1. Introduction
Astrophysical observations indicate the existence of super-massive objects at the centers of galaxies. In our Milky Way, there is an invisible object located in the same place as the radio source “Sagittarius A”. According to the orbital motion of stars, astronomers estimate its mass to be 4.3 × 10
6 M
ʘ, and its radius to be less than 0.002 light years [
1]. The mass of the Sun is M
ʘ = 1.989 × 10
33 g. If the mass of an object in the center of a galaxy is six orders of magnitude greater than the maximum equilibrium mass of a neutron star, then it is natural to consider black holes as the most likely candidates for super-massive objects in the centers of galaxies.
According to modern concepts, a black hole is a process of unlimited compression (collapse) of matter under the action of dominant forces from its own gravitational field. In this case, the density grows indefinitely, and the energy per particle inevitably reaches and exceeds the binding energy of “elementary” particles in neutrons, or in their composite components.
The time of existence of galaxies (and, consequently, of black holes in their centers) is of the order of lifetime of the universe. With a very slow evolution of a black hole, the local equilibrium concentrations of particles, transforming one into another via “chemical reactions”, depend on temperature and pressure, and do not depend on specific reaction channels (Reference [
2], Section 101). One of the most important issues, that is not yet clarified, is the reverse effect of mutual transformations of particles on the process of gravitational self-compression. The fact that the lifetime of a galaxy with a collapsing object in the center is of the order of the universe’s lifetime suggests that the mutual transformation of particles one into another can slow down the compression process, or even stop it. This is the main reason for searching and analyzing static configurations of gravitating objects in the general theory of relativity.
In the macroscopic theory of dark sector [
3], the contribution to the galaxy rotation curves by dark matter, described by a longitudinal vector field, is expressed in terms of the non-zero covariant divergence
of the field at the center.
is a free parameter of the theory [
3]. In fact, the non-zero value
depends on the interaction of dark matter with a black hole in the center of a galaxy. To establish this connection, it is necessary to analyze the internal structure of a black hole in balance with the dark matter outside of it. In a vacuum (with no stabilizing action by dark matter) an equilibrium state of a super-heavy black hole would be impossible.
In the process of gravitational collapse, the density of matter increases continuously. The stage of compression, when massive Bose particles (Z bosons, W bosons, Higgs scalar bosons, and/or boson pairs of fermions) are dominating, is inevitable. At low temperatures, a degenerate boson gas is energetically more favorite than a degenerate fermion gas. At zero temperature, bosons are located on the ground level. This state of boson matter is named Bose–Einstein condensate. Its wave function is a scalar field (Reference [
4], Section 26). At the border of a super-massive scalar field, the energy density vanishes, but the pressure does not. If a degenerate Fermi gas is considered as the ordinary matter of a black hole, then both the energy density and pressure are zeros on the surface. Therefore, a realization of phase equilibrium on the interface of a super-massive black hole and dark matter is more likely at the stage of domination of boson matter in a collapsing black hole.
The main properties of a gravitating Bose–Einstein condensate in an equilibrium state are summarized in my review article [
5], published in an issue of the Journal of Experimental and Theoretical Physics (JETP), dedicated to the 85th anniversary of academician Pitaevskii. Applying a homogeneous Klein–Gordon equation, we deal with a nonlinear eigenvalue problem. At a slightest deviation of parameters from eigenvalue, the integral curves go into the signature area of the opposite sign, which is tacitly considered to be “non-physical”. With a restriction of a constant signature, it is not possible to find static solutions that adequately describe the equilibrium state of gravitating super-massive objects.
A possible equilibrium structure of a gravitating scalar field in the region of space of a black hole with the signature changing sign (within the Schwarzschild sphere, where ) is considered in this article. Instead of representing the metric tensor component , which fixes the sign, I use a less rigorous condition of regularity. A change in the metric signature does not lead to logical contradictions, since it occurs beyond the event horizon in an inaccessible area to a remote observer.
The condition of phase equilibrium at the interface between a scalar field (black hole) and a longitudinal vector field (dark matter) determines the covariant divergence of the longitudinal vector field. It becomes possible to establish a connection between the plateau velocity of a galaxy rotation curve and the black hole’s mass (see Equation (68) below).
2. Space-Time, Strongly Curved Up to Changing the Signature of Metric Tensor
In the framework of the standard approach, a static space-time metric,
with a spherically symmetric distribution of matter, is considered.
In the general theory of relativity, one can introduce a local Galilean coordinate system at any point of a smoothly curved space-time. In a curved space-time, a transition from one coordinate system to another, and back, is unambiguous if the Jacobean
j of transformation is non-zero. The Jacobean of a transition from coordinates
to
can be expressed in terms of the determinants of metric tensors
and
:
where
(Reference [
6], Section 83). The signature of a Galilean coordinate system is (+, −, −, −). The Jacobean of transition from Galilean coordinates
, where
, to curved ones
is
.
In our case of interest, the metric tensor in Equation (1) is diagonal, and there is a Schwarzschild’s sphere inside a gravitating object, where the component changes sign, while does not. When only changes sign, the signature of metric tensor (1) becomes (+, +, −, −). Accordingly, . On the Schwarzschild radius, the Jacobean . Due to different signatures on the opposite sides, the space-time can be either locally flat, or locally non-flat on the hyper-sphere .
Formally, derivation of Einstein equations at
, and at
should be carried out separately, with
where
, and
in the area with
. For both sides, the static gravitational field, created by a spherically symmetric distribution of matter, satisfies the same Einstein equations (see Equations (100.4) and (100.6) in Reference [
6]). We write them down in the following form:
Gravitational constant
,
,
.
The general solution of Equation (3) is
r0 is a constant of integration, and
is a regular function, provided that the integral in Equation (5) converges. The convergence of this integral means that the mass within the layer
,
(with account of gravitational mass-defect) is finite. If the total mass
is finite, the metric component
is a continuous function within the whole space
. The regularity of
follows from the finiteness of the total mass of matter, regardless of its physical nature.
A static solution to Einstein’s equations for a gravitating point-size body was found by Schwarzschild back in 1916 [
7].
Actually Equation (7) is applicable outside a gravitating body, including its boundary, provided that, on the surface,
. At
Newton’s law is applicable, and the Schwarzschild radius
is proportional to the total mass
M of a gravitating point-like object in the center.
In a vacuum, outside a gravitating body, there is no physical reason for the existence of irregularities. Naturally, the determinant of a Schwarzschild’s static metric (Equation (100.14) in Reference [
6]), as well as the determinant of a stationary Kerr’s metric (Equation (104.2) in Reference [
6]), does not change signs. In a vacuum, metric signatures remain unchanged, and the space-time on the hyper-surface
is locally flat.
Analyzing the equilibrium states of spherically symmetric gravitating matter, it is usually assumed that the energy density at the center is finite, and the metric tends to the Galilean metric at infinity. Bearing in mind a regular solution in the whole space, people choose two equations,
as boundary conditions for the Einstein Equations (3) and (4). These kinds of solutions exist only if the total mass of matter does not exceed a critical value
. Critical mass is of the order of the Sun mass in case of a degenerate neutron Fermi gas. For a Bose–Einstein condensate (see Reference [
5]),
Plank mass
. For bosons with a mass
, the critical mass
is about one million tons. In the solutions with boundary conditions from Equation (9) and total mass
, the metric component
does not change sign at all.
Formally, at the absolute zero temperature in a vacuum, there could exist static states of boson condensate having an arbitrarily large mass with
not changing sign. However, these states are metastable. An aggregate of numerous isolated centers, each containing a condensate with the maximum possible mass of the order of Equation (10), is energetically more advantageous [
5].
As applied to the gravitating Bose–Einstein condensate, the Klein–Gordon Equation (16) for the condensate wave function, together with Einstein Equations (3) and (4), form a complete set of equations. Equation (16) is a homogeneous second-order equation. We face a nonlinear eigenvalue problem [
5]. If the total mass
M is less than the critical
, then, starting from −1, the function
grows with increasing
r, achieves its maximum value, and decreases back to −1 as
. With
the function
remains negative within the whole interval
.
At the slightest deviation from an eigenvalue, the integral curve grows monotonically and inevitably changes sign at some . Note that . If the total mass , then the integral in Equation (6) converges at the upper limit, and the energy density decreases faster than r−3 as . Looking at Equation (3), one can see that, in the region of changed signature , the derivative (grr)′ becomes negative when (1 + grr)/r exceeds with growing r. It means, that there may exist (and does exist, see below) a solution where, with increasing r, the metric component gets into the region of violated signature, passes through a maximum, and returns back into the region of the Galilean signature. intersects the x-axis twice.
Suppose that the metric signature is changed to (+, +, −, −) within a spherical layer
. Then,
If
, then
and the Jacobean in Equation (2) are zeros at
, and at
. The area
of the modified metric signature is tacitly considered “non-physical”. I do not see a reason why (unlike the Schwarzschild’s and/or Kerr’s metrics in a vacuum) a super-massive gravitating object cannot curve the space-time very strongly, up to changing the signature of the metric tensor inside a material medium.
Yes, there can exist selected regions, like hyper-spheres
and
, where space-time is not locally flat. As a rule, it happens, if there is a physical reason. In our case, the radial function
of the gravitating Bose–Einstein condensate satisfies the Klein–Gordon Equation (19).
is the coefficient at the highest derivative. At
, Equation (19) is not defined, since the coefficient at the highest derivative is zero. I used a standard approach to the problem of a small parameter at the highest derivative in
Section 3.
From the general solution in Equation (5) to Equation (3), we get the identity
If, in the region , the metric signature (+, −, −, −) remains unchanged, then is the event horizon for a remote observer. In the Schwarzschild solution in Equation (7), turning to zero can happen at any radius outside the gravitating body. However, in a vacuum, the determinant of metric tensor is non-zero, and the metric signature remains unchanged. Turning to zero, followed by the change of the metric signature, can take place inside a body, or on its surface, or on the interface of two gravitating objects.
As it follows from the identity in Equation (12), in the presence of a layer with a different signature, the radius
is not connected with the whole mass of a gravitating object, but only with its part inside the layer. The total mass
M is connected with the horizon
by Equation (8) only if the integral in Equation (12) converges at
. In this case the relation in Equation (7) takes place for
. At
, we have
In view of Equation (13), it follows from Einstein Equations (3) and (4) that
Turning the energy density to zero at means that the surface of a gravitating body can be an event horizon. However, since , the pressure on the surface does not disappear. It means that (without taking into account interactions of a non-gravitational nature) there can be no static equilibrium state of a gravitating body in a vacuum with a broken signature of the metric tensor. Nevertheless, one cannot exclude a possibility that the event horizon takes place on the interface between two gravitating objects, for instance, if the inside pressure of a black hole is balanced by the pressure of dark matter from outside.
According to modern concepts, the amount of dark matter in the universe is several times larger than the amount of the ordinary (baryonic) matter. The presence of dark matter opens up a possibility for existence of static solutions to the Einstein equations describing the gravitational field of spherically symmetric matter with no limitation of mass. An example of such a solution, in which the metric component changes sign twice (firstly, inside a black hole and, secondly, on the interface with dark matter), is investigated below. The equilibrium state of a super heavy black hole becomes possible due to the presence of dark matter. The balance of pressures at the interface of a black hole and dark matter is able to support a static equilibrium of these two phases. Let us demonstrate this considering a gravitating scalar field as an example of ordinary (not dark) matter inside a black hole.
