Finsler geometry and
quantum field theory
Howard
E.Brandt
U.S. Army
Research Laboratory, Adelphi, Maryland, U.S.
Abstract. Finsler geometry motivates a generalization of
the Riemannian structure of space-time to include dependence of the space-time
metric and associated invariant tensor fields on the four-velocity coordinates
as well as the space-time coordinates of the observer. It is then useful to
consider the tangent bundle of space-time with space-time in the base manifold
and four-velocity space in the fiber. A physical basis for the differential
geometric structure of the space-time tangent bundle is provided by the
universal upper limit on proper acceleration relative to the vacuum. It is then
natural to consider a quantum field having a vanishing eigenvalue when acted on
by the Laplace-Beltrami operator of the space-time tangent bundle. On this
basis a quantum field theory can be constructed having a built-in intrinsic
regularization at the Planck scale, and finite vacuum energy density.
Finslerian fields
A physical Finslerian field F(x.v) is one that depends not only on the observer's space-time coordinates,
x ≡ {xm}= {x0, x1, x2, x3}, (1)
but also on the observer's four-velocity coordinates,
v ≡ {vm}= {dxm/ds} = {v0,
v1,
v2,
v3}, (2)
where ds is the infinitesimal interval along the world line of the observer [1]-[3]. The four-velocity coordinates play here the role of the tangent space coordinates of Finsler geometry. It can be argued that the space-time-metric field gmn itself must in general depend not only on where it is observed in space-time, but also on the four-velocity of the 'observer', namely it is a Finslerian field [1]-[3]:
gmn = gmn (x,v). (3)
(The reader may prefer to replace the word 'observer' by 'measuring device', 'object acted upon by the field', or 'some other field interacting locally with the field'.) The space-time metric in a canonical Finsler space-time is not only Finslerian, but also satisfies special homogeneity conditions involving the dependence of the metric on the tangent space coordinates, v. In particular, one has [4], [5]
ds = L(x, dx), (4)
where L is the fundamental Finsler function, and
L(x, adx) = aL(x, dx), (5)
from which it follows that
(6)
L2(x,v) = gmnvmvn, (7)
(8)
and
(9)
In considering Finslerian space-time and associated embedded Finslerian fields, it is useful to consider the tangent bundle of space-time with space-time in the base manifold and four-velocity space in the fiber. Using the homogeneity relations, Eqs. (4)-(9), then the connection and Riemann curvature scalar of the space-time tangent bundle [6] can be significantly reduced for the case of a Finsler-space-time base manifold [4], [7]. However, the special homogeneity requirements may not hold physically in general, but may only hold in certain special space-time models.
A physical basis for the differential geometric structure of the space-time tangent bundle is provided by the universal upper limit a0 on proper acceleration a relative to the vacuum [8]-[11]. If the proper acceleration were sufficiently large, then, because of vacuum radiation in an accelerated frame (in which particles are produced with average energy proportional to the proper acceleration), particles would be produced with mass such that their Schwarzschild radius exceeds their extent (Compton wave-length). Copious production of black-hole anti-black-hole pairs would ensue, accompanied by breakdown of the classical space-time structure, and the very concept of acceleration would loose any meaning because of the resulting complex topology. Explicitly, the maximal proper acceleration, a0, is given by [8]
(10)
where a is a
number of order unity, c is the speed
of light,
is Planck's constant
divided by 2p, and G is the universal
gravitational constant. This is the maximum possible proper acceleration
relative to the vacuum and is taken to be universal. Hence for any proper
acceleration a, one requires
(11)
But, according to the differential geometry of space-time, the proper acceleration, a, along a world line in curved space-time is given by
(12)
where the four-velocity vm is given by
(13)
and
denotes the covariant
derivative of the four-velocity with respect to the interval along the world
line, namely,
(14)
in which
is the space-time
affine connection, and
ds2 ≡ gmn dxm dxn (15)
is the line element of space-time. Then substituting Eqs.(12) and (14) in
Eq.(11), one obtains
(16)
Next substituting Eqs. (13) and (15) in Eq.(16), one obtains [10]
(17)
where
(18)
is the minimum radius of curvature of world lines. Equation (17) defines the eight-dimensional quadratic form ds2, which is non-negative along the world line. The inequality, Eq. (17), simply expresses the fact that the proper acceleration can never exceed the maximal proper acceleration. By analogy with the construction of the space-time line element of general relativity from the limiting speed of light, it is natural to take ds2 to be the line element in the tangent bundle of space-time, in which the space-time coordinates xm are the coordinates in the space-time base manifold, and the four-velocity coordinates r0vm (modulo a factor of r0) are the tangent space coordinates.
The bundle line element, Eq.(17), can be rewritten as follows [10], [1]:
ds2 ≡ GMNdxMdxN, {M,N =
0,2,...,7}, (19)
where the bundle coordinates are
{xM} ≡
{xm, r0vm} , {M = 0,2,...,7; m = 0, 1, 2, 3}, (20)
and the
metric of the tangent bundle of space-time is
(21)
in which
(22)
The bundle
metric GMN, given by
Eq.(21), has a structure similar to that of an eight-dimensional Kaluza-Klein
gauge theory in which the higher dimensions are in four-velocity space, and
is the gauge potential.
Eqs.( 19)-(22) served as the starting point for investigating possible
implications of a limiting proper acceleration for the differential geometric
structure of the tangent bundle of space-time [1]-[4], [6], [8]-[24]. Possible
forms for the bundle connection, curvature, and action were explored, including
those based on Riemann and Finsler space-times, and also Kähler and
complex space-time tangent bundles. Among the many differential geometric
invariants of the space-time tangent bundle, important for the present
discussions is the Laplace-Beltrami operator:
(23)
This is the invariant generalization of the wave operator, or d'Alembertian, of field theory. A simple invariant field equation for a Finslerian scalar field f(x,v) is then given by [20]
Lf (x,v) = 0. (24)
Finslerian scalar quantum fields
When the space-time is Minkowskian, the ordinary inhomogeneous Lorentz group (or Poincaré group) is a subgroup of the invariance group of the space-time tangent bundle [12], [20]. It is of interest to examine quantum field solutions to Eq.(24) for this simple case, in order to establish connections between the Finslerian framework and canonical relativistic quantum field theory. To this end, consider a simple case in which the space-time is completely flat. In particular, take the space-time metric to be Minkowskian. For this case, it was argued in earlier work that the scalar field satisfying Eq.(24) is given by [20], [25], [26]:
(25)
where p denotes the four-momentum pm = {p0, p1, p2, p3} of a particle excitation of the scalar field, a(p) and a(p) are the particle creation and annihilation operators satisfying the commutation relations
[a(p), a(p′)] = d 3(p - p′), [a(p), a(p′)] = 0, [a(p), a(p′)] = 0, (26)
d 3(z) is the three-dimensional Dirac delta function, and q1(z) is the Heaviside function [28],
(27)
Also in
Eq.(25), N is a normalization factor
such that the field operator acting on the vacuum state, namely
, is the state of the field corresponding to a
single-particle excitation appearing at some point (x,v) in the space-time
tangent bundle, namely [25], [26],
(28)
where V denotes the volume of space, m is the mass of the scalar particle excitation of the quantum field, and K1(z) is the modified Bessel function of the third kind of order one.
It can be shown that both the positive and negative frequency terms in Eq.(25) are proportional to [20], [9]:
(29)
where dx/dt is the ordinary velocity of the observer, and
(30)
As can be seen from Eq.(29), there occurs an intrinsic Planck-scale regularization of the quantum field, with an exponential energy cutoff beyond the Planck energy.
It is important to note that for an observer with ordinary velocity
(31)
in Minkowski space-time, one has
(32)
and therefore
(33)
Thus the
ordinary velocity
corresponds to both
future and past directed four velocity. This is essential in order that both
the positive-frequency particle modes and the negative-frequency antiparticle
modes in Eq.(25) and in the following have non-vanishing support. One is
reminded of the Stückelberg-Feynman idea that antiparticles can be
represented as particles with proper time reversed relative to true time [21],
[27].
Hamiltonian of scalar quantum field
On the basis of microcausality, it was argued in earlier work [21] that it is logical to define the generalized adjoint (f (x,v)) of the quantum field f (x,v) by
(34)
in which the ordinary adjoint is taken, but also the sign of the four-velocity v is changed. This is connected, through charge-conjugation, with the fact that the positive-frequency particle component of the field in Eq.(25) has non-vanishing support for positive v0 (or equivalently, positive pv), and the negative-frequency anti-particle component has non-vanishing support for negative v0 (equivalently, negative pv) [21]. In order that the Hamiltonian be Hermitian (in terms of the generalized adjoint), that it reduce to the canonical form in Fock space, and that the observer's four-velocity lie on the four-velocity shell (v2 = 1), it is natural to define the Hamiltonian for the scalar quantum field as follows [25], [26]:
(35)
in which the bundle energy density T00(x,v) at point (x,v) in the bundle is integrated over the entire space-time tangent bundle with large spatial volume V, d (z) is the one-dimensional Dirac delta function, and the bundle energy density is given by
(36)
To see that the Hamiltonian is Hermitian, one first notes that
(37)
Next, taking the generalized adjoint of T00(x,v) and using Eq.(34), one obtains
(38)
Equivalently, using Eq.(25), Eq.(38) becomes
(39)
or comparing Eqs.(39) and (36), one obtains
T00(x,v) = T00(x, -v). (40)
Next substituting Eq.(40) in Eq.(37), one has
(41)
and replacing the dummy variable of integration v by -v, Eq.(41) becomes
(42)
Finally, substituting Eq.(35) in Eq.(42), one concludes that
H = H. (43)
Thus H is in fact Hermitian.
Next, using (25), (27) and (36) in (35), integrating over space, and substituting the integral [25], [29]
(44)
one obtains the canonical expression for the Hamiltonian operator for a scalar quantum field in Fock space [25]:
(45)
in
which
(46)
For consistency, it is also well to verify that the Heisenberg equation of motion for the field f is satisfied, namely [30],
(47)
To see that (47) holds, one uses (45) and (25) to evaluate the commutator [H,f]. Thus one has
(48)
or equivalently,
(49)
Then substituting Eqs.(26) in Eq.(49), one obtains
(50)
Finally, substituting Eq.(25) in Eq.(50), one obtains the Heisenberg equation of motion for the field, Eq.(47).
Vacuum energy density
Using
Eqs.(35) and (36), and denoting the vacuum state by
, it follows that the vacuum energy in the bundle, for
spatial volume V, is given by [25]:
(51)
which is the
canonical divergent result [30]. However, a particular observer has
four-velocity v, and his world line
at any time is confined to the neighborhood of v in the fiber (See Eq.(33)). The vacuum energy which he observes
is therefore given by [25]
(52)
in which D3v is defined by
(53)
where d v is the spread in the spatial components
of the four-velocity of the observer, and because of the Dirac delta function
in Eq.(52),
(54)
One notes that in Eq.(52), one can use the well known identity:
(55)
Equation (53), defining the spread, is true because the expression following D3v in Eq.(52) turns out to be proportional to (1/v0) = (1+|v|2)-1/2, with no other dependence on the spatial component v of four-velocity (See Eqs.(54) and (62)-(67)).
Next,
denoting the three respective contributions to Eq.(52) of the three terms of
Eq.(36) by
, and
, respectively, then
(56)
Then using Eqs.(36), (52), and (56), one has
(57)
![]()
.
and substituting Eqs.(25) and (55) in Eq.(57), one obtains
![]()
(58)
Here q1(+r0pv/ÿ) has been replaced by q1(+v0), since it can be shown that pv > mc when v0 = g > 0, and pv < mc when v0 = -g < 0 [20], [21]. Then, according to Eqs.(27), (30) and (54), one has
q1(v0)q1(-v0) = 0, [q1(v0)]2 = q1(v0). (59)
Next using Eq.(59) in Eq.(58), one obtains
![]()
(60)
One also has
(61)
Therefore using Eqs.(61) in Eq.(60), one obtains
![]()
(62)
or
equivalently,
(63)
Analogously, corresponding to the contributions to Eq.(56) of the two remaining terms of Eq.(38), containing derivatives with respect to space and four-velocity, respectively, one obtains
(64)
and
(65)
Then substituting Eqs.(63)-(65) in Eq.(56), noting that
p02 = |p|2 + m2c2, (66)
evaluating the following integral [26], [29],
(67)
where K2(z) is the modified Bessel function of the third kind of order 2, substituting Eq.(28), and dividing by the spatial volume V, one finally obtains the vacuum energy density seen by an observer at (x,v) in the space-time tangent bundle:
(68)
Thus any one observer sees a finite vacuum energy density given by Eq.(68). One notes that for sufficiently small spread in the spatial part of the observer's four-velocity (Eq.(53)), the observed vacuum energy density may be near vanishing. This may be consistent with a near vanishing cosmological constant.
References
1. H.E.Brandt. Finslerian Spacetime. Contemporary Math., 196, 273-287, 1996.
2. H.E.Brandt. Finslerian Fields in the Spacetime Tangent Bundle. Chaos, Solitons Fractals, 10, 267-282, 1999.
3. H.E.Brandt. Finslerian Fields, in Finslerian Geometries: A Meeting of Minds. P.L.Antonelli, ed., Kluwer, Dordrecht, pp. 131-138, 2000.
4. H.E.Brandt. Finsler-Spacetime Tangent Bundle. Found. Phys. Lett., 5, 221-248, 1992.
5. H.Rund. The Differential Geometry of Finsler Spaces. Berlin, Springer, 1959.
6. H.E.Brandt. Riemann Curvature Scalar of Spacetime Tangent Bundle. Found. Phys. Lett., 5, 43-55, 1992.
7. K.Yano, E.T.Davies. On the Tangent Bundles of Finsler and Riemannian Manifolds. Rend. Circ. Mat., Palermo, 12, 211-228, 1963.
8. H.E.Brandt. Maximal Proper Acceleration Relative to the Vacuum. Lett. Nuovo Cim., 38 (1983), 522-524; 39 (1984), 192.
9. H.E.Brandt. Quantum Vacuum Heuristics. J. Mod. Optics, 50, 2455-2463, 2003.
10. H.E.Brandt. Maximal Proper Acceleration and the Structure of Spacetime. Found. Phys. Lett., 2, 39-58, 405, 1989.
11. H.E.Brandt. Structure of Spacetime Tangent Bundle. Found. Phys. Lett., 4, 523-536, 1991.
12. H.E.Brