magnetization relaxation and switching in spintronics devices are profoundly affected by Gilbert damping. In particular, it affects the velocity of domain-wall in current-carrying domain-wall structures where fast propagation of domain-walls is essential for applications in high-speed spintronic devices such as magnetic race track memory [32]. Now, the evolution of magnetization density associated with an anisotropic ferromagnetic medium where the EM wave propagates through it along the z-direction can be redefined by making use of expression for the effective field 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT. The LLG equation demands that the length of the magnetization vector does not change with time (𝐌𝟐=𝟏superscript𝐌21\bf{M}^{2}=1bold_M start_POSTSUPERSCRIPT bold_2 end_POSTSUPERSCRIPT = bold_1). The one-dimensional form of the evolution of magnetization density after substituting the effective field 𝐅𝐞𝐟𝐟subscript𝐅𝐞𝐟𝐟{\bf{F}_{eff}}bold_F start_POSTSUBSCRIPT bold_eff end_POSTSUBSCRIPT in the resultant equation is expressed as

𝐌t=𝐌×[J2𝐌z22AMz𝐧^+2β𝐇+λ[J2𝐌z22AMz𝐧^+2β𝐇\displaystyle\frac{\partial{\bf{M}}}{\partial t}={\bf{M}}\times\Bigl{[}J\frac{% \partial^{2}{\bf{M}}}{\partial z^{2}}-2AM^{z}{\bf{\hat{n}}}+2\beta{\bf{H}}+% \lambda\bigl{[}J\frac{\partial^{2}{\bf{M}}}{\partial z^{2}}-2AM^{z}{\bf{\hat{n% }}}+2\beta{\bf{H}}divide start_ARG ∂ bold_M end_ARG start_ARG ∂ italic_t end_ARG = bold_M × [ italic_J divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT bold_M end_ARG start_ARG ∂ italic_z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - 2 italic_A italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT over^ start_ARG bold_n end_ARG + 2 italic_β bold_H + italic_λ [ italic_J divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT bold_M end_ARG start_ARG ∂ italic_z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - 2 italic_A italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT over^ start_ARG bold_n end_ARG + 2 italic_β bold_H
J(𝐌.2𝐌z2)𝐌+2A(𝐌.Mz𝐧^)𝐌2β(𝐌.𝐇)𝐌]],\displaystyle-J({\bf{M}}.\frac{\partial^{2}{\bf{M}}}{\partial z^{2}}){\bf{M}}+% 2A({\bf{M}}.M^{z}{\bf{\hat{n}}}){\bf{M}}-2\beta({\bf{M}}.{\bf{H}}){\bf{M}}% \bigr{]}\Bigr{]},- italic_J ( bold_M . divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT bold_M end_ARG start_ARG ∂ italic_z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG ) bold_M + 2 italic_A ( bold_M . italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT over^ start_ARG bold_n end_ARG ) bold_M - 2 italic_β ( bold_M . bold_H ) bold_M ] ] , (2)

where z𝑧zitalic_z is now the propagation direction. The magnetization 𝐌𝐌\bf{M}bold_M of the ferromagnetic medium is related to the magnetic induction 𝐁𝐁\bf{B}bold_B and the magnetic field 𝐇𝐇\bf{H}bold_H as 𝐁=μ(𝐇+𝐌)𝐁𝜇𝐇𝐌{\bf{B}}=\mu(\bf{H}+\bf{M})bold_B = italic_μ ( bold_H + bold_M ), where μ𝜇\muitalic_μ is the magnetic permeability of the medium. The variations of the electric field 𝐄𝐄\bf{E}bold_E and magnetic field 𝐇𝐇\bf{H}bold_H components of an EM wave as it propagates through a ferromagnetic medium in the absence of static and moving charges are governed by the Maxwell equations as

𝐄=𝟎,𝐄0\nabla\cdot\bf{E}=0,∇ ⋅ bold_E = bold_0 , (3)
𝐁=𝟎,𝐁0\nabla\cdot\bf{B}=0,∇ ⋅ bold_B = bold_0 , (4)
×𝐄=𝐁𝐭,𝐄𝐁𝐭\nabla\times\bf{E}=-\frac{\partial\bf{B}}{\partial t},∇ × bold_E = - divide start_ARG ∂ bold_B end_ARG start_ARG ∂ bold_t end_ARG , (5)
×𝐇=𝐃𝐭.𝐇𝐃𝐭\nabla\times\bf{H}=\frac{\partial\bf{D}}{\partial t}.∇ × bold_H = divide start_ARG ∂ bold_D end_ARG start_ARG ∂ bold_t end_ARG . (6)

The vector representations of the electric field, magnetic field and magnetic induction are 𝐄=(Ex,Ey,Ez)𝐄superscript𝐸𝑥superscript𝐸𝑦superscript𝐸𝑧{\bf{E}}=(E^{x},E^{y},E^{z})bold_E = ( italic_E start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT , italic_E start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT , italic_E start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ), 𝐇=(Hx,Hy,Hz)𝐇superscript𝐻𝑥superscript𝐻𝑦superscript𝐻𝑧{\bf{H}}=(H^{x},H^{y},H^{z})bold_H = ( italic_H start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT , italic_H start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT , italic_H start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) and 𝐁=(Bx,By,Bz)𝐁superscript𝐵𝑥superscript𝐵𝑦superscript𝐵𝑧{\bf{B}}=(B^{x},B^{y},B^{z})bold_B = ( italic_B start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT , italic_B start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT , italic_B start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) respectively. The vector 𝐃𝐃\bf{D}bold_D represents the electric induction 𝐃=ε𝐄𝐃𝜀𝐄\bf{D}={\varepsilon}\bf{E}bold_D = italic_ε bold_E, where ε𝜀{\varepsilon}italic_ε denotes the dielectric constant of the medium. Using curl on Eq. (6) and the relation that connects magnetization of the medium, finally, we obtain the evolution of the magnetic field component of an EM wave as it propagates through an anisotropic ferromagnetic medium as

2t2(𝐇+𝐌)=v2[2𝐇z22Hzz2𝐧^],superscript2superscript𝑡2𝐇𝐌superscript𝑣2delimited-[]superscript2𝐇superscript𝑧2superscript2superscript𝐻𝑧superscript𝑧2^𝐧\frac{\partial^{2}}{\partial t^{2}}({\bf{H}+\bf{M}})=v^{2}\Bigl{[}\frac{% \partial^{2}\bf{H}}{\partial z^{2}}-\frac{\partial^{2}H^{z}}{\partial z^{2}}{% \bf{\hat{n}}}\Bigr{]},divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_t start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG ( bold_H + bold_M ) = italic_v start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT [ divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT bold_H end_ARG start_ARG ∂ italic_z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_H start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG over^ start_ARG bold_n end_ARG ] , (7)

where v=1με𝑣1𝜇𝜀v={1\over{\sqrt{\mu\varepsilon}}}italic_v = divide start_ARG 1 end_ARG start_ARG square-root start_ARG italic_μ italic_ε end_ARG end_ARG is the phase velocity of the EM wave in a ferromagnetic medium. Eqs. (2) and (7) are in vector forms of partial differential equations along with the constraint on the magnitude of the spin and solving them analytically is a quite challenging task.

3 Perturbation scheme and evolution equation

We attempt to solve the one-dimensional version of the coupled dynamical Eqs. (2) and (7) using the reductive perturbation method developed by Taniuti and Yajima [33] in order to study the nature of propagation of EM waves in an anisotropic charge-free ferromagnetic medium. The spatial and time variables are stretched in the reductive perturbation method as ζ=ϵ(zct)𝜁italic-ϵ𝑧𝑐𝑡\zeta=\epsilon(z-ct)italic_ζ = italic_ϵ ( italic_z - italic_c italic_t ), τ=ϵ2t𝜏superscriptitalic-ϵ2𝑡\tau={\epsilon^{2}}titalic_τ = italic_ϵ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_t which characterize the shape of the pulse propagating at the speed c𝑐citalic_c and the time variable accounts for the evolution of the propagating pulse, where ϵitalic-ϵ\epsilonitalic_ϵ is a small parameter. We rescale J𝐽Jitalic_J as ϵ1Jsuperscriptitalic-ϵ1𝐽\epsilon^{-1}Jitalic_ϵ start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT italic_J, A𝐴Aitalic_A as ϵAitalic-ϵ𝐴{\epsilon}Aitalic_ϵ italic_A and λ𝜆\lambdaitalic_λ as λ𝜆\lambdaitalic_λ on the assumption that the bilinear exchange interaction is stronger than the anisotropic interaction. Additionally, because of the fact that the ferromagnetic medium is anisotropic in nature with an easy axis of magnetization along the z𝑧zitalic_z direction, we expand the components of the magnetization 𝐌𝐌\bf{M}bold_M of the medium and the magnetic induction 𝐁𝐁\bf{B}bold_B in terms of the small parameter ϵitalic-ϵ\epsilonitalic_ϵ in a non-uniform way about the uniform values M0subscript𝑀0M_{0}italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT and B0subscript𝐵0B_{0}italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT respectively by treating the small parameter ϵitalic-ϵ\epsilonitalic_ϵ as the perturbation parameter [7] (For more details, refer Appendix-I). By substituting the non-uniform expansions of 𝐌𝐌{\bf{M}}bold_M and 𝐁𝐁{\bf{B}}bold_B in the component form of Eq. (2) and Eq. (7) and after collecting the coefficients of ϵitalic-ϵ\epsilonitalic_ϵ at the order ϵ0superscriptitalic-ϵ0\epsilon^{0}italic_ϵ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT, we obtain a system of equations for 𝐁𝟏subscript𝐁1\bf{B}_{1}bold_B start_POSTSUBSCRIPT bold_1 end_POSTSUBSCRIPT and 𝐌𝟏subscript𝐌1\bf{M}_{1}bold_M start_POSTSUBSCRIPT bold_1 end_POSTSUBSCRIPT. On solving the resultant equations, we obtain B1x=kM1xsuperscriptsubscript𝐵1𝑥𝑘superscriptsubscript𝑀1𝑥B_{1}^{x}=kM_{1}^{x}italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT = italic_k italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT, B1y=kM1ysuperscriptsubscript𝐵1𝑦𝑘superscriptsubscript𝑀1𝑦B_{1}^{y}=kM_{1}^{y}italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT = italic_k italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT and k=B0M0=1ε(v2c2)𝑘subscript𝐵0subscript𝑀01𝜀superscript𝑣2superscript𝑐2k={B_{0}\over{M_{0}}}={1\over{\varepsilon(v^{2}-c^{2})}}italic_k = divide start_ARG italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG start_ARG italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG = divide start_ARG 1 end_ARG start_ARG italic_ε ( italic_v start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT - italic_c start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) end_ARG. In the zeroth order of the perturbation, the steady-state solutions of the xlimit-from𝑥x-italic_x - and ylimit-from𝑦y-italic_y - components of the LLG equation (2) and the Maxwell equation (7) are identically satisfied. We finally obtain the following equations at order ϵ1superscriptitalic-ϵ1\epsilon^{1}italic_ϵ start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT, by using the results at O(ϵ0superscriptitalic-ϵ0\epsilon^{0}italic_ϵ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT), as B1z=0superscriptsubscript𝐵1𝑧0B_{1}^{z}=0italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = 0 and

cM1xζ=JM02M1yζ22AM0M1y+2βμ[B1zM1yB1yM1z+B0M2yM0B2y]𝑐superscriptsubscript𝑀1𝑥𝜁𝐽subscript𝑀0superscript2superscriptsubscript𝑀1𝑦superscript𝜁22𝐴subscript𝑀0superscriptsubscript𝑀1𝑦2𝛽𝜇delimited-[]superscriptsubscript𝐵1𝑧superscriptsubscript𝑀1𝑦superscriptsubscript𝐵1𝑦superscriptsubscript𝑀1𝑧subscript𝐵0superscriptsubscript𝑀2𝑦subscript𝑀0superscriptsubscript𝐵2𝑦\displaystyle-c{\partial{M_{1}^{x}}\over{\partial\zeta}}=-JM_{0}{\partial^{2}{% M_{1}^{y}}\over{\partial\zeta^{2}}}-2AM_{0}M_{1}^{y}+{2\beta\over{\mu}}\bigl{[% }B_{1}^{z}M_{1}^{y}-B_{1}^{y}M_{1}^{z}+B_{0}M_{2}^{y}-M_{0}B_{2}^{y}\bigr{]}- italic_c divide start_ARG ∂ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ end_ARG = - italic_J italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG [ italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT - italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT - italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ]
+λ[J2M1xζ2+2AM02M1x+2βμB2x2βμ{(B1xM1x+B1yM1y+M0B1z\displaystyle+\lambda\Bigl{[}J{\partial^{2}{M_{1}^{x}}\over{\partial\zeta^{2}}% }+2AM_{0}^{2}M_{1}^{x}+{2\beta\over{\mu}}B_{2}^{x}-{2\beta\over{\mu}}\Bigl{\{}% \bigl{(}B_{1}^{x}M_{1}^{x}+B_{1}^{y}M_{1}^{y}+M_{0}B_{1}^{z}+ italic_λ [ italic_J divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG { ( italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT
+B0M1z)M1x+B0M0M2x}],\displaystyle+B_{0}M_{1}^{z}\bigr{)}M_{1}^{x}+B_{0}M_{0}M_{2}^{x}\Bigr{\}}% \Bigr{]},+ italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT } ] , (8)
cM1yζ=JM02M1xζ2+2AM0M1x+2βμ[B1xM1zB1zM1x+M0B2xB0M2x]𝑐superscriptsubscript𝑀1𝑦𝜁𝐽subscript𝑀0superscript2superscriptsubscript𝑀1𝑥superscript𝜁22𝐴subscript𝑀0superscriptsubscript𝑀1𝑥2𝛽𝜇delimited-[]superscriptsubscript𝐵1𝑥superscriptsubscript𝑀1𝑧superscriptsubscript𝐵1𝑧superscriptsubscript𝑀1𝑥subscript𝑀0superscriptsubscript𝐵2𝑥subscript𝐵0superscriptsubscript𝑀2𝑥\displaystyle-c{\partial{M_{1}^{y}}\over{\partial\zeta}}=JM_{0}{\partial^{2}{M% _{1}^{x}}\over{\partial\zeta^{2}}}+2AM_{0}M_{1}^{x}+{2\beta\over{\mu}}\bigl{[}% B_{1}^{x}M_{1}^{z}-B_{1}^{z}M_{1}^{x}+M_{0}B_{2}^{x}-B_{0}M_{2}^{x}\bigr{]}- italic_c divide start_ARG ∂ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ end_ARG = italic_J italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG [ italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT - italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ]
+λ[J2M1yζ2+2AM02M1y+2βμB2y2βμ{(B1xM1x+B1yM1y+M0B1z\displaystyle+\lambda\Bigl{[}J{\partial^{2}{M_{1}^{y}}\over{\partial\zeta^{2}}% }+2AM_{0}^{2}M_{1}^{y}+{2\beta\over{\mu}}B_{2}^{y}-{2\beta\over{\mu}}\Bigl{\{}% \bigl{(}B_{1}^{x}M_{1}^{x}+B_{1}^{y}M_{1}^{y}+M_{0}B_{1}^{z}+ italic_λ [ italic_J divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT - divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG { ( italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT
+B0M1z)M1y+B0M0M2y}],\displaystyle+B_{0}M_{1}^{z}\bigr{)}M_{1}^{y}+B_{0}M_{0}M_{2}^{y}\Bigr{\}}% \Bigr{]},+ italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT } ] , (9)

and also

ζ[B2xkM2x]=2ckεB1xτ,𝜁delimited-[]superscriptsubscript𝐵2𝑥𝑘superscriptsubscript𝑀2𝑥2𝑐𝑘𝜀superscriptsubscript𝐵1𝑥𝜏{\partial\over{\partial\zeta}}\bigl{[}B_{2}^{x}-kM_{2}^{x}\bigr{]}=-2ck% \varepsilon{\partial{B_{1}^{x}}\over{\partial\tau}},divide start_ARG ∂ end_ARG start_ARG ∂ italic_ζ end_ARG [ italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - italic_k italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ] = - 2 italic_c italic_k italic_ε divide start_ARG ∂ italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_τ end_ARG , (10)
ζ[B2ykM2y]=2ckεB1yτ.𝜁delimited-[]superscriptsubscript𝐵2𝑦𝑘superscriptsubscript𝑀2𝑦2𝑐𝑘𝜀superscriptsubscript𝐵1𝑦𝜏{\partial\over{\partial\zeta}}\bigl{[}B_{2}^{y}-kM_{2}^{y}\bigr{]}=-2ck% \varepsilon{\partial{B_{1}^{y}}\over{\partial\tau}}.divide start_ARG ∂ end_ARG start_ARG ∂ italic_ζ end_ARG [ italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT - italic_k italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ] = - 2 italic_c italic_k italic_ε divide start_ARG ∂ italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_τ end_ARG . (11)

Making use of Eqs. (10) and (11) in Eqs. (8) and (9), we get

cM1xζ=4cεβk2M0μζM1yτ𝑑ζJM02M1yζ22kβμM1yM1z2AM0M1y𝑐superscriptsubscript𝑀1𝑥𝜁4𝑐𝜀𝛽superscript𝑘2subscript𝑀0𝜇superscriptsubscript𝜁superscriptsubscript𝑀1𝑦𝜏differential-d𝜁𝐽subscript𝑀0superscript2superscriptsubscript𝑀1𝑦superscript𝜁22𝑘𝛽𝜇superscriptsubscript𝑀1𝑦superscriptsubscript𝑀1𝑧2𝐴subscript𝑀0superscriptsubscript𝑀1𝑦\displaystyle-c{\partial{M_{1}^{x}}\over{\partial\zeta}}={4c\varepsilon{\beta}% k^{2}M_{0}\over{\mu}}\int_{-\infty}^{\zeta}{{\partial}M_{1}^{y}\over{\partial% \tau}}d\zeta-JM_{0}{\partial^{2}M_{1}^{y}\over{\partial\zeta^{2}}}-{2k\beta% \over{\mu}}M_{1}^{y}M_{1}^{z}-2AM_{0}M_{1}^{y}- italic_c divide start_ARG ∂ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ end_ARG = divide start_ARG 4 italic_c italic_ε italic_β italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG start_ARG italic_μ end_ARG ∫ start_POSTSUBSCRIPT - ∞ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_ζ end_POSTSUPERSCRIPT divide start_ARG ∂ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_τ end_ARG italic_d italic_ζ - italic_J italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG - divide start_ARG 2 italic_k italic_β end_ARG start_ARG italic_μ end_ARG italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT - 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT
+λ[J2M1xζ2+2AM02M1x+2βμB2x2βμ{(B1xM1x+B1yM1y+B0M1z)M1x\displaystyle+\lambda\Bigl{[}J{\partial^{2}{M_{1}^{x}}\over{\partial\zeta^{2}}% }+2AM_{0}^{2}M_{1}^{x}+{2\beta\over{\mu}}B_{2}^{x}-{2\beta\over{\mu}}\Bigl{\{}% \bigl{(}B_{1}^{x}M_{1}^{x}+B_{1}^{y}M_{1}^{y}+B_{0}M_{1}^{z}\bigr{)}M_{1}^{x}+ italic_λ [ italic_J divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG { ( italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT
+B0M0M2x}]\displaystyle+B_{0}M_{0}M_{2}^{x}\Bigr{\}}\Bigr{]}+ italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT } ] (12)

and

cM1yζ=4cεβk2M0μζM1xτ𝑑ζ+JM02M1xζ2+2kβμM1xM1z+2AM0M1x𝑐superscriptsubscript𝑀1𝑦𝜁4𝑐𝜀𝛽superscript𝑘2subscript𝑀0𝜇superscriptsubscript𝜁superscriptsubscript𝑀1𝑥𝜏differential-d𝜁𝐽subscript𝑀0superscript2superscriptsubscript𝑀1𝑥superscript𝜁22𝑘𝛽𝜇superscriptsubscript𝑀1𝑥superscriptsubscript𝑀1𝑧2𝐴subscript𝑀0superscriptsubscript𝑀1𝑥\displaystyle-c{\partial{M_{1}^{y}}\over{\partial\zeta}}=-{4c\varepsilon{\beta% }k^{2}M_{0}\over{\mu}}\int_{-\infty}^{\zeta}{{\partial}M_{1}^{x}\over{\partial% \tau}}d\zeta+JM_{0}{\partial^{2}M_{1}^{x}\over{\partial\zeta^{2}}}+{2k\beta% \over{\mu}}M_{1}^{x}M_{1}^{z}+2AM_{0}M_{1}^{x}- italic_c divide start_ARG ∂ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ end_ARG = - divide start_ARG 4 italic_c italic_ε italic_β italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG start_ARG italic_μ end_ARG ∫ start_POSTSUBSCRIPT - ∞ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_ζ end_POSTSUPERSCRIPT divide start_ARG ∂ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_τ end_ARG italic_d italic_ζ + italic_J italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + divide start_ARG 2 italic_k italic_β end_ARG start_ARG italic_μ end_ARG italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT
+λ[J2M1yζ2+2AM02M1y+2βμB2y2βμ{(B1xM1x+B1yM1y+B0M1z)M1y\displaystyle+\lambda\Bigl{[}J{\partial^{2}{M_{1}^{y}}\over{\partial\zeta^{2}}% }+2AM_{0}^{2}M_{1}^{y}+{2\beta\over{\mu}}B_{2}^{y}-{2\beta\over{\mu}}\Bigl{\{}% \bigl{(}B_{1}^{x}M_{1}^{x}+B_{1}^{y}M_{1}^{y}+B_{0}M_{1}^{z}\bigr{)}M_{1}^{y}+ italic_λ [ italic_J divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_ζ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + 2 italic_A italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT - divide start_ARG 2 italic_β end_ARG start_ARG italic_μ end_ARG { ( italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT
+B0M0M2y}].\displaystyle+B_{0}M_{0}M_{2}^{y}\Bigr{\}}\Bigr{]}.+ italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT } ] . (13)

To account for the conservation of length of the magnetization M1z=12M0[(M1x)2+(M1y)2]superscriptsubscript𝑀1𝑧12subscript𝑀0delimited-[]superscriptsuperscriptsubscript𝑀1𝑥2superscriptsuperscriptsubscript𝑀1𝑦2M_{1}^{z}=-{1\over{2M_{0}}}\bigl{[}(M_{1}^{x})^{2}+(M_{1}^{y})^{2}\bigr{]}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = - divide start_ARG 1 end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG [ ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ], we define a new complex field ψ=M1xiM1y𝜓superscriptsubscript𝑀1𝑥𝑖superscriptsubscript𝑀1𝑦\psi=M_{1}^{x}-iM_{1}^{y}italic_ψ = italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - italic_i italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT and |ψ|2=2M0M1zsuperscript𝜓22subscript𝑀0superscriptsubscript𝑀1𝑧|\psi|^{2}=-2M_{0}M_{1}^{z}| italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = - 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT (Refer Appendix-I for more details). The amplitude of the complex field ψ𝜓\psiitalic_ψ or |ψ|2superscript𝜓2|\psi|^{2}| italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT corresponds to small deviations from M0subscript𝑀0M_{0}italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT (constant, see Eqs. (A2) and (A9)) and phase invariance of ψ𝜓\psiitalic_ψ corresponds to the isotropy in the MxMysuperscript𝑀𝑥superscript𝑀𝑦M^{x}-M^{y}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT plane (see Eq. (A8)). By taking a single differentiation of Eqs. (12) and (13) and using the complex field ψ𝜓\psiitalic_ψ with the appropriate transformations of Z=ζ+Aμ2cβεk2τ𝑍𝜁𝐴𝜇2𝑐𝛽𝜀superscript𝑘2𝜏Z=\zeta+{A\mu\over{2c{\beta}\varepsilon k^{2}}}\tauitalic_Z = italic_ζ + divide start_ARG italic_A italic_μ end_ARG start_ARG 2 italic_c italic_β italic_ε italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG italic_τ and τ=τ𝜏𝜏\tau=\tauitalic_τ = italic_τ, we obtain the following nonlinear evolution equation after extensive algebra as

iηψτ+2ψZ2+iσ(|ψ|2ψ)Ziδ3ψZ3+λ1ψτ+λ23ψZ3+λ3(|ψ|2ψ)Z=0,𝑖𝜂𝜓𝜏superscript2𝜓superscript𝑍2𝑖𝜎superscript𝜓2𝜓𝑍𝑖𝛿superscript3𝜓superscript𝑍3subscript𝜆1𝜓𝜏subscript𝜆2superscript3𝜓superscript𝑍3subscript𝜆3superscript𝜓2𝜓𝑍0i\eta{\partial\psi\over{{\partial}\tau}}+{\partial^{2}\psi\over{{\partial}Z^{2% }}}+i\sigma{\partial\bigl{(}|\psi|^{2}\psi\bigr{)}\over{{\partial}Z}}-i\delta{% \partial^{3}\psi\over{\partial Z^{3}}}+\lambda_{1}{\partial\psi\over{{\partial% }\tau}}+\lambda_{2}{\partial^{3}{\psi}\over{\partial Z^{3}}}+\lambda_{3}{% \partial\bigl{(}|\psi|^{2}\psi\bigr{)}\over{{\partial}Z}}=0,italic_i italic_η divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG + divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + italic_i italic_σ divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ ) end_ARG start_ARG ∂ italic_Z end_ARG - italic_i italic_δ divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG + italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG + italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG + italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ ) end_ARG start_ARG ∂ italic_Z end_ARG = 0 , (14)

where η=4εβk2M0μ𝜂4𝜀𝛽superscript𝑘2subscript𝑀0𝜇\eta={4\varepsilon{\beta}k^{2}\over{M_{0}\mu}}italic_η = divide start_ARG 4 italic_ε italic_β italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_μ end_ARG, σ=kβM0cμ𝜎𝑘𝛽subscript𝑀0𝑐𝜇\sigma={k{\beta}\over{M_{0}c\mu}}italic_σ = divide start_ARG italic_k italic_β end_ARG start_ARG italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_c italic_μ end_ARG, δ=JM0c𝛿𝐽subscript𝑀0𝑐\delta={J\over{M_{0}c}}italic_δ = divide start_ARG italic_J end_ARG start_ARG italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_c end_ARG, λ1=4εβk2μλsubscript𝜆14𝜀𝛽superscript𝑘2𝜇𝜆\lambda_{1}=-{4\varepsilon{\beta}k^{2}\over{\mu}}\lambdaitalic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = - divide start_ARG 4 italic_ε italic_β italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG italic_μ end_ARG italic_λ, λ2=Jcλsubscript𝜆2𝐽𝑐𝜆\lambda_{2}={J\over{c}}\lambdaitalic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = divide start_ARG italic_J end_ARG start_ARG italic_c end_ARG italic_λ and λ3=kβcμλsubscript𝜆3𝑘𝛽𝑐𝜇𝜆\lambda_{3}=-{k{\beta}\over{c\mu}}\lambdaitalic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT = - divide start_ARG italic_k italic_β end_ARG start_ARG italic_c italic_μ end_ARG italic_λ. After redefinition of Eq. (14), it becomes

ϱψτ+2ψZ2+φ3ψZ3+ς(|ψ|2ψ)Z=0,italic-ϱ𝜓𝜏superscript2𝜓superscript𝑍2𝜑superscript3𝜓superscript𝑍3𝜍superscript𝜓2𝜓𝑍0\varrho{\partial\psi\over{{\partial}\tau}}+{\partial^{2}\psi\over{{\partial}Z^% {2}}}+\varphi{\partial^{3}\psi\over{\partial Z^{3}}}+\varsigma{\partial\bigl{(% }|\psi|^{2}\psi\bigr{)}\over{{\partial}Z}}=0,italic_ϱ divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG + divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + italic_φ divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG + italic_ς divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ ) end_ARG start_ARG ∂ italic_Z end_ARG = 0 , (15)

where ϱ=iη+λ1italic-ϱ𝑖𝜂subscript𝜆1\varrho=i\eta+\lambda_{1}italic_ϱ = italic_i italic_η + italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT, φ=iδ+λ2𝜑𝑖𝛿subscript𝜆2\varphi=-i\delta+\lambda_{2}italic_φ = - italic_i italic_δ + italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT and ς=iσ+λ3𝜍𝑖𝜎subscript𝜆3\varsigma=i\sigma+\lambda_{3}italic_ς = italic_i italic_σ + italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT. Eq.(15) is known as the time-dependent damped derivative nonlinear Schrödinger (TDDNLS) equation. The real part of the first term in Eq. (15) represents the time-dependent damping factor which arisesin the anisotropic ferromagnetic medium when it is subjected to intense EM field in the presence of intrinsic Gilbert damping. This time-dependent damping factor highly influences the nature of nonlinear spin excitations. The second and third terms in Eq. (15) are the dispersion terms that cause the solution ψ𝜓\psiitalic_ψ to broaden as it evolves. The third term originates essentially from the exchange interaction between the spins, wherein the real and imaginary components arise due to the presence and absence of Gilbert damping, respectively. The fourth term represents electromagnetically induced higher-order derivative cubic nonlinearity in the material medium. The compensation between the dispersion terms and the electromagnetically induced higher-order derivative cubic nonlinearity admits very interesting nonlinear dynamical structures. The nonlinear evolution Eq. (15) with δ=λ1=λ2=λ3=0𝛿subscript𝜆1subscript𝜆2subscript𝜆30\delta=\lambda_{1}=\lambda_{2}=\lambda_{3}=0italic_δ = italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT = 0 is the completely integrable Kaup-Newell derivative nonlinear Schrödinger (DNLS) equation admitting N-soliton solutions [34, 35]. It has been identified in optical fiber that the propagation of optical pulses with asymmetric self-phase modulation and self-steepening is governed by DNLS equation [36]. The nonlinear evolution of Alfv́en waves that admit completely integrable DNLS equation was first shown by Rogister [37] in space plasma. It is found that the magnetization of an anisotropic ferromagnetic medium is excited in the form of an oscillating electromagnetic soliton as a breathing mode of excitations when it is subjected to a intense EM wave in the absence of Gilbert damping λ=0𝜆0\lambda=0italic_λ = 0 [12]. Therefore, the investigation of nonlinear spin excitations in ferromagnetic media is not only interesting and significant from a mathematical point of view, but also has important physical applications.

4 Variational methods: Exact soliton solutions

The existence of the time-derivative factor in the form of ψτ𝜓𝜏\partial\psi\over{{\partial}\tau}divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG which is proportional to λ1subscript𝜆1\lambda_{1}italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT in the TDDNLS equation (15) that becomes a disruption to the system causes the parameters associated with the underlying soliton such as amplitude, position, velocity and phase to be time-dependent [38]. The variational method can find the time-dependent parameters associated with the systems under consideration. In order to obtain them through the variational method, the Lagrangian density function \mathcal{L}caligraphic_L is constructed by using Eq. (15) as follows:

=iη[ψψτψψτ]+[ψ2ψZ2+ψ2ψZ2]+iσ[ψ(|ψ|2ψ)Zψ(|ψ|2ψ)Z]𝑖𝜂delimited-[]superscript𝜓𝜓𝜏𝜓superscript𝜓𝜏delimited-[]superscript𝜓superscript2𝜓superscript𝑍2𝜓superscript2superscript𝜓superscript𝑍2𝑖𝜎delimited-[]superscript𝜓superscript𝜓2𝜓𝑍𝜓superscript𝜓2superscript𝜓𝑍\displaystyle\mathcal{L}=i\eta\Bigl{[}\psi^{*}{\partial\psi\over{{\partial}% \tau}}-\psi{\partial\psi^{*}\over{{\partial}\tau}}\Bigr{]}+\Bigl{[}\psi^{*}{% \partial^{2}\psi\over{{\partial}Z^{2}}}+\psi{\partial^{2}\psi^{*}\over{{% \partial}Z^{2}}}\Bigr{]}+i\sigma\Bigl{[}\psi^{*}{\partial\bigl{(}|\psi|^{2}% \psi\bigr{)}\over{{\partial}Z}}-\psi{\partial\bigl{(}|\psi|^{2}\psi^{*}\bigr{)% }\over{{\partial}Z}}\Bigr{]}caligraphic_L = italic_i italic_η [ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG - italic_ψ divide start_ARG ∂ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_τ end_ARG ] + [ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG + italic_ψ divide start_ARG ∂ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG ] + italic_i italic_σ [ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ ) end_ARG start_ARG ∂ italic_Z end_ARG - italic_ψ divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ) end_ARG start_ARG ∂ italic_Z end_ARG ]
iδ[ψ3ψZ3ψ3ψZ3]+λ1[(ψψ)τ]+λ2[ψ3ψZ3+ψ3ψZ3]𝑖𝛿delimited-[]superscript𝜓superscript3𝜓superscript𝑍3𝜓superscript3superscript𝜓superscript𝑍3subscript𝜆1delimited-[]𝜓superscript𝜓𝜏subscript𝜆2delimited-[]superscript𝜓superscript3𝜓superscript𝑍3𝜓superscript3superscript𝜓superscript𝑍3\displaystyle-i\delta\Bigl{[}\psi^{*}{\partial^{3}\psi\over{\partial Z^{3}}}-% \psi{\partial^{3}\psi^{*}\over{\partial Z^{3}}}\Bigr{]}+\lambda_{1}\Bigl{[}{% \partial\bigl{(}\psi\psi^{*}\bigr{)}\over{{\partial}\tau}}\Bigr{]}+\lambda_{2}% \Bigl{[}\psi^{*}{\partial^{3}{\psi}\over{\partial Z^{3}}}+\psi{\partial^{3}{% \psi^{*}}\over{\partial Z^{3}}}\Bigr{]}- italic_i italic_δ [ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG - italic_ψ divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG ] + italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT [ divide start_ARG ∂ ( italic_ψ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ) end_ARG start_ARG ∂ italic_τ end_ARG ] + italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT [ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG + italic_ψ divide start_ARG ∂ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT end_ARG start_ARG ∂ italic_Z start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT end_ARG ]
+λ3[ψ(|ψ|2ψ)Z+ψ(|ψ|2ψ)Z]=0.subscript𝜆3delimited-[]superscript𝜓superscript𝜓2𝜓𝑍𝜓superscript𝜓2superscript𝜓𝑍0\displaystyle+\lambda_{3}\Bigl{[}\psi^{*}{\partial\bigl{(}|\psi|^{2}\psi\bigr{% )}\over{{\partial}Z}}+\psi{\partial\bigl{(}|\psi|^{2}\psi^{*}\bigr{)}\over{{% \partial}Z}}\Bigr{]}=0.+ italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT [ italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ ) end_ARG start_ARG ∂ italic_Z end_ARG + italic_ψ divide start_ARG ∂ ( | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ) end_ARG start_ARG ∂ italic_Z end_ARG ] = 0 . (16)

The Lagrangian function L𝐿\it{L}italic_L is obtained by substituting Eq. (16) into the following equation

L=𝑑𝑍.𝐿superscriptsubscript𝑑𝑍\it{L}=\int_{-\infty}^{\infty}\mathcal{L}dZ.italic_L = ∫ start_POSTSUBSCRIPT - ∞ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ∞ end_POSTSUPERSCRIPT caligraphic_L italic_dZ . (17)

4.1 Gilbert damping managing symmetrically breathing solitons

Let us assume that the TDDNLS equation (15) admits a special type of soliton solution that begins to evolve as a function of time with the time-dependent arbitrary functions β(τ)𝛽𝜏\beta(\tau)italic_β ( italic_τ ), α(τ)𝛼𝜏\alpha(\tau)italic_α ( italic_τ ) and θ(τ)𝜃𝜏\theta(\tau)italic_θ ( italic_τ ) of the form

ψ(Z,τ)=β(τ)sech[sZ+α(τ)]exp[i[θ(τ)Z+r]].𝜓𝑍𝜏𝛽𝜏sech𝑠𝑍𝛼𝜏𝑖delimited-[]𝜃𝜏𝑍𝑟\psi(Z,\tau)=\beta(\tau)\operatorname{sech}\Bigl{[}sZ+\alpha(\tau)\Bigr{]}\exp% \Bigl{[}i\bigl{[}\theta(\tau)Z+r\bigr{]}\Bigr{]}.italic_ψ ( italic_Z , italic_τ ) = italic_β ( italic_τ ) roman_sech [ italic_s italic_Z + italic_α ( italic_τ ) ] roman_exp [ italic_i [ italic_θ ( italic_τ ) italic_Z + italic_r ] ] . (18)

After integrating Eq. (17) with respect to Z𝑍Zitalic_Z by making use of Eq. (18), we obtained the following dynamical equation,

dβdτ=13λ1[s2β+3δs2θβ+2σθβ3+3βθ2+3δβθ3].𝑑𝛽𝑑𝜏13subscript𝜆1delimited-[]superscript𝑠2𝛽3𝛿superscript𝑠2𝜃𝛽2𝜎𝜃superscript𝛽33𝛽superscript𝜃23𝛿𝛽superscript𝜃3{d\beta\over{d\tau}}={1\over{3\lambda_{1}}}\Bigl{[}s^{2}\beta+3{\delta}s^{2}% \theta\beta+2\sigma\theta\beta^{3}+3\beta\theta^{2}+3{\delta}\beta\theta^{3}% \Bigr{]}.divide start_ARG italic_d italic_β end_ARG start_ARG italic_d italic_τ end_ARG = divide start_ARG 1 end_ARG start_ARG 3 italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_ARG [ italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_β + 3 italic_δ italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_θ italic_β + 2 italic_σ italic_θ italic_β start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT + 3 italic_β italic_θ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + 3 italic_δ italic_β italic_θ start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ] . (19)

Next, we multiply both sides of Eq. (17) by tanh(sZ+α(τ))𝑠𝑍𝛼𝜏\tanh(sZ+\alpha(\tau))roman_tanh ( italic_s italic_Z + italic_α ( italic_τ ) ) and integrate it once to obtain

5sλ1dαdτ15ηdθdτ=7λ2s46λ3s2β2+15λ2s2θ2.5𝑠subscript𝜆1𝑑𝛼𝑑𝜏15𝜂𝑑𝜃𝑑𝜏7subscript𝜆2superscript𝑠46subscript𝜆3superscript𝑠2superscript𝛽215subscript𝜆2superscript𝑠2superscript𝜃25s\lambda_{1}{d\alpha\over{d\tau}}-15\eta{d\theta\over{d\tau}}=7\lambda_{2}s^{% 4}-6\lambda_{3}s^{2}\beta^{2}+15\lambda_{2}s^{2}\theta^{2}.5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT divide start_ARG italic_d italic_α end_ARG start_ARG italic_d italic_τ end_ARG - 15 italic_η divide start_ARG italic_d italic_θ end_ARG start_ARG italic_d italic_τ end_ARG = 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT - 6 italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_β start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + 15 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_θ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT . (20)

After rearranging the above Eq. (20), we get the following equation in which both LHS and RHS are equal if they evolve independently at a constant rate m𝑚mitalic_m, and hence

15ηdθdτ+15λ2s2θ2=5sλ1dαdτ7λ2s4+6λ3s2β2=m.15𝜂𝑑𝜃𝑑𝜏15subscript𝜆2superscript𝑠2superscript𝜃25𝑠subscript𝜆1𝑑𝛼𝑑𝜏7subscript𝜆2superscript𝑠46subscript𝜆3superscript𝑠2superscript𝛽2𝑚15\eta{d\theta\over{d\tau}}+15\lambda_{2}s^{2}\theta^{2}=5s\lambda_{1}{d\alpha% \over{d\tau}}-7\lambda_{2}s^{4}+6\lambda_{3}s^{2}\beta^{2}=m.15 italic_η divide start_ARG italic_d italic_θ end_ARG start_ARG italic_d italic_τ end_ARG + 15 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_θ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = 5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT divide start_ARG italic_d italic_α end_ARG start_ARG italic_d italic_τ end_ARG - 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT + 6 italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_β start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = italic_m . (21)

The LHS and RHS of the aforementioned equation now become

15ηdθdτ+15λ2s2θ2=m15𝜂𝑑𝜃𝑑𝜏15subscript𝜆2superscript𝑠2superscript𝜃2𝑚15\eta{d\theta\over{d\tau}}+15\lambda_{2}s^{2}\theta^{2}=m15 italic_η divide start_ARG italic_d italic_θ end_ARG start_ARG italic_d italic_τ end_ARG + 15 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_θ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = italic_m (22)

and

5sλ1dαdτ7λ2s4+6λ3s2β2=m.5𝑠subscript𝜆1𝑑𝛼𝑑𝜏7subscript𝜆2superscript𝑠46subscript𝜆3superscript𝑠2superscript𝛽2𝑚5s\lambda_{1}{d\alpha\over{d\tau}}-7\lambda_{2}s^{4}+6\lambda_{3}s^{2}\beta^{2% }=m.5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT divide start_ARG italic_d italic_α end_ARG start_ARG italic_d italic_τ end_ARG - 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT + 6 italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_β start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = italic_m . (23)

If the time-dependent function β(τ)𝛽𝜏\beta(\tau)italic_β ( italic_τ ) remains constant (β(τ)=β𝛽𝜏𝛽\beta(\tau)=\betaitalic_β ( italic_τ ) = italic_β), the following solutions can be obtained by solving Eqs. (22 &\&& 23),

α(τ)=15sλ1[9δλ3s4σ+7λ2s4+m]τ+c2,𝛼𝜏15𝑠subscript𝜆1delimited-[]9𝛿subscript𝜆3superscript𝑠4𝜎7subscript𝜆2superscript𝑠4𝑚𝜏subscript𝑐2\displaystyle\alpha(\tau)={1\over{5s\lambda_{1}}}\Bigl{[}{9\delta\lambda_{3}s^% {4}\over{\sigma}}+7\lambda_{2}s^{4}+m\Bigr{]}\tau+c_{2},italic_α ( italic_τ ) = divide start_ARG 1 end_ARG start_ARG 5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_ARG [ divide start_ARG 9 italic_δ italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT end_ARG start_ARG italic_σ end_ARG + 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT + italic_m ] italic_τ + italic_c start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , (24a)
θ(τ)=1sm15λ2tanh[sηmλ215(τ+c1)],𝜃𝜏1𝑠𝑚15subscript𝜆2𝑠𝜂𝑚subscript𝜆215𝜏subscript𝑐1\displaystyle\theta(\tau)={1\over{s}}\sqrt{{m\over{15\lambda_{2}}}}\tanh\Biggl% {[}{s\over{\eta}}\sqrt{{m\lambda_{2}\over{15}}}\bigl{(}\tau+c_{1}\bigr{)}% \Biggr{]},italic_θ ( italic_τ ) = divide start_ARG 1 end_ARG start_ARG italic_s end_ARG square-root start_ARG divide start_ARG italic_m end_ARG start_ARG 15 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_ARG end_ARG roman_tanh [ divide start_ARG italic_s end_ARG start_ARG italic_η end_ARG square-root start_ARG divide start_ARG italic_m italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT end_ARG start_ARG 15 end_ARG end_ARG ( italic_τ + italic_c start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT ) ] , (24b)
β(τ)=β=±s26δσ.𝛽𝜏𝛽plus-or-minus𝑠26𝛿𝜎\displaystyle\beta(\tau)=\beta=\pm{s\over{2}}\sqrt{{-6\delta\over{\sigma}}}.italic_β ( italic_τ ) = italic_β = ± divide start_ARG italic_s end_ARG start_ARG 2 end_ARG square-root start_ARG divide start_ARG - 6 italic_δ end_ARG start_ARG italic_σ end_ARG end_ARG . (24c)

where c1subscript𝑐1c_{1}italic_c start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and c2subscript𝑐2c_{2}italic_c start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT are constants of integration. Using the complex field ψ=M1xiM1y𝜓superscriptsubscript𝑀1𝑥𝑖superscriptsubscript𝑀1𝑦\psi=M_{1}^{x}-iM_{1}^{y}italic_ψ = italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - italic_i italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT, |ψ|2=2M0M1zsuperscript𝜓22subscript𝑀0superscriptsubscript𝑀1𝑧|\psi|^{2}=-2M_{0}M_{1}^{z}| italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = - 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT and Eqs. (24), the components M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT, M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT and M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT of the magnetization of the anisotropic ferromagnetic material exposed by intense EM field can be constructed as

M1x=±s26δσsech[sZ+15sλ1[9δλ3s4σ+7λ2s4+m]τ+c2]cos[θ(τ)Z+r],superscriptsubscript𝑀1𝑥plus-or-minus𝑠26𝛿𝜎sech𝑠𝑍15𝑠subscript𝜆1delimited-[]9𝛿subscript𝜆3superscript𝑠4𝜎7subscript𝜆2superscript𝑠4𝑚𝜏subscript𝑐2𝜃𝜏𝑍𝑟\displaystyle M_{1}^{x}=\pm{s\over{2}}\sqrt{{-6\delta\over{\sigma}}}% \operatorname{sech}\Biggl{[}sZ+{1\over{5s\lambda_{1}}}\Bigl{[}{9\delta\lambda_% {3}s^{4}\over{\sigma}}+7\lambda_{2}s^{4}+m\Bigr{]}\tau+c_{2}\Biggr{]}\cos\Bigl% {[}\theta(\tau)Z+r\Bigr{]},italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT = ± divide start_ARG italic_s end_ARG start_ARG 2 end_ARG square-root start_ARG divide start_ARG - 6 italic_δ end_ARG start_ARG italic_σ end_ARG end_ARG roman_sech [ italic_s italic_Z + divide start_ARG 1 end_ARG start_ARG 5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_ARG [ divide start_ARG 9 italic_δ italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT end_ARG start_ARG italic_σ end_ARG + 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT + italic_m ] italic_τ + italic_c start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT ] roman_cos [ italic_θ ( italic_τ ) italic_Z + italic_r ] , (25a)
M1y=s26δσsech[sZ+15sλ1[9δλ3s4σ+7λ2s4+m]τ+c2]sin[θ(τ)Z+r],superscriptsubscript𝑀1𝑦minus-or-plus𝑠26𝛿𝜎sech𝑠𝑍15𝑠subscript𝜆1delimited-[]9𝛿subscript𝜆3superscript𝑠4𝜎7subscript𝜆2superscript𝑠4𝑚𝜏subscript𝑐2𝜃𝜏𝑍𝑟\displaystyle M_{1}^{y}=\mp{s\over{2}}\sqrt{{-6\delta\over{\sigma}}}% \operatorname{sech}\Biggl{[}sZ+{1\over{5s\lambda_{1}}}\Bigl{[}{9\delta\lambda_% {3}s^{4}\over{\sigma}}+7\lambda_{2}s^{4}+m\Bigr{]}\tau+c_{2}\Biggr{]}\sin\Bigl% {[}\theta(\tau)Z+r\Bigr{]},italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT = ∓ divide start_ARG italic_s end_ARG start_ARG 2 end_ARG square-root start_ARG divide start_ARG - 6 italic_δ end_ARG start_ARG italic_σ end_ARG end_ARG roman_sech [ italic_s italic_Z + divide start_ARG 1 end_ARG start_ARG 5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_ARG [ divide start_ARG 9 italic_δ italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT end_ARG start_ARG italic_σ end_ARG + 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT + italic_m ] italic_τ + italic_c start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT ] roman_sin [ italic_θ ( italic_τ ) italic_Z + italic_r ] , (25b)
M1z=u(Z,τ)22M0=3s2δ4σM0sech2[sZ+15sλ1[9δλ3s4σ+7λ2s4+m]τ+c2].superscriptsubscript𝑀1𝑧𝑢superscript𝑍𝜏22subscript𝑀03superscript𝑠2𝛿4𝜎subscript𝑀0superscriptsech2𝑠𝑍15𝑠subscript𝜆1delimited-[]9𝛿subscript𝜆3superscript𝑠4𝜎7subscript𝜆2superscript𝑠4𝑚𝜏subscript𝑐2\displaystyle M_{1}^{z}=-{u(Z,\tau)^{2}\over{2M_{0}}}={3s^{2}\delta\over{4{% \sigma}M_{0}}}\operatorname{sech}^{2}\Biggl{[}sZ+{1\over{5s\lambda_{1}}}\Bigl{% [}{9\delta\lambda_{3}s^{4}\over{\sigma}}+7\lambda_{2}s^{4}+m\Bigr{]}\tau+c_{2}% \Biggr{]}.italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = - divide start_ARG italic_u ( italic_Z , italic_τ ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG = divide start_ARG 3 italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_δ end_ARG start_ARG 4 italic_σ italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG roman_sech start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT [ italic_s italic_Z + divide start_ARG 1 end_ARG start_ARG 5 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT end_ARG [ divide start_ARG 9 italic_δ italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT end_ARG start_ARG italic_σ end_ARG + 7 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT + italic_m ] italic_τ + italic_c start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT ] . (25c)

When λ=0𝜆0\lambda=0italic_λ = 0, the propagation of intensely varying EM wave along the direction of easy axis of an anisotropic ferromagnetic medium (M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT) is governed by a highly localized magnetized state in the form of soliton which makes some instabilities in the transmission character of the M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components [12]. Therefore, higher-order local instabilities arise in the form of derivative cubic nonlinearity in the medium which may result from the periodically varying refractive index of the medium. As a result, the M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization of the medium exhibit oscillating solitary modes in the form of breathers with constant amplitude due to the strong counter-balance between the higher-order dispersion and the electromagnetically induced derivative cubic nonlinearity. In order to realize the influence of Gilbert damping with third-order dispersion and derivative cubic nonlinearity when intense varying EM wave propagates through an anisotropic ferromagnetic medium, we have plotted Eqs. (25) for the x, y and z-components of the magnetization M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT, M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT with parametric choice of η=0.5𝜂0.5\eta=-0.5italic_η = - 0.5, σ=0.3𝜎0.3\sigma=-0.3italic_σ = - 0.3, δ=0.3𝛿0.3\delta=0.3italic_δ = 0.3, M0=0.153subscript𝑀00.153M_{0}=0.153italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.153, r=0.2, s=-0.1, C1=0.2subscript𝐶10.2C_{1}=0.2italic_C start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = 0.2, C2=0.01subscript𝐶20.01C_{2}=0.01italic_C start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = 0.01, m=0.04𝑚0.04m=0.04italic_m = 0.04 for various values of λ𝜆\lambdaitalic_λ.

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 1: Evolution of symmetrically breathing solitons in (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT, and soliton in (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT for η=0.5𝜂0.5\eta=-0.5italic_η = - 0.5, σ=0.3𝜎0.3\sigma=-0.3italic_σ = - 0.3, δ=0.3𝛿0.3\delta=0.3italic_δ = 0.3, M0=0.153subscript𝑀00.153M_{0}=0.153italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.153, r=0.2, s=-0.1, C1=0.2subscript𝐶10.2C_{1}=0.2italic_C start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = 0.2, C2=0.01subscript𝐶20.01C_{2}=0.01italic_C start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = 0.01, m = 0.04 and λ=0.03𝜆0.03\lambda=0.03italic_λ = 0.03.

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 2: Evolution of symmetrically breathing solitons in (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT, and soliton in (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT for η=0.5𝜂0.5\eta=-0.5italic_η = - 0.5, σ=0.3𝜎0.3\sigma=-0.3italic_σ = - 0.3, δ=0.3𝛿0.3\delta=0.3italic_δ = 0.3, M0=0.153subscript𝑀00.153M_{0}=0.153italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.153, r=0.2, s=-0.1, C1=0.2subscript𝐶10.2C_{1}=0.2italic_C start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = 0.2, C2=0.01subscript𝐶20.01C_{2}=0.01italic_C start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = 0.01, m = 0.04 and λ=0.05𝜆0.05\lambda=0.05italic_λ = 0.05.

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 3: Evolution of symmetrically breathing solitons in (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT, and soliton in (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT for η=0.5𝜂0.5\eta=-0.5italic_η = - 0.5, σ=0.3𝜎0.3\sigma=-0.3italic_σ = - 0.3, δ=0.3𝛿0.3\delta=0.3italic_δ = 0.3, M0=0.153subscript𝑀00.153M_{0}=0.153italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.153, r=0.2, s=-0.1, C1=0.2subscript𝐶10.2C_{1}=0.2italic_C start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT = 0.2, C2=0.01subscript𝐶20.01C_{2}=0.01italic_C start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT = 0.01, m = 0.04 and λ=0.1𝜆0.1\lambda=0.1italic_λ = 0.1.

When an anisotropic ferromagnetic medium is exposed to a strong EM field with λ=0.03𝜆0.03\lambda=0.03italic_λ = 0.03, the z-component of magnetization M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT experiences nonlinear spin excitations in the form of soliton, as illustrated in Fig. 1c. This soliton spin excitations in M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT supports Gilbert damping managing symmetrically oscillating solitary modes in the form of breathing soliton. It has been found that the x-component of magnetization M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT of the ferromagnetic medium experiences nonlinear spin excitations in the form of symmetrically breathing soliton in the presence of intense EM field and Gilbert damping as seen in Fig. 1a. That is, in the initial stage of breathing soliton, the width of the central peak of breathing soliton is maximum. As time goes on, the Gilbert damping causes the width of the breathing soliton to slightly decrease while maintaining its amplitude. This breathing soliton absorbs the electromagnetic energy from the intense EM field during the propagation in order to manage the influence of inherent Gilbert damping present in the system which results in lossless amplitude of breathing soliton. Also the y-component of magnetization M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT of the ferromagnetic medium exhibits the same behaviors without the central maxima peak as depicted in Fig. 1b. This is due to the fact that the counterbalance between the electromagnetically induced derivative cubic nonlinearity and time-dependent damping factor ψτ𝜓𝜏{\partial\psi\over{{\partial}\tau}}divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG in the anisotropic ferromagnetic medium having intrinsic Gilbert damping leads to symmetrically breathing solitons with varying width. On the other hand, it has been found that, in the absence of Gilbert damping, electromagnetically induced derivative cubic nonlinearity in an anisotropic ferromagnetic medium admits breathing soliton with constant amplitude and width due to intense EM field [12]. However, an isotropic Heisenberg ferromagnetic spin chain with Gilbert damping in the absence intense EM field is governed by a damped nonlinear Schrödinger equation which admits decaying solitary wave solution for the energy density and magnetization density [39]. When natural Gilbert damping λ𝜆\lambdaitalic_λ is 0.05, the z-component of magnetization shows the same behavior as described previously for λ=0.03𝜆0.03\lambda=0.03italic_λ = 0.03 with small change in phase shift. Meanwhile, the x and y-components of magnetization exhibit low breathing frequency of symmetrically breathing solitons with a small change in phase shift as shown in Figs. 2. As the value of Gilbert damping is increased to λ=0.1𝜆0.1\lambda=0.1italic_λ = 0.1, the x, y and z components of magnetization undergo nonlinear spin excitations with small changes in phase shifts as well as breathing frequencies as depicted in Figs. 3. From the observations, it is inferred that when the value of Gilbert damping is increased in the presence of intense EM field, both the x and y components of magnetization of the anisotropic ferromagnetic medium show Gilbert damping managing symmetrically breathing solitons. Thus, the Gilbert damping managing symmetrically breathing soliton could open new opportunities for robust information processing technologies.

4.2 Breathing dromions-like spin excitations

Next, we assume that the solution with the time-dependent arbitrary function α(τ)𝛼𝜏\alpha(\tau)italic_α ( italic_τ ) as

ψ(Z,τ)=α(τ)sech[sZ+α(τ)]exp[i[α(τ)Z+r]],𝜓𝑍𝜏𝛼𝜏sech𝑠𝑍𝛼𝜏𝑖delimited-[]𝛼𝜏𝑍𝑟\psi(Z,\tau)=\alpha(\tau)\operatorname{sech}\Bigl{[}sZ+\alpha(\tau)\Bigr{]}% \exp\Bigl{[}i\bigl{[}\alpha(\tau)Z+r\bigr{]}\Bigr{]},italic_ψ ( italic_Z , italic_τ ) = italic_α ( italic_τ ) roman_sech [ italic_s italic_Z + italic_α ( italic_τ ) ] roman_exp [ italic_i [ italic_α ( italic_τ ) italic_Z + italic_r ] ] , (26)

where, r𝑟ritalic_r and s𝑠sitalic_s represent arbitrary constants. Note that the form (26) is a special case of (18). Allowing the time-dependent parameter α(τ)𝛼𝜏\alpha(\tau)italic_α ( italic_τ ) to evolve as a function of time is necessary to understand the nature of spin excitations associated with the TDDNLS equation. The following evolution equation can be obtained by using Eq. (26) and multiplying tanh(sZ+α(τ))𝑠𝑍𝛼𝜏\tanh(sZ+\alpha(\tau))roman_tanh ( italic_s italic_Z + italic_α ( italic_τ ) ) on both sides of Eq. (17), and then integrating it with regard to Z𝑍Zitalic_Z as

15ηdαdτ10sλ1dαdτ+14s4λ212s2λ3α2+30s2λ2α2=0.15𝜂𝑑𝛼𝑑𝜏10𝑠subscript𝜆1𝑑𝛼𝑑𝜏14superscript𝑠4subscript𝜆212superscript𝑠2subscript𝜆3superscript𝛼230superscript𝑠2subscript𝜆2superscript𝛼2015\eta{d\alpha\over{d\tau}}-10s\lambda_{1}{d\alpha\over{d\tau}}+14s^{4}\lambda% _{2}-12s^{2}\lambda_{3}\alpha^{2}+30s^{2}\lambda_{2}\alpha^{2}=0.15 italic_η divide start_ARG italic_d italic_α end_ARG start_ARG italic_d italic_τ end_ARG - 10 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT divide start_ARG italic_d italic_α end_ARG start_ARG italic_d italic_τ end_ARG + 14 italic_s start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT - 12 italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT italic_α start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + 30 italic_s start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_α start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = 0 . (27)

By solving the nonlinear ordinary differential Eq. (27), the following solution can be obtained:

α(τ)=Aptanh[Ω(τ+C)],𝛼𝜏subscript𝐴𝑝Ω𝜏𝐶\alpha(\tau)=A_{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]},italic_α ( italic_τ ) = italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] , (28)

where Ap=s(42λ2λ3105λ22)(15λ26λ3)A_{p}={s\sqrt{(42\lambda_{2}\lambda_{3}-105\lambda_{2}^{2}})\over{(15\lambda_{% 2}-6\lambda{3})}}italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT = divide start_ARG italic_s square-root start_ARG ( 42 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT - 105 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG ) end_ARG start_ARG ( 15 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT - 6 italic_λ 3 ) end_ARG, Ω=25s3(42λ2λ3105λ22)(3η2sλ1)\Omega={2\over{5}}s^{3}{\sqrt{(42\lambda_{2}\lambda_{3}-105\lambda_{2}^{2}})% \over{(3\eta-2s\lambda_{1})}}roman_Ω = divide start_ARG 2 end_ARG start_ARG 5 end_ARG italic_s start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT divide start_ARG square-root start_ARG ( 42 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT italic_λ start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT - 105 italic_λ start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG ) end_ARG start_ARG ( 3 italic_η - 2 italic_s italic_λ start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT ) end_ARG and C𝐶Citalic_C represents integration constant. The evolution of magnetization components of the anisotropic ferromagnetic medium when subjected to an EM field can be constructed using the solution Eq. (28) in the soliton solution (26) through the complex field ψ𝜓\psiitalic_ψ as

M1x=Aptanh[Ω(τ+C)]sech[sZ+Aptanh[Ω(τ+C)]]cos[Aptanh[Ω(τ+C)]Z+r],superscriptsubscript𝑀1𝑥subscript𝐴𝑝Ω𝜏𝐶sech𝑠𝑍subscript𝐴𝑝Ω𝜏𝐶subscript𝐴𝑝Ω𝜏𝐶𝑍𝑟\displaystyle M_{1}^{x}=A_{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}\operatorname{% sech}\Bigl{[}sZ+A_{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}\Bigr{]}\cos\Bigl{[}A_% {p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}Z+r\Bigr{]},italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT = italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] roman_sech [ italic_s italic_Z + italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] ] roman_cos [ italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] italic_Z + italic_r ] , (29a)
M1y=Aptanh[Ω(τ+C)]sech[sZ+Aptanh[Ω(τ+C)]]sin[Aptanh[Ω(τ+C)]Z+r],superscriptsubscript𝑀1𝑦subscript𝐴𝑝Ω𝜏𝐶sech𝑠𝑍subscript𝐴𝑝Ω𝜏𝐶subscript𝐴𝑝Ω𝜏𝐶𝑍𝑟\displaystyle M_{1}^{y}=-A_{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}\operatorname% {sech}\Bigl{[}sZ+A_{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}\Bigr{]}\sin\Bigl{[}A% _{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}Z+r\Bigr{]},italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT = - italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] roman_sech [ italic_s italic_Z + italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] ] roman_sin [ italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] italic_Z + italic_r ] , (29b)
M1z=u(Z,τ)22M0=Ap22M0[tanh2[Ω(τ+C)]sech2[sZ+Aptanh[Ω(τ+C)]]].superscriptsubscript𝑀1𝑧𝑢superscript𝑍𝜏22subscript𝑀0superscriptsubscript𝐴𝑝22subscript𝑀0delimited-[]superscript2Ω𝜏𝐶superscriptsech2𝑠𝑍subscript𝐴𝑝Ω𝜏𝐶\displaystyle M_{1}^{z}=-{u(Z,\tau)^{2}\over{2M_{0}}}=-{A_{p}^{2}\over{2M_{0}}% }\Biggl{[}\tanh^{2}\bigl{[}\Omega(\tau+C)\bigr{]}\operatorname{sech}^{2}\Bigl{% [}sZ+A_{p}\tanh\bigl{[}\Omega(\tau+C)\bigr{]}\Bigr{]}\Biggr{]}.italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = - divide start_ARG italic_u ( italic_Z , italic_τ ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG = - divide start_ARG italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG [ roman_tanh start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT [ roman_Ω ( italic_τ + italic_C ) ] roman_sech start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT [ italic_s italic_Z + italic_A start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT roman_tanh [ roman_Ω ( italic_τ + italic_C ) ] ] ] . (29c)

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 4: Evolution of localized erupting electromagnetic (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT breathing dromion-like modes of excitations, and (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT breathing dromion-like soliton for η=0.01𝜂0.01\eta=0.01italic_η = 0.01, σ=0.2𝜎0.2\sigma=0.2italic_σ = 0.2, δ=0.31𝛿0.31\delta=0.31italic_δ = 0.31, M0=0.1subscript𝑀00.1M_{0}=0.1italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.1, r=0.1, s=0.13 and λ=0.05𝜆0.05\lambda=0.05italic_λ = 0.05.

The x, y and z-components of the magnetization M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT, M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT have been plotted using Eqs. (29) to understand the nature of nonlinear spin excitations with a parametric choice of η=0.01𝜂0.01\eta=0.01italic_η = 0.01, σ=0.2𝜎0.2\sigma=0.2italic_σ = 0.2, δ=0.31𝛿0.31\delta=0.31italic_δ = 0.31, M0=0.1subscript𝑀00.1M_{0}=0.1italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.1, r=0.1𝑟0.1r=0.1italic_r = 0.1 and s=0.13𝑠0.13s=0.13italic_s = 0.13 for the different values of λ𝜆\lambdaitalic_λ. The propagation of varying EM wave through an anisotropic ferromagnetic medium in the presence of Gilbert damping is governed by TDDNLS equation in which the time-dependent damping factor ψτ𝜓𝜏{\partial\psi\over{{\partial}\tau}}divide start_ARG ∂ italic_ψ end_ARG start_ARG ∂ italic_τ end_ARG disrupts the periodically varying refractive index of the medium. More interestingly, it is found that when λ=0.05𝜆0.05\lambda=0.05italic_λ = 0.05, spatially localized and temporally periodic dromion-like electromagnetic soliton is observed in the M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT component of the anisotropic ferromagnetic system as shown in Fig. 4c. Generally, the dromions decay exponentially and are localized in all the spatial directions which is a special type of higher dimensional soliton solutions. Here, the M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization experiences oscillating electromagnetic dromion-like modes of excitations with erupting profiles as shown in Figs. 4a &\&& 4b. The evolution of erupting soliton starts from a localized stationary solution which has a perfect dromion-like shape. After a while its profiles become covered with small ripples due to internal instability which seem to move downwards and very soon the pulse is covered with this seemingly fluctuating structure. As the ripples grow in size, the oscillating electromagnetic dromion splits into fragments, analogous to a mountain after a powerful volcanic eruption or an earthquake. This fluctuating structure, but well-localized, repeats itself exactly in successive periods with constant amplitude. As a result of the nonlinear Gilbert damping effect in the presence of EM wave interaction with spins, soliton eruptions or explosions may occur during the transition from one stable spin configuration to another spin configuration.

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 5: Evolution of localized erupting electromagnetic (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT breathing dromion-like modes of excitations, and (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT breathing dromion-like soliton with varying amplitudes for η=0.01𝜂0.01\eta=0.01italic_η = 0.01, σ=0.2𝜎0.2\sigma=0.2italic_σ = 0.2, δ=0.31𝛿0.31\delta=0.31italic_δ = 0.31, M0=0.1subscript𝑀00.1M_{0}=0.1italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.1, r=0.1, s=0.13 and λ=0.1𝜆0.1\lambda=0.1italic_λ = 0.1.

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 6: Evolution of localized erupting electromagnetic (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT breathing dromion-like modes of excitations, and (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT breathing dromion-like soliton with decaying amplitudes for η=0.01𝜂0.01\eta=0.01italic_η = 0.01, σ=0.2𝜎0.2\sigma=0.2italic_σ = 0.2, δ=0.31𝛿0.31\delta=0.31italic_δ = 0.31, M0=0.1subscript𝑀00.1M_{0}=0.1italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.1, r=0.1, s=0.13 and λ=0.3𝜆0.3\lambda=0.3italic_λ = 0.3.

(a)Refer to caption (b)Refer to caption (c)Refer to caption

Figure 7: Evolution of localized erupting electromagnetic (a) M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and (b) M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT breathing dromion-like modes of excitations, and (c) M1zsuperscriptsubscript𝑀1𝑧M_{1}^{z}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT breathing dromion-like modes of excitations with growing-decaying amplitudes for η=0.01𝜂0.01\eta=0.01italic_η = 0.01, σ=0.2𝜎0.2\sigma=0.2italic_σ = 0.2, δ=0.31𝛿0.31\delta=0.31italic_δ = 0.31, M0=0.1subscript𝑀00.1M_{0}=0.1italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 0.1, r=0.1, s=0.13 and λ=0.5𝜆0.5\lambda=0.5italic_λ = 0.5.

When λ=0.1𝜆0.1\lambda=0.1italic_λ = 0.1 with intense EM wave, the amplitude of the oscillating electromagnetic dromions is suppressed in the M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT component and become localized solitary modes with varying amplitude, while the M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization appear as erupting dromion-like modes with varying amplitude as shown in Figs. 5. When λ=0.3𝜆0.3\lambda=0.3italic_λ = 0.3, the amplitude of the localized solitary modes of the M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization decreases exponentially as the time goes. However, the M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization exhibit erupting solitary excitations with decaying amplitude as shown in Figs. 6. From the figures, it is inferred that when the strength of the Gilbert damping is high in the presence of EM wave, the precessional motion of the spin vector experiences strong nonlinear force along the z-direction in the anisotropic ferromagnetic medium which causes instability in the x-y plane. This strong nonlinear force may be developed physically from the imperfections, impurity atoms, or defects in the regular lattice sites of the anisotropic ferromagnetic medium, which leads to the instability in the precessional motion of the spin vector. A fluctuating or nonlinear precessional motion of the spin vector due to the instability will be relaxed along the z-direction resulting in the decaying amplitude of dromion-like modes of excitations with erupting profiles in the M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization.

When the Gilbert damping is increased to λ=0.5𝜆0.5\lambda=0.5italic_λ = 0.5 in the presence of an EM wave, the amplitude of the nonlinear excitations in the M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization progressively grows to a maximum value and then subsequently diminishes to a minimum value. Here, the energy of the nonlinear spin excitations increases gradually to create dromion-like structure until it reaches a maximum level and it experiences intermittency at that point and then it begins to dissipate its energy as decaying or annihilating dromions. It is clearly seen in the M1zsubscriptsuperscript𝑀𝑧1M^{z}_{1}italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization as in Figs .7. The M1xsubscriptsuperscript𝑀𝑥1M^{x}_{1}italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and M1ysubscriptsuperscript𝑀𝑦1M^{y}_{1}italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT components of magnetization also exhibit the same behavior with erupting profiles. To preserve their originality, the dromion-like modes of excitation restore themselves after each eruption. Therefore, when an anisotropic ferromagnetic medium is subjected to a intense EM wave in the presence of a strong dissipation in the form of Gilbert damping, an unexpected event of creation and annihilation of nonlinear spin excitations in the form of growing and decaying dromion-like mode is observed in it which may also offer potentially attractive platforms for robust information processing. We also wish to finally state the following limitation of our study. Even though our study brings out the existence of a very interesting localized dromion-like structure, we have not established the stability of these structures. We hope to pursue such studies in the near future.

5 Conclusions

The propagation of EM waves in an anisotropic ferromagnetic medium under the influence of Gilbert damping has been studied theoretically. The interaction of the magnetic field component of the intense EM wave with the magnetization of a ferromagnetic medium has been studied by mapping the associated Maxwell’s equations coupled with a LLG equation. When the magnetization of the ferromagnetic medium and magnetic field component of the EM wave are perturbed in a non-uniform way by using the reductive perturbation method, the associated nonlinear spin excitations are governed by a time-dependent damped derivative nonlinear Schrödinger equation. The Lagrangian density function has been constructed by using the variational method to understand the nature of nonlinear spin excitations under the influence of the time-dependent damping on the system under consideration. It has been demonstrated for the solution (18) that Gilbert damping managing symmetrically breathing solitons is admitted by the compensation between the electromagnetically induced derivative cubic nonlinearity and the time-dependent damping factor in an anisotropic ferromagnetic medium. In particular, for the solution (26), it has been shown that the z-component of magnetization of the medium is governed by spatially localized and temporally periodic electromagnetic breathing dromion-like soliton, while the x- and y-components of the medium are excited in the form of erupting electromagnetic breathing dromion-like modes of excitations when the strength of Gilbert damping is low. When the strength of Gilbert damping is high, it is found that the propagation of EMW in a ferromagnetic medium is governed by decaying breathing dromion-like modes of excitations and an unexpected creation-annihilation modes of excitations in the form of growing-decaying dromion-like modes. Therefore, it is theoretically understood from the present study that the propagation of EM wave in an anisotropic ferromagnetic medium with Gilbert damping admits very interesting nonlinear dynamical structures and phenomenon which may have potentially attractive platforms for robust information processing.

Acknowledgment

The author M. Lakshmanan wishes to thank the Science and Engineering Research Board, Department of Science and Technology, Government of India for the award of a National Science Chair position under Grant No. NSC/2020/00029.

Appendix-I

The non-uniform expansion of the components of the magnetization 𝐌𝐌{\bf{M}}bold_M of the medium and the magnetic induction 𝐁𝐁{\bf{B}}bold_B in terms of the small parameter ϵitalic-ϵ{\epsilon}italic_ϵ as the perturbation parameter about the uniform values M0subscript𝑀0M_{0}italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT and B0subscript𝐵0B_{0}italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT respectively as

MΓ=ϵ[M1Γ+ϵM2Γ+],superscript𝑀Γitalic-ϵdelimited-[]superscriptsubscript𝑀1Γitalic-ϵsuperscriptsubscript𝑀2ΓM^{\Gamma}=\sqrt{\epsilon}\Bigl{[}M_{1}^{\Gamma}+{\epsilon}M_{2}^{\Gamma}+...% \Bigr{]},italic_M start_POSTSUPERSCRIPT roman_Γ end_POSTSUPERSCRIPT = square-root start_ARG italic_ϵ end_ARG [ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_Γ end_POSTSUPERSCRIPT + italic_ϵ italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_Γ end_POSTSUPERSCRIPT + … ] , (A1)𝐴1( italic_A 1 )
Mz=M0+ϵM1z+ϵ2M2z+,superscript𝑀𝑧subscript𝑀0italic-ϵsuperscriptsubscript𝑀1𝑧superscriptitalic-ϵ2superscriptsubscript𝑀2𝑧M^{z}=M_{0}+{\epsilon}M_{1}^{z}+{\epsilon}^{2}M_{2}^{z}+...,italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT + italic_ϵ italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + italic_ϵ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_M start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + … , (A2)𝐴2( italic_A 2 )

and

BΓ=ϵ[B1Γ+ϵB2Γ+],superscript𝐵Γitalic-ϵdelimited-[]superscriptsubscript𝐵1Γitalic-ϵsuperscriptsubscript𝐵2ΓB^{\Gamma}=\sqrt{\epsilon}\Bigl{[}B_{1}^{\Gamma}+{\epsilon}B_{2}^{\Gamma}+...% \Bigr{]},italic_B start_POSTSUPERSCRIPT roman_Γ end_POSTSUPERSCRIPT = square-root start_ARG italic_ϵ end_ARG [ italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_Γ end_POSTSUPERSCRIPT + italic_ϵ italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT roman_Γ end_POSTSUPERSCRIPT + … ] , (A3)𝐴3( italic_A 3 )
Bz=B0+ϵB1z+ϵ2B2z+,superscript𝐵𝑧subscript𝐵0italic-ϵsuperscriptsubscript𝐵1𝑧superscriptitalic-ϵ2superscriptsubscript𝐵2𝑧B^{z}=B_{0}+{\epsilon}B_{1}^{z}+{\epsilon}^{2}B_{2}^{z}+...,italic_B start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = italic_B start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT + italic_ϵ italic_B start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + italic_ϵ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_B start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + … , (A4)𝐴4( italic_A 4 )

where Γ=x,yΓ𝑥𝑦\Gamma=x,yroman_Γ = italic_x , italic_y. To account for the conservation of length of the magnetization, consider the magnetization as

𝐌=iMx+jMy+kMz,𝐌𝑖superscript𝑀𝑥𝑗superscript𝑀𝑦𝑘superscript𝑀𝑧{\bf{M}}=\vec{i}M^{x}+\vec{j}M^{y}+\vec{k}M^{z},bold_M = over→ start_ARG italic_i end_ARG italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT + over→ start_ARG italic_j end_ARG italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT + over→ start_ARG italic_k end_ARG italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT , (A5)𝐴5( italic_A 5 )
𝐌𝟐=(Mx)2+(My)2+(Mz)2,superscript𝐌2superscriptsuperscript𝑀𝑥2superscriptsuperscript𝑀𝑦2superscriptsuperscript𝑀𝑧2{\bf{M^{2}}}=(M^{x})^{2}+(M^{y})^{2}+(M^{z})^{2},bold_M start_POSTSUPERSCRIPT bold_2 end_POSTSUPERSCRIPT = ( italic_M start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT , (A6)𝐴6( italic_A 6 )

where i𝑖\vec{i}over→ start_ARG italic_i end_ARG, j𝑗\vec{j}over→ start_ARG italic_j end_ARG and k𝑘\vec{k}over→ start_ARG italic_k end_ARG are unit vectors. After substituting the non-uniform expansions Eqs. (A1 &\&& A2) in Eq. (A6), we get

𝐌𝟐=M02+ϵ[2M0M1z+(M1x)2+(M1y)2]+superscript𝐌2superscriptsubscript𝑀02italic-ϵdelimited-[]2subscript𝑀0superscriptsubscript𝑀1𝑧superscriptsuperscriptsubscript𝑀1𝑥2superscriptsuperscriptsubscript𝑀1𝑦2{\bf{M^{2}}}={M_{0}}^{2}+{\epsilon}\Bigl{[}2M_{0}M_{1}^{z}+(M_{1}^{x})^{2}+(M_% {1}^{y})^{2}\Bigr{]}+...bold_M start_POSTSUPERSCRIPT bold_2 end_POSTSUPERSCRIPT = italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + italic_ϵ [ 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ] + … (A7)𝐴7( italic_A 7 )

Collecting the coefficients of ϵitalic-ϵ\epsilonitalic_ϵ at different orders of ϵitalic-ϵ\epsilonitalic_ϵ since 𝐌𝟐=𝟏superscript𝐌21\bf{M}^{2}=1bold_M start_POSTSUPERSCRIPT bold_2 end_POSTSUPERSCRIPT = bold_1,
ϵ0superscriptitalic-ϵ0\epsilon^{0}italic_ϵ start_POSTSUPERSCRIPT 0 end_POSTSUPERSCRIPT    :  M02=1superscriptsubscript𝑀021{M_{0}}^{2}=1italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = 1
and
ϵ1superscriptitalic-ϵ1\epsilon^{1}italic_ϵ start_POSTSUPERSCRIPT 1 end_POSTSUPERSCRIPT    :  2M0M1z+(M1x)2+(M1y)2=02subscript𝑀0superscriptsubscript𝑀1𝑧superscriptsuperscriptsubscript𝑀1𝑥2superscriptsuperscriptsubscript𝑀1𝑦202M_{0}M_{1}^{z}+(M_{1}^{x})^{2}+(M_{1}^{y})^{2}=02 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT = 0.
On solving the above equation, we obtain

M1z=12M0[(M1x)2+(M1y)2].superscriptsubscript𝑀1𝑧12subscript𝑀0delimited-[]superscriptsuperscriptsubscript𝑀1𝑥2superscriptsuperscriptsubscript𝑀1𝑦2M_{1}^{z}=-{1\over{2M_{0}}}\Bigl{[}(M_{1}^{x})^{2}+(M_{1}^{y})^{2}\Bigr{]}.italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = - divide start_ARG 1 end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG [ ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ] . (A8)𝐴8( italic_A 8 )

By making use of the expression ψ=M1xiM1y𝜓superscriptsubscript𝑀1𝑥𝑖superscriptsubscript𝑀1𝑦\psi=M_{1}^{x}-iM_{1}^{y}italic_ψ = italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT - italic_i italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT, we get

M1z=12M0[(M1x)2+(M1y)2]=12M0|ψ|2.superscriptsubscript𝑀1𝑧12subscript𝑀0delimited-[]superscriptsuperscriptsubscript𝑀1𝑥2superscriptsuperscriptsubscript𝑀1𝑦212subscript𝑀0superscript𝜓2M_{1}^{z}=-{1\over{2M_{0}}}\Bigl{[}(M_{1}^{x})^{2}+(M_{1}^{y})^{2}\Bigr{]}=-{1% \over{2M_{0}}}|\psi|^{2}.italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_z end_POSTSUPERSCRIPT = - divide start_ARG 1 end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG [ ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT + ( italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ] = - divide start_ARG 1 end_ARG start_ARG 2 italic_M start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG | italic_ψ | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT . (A9)𝐴9( italic_A 9 )

Similarly, using the complex field ψ𝜓\psiitalic_ψ, the components M1xsuperscriptsubscript𝑀1𝑥M_{1}^{x}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT and M1ysuperscriptsubscript𝑀1𝑦M_{1}^{y}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT of the magnetization of the anisotropic ferromagnetic medium can be constructed as M1x=[ψ+ψ2]superscriptsubscript𝑀1𝑥delimited-[]𝜓superscript𝜓2M_{1}^{x}=\Bigl{[}{\psi+\psi^{*}\over{2}}\Bigr{]}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_x end_POSTSUPERSCRIPT = [ divide start_ARG italic_ψ + italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT end_ARG start_ARG 2 end_ARG ] and M1y=[ψψ2i]superscriptsubscript𝑀1𝑦delimited-[]𝜓superscript𝜓2𝑖M_{1}^{y}=-\Bigl{[}{\psi-\psi^{*}\over{2i}}\Bigr{]}italic_M start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_y end_POSTSUPERSCRIPT = - [ divide start_ARG italic_ψ - italic_ψ start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT end_ARG start_ARG 2 italic_i end_ARG ].

References

  • [1] Fleischer JW, Segev M, Efremidis NK, Christodoulides DN. Observation of two-dimensional discrete solitons in optically induced nonlinear photonic lattices. Nature 2003;422:147.
  • [2] Fert A, Reyren N, Cros V. Magnetic skyrmions: advances in physics and potential applications. Nat Rev Mater 2017;2:17031.
  • [3] Stanciu CD, Hansteen F, Kimel AV, Kirilyuk A, Tsukamoto A, Itoh A, Rasing T. All-optical magnetic recording with circularly polarized light. Phys Rev lett 2007;99:047601.
  • [4] Chien CL, Gornakov VS, Nikitenko VI, Shapiro AJ, Shull RD. Hybrid domain walls and antiferromagnetic domains in exchange-coupled ferromagnet/antiferromagnet bilayers. Phys Rev B 2003;68:014418.
  • [5] Zhang J, Wen S, Xiang Y, Wang Y, Luo H. Spatiotemporal electromagnetic soliton and spatial ring formation in nonlinear metamaterials. Phys Rev A 2010;81:023829.
  • [6] Leblond H, Manna M. Two-dimensional electromagnetic solitons in a perpendicularly magnetized ferromagnetic slab. Phys Rev B 2009;80:064424.
  • [7] Veerakumar V, Daniel M. Electromagnetic soliton in an anisotropic ferromagnetic medium under nonuniform perturbation. Phys Lett A 2001;278:331.
  • [8] Leblond H. Interaction of two solitary waves in a ferromagnet. J Phys A Math Gen 1995;28:3763.
  • [9] Lakshmanan M, Ruijgrok TW, Thompson CJ. On the dynamics of a continuum spin system. Phys A Stat Mec Appl 1976;84:577.
  • [10] Lakshmanan M. Continuum spin system as an exactly solvable dynamical system. Phys Lett A 1977;61:53.
  • [11] Daniel M, Veerakumar V, Amuda R. Soliton and electromagnetic wave propagation in a ferromagnetic medium. Phys Rev E 1997;55:3619.
  • [12] Sathishkumar P, Senjudarvannan R. Oscillating electromagnetic soliton in an anisotropic ferromagnetic medium. J Magn Magn Mater 2017;429:379.
  • [13] Veerakumar V, Daniel M. Electromagnetic soliton in an anisotropic ferromagnetic medium under nonuniform perturbation. Phys Lett A 2001;278:331.
  • [14] Veerakumar V, Daniel M. Electromagnetic soliton damping in a ferromagnetic medium. Phys Rev E 1998;57:1197.
  • [15] Matsubara M, Schmehl A, Mannhart J, Schlom DG, Fiebig M. Large nonlinear magneto-optical effect in the centrosymmetric ferromagnetic semiconductor EuO𝐸𝑢𝑂EuOitalic_E italic_u italic_O. Phys Rev B 2010;81:214447.
  • [16] Ogawa N, Satoh T, Ogimoto Y, Miyano K. Half-metallic spin dynamics at a single LaMnO3/SrMnO3𝐿𝑎𝑀𝑛subscript𝑂3𝑆𝑟𝑀𝑛subscript𝑂3LaMnO_{3}/SrMnO_{3}italic_L italic_a italic_M italic_n italic_O start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT / italic_S italic_r italic_M italic_n italic_O start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT interface studied with nonlinear magnetooptical kerr effect. Phys Rev B 2009;80:241104.
  • [17] Demidov VE, Demokritov SO, Rott K, Krzysteczko P, Reiss G. Mode interference and periodic self-focusing of spin waves in permalloy microstripes. Phys Rev B 2008;77:064406.
  • [18] Logoboy NA, Sonin EB. Two-dimensional domain-wall magnon waves in superconducting ferromagnets. Phys Rev B 2007;75:153206.
  • [19] Kavitha L, Saravanan M, Srividya S, Gopi D. Breatherlike electromagnetic wave propagation in an antiferromagnetic medium with Dzyaloshinsky-Moriya interaction. Phys Rev E 2011;84:066608.
  • [20] Leblond H, Manna M. Benjamin-feir-type instability in a saturated ferrite: Transition between focusing and defocusing regimes for polarized electromagnetic waves. Phys Rev E 1994;50:2275.
  • [21] MacKay RS, Sepulchre JA. Stability of discrete breathers. Phys D Nonl Phen 1998;119:148.
  • [22] Lakshmanan M, Saxena A. Dynamic and static excitations of a classical discrete anisotropic heisenberg ferromagnetic spin chain. Phys D Nonl Phen 2008;237:885.
  • [23] Lakshmanan M, Subash B, Saxena A. Intrinsic localized modes of a classical discrete anisotropic heisenberg ferromagnetic spin chain. Phys Lett A 2014;378:1119.
  • [24] Runge AFJ, Broderick NGR, Erkintalo M. Observation of soliton explosions in a passively mode-locked fiber laser. Optica 2015;2:36.
  • [25] Braun OM, Kivshar YS. The Frenkel-Kontorova model: concepts, methods, and applications. Springer-Verlag: Berlin; 2004.
  • [26] Gonz\a’lez J, Bellorín A, Guerrero LE. Controlling soliton explosions. Phys Lett A 2005;338:60.
  • [27] Chua LO, Desoer CA, Kuh ES. Linear and nonlinear circuits. McGraw-Hill:New York;1987.
  • [28] Panfilov A, Holden A. Self-generation of turbulent vortices in a two-dimensional model of cardiac tissue. Phys Lett A 1990;151:463.
  • [29] Lakshmanan M. The fascinating world of the landau–lifshitz–gilbert equation: an overview. Phil Trans Roy Soc A Math, Phys and Eng Sci 2011;369:1280.
  • [30] Zakeri K, Lindner J, Barsukov I, Meckenstock R, et al. Spin dynamics in ferromagnets: Gilbert damping and two-magnon scattering. Phys Rev B 2007;76:104416.
  • [31] Hickey MC, Moodera JS. Origin of intrinsic gilbert damping. Phys Rev lett 2009;102:137601.
  • [32] Barla P, Joshi VK, Bhat S. Spintronic devices: a promising alternative to cmos devices. J Comp Elec 2021;20:805.
  • [33] Taniuti T, Yajima N. Perturbation method for a nonlinear wave modulation I. J Math Phys 1969;10:1369.
  • [34] Kaup DJ, Newell AC. An exact solution for a derivative nonlinear schrø”dinger equation. J Math Phys 1978;19:798.
  • [35] Liu SL, Wang WZ. Exact n-soliton solution of the modified nonlinear schrø”dinger equation. Phys Rev E 1993;48:3054.
  • [36] Anderson D, Lisak M. Nonlinear asymmetric self-phase modulation and self-steepening of pulses in long optical waveguides. Phys Rev A 1983;27:1393.
  • [37] Rogister A. Parallel propagation of nonlinear low-frequency waves in high-β𝛽\betaitalic_β plasma. Phys Fluids 1971;14:2733.
  • [38] Mertens FG, Quintero NR, Bishop AR. Nonlinear schrø”dinger equation with spatiotemporal perturbations. Phys Rev E 2010;81:016608.
  • [39] Lakshmanan M, Daniel M. Soliton damping and energy loss in the classical continuum heisenberg spin chain. Phys Rev B 1981;24:6751.