3. Gravitating Scalar Field behind the Horizon
The Lagrangian of a complex scalar field
ψ in a curved space-time with the metric tensor
(Equation (1)) has the form
In accordance with the least action principle,
ψ and
ψ* satisfy the Klein–Gordon equation.
From the point of view of equilibrium in its own gravitational field, it is implied that the number of quanta of the field is large, and all interactions, except the gravitational one, are not significant. Today, the interaction of dark and ordinary matter is observed through the curvature of space-time only.
The main characteristic determining the gravitational properties of a scalar field is the mass of a quantum
m. In the power series of the potential,
term
and higher degrees are corrections for collisions of particles and/or interactions of non-gravitational nature. Omitting these corrections,
in Equation (16) is a constant having the dimension cm
−2. It is related to the mass of the quantum
m:
.
Time is a cyclic coordinate in a static field. The energy of a single quantum
E =
ω is the integral of motion. In a flat space-time, the Klein–Gordon Equation (16) is a linear one. Its solution is a plane wave
, describing the motion of a particle with relativistic spectrum
. In a curved space-time,
E is the conserved energy of the field per one quantum. A scalar field in the state of definite energy
E has the form
Radial function
ψ(
r) obeys the equation
Note that
is the coefficient at the highest derivative in Equation (19). Since
, the Klein–Gordon equation is not determined on the hyper-sphere
.
The Lagrangian (15) of a scalar field does not depend on the derivatives of the metric tensor. The energy–momentum tensor is derived by the formula
For the components of the mixed energy–momentum tensor, operating in the Einstein Equations (3) and (4), we find
The relation
follows from Equation (21). Using the relations in Equations (21) and (22), it is convenient to present the set of field and Einstein Equations (19), (3) and (4) in the normal form. In the dimensionless variables
,
we have a system of four first-order equations, resolved with respect to derivatives:
The set of Equations (24)–(27) contains no parameters. In Equation (23) and below, g(x) is the component of the metric tensor. Please do not confuse it with the determinant in the previous formulas.
Denote a dimensionless gravitational radius . At the component of metric tensor . Equations (24)–(27) are determined at , and at separately. They are not determined at because, at this point, the coefficient at the highest derivative in the scalar field Equation (19) is zero.
It follows from Equation (27) that, at
, (+0 means from above, and −0 means from below)
Here,
and
are one-sided limits either from above or from below. Substituting Equation (28) into Equation (26), we obtain:
Note that the derivative does not depend on the behavior of and at . The assumption of linearity for at is not required in advance. The linearity of at follows directly from Equation (3) provided that is a regular function. Energy density is a continuous function inside a homogeneous medium and, therefore, one can put in Equation (29).
The sign of depends on the difference . If , then the function itself , and its derivative at . If , the derivative and is a growing function in the vicinity of . In the case , we have , and is a decreasing function in the vicinity of the horizon .
It follows from Equations (28) and (29) that
On the left-hand side
; thus, the right-hand side of Equation (30) is also a positive quantity. The derivative
in Equation (29) is a non-zero constant. The factor
in Equation (30) changes sign at
. Therefore, the combination
has different signs at
and
. In the case when
Equation (30) makes sense if
at
, and
at
.
3.1. Regular Gravitational Radius
In the case of exact connection between the parameters,
all four functions,
, as well as their derivatives, are continuous at
. In accordance with Equation (28),
at
. From Equations (25) and (27), we find
With the exact connection in Equation (31), we have a regular solution of Equations (24)–(27), continuous on the boundary
between the regions of different signatures of the metric tensor:
It is convenient to use these relations as boundary conditions for numerical integration. There are two free dimensionless parameters
and
.
In the absence of the relation in Equation (31), function h(x) would have a gap at . Without a physical reason, just like in a vacuum, there should be no discontinuity of the metric component inside a homogeneous medium.
3.2. Horizon
A sphere, where and , can be the surface of a gravitating body. In this case, at . The scalar field terminates at in a square-root manner. It is the event horizon for a remote observer. In order to avoid confusion, the event horizon where w(x) tends to zero as a square root we denote . We leave the notation for a regular gravitational radius inside a black hole. The horizon with a root feature is the upper border of the region occupied by the scalar field.
Solutions with the metric function g(x), changing signs twice deserve attention: once inside a black hole at a regular gravitational radius and the second time at the horizon .
The area is a layer with a violated signature of the metric tensor. is positive in this zone. Intervals and , where the metric function , are ordinary space-like zones. In the region (from the center to the horizon ) the scalar field is a regular function of the radius. In this region, the unknowns satisfy Equations (24)–(27) with the boundary conditions in Equation (33) in the close vicinity of .
At the boundary of the scalar field, the energy density vanishes (Equation (14)). Comparing
in Equation (13) with Equation (29), valid for both cases
and
, we find that, at the termination point (on the horizon
), the scalar field
is non-zero:
. In the vicinity of the horizon, we get
The series expansion near the horizon goes in powers of
. Determining
and
at
from Equation (27), we see that the main contribution comes from the combination
. In accordance with Equation (28),
. The main term of expansion is
Inside the horizon sphere
the derivative
(see Equation (35)). If a test body is at rest, then the force acting on it is directed toward the center (Reference [
6], problem 1 at the end of Section 88).
5. Longitudinal Vector Field
The wave function of Bose–Einstein condensate, being a scalar, satisfies the Klein–Gordon Equation (16). Its derivative is a covariant vector. In the case of a longitudinal vector field, it is the opposite situation. The wave function
is a vector, and its covariant divergence
, being a scalar, satisfies the Klein–Gordon Equation (16). At the same time, the wave function
itself is a gradient of the scalar
[
3],
The mass of a quantum of a longitudinal vector field is denoted by μ, so as not to be confused with the mass
m of a quantum of a scalar field. A longitudinal vector field has a potential. The covariant derivative of the field is its potential at the same time.
Lagrangian of a longitudinal vector field
,
depends on the derivatives of the metric tensor via
. The energy–momentum tensor is derived using the general Equation (94.4) in Reference [
6]. The detailed derivation of the energy–momentum tensor of a longitudinal vector field,
is given in Reference [
3] (Section 2.4).
For a super-heavy black hole, the inequality
>>
is carried out with a huge margin. The function
h(
x) at
is negligible in comparison with unity. One can see it in
Figure 10. Energy–momentum tensors in Equations (21) and (46) have the same structure on the interface region. Only the wave function and the derivative are swapped, and masses of quanta are different. In dimensionless variables with
as a unit of length,
we have the following system of equations describing the longitudinal vector field and the gravitational field outside a black hole
Equations (48)–(51) are similar to Equations (24)–(27), presented in
Section 3 in the reference frame with
as a unit of length. Doing the same derivations as in
Section 3, we get the derivative of the metric function
at the interface
from the side of dark matter:
Equations (52) and (29) are derived in different reference systems. Turning from Equation (47) to the reference system used inside a black hole (Equation (23)),
we see that Equation (52) is invariant against this transformation:
Regularity of the gravitational field means that the metric function
and its derivative are continuous at the horizon
. From Equations (29), (54), and (34), it follows that
In this case, the components of the energy–momentum tensors of the scalar and vector fields coincide at the interface
The equality in Equation (55) provides the balance of pressures on both sides of the horizon . This is the phase equilibrium of the scalar field inside the horizon, and the longitudinal vector field describing dark matter outside the horizon.
6. Galaxy Rotation Curve: Dependence of Plateau Velocity on the Mass of a Black Hole
In Equations (24)–(27), the unit of length along the
x-axis is the de Broglie wavelength
of a quantum of the scalar field (~10
−16 cm for a quantum with the rest mass ~ 100 GeV/
c2). The argument of covariant divergence
and other functions in Equations (48)–(51) is
. On the galactic scale, a natural unit of length is the de Broglie wavelength
of a quantum of the longitudinal vector field. According to damped oscillations on the plateau of galaxy rotation curve (see
Figure 11),
10 kpc ≈ 3 × 10
22 cm.
In References [
3] and [
10], the wavelength
is considered as a unit of length. The argument of the covariant divergence
is
It follows from Equations (48)–(51) that
on the horizon
. Therefore, on the interface
, the covariant divergence gets reduced to the usual one:
When the scale is changed by
q times,
the ordinary derivative is transformed as
Thus,
The balance of forces of gravitational attraction and centrifugal repulsion determines dependence of the velocity
V(
r) of a star, rotating around the center of a galaxy, on the radius. In a static spherical gravitational field, the centripetal acceleration is
(Reference [
6], problem 1 at the end of Section 88). Equality of centripetal acceleration to centrifugal
V2/
r gives
In the macroscopic theory of the dark sector [
3], dark matter is described by the longitudinal vector field
A black hole in the center is considered as a point-like object due to the small size of its radius when compared to the large size of a galaxy. The derivative
dν/
dr is expressed in terms of the covariant divergence of the field in the center
(see Reference [
3], Equation (86). This non-zero phenomenological parameter indicates the presence of a connection of dark matter with a black hole. Earlier, it was not possible to determine this connection without knowing what was going on inside a black hole. Now, the condition of phase equilibrium in Equation (55) at the interface
between the longitudinal vector field (dark matter) and the gravitating scalar field (black hole) allows us to connect the covariant divergence
of the longitudinal vector field to the parameters of a black hole. In particular, it allows us to express the velocity of rotation on the plateau as a function of the mass of a black hole.
Figure 9b,c show that, in the case of
, the metric function
g(
x) (red curve) practically coincides with the Schwarzschild asymptote (dashed line) at
. It means that, at the interface
the gravitational field is weak. At
, metric function
The gravitational field is described by linearized Einstein equations,
Divergence
of the longitudinal vector field obeys a linear Klein–Gordon equation,
In accordance with Equation (44),
The condition of phase equilibrium in Equation (55) at the interface
between a scalar field and a longitudinal vector field determines the value
. In the macroscopic theory [
3],
is a free parameter. In view of insignificance of the gravitational radius of a black hole
in the scale of a galaxy ∼10
22 cm, the value
can be used as a boundary condition at
The solution to Equations (61) and (62) is
Equations (8) and (63) determine the dependence of the dark matter wave function on the black hole mass: Local dark matter density is proportional to
Substituting Equation (63) into Equation (60), and using the identity,
we find
Substituting Equations (63) and (65) into Equation (59) and taking into account Equation (57), we get
, which is needed for Equation (58).
In accordance with Equation (58), the dependence of the velocity
V(
r) of a circulating star on the distance
r from the center of a galaxy is as follows:
where
Vpl,
is the velocity on the plateau of a galaxy rotation curve.
is the Plank mass.
is the mass of a black hole, visible to a remote observer.
m is the mass of a quantum of a scalar field, and
μ is the mass of a quantum of a longitudinal vector field (dark matter). Equation (67) shows that the saturation of the function
V(
r) on the plateau is accompanied by damped oscillations [
10].
Figure 11 shows the rotation curve of the spiral galaxy NGC 3769 of the Ursa Major cluster [
3,
10]. Dots with error bars are observations, copied from Reference [
11]. The solid curve is the analytical dependence of the velocity on the radius (Equation (67)).
The plateau velocity of this galaxy is
The period of damping oscillations is
This corresponds to the de Broglie wavelength of a particle with a mass
. On the basis of the standard model of elementary particles, the mass of a quantum of a scalar field with the rest energy about 100 GeV is
. If the black hole at the center of the galaxy NGC 3769 in the Ursa Major cluster is really compressed to the state of condensation of scalar particles, then, according to Equation (68), its mass is three times larger than the solar mass: