To provide an analysis of the above equation for gravitational waves and the analysis of the resulting classes of metrics, a representation using null-vectors will be useful. Therefore, in the next few steps it will be shown that Alena Tensor allows representing the energy-momentum tensor of the electromagnetic field with the use of two null-vectors.
3.1. Decomposition of the Electromagnetic Field Using Null Vectors
At first step one may recall equation (
15) and define new four-vector
obtaining
Since it is know from previous publications, that
and
, therefore above definition also yields
. This property allows to represent
and
using two null-vectors
and
as follows
thus
Next, one may define auxiliary parameter
as
and subtract the linear combination of
and
from both sides
Next, using
simplicity, one may recall from [
9] coefficients related to the electromagnetic field
relative permeability
volume magnetic susceptibility
relative permittivity
electric susceptibility
and notice, that one obtains Alena Tensor
as
where electromagnetic stress-energy tensor is equal to
and where
actually simplifies, as shown in introduction, to
Completing the definition of the first invariant of the electromagnetic field tensor
, one may define the second invariant
by electric
and magnetic
fields as
where it is known [
28], that
Therefore from (
25) one obtains simplifications
and by defining a useful auxiliary variable
one gets
Finally, defining for simplicity as below
then calculating
and expressing
, one gets further useful expressions
To simplify further analysis, one may also normalize four-vectors
and
using (
29) as follows
where
was introduced to avoid confusion related to the previously defined perturbation
. After few calculations using previously derived relationships in (
25)
one may now rewrite the electromagnetic field tensor
as
As shown in [
9] element
is responsible for electric field energy density carried by light, where
was shown as describing energy density of magnetic moment and was linked to charged matter in motion. The element
is a new term and part of equations related to this term may be expressed as
with help of (
33) as
Since
does not actually carry energy but only momentum, it can be associated with some description of spin field effects by analogy to (
20). Using (
29), (
30) (
33) and (
36), electromagnetic field tensor
may be, however, expressed in more useful form. Since
thus
Since the variables
and
are merely auxiliary variables (used only to highlight certain relationships) and can be expressed in terms of
one may further simplify the description of the system. As one may notice, (
34) also allows to simplify (
19) by introducing
what using (
30) yields
and therefore allows to analyze the system using hyperbolic (and trigonometric) functions
One may thus denote normalized Alena Tensor in flat spacetime as
with help of (
24)
and notice, that it may be presented after simple calculations using (
42) - (
45) as
Since the expression in the brackets must be equal to
, therefore
Therefore, according to (
37), (
38) there must occur
. Indeed, with help of (
30) and (
33) one obtains
and the same equality may be calculated for
what confirms compliance with (
40). This shows that further, in-depth analysis of the system is also possible, however, modeling and simplifying the description of the electromagnetic field or searching for elementary particles that provide stable solutions requires a separate article (probably several articles). From the perspective of describing gravitational waves, other elements of the description are crucial, which will be discussed next.
Finally, one may notice, that property
in (
34) requires analysis in a complex basis. An example of such a basis are four-vectors
where the angle
was introduced to facilitate further analysis. A cursory examination shows that this basis describes the electromagnetic field very well indeed. It has a good representation in conformal geometry (null vectors correspond to points on the equator of the Penrose sphere), where the propagation directions are perpendicular to the time axis (purely spatial), ideal for describing a circularly or elliptically polarized wave in the direction of Poyting vector
, where
represents the constant phase relation between the electric and magnetic fields. The proposed basis naturally enters the Newman–Penrose formalism, allows for a full spinor representation of the electromagnetic field, where
is a typical massless wave, satisfies the wave equation
, and since it provides a spin-helicity of +/- 1, it is well suited for further analysis in the QFT framework as a photon wave representation, describing a single-particle state. However, detailed analysis of these issues is beyond the scope of this article.
The null basis product is a simple consequence of equation (
28), i.e., taking only the electromagnetic field into account in the analysis. Since reality requires other fields (e.g. electroweak), it can be assumed that changing
into the field tensor corresponding to reality will probably provide
, which makes it possible to assume a basis in real numbers. However, since in this paper it is considered the Alena Tensor with the electromagnetic field only, the basis (
50) will be retained for further analysis as an example, especially since the transition to the generalized field in the discussed approach is a fairly simple procedure.
3.2. Covariant Metric, Higgs Field, Riemann Tensor, Weyl Tensor and Gravitational Waves
Substituting (
37) into (
17) using (
33) and using Alena Tensor properties one gets expression for metric
describing the system in curved spacetime
It turns out that using properties of metrics, one may find a general solution for the inverted metric
. Summarizing the key properties one obtains
where in the last condition it is enough to check the index (0,0), because the null vectors are normalized (
). To simplify the calculations, it is easiest to define auxiliary variable q and start from anstaz with unknown
in
By eliminating the subsequent variables to provide equations (
52), one obtains covariant metric in the form
where invariants of electromagnetic field turns out to be related to the trace
which means that the trace
is also invariant. Using arc
⌒ for simplicity, to emphasize the change to curvilinear description (since values in curvilinear description may be diffrent), one may notice, that considered in curved spacetime trace
yields
. Therefore, the transition to curved spacetime can be understood as solutions with imaginary magnetic field
. It is also worth notice, that for the null basis example (
50), the above metric seems to describe a gravitational wave in conformal geometry, where
plays the role of a conformal factor
. Further analysis in this direction should allow to isolate both the polarization and the relation of
to the Ricci scalar by classical relation
.
Expressing
by null-vectors as in (
45) and requesting
one obtains ugly expression linking
and
q. However, substituting q as the
function according to (
55) and denoting
, it appears that this ugly expression actually expresses following dependence
where V is the classical Higgs potential and the broken symmetry in the system can be interpreted as an apparent mismatch of coefficients related to
. The
Figure 1 shows the classic "Mexican sombrero" potential
. Since
is related in equation (43) to the angle describing the spin, this means that in this solution the spin effect of the field is described by invariant angle (depends only on
) and, according to (
55) it is closely related to the trace of the metric and, consequently, to the Ricci scalar in the curvilinear description and thus to spacetime curvature. This also at least means that to obtain correct results in curved spacetime in the Alena Tensor model, the Higgs-like field is needed. Perhaps the above result will help explain the existence of the Higgs field and its relationship to curvature and spin.
For
the whole metric becomes significantly simplified and symmetric
which means that it may be assumed that the Higgs boson is probably related to the curvature of spacetime, corresponding to this configuration.
Using hyperbolic identities and expressing
then using (
33) one obtains
Using (
37) and (
38), the releation between electromagnetic energy density
, angle
and spin four-vector
zero component is therefore
which explains the origin of
and the connection of spin and Lagrangian with the energy density of the electromagnetic field. Since, according to the considered approach of Alena Tensor, in curved spacetime the energy density expressed by
must vanish (the entire stress-energy tensor of the electromagnetic field vanishes), this means, that
and thus
becomes invariant, which allows to analyze the transition to curved spacetime.
One may now consider what value the Alena Tensor takes in curved spacetime. To do this, it is easiest to analyze the behavior of
. As one may calculate, the determinant of this tensor is 0 and the matrix rank is 2, but, as described in the introduction, it degenerates to
in curved spacetime. It can be seen that in the Alena Tensor approach the metric follows from propagation. However, in curved spacetime the electromagnetic field according to this approach should vanish, remaining present only in the metric and determining geodetic. This can be achieved by degenerating vectors
and
to a single vector in curved spacetime
. However, in such situation the equation (
45) forces this vector to be the four-velocity, divided by the Lorenz factor
which degenerates
from (
46) to the form
Since the metric is known and
may be expressed by values of
,
in flat spacetime thanks to (
45) this allows to easy obtain the Einstein tensor and Ricci tensor using the equation (
10).
The described system seems to be Petrov type D or II [
29], although to be sure, the Weyl tensor should be calculated. This does not seem possible for the general case (without auxiliary assumptions about symmetries), but one could simplify the obtained description considerably, based on the following observation. One may notice, that the vanishing Lorentz factor
in
can be interpreted as an important suggestion for the description of motion in curved spacetime. Such motion would correspond to a stream of particles moving without dilation, a strictly ordered flow without local perturbations, resembling a perfect, infinitely stiff fluid. This means that
should be the Killing tensor and the system schould have hidden symmetry (similar to the Carter constant in Kerr solutions).
As shown previously, product of the basis is in considered case , which means that null vectors have global significance for spacetime geometry, thus Killing tensor should have a strong connection with propagation along null vectors (it is not a random symmetry, but a deep feature of spacetime) and this would mean the existence of special wave surfaces, e.g. electromagnetic waves and/or gravitational radiation. This would be a also clear indication that spacetime belongs to the Petrov D or II class and is associated with wave propagation solutions.
Also, the analysis of obtained equations drive to conclusion, that energy (energy density) is not something external to geometry in Alena Tensor approach, but energy is defined by the geometry of spacetime itself. This is a result in the spirit of General Relativity, but it goes even deeper: metric
depends on
, but at the same time
depends on the metric because it is trace of
in this metric. The system itself defines its own energy through the structure of the field, which is similar to the idea of self-consistent field [
30] - where the field and the source are inseparable, or emergent gravity [
31] - where energy, gravity and geometry arise from a common structure, or induced geometry [
32] - where energy comes from deformation of spacetime itself, as e.g. in Sakharov’s theory [
33]. It is impossible to “decree” energy in Alena Tensor, it must be calculated from geometry and the system works as a closed causal cycle.
The dependence for
of its norm and trace in curved spacetime (the norm is the square of the trace) is a key property for null space and suggests that
describes isometries related to null wave propagation, similar to pp-wave [
34] and Robinson-Trautman solutions [
35], what is actually expected in considered approach based on electromagnetic stress-energy tensor and result (
51). This would also mean that the Killing tensor is directly related to the energy distribution in spacetime as expected, similar to other GR solutions (eg. Kerr solution), and lead to a rather groundbreaking but also expected result in the context of the discussed approach, that the Killing tensor directly determines the Einstein tensor in main GR equation.
Since is simply the Alena Tensor (stress-energy tensor for the system) divided by , it gives correct conserved values in the Noether formalism (conserved density of energy and momentum). From the definition of the Alena Tensor as the energy-momentum tensor for a system it also follows that is symmetric and vanishes.
Since
this implies that along a geodesic parametrized by proper time
, its total derivative vanishes.
Using the symmetry of
one may note that
Therefore, the condition
holds for arbitrary tangent vectors
, and it follows that
. Therefore normalized Alena Tensor is Killing tensor for considered system.
In this way Alena Tensor theory becomes equivalent to some specific case of General Relativity equation expressed in the following form
where according to (
10) in the above equation
is the classical Einstein tensor
. Thanks to the above, since the Weyl tensor is defined as the part of the Riemann tensor that does not depend on the Ricci tensor, this implies that since Killing tensor determines the Einstein tensor, and the Einstein tensor determines the Ricci tensor, then one could calculate the Weyl tensor by extracting the part of the curvature that does not contain the Ricci tensor. One may expect that the Weyl tensor calculated in this way should depend on the energy density, the cosmological constant and the null vectors, which would mean that the spacetime geometry is strongly related to the energy of the electromagnetic field, exactly as seen in the obtained equations. This would also mean that the Christoffel symbols can be expressed as a function of the Killing and Einstein tensors, and the Riemann tensor can be written directly as a function of the Ricci and Killing tensors.
For the system under consideration, one may therefore construct a general Weyl tensor ansatz, which must contain only components that do not become zero when the trace part is subtracted from the Riemann tensor. It should be of the form
where
It is easy to check that the above tensors are linearly independent and form the basis for the representation of any Weyl tensor in the system under consideration. Analysis of their behavior indicates that
is responsible for "pure" directional propagation - e.g. a gravitational wave propagating along null directions (purely conformal part of the Weyl tensor — described solely by null geometry),
describes non-radiating, "axial" deformation of space - e.g. tidal sequences, consistent with mass motion without undulations,
describes conformal distortion of the background metric itself.
The coefficients
,
,
are not known a priori, but one may determine them with help of Riemann tensor. To calculate the Riemann tensor
in the system under consideration, it is enough to assume the following ansatz
It has the following justification
The Riemann tensor satisfies the known algebraic symmetries: The above ansatz satisfies them automatically.
There are only two tensor objects available in the system: the metric and the Killing tensor . The Riemann tensor must be constructed exclusively from them.
The first term with corresponds to the geometry of a space with constant curvature, as in de Sitter space:
The second term with is the minimal geometrically correct extension that takes into account the presence of non-null energy (represented by ). Its construction provides correct symmetries and enables the reproduction of a non-null Ricci tensor
Other possible combinations (e.g. ) are linearly dependent or asymmetric with respect to the required properties of the Riemann tensor — they do not provide new information in the case under consideration.
The whole creates the most general fourth-order tensor with Riemann symmetries, which can be constructed from available geometric objects.
Comparing the Ricci scalar and obtained from anstaz and the one obtained from GR equation with Alena Tensor, one obtains coefficients . Their complex structure does not allow for inclusion in this article, but they can be read in the supplementary files, which allows for further analysis. Given the Riemann tensor and computing the trace part of the Riemann tensor, one obtains the Weyl tensor in generalized form for the system under consideration. Normalized Riemann and Weyl tensors (divided by the cosmological constant) were computed in the supplementary files.
Having calculated the above-mentioned tensors, one may substitute an example basis (
50) depending on
and solve the system of equations (
33,
41,
56) presenting the results only as a function of
and
. Due to the complex relationships between variables, analytical solution is presented in supplementary file, and Weyl tensor component values were calculated for an example value of
from (
57). The
Figure 2 below clearly shows the wave nature of the
component of the Weyl tensor, presenting the result of the calculations performed and it actually seems, that the Weyl tensor component
encodes the variation of some curvature field along the propagating direction
.
Table 1 presents the values of the key components of the Weyl tensor calculated for the value
adopted for the analysis.
By calculating the Penrose–Newman scalars one can conclude that the example system described by the above Weyl tensor is Petrov type II and does indeed contain gravitational waves with two polarizations (+ and x), as well as some observable symmetries. The next two graphs show the oscillations of these scalars and the deformation effects for an example arrangement of test particles on a circle. In the supplementary materials there is a gif file presenting an animation of this distortion. This animation reveals the typical quadrupolar nature of spacetime distortions induced by gravitational waves (characteristic twisting and stretching).
The above Weyl tensor components are actually normalized. One may thus present the main GR equation in the Alena Tensor notation (
10) and represent it using the vacuum pressure
as follows
Since the obtained Weyl tensor is normalized by
, this means that the interpretation from (
18) that gravitational waves in the Alena Tensor are in fact propagating vacuum pressure disturbances also seems to be correct.
It is also worth noting that the obtained metric term
can in principle be interpreted as a vacuum energy contribution (effective cosmological constant) as in [
36,
37] playing the role of a metric scaling factor, as e.g. described in [
38] which allows to conclude about the value of the second invariant of the electromagnetic field.
Additionally, one may invoke the scalar field
associated with the presence of matter, where
. It is known from
2.2 that
is responsible for the presence of sources and in their absence
. Therefore, interpreting whole
as the wave amplitude tensor one would get representation
which would also allow to search for
as a certain wave function.
Figure 4.
Deformation visualization for particles on a circle.
Figure 4.
Deformation visualization for particles on a circle.
This approach allows for two simplifications related to the analysis of gravitational waves. Considering the force
responsible for effects related to gravity as shown in (
12) and extracting the acceleration
from it, one gets
since according amendment from [
10]
.
As shown in [
9],
is directly related to the effective potential in gravitational systems which can be calculated from the GR equations. This would allow searching for propagating changes of the effective potential itself (
) similarly as was postulated in [
39]. It would significantly simplify both the calculations and perhaps the methods of detecting gravitational waves.
The second potential simplification results from the possibility of analyzing only the Poynting four-vector as which might also help simplify the calculations and look for experimental proof of correctness for the Alena Tensor approach.