betaFeedback

WorksPhysics and Astronomy

Lienard-Wiechert Potentials due to a 'classically' spinning point-charge in Non-Relativistic arbitrary motion

Abstract

Li'enard-Wiechert potentials have been derived for a moving and 'classically' spinning point-charge; assuming it to be a small rigid charged-sphere in combined non-relativistic translational and rotational motion, and subsequently reducing its dimensions to 'point-particle' limit. The paper demonstrates that when the effect of rotation were taken into account, together with causality, expressions for the LW potentials accompany additional correction terms that contain spin-angular-momentum (or simply the 'classical-spin') of the point-charge.

Introduction

The principle of ‘causality’ states that no effect can occur outside the future light cone of its ‘cause’. [1]

However, because “classical physics” lacks a “arrow of time” (except “law of entropy” for macroscopic systems), ’retardation’ in classical field theories is introduced by hand to obtain causal solutions. [1] “Retarded Potentials” are such a set of electromagnetic potentials that are used to obtain causal solutions in the field theory of classical electrodynamics.

In classical electrodynamics, retarded potentials appear because “electromagnetic news travels with speed of light”, [2] which causes the information from different parts of a dynamic charge-configuration to reach the observer at different times; and forces the volume integration of charge and current densities to be evaluated at different times. [2]

Such a set of ‘causal’ potentials are known as the “Retarded Potentials”, given by[2] [3]:

Φ(R,t)=14πϵ0∭Vρ(r′,tret)R dV′A(R,t)=μ04π∭Vj(r′,tret)R dV′(1)\begin{aligned} \Phi(\mathbf{R},t) &= \frac{1}{4\pi\epsilon_0}\iiint_V \frac{\rho(\mathbf{r}',t_{\mathrm{ret}})}{R}\,dV' \\ \mathbf{A}(\mathbf{R},t) &= \frac{\mu_0}{4\pi}\iiint_V \frac{\mathbf{j}(\mathbf{r}',t_{\mathrm{ret}})}{R}\,dV' \tag*{(1)} \end{aligned}

Where, ρ(r′,tret)\rho(\mathbf{r}',t_{\mathrm{ret}}) and j(r′,tret)\mathbf{j}(\mathbf{r}',t_{\mathrm{ret}}) being the charge and current densities that prevailed at point r′\mathbf{r}' and retarded time trett_{\mathrm{ret}}, respectively.

For a point charge, the charge and current density functions ρ(r′,tret)\rho(\mathbf{r}',t_{\mathrm{ret}}) and j(r′,tret)\mathbf{j}(\mathbf{r}',t_{\mathrm{ret}}) are either replaced by direc-delta functions [3], or they are evaluated for an extended charge distribution in the limit of its size approaching to zero.[2]

The potentials thus obtained for the moving point charge are given by[2] [3]:

Φ(R,t)=q4πϵ0cRc−R⋅v=q4πϵ0R11−n^⋅βA(R,t)=q4πϵ0cvRc−R⋅v=q4πϵ0Rcβ1−n^⋅β=vc2Φ(R,t)(2)\begin{aligned} \Phi(\mathbf{R},t) &= \frac{q}{4\pi\epsilon_0}\frac{c}{Rc-\mathbf{R}\cdot\mathbf{v}} = \frac{q}{4\pi\epsilon_0 R}\frac{1}{1-\hat{\mathbf{n}}\cdot\boldsymbol{\beta}} \\ \mathbf{A}(\mathbf{R},t) &= \frac{q}{4\pi\epsilon_0 c}\frac{\mathbf{v}}{Rc-\mathbf{R}\cdot\mathbf{v}} = \frac{q}{4\pi\epsilon_0 Rc}\frac{\boldsymbol{\beta}}{1-\hat{\mathbf{n}}\cdot\boldsymbol{\beta}} \\ &= \frac{\mathbf{v}}{c^2}\Phi(\mathbf{R},t) \tag*{(2)} \end{aligned}

These are famously known as Liénard-Wiechert (LW) potentials, developed in part by Alfred-Marie Liénard in 1898 and independently by Emil Wiechert in 1900.

In 4-vector form, with flat space-time (Minkowski) metric being ηαβ=diag⁡[−1;1;1;1]\eta_{\alpha\beta} = \operatorname{diag}[-1;1;1;1], LW 4-potentials may be written as:

Aμ(Rμ)=q4πϵ0c2uμ(−Rμuμ)(3)A^\mu(R^\mu) = \frac{q}{4\pi\epsilon_0c^2}\frac{u^\mu}{(-R_\mu u^\mu)} \tag*{(3)}

Where: Rμ=[ctret;R]R^\mu= [ct_{\mathrm{ret}};\mathbf{R}] is the instantaneous 4-distance connecting the ‘retarded’ position of the point-charge, at time t−trett-t_{\mathrm{ret}}, to the distant observer at present time tt. And uμ=[c;v]u^\mu= [c;\mathbf{v}] being the instantaneous 4-velocity of the point-charge at the ‘retarded’ position.

Since equations (2) and (3) make no reference to the dimensions of the charge, they do hold equally for extended charges as well, provided their dimensions are negligible in comparison to the distances at which the fields are being evaluated. [4] [5]

In classical electrodynamics the only thing known about a point charge (e.g. electron) is that it has a specific ‘total’ charge, and any calculations of its electromagnetic fields do not take into account how this charge is distributed geometrically within it. [4]

Further, it is impossible to assume that the charge could have ‘zero’ physical extent without introducing various mathematical divergences. However, most field characteristics are independent of the charge’s physical geometry and dimensions, as long as they are tiny in comparison to the distances at which the fields are being observed. [4]

In this paper, we assume that our point-charge has a finite size; though, negligible compared to the distance from which its fields are being observed. In other words, it appears like a ‘point’ to the distant observer. Thus, in this paper mathematically we mean the “point-particle limit” as: “dimensions are sufficiently close to zero”, rather than “dimensions are zero”.

The purpose of this paper is to extend equations (2) for a moving point-charge, that is also rotating about its center-line; or in other words “spinning classically”.

Special relativity further limits the theoretical size of a ‘spinning’ ‘point-particle’ to constraint the tangential velocities of its surface below the speed of light, though the same is not in the scope of this paper, which is limited to only “non-relativistic” motion, including rotation.

In this work ‘classical-spin’ (“Spin Angular-Momentum” of a point-charge) has been considered a ‘bulk’ ‘invariant’ observable property of the point-charge independent of whatever geometry and velocities its different parts might have inside; much like the ‘total-charge’ which is independent of whatever geometry and charge-densities its parts might have inside.

Arakelyan et al. [6], evaluated LW potentials for “pseudo-classical spinning point-particle” in 2000. However, in their work spin terms appear alongside 1/R21/R^2; which is essentially due to the fact that in their work ( and their earlier cited works [6]) ‘spin’ was described in terms of “magnetic dipole-moment” that appeared from “Multi-pole expansion” of potentials.

In contrast, our work shows ‘spin’ as the fundamental quantity: “spin angular-momentum”. Also, In our work, spin terms appear alongside 1/R1/R as well.

To further comprehend, we assume that our ‘spinning’ point-charge is spherically symmetrical. Though, several other configurations may be feasible, we believe that as long as the charge’s physical extent is negligibly small, the outcome will be the same (or of the same order of magnitude).

Griffiths [2] and Lorentz [7] performed such calculations to determine the ‘Self-Force’ for a point charge. While Lorentz evaluated the self-force for a spherical charge, Griffiths demonstrated it for ‘dumbbell’ configuration in which the total charge was divided into two lobes separated by a short distance. However, in ‘point-particle’ limit both the cases produced the same result.

Lastly, we approximate our charge to be ‘rigid’; so that all parts of the charge would be co-moving with the center and accelerate with the same amount with time, in the observer’s rest-frame. This assumption greatly simplifies our mathematics, because in the observer’s rest-frame the charge would retains its spherical shape (though, apparently it would looks deformed due to retardation effects.)

Also, as our charged sphere has negligible physical extent, it would perceive the force-field, through which it is moving, as nearly ‘uniform’; thus, causing the same acceleration for all the points within the sphere.

In short, our work seeks to demonstrate ‘retardation’ (not ‘relativistic’) effects measured by a distant observer, due to a non-relativistically moving and rotating (classically spinning), rigid and negligibly small charged sphere.

The organization of this paper is as follows:

Section- 2 demonstrates how a body’s acceleration also significantly distorts the apparent geometry that a distant observer perceives.

Section- 3 analyses our “rigid body assumption” and fixes further constraints on non-relativistic motion.

Section- 4 sums up the postulate for a spinning, rigid, spherical volume charge distribution in arbitrary motion.

Section- 5 continues with the derivation of LW 4-potentials for a non-relativistically pinning and moving point-charge; and demonstrates the presence of extra terms that contain ‘classical-spin’ of the point-particle.

Section- 6 attempts to analyze significance of the obtained extra terms.

Section- 7 confirms the validity of obtained expression.

Section- 8 is the conclusion, followed by references.

Electromagnetic news received from an Accelerated Object:

“Electromagnetic news travels with speed of light cc”. Thus, in non-static case, it is not the status of the source “right now” that matters, but rather its condition at some earlier time trett_{ret} when the ‘message’ departed. Since, this message must travel a distance RR to the observer, it gets delayed by R/cR/c.[2] This ‘earlier’ time trett_{ret} is called “Retarded Time”.

When a non-static charge-configuration is viewed by a distant observer, it rather presents a distorted picture of total charge and current. This is because the retardation tret=t−R/ct_{ret}=t-R/c obliges evaluation of charge and current densities ρ\rho and j\mathbf{j}, respectively, for different parts of the configuration at different times. [2]

This is a purely geometrical effect.[2]

Considering a rigid rod ABAB of length LL (measured in observer’s rest frame) lying along the X-axis and moving in the direction of positive X-axis with constant velocity vv, as in spacetime diagram of figure-Figure 1.

A rigid rod moving in 1-dimension with constant velocity

Figure 1. (Color Online) A rigid rod moving in 1-dimension with constant velocity.

The ‘knowledge’ about the rod that the distant observer at x=0x = 0 perceives, is due to ‘information’ that he receives from the rod at “one instant of time”. However, because of retardation, by the time the light-ray from far end BB meets the opposite end AA on its journey to the distant observer, it will have moved to A′A', by amount vΔtv\Delta t. While Δt\Delta t being the time it takes the light-ray from one end of the moving rod to other, “as viewed by the observer”.

Since, the velocity is constant, locus of any point on the rod, connecting its retarded-position at T=tretT = t_{ret} to its present position T=tret+ΔtT = t_{ret} + \Delta t (e.g. AA′AA'), will be a straight line on the spacetime diagram. The slope of this line is determined by the magnitude of the velocity as tan⁡θ=v/c=β\tan\theta= v/c = \beta. [8].

Therefore, the apparent length perceived by the distant observer comes out to be L′=L−vΔtL' = L - v\Delta t.

Solving for L′L' and Δt\Delta t, with L′=cΔtL' = c\Delta t, we get:

L′L=c(c+v)=1(1+β)(4)\frac{L'}{L} = \frac{c}{(c+v)} = \frac{1}{(1+\beta)} \tag*{(4)}

Which may be generalized for motion in arbitrary direction as:

L′=L(1−n^⋅β)=(cc−n^⋅v)L=(RcRc−R⋅v)L(5)L' = \frac{L}{(1-\hat{\mathbf{n}}\cdot\boldsymbol{\beta})} = \left(\frac{c}{c-\hat{\mathbf{n}}\cdot\mathbf{v}}\right)L = \left(\frac{Rc}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)L \tag*{(5)}

While, R\mathbf{R} is the vector distance pointing from the particle’s ‘retarded’ position to the observer’s ‘current’ position; and n^\hat{\mathbf{n}} is the unit vector in the direction of R\mathbf{R}.

The quantity (1−n^⋅β)(1-\hat{\mathbf{n}}\cdot\boldsymbol{\beta}) or (Rc−R⋅v)(Rc-\mathbf{R}\cdot\mathbf{v}) or −Rμuμ-R_\mu u^\mu in denominator appears purely due to the apparent geometry as perceived by the distant observer.

Thus, depending on the particle’s velocity vector, the observer perceives the retarded charge and current densities that are either in-access or less than the actual. [4]

Nevertheless, this geometrical effect has nothing to do with “Lorentz Contraction”, as LL is length of the ‘moving’ rod, and its ‘rest’ length is not an issue. [2]

When the same logic is extended to the accelerating rod, the picture gets distorted further.

With the assumption that the rod is ‘rigid’, and is acted upon by a uniform force field, the entire rod (i.e. both the ends as well as all the parts of the rod) would accelerate by same amount in a given “coordinate time difference”.

However, the distant observer would see different parts of the rod moving with different velocities. As a result, to the distant observer, the rod would not appear rigid; but would appear to expand or shrink over time (as long as the acceleration persists), depending on the direction of motion and strength of acceleration. This may be visualized from figure- Figure 2 below:

Image not extracted

Figure 2. (Color Online) Accelerating rigid rod in 1-dimension.

From the light-cone carrying the combined information form both the ends, the distant observer would infer that, the far end BB and the near end AA were moving with instantaneous velocities vBv_B and vA′v_{A'} respectively, at the same instance; while vA′≠vBv_{A'} \ne v_B.

Assuming the velocities and acceleration being small (vA′,vB≪cv_{A'}, v_B \ll c; aΔt≪ca\Delta t \ll c), non-relativistic (Newtonian) equations of motion may be used. Thus, the distorted length as perceived by the observer now comes:

L′=L−vBΔt−a2Δt2=L−(vB+vA′)2Δt(6)L' = L - v_B\Delta t - \frac{a}{2}\Delta t^2 = L - \frac{(v_B + v_{A'})}{2}\Delta t \tag*{(6)}

Though, it seems that for a point-particle with L≈0L \approx0, vA′→vBv_{A'} \to v_B; and the expressions for LW potentials (2) would remain unchanged; however, things are not this simple if the particle is rotating or ‘spinning’.

Jerrold Franklin [9] in 2010 shown that, in context of special relativity, for a rigid body to remain ’rigid’ under accelerated motion, acceleration in its own frame must vary throughout it’s volume to preserve its rest frame dimensions.

However, ‘retardation’ is a phenomenon that is applicable purely in observer’s perspective. “Retarded-time” refers to a ‘past’ coordinate-time that lies in the observer’s rest-frame Therefore, the accelerations experienced by the moving rod within its own frame are irrelevant.

In the upcoming section, we will further analyze the validity of “rigid body” assumption.

Rigid Body Approximation:

‘Rigid-body’ assumption is not a covariant concept as per relativity; or this would allow forces traveling within the rigid body at infinite speed. [10]

When acceleration or rotation is applied to a body’s one part, the information does not reach its other parts instantaneously. Rather, an elastic wave (of speed less than the speed of light) spreads within the body carrying the information to its other parts; causing phase lag among motion of the particles. [10]

Thus, according to relativity, all matter is non-rigid.

Considering a non-rigid rod ABAB of length Lno−rgdL_{no-rgd}, measured in observer’s rest frame; lying along the XX-axis and moving in the direction of positive XX-axis with velocity V1V_1, as shown in the space-time diagram of figure- Figure 3 below:

Image not extracted

Figure 3. (Color Online) Accelerated non-rigid rod.

At time T=0T = 0 the end BB of the rod is dragged in the direction of motion, causing its velocity change to V2V_2. The information regarding the change in motion does not reach the other end instantaneously; rather, it is carried by an elastic wave of speed WW that would propagate towards the end AA.

The end AA continues to move with the earlier velocity V1V_1 until the elastic wave encounters it at A′′A^{\prime\prime}, at time t2t_2. Thus, the rod gets elongated in observer’s rest frame.

However, as W<cW < c, when the light ray from A′′A^{\prime\prime} reaches the distant observer, it also carries the information from the opposite end when that was at B′′B^{\prime\prime}, at time t1t_1. Thus, the distant observer perceives the rod as elongated with length Lno−rgd′=Lno−rgd+(V2−V1) t1L'_{no-rgd} = L_{no-rgd} + (V_2-V_1)\,t_1.

Also, as Lno−rgd′=c (t2−t1)L'_{no-rgd} = c\,(t_2-t_1) and Lno−rgd=(W+V1) t2L_{no-rgd} = (W+V_1)\,t_2, the apparent elongated length as perceived by the observer can be evaluated as:

Lno−rgd′Lno−rgd=c(c+V2)(W+V2W+V1)(7)\frac{L'_{no-rgd}}{L_{no-rgd}} = \frac{c}{(c+V_2)}\left(\frac{W+V_2}{W+V_1}\right) \tag*{(7)}

The change in velocity (V2−V1)(V_2-V_1) is not straightaway proportional to the applied force; but is also affected by the elastic resistance offered by the rod’s material. Further, in the end, the rod may either retain the elongation, or revert to its original length following several to-and-fro elastic wave oscillations between its ends; depending upon the elastic properties of the rod’s material.

Though, we don’t know how fast elastic waves move within the matter of point-charges, e.g. electron; data on “speed of sound” in elements [11] suggests that it could be several thousand times the “speed of sound” in air.

A comparison of equation (7) and (4) shows that the non-rigid condition introduces an elastic factor (W+V2)/(W+V1)(W+V_2)/(W+V_1) to the apparent retarded length.

For V1,V2≪WV_1,V_2 \ll W, the elastic factor approaches unity and the retarded length of non-rigid rod approaches to that of a rigid-rod. Thus, for very slow motion, elastic deformation would be negligible; and the body may be considered as “near rigid”; or may be ‘approximated’ as ‘rigid’.

The next section demonstrates the effect of acceleration on the geometry of a spinning rigid sphere.

Effect of Acceleration and Spin on apparent geometry of a point-sphere:

Considering a spherical volume charge distribution of total charge qq, radius rqr_q and spinning with angular-velocity ωs\omega_s. The center of the charged-sphere is moving in positive Z-direction, with instantaneous velocity v\mathbf{v} and acceleration a\mathbf{a}.

For the sake of simplicity, we assume that the charge distribution’s spherical shape is in equilibrium with the rotation; that is, it has a spherical shape while rotating with angular velocity ωs\omega_s rather than at ‘rest’.

We also assume that the sphere is ‘near-rigid’, which means that all transnational and rotational velocities are significantly lower than the speed of an elastic wave sustained by the sphere’s matter. Thus, any elastic deformations are negligibly smaller than the various transnational and angular displacements.

Finally, we assume our sphere is small enough, so that the “force-causing field” in which it is moving appears ‘uniform’ to it; as a result, every point within it feels the same acceleration in a given “coordinate time difference”.

Considering an infinitesimal volume-element r2dΩdrr^2 d\Omega dr at position r\mathbf{r} inside the sphere, with reference to its center, (drdr is the thickness, and dΩd\Omega is the solid-angle that the volume-element subtends at the center of the sphere) as shown in figure- Figure 4:

Spinning point sphere in arbitrary motion

Figure 4. (Color Online) Spinning point sphere in arbitrary motion.

Due to linear and rotational velocities and accelerations, the volume-element would be subjected to retardation with reference to the center of the sphere.

Thus, when a distant observer detects a light-ray from the center of the sphere, the ray that arrives from the volume-element appears to have originated from a slightly displaced position r′\mathbf{r}'.

While, the amount of retardation satisfies the equation:

cΔt=n^⋅r′(8)c\Delta t = \hat{\mathbf{n}} \cdot\mathbf{r}' \tag*{(8)}

Where, Δt\Delta t is the coordinate-time difference in which the volume-element moves from position r\mathbf{r} to r′\mathbf{r}'; and n^\hat{\mathbf{n}} is the unit-vector pointing in the direction of faraway observer.

If vrv_r and ara_r being the net velocity and acceleration of the volume-element at position r\mathbf{r}, respectively; the retarded-position r′\mathbf{r}' and velocity at retarded-position vr′v_{r'}, are evaluated using the Taylor series as:

r′=r+vrΔt+12arΔt2+16a˙rΔt3+⋯(9)\mathbf{r}' = \mathbf{r} + v_r\Delta t + \frac{1}{2}a_r\Delta t^2 + \frac{1}{6}\dot{a}_r\Delta t^3 + \cdots \tag*{(9)}

and,

vr′=vr+arΔt+12a˙rΔt2+16a¨rΔt3+⋯(10)v_{r'} = v_r + a_r\Delta t + \frac{1}{2}\dot{a}_r\Delta t^2 + \frac{1}{6}\ddot{a}_r\Delta t^3 + \cdots \tag*{(10)}

Where, a˙r=dar/dtret\dot{a}_r = da_r/dt_{\mathrm{ret}} and a¨r=d2ar/dtret2\ddot{a}_r = d^2a_r/dt_{\mathrm{ret}}^2 being the acceleration and jerk of the volume-element, respectively, evaluated at retarded time trett_{\mathrm{ret}}.

Further, taking spin into account, vrv_r would be the sum of the center’s velocity and the rotational velocity of the volume-element with respect to the center, as given by:

vr=v+ωs×r(11)v_r = \mathbf{v} + \boldsymbol{\omega}_s \times\mathbf{r} \tag*{(11)}

From, vrv_r above, ara_r may be evaluated as:

ar=dvrdtret=a+dωsdtret×r+ωs×(ωs×r)=a+ω˙s×r+{(ωs⋅r)ωs−ωs2r}(12)\begin{aligned} a_r &= \frac{dv_r}{dt_{\mathrm{ret}}} = \mathbf{a} + \frac{d\boldsymbol{\omega}_s}{dt_{\mathrm{ret}}} \times\mathbf{r} + \boldsymbol{\omega}_s \times(\boldsymbol{\omega}_s \times\mathbf{r}) \\ &= \mathbf{a} + \dot{\boldsymbol{\omega}}_s \times\mathbf{r} + \{(\boldsymbol{\omega}_s \cdot\mathbf{r})\boldsymbol{\omega}_s - \omega_s^2\mathbf{r}\} \tag*{(12)} \end{aligned}

The first term in equation (12) is instantaneous linear acceleration of the center of the sphere. The second term is the acceleration component caused by angular-acceleration. The third-term represents the centrifugal-acceleration.

Furthermore, the jerk on the volume-element at position r\mathbf{r} can be evaluated as:

a˙r=dardtret=dadtret+d2ωsdtret2×r+2dωsdtret×(ωs×r)+ωs×(dωsdtret×r)+ωs×{ωs×(ωs×r)}=a˙+ω¨s×r+2ω˙s×(ωs×r)+ωs×(ω˙s×r)−ωs2(ωs×r)(13)\begin{aligned} \dot{a}_r &= \frac{da_r}{dt_{\mathrm{ret}}} \\ &= \frac{d\mathbf{a}}{dt_{\mathrm{ret}}} + \frac{d^2\boldsymbol{\omega}_s}{dt_{\mathrm{ret}}^2} \times\mathbf{r} + 2\frac{d\boldsymbol{\omega}_s}{dt_{\mathrm{ret}}} \times(\boldsymbol{\omega}_s \times\mathbf{r}) + \boldsymbol{\omega}_s \times\left(\frac{d\boldsymbol{\omega}_s}{dt_{\mathrm{ret}}} \times\mathbf{r}\right) \\ &\qquad+ \boldsymbol{\omega}_s \times\{\boldsymbol{\omega}_s \times(\boldsymbol{\omega}_s \times\mathbf{r})\} \\ &= \dot{\mathbf{a}} + \ddot{\boldsymbol{\omega}}_s \times\mathbf{r} + 2\dot{\boldsymbol{\omega}}_s \times(\boldsymbol{\omega}_s \times\mathbf{r}) + \boldsymbol{\omega}_s \times(\dot{\boldsymbol{\omega}}_s \times\mathbf{r}) \\ &\qquad- \omega_s^2(\boldsymbol{\omega}_s \times\mathbf{r}) \tag*{(13)} \end{aligned}

If R\mathbf{R} is the vector distance of the distant observer from the center of the point-charge, the vector distance of the observer from the retarded position of the volume-element is given by:

R′=R−r′(14)\mathbf{R}' = \mathbf{R} - \mathbf{r}' \tag*{(14)}

and, scalar-distance R′R' is given by:

R′=R2+r′2−2R⋅r′=R{1+r′2R2−2R(n^⋅r′)}12(15)R' = \sqrt{R^2 + r'^2 - 2\mathbf{R}\cdot\mathbf{r}'} = R\left\{1 + \frac{r'^2}{R^2} - \frac{2}{R}(\hat{\mathbf{n}}\cdot\mathbf{r}')\right\}^{\frac{1}{2}} \tag*{(15)}

Equations (14) and (15) along with (9) show that due to retardation, the sphere appears deformed to the distant observer.

Moreover, the amount of apparent-deformation varies in complex manner depending on magnitudes of velocity, acceleration, angular-velocity of the spin, their relative orientations with reference to the observer, and their mutual interaction.

In the following section we evaluate LW potentials for our spinning small rigid sphere in non-relativistic motion. We shall also demonstrate that in the point-particle limit rq→0r_q \to0 if spin-angular-momentum were considered a bulk property of the point-charge, independent of its dimensions, the terms containing ‘classical-spin’ survive and show up in the expression of LW potentials.

LW 4-Potentials due to Spinning Point-Charge in motion:

Using equation (5) for apparently deformed dimensions and with ρ\rho being the charge-density, the retarded 4-potentials due to the volume-element is given by:

dAμ(Rμ)=ρuμ(r′)4πϵ0c2r2dΩdr{−Rμ′uμ(r′)}=ρ4πϵ0c2[c;vr′]r2dΩdr{R′c−R′⋅vr′}(16)\begin{aligned} dA^\mu(R^\mu) &= \frac{\rho u^\mu(\mathbf{r}')}{4\pi\epsilon_0 c^2}\frac{r^2d\Omega dr}{\{-R'_\mu u^\mu(\mathbf{r}')\}} \\ &= \frac{\rho}{4\pi\epsilon_0 c^2}\frac{[c;\mathbf{v}_{r'}]r^2d\Omega dr}{\{R'c-\mathbf{R}'\cdot\mathbf{v}_{r'}\}} \tag*{(16)} \end{aligned}

Integrating equation (16) over the entire volume of the sphere and applying the point-particle limit rq→0r_q \to0 would yield LW potentials due to “spinning point-charge in arbitrary motion”.

However, the 4-vector terms uμ(r′)/{−Rμ′uμ(r′)}u^\mu(\mathbf{r}')/\{-R'_\mu u^\mu(\mathbf{r}')\} for the volume-element, which represent the apparent deformed geometry, need to be evaluated first.

From equation (8), Δt\Delta t seems to be of the order of r′/cr'/c. In the non-relativistic limit, i.e. for ’slow enough’ velocity and acceleration (v≪cv \ll c, aΔt≪ca\Delta t \ll c), r′r' approaches rr.

Further, rr being the radial distance from the center of point-particle, is too small as compared to the distance from the observer (i.e. r≪Rr \ll R). Thus, in point-particle limit rq→0r_q \to0, Δt\Delta t would also be negligibly small; and Δt≈r/c\Delta t \approx r/c.

However, as the order of angular-momentum goes as r2r^2, to examine spin dependence of LW potentials, we must include terms of at least the order Δt2\Delta t^2 in our approximation.

Thus, equation (9) for the apparent retardation and equation (10) for the apparent velocity at retarded position, as sensed by the faraway observer, are approximated to the second powers of Δt\Delta t, as:

r′≈r+vrΔt+12arΔt2(17)r' \approx r+v_r\Delta t+\frac{1}{2}a_r\Delta t^2 \tag*{(17)}

and,

vr′≈vr+arΔt+12a˙rΔt2(18)v_{r'} \approx v_r+a_r\Delta t+\frac{1}{2}\dot{a}_r\Delta t^2 \tag*{(18)}

From (14) and (15), along with (17) and (18), expanding the square-root term of (15) by binomial theorem, and keeping only up to r2r^2 like terms (i.e. up to r2r^2, rΔtr\Delta t and Δt2\Delta t^2), the quantity in denominator of (16) −Rμ′uμ(r′)-R'_\mu u^\mu(r') may be evaluated as:

−Rμ′μμ(r′)=R′c−R′⋅vr′=R[c−n^⋅vr−(n^⋅ar)Δt−12(n^⋅a˙r)Δt2+[(r⋅vr)−c(n^⋅r)−c(n^⋅vr)Δt+vr2Δt+(r⋅ar)Δt−c2(n^⋅ar)+32(vr⋅ar)]+cR[r22−(r⋅vr)Δt+12vr2Δt2−12(n^⋅vr)2Δt2−12(n^⋅r)2−(n^⋅vr)(n^⋅r)Δt](19)\begin{aligned} -R'_{\mu}\mu^\mu(\mathbf{r}') &= R'c-\mathbf{R}'\cdot\mathbf{v}_{r'} \\ &= R[c-\hat{\mathbf{n}}\cdot\mathbf{v}_r-(\hat{\mathbf{n}}\cdot\mathbf{a}_r)\Delta t-\frac{1}{2}(\hat{\mathbf{n}}\cdot\dot{\mathbf{a}}_r)\Delta t^2\\ &\quad+[(\mathbf{r}\cdot\mathbf{v}_r)-c(\hat{\mathbf{n}}\cdot\mathbf{r})-c(\hat{\mathbf{n}}\cdot\mathbf{v}_r)\Delta t+v_r^2\Delta t\\ &\quad+(\mathbf{r}\cdot\mathbf{a}_r)\Delta t-\frac{c}{2}(\hat{\mathbf{n}}\cdot\mathbf{a}_r)+\frac{3}{2}(\mathbf{v}_r\cdot\mathbf{a}_r)]\\ &\quad+\frac{c}{R}[\frac{r^2}{2}-(\mathbf{r}\cdot\mathbf{v}_r)\Delta t+\frac{1}{2}v_r^2\Delta t^2\\ &\quad-\frac{1}{2}(\hat{\mathbf{n}}\cdot\mathbf{v}_r)^2\Delta t^2-\frac{1}{2}(\hat{\mathbf{n}}\cdot\mathbf{r})^2-(\hat{\mathbf{n}}\cdot\mathbf{v}_r)(\hat{\mathbf{n}}\cdot\mathbf{r})\Delta t] \tag*{(19)} \end{aligned}

Meanwhile, Δt\Delta t may be evaluated by using equation (17) with (8), which yields the quadratic equation in Δt\Delta t:

12(n^⋅ar)Δt2−(c−n^⋅vr)Δt+n^⋅r=0(20)\frac{1}{2}(\hat{\mathbf{n}}\cdot\mathbf{a}_r)\Delta t^2-(c-\hat{\mathbf{n}}\cdot\mathbf{v}_r)\Delta t+\hat{\mathbf{n}}\cdot\mathbf{r}=0 \tag*{(20)}

solution of which comes out to be:

Δt=c−n^⋅vrn^⋅ar±1n^⋅ar(c−n^⋅vr)2−2(n^⋅ar)(n^⋅r)(21)\Delta t=\frac{c-\hat{\mathbf{n}}\cdot\mathbf{v}_r}{\hat{\mathbf{n}}\cdot\mathbf{a}_r}\pm\frac{1}{\hat{\mathbf{n}}\cdot\mathbf{a}_r}\sqrt{(c-\hat{\mathbf{n}}\cdot\mathbf{v}_r)^2-2(\hat{\mathbf{n}}\cdot\mathbf{a}_r)(\hat{\mathbf{n}}\cdot\mathbf{r})} \tag*{(21)}

As Δt\Delta t must vanish at r=0r=0, solving with negative sign and expanding the square-root with binomial theorem, we get the expression for Δt\Delta t, approximated up to r2r^2 terms as:

Δt≈n^⋅rc−n^⋅vr+12(n^⋅r)2(n^⋅ar)(c−n^⋅vr)3(22)\Delta t\approx\frac{\hat{\mathbf{n}}\cdot\mathbf{r}}{c-\hat{\mathbf{n}}\cdot\mathbf{v}_r}+\frac{1}{2}\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})^2(\hat{\mathbf{n}}\cdot\mathbf{a}_r)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v}_r)^3} \tag*{(22)}

Using expression of Δt\Delta t of equation (22) in equation (18), velocity of the volume-element at retarded position r′\mathbf{r}', approximated up to r2r^2 terms, comes out to be:

vr′≈vr+(n^⋅r)arc−n^⋅vr+12(n^⋅r)2(n^⋅ar)ar(c−n^⋅vr)3+12(n^⋅r)2a˙r(c−n^⋅vr)2(23)\mathbf{v}_{r'}\approx\mathbf{v}_r+\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})\mathbf{a}_r}{c-\hat{\mathbf{n}}\cdot\mathbf{v}_r}+\frac{1}{2}\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})^2(\hat{\mathbf{n}}\cdot\mathbf{a}_r)\mathbf{a}_r}{(c-\hat{\mathbf{n}}\cdot\mathbf{v}_r)^3}+\frac{1}{2}\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})^2\dot{\mathbf{a}}_r}{(c-\hat{\mathbf{n}}\cdot\mathbf{v}_r)^2} \tag*{(23)}

Using equations (11), (12) and (13) in (23) and up to the r2r^2 terms, we get:

vr′≈v+ωs×r+(n^⋅r)ac−n^⋅v+(n^⋅r)(r⋅[n^×ωs])a(c−n^⋅v)2−(n^⋅r)(r×ω˙s)c−n^⋅v+(n^⋅r)(ωs⋅r)ωsc−n^⋅v−(n^⋅r)ωs2rc−n^⋅v+12(n^⋅r)2{n^⋅a}a(c−n^⋅v)3+12(n^⋅r)2a˙(c−n^⋅v)2(24)\begin{aligned} \mathbf{v}_{r'}\approx{}&\mathbf{v}+\boldsymbol{\omega}_s\times\mathbf{r}+\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})\mathbf{a}}{c-\hat{\mathbf{n}}\cdot\mathbf{v}}+\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot[\hat{\mathbf{n}}\times\boldsymbol{\omega}_s])\mathbf{a}}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}\\ &-\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\times\dot{\boldsymbol{\omega}}_s)}{c-\hat{\mathbf{n}}\cdot\mathbf{v}}+\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})(\boldsymbol{\omega}_s\cdot\mathbf{r})\boldsymbol{\omega}_s}{c-\hat{\mathbf{n}}\cdot\mathbf{v}}-\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})\omega_s^2\mathbf{r}}{c-\hat{\mathbf{n}}\cdot\mathbf{v}}\\ &+\frac{1}{2}\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})^2\{\hat{\mathbf{n}}\cdot\mathbf{a}\}\mathbf{a}}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}+\frac{1}{2}\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})^2\dot{\mathbf{a}}}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2} \tag*{(24)} \end{aligned}

Further, using expression of Δt\Delta t from equation (22), and inverting equation (19) with binomial-expansion in fractions of (c−n^⋅vr)(c-\hat{\mathbf{n}}\cdot\mathbf{v}_r), up to r2r^2 like terms; we get:

1{−Rμ′uμ(r′)}≈1R[1(c−n^⋅vr)+(n^⋅ar)(n^⋅r)(c−n^⋅vr)3+32(n^⋅ar)2(n^⋅r)2(c−n^⋅vr)5+12(n^⋅ar)(n^⋅r)2(c−n^⋅vr)4]+1R2[c2(n^⋅r)(c−n^⋅vr)3−(r⋅vr)(c−n^⋅vr)2+5c2(n^⋅ar)(n^⋅r)22(c−n^⋅vr)5−vr2(n^⋅r)(c−n^⋅vr)3−2(n^⋅ar)(r⋅vr)(n^⋅r)(c−n^⋅vr)3−(r⋅ar)(n^⋅r)(c−n^⋅vr)3−52vr2(n^⋅ar)(n^⋅r)2(c−n^⋅vr)5−32(vr⋅ar)(n^⋅r)2(c−n^⋅vr)4]+cR3[(r⋅vr)(n^⋅r)(c−n^⋅vr)3−12r2(c−n^⋅vr)2−12(vr)2(n^⋅r)2(c−n^⋅vr)4+c2(n^⋅r)22(c−n^⋅vr)4−2c(r⋅vr)(n^⋅r)(c−n^⋅vr)4+2c(n^⋅vr)(n^⋅r)2(c−n^⋅vr)4−2cvr2(n^⋅r)2(c−n^⋅vr)5−2vr2(r⋅vr)(n^⋅r)(c−n^⋅vr)4](25)\begin{aligned} \frac{1}{\left\{-R'_{\mu}u^{\mu}(r')\right\}} &\approx\frac{1}{R}\left[\frac{1}{(c-\hat{n}\cdot v_r)}+\frac{(\hat{n}\cdot a_r)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^3}+\frac{3}{2}\frac{(\hat{n}\cdot a_r)^2(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^5}\right.\\ &\left.+\frac{1}{2}\frac{(\hat{n}\cdot a_r)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^4}\right]\\ &+\frac{1}{R^2}\left[\frac{c^2(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^3}-\frac{(r\cdot v_r)}{(c-\hat{n}\cdot v_r)^2}+\frac{5c^2(\hat{n}\cdot a_r)(\hat{n}\cdot r)^2}{2(c-\hat{n}\cdot v_r)^5}\right.\\ &\left.-\frac{v_r^2(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^3}-\frac{2(\hat{n}\cdot a_r)(r\cdot v_r)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^3}-\frac{(r\cdot a_r)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^3}\right.\\ &\left.-\frac{5}{2}\frac{v_r^2(\hat{n}\cdot a_r)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^5}-\frac{3}{2}\frac{(v_r\cdot a_r)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^4}\right]\\ &+\frac{c}{R^3}\left[\frac{(r\cdot v_r)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^3}-\frac{1}{2}\frac{r^2}{(c-\hat{n}\cdot v_r)^2}-\frac{1}{2}\frac{(v_r)^2(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^4}\right.\\ &\left.+\frac{c^2(\hat{n}\cdot r)^2}{2(c-\hat{n}\cdot v_r)^4}-\frac{2c(r\cdot v_r)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^4}+\frac{2c(\hat{n}\cdot v_r)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^4}\right.\\ &\left.-\frac{2cv_r^2(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v_r)^5}-\frac{2v_r^2(r\cdot v_r)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v_r)^4}\right] \tag*{(25)} \end{aligned}

Now, inserting spin dependent quantities by using equations (11), (12) and (13), in (25), and carrying out binomial-expansion of the denominator in fractions of (c−n^⋅v)(c-\hat{n}\cdot v), and keeping the terms up to r2r^2, we get:

1{−Rμ′uμ(r′)}≈1R[1(c−n^⋅v){1+n^⋅[r×ωs](c−n^⋅v)+(n^⋅[r×ωs])2(c−n^⋅v)2}+(n^⋅r)(c−n^⋅v)3{n^⋅(a+ωs×r+(ωs⋅r)ωs−ωs2r)}{1+3(n^⋅[ωs×r])(c−n^⋅v)}+32(n^⋅a)2(n^⋅r)2(c−n^⋅v)5+12(n^⋅a)(n^⋅r)2(c−n^⋅v)4]+1R2[c2(n^⋅r)(c−n^⋅v)3{1+3(n^⋅[ωs×r])(c−n^⋅v)}−1(c−n^⋅v)2{r⋅(v+ωs×r)}{1+2(n^⋅[ωs×r])(c−n^⋅v)}+5c2(n^⋅a)(n^⋅r)22(c−n^⋅v)5−(n^⋅r)(c−n^⋅v)3{v+ωs×r}2{1+3(n^⋅[ωs×r])(c−n^⋅v)}−2(n^⋅a)(r⋅v)(n^⋅r)(c−n^⋅v)3−(r⋅a)(n^⋅r)(c−n^⋅v)3−52v2(n^⋅a)(n^⋅r)2(c−n^⋅v)5−32(v⋅a)(n^⋅r)2(c−n^⋅v)4]+cR3[(r⋅v)(n^⋅r)(c−n^⋅v)3−12r2(c−n^⋅v)2−12v2(n^⋅r)2(c−n^⋅v)4+c2(n^⋅r)22(c−n^⋅v)4−2c(r⋅v)(n^⋅r)(c−n^⋅v)4+2c(n^⋅v)(n^⋅r)2(c−n^⋅v)4−2cv2(n^⋅r)2(c−n^⋅v)5−2v2(r⋅v)(n^⋅r)(c−n^⋅v)4](26)\begin{aligned} \frac{1}{\left\{-R'_{\mu}u^{\mu}(r')\right\}} &\approx\frac{1}{R}\left[\frac{1}{(c-\hat{n}\cdot v)}\left\{1+\frac{\hat{n}\cdot\left[r\times\omega_s\right]}{(c-\hat{n}\cdot v)}+\frac{\left(\hat{n}\cdot\left[r\times\omega_s\right]\right)^2}{(c-\hat{n}\cdot v)^2}\right\}\right.\\ &\left.+\frac{(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^3}\left\{\hat{n}\cdot\left(a+\omega_s\times r+(\omega_s\cdot r)\omega_s-\omega_s^2r\right)\right\}\left\{1+\frac{3(\hat{n}\cdot[\omega_s\times r])}{(c-\hat{n}\cdot v)}\right\}\right.\\ &\left.+\frac{3}{2}\frac{(\hat{n}\cdot a)^2(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^5}+\frac{1}{2}\frac{(\hat{n}\cdot a)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^4}\right]\\ &+\frac{1}{R^2}\left[c^2(\hat{n}\cdot r)(c-\hat{n}\cdot v)^3\left\{1+\frac{3(\hat{n}\cdot[\omega_s\times r])}{(c-\hat{n}\cdot v)}\right\}\right.\\ &\left.-\frac{1}{(c-\hat{n}\cdot v)^2}\{r\cdot(v+\omega_s\times r)\}\left\{1+\frac{2(\hat{n}\cdot[\omega_s\times r])}{(c-\hat{n}\cdot v)}\right\}\right.\\ &\left.+\frac{5c^2(\hat{n}\cdot a)(\hat{n}\cdot r)^2}{2(c-\hat{n}\cdot v)^5}-\frac{(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^3}\{v+\omega_s\times r\}^2\left\{1+\frac{3(\hat{n}\cdot[\omega_s\times r])}{(c-\hat{n}\cdot v)}\right\}\right.\\ &\left.-\frac{2(\hat{n}\cdot a)(r\cdot v)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^3}-\frac{(r\cdot a)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^3}\right.\\ &\left.-\frac{5}{2}\frac{v^2(\hat{n}\cdot a)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^5}-\frac{3}{2}\frac{(v\cdot a)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^4}\right]\\ &+\frac{c}{R^3}\left[\frac{(r\cdot v)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^3}-\frac{1}{2}\frac{r^2}{(c-\hat{n}\cdot v)^2}-\frac{1}{2}\frac{v^2(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^4}+\frac{c^2(\hat{n}\cdot r)^2}{2(c-\hat{n}\cdot v)^4}\right.\\ &\left.-\frac{2c(r\cdot v)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^4}+\frac{2c(\hat{n}\cdot v)(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^4}-\frac{2cv^2(\hat{n}\cdot r)^2}{(c-\hat{n}\cdot v)^5}-\frac{2v^2(r\cdot v)(\hat{n}\cdot r)}{(c-\hat{n}\cdot v)^4}\right] \tag*{(26)} \end{aligned}

Now LW Potentials can be simply evaluated calculating uμ(r′)/{−Rμ′uμ(r′)}u^{\mu}(r')/\left\{-R'_{\mu}u^{\mu}(r')\right\} using product of equations (26) above, and (24), plugging it into equation (16), performing integration over the entire volume of the sphere, and lastly applying the point-particle limit rq→0r_q\to0.

Though, it appears that the resulting expression of 4-vector uμ(r′)/{−Rμ′uμ(r′)}u^{\mu}(r')/\left\{-R'_{\mu}u^{\mu}(r')\right\} such obtained, would be be fairly lengthy; many terms may be omitted by inspection, using formulae for known “spherical integrals”, and considering the terms in the context of the point-particle limit rq→0r_q\to0.

Following are the known spherical integrals:

∬Ω(r⋅A) dΩ=0;∬Ω(r×A) dΩ=0;∬Ω(r⋅A)(r⋅B) dΩ=4π3r2(A⋅B);∬Ω(r⋅A)(r×B) dΩ=4π3r2(A×B);∬Ω(r⋅A)r dΩ=4π3r2A.(27)\begin{aligned} \iint_{\Omega}(\mathbf{r}\cdot\mathbf{A})\,d\Omega&= 0;\\ \iint_{\Omega}(\mathbf{r}\times\mathbf{A})\,d\Omega&= 0;\\ \iint_{\Omega}(\mathbf{r}\cdot\mathbf{A})(\mathbf{r}\cdot\mathbf{B})\,d\Omega&= \frac{4\pi}{3}r^2(\mathbf{A}\cdot\mathbf{B});\\ \iint_{\Omega}(\mathbf{r}\cdot\mathbf{A})(\mathbf{r}\times\mathbf{B})\,d\Omega&= \frac{4\pi}{3}r^2(\mathbf{A}\times\mathbf{B});\\ \iint_{\Omega}(\mathbf{r}\cdot\mathbf{A})r\,d\Omega&= \frac{4\pi}{3}r^2\mathbf{A}. \tag*{(27)} \end{aligned}

Equations (27) show that the spherical-integrals of terms that contain single rr vanish. Further, the terms that contain rr but are without ωs\omega_s would approach 0 in the limit rq→0r_q \to0.

Thus, by identifying and writing only non-vanishing terms from the product of equations (26) and (24), and inserting them into (16), LW 4-potentials due to the spinning point-charge in arbitrary motion can be evaluated by:

Aμ(Rμ)=lim⁡rq→0ρ4πϵ0c2∫rr2 dr∬Ω[[c;v]R{1(c−n^⋅v)+(r⋅⟨n^×ωs⟩)2(c−n^⋅v)3+(n⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)3+(n^⋅ωs)(n⋅r)(r⋅ωs)(c−n^⋅v)3−ωs2(n^⋅r)2(c−n^⋅v)3+3(n^⋅a)(n^⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)4}+[c;v]R2{3c2(n⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)4−2(r⋅v)(r⋅⟨n^×ωs⟩)(c−n^⋅v)3−2(n^⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)3−3v2(n^⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)4}−[0;r×ωs]R{(n^⋅r)(c−n^⋅v)2+(r⋅⟨n^×ωs⟩)(c−n^⋅v)2+(n^⋅a)(n^⋅r)(c−n^⋅v)3}+[0;r×ωs]R2{−c2(n^⋅r)(c−n^⋅v)3+(r⋅v)(c−n^⋅v)2+v2(n^⋅r)(c−n^⋅v)3}+[0;a]R{(n^⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)3+(n^⋅r)(r⋅⟨n^×ωs⟩)(c−n^⋅v)3}+1R{(n^⋅r)(r⋅ωs)(c−n^⋅v)2[0;ωs]−ωs2(n^⋅r)(c−n^⋅v)2[0;r]}] dΩ(28)\begin{aligned} A^\mu(R^\mu)=\lim_{r_q\to0}\frac{\rho}{4\pi\epsilon_0c^2}\int_r r^2\,dr\iint_{\Omega}\bigg[& \frac{[\mathbf{c};\mathbf{v}]}{R}\bigg\{ \frac{1}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})} +\frac{(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)^2}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}\\ &+\frac{(\mathbf{n}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} +\frac{(\hat{\mathbf{n}}\cdot\boldsymbol{\omega}_s)(\mathbf{n}\cdot\mathbf{r})(\mathbf{r}\cdot\boldsymbol{\omega}_s)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}\\ &-\frac{\omega_s^2(\hat{\mathbf{n}}\cdot\mathbf{r})^2}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} +\frac{3(\hat{\mathbf{n}}\cdot\mathbf{a})(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^4} \bigg\}\\ &+\frac{[\mathbf{c};\mathbf{v}]}{R^2}\bigg\{ \frac{3c^2(\mathbf{n}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^4} -\frac{2(\mathbf{r}\cdot\mathbf{v})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}\\ &-\frac{2(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} -\frac{3v^2(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^4} \bigg\}\\ &-\frac{[0;\mathbf{r}\times\boldsymbol{\omega}_s]}{R}\bigg\{ \frac{(\hat{\mathbf{n}}\cdot\mathbf{r})}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2} +\frac{(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2} +\frac{(\hat{\mathbf{n}}\cdot\mathbf{a})(\hat{\mathbf{n}}\cdot\mathbf{r})}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} \bigg\}\\ &+\frac{[0;\mathbf{r}\times\boldsymbol{\omega}_s]}{R^2}\bigg\{ -\frac{c^2(\hat{\mathbf{n}}\cdot\mathbf{r})}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} +\frac{(\mathbf{r}\cdot\mathbf{v})}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2} +\frac{v^2(\hat{\mathbf{n}}\cdot\mathbf{r})}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} \bigg\}\\ &+\frac{[0;\mathbf{a}]}{R}\bigg\{ \frac{(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} +\frac{(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3} \bigg\}\\ &+\frac{1}{R}\bigg\{ \frac{(\hat{\mathbf{n}}\cdot\mathbf{r})(\mathbf{r}\cdot\boldsymbol{\omega}_s)}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}[0;\boldsymbol{\omega}_s] -\frac{\omega_s^2(\hat{\mathbf{n}}\cdot\mathbf{r})}{(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}[0;\mathbf{r}] \bigg\}\bigg]\,d\Omega \tag*{(28)} \end{aligned}

Here, we have used the triple product identities: n^⋅⟨ωs×r⟩=r⋅⟨n^×ωs⟩\hat{\mathbf{n}}\cdot\langle\boldsymbol{\omega}_s\times\mathbf{r}\rangle=\mathbf{r}\cdot\langle\hat{\mathbf{n}}\times\boldsymbol{\omega}_s\rangle and ωs×⟨ωs×r⟩=(r⋅ωs)ωs−ωs2r\boldsymbol{\omega}_s\times\langle\boldsymbol{\omega}_s\times\mathbf{r}\rangle=(\mathbf{r}\cdot\boldsymbol{\omega}_s)\boldsymbol{\omega}_s-\omega_s^2\mathbf{r}.

Evaluating, spherical-integrals by using the set of equation (27) first, we get:

\begin{aligned} A^\mu(R^\mu)={}&\lim_{r_q\to0}\frac{\rho}{4\pi\epsilon_0c^2}\int_{r=0}^{r_q}4\pi r^2\,dr\left\{\frac{[c;v]}{R}\left(\frac{4\pi}{c-\hat n\cdot v}\right)+\frac{4\pi r^2(\hat n\times\omega_s)^2}{3(c-\hat n\cdot v)^3}+\frac{4\pi r^2(\hat n\cdot(\hat n\times\dot\omega_s))}{3(c-\hat n\cdot v)^3}+\frac{4\pi r^2(\hat n\cdot\omega_s)^2}{3(c-\hat n\cdot v)^3}\\ &-\frac{4\pi r^2\omega_s^2}{3(c-\hat n\cdot v)^3}+\frac{4\pi r^2(\hat n\cdot a)(\hat n\cdot(\hat n\times\omega_s))}{(c-\hat n\cdot v)^4}\right\}\\ &+\frac{[c;v]}{R^2}\left\{\frac{4\pi r^2c^2(\hat n\cdot(\hat n\times\dot\omega_s))}{(c-\hat n\cdot v)^4}-\frac{8\pi r^2(v\cdot(\hat n\times\omega_s))}{3(c-\hat n\cdot v)^3}-\frac{8\pi r^2(\hat n\cdot(v\times\omega_s))}{3(c-\hat n\cdot v)^3}-\frac{4\pi r^2v^2(\hat n\cdot(\hat n\times\omega_s))}{(c-\hat n\cdot v)^4}\right\}\\ &-\frac{[0;\hat n\times\omega_s]}{R}\left\{\frac{4\pi r^2(\hat n\cdot a)}{3(c-\hat n\cdot v)^3}\right\}\\ &+\frac{[0;\hat n\times\omega_s]}{R^2}\left\{-\frac{4\pi r^2c^2}{3(c-\hat n\cdot v)^3}+\frac{4\pi r^2v^2}{3(c-\hat n\cdot v)^3}\right\}\\ &+\frac{[0;a]}{R}\left\{\frac{4\pi r^2(\hat n\cdot(\hat n\times\omega_s))}{3(c-\hat n\cdot v)^3}+\frac{4\pi r^2(\hat n\cdot(\hat n\times\omega_s))}{3(c-\hat n\cdot v)^3}\right\}\\ &-\frac{[0;\hat n\times\dot\omega_s]}{R}\left\{\frac{4\pi r^2}{3(c-\hat n\cdot v)^2}\right\}+\frac{[0;\omega_s]}{R}\left\{\frac{4\pi r^2(\hat n\cdot\omega_s)}{3(c-\hat n\cdot v)^2}\right\}\\ &-\frac{[0;\hat n]}{R}\left\{\frac{4\pi r^2\omega_s^2}{3(c-\hat n\cdot v)^2}\right\}-\frac{[0;(\hat n\times\omega_s)\times\omega_s]}{R}\left\{\frac{4\pi r^2}{3(c-\hat n\cdot v)^2}\right\}\\ &+\frac{[0;v\times\omega_s]}{R^2}\left\{\frac{4\pi r^2}{3(c-\hat n\cdot v)^2}\right\} \tag*{(29)} \end{aligned}

All terms like n^⋅(n^×ωs)\hat n\cdot(\hat n\times\omega_s) vanish. Further, second term cancels-out with fourth and fifth, as (n^×ωs)2=ωs2−(n^⋅ωs)2(\hat n\times\omega_s)^2=\omega_s^2-(\hat n\cdot\omega_s)^2. The eighth term cancels-out with ninth. Lastly, the seventeenth and eighteenth terms cancel-out with nineteenth as (n^×ωs)×ωs=−ωs2n^+(n^⋅ωs)ωs(\hat n\times\omega_s)\times\omega_s=-\omega_s^2\hat n+(\hat n\cdot\omega_s)\omega_s.

Thus, writing and rearranging only the surviving term out of equation (29), we have:

\begin{aligned} A^\mu(R^\mu)={}&\lim_{r_q\to0}\frac{\rho}{4\pi\epsilon_0c^2}\int_{r=0}^{r_q}4\pi\left\{[c;v]\left(\frac{r^2}{R(c-\hat n\cdot v)}\right)\\ &-\frac{[0;\hat n\times\omega_s]}{3}\left\{\frac{r^4}{R(c-\hat n\cdot v)^2}+\frac{r^4(\hat n\cdot a)}{R(c-\hat n\cdot v)^3}+\frac{r^4(c^2-v^2)}{R^2(c-\hat n\cdot v)^3}\right\}\\ &+\frac{[0;v\times\omega_s]}{3}\left\{\frac{4\pi r^2}{R^2(c-\hat n\cdot v)^2}\right\}\right\}\,dr \tag*{(30)} \end{aligned}

Now, performing rr integrals and re-arranging the resultant terms, we get:

Aμ(Rμ)=lim⁡rq→0ρ4πϵ0c2(43πrq3){1R[c;v]c−n^⋅v−rq2[0;n^×ωs]5(c−n^⋅v)2−rq2(n^⋅a)[0;n^×ωs]5(c−n^⋅v)3+1R2{rq2[0;v×ωs]5(c−n^⋅v)2−rq2(c2−v2)[0;n^×ωs]5(c−n^⋅v)3}}(31)\begin{aligned} A^\mu(R^\mu)={}&\lim_{r_q\to0}\frac{\rho}{4\pi\epsilon_0c^2}\left(\frac{4}{3}\pi r_q^3\right)\left\{\frac{1}{R}\frac{[c;v]}{c-\hat n\cdot v}-\frac{r_q^2[0;\hat n\times\omega_s]}{5(c-\hat n\cdot v)^2}-\frac{r_q^2(\hat n\cdot a)[0;\hat n\times\omega_s]}{5(c-\hat n\cdot v)^3}\right.\\ &\left.+\frac{1}{R^2}\left\{\frac{r_q^2[0;v\times\omega_s]}{5(c-\hat n\cdot v)^2}-\frac{r_q^2(c^2-v^2)[0;\hat n\times\omega_s]}{5(c-\hat n\cdot v)^3}\right\}\right\} \tag*{(31)} \end{aligned}

In the point-particle limit rq→0r_q \to0, the total charge of the sphere 4/3πrq3ρ4/3\pi r_q^3 \rho becomes qq, the specific charge of the point-particle.

Further, in rest of the terms, quantity rq2ωsr_q^2\omega_s can be represented as the particle’s spin-angular-momentum s\mathbf{s}.

Spin-Angular-Momentum of a solid sphere is given by:

s=25mrq2ωs(32)\mathbf{s} = \frac{2}{5}mr_q^2\omega_s \tag*{(32)}

In point-particle limit rq→0r_q \to0, s\mathbf{s} is known as ’Classical-spin’ of the point-particle.

Thus, using equation (31) in point-particle limit, and writing equation (31) for Φ(R,t)\Phi(\mathbf{R},t) and A(R,t)\mathbf{A}(\mathbf{R},t) individually, we have:

Φ(R,t)=q4πϵ0Rcc−n^⋅v(33)\Phi(\mathbf{R},t)=\frac{q}{4\pi\epsilon_0R}\frac{c}{c-\hat{\mathbf{n}}\cdot\mathbf{v}} \tag*{(33)}

and:

A(R,t)=q4πϵ0c[1R{vc−n^⋅v−n^×s˙2m(c−n^⋅v)2}−(n^⋅a)(n^×s)2m(c−n^⋅v)3+1R2{v×s2m(c−n^⋅v)2−(c2−v2)(n^×s)2(c−n^⋅v)3}](34)\mathbf{A}(\mathbf{R},t)=\frac{q}{4\pi\epsilon_0c}\left[\frac{1}{R}\left\{\frac{\mathbf{v}}{c-\hat{\mathbf{n}}\cdot\mathbf{v}}-\frac{\hat{\mathbf{n}}\times\dot{\mathbf{s}}}{2m(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}\right\}-\frac{(\hat{\mathbf{n}}\cdot\mathbf{a})(\hat{\mathbf{n}}\times\mathbf{s})}{2m(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}+\frac{1}{R^2}\left\{\frac{\mathbf{v}\times\mathbf{s}}{2m(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}-\frac{(c^2-v^2)(\hat{\mathbf{n}}\times\mathbf{s})}{2(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}\right\}\right] \tag*{(34)}

The scalar LW potential remains the same as in equation (2). However, the vector LW potential arrives with several additional spin-dependent terms.

Interpretation:

A comparison of equations (33) and (34) with (2) reveals that the ‘original’ LW potentials are, the zeroth order terms caused only by the motion of a ‘non-spinning’ point-charge.

Further, a firsthand inspection of other terms of equation (34) appears to yield the following interpretation:

The second term in equation (34) contains time rate-of-change of spin. Thus, it is a transient term, referring to some mechanism by which the point-charge negotiates its own spin (orientation) with the system outside during transients, e.g. shifting orbits.

The third term in equation (34) indicates some sort of interaction between particle’s acceleration and it’s spin. The fourth term in equation (34) resembles the Coriolis acceleration term. However, in this case, the particle’s velocity interacts with its own spin rather than external revolving frame. We may call it “Self Coriolis Acceleration” term.

The fifth, and the last, term, simply, is the ’Magnetic-Dipole’ term that results from the multi-pole expansion of A(R,t)\mathbf{A}(\mathbf{R},t). [2]

However, the form of equation (34) and the numerous terms do not seem to be in meaningful pattern and need further investigation to properly comprehend.

Thus, to investigate further, we first formulate the equations (33) and (34) as sums of original LW potentials of equation (2) and acquired spin-dependent additional (correction) terms, as:

Φ(R,t)=Φ0(R,t)+δΦ(R,t,s)A(R,t)=A0(R,t)+δA(R,t,s)(35)\begin{aligned} \Phi(\mathbf{R},t) &= \Phi_0(\mathbf{R},t) + \delta\Phi(\mathbf{R},t,s) \\ \mathbf{A}(\mathbf{R},t) &= \mathbf{A}_0(\mathbf{R},t) + \delta\mathbf{A}(\mathbf{R},t,s) \tag*{(35)} \end{aligned}

While, Φ0(R,t)\Phi_0(\mathbf{R},t) and A0(R,t)\mathbf{A}_0(\mathbf{R},t) being the original LW potentials given by equations [2]; i.e. due to a ’non-spinning’ point-charge. Here ’0’ subscript indicates s=0s=0.

Then, the spin-dependent correction terms are:

δΦ(R,t,s)=0δA(R,t,s)=q4πϵ0c[1R{−(n^×s˙)2m(c−n^⋅v)2−(n^⋅a)(n^×s)2m(c−n^⋅v)3}+1R2{(v×s)2m(c−n^⋅v)2−(c2−v2)(n^×s)2(c−n^⋅v)3}](36)\begin{aligned} \delta\Phi(\mathbf{R},t,s) &= 0 \\ \delta\mathbf{A}(\mathbf{R},t,s) &= \frac{q}{4\pi\epsilon_0 c}\left[\frac{1}{R}\left\{-\frac{(\hat{\mathbf{n}}\times\dot{\mathbf{s}})}{2m(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}-\frac{(\hat{\mathbf{n}}\cdot\mathbf{a})(\hat{\mathbf{n}}\times\mathbf{s})}{2m(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}\right\}\right.\\ &\qquad\left.+\frac{1}{R^2}\left\{\frac{(\mathbf{v}\times\mathbf{s})}{2m(c-\hat{\mathbf{n}}\cdot\mathbf{v})^2}-\frac{(c^2-v^2)(\hat{\mathbf{n}}\times\mathbf{s})}{2(c-\hat{\mathbf{n}}\cdot\mathbf{v})^3}\right\}\right] \tag*{(36)} \end{aligned}

The expression for δA(R,t,s)\delta\mathbf{A}(\mathbf{R},t,s) from equation (36) may be re-written as:

δA(R,t,s)=q8πϵ0mc[−(R×s˙)(Rc−R⋅v)2−(R⋅a)(R×s)(Rc−R⋅v)3+(v×s)(Rc−R⋅v)2−(c2−v2)(R×s)(Rc−R⋅v)3](37)\delta\mathbf{A}(\mathbf{R},t,s)=\frac{q}{8\pi\epsilon_0mc}\left[-\frac{(\mathbf{R}\times\dot{\mathbf{s}})}{(Rc-\mathbf{R}\cdot\mathbf{v})^2}-\frac{(\mathbf{R}\cdot\mathbf{a})(\mathbf{R}\times\mathbf{s})}{(Rc-\mathbf{R}\cdot\mathbf{v})^3}+\frac{(\mathbf{v}\times\mathbf{s})}{(Rc-\mathbf{R}\cdot\mathbf{v})^2}-\frac{(c^2-v^2)(\mathbf{R}\times\mathbf{s})}{(Rc-\mathbf{R}\cdot\mathbf{v})^3}\right] \tag*{(37)}

The various terms may be re-arranged as:

δA(r,t,s)=q8πϵ0mc[(−RRc−R⋅v)×(s˙Rc−R⋅v)+{v+(c2−v2+R⋅a)(−RRc−R⋅v)}×{s(Rc−R⋅v)2}](38)\begin{aligned} \delta\mathbf{A}(\mathbf{r},t,s)={}&\frac{q}{8\pi\epsilon_0mc}\left[\left(\frac{-\mathbf{R}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\times\left(\frac{\dot{\mathbf{s}}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\right.\\ &\left.+\left\{\mathbf{v}+(c^2-v^2+\mathbf{R}\cdot\mathbf{a})\left(\frac{-\mathbf{R}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\right\}\times\left\{\frac{\mathbf{s}}{(Rc-\mathbf{R}\cdot\mathbf{v})^2}\right\}\right] \tag*{(38)} \end{aligned}

Using expression for gradient of retarded-time ∇tret\nabla t_{ret} [2]:

∇tret=−RRc−R⋅v(39)\nabla t_{ret}=\frac{-\mathbf{R}}{Rc-\mathbf{R}\cdot\mathbf{v}} \tag*{(39)}

And [2]:

∇(Rc−R⋅v)=(v2−c2−R⋅a)∇tret−v(40)\nabla(Rc-\mathbf{R}\cdot\mathbf{v})=(v^2-c^2-\mathbf{R}\cdot\mathbf{a})\nabla t_{ret}-\mathbf{v} \tag*{(40)}

Equation (38) may be written as:

δA(R,t,s)=q8πϵ0mc[∇tret×s˙(Rc−R⋅v)−∇(Rc−R⋅v)(Rc−R⋅v)2×s](41)\delta\mathbf{A}(\mathbf{R},t,s)=\frac{q}{8\pi\epsilon_0mc}\left[\frac{\nabla t_{ret}\times\dot{\mathbf{s}}}{(Rc-\mathbf{R}\cdot\mathbf{v})}-\frac{\nabla(Rc-\mathbf{R}\cdot\mathbf{v})}{(Rc-\mathbf{R}\cdot\mathbf{v})^2}\times\mathbf{s}\right] \tag*{(41)}

Now [2]:

∇×s=−(dsdtret)×∇tret=∇tret×s˙(42)\nabla\times\mathbf{s}=-\left(\frac{d\mathbf{s}}{dt_{ret}}\right)\times\nabla t_{ret}=\nabla t_{ret}\times\dot{\mathbf{s}} \tag*{(42)}

And [2]:

∇(1Rc−R⋅v)=−∇(Rc−R⋅v)(Rc−R⋅v)2(43)\nabla\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)=\frac{-\nabla(Rc-\mathbf{R}\cdot\mathbf{v})}{(Rc-\mathbf{R}\cdot\mathbf{v})^2} \tag*{(43)}

Making use of equations (42) and (43), equation (41) becomes:

δA(R,t,s)=q8πε0mc[∇×sRc−R⋅v+∇(1Rc−R⋅v)×s](44)\delta\mathbf{A}(\mathbf{R},t,\mathbf{s}) = \frac{q}{8\pi\varepsilon_{0}mc}\left[\frac{\nabla\times\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}+\nabla\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\times\mathbf{s}\right] \tag*{(44)}

Which finally simplifies to:

δA(R,t,s)=q8πε0mc∇×(sRc−R⋅v)(45)\delta\mathbf{A}(\mathbf{R},t,\mathbf{s}) = \frac{q}{8\pi\varepsilon_{0}mc}\nabla\times\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right) \tag*{(45)}

Thus, summing up equation (44) with (35), and along with (36) and (2), we have the final expression for LW potentials:

Φ(R,t)=qc4πε0[1Rc−R⋅v]A(R,t)=q4πε0c[vRc−R⋅v+12m∇×(sRc−R⋅v)](46)\begin{aligned} \Phi(\mathbf{R},t) &= \frac{qc}{4\pi\varepsilon_{0}}\left[\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right] \\ \mathbf{A}(\mathbf{R},t) &= \frac{q}{4\pi\varepsilon_{0}c}\left[\frac{\mathbf{v}}{Rc-\mathbf{R}\cdot\mathbf{v}}+\frac{1}{2m}\nabla\times\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\right] \tag*{(46)} \end{aligned}

As the quantity Rc−R⋅v=−RμuμRc-\mathbf{R}\cdot\mathbf{v}=-R_{\mu}u^{\mu} in denominator represents apparent geometry due to ’retardation’, the vector s/(Rc−R⋅v)\mathbf{s}/(Rc-\mathbf{R}\cdot\mathbf{v}) may be referred to as “apparent spin”: the classical-spin that is ’perceived’ by a distant observer.

Therefore, the correction terms obtained in A(R,t)\mathbf{A}(\mathbf{R},t) are collectively proportional to the ’rotation’ (curl) of the apparent spin.

However, in view of equations (46) there is an another important observation, that the previously known coupling between LW scalar and vector potentials would no longer hold; i.e.:

A(R,t)≠vc2Φ(R,t)(47)\mathbf{A}(\mathbf{R},t)\ne\frac{\mathbf{v}}{c^{2}}\Phi(\mathbf{R},t) \tag*{(47)}

Rather, another intuitive relation between LW scalar and vector potentials appears:

n^⋅A(R,t)=n^⋅vc2Φ(R,t)(48)\hat{\mathbf{n}}\cdot\mathbf{A}(\mathbf{R},t)=\frac{\hat{\mathbf{n}}\cdot\mathbf{v}}{c^{2}}\Phi(\mathbf{R},t) \tag*{(48)}

Or, in other words, only the ’parallel’ components of A(R,t)\mathbf{A}(\mathbf{R},t) and v\mathbf{v}, in the direction of faraway observer satisfy:

A∥(R,t)=v∥c2Φ(R,t)(49)A_{\parallel}(\mathbf{R},t)=\frac{v_{\parallel}}{c^{2}}\Phi(\mathbf{R},t) \tag*{(49)}

In upcoming section we shall check the validity of equations (46).

Validation:

Lié​nard-Wiechert potentials are Retarded-Potentials, which being in Lorenz Gauge, satisfy Lorenz Gauge condition [2] [3]:

∇⋅A(R,t)=−1c2∂Φ(R,t)∂t(50)\nabla\cdot\mathbf{A}(\mathbf{R},t)=-\frac{1}{c^{2}}\frac{\partial\Phi(\mathbf{R},t)}{\partial t} \tag*{(50)}

In addition, they also satisfy the homogeneous wave-equations in vacuum [2] [3]:

□2Φ(R,t)=∇2Φ(R,t)−1c2∂2Φ(R,t)∂t2=0□2A(R,t)=∇2A(R,t)−1c2∂2A(R,t)∂t2=0(51)\begin{aligned} \Box^{2}\Phi(\mathbf{R},t) &= \nabla^{2}\Phi(\mathbf{R},t)-\frac{1}{c^{2}}\frac{\partial^{2}\Phi(\mathbf{R},t)}{\partial t^{2}} = 0\\ \Box^{2}\mathbf{A}(\mathbf{R},t) &= \nabla^{2}\mathbf{A}(\mathbf{R},t)-\frac{1}{c^{2}}\frac{\partial^{2}\mathbf{A}(\mathbf{R},t)}{\partial t^{2}} = 0 \tag*{(51)} \end{aligned}

If the derived LW potentials of equations (46) are valid, they must also meet the criteria of equations (50) and (51).

The part of equations (35), Φ0(R,t)\Phi_{0}(\mathbf{R},t) and A0(r,t)\mathbf{A}_{0}(\mathbf{r},t) are simply the original LW potentials, thus they already satisfy (50) and (51).

Further, as there is no correction term in Φ(R,t)\Phi(\mathbf{R},t) (i.e. δΦ(R,t,s)=0\delta\Phi(\mathbf{R},t,s)=0) according to equation (46), the only test remains is for δA(R,t,s)\delta\mathbf{A}(\mathbf{R},t,s), which must satisfy equations (50) and (51).

As from equation (36), δA(R,t,s)\delta\mathbf{A}(\mathbf{R},t,s) being the curl of a vector; its divergence vanishes. And, since, δΦ(R,t,s)=0\delta\Phi(\mathbf{R},t,s)=0, it immediately satisfies the Lorenz gauge condition:

∇⋅δA(R,t)=0=−1c2∂δΦ(R,t)∂t(52)\nabla\cdot\delta\mathbf{A}(\mathbf{R},t)=0=-\frac{1}{c^{2}}\frac{\partial\delta\Phi(\mathbf{R},t)}{\partial t} \tag*{(52)}

Now, writing the L.H.S. Of the vacuum homogeneous wave-equation for δA(R,t,s)\delta\mathbf{A}(\mathbf{R},t,s) and using equation (45), we have:

∇2δA−1c2∂2δA∂t2=q8πϵ0mc⟨∇2−1c2∂2∂t2⟩[∇×(sRc−R⋅v)](53)\nabla^{2}\delta\mathbf{A}-\frac{1}{c^{2}}\frac{\partial^{2}\delta\mathbf{A}}{\partial t^{2}}=\frac{q}{8\pi\epsilon_{0}mc}\left\langle\nabla^{2}-\frac{1}{c^{2}}\frac{\partial^{2}}{\partial t^{2}}\right\rangle\left[\nabla\times\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\right] \tag*{(53)}

Curl and Laplacian operators are orthogonal and thus are interchangeable. Also the curl operator, being a space-derivative, is independent of time; so the curl operator and time-derivative are also interchangeable. Therefore, equation (53) may be written as:

∇2δA−1c2∂2δA∂t2=q8πϵ0mc∇×[⟨∇2−1c2∂2∂t2⟩(sRc−R⋅v)](54)\nabla^{2}\delta\mathbf{A}-\frac{1}{c^{2}}\frac{\partial^{2}\delta\mathbf{A}}{\partial t^{2}}=\frac{q}{8\pi\epsilon_{0}mc}\nabla\times\left[\left\langle\nabla^{2}-\frac{1}{c^{2}}\frac{\partial^{2}}{\partial t^{2}}\right\rangle\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\right] \tag*{(54)}

Equation (54) says that if the vector s/(Rc−R⋅v)\mathbf{s}/(Rc-\mathbf{R}\cdot\mathbf{v}) satisfies the vacuum homogeneous wave-equation, δA(r,t,s)\delta\mathbf{A}(\mathbf{r},t,s) would also satisfy it.

Thus, for δA(r,t,s)\delta\mathbf{A}(\mathbf{r},t,s) to be valid:

∇2(sRc−R⋅v)−1c2∂2∂t2(sRc−R⋅v)=0(55)\nabla^{2}\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)-\frac{1}{c^{2}}\frac{\partial^{2}}{\partial t^{2}}\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)=0 \tag*{(55)}

Solving the L.H.S. of equation (55):

⟨∇2−1c2∂2∂t2⟩(sRc−R⋅v)=1(Rc−R⋅v)(∇2s−1c2∂2s∂t2)+s⟨∇2−1c2∂2∂t2⟩(1Rc−R⋅v)+2(∂s∂tret){(∇tret)⋅∇(1Rc−R⋅v)−1c2(∂tret∂t)∂∂t(1Rc−R⋅v)}(56)\begin{aligned} \left\langle\nabla^{2}-\frac{1}{c^{2}}\frac{\partial^{2}}{\partial t^{2}}\right\rangle\left(\frac{\mathbf{s}}{Rc-\mathbf{R}\cdot\mathbf{v}}\right) &=\frac{1}{(Rc-\mathbf{R}\cdot\mathbf{v})}\left(\nabla^{2}\mathbf{s}-\frac{1}{c^{2}}\frac{\partial^{2}\mathbf{s}}{\partial t^{2}}\right)\\ &+\mathbf{s}\left\langle\nabla^{2}-\frac{1}{c^{2}}\frac{\partial^{2}}{\partial t^{2}}\right\rangle\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\\ &+2\left(\frac{\partial\mathbf{s}}{\partial t_{ret}}\right)\left\{\left(\nabla t_{ret}\right)\cdot\nabla\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)-\frac{1}{c^{2}}\left(\frac{\partial t_{ret}}{\partial t}\right)\frac{\partial}{\partial t}\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)\right\} \tag*{(56)} \end{aligned}

The middle term vanishes, because the function 1/(Rc−R⋅v)1/(Rc-\mathbf{R}\cdot\mathbf{v}) defines the geometry of original Scalar LW potential of equation (2); thus satisfies homogeneous vacuum wave equation:

[∇2−1c2∂2∂t2]Φ(R,t)=[∇2−1c2∂2∂t2][q4πϵ0(Rc−R⋅v)]=qc4πϵ0[∇2−1c2∂2∂t2](1Rc−R⋅v)=0(57)\begin{aligned} \left[\nabla^2-\frac{1}{c^2}\frac{\partial^2}{\partial t^2}\right]\Phi(\mathbf{R},t)=\left[\nabla^2-\frac{1}{c^2}\frac{\partial^2}{\partial t^2}\right]\left[\frac{q}{4\pi\epsilon_0\left(Rc-\mathbf{R}\cdot\mathbf{v}\right)}\right] \\ &=\frac{qc}{4\pi\epsilon_0}\left[\nabla^2-\frac{1}{c^2}\frac{\partial^2}{\partial t^2}\right]\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)=0 \tag*{(57)} \end{aligned}

Further, it may be shown via evaluation of various space and time derivatives that:

∇2s−1c2∂2s∂t2=−2(Rc−R⋅v)∂s∂tret(58)\nabla^2s-\frac{1}{c^2}\frac{\partial^2s}{\partial t^2}=\frac{-2}{\left(Rc-\mathbf{R}\cdot\mathbf{v}\right)}\frac{\partial s}{\partial t_{\mathrm{ret}}} \tag*{(58)}

And:

(∇tret)⋅∇(1Rc−R⋅v)−1c2(∂tret∂t)∂∂t(1Rc−R⋅v)=1(Rc−R⋅v)2(59)\left(\nabla t_{\mathrm{ret}}\right)\cdot\nabla\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)-\frac{1}{c^2}\left(\frac{\partial t_{\mathrm{ret}}}{\partial t}\right)\frac{\partial}{\partial t}\left(\frac{1}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)=\frac{1}{\left(Rc-\mathbf{R}\cdot\mathbf{v}\right)^2} \tag*{(59)}

Thus, with use of equations (58), (59) and (57) in (56), we get:

⟨∇2−1c2∂2∂t2⟩(sRc−R⋅v)=0(60)\left\langle\nabla^2-\frac{1}{c^2}\frac{\partial^2}{\partial t^2}\right\rangle\left(\frac{s}{Rc-\mathbf{R}\cdot\mathbf{v}}\right)=0 \tag*{(60)}

Therefore, from equation (60) and (54), it is proved that δA(R,t,s)\delta\mathbf{A}(\mathbf{R},t,s) satisfies vacuum homogeneous wave-equation.

Thus, the derived expressions (34) and (46) for LW potentials due to a spinning and non-relativistically moving point-charge, are valid in the context of classical electrodynamics.

Conclusion:

In this paper LW potentials due to a spinning point-charge in arbitrary motion are derived, beginning with a spinning and non-relativistically moving charged-sphere and subsequently reducing its dimension to the ‘point-charge’ limit.

Though, expression for LW scalar-potential stays the same, several, spin-dependent correction terms are discovered in the expression for the LW vector-potential (34) that coalesce into a single correction term in equation (46), which is proportional to the ‘rotation’ of apparent classical-spin.

However, equation (47) also reveals that the previously known v/c2v/c^2 coupling between the scalar and vector LW potentials is not valid generally; but only for the components that lie in the observer’s direction.

The importance of the acquired correction term(s) needs to be further studied by deriving electromagnetic-fields and radiated power from them. Though, the same is not in this paper’s scope.

We believe that these ‘spin’ dependent terms, in LW vector-potential may alter (or extend) the physics of EM fields and radiation (from a point-charge); nevertheless, further investigation is required.

One observation is quite intriguing: in Quantum-Mechanics spin of elementary charged-particles are seen as h/2h/2 (or odd multiple of h/2h/2); while hh being the Plank’s constant, which bears the dimensions of angular-momentum. Whereas, in our expressions (34) and (46), additional terms in LW potentials are proportional to s/2s/2. Thus we believe the work of this paper will be crucial in establishing some link between the Classical and Quantum Physics.

Lastly, in this paper, the LW potentials of equations (46) are the outcomes of series approximation in the non-relativistic case; where only the terms up to r2r^2 are examined. However, to acquire a better understanding of the classical-spin dependence of potentials (and fields), complete analytical solution needs to be evaluated.

Data availability: No data has been analyzed or generated, because, our work is a purely theoretical and mathematical approach.

References

  1. [1]Rohrlich F. “Causality, the Coulomb field, and Newton’s law of gravitation.” In: American Journal of Physics 70.4 (2002), pp. 411–414.DOI
  2. [2]Griffiths D. “Introduction to Electrodynamics”. 3rd ed. New Delhi, India: PHI Learning Pvt. Ltd., 2011, pp. 242–470.
  3. [3]Jackson J. D. “Classical Electrodynamics”. New York, USA: John Wiley & Sons, Inc., 1962, pp. 464–589.
  4. [4]Panofsky W.K.H. and Phillips M. “Classical Electricity and Magnetism”. 2nd ed. Reading, MA: Addison-Wesley, 1962, pp. 341–343.
  5. [5]Vanderlinde J. “Classical Electromagnetic Theory”. 2nd ed. Dordrecht: KAP, 2004, pp. 294–296.
  6. [6]Arakelyan S.A., Grigoryan G.V., and Grigoryan R.P. “Lienard-Wiechert potentials and synchrotron radiation from a relativistic spinning particle in pseudo-classical theory”. In: Physics of Atomic Nuclei 63 (2000), pp. 2115–2122.
  7. [7]Lorentz H. A. “Theory of Electrons”. 2nd ed. New York, USA: G. E. Stechert & Co., 1916, p. 252.
  8. [8]Mermin Devid N. “Space & Time in Special Relativity”. New York, USA: McGraw-Hill, 1968, pp. 63–170.
  9. [9]Jerrold Franklin. “Lorentz contraction, Bell’s spaceships and rigid body motion in special relativity”. In: European Journal of Physics 31 (2010), pp. 291–298.DOI
  10. [10]Aniket Basu, R.S. Saraswat, Kedar B. Khare, G.P. Sastry, and Sougato Bose. “Tippe top paradox in relativity”. In: European Journal of Physics 23 (2002), pp. 295–305.DOI
  11. [11]Wikipedia contributors. “Speeds of sound of the elements — Wikipedia, The Free Encyclopedia”. 2024. URL: https://en.wikipedia.org/w/index.php?title=Speeds_of_sound_of_the_elements&oldid=1226

Paper details

vlnc:W_z9r6cd6a
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents