The Modified Electromagnetism and The Emergent Longitudinal Wave
The Modified Electromagnetism and The Emergent Longitudinal Wave
1
Department of Physics, Faculty of Science, University of Khartoum, P.O.
arXiv:1403.2687v2 [physics.gen-ph] 25 Oct 2018
Abstract
The classical theory of electromagnetism has been revisited and the possibil-
ity of longitudinal photon wave is explored. It is shown that the emergence
of longitudinal wave is a consequence of Lorenz gauge (condition) violation.
Proca, Vlaenderen & Waser and Arbab theories are investigated. The dif-
ferent approaches are compared to each other and the relevant equations are
combined. The telegrapher’s equation can be obtained and with a specific
choice of the function a Klein-Gordon equation of massive scalar field can be
obtained. When the Lorentz gauge is violated by introducing the first order
time derivative the emergence of the photon mass and the relevant longitu-
dinal wave for the electromagnetic wave is apparnt.
1 Introdution
Maxwell’s equations are one of the biggest edifices that mankind had come to
establish. Maxwell unified electromagnetism with light in a set of consistent
equation dubbed in Maxwell equation. These equation prematurely predicted
that speed of light in vacuum is constant. In an elegant theory, Salam,
Weinberg and Glashow unified electromagnetism with weak interaction [1, 2,
3]. The ultimate goal of unifying all interactions into a single universal theory
1
is underway. Some theoretical prejudices advocated by Proca in extending
Maxwell’s theory to encompass massive photons was expounded in 1967 [4].
This later theory will be of importance, at cosmic scale, as some compelling
investigation usher in that direction. Of these investigation is the missing
mass (dark matter) in the universe. It is remarked by Schrodinger that if the
photon had a mass then the Black-Body formula will be modified and the
energy density of these photons would be 3/2 times the ordinary one [5]. This
is because these photon would have a longitudinal polarization besides the
transverse one. Longitudinal waves have been shown to have high penetration
power. This character rendered them to be of great importance and will find
wider applications in communication systems. Some endeavors to generalize
Maxwell’s equations to remedy these shortcoming have been proposed by
following a quaternionic generalizations. In the framework of these attempts,
the shortcoming present in the original Maxwell’s formulation can be avoided
[6]. In this work, we will outline the different approaches and explore these
interrelations. We will see that whenever the Lorenz gauge (condition) is
violated, longitudinal waves will be inevitable. We also analyze the status of
the experimental evidence supporting the presence of these waves.
2 Maxwell’s equation
Maxwell’s equations describe the behavior of the electric and magnetic fields.
These equations are [7]
∇ ~ = ρ ,
~ ·E ~ ·B
∇ ~ = 0, (1)
ε0
and
~ ~
∇ ~ = − ∂B ,
~ ×E ~ = µ0 J~ + 1 ∂ E .
~ ×B
∇ (2)
∂t c2 ∂t
2
3 Vector and scalar potential
The electric and magnetic fields are defined as
~
E ~ − ∂A ,
~ = −∇ϕ ~ =∇
B ~ ×A
~. (3)
∂t
~ and ϕ, are more fun-
It is believed that vector and scalar potentials, A
damental than E ~ and B.
~ For this reason, we describe the electromagnetic
interaction in terms of these potentials, rather than the electric and magnetic
fields. This is apparent, at the quantum level, in the Aharonov-Bohm effect
where when two beams of electrons are sent through region with same B, ~
~ experience an interference pattern when collected at the de-
but different A,
tector [8]. This occurs because the phase factor of the electron wavefunction
depends on A.~
Substituting eq.(3) in (1) and (2), we obtain [7]
1 ∂2ϕ
2 ∂ ~ ~ 1 ∂ϕ ρ
2 2
−∇ ϕ− ∇·A+ 2 = , (4)
c ∂t ∂t c ∂t ε0
and
1 ∂2A~
1 ∂ϕ
2~ ~ ∇ ~ ·A
~+
−∇ A+∇ = µ0 J~ . (5)
c2 ∂t2 c2 ∂t
These are the Maxwell’s equations in terms of the vector and scalar poten-
tials.
∇ ~ + 1 ∂ϕ = 0 .
