You are on page 1of 33

Vortex particle model in a renormalizable quantum

gravity
T. D. Andersen

E-mail: andert at alum dot rpi dot edu

Abstract. The search for a quantum theory of gravity has become one of the most
well-known problems in theoretical physics. Problems quantizing general relativity
because it is not renormalizable have led to a search for a new theory of gravity that,
while still agreeing with measured observations, is renormalizable. In this paper, I show
that, given a “vortex” model of elementary particles in which rest mass derives from
intrinsic spin and polarization, a Yang-Mills force with a U(1)⊗SU(2)⊗SU(2) group
symmetry (which is the symmetry group of conservation of energy, linear momentum,
and angular momentum) predicts solar system observations of gravitational behavior as
well as binary pulsar precession and orbital speed-up caused by gravitational radiation-
reaction. Using a homogeneous, isotropic universe model, I show that this theory
explains the accelerating expansion of the universe directly from group symmetry with
no ad hoc constants. In addition, because it is a generic massless Yang-Mills theory,
it is a renormalizable quantum theory.

PACS numbers: 04.60.-m, 95.36.+x

Submitted to: Classical and Quantum Gravity


Vortex particle model in a renormalizable quantum gravity 2

Keywords: quantum gravity, yang-mills theory of gravity, alternatives to general


relativity, accelerating expansion of the universe.

1. Introduction

That general relativity is able to subsume Newtonian gravity and explain phenomena
that do not fit into the Newtonian framework such as Mercury’s perihelion precession,
light deflection, and gravitational red-shift as well as being compatible with special
relativity has brought it wide acceptance. In recent years, however, as measurement
tools have grown more accurate and new observations made, the necessity of introducing
either tunable parameters or exotic forms of matter and energy to make general relativity
fit those measurements has brought the theory into question. Observations of galactic
rotational curves and gravitational lensing [1, 2, 3] have demanded the introduction of
dark matter and accelerating expansion of the universe [4] dark energy. At present there
is no consensus on what these substances are, whether they even exist, or whether they
are features of a more complete theory of gravity.
Besides the problems with general relativity at the macroscopic scale, achieving a
quantum theory of gravity has become one of the most significant unsolved problems in
physics. Attempts to place the Einstein-Hilbert action,

Z
SEH = d4 x −gR, (1)

where R is the Ricci scalar, into a functional path integral ensemble, have all failed,
giving nonsense results. The immediate source of the difficulty is the nonlinear

interaction of the volume element, −g, and the inverse metric, g ab , which create an
infinite number of coupling constants in perturbation expansions. Therefore, unlike
the actions of other forces, the gravitational action is not finite polynomial. A similar
disaster occurs quantizing the weak force but disappears when it is unified with the
electromagnetic force, motivating the quest to unify gravity with the other forces in the
hope that it will become renormalizable [5]. This unified theory, however, would not
solve all of gravity’s problems because, by design, it would become general relativity at
galactic and cosmological scales where dark matter and energy have an effect. Therefore,
a unified theory would not necessarily solve these large scale, classical problems without
additional, low-energy features.
Because the curved spacetime assumption causes such deep, unresolvable problems
at the quantum level, there is considerable motivation to find an alternative to spacetime
curvature as the source of gravity. Despite its current acceptance, curved spacetime was
a counterintuitive choice for a theory of gravity given the success of electromagnetism
in the 19th century. As Wald points out [6],
Maxwell’s theory is a remarkably successful theory of electricity, magnetism,
and light which is beautifully incorporated into the framework of special
relativity. Therefore, one might expect that the next logical step would
have been to develop a new theory of the other classical force, gravitation,
Vortex particle model in a renormalizable quantum gravity 3

which would generalize Newton’s theory and make it compatible with special
relativity in the same way that Maxwell’s theory generalized Coulomb’s
electrostatics. However, Einstein chose an entirely different path[.]
Because curved spacetime has never been precisely measured, the path not taken still
lies open.
Indeed, we can ask how precisely general relativity has been confirmed, and, as this
paper explores in detail, the answer is two-fold: (1) if general relativity’s field equations
were as simple as Maxwell’s equations or not very much more complicated, then it could
be considered a well-confirmed theory because its linearized version has a great deal of
evidence behind it, and (2) although observation has confirmed many effects, the non-
linear portions of the theory have been unobservable in any precise way because of the
weakness of gravity. All precision experiments (classic tests) of general relativity have
been done within the solar system (with the exception of binary pulsar precession) where
spherically symmetric, weak gravity prevails and only the first order post-Newtonian
Einstein-Infeld-Hoffman (EIH) equations have been confirmed [7, 6]. An example of
the best recent evidence for strong field general relativity is the periastron precession
of the double pulsar system PSR J0737-3039A/B [8], but these measurements do not
contradict a flat spacetime Lagrangian treatment of the binary system as shown in Sec.
6.3. Furthermore, recent measurements of the cosmic microwave background, where
spacetime curvature is most likely to appear, show none at all [9] with the inflationary
theory as a potential explanation [10].
The difference between the EIH equations and the full equations of GR has invited
a large number of alternative theories of gravity such as are discussed in [11], but
virtually all are metric theories and retain the curved spacetime picture with the
associated problems mentioned above. Non-metric theories, by contrast, offer the hope
of eliminating gravity’s problems with quantization at the cost of also eliminating the
elegant but unconfirmed curved spacetime approach and its intractable nonlinearities.
It has been suggested that the Standard Model approach, where forces have Yang-Mills
actions, may be more suitable for a quantum theory of gravity. The tetrad or veirbein
formulation of the Einstein-Hilbert action, for example, has a structure similar to the
Yang-Mills [5]. Conformal gravity has a Yang-Mills structure as well [12]. None of
these theories, however, have achieved significant success in making predictions that
general relativity is not capable of making nor in resolving, entirely, the problems
with the quantum theory. Thus, there has been, as yet, no motivation for replacing
general relativity with any of these nor of accepting any as quantum gravity’s true
representation.
One of the problems with many of these approaches is that they attempt to
recast general relativity into the non-Abelian gauge format, proceeding downward from
relativity to quantum gravity, rather than proceeding from the known symmetries of
spacetime and the properties of macroscopic bodies and light and proceeding upward to a
theory. The latter is the more difficult task because it necessitates tediously recalculating
all the observed consequences of GR from the new theory rather than subsuming them
Vortex particle model in a renormalizable quantum gravity 4

as Einstein did with Newtonian theory. Given the decades long search for a theory of
quantum gravity, however, the time for a such an approach has come. In essence, if GR
is the child of Newtonian theory, then instead of arriving at the child of GR, we arrive
at a sibling with similar features and origins but no direct inheritance.
In this paper, I present a Yang-Mills theory of gravity with a U(1)⊗SU(2)⊗SU(2)
symmetry group. The U(1) symmetry group implies conservation of a scalar quantity,
mass/energy. The SU(2)⊗SU(2) group is a double cover of SO(3,1), the Lorentz group.
This symmetry implies conservation of linear momentum by the Lorentz boost members
and angular momentum by the Lorentz rotation members of the SO(3,1) symmetry
group. Because a non-Abelian SO(3,1) theory has local Lorentz symmetry, the theory
complies with the Einstein Equivalence Principle (EEP) and the Strong Equivalence
Principle (SEP) (which even most alternate metric theories do not satisfy) but has a
finite, polynomial action with a renormalizable quantization. Translations in spacetime,
which extend the Lorentz group to the Poincaré group, are not included because
all physical particles are observed to trace curves in a 3+1-D spacetime and, thus,
translational invariance is not an observed effect.
Rest mass is derived from the energy of spin and polarization and this is the second
component to the theory. In any frame, these vectors form an anti-symmetric tensor
S µν such that polarization is pi = S 0i and spin is si = ijk S jk where summation is
implied. Spin and polarization have a conserved current (similar to the symmetric
electromagnetic stress energy tensor [13]),
 
µν µλ ν 1 µν αβ
T = −m S Sλ + η Sαβ S , (2)
4
where η µν = diag(−1, 1, 1, 1) is the Minkowski metric. This tensor is both symmetric
and traceless. This implies that mass derives relativistically from an intrinsic kinetic
energy of spin and oscillation.
With these definitions, this Yang-Mills theory, called vortex gravity because of the
vortex particle interpretation that spin and oscillation implies, predicts the outcomes
of all Solar System gravitational experiments and observations and binary pulsar
observations including radiation reaction, all on a flat (Minkowski) spacetime with a
finite polynomial, positive definite action that is renormalizable. Indeed, as discussed
in section 7, no experiment or observation to date contradicts the theory.
In addition, the theory predicts the accelerating expansion of the universe directly
from group symmetry. The accelerated expansion of the universe is a phenomena
that is now beyond doubt. This phenomenon has been managed by adding an ad
hoc, empirical parameter to the Einstein field equations, the cosmological constant,
Λ. The constant has been dissatisfying, however, not only because it is ad hoc and
unexplained but because it is necessarily constant by conservation of energy. The
classical Yang-Mills field equations, derived in Section 3, predict, meanwhile, that the
accelerated expansion is a consequence of group symmetry. In section 6.4, I show that
the gravitational field itself creates negative pressure in a homogeneous-isotropic model,
Vortex particle model in a renormalizable quantum gravity 5

resulting in acceleration. Observations currently attributed to dark matter such as


galactic rotational curves and anomalous lensing are not addressed in this paper.
The paper is organized as follows: Sec. 2 describes the conserved quantities of mass,
linear momentum, and angular momentum, how these are represented in the theory, and
how they appear intrinsically through the vortex model; Sec. 3 shows a derivation of
the field equations for the U(1)⊗SU(2)⊗SU(2) Yang-Mills theory and its formulation
as two covariant field equations; Sec. 4 derives the equation for particle motion and
shows that it is the same as the geodesic equation in the traceless gauge; Sec. 5 gives a
discussion for how the theory satisfies the Strong Equivalence Principle (SEP) which GR
also satisfies; Sec. 6 shows that the theory agrees with measurements within the solar
system including redshift, perihelion precession, time dilation and light bending (6.2),
binary pulsar precession (6.3), the quadrupole formula for gravitational radiation as well
as showing that there are no dipole/monopole moments (6.3.1), and the equations for
an expanding Robertson-Walker universe (6.4) including the accelerating expansion of
the universe; Sec. 7 explores any possible experimental disagreement with the theory;
Sec. 8 discusses novel predictions made by the theory; Sec. 9 briefly mentions the
renormalizable quantization; and Sec. 10 discusses related work. There are also three
Appendices.

2. Conserved quantities of the vortex model in the classical limit

The vortex model of elementary particles was first introduced by Lord Kelvin for atoms
and since expanded with modern intrinsic quantum properties of particles [14]. Even
predating the development of special relativity, the mass of the electron has been
presumed to derive from its electrical energy [15], and hence the stress-energy tensor of
an electron, and by extension all matter, must, classically, take the general form of the
Maxwell stress-energy tensor. For neutral particles, their stress energy is based on spin
and oscillation rather than magnetic and electric fields, but the critical point is that
spin/oscillation has, not only mass, but kinetic energy in the form of static pressure. In
other words, particles cannot be represented as “point masses”, blobs of matter with the
only non-zero components of their stress-energy tensor T 00 = m in the rest frame. They
have space-space components as well, such that δjk T jk = m (derived below). Because
they are additive, these space-space components cannot be thrown out in scaling up to
macroscopic bodies.
The vortex model proposes that all mass derives from intrinsic quantum properties
of spin and oscillation. Special relativity shows that spin and oscillation are not separate
entities and that each particle has a covariant, anti-symmetric spin tensor, sµν = −sνµ ,
forming a single 4-dimensional spin in the six planes of a Minkowski space. In any
frame, oscillation occurs in the direction of the polarization vector P i = s0i and spin in
the plane with spin vector S i = ijk sjk .
The 5×5 matrix current of the U(1)⊗SU(2)⊗SU(2) Yang-Mills theory, Jµ , describes
the motion of particles (see Appendix Appendix A for an overview of the Lorentz group).
Vortex particle model in a renormalizable quantum gravity 6

This current, written in component form, is,


Jµ = Jµ 0 (J 0 ) + Jµ i J+i + Jµ0 i J−i , (3)
0
where Jµi represents local Lorentz boosts plus rotations, Jµi represents boosts minus
rotations, and Jµ0 represents mass and energy flux. The energy current, Jµ 0 plus the
boost plus rotation current, Jµ i , (which, from the previous section, is the coefficient of
J+i = K i + J i ) forms a tensor, Jµa where a is a rest frame Lorentz index. This is the
sum of the energy of oscillation and the energy of spin, and, because the particle can
0
have no other internal motion, Jµa describes its total energy. The second, Jµi , is the
coefficient to the boost generator minus the rotation generator, J− = K − J i , and,
i i

because, the energy of spin and oscillation are both classically equal, by subtracting the
two, it vanishes in the classical limit.
The energy of spin and oscillation associated with the first current, Jµa , arises
because classical particles oscillate and spin at the speed of light, c = 1; therefore, we
have rest frame normalized polarization vector pi = P i /P and rest frame normalized
p p
spin vector si = S i /S where P = δij P i P j and S = pδij S i S j which describe their
internal velocity. The spin angular momentum, Lspin = l(l + 1)~ where l is the spin
number of the particle, suggests that the particle has a classical radius somewhat larger
or smaller than its Compton wavelength, λ = ~/mc. Since it is rotating at the speed of
light, if r0 is the radius, we have,
c c
Lspin = ~ = mc2 , (4)
r0 λ
in which case
p
r0 = λ l(l + 1) (5)

is the particle’s classical radius, e.g., r0 = 3λ/2 for spin-1/2 fermions. From the
normalized spin/polarization vectors, a normalized tensor is formed which can then be
in any frame,
 
0 p1 p2 p3
 −p1 0 s3 s2 
 
Sµν =  . (6)
 −p2 −s3 0 s1 
−p3 −s2 −s1 0
The first current, total energy, is given in any frame by,
1
Jµ ν = −m[Sµλ S λν + δµ ν Sαβ S αβ ]. (7)
4
Like the electromagnetic stress energy tensor, this has the form (in Cartesian coordinates
(x, y, z)) [13],
 
1 2 2
(p + s ) Ux U y U z
 2
Ux −Pxx −Pxy −Pxz 

Jµν = m  , (8)

 Uy −Pyx −Pyy −Pyz 
Uz −Pzx −Pzy −Pzz
Vortex particle model in a renormalizable quantum gravity 7

where U~ = p~ × ~s, Pij = pi pj + si sj − 1 (p2 + s2 )δij , p2 = (px )2 + (py )2 + (pz )2 , and


2
s2 = (sx )2 + (sy )2 + (sz )2 .
Similar to the electromagnetic Poynting vector, S ~Poynting = 1 E
~ × B,
~ the vector U ~
2
is a particle’s time-space flux. The energy flux is mU ~ . Particles in the rest frame have
no energy flux. This gives a definition of a massive particle’s rest frame: it is the frame
where the polarization and spin vectors are parallel, p~ × ~s = 0. This means that in
every particle’s rest frame, the plane of oscillation is orthogonal to the plane of spin.
Classically, one can visualize this as a point particle pursuing a periodic helical motion
revolving about its axis of spin while oscillating normal to the plane of spin like a tiny
vortex. As the particle increases velocity, the angle between the planes decreases until
they become parallel at the speed of light. This also implies that, not only are the
polarization and spin velocities of the same magnitude but they are plus or minus the
same direction, pi = ±si .
The second conserved current in the rest frame is,
 
0 0 0
0 0 0 
m  −Pxx −Pxy −Pxz
Jai = 0 , (9)

0 0
2  −Pyx −Pyy −Pyz 

0 0 0
−Pzx −Pzy −Pzz
where Pij0 = pi pj − si sj . Because pi = ±si , which follows from p2 = s2 and p~ × ~s = 0,
Pij0 = 0. This implies that the current vanishes in all frames.
The conserved current for massive particles in the rest frame has a convenient
representation in spherical coordinates,
m
J00 = Jrr = (p2 + s2 ) (10)
2
while the second current is
0 m
Jrr = (p2 − s2 ) = 0. (11)
2
Because the kinetic energy of spin and polarization equal, we have p2 = s2 = 1 in the
rest frame which implies
J00 = Jrr = m (12)
Note that matter and antimatter both have positive relativistic mass (as do
photons) as a natural consequence of the positiveness of the polarization and spin energy.
This energy does not change its sign. For example, in a conversion experiment from a
electron-positron pair in a head-on collision to a pair of photons emerging in opposite
directions, e− + e+ = 2γ, because the total momenta sum to zero (neglecting spins), the
relativistic masses going in must sum to those coming out. Hence,
e− e + γ
J00 + J00 = 2J00 . (13)
Because electrons and positrons have the same relativistic mass if travelling at the same
e− γ
speed, we have J00 = J00 . This implies that matter, antimatter, and photons are all
gravitationally attracted to one another and, further, that antimatter behaves the same
way in a gravitational field as matter. (This result is already predicted by general
relativity.)
Vortex particle model in a renormalizable quantum gravity 8

3. Yang-Mills theory and derivation of the field equations

Yang-Mills theory describes forces as exchanges of gauge bosons based on a group


symmetry. The Standard Model of quantum field theory currently contains three forces,
electromagnetism, the weak force, and the strong force in a U(1)⊗SU(2)⊗SU(3) group
symmetry. In this section, I show how gravity can also be incorporated into the Yang-
Mills framework.
A conventional SU(N) Yang-Mills theory has action,
Z
1
S = − 2 d4 x Fµν F µν , (14)
4g
where g is the coupling constant and
Fµν = ∂µ Aν − ∂ν Aµ − i[Aµ , Aν ], (15)
is the “force” where Aµ is a matrix potential from the SU(N) group, written with indexes
as Aµab . For the U(1)⊗SU(2)⊗SU(2) group these are 5 × 5 matrices.
It is more convenient, however, to express Yang-Mills equations in group component
form, where the action is,
Z
1
S=− d4 x Fµν
a
F µνa , (16)
4
a
Fµν = ∂µ Aaν − ∂ν Aaµ + f abc Abµ Acν , (17)
and f abc is the group symbol and g = 1.
The U(1)⊗SU(2)⊗SU(2) potential can be written as a sum of generators,
Aµ = Gµ 0 (J 0 ) + Gµ i (J+i ) + Hµ i (J−i ), (18)
where the potential Gµ 0 couples to Jµ 0 , Gµ i to Jµ i , and Hµ i to Jµ0 i .
Following the Lorentz algebra of Appendix Appendix A with group symbol defined
by A.10, the “forces” of gravity are,
Fµν 0 = ∂µ Gν 0 − ∂ν Gµ 0 , (19)
Fµν i = ∂µ Gν i − ∂ν Gµ i + ijk Gµ j Gν k , (20)
and
0 i
Fµν = ∂µ Hν i − ∂ν Hµ i + ijk Hµ j Hν k , (21)
where i, j, k = 1, 2, 3. Thus, the action of gravity is,
Z
1 0 i 0µνi
Sgravity = − d4 x Fµν b F µνb + Fµν F . (22)
4
Here the group indexes b = 0, 1, 2, 3 and i = 1, 2, 3 refer to rest frame coordinates of
momentum.
The classical equations of motion of any Yang-Mills theory can be found by the
variation of the action, δS = 0, via the Euler-Lagrange equations,
 
µ ∂L ∂L
∂ − = 0, (23)
∂(∂ µ Aν a ) ∂Aν a
Vortex particle model in a renormalizable quantum gravity 9
a
where L = Fµν Faµν is the Lagrangian. Evaluating the Euler-Lagrange equations for the
Yang-Mills Lagrangian, the equations of motion are,
∂ µ Fµν a + f abc Fµν b Aµc = −4πJν a , (24)
where Jν a is the conserved current.
Applying the definition of the group symbol again and the previous definitions 19,
20, and 21 to the field equations (24) yields the following equations:
∂ µ Fµν 0 = 4πJν 0 , (25)
∂ µ Fµν i + ijk Fµν j Gµk = 4πJν i , (26)
0 i 0 j µk
∂ µ
Fµν ijk
+ Fµν H = 4πJν0 i . (27)
where the usual negative sign has been absorbed into the current. This will require
another flipped sign in the equation for test particle motion in Sec. 4. Again, the group
indexes 0 and i refer to rest frame spacetime coordinates.
Using the definition of the cross-product the last two equations (26 and 27) can be
rewritten in a vector form [5],
∂ µ F~µν + F~µν × G
~ µ = 4π J~ν , (28)
∂ µ F~µν
0
+ F~µν
0
×H~ µ = 4π J~0 ,
ν (29)
where
F~µν = ∂µ G
~ ν − ∂ν G
~µ + G
~µ × G
~ ν, (30)
F~µν
0 ~ ν − ∂ν H
= ∂µ H ~µ + H
~µ × H
~ ν, (31)
are the vector forms of the force fields.
Because the group indexes correspond to spatial indexes in the rest frames of each
of the source particles, we can derive a generally covariant form of these three equations
and combine them into two covariant equations. Gauge transformations are of the form,
G0µ a = Gµ a − ∂µ θa , (32)
for some set of functions θa . Therefore, constant Lorentz gauge transformations do not
change the potential and all gauge transformations represent accelerations in the rest
frame of the particle. Because energy/momentum is conserved in all frames, including
those at a non-zero velocity relative to any matter particles, the group indexes must be
converted to spacetime indexes, a → µ. Given a Lorentz transform, Λαβ , with 4-velocity
uβ = Λ0β , the 3-vector permutation symbol transforms as, ijk → αβρσ uβ where αβρσ is
the Levi-Civita symbol. In the rest frame, uβ = (1, 0, 0, 0), the four-space permutation
symbol reduces to the three-space. Therefore, the covariant field equations are,
∂ µ Fµν α + αβρσ uβ Fµνρ Gµ σ = 4πJν α , (33)
0 α 0
∂ µ
Fµν + αβρσ
uβ Fµνρ H µσ = 4πJν0 α , (34)
where
Fµν α = ∂µ Gν α − ∂ν Gµ α + αβσρ uβ Gµσ Gνρ , (35)
0 α
Fµν = ∂µ Hν α − ∂ν Hµ α + αβσρ uβ Hµσ Hνρ , (36)
Vortex particle model in a renormalizable quantum gravity 10

and the velocity uβ is the 4-velocity of the source of the conserved current and, thus, is
a velocity field. In the rest frame, Hν 0 = 0, Jν0 0 = 0, and these equations return to the
forms given in 25,26, and 27.
By Noether’s theorem, for any matter action, SM , e.g., Dirac’s, we have the relation,
δSM
J µν = , (37)
δ(Gµν )
δSM
J 0µν = , (38)
δ(Hµν )
and the continuity equation is,
∂ν j να = 0, (39)
where
j να = J να − αβσρ uβ Gµσ F µν ρ , (40)
[16].
A relaxed version of the equations of motion can be achieved by assuming the
appropriate gauge (as with the de Donder or harmonic coordinates of GR [17]), giving
the following equations,
Gµ ν = 4πΩµ ν , (41)
Hµ ν = 4πΩ0µ ν , (42)
where  = −∂ 2 /∂t2 +∇2 is the ordinary d’Alembertian operator; Ωµ ν and Ω0µ ν represent
the total contribution to the potential by both the gravitational field itself and any
matter sources. These can be solved with the usual iterative integral methods [17] as
shown in Appendix Appendix B.

4. Equation of test particle motion

The geodesic equation of general relativity is its defining feature, and, in this section,
I derive the same equation of test particle motion (or compact spherically symmetric
body) for the Yang-Mills theory.
For a particle of small mass m, following a curve xµ (σ) (in the observer’s frame)
the action is,
Z σ1
S= dσ L, (43)
σ0
where L is the Lagrangian for the appropriate theory.
In general, for a Yang-Mills potential, Aµ a , and particle current Jµ a , the Lagrangian
is,
dxµ dxν 1
L = gµν − Aµ a J µa . (44)
dσ dσ m
Often the square root of the metric appears in the Lagrangian, but the first term, the
one containing the metric, is proportional to the kinetic energy of the particle rather
than the arclength as is standard in quantum field Lagrangians [5].
Vortex particle model in a renormalizable quantum gravity 11

The equations of motion are given by the Euler-Lagrange equation,


 
d ∂L ∂L
− = 0, (45)
dσ ∂(dxν /dσ) ∂xν
where Jν a is a function of dxν /dσ. This minimizes the action over the course of the
particle. For electromagnetism, Jν = quν where q is the rest charge and uν = dxν /dσ.
Applying the Euler-Lagrange equations in this case, yields the Lorentz force equation,
duµ /dσ = q/mF µν uν , [13].
General relativity has a Lagrangian for test particle motion that reflects curved
spacetime,
dxµ dxν
L = gµν , (46)
dσ dσ
where gµν = ηµν + hµν , and, applying the Euler-Lagrange equation to this yields the
geodesic equation,
d2 xλ 1 λν dxµ dxρ
+ g [∂ρ hµν + ∂µ hρν − ∂ν hµρ ] = 0. (47)
dτ 2 2 dτ dτ
In order to derive the geodesic equation from the Yang-Mills theory of the previous
section, let the Lagrangian for particle motion be,
dxµ dxν 1
L = gµν + Gµν J µν , (48)
dσ dσ m
where gµν is the flat spacetime metric and J µν is the particle’s current given its energy
of spin and polarization as defined in the section on conserved currents (Sec. 2). The
second current is always zero, J 0µν = 0, in the classical limit and, hence, the potential
Hµν = 0 as well. These tensors will be neglected henceforth. Also, normally, in the
Lagrangian, the potential is subtracted from the kinetic energy, but, because the right
hand side of the field equation (33) is positive, the second term in 48 containing the
potential is added. The change of sign, Jµν → −Jµν , puts both the field equation and
the Lagrangian into standard Yang-Mills form.
To derive the geodesic equation, 48 must be in the same form as 46. This can
be shown from the following two properties of the conserved currents and potentials
described in Sec. 2 and 3: (1) both currents and potentials are traceless and (2) currents
are spherically symmetric in the rest frame.
Let the rest frame of the particle (specifically, the frame where the energy flux,
J0µ and Jµ0 , vanishes) be x̄µ = (σ, r̄, θ̄, φ̄) where the particle is at (σ, 0, 0, 0). Let
ḡµν = diag(−1, 1, r̄2 , r̄2 sin2 θ̄) be the metric, Ḡµν , the potential, and J¯µν , the current in
frame x̄µ . From 12 the rest frame current is J¯00 = J¯r̄r̄ = m; therefore, 48 is,
L = ḡσσ + Ḡσσ + Ḡr̄r̄ . (49)
At the particle’s location, r̄ = 0, the metric degenerates into a 2-D metric,
ḡµν = (−1, 1, 0, 0). Let the potential be traceless, i.e. ḡµν Ḡµν = 0, Ḡσσ = Ḡr̄r̄ .
Remark: The assumption that the potential is traceless is not a restrictive
assumption because one can always find a gauge where the potential is traceless.
Without loss of generality, let us assume a space in Cartesian coordinates (t, x, y, z),
Vortex particle model in a renormalizable quantum gravity 12

which in a Minkowski spacetime, one may always do. Given a potential Gµν with trace
Rt
η µν Gµν = f (t, x, y, z), choose a function θ0 = 0 dt0 f (t0 , x, y, z). Now, ∂0 θ0 = f . Change
the gauge, G0µ0 = Gµ0 − ∂µ θ0 . Since off-diagonals do not affect the trace, the resulting
potential is traceless.
Replacing Ḡr̄r̄ with Ḡσσ in the Lagrangian, we have,
L = ḡσσ + 2Ḡσσ , (50)
or, in the observer’s frame,
dxµ dxν
L = (gµν + 2Gµν ) , (51)
dσ dσ
which, as desired, is the same as 46.
The meaning of the doubling of the potential comes from the definition of mass.
Since all matter derives its mass from intrinsic kinetic energy of spin and polarization, if
two bodies are stationary with respect to one another, A and B, with stress-energy
tensors JAµν and JBµν , then the gravitational field generated by JA00 attracts B by
generating a potential that couples to JB00 and vice versa. The same result occurs
independently (since they are stationary) in that JAij generates a field that attracts JBij
and that force is the same magnitude and direction.
Taking the variation of the action, δS = 0, gives the equation of motion for the
particle (the “geodesic”),
d2 xλ 1 λν dxµ dxρ
+ γ [∂ρ γµν + ∂µ γρν − ∂ν γµρ ] = 0, (52)
dτ 2 2 dτ dτ
where γµν = gµν + 2Gµν , which is the geodesic equation, also as desired.
The choice of parameterization determines whether the spacetime arclength,
r
dxµ dxν
−gµν , (53)
dτ dτ
is constant. In any flat spacetime motion model, there are at least two choices of
parameterization, τ : (1) the proper time of the particle set by the ticks of a clock
moving with it or (2) the constant velocity parameterization where 53 is the speed of
light in vacuum, c. General relativity is constructed so that these parameterizations are
one and the same, but, in the case of all other forces, the two parameterizations are the
same only when a particle is moving at constant velocity. In this paper, I use the proper
time/affine parameterization where 53 is not constant.
The combined metric and potential,
r
dxµ dxν
−γµν = C, (54)
dτ dτ
is constant however. The Legendre transform of the integrand of S gives the Hamiltonian
[13],
dxµ
H = Pµ −L (55)

Vortex particle model in a renormalizable quantum gravity 13

where
∂L dxν
Pµ =   = 2 (gνµ + 2Gνµ ) , (56)
∂ dxµ dτ

is the conjugate momentum. Combining 51, 55, and 56, the Hamiltonian is the same as
the Lagrangian,
dxµ dxν dxµ dxν
gµν + 2Gµν = E, (57)
dσ dσ dσ dσ
and is a constant of motion where E = 1 for timelike and E = 0 for null trajectories.
Thus, the Lagrangian and Hamiltonian of gravity are both constants of motion because
of the two index nature of the G-field.

5. Strong Equivalence Principle

Equivalence principles are defining features of gravitational theories. Newton’s theory


included the Weak Equivalence Principle (WEP) where an object’s weight was assumed
to be proportional to its mass. Einstein then developed the equivalence principle named
for him, the Einstein Equivalence Principle (EEP), which stated that local experiments
in free fall are independent of position and velocity. When strongly gravitating bodies are
taken into account, the validity of the equivalence principles come into doubt however as
internal structure may cause violations of WEP such as the (never observed) Nordtvedt
effect [11].
The Strong Equivalence Principle (SEP) is a natural extension of early equivalence
principles. It has three conditions: (a) an object’s weight is proportional to its mass
(WEP) for self-gravitating bodies as well as test bodies, (b) the outcome of any local
experiment is independent of the velocity of the (freely falling) apparatus, and (c) the
outcome of any local experiment is independent of where and when in the universe it is
performed.
Although a rigorous proof has not been found, GR is the only known metric theory
that strictly satisfies SEP. Because experiment has never found a violation of SEP,
any new gravitational theory must satisfy strict constraints on SEP violations. In the
following I show that the Yang-Mills theory satisfies SEP strictly (1) because it is not a
metric theory and its prior geometry is not affected by gauge transformations and (2)
because of its gauge symmetry.
By satisfying all three conditions, the Yang-Mills theory satisfies SEP. The Yang-
Mills theory satisfies (a) because it fully conserves mass/energy, linear and angular
momentum (by design) and because it is always possible to find a gauge that eliminates
the gravitational field at the boundary between a local, spherically symmetric, compact
system and the external environment, an essential ingredient for a spherically symmetric
self-gravitating body, even a neutron star or black hole, to be regarded as a point particle.
Unlike bimetric theories where this cannot be done, here the gauge can be changed
without changing the coordinate system. Thus, the Minkowski metric field ηµν does
Vortex particle model in a renormalizable quantum gravity 14

not change when the gauge of the gravitational field Gµν changes. By the arguments of
Section 20.6 of [7] also found in [17], if the field of a self-gravitating body asymptotically
approaches “flatness” (Gµν (R) = 0) at some distance, R, considered to be the boundary
of the local system, this is sufficient to guarantee that a body’s self-gravitation and
other internal structure does not affect its motion. Therefore, (a) is satisfied.
It also satisfies (b) by local Lorentz covariance which means that experimental
outcomes are independent (by gauge covariance) of Lorentz boosts and rotations and
satisfies (c) by having no preferred location/time in the field equations. Unlike Rosen’s
bimetric theory which has a local gravitational constant that depends on the field, the
Yang-Mills theory has a non-location specific gravitational constant, G, which is part
of the Yang-Mills coupling constant G/c4 . Thus, all three conditions are met, and the
Yang-Mills theory satisfies SEP.

6. Experiments and Observations

Numerous observations, starting with light bending in 1919, have been made to attempt
to confirm predictions of general relativity. These include gravitational time dilation,
redshift, and light bending which have been measured within the solar system as well as
perihelion precession of the planets and binary pulsar precession [6]. Other observations
do not agree with the original theory and have required ad hoc modifications to the
Einstein field equations, yielding empirical models. In order to show that the theory
agrees with as many observations as possible, in the following, I derive (1) a static,
spherically symmetric solution to the field equations, (2) the 1PN equations of motion
for a binary system, (3) the radiation reaction for binary pulsar inspiral, and (4) predict
the accelerating expansion of the universe via a homogeneous, isotropic model.

6.1. Exact spherically symmetric solution to the field equations


The static, spherically symmetric Schwarzchild solution is one of the most important
solutions to the Einstein field equations. The corresponding solution to the field
equations, 33 and 34, is identical up to linear order. Let spacetime be covered by
spherical coordinates (t, r, φ, θ) where θ is colatitude and let the metric be the flat
spacetime metric [6],
ds2 = −dt2 + dr2 + r2 dθ2 + r2 sin2 (θ)dφ2 . (58)
The only allowable non-zero components of the potentials are Grr and Hrr . Converting
to Cartesian coordinates by tensor calculus,
∂r ∂r
Gij = Grr i j , (59)
∂x ∂x
~ µ ×G
the cross-products G ~ ν are zero. Evaluating F~µν , taking the cross-products F~µν × G
~ µ,
and converting back to spherical coordinates, we find that these also evaluate to zero.
Therefore, applying the same procedure to the equation for Hrr , in the spherically
Vortex particle model in a renormalizable quantum gravity 15

symmetric case, the field equations (33 and 34) are linear. Choosing a harmonic gauge
we have,
∇2 Grr = 4πJrr , (60)
2 0
∇ Hrr = 4πJrr . (61)
Placing the source at the origin, in spherical coordinates, its tensor has components
0 0
J00 = Jrr = ρ such that its density is ρ = M δ(r). From Sec. 2, we have J00 = Jrr =0
and Hµν = 0. The solutions to the field equations are,
M M
G00 = , Grr = . (62)
r r
Adding 2Gµν to the metric gives,
 
2M 2M
γ00 = − 1 − , γrr = 1 + , γθθ = r2 , γφφ = r2 sin2 (θ), (63)
r r
where γµν = gµν + 2Gµν . This is the linearized Schwarzchild solution to the Einstein
equations [6] but, for this Yang-Mills field theory, it is exact for all spherically symmetric
bodies, including black holes.

6.2. Particle trajectories


The spherically symmetric vacuum solution 63 combined with the equations of motion
in a negligible cycleton field are clearly identical to those of general relativity within
the solar system, neglecting terms of quadratic order in M/r. Calculations of particle
trajectories in a static, spherically symmetric, spinless potential field are identical
to those of general relativity. The only difference is in the radial potential. The
Schwarzchild radial potential grr = (1−2M/r)−1 differs from the radial potential derived
in Section 6.1 to quadratic order,
γrr = 1 + 2M/r ≈ (1 − 2M/r)−1 + O((M/r)2 ). (64)
No tests within the Solar System, whether with respect to the Sun, Earth, or
another body (the most precise tests have been done near the Earth) have achieved
better than linear order in Schwarzschild coordinates [7, 6] because of the weak fields
involved. If M = Gm/c2 is the Schwarzchild radius and m is the mass in other units
(e.g., kilograms) at the surface of the Sun, 2M /r ≈ 4.25 × 10−6 and at the surface of
the Earth 2M♁ /r ≈ 1.4×10−9 . Both are much smaller than necessary to measure higher
order effects where imperfections in density (e.g., mountains) would make measurements
difficult to verify.
Predictions of gravitational time dilation and redshift, meanwhile, are mathemati-
cally identical in both theories (to any order) given a static, spherically symmetric field
with redshift proportional to 1 + 2M/r. Recent high precision measurements of red-
shift using cesium atoms in a laboratory agree with predictions of both theories [18].
Therefore, no tests done within the Solar system to date disprove either theory. For
strong field tests, we look outside the solar system to binary pulsars and, further on,
cosmology.
Vortex particle model in a renormalizable quantum gravity 16

6.3. Post-Newtonian Equations of Motion


The binary pulsar system B1913+16 discovered in 1974 provided one of the first tests
of strong field general relativity [17]. Because of the regular radio pulses of the star
which is orbiting a relatively inert body such as a dead pulsar, henceforth called the
companion body, allowed extremely precise measurements of the pulsar’s orbit, the small
deviations of the orbit from Kepler’s laws can be measured, including orbital damping
caused by gravitational radiation, which is a higher order effect than orbital precession
[17]. Observation of this system over the past thirty-five years have only increased the
accuracy of the measurements. Recently, the double binary pulsar system PSR J0737-
3039A/B with two active pulsars orbiting each other has been observed providing even
greater confirmation of GR’s predictions [8]. The following demonstrates that the field
equations of Sec. 3 also predict the observations of these pulsar systems.
Although it is not a metric theory, its close resemblance to a metric allows the
G-field to be expressed in a form similar to PPN form. Therefore, in this section, I
apply this framework to show that the many-body equations of motion agree with those
of general relativity up to first post-Newtonian corrections. The parameterized post-
Newtonian (PPN) equations of motion of gravity are the primary vehicle by which not
only different theories of gravity (including GR) are compared but provide the equations
for predicting the motion of two or more gravitating bodies. Tests of these equations
include the measurement of planetary motion within the Solar system for weak fields
and binary pulsar precession and gravitational radiation for somewhat stronger fields
[7, 11].
For a binary pulsar system the relevant metric in the post-Newtonian coordinate
system for general relativity is ([11], 11.52)
X
g00 = − 1 + 2 ma /|~x − ~xa (t)| + O(4 ), (65)
a=1,2
3
g0j = O( ), (66)
!
X
gij = δij 1+2 ma /|~x − ~xa (t)| + O(4 ), (67)
a=1,2

with m1 and m2 being the masses of the bodies and ~x1 and ~x2 their positions in the
appropriate harmonic coordinate system. From the relaxed field equations (41),
X ma
G00 = + 2Ψ + O(6 ), (68)
a=1,2
|~
x − ~
x a |
where
X ma va2 X ma mb
Ψ(t, ~x) = + , (69)
a=1,2
|~
x − ~
x a | a,b=1,2,a6=b
|~
x − ~
x a | |~
x a − ~
x b |
X ma vaj
G0j = −2 + O(5 ), (70)
a=1,2
|~
x − ~
x a |
Vortex particle model in a renormalizable quantum gravity 17

and
X ma
Gij = δij + O(4 ). (71)
a=1,2
|~
x − ~
x a |

The velocities of the bodies are ~v1 = (v11 , v12 , v13 ) and ~v2 = (v21 , v22 , v23 ) and va = k~va k.
(See Appendix Appendix B for a derivation of these potentials.)
Dropping the extra terms from Gµν for this system that exceed the required order,
the above metric easily matches γµν = ηµν + 2Gµν . Since the theory satisfies SEP, for
any set of compact nearly-spherical bodies where tidal forces may be neglected, the
geodesic equation applies. Because the geodesic equation of general relativity and 52
are identical, the post-Keplerian parameters of periastron advance, hω̇i, time delay, γ 0 ,
and Shapiro delay parameters, r and s, are the same as well. Thus, the equations for
these parameters are the same as in general relativity.
The final post-Keplerian parameter, orbital speed-up caused by gravitational
radiation, Ṗb , derives from the quadrupole formula from the Yang-Mills field equations
discussed in the following section.

6.3.1. Orbital speed-up of a binary pulsar system Gravitational radiation was first
addressed by Einstein shortly after the publication of general relativity, where he
demonstrated that the radiation is primarily quadrupolar in contrast to the primarily
dipolar radiation from electromagnetic sources [19]. The energy loss of a binary system
of compact stars was first demonstrated in a paper by Peters and Matthews [20], who
derived the energy loss of binary stars in Keplerian orbit leading to the formula for the
orbital speed-up,
 −5/3
192πG5/3 Pb
 
2 −7/2 73 2 37 4
Ṗb = − (1−e ) 1 + e + e mp mc (mp +mc )−1/3 ,(72)
5c5 2π 24 96
with mp and mc the masses of the pulsar and companion body respectively and e the
orbital eccentricity [21]. Although this equation was developed for Keplerian orbits, it
has been applied to post-Keplerian orbits since the only requirement is that the orbit be
elliptical, and, although much higher order derivations have been made [11], only this
equation has been tested. In this section, I demonstrate an identical increase in orbital
speed for the Yang-Mills theory via the same techniques but omit a general discussion
of gravitational waves. A detailed discussion of multipole expansions for gravitational
radiation in general relativity can be found in [22].
Orbital speed-up is a function of energy loss, i.e. radiation-reaction, which to lowest
multipole order in General Relativity is ([7], 36.31),
dE 1 d3 Qjk d3 Qjk
=− h i, (73)
dt 5 dt3 dt3
where Qµν is the reduced quadrupole moment (the trace free part of the second moment
of the mass distribution) such that Qµν = qµν − 31 δµν q,
Z
qµν = d3 x ρ0 xµ xν , (74)
Vortex particle model in a renormalizable quantum gravity 18

and q = qµ µ . The Peters-Matthews formula for energy loss of a binary system ([11],
10.80),
 2 2 
dE µm 8 2 2

=− 12v − 11ṙ , (75)
dt r4 15
where m = m1 + m2 , µ = m1 m2 /m, v = |~v1 − ~v2 |, and r = |~x1 − ~x2 |, can be found
from the linearized vacuum equations of general relativity, h̄µν = 0, [20, 7] by the
quadrupole relation,
2
h̄jk (t, ~x) = q̈jk (t − r), (76)
r
for the spatial part of the radiation field ([7], 36.50) where r is the distance from the
source.
To show that both GR and the Yang-Mills theory have the same orbital speed-up,
we compare the equations for the two theories. Peters and Matthews derives the energy
loss using the linearized Einstein vacuum equations:
gµν = ηµν + hµν , (77)
such that
h̄µν = 0, (78)
where h̄µν = hµν − 12 ηµν hλ λ .
The linearized Yang-Mills equation,
Gµν = 4πJµν , (79)
is similar to the linearized equation for GR but with a coupling constant half as large
because of the doubling of the potential that occurs in the geodesic equation.
A plane wave solution to the wave equation,
Gµν = 0, (80)
is
Gµν = aeµν cos(ωt − ~k · ~x), (81)
where a is the amplitude, eµν is a symmetric, traceless, transverse, and unitary
polarization tensor (as defined in [20]), and ω is the frequency.
An important difference between electromagnetism and gravity is that, whereas the
primary electromagnetic radiation for charged bodies in motion is dipole, the primary
gravitational radiation for massive bodies in motion is quadrupole. The following is
a demonstration from the YM theory, following the outline in [11], Chapter 10. The
multipole expansion of the plane wave solution (81) is,
X∞ Z
−1
Gµν = 2r (1/m!)(∂/∂t) m
d3 x0 Jµν (t − r, ~x0 )(~n · ~x0 )m . (82)
m=0

to 1PN order, where ~n = ~x/r. Because of the gauge condition, ∂µ Gµν = 0, and the
retarded potential, we can use the following relations,
∂0 G0k = nj ∂0 Gjk , ∂0 G00 = nj nk ∂0 Gjk , (83)
Vortex particle model in a renormalizable quantum gravity 19

and only need to determine the Gjk components [11]. Because, to order, ∂µ J µν = 0 by
conservation and that the source is symmetric, a useful relation is,
Z Z
(∂ /∂t ) d xJ x x = 2 d3 x J jk
2 2 3 00 j k
(84)
Therefore, 82 becomes,
Z 
−1 2 2 3 00
Gij = r (∂ /∂t ) d x J (t − r, ~x)xi xj + higher order. (85)

This means that monopole and dipole moments of J ij may be expressed as time
derivatives of quadrupole moments of J 00 . To post-Newtonian order J 00 = ρ, the mass
density, and we now have the quadrupole approximation used to derive the reaction.
Note that at orders higher than 1PN, the source is no longer symmetric.
For the binary system the integral simplifies to,
1
Gij = q̈jk (t − r). (86)
r
Since from 76 we have 2Gij = h̄ij and because h̄ij solves the vacuum equation we also
have h̄ij = hij , the speed-up formula (72) follows as in general relativity: Taking a Taylor
expansion of 86 in powers of r to extract the radiation-reaction potential and applying
the geodesic equation 52, one determines the reaction acceleration of the bodies ([7],
Sec. 36.11). These demonstrations imply that the inspiral of orbiting binary pulsars
is the same in both theories. In the next section, I will discuss an area where the two
theories fundamentally diverge: the accelerating expansion of the universe.

6.4. Cosmology
While general relativity is unable to account for the accelerating expansion of the
universe without a cosmological constant of uncertain origin, the field equations 26
contain a built-in cosmological term (which is not necessarily constant) than derives
from the gauge symmetry of gravity itself. In this section, I derive the equations for an
isotropic, homogeneous Robertson-Walker universe for the Yang-Mills theory.

6.4.1. Isotropic, Homogeneous Universe Observations of the cosmic microwave


background and statistical counts of the distribution of galaxies imply that the universe
is statistically isotropic [6]. By the Copernican principle that we do not occupy a
privileged location in the universe, the universe must also be homogeneous. An isotropic,
homogeneous universe has a very simple dynamical description. For a perfect fluid
universe, the stress-energy-momentum tensor in special relativity has the standard form,
T µν = ρuµ uν + p(uµ uν + g µν ), (87)
where ρ is mass density and p is pressure [6]. The fundamental Friedmann equations
for general relativity are,
4πρa2
ȧ2 + k = , (88)
3
3ä
= − 4π(ρ + 3p), (89)
a
Vortex particle model in a renormalizable quantum gravity 20

where k is the curvature constant [10] and


a2 (τ )
g00 = −1, grr = , gθθ = a2 (τ )r2 , gφφ = a2 (τ )r2 sin2 θ, (90)
1 − kr2
is the metric. Recent observations that the universe’s density, the sum of baryon, dark
matter, and dark energy densities, is near or at the critical value, Ωb + Ωc + ΩΛ ' Ωcrit ,
and the universe is spatially flat, k ≈ 0 [4, 9]. Therefore, for the rest of this section,
k = 0. From the conservation of energy,
∂ρ 3ȧ
+ (ρ + p) = 0, (91)
∂τ a
is a fundamental equation for the evolution of matter and energy.
The Robertson-Walker assumptions of isotropy and homogeneity are essential
starting points for the study of cosmology, but the equations for the Robertson-Walker
universe derived from the Einstein field equations are unconfirmed to a large extent
because the scale factor a cannot be measured directly and, even though precision
measurements are now possible of derived quantities such as the Hubble constant
H = ȧ/a, whether the universe obeys the quantitative dynamics these equations imply
(as opposed to qualitative dynamics, e.g., that the universe is expanding which any
number of theories could predict) is unknown. We only know, to the grossest degree,
how the universe as a whole behaves and has behaved in the past, and these results have
only arrived within the past decade and are still the subject of a great deal of study.
Thus, the era of precision cosmological dynamics is still some distance in the future and
much of current observations are intended only to determine the values of theoretical
parameters in the present day and at what time in the past certain events, such as
reionization, occurred. These values necessarily depend on the theory being assumed
and the intention of the observations is to rule out competing ideas such as whether the
universe has curvature.
Turning now to the Yang-Mills theory. In a comoving coordinate system, let
g00 = −1, gij = a2 (t)δij and choose a gauge such that G00 = 0 and Gij = b(t)a2 (t)δij .
This means that, as with general relativity, the coordinate system moves with matter
as it expands so that distances remain constant on average.
Because of the comoving coordinate system, we make the substitution in the field
equations ∂µ → ∇µ , where ∇µ is the covariant derivative,
∇µ Fµν α + αβρσ uβ Fµνρ Gµ σ = 4πJν α , (92)
where
Fµν α = ∇µ Gν α − ∇ν Gµ α + αβσρ uβ Gµσ Gνρ , (93)
the velocity, uβ , is the 4-velocity of the source,
∇µ Gν α = ∂ µ Gν α − Γλ µν Gλ α + Γα µλ Gν λ , (94)
and
∇µ Fµν α = ∂ µ Fµν α − Γλµ µ Fλν α − Γλµ ν Fµλ α + Γαµ λ Fµν λ . (95)
Vortex particle model in a renormalizable quantum gravity 21

In the comoving coordinates, the conserved current is J00 = ρ and Jij = pa2 δij .
Evaluating the field equations 33, we have,
3ȧ(aḃ + bȧ)
= 4πρ, (96)
a2
and the next three (all the same) are,
ȧḃ äb ȧ2 b
−3 − b̈ − − 2 + 2b3 = 4πp. (97)
a a a
From 39 we have the equation for matter,
3ȧ
ρ̇ + (ρ + p − 2b3 ) = 0. (98)
a
Therefore, the gravitational potential has negative pressure.
By the vortex model all matter has p = ρ/3 which simplifies 98 to
 
4ȧ 3 3
ρ̇ + ρ − b = 0. (99)
a 2
The system of three differential equations (96,97, and 99) govern the gravitational
behavior of matter in this universe, but they have no direct solution because of the
nonlinearity, b3 . For the linearized equations, where this term is neglected, 97 becomes,
ȧḃ äb ȧ2 b
−3 − b̈ − − 2 = 4πρ/3, (100)
a a a
and 99 becomes
4ȧ
ρ̇ + ρ = 0. (101)
a
The equations 96,100, and 101 have the non-trivial solution,
a(t) = C1 t + C2 , (102)
C3
ρ(t) = 4 , (103)
a
C4 a − 4πC
3C12
3

b(t) = . (104)
a2
The boundary condition are given in terms of present day, t0 , and origin, t = 0,
parameters. The boundary conditions are a(0) = 0 and ρ(t0 ) = C3 /a4 (t0 ) = ρ0 or
C3 = ρ0 a40 for the present matter density where a0 = a(t0 ). (Note that a(0) = 0 implies
a singularity in the coordinate system and not spacetime itself.) Let C1 = a0 /t0 . The
value of C4 is determined by the current value of the gravitational potential, b0 = b(t0 ).
The rate of expansion of the universe, v, is related to the Hubble constant by,
dR
v≡ = HR, (105)

where R is the distance between two isotropic observers [6], and

H = = C1 /C1 t = 1/t, (106)
a
Vortex particle model in a renormalizable quantum gravity 22

where currently accepted value of the Hubble constant is 70.1 ± 1.3(km/s)/M pc [4].
The final equations are,
a(t) = a0 t/t0 , (107)
 a 4
0
ρ(t) = ρ0 , (108)
a
(b0 + 4πρ0 t20 /3)a0 4πρ0 t20
b(t) = − . (109)
a 3(a/a0 )2
Note that the gravitational potential, Gij = b(t)a2 δij increases linearly in a in the
comoving frame which means that, because it is a tensor, in the non-comoving frame it
decays like 1/a. Thus, as t → ∞, Gµν → 0 as the universe expands to infinity. Because
it is inversely proportional to the scale factor, for most of the history of the universe, the
gravitational field has been considerably smaller than the scale factor. Thus, redshift
correlates well to expansion rate.
Critically, the Hubble constant remains proportional to the inverse of the age of
the universe even as the scale factor expands linearly with time (this is true of any
scale factor of the form a(t) = Ctx ). Thus, age estimates remain the same as for GR
with the universe at about 13.73 billion years old. The vortex model also eliminates the
distinction between radiation-dominated and matter-dominated eras because matter is
simply localized “radiation”. This does not mean that there is no distinction because
radiation, by travelling at the speed of light, naturally has different features than matter.
Indeed, the cessation of the thermal equilibrium of radiation with matter in the early
universe resulting in the emission of the CMB [10] is a consequence of this difference.
The differences, however, are eliminated by the isotropic-homogeneous assumptions.
Also crucial is that the temperature, T ∼ 1/a, be large enough in the early universe
to form elements (nucleosynthesis). This occurs, in the currently accepted theory,
between 2 and 1000 seconds after the initial explosion. Since the YM theory predicts an
expansion (in the sense of decreasing density) with no other dynamical behavior such
as bounces or contractions, it requires that the universe’s initial condition be similar to
that of general relativity.
The approximate scale factor solution (107) predicts a linear expansion, meaning
that the universe expands at constant rate, which is consistent with an isotropic-
homogeneous universe of undefined size. Because the metric is fixed and the expansion
represents, not the expansion of space itself, but the expansion of the coordinate system
as matter expands, space is infinite (or at least is larger than the past lightcone of
the Earth back to the initial moment of the Big Bang). Therefore, by the isotropic-
homogeneous assumptions, the Big Bang itself was not a singularity of space, but a
period of infinite or nearly infinite matter density filling all of infinite or near infinite
space. This is not significantly different from general relativity since general relativity
also assumes that the Big Bang filled all of space. The only difference is the size of
space.
The cubic term in 99 represents negative pressure created by gravitons themselves,
discussed in the next section, but, neglecting this nonlinear term, gravity neither slows
Vortex particle model in a renormalizable quantum gravity 23

nor increases the expansion of the universe. This appears counter-intuitive but is clearly
a feature of a universe with no “outside” which is isotropic and homogeneous. It can
be proved from symmetry conditions: Because any point, ~x0 , is equidistance from all
other points in the universe, ~x ∈ R3 , and all points in the universe have the same
matter density, ρ(t, ~x) = ρ(t), every point experiences an equal gravitational force in all
directions. This indicates that gravity would not, in fact, cause the universe to slow its
expansion because every point in the universe is a Lagrange point. Hence there is no
force to cause the slowing.
A further demonstration of this fact is as follows: if we consider a particle of matter
in a homogeneous, spherical body of matter of finite size that is at the exact center,
this particle must, by symmetry conditions, experience the same force in all directions.
Thus, it experience no net force. Expanding this sphere of matter to infinite size, all
points become the “center”‡, hence no points experience a net force. Thus, for a linear
theory of gravity, the expansion of the universe proceeds unimpeded. This is not true,
however, for a nonlinear theory.

6.4.2. Accelerating Expansion Measurements of type Ia supernovae have given strong


evidence that the universe is not only expanding but accelerating this expansion. This
acceleration has been attributed to a cosmological parameter of uncertain origin, Λ,
known as the cosmological constant. Several explanations for the constant have been
proposed, the most common being vacuum energy. This is despite that the vacuum
energy of the universe has a density on the order of 1 in Planck units and the actual
value is about 10−120 , virtually, but not quite, zero [5]. The vacuum energy is essentially

an additive constant that couples to the metric’s volume element, −g, that appears in
every integral over spacetime. Because it is a standard non-Abelian gauge theory, the
YM theory’s potential does not couple to additive constants. Hence vacuum energy is
not the source of the acceleration. The YM theory however has a natural choice for the
cause of the acceleration.
Yang-Mills theories such as the U(1)⊗SU(2) unbroken electroweak theory have
nonlinear elements. In electroweak theory, gauge bosons themselves carry charge.
Because they couple to charge, they couple to one another and create nonlinearities.
The same is true in the gravitational Yang-Mills theory. Gravitons carry momentum
(but, as demonstrated by the linear U(1) equations, not mass which would shorten the
range of gravity). Because gravity also couples to momentum, gravitons couple to one
another. This creates kinetic energy in the form of pressure in the isotropic-homogeneous
universe. As 99 shows, however, this pressure,
3
pgrav = − b3 , (110)
2
‡ Classically, all points are only affected by the matter within their past lightcone back to the Big Bang,
and this forms the actual sphere. This assumes that no particles in the universe, including gravitons,
began traveling before the Big Bang.
Vortex particle model in a renormalizable quantum gravity 24

is negative resulting in acceleration. The actual value of b at the present time b(t0 ) = b0
must be determined by observation to give the rate of acceleration.
The value of this result over an ad hoc constant, Λ, is that it is entirely self-
contained within Yang-Mills theory, hence nothing has been added to the theory to
force it to agree with the empirical result that the universe is accelerating. YM theory
itself predicts that it must be and cannot be contrived to predict otherwise without
violating the symmetry group.

7. Potential Disagreements

Observations of gravity can be classed into a four basic types: (1) Newtonian gravity
which applies to very weak fields and slow velocities that pertain to most galaxies, stars,
and orbits of the outer planets, (2) weak field gravity responsible for classic effects such as
redshift, light bending, time dilation, and perihelion precession, (3) strong field gravity
responsible for binary pulsar precession, orbital damping, and gravitational waves, and
(4) cosmology where assumptions such as homogeneity and isotropy greatly simplify
equations. Included are tests of the Strong Equivalence Principle (SEP).
Going through each of the four types of tests:
(i) Newtonian gravity: Both theories subsume Newtonian gravity at very weak field
strengths and slow velocities, and this is easily demonstrated from the field
equations.
(ii) Weak Fields: As shown in section 6.2, both theories agree with classic effects
up to linear order in the Schwarzschild radius over the distance. Measurements
of redshift, light bending, perihelion precession, and time dilation have not been
made to quadratic accuracy in Schwarzschild coordinates (as opposed to isotropic
where the coordinate change introduces an artificial “measurable” quadratic term);
therefore, none of these experiments contradict the YM theory.
(iii) Strong Fields: As [8] mentions, measurements of binary pulsar precession, while
some of the most precise measurements of relativistic gravity ever made, are not
sufficiently accurate to confirm beyond the 1PN equations of motion. Orbital
damping caused by radiation reaction is a higher order effect but also not confirmed
beyond the quadrupole approximation, and neither GR nor the YM theory predict
a dipole moment. Therefore, as shown in section 6.3, none of these measurements
contradicts either theory nor any other measurements of multibody systems which
also rely on the 1PN equations. The geodesic equation 52 is also the same as in
relativity; therefore, since SEP is satisfied (Sec. 5) and compact self-gravitating
bodies behave as if they are test bodies, it agrees with the 1PN equations of motion
of general relativity. Gravitational wave measurements promise to provide higher
order estimates which may show a violation of one of the theories, but these, as
yet, are not available.
(iv) Cosmology: The highest order measurements available are cosmological. In this
Vortex particle model in a renormalizable quantum gravity 25

case, several sources including Type Ia supernovae (SN), baryon acoustic oscillation
(BAO), and the cosmic microwave background (CMB) provide data that can be
used to constrain various theories [9]. For example, first order perturbation of
the field equations about the Robertson-Walker solution combined with numerical
solvers can indicate whether observed fluctuations in the CMB agree with the
theory, and this is an important future direction of research for the YM theory.
Studies of Big Bang Nucleosynthesis (BBN) where the predicted quantities of light
elements such as Helium, Lithium, and Deuterium are compared with predicted
quantities can also constrain some theories and several studies of Tensor-Scalar
theory have been done [23, 24]. These studies tend to focus on the effects of the
scalar field on nucleosynthesis and are not directly applicable to the YM theory
which has only a single tensor gauge field. While measurements of cosmological
parameters have become remarkably precise in recent years, quantitative cosmology
is still in its infancy when compared to measurements within the solar system and
of binary pulsars and lensing. The minimum requirements, at present, are that the
theory agree with measurements of BBN and the existence, mean temperature, and
variation of the CMB. As demonstrated in Sec. 6.4, the YM theory meets these
requirements and, additionally, explains the acceleration of the universe’s expansion
without a cosmological constant. Another advantage of the YM theory is that it
predicts, indeed must assume, that the universe is flat, which is in good agreement
with observation.

8. Novel Predictions

Because the YM theory contradicts rather than modifies or extends General Relativity
but agrees with all of GR’s predictions to date, the validity of the theory requires
additional measurements of gravitational effects.
General Relativity and the Yang-Mills theory agree for the linearized Schwarzschild
metric predictions and 1PN equations of motion but, with the exception of radiation-
back-reaction where the same degree of inspiral is predicted for binary pulsar systems,
they fundamentally diverge at higher order motion. The 2PN order should be
measurable for binary pulsars within the next few decades as additional data is
collected about known systems (the more data collected the smaller the error bound).
Gravitational wave detectors such as LIGO and VIRGO may also be capable of detecting
much higher order effects (well into the quadratic or even cubic realm) from inspiraling
black holes and other closely spaced, highly massive objects. Because general relativity
and the Yang-Mills theory no longer agree at this level, these observations will determine
whether gravity is truly the curvature of spacetime or a non-Abelian gauge theory like
the other known forces.
Vortex particle model in a renormalizable quantum gravity 26

9. Quantum Theory

The details of the quantum theory of gravity are beyond the scope of this paper, but,
because it is a massless gauge boson Yang-Mills theory, the theory given in this paper
has a finite polynomial, positive definite action (16). Since generic Yang-Mills theories
are renormalizable [5], this one is as well. There is, indeed, no fundamental difference
between gravity and any other spin-1 Yang-Mills theory such as unbroken electroweak
theory.
The standard quantization scheme is appropriate with Fadeev-Popov ghost fields
used to derive Feynman rules including the graviton propagator,
−iδ ab
 
ab (1 − ξ)pµ pν
Dµν (p) = 2 ηµν − , (111)
p + i p2 + i
where a, b are group indexes [5]. In practice, however, the gravitational constant is too
small (coupling too strong) to make Feynman diagrams useful in studying black hole
singularities§ or the early universe. Lattice gauge simulations of the full theory, however,
may yield useful results.

10. Related Work

While this paper has focused on the problems with the prevailing theory of gravity,
general relativity, several other theories have been proposed to solve quantum gravity
and explain macroscopic observations. The most important empirical theory is the
Lambda-CDM model which was used in the WMAP survey [4] and combines the Einstein
field equations with a cosmological constant and cold dark matter. The field equations
33 and 34 do away with the cosmological constant and much of the complexity of the
curved spacetime model, but the YM theory does not challenge the prevailing theory of
dark matter.
The most prominent theory of quantum gravity is string theory which, rather than
being a simple theory of gravity, is an extensive modification of physical law positing
that all matter is composed of strings, attempting to unify all forces [5, 25]. Although
unification is not its purpose, in representing gravity identically to the other three forces,
the theory may possibly be unified with them at a high enough symmetry group without
requiring any significant additional physical assumptions such as strings or additional
spacetime dimensions.
The next most significant model of quantum gravity is loop quantum gravity [25].
Loop quantum gravity has a different approach from that given in this paper, further
marrying gravity to curved spacetime. In making the assumption that gravity has
only diffeomorphism covariance and is dominated by a metric geometry, it retains
the complexity of general relativity and adds to it by introducing spin foams, i.e.
discretizations of spacetime, in order to avoid blow-ups of the quantum variables. It
§ Gravitational singularities are errors of the classical theory, and quantum fluid properties likely take
over to prevent a true singularity.
Vortex particle model in a renormalizable quantum gravity 27

also makes no predictions that are currently testable. The Yang-Mills theory, on the
other hand, assumes that, like other forces, the only necessary metric is Minkowski,
requires no discretization of spacetime, is a renormalizable quantum theory, and explains
a macroscopic phenomenon, the accelerating expansion of the universe. Although
philosophical arguments exist proposing that prior geometry goes against the spirit of
relativity, there are no observational data that preclude a prior geometry, and there is
reason to suppose that the geometry of the universe is a non-gravitational phenomenon
with a non-gravitational explanation. Also, as shown above (Sec. 5), the prior geometry
of a Yang-Mills theory is not as problematic as for a bimetric theory where that geometry
must, by diffeomorphism covariance, be interpreted as a gauge field.
The main Yang-Mills approach to gravity is conformal gravity, which, although
proposing a Yang-Mills style action, retains much of general relativity, including the
cosmological constant and the spacetime metric and volume elements [12], and is not
directly relevant to the theory in this paper which makes significantly different physical
and mathematical assumptions.

11. Conclusion

The assumption of curved spacetime, while elegant, has never been a necessary
component given current experimental evidence, and, therefore, there is no need to
retain it in a theory of gravity. By assuming a Minkowski spacetime, vortex gravity
not only agrees with the observations confirming general relativity’s major predictions
but also more, including the accelerating expansion of the universe, a renormalizable
quantum theory of gravity, and a definition of mass as relativistic energy. It does
this by establishing gravity on the theoretical foundation of the Standard Model which
has far more supporting evidence than general relativity. Further confirmations of
vortex gravity may be obtained from detection of gravitational waves, observations of
supermassive objects such as black holes, and additional cosmological measurements.

References

[1] Zwicky F. On the masses of nebulae and of clusters of nebulae. Astrophysical J. 1937;86:217.
[2] Rubin V. Rotation of the Andromeda nebula from a spectroscopic survey of emission regions.
Astrophysical J. 1970;159:379.
[3] Clowe D, et al. A Direct empirical proof of the existence of dark matter. Astrophysical Journal
Letters. 2006;648:109–114.
[4] NASA. WMAP Mission Results; 2008. Updated: March 7, 2008. Accessed: March 18, 2009.
http://map.gsfc.nasa.gov/news.
[5] Zee A. Quantum Field Theory in a Nutshell. Princeton: Princeton UP; 2003.
[6] Wald R. General Relativity. Chicago: U. Chicago UP; 1984.
[7] Misner CW, Thorne KS, Wheeler JA. Gravitation. San Francisco: W. H. Freeman; 1973.
[8] Kramer M, Stairs IH, et al. Tests of general relativity from timing double pulsar. Science.
2006;314(5796):97–102.
[9] Komatsu E, et al. Five-Year Wilkinson Microwave Anisotropy Probe Observations: Cosmological
Interpretation. Astrophysical J Sup Ser. 2009;180:330–376.
Vortex particle model in a renormalizable quantum gravity 28

[10] Weinberg S. Cosmology. Oxford, UK: Oxford UP; 2008.


[11] Will C. Theory and experiment in gravitational physics. Cambridge, UK: Cambridge UP; 1993.
[12] Riegert RJ. Classical and Quantum Conformal Gravity. UCSD; 1986.
[13] Jackson JD. Classical Electrodynamics. Hoboken, NJ: Wiley; 1999.
[14] Levitt LS. The spin kinetic energy and intrinsic mass of elementary particles. Lett Al Nuovo
Cimento. 1983;36:167–170.
[15] Lorentz HA. Electromagnetic phenomena in a system moving with any velocity smaller than that
of light. Proceedings of the Royal Netherlands Academy of Arts and Sciences. 1904;6:809–831.
[16] Joshi DC, Benjwal MP, Kumar S. Classical yang mills theory in presence of electric and magnetic
charges. Acta Physica Polonica. 1985;B16(10):901–908.
[17] Will C. The Confrontation between General Relativity and Experiment. Living Review:
Relativity. 2006;9.
[18] Müller H, Peters A, Chu S. A precision measurement of the gravitational redshift by the
interference of matter waves. Nature. 2010;463:926.
[19] Einstein A. Preuss Akad Wiss Berlin, Sitzber. 1918;154.
[20] Peters PC, Matthews J. Gravitational Rational from Point Masses in a Keplerian Orbit. Phys
Rev. 1963;131(1):435–440.
[21] Weisberg JM, Taylor JH. Relativistic Binary Pulsar B1913+16: Thirty Years of Observations and
Analysis. ASP Conf Series. 2005;328:25–31.
[22] Thorne KS. Multipole expansions of gravitational radiation. Rev Mod Phys. 1980;52:299–324.
[23] Damour T, Pichon B. Big Bang nucleosynthesis and tensor-scalar gravity. Phys Rev D.
1999;59(123502).
[24] Coc A, Olive K, Uzan JP, Vangioni E. Big bang nucleosynthesis constraints on scalar-tensor
theories of gravity. Phys Rev D. 2006;73(083525).
[25] Rovelli C. Quantum Gravity. Cambridge, UK: Cambridge UP; 2004.

Appendix A. The Lorentz Group Lie Algebra

The vacuum has six degrees of symmetry that form the Lorentz group and all physical
theories must have Lorentz symmetry but only gravity has local Lorentz symmetry,
i.e. it is the field that makes observations by all observers consistent no matter their
relative motion. The Lorentz group has three rotation generators J1 , J2 , J3 and three
boost generators K1 , K2 , K3 . Classical linear and angular momentum are conserved
because of the latter symmetry while angular momentum is conserved because of the
former. The Lorentz group generators satisfy the following algebra [5],
[Ji , Jj ] = iijk Jk , (A.1)
[Ji , Kj ] = iijk Kk , (A.2)
[Ki , Kj ] = − iijk Jk . (A.3)
(For completeness, also note that [Ki , Jj ] = −iijk Kk .) There is some ambiguity in texts
about which generators are imaginary and which are real. In this algebra, the rotation
matrices are imaginary while the boosts are real. (Thus, in converting to imaginary
time all become imaginary rotations in the 4-D Euclidean space.) The boost generator
matrices are
(
1 j=k
(K j )0k = (K j )k0 = (A.4)
0 otherwise
Vortex particle model in a renormalizable quantum gravity 29

and the rotation generators are


(
−ijkl
(J j )kl = (J j )kl = (A.5)
0 otherwise
with other elements zero. A general Lorentz transform has the form,
Λab = exp[(ζ i (K i )ab + θi (J i )ab )], (A.6)
where ζ i is called the rapidity in each of the spatial directions and θi is the degree of
rotation in each of the spatial planes, i = 1, 2, 3. For   1,
Λab = δ ab + ζ i (K i )ab + θi (J i )ab + O(2 ), (A.7)
with δ ab the identity. The small parameter  depends on the size of a discrete element
of spacetime. For continuous spacetime,  → 0 [5] and −1 Λab → ζ i (K i )ab + θi (J i )ab
becomes an infinitesimal Lorentz transform.
Another, more useful way to represent the SO(3,1) group is as an SU(2)⊗SU(2)
group which, as shown in the next section, allows the field equations to decouple into
two sets. Let
J+i = Ki + Ji , (A.8)
and
J−i = Ki − Ji . (A.9)
One can then show that
[J+i , J+j ] = iijk J+k , (A.10)
and likewise for J−i but
[J+i , J−j ] = 0, (A.11)
meaning that in this form the two sets of three generators decouple [5].

Appendix B. Derivation of the 1PN potential

The potential at a spacetime point (t, ~x) in the given inertial frame can be solved by
the integral equations [17],
Ωµν (t − |~x − ~x0 |, ~x0 )
Z
Gµν (t, ~x) = d3 x0 , (B.1)
C |~x − ~x0 |
where C is the past-light cone,
Ωµν = Jµν − (4π)−1 Λµν , (B.2)
and
Λµν = ∂ λ (µ βρσ uβ Gλρ Gνσ )+µ βρσ uβ (∂λ Gνρ −∂ν Gλρ +ρ βκτ uβ Gλκ Gντ )Gλ σ .(B.3)
0
We assume that Hµν = 0 because Jµν = 0 by the vortex model discussed in Sec. 2.
The integral equation may be solved by iteration in which a trial potential Gµν = 0
is inserted into the integrand and the integral solved to arrive at a new potential G0µν
Vortex particle model in a renormalizable quantum gravity 30

which is then reinserted to achieve a potential G00µν and so on until the required accuracy
is achieved. In order to derive the 1PN equations of motion general relativity requires
two iterations but here we only require one, meaning that the 1PN equations for this
theory are linear.
Let the baryon density be ρ0 , and the Newtonian potential is given by,
ρ0 (t, ~x0 )
Z
U = d3 x0 . (B.4)
|~x − ~x0 |
Let U ∼ 2 and v/c ∼  where v is the average velocity and c is the speed of light in
vacuum. The first post-Newtonian corrections require that G00 be given up through
order 4 , G0j through order 3 and Gij through 2 , correcting the Newtonian order,
G00 = U, G0j = 0, Gij = 0. (B.5)
The G-field derives from the equations of Sec. 3 above and the stress-energy-
momentum tensor for baryon dust. Each baryon particle or compact spherical mass
(such as a neutron star or black hole) p has a stress-energy-momentum tensor in its rest
frame in a spherical coordinate system with the particle at the origin,
J00 = mp δ 3 (~x), Jrr = mp δ 3 (~x). (B.6)
The latter is related to the potential Grr which has a transformation,
xi xj
Gij = Grr 2 , (B.7)
|~x|
which is
(xi − x0i )(xj − x0j )
Gij = mp (B.8)
|~x − ~x0 |3
for a particle at ~x0 . To obtain the potential for all the dust, sum all the potentials for
all the baryons by superposition of the potentials,
ρ0 (t, ~x0 )(xi − x0i )(xj − x0j )
Z
G̃ij = d3 x0 + O(4 ). (B.9)
|~x − ~x0 |3
By a non-Abelian gauge transformation [7],
Gij = G̃ij + ∂i θj , (B.10)
where θµ = ∂µ χ and
Z
χ = d3 x0 ρ0 (t, ~x0 )|~x − ~x0 |, (B.11)

we have
Gij = δij U + O(4 ). (B.12)
The next two parts of the G-field, G0j and G00 , require particle velocities and time
dilation to be included. The current tensors, Jµν (p) of each baryon p, are in the rest
frame, but, because the bodies are moving with respect to the observer at infinity and
subject to the gravitational fields of one another, the tensors must be (1) boosted from
comoving frame x̄µ to the observer’s frame, xµ , and (2) subject to a non-Abelian gauge
Vortex particle model in a renormalizable quantum gravity 31

transformation from the zero gravitational field (free fall) in the rest frame of the body
to the gravitational field in the rest frame of the observer,
dx̄µ dx̄ν
Jµν (p) = α β Jαβ (p) + Gµν mp δ 3 (~x − ~xp ), (B.13)
dx dx
where Λ is a Lorentz transformation matrix for a boost density ~v (t, ~x) (c = 1)[7] and
Gµν is the G-field not including the contribution of the mass at that point. These two
transformations are equivalent to the two step, boost-and-coordinate-transform method
of general relativity [7] which reflects the effects of both velocity and gravitational fields
on measurements of stress-energy-momentum tensors.
For a baryon p with mass mp at ~x0 with a rest frame potential Ḡµν and velocity ~v
(c = 1),
∂ x̄0 ∂ x̄0 ∂ x̄i ∂ x̄j
G0j = Ḡ00 + Ḡij , (B.14)
∂x0 ∂xj ∂x0 ∂xj
where
mp (xi − x0i )(xj − x0j )
Ḡij = . (B.15)
|~x − ~x0 |3
To order this becomes,
[(~x − ~x0 ) · ~v ](xj − x0j )
 
mp
G̃0j = − vj + + O(5 ). (B.16)
|~x − ~x0 | |~x − ~x0 |2
Under the same gauge transformation as above,
G0j = G̃0j − ∂0 θj , (B.17)
and summing over all baryons, the potential is,
G0j = −2Vj + O(5 ), (B.18)
where
ρ0 (t, ~x0 )vj (t, ~x0 )
Z
Vj = d3 x0 . (B.19)
|~x − ~x0 |
The time-time field is,
G̃00 = U + 2Ψ + Φ + O(6 ), (B.20)
where
ρ0 (t, ~x0 )(v 2 + U )
Z
Ψ(t, ~x) = d3 x0 , (B.21)
|~x − ~x0 |
and
ρ0 (t, ~x0 ) [(~x − ~x0 ) · ~v ]2
Z  
3 0 2
Φ(t, ~x) = dx −v . (B.22)
|~x − ~x0 | |~x − ~x0 |2
The gauge transformation, G00 = G̃00 + ∂02 χ (which we have already applied to the other
potentials and hence must carry over to this one), eliminates the last term, and gives
G00 = U + 2Ψ + O(6 ). (B.23)
Vortex particle model in a renormalizable quantum gravity 32

Missing is a term of order U 2 that appears in metric PPN formalism which indicates
the nonlinearity in the superposition of the potential. In GR, however, this term is an
artifact of the choice of isotropic coordinates and not a true nonlinearity such as would
be associated with massive gravitons.
In general relativity, the change of coordinates, r = r̄(1 + M/2r̄)2 , switches the
Schwarzschild solution to isotropic coordinates [7],
 2
2 1 − M/2r̄
ds = − dt2 + (1 + M/2r̄)4 [dr̄2 + r̄2 (dθ2 + sin2 θdφ2 )].(B.24)
1 + M/2r̄
Because the change in coordinates, however, depends on the potential itself, this
introduces a nonlinearity that previously did not exist in the solution. Like the
Schwarzschild solution in Schwarzschild coordinates, the Yang-Mills theory 1PN solution
does not contain any nonlinearity because it is not in this nonlinear coordinate system.
The exact, Yang-Mills spherically symmetric solution given in spherical coordinates,
G00 = Grr = M/r, (B.25)
under the same change of variables, becomes to 1PN order,
G00 = M/r̄ − 2(M/r̄)2 + O(6 ), (B.26)

G0i = Gi0 = O(5 ) (B.27)

and
Gij = δij M/r̄ + O(4 ). (B.28)
While one could make a change of the flat spacetime coordinates to bring the YM
potential in line with the GR metric, there is no purpose in doing so. Since both GR
and the Yang-Mills theory are diffeomorphism covariant (generally covariant), changing
coordinates does not change predictions. It does, however, require that observations be
input in the appropriate coordinate system since the Schwarzschild r is not the same as
the 1PN r̄. Indeed, because distances are defined by the metric, gµν , for the YM theory
r is the true distance between a point and the origin while, in General Relativity, the
Schwarzschild r is simply a coordinate. Meanwhile, because of the potential dependent
metric in “isotropic” coordinates, r̄ is just a coordinate in both theories and does not
refer to distance directly.
The remaining equations (which can be derived from the field equations) are the
equation of motion. The continuity equations (39) to 1PN order are the four equations,
∂ν J να = 0, (B.29)
each of which is the same as the electromagnetic continuity equation. The current is
symmetric and hence contains ten unknowns. In our simplified view, however, where
the polarization and spin are abstracted into a single, constant scalar mass, m, these
equations completely specify the four velocity as a straight line. Linearized general
relativity has the same problem [6] and care needs to be taken that the nonlinear,
gauge covariant equations of motion (39) be used instead. For N -body problems, which
Vortex particle model in a renormalizable quantum gravity 33

are generally the ones that gravitation deals with, however, the geodesic equation is
sufficient (provided that potential is in a gauge such that it is traceless to 1PN order).
This is because the Yang-Mills theory satisfies the Strong Equivalence Principle (SEP)
and internal structure does not affect the motion of bodies. Therefore, any spherical,
compact body moves as a test particle.

Appendix C. Proof of gauge transformation for index potentials

For a generic SU(N) Yang-Mills theory, for a non-Abelian gauge transformation U (x),
a matrix potential Aµ (x) and generators T a we have,
A0µ = U AU † − iU ∂µ U † , (C.1)
or, more simply,
A0µ = U Dµ U † , (C.2)
where Dµ = ∂µ − iAµ is the covariant derivative. This shows how the matrix potential
changes but not how the components change. Let Aaµ , where Aµ = Aaµ T a and summation
is implied. These change like electromagnetic fields,
A0a a a
µ = Aµ − i∂µ θ , (C.3)
a a
for an arbitrary set of functions θa (x) such that U = eiθ T . This can be built from
infinitesimal changes of gauge as follows: Let U = I + iχa T a be an infinitesimal gauge
transformation, then,
A0a a b b c c d d
µ T = (I + iχ T )(∂µ + Aµ T )(I − iχ T ). (C.4)
which is,
A0a a c c d d c c d d b b c c 2
µ T = Aµ T +∂µ (−iχ T )+Aµ T (−iχ T )+(iχ T )(Aµ T )+O( ).(C.5)

All but two terms cancel and we have,


A0a a c c d d 2
µ T = Aµ T − i(∂µ χ )T + O( ). (C.6)
Now, if we multiply by any generator T b and take the trace we have tr(T a T b ) = δ ab [5];
therefore,
A0a a a 2
µ = Aµ − i∂µ χ + O( ). (C.7)
Because the non-Abelian gauge symmetry is continuous, any gauge transformation can
be built from infinitesimal gauge transformations such as this. Let θa = χa / and let
 → 0, then,
A0a a a
µ = Aµ − i∂µ θ , (C.8)
is proved.

You might also like