~ ·A (7)
c2 ∂t
3
In this case one has the condition that
1 ∂Λ
2
− ∇2 Λ = 0 . (8)
c ∂t
It is interesting to see that the electric and magnetic fields can not be uniquely
specified by A ~ and ϕ. For this reason these two fields are treated as mere
~ and B
mathematical constructs and bear no physical meaning, but E ~ do.
∇ ~ + 1 ∂ϕ = S .
~ ·A (9)
c2 ∂t
Accordingly, they obtain the following equations
∇ ~ = ρ − ∂S ,
~ ·E ~ ·B
∇ ~ = 0. (10)
ε0 ∂t
and
~ ~
∇ ~ = − ∂B ,
~ ×E ∇ ~ = µ0 J~ + 1 ∂ E + ∇S.
~ ×B ~ (11)
∂t c2 ∂t
The scalar S satisfies the wave equation
1 ∂2S
2 ∂ρ ~ ~
− ∇ S = µ0 +∇·J = 0, (12)
c2 ∂t2 ∂t
upon using the continuity equation. Besides this, S is found to represent an
~ i.e., along the electric field
electroscalar wave having a Poynting vector S E,
direction.
4
5 Proca’s equations
Proca generalized Maxwell’s equations by introducing a massive photon filed.
The resulting equations are [4]
∇ ~ = ρ − µ2 ϕ ,
~ ·E ~ ·B
∇ ~ = 0, (13)
γ
ε0
and
~ ~
∇ ~ = − ∂B ,
~ ×E ∇ ~ = µ0 J~ + 1 ∂ E − µ2 A
~ ×B ~. (14)
γ
∂t c2 ∂t
5
7 The analogy among these paradigms
1
Comparing eqs.(10) & (11) with eqs.(13) & (14) yields
∂S ~ = ∇S
~ .
µ2γ ϕ = − , µ2γ A (17)
∂t
Differentiation partially the first equation w.r.t. t, taking the divergence of
the second equation and subtracting the resulting two equations yield
1 ∂2S
− ∇2 S + µ2γ S = 0 . (18)
c2 ∂t2
This is the familiar Klein-Gordon equation of massive spin zero particle.
Thus, if the photon had a non-zero mass, as hypothesized by Proca, then
this photon wave must be governed by eq.(18).
Integrating the second equation in eq.(17) yields
Z
S = µ2γ A ~ .
~ · dr (19)
~ = ∇
~ · dr ~ ×A
~ · dS
~ = φm , where φm is the
R R
But from Stokes theorem, A
magnetic flux, one has
S = µ2γ φm + f (t) . (20)
6
now the scalar and vector potentials satisfy the matter wave equation, i.e.,
eq.(15), then
∇ ~ + 1 ∂ϕ = − 2m ϕ .
~ ·A (22)
c2 ∂t h̄
This equation can be compared with eq.(9), in which case, S = − 2m
h̄
ϕ. If
we compare eqs.(9), (17), (18) and (22) we will get
1 ∂2ϕ
2 2
− ∇2 ϕ + µ2γ ϕ = 0 . (23)
c ∂t
This implies that the scalar potential of the massive photon, ϕ, satisfies the
Klein-Gordon equation.
Now differentiating partially eq.(22) with respect to time and using (3) yield
1 ∂2ϕ 2m ∂ϕ ~ ~
2 2
− ∇2 ϕ + − ∇·E = 0. (24)
c ∂t h̄ ∂t
Taking the gradient of eq.(22) and using eq.(3) and the second equation in
eq.(2), we obtain
1 ∂2A~ ~ 2m
2m ∂ A
2~ ~ − µ0 J~ = 0 ,
− ∇ A + + E (25)
c2 ∂t2 h̄ ∂t h̄
~ × (∇
where we have used the vector identity, ∇ ~ × A)
~ = ∇(
~ ∇~ · A)
~ − ∇ 2 A.
~
It is interesting to see that if the photon had a mass (m 6= 0) in vacuum,
then the Lorenz gauge in eq.(22) would not be satisfied. This encourages us
to associate a non-zero mass of the photon to the breaking of Lorenz gauge.
In vacuum, ρ = 0 and J~ = 0, so that eqs.(24) and (25), using eq.(3), become
1 ∂2ϕ 2m ∂ϕ
2 2
− ∇2 ϕ + = 0. (26)
c ∂t h̄ ∂t
Taking the gradient of eq.(22) and using eq.(3) and the second equation in
eq.(2), we obtain
1 ∂2A~
~ − 2m ∇ϕ
− ∇2 A ~ = 0. (27)
2
c ∂t2 h̄
Therefore, eqs.(26) and (27) are the corresponding equations describing mas-
sive photon in vacuum. Adding the time derivative of eq.(27) to the gradient
7
~ and B
of eq.(26), using eq.(3), imply that both E ~ travel at speed of light,
viz.,
1 ∂2E~ ~
1 ∂2B
2~ ~ = 0.
− ∇ E = 0 , − ∇2 B (28)
c2 ∂t2 c2 ∂t2
This is a quite surprising result. It is evident from eqs.(26) and (27) that
spatial and temporal variations of the scalar field is the source of the ϕ and
~ waves.
A
Now apply the gauge transformation, eq.(6), in eq.(22) to get
1 ∂2Λ 2m ∂Λ
2 2
− ∇2 Λ + = 0. (29)
c ∂t h̄ ∂t
This equation is know as the Telegrapher’s equation. It describes the electric
signals in telephone and telegraph lines. However, by making the substitu-
2 t/h̄)
tion, Λ(r, t) = e−(mc χ(r, t), eq.(29) can be written as
1 ∂2χ 2 m2 c2
− ∇ χ + χ = 0. (30)
c2 ∂t2 h̄2
which is the ordinary Klein-Gordon equation of massive scalar field χ. This
equation is a consequence of gauge invariance.
~′ =A
~ + ∇Λ
~ , ∂Λ
A ϕ′ = ϕ− , (31)
∂t
and
∂Λ
J~ ′ = J~ + µ2γ ε0 c2 ∇Λ
~ , ρ ′ = ρ − µ2γ ε0
. (32)
∂t
Under these gauge transformations, Proca’s equations become gauge invari-
ant and that the mass term is no longer problematic. Moreover, these trans-
formations leave the system of the continuity equations invariant as well, as
long as Λ satisfies the wave equation in eq.(8) [11].
8
8 The status of experimental evidence of lon-
gitudinal wave and photon mass
There are some appealing experiments that amount to the existence of lon-
gitudinal wave. An electric longitudinal wave has been observed by Miyaji
et al. while studying the transverse wave[12]. This wave is characterized by
the vanishing magnetic component.
The enormous success in quantum electrodynamics has lead to the acceptance
of the concept of massless photons. Also, the nonzero photon mass would
give rise to a wavelength dependence of the speed of light in free space, a
more rapid (exponential or Yukawa) falloff of magnetic dipole fields with dis-
tance than the usual inverse cube dependence [13]. This could be attributed
to the possibility of existing longitudinal electromagnetic waves. This wave
is characterized by the vanishing magnetic and electric field components [14].
A massive photon is equivalent to a violation of Coulomb’s Law [15]. Hence,
a Yukawa factor e−r/λ in the 1/r electrostatic and magnetostatic potential
terms would appear. The magnetic field term will contain correction terms
related to the de Broglie wavelength λγ = h̄/mγ c. Another way to test the
existence of the photon mass and hence the longitudinal waves is the limit
of the power in the inverse of the distance distance in Coulomb law, namely
1/r 2+ε . One of the consequences of ε 6= 0 is the decrease of the electromag-
netic waves as the wavelength increases which is due to the smallness of ε.
A limit of ε = 5.1 × 10−20 was set by Lou et al [16]. According to the un-
certainty principle the upper limit on the photon mass can be obtained from
mγ ≈ h̄/(∆t c2 ). Using the age of the universe as 1010 years one obtains
mγ ≈ 10−66 g.
De Broglie noticed that photon mass leads to a larger speed of violet light
than that of the red one [17]. He concluded that during the eclipse in a
double star system the color of the appearing star would change from violet
to red. On the other hand, Schrodinger pointed out that magnetic field of
9
the Earth would be exponentially cut off at distances of the order of the
photon Compton wave length [18]. From the observed altitude of auroras he
concluded that λγ > 104. Other data have shown more limits also.
Because of limitations involved in experiments performed on the Earth, as-
trophysical measurement could promise better limits on photon masses. One
of the consequences of photons being massive is the dispersion of star light.
According to measurements performed on planet Jupiter the upper limit of
Compton wavelength is λγ = 3.14 × 1011 cm [19]. Based on analyzing the
Earth’s magnetic field the upper limit on photon mass is mγ = 1 × 10−48 g
[20]. Also from geomagnetic field Fischbach et al. found a limit on the upper
mass of the photon as mγ ≤ 10−48 g [21]. According to Proca equation, the
velocity of light depends on its frequency. From the measurement on the
radio signals from the Crab Nebula it was found that the upper limit on the
mass of the photon is mγ ≤ 10−44 g [22]. Also, hydrodynamic waves coupling
to the interplanetary magnetic fields could be useful in finding the mass of
the photon. Ryutov (1997) set a limit of mγ ≤ 10−51 g based on these studies
[23]. Other limits on the photon mass are obtained from studying the current
density and plasma of the interstellar media (mγ ≤ 10−53 g) [24].
Modern theories that appeal to the concept of spontaneous symmetry break-
ing assume that particles, which are massless above a certain critical tem-
perature Tc , acquire mass below this temperature. This can be applied to
photons. From the cryogenic photon mass experiment performed by Ryan et
al an upper limit mass on photons were found to be mγ = 1.5 × 10−42 . From
the symmetry breaking point of view and the experiment, one can speculate
that photons can be massless above a certain temperature [25].
Superconductivity is characterized by perfect conductivity and perfect dia-
magnetism. BCS theory can explain both of these fundamental properties of
superconductors [26]. However, the London equations can be easily derived
and used to understand the expulsion of flux within the material without
appealing to the broader BCS theory [27]. Such a derivation relies solely
10
on postulating that the Cooper pairs in the superconducting state maintain
their ground state momentum of p = 0. It has been possible to identify
an effective photon mass that is directly related to the London penetration
depth of the applied field [28].
In 2008 a Russian group performed a measurement on the volt value triggered
by the radiation coming from solar eclipse. It was noticed the registered
signal does not depend on the orientation of the tourmaline plates used in
front of the detector [29]. This could be an indication of the longitudinal
wave component and, hence, the non-zero mass of the photon.
Another experimental evidence for the existence of the massive photons is
the Tesla coil. The coil is a device which produces high frequency alternating
current with high voltage and low current. The coil is used to produce energy
which can be transmitted to a device without any wires. A neon light can lit
without wiring connection. This can be explained in terms of the longitudinal
component of the light wave.
9 Conclusion
In this paper, the possibility of longitudinal photon wave is explored from
the electromagnetic theory. We revisited Proca, Vlaenderen & Waser and
Arbab theories. We studied the equations obtained from these approaches
and combined them together. The different approaches are compared to
each other and the relevant equations are combined. It is shown, then, that
if the Lorentz gauge invariance is violated the photon can gain mass and the
longitudinal part of the electromagnetic wave can appear.
11
References
[1] Glashow, S. L, Nucl. Phys,22 (1961) 579.
[5] Bass, L. and Schrodinger, E, Must the Photon Mass be Zero?, Proc. R.
Soc. London A, 232 (1955) 1.
12
[12] Godai Miyaji, Noriaki Miyanaga, Koji Tsubakimoto, Keiichi Sueda, and
Ken Ohbayashi, Intense longitudinal electric elds generated from trans-
verse electromagnetic waves, Appl.Phys. Lett, 84 (2004) 3855.
[19] M. A. Gintsburg, Astron. 40, 703 (1963); Sov. Astron. AJ 7, 563 (1963).
[22] Liang-Cheng Tu, Jun Luo and G. T. Gillie, The mass of the photon,
Institute of Physics Publishing Rep. Prog. Phys. 68 (2005) 77.
13
[24] D. D. Ryutov, Using plasma physics to weigh the photon, 2007 Plasma
Phys. Control. Fusion 49 B 429.
[25] Ryan, J. J., Accetta, F. and Austin, R. H. 1985 Phys. Rev.D 32 802.
[28] M. Tajmar, C.J. Matos, Local Photon and Graviton Mass and its Con-
sequences, arxiv.org/pdf/gr-qc/0603032.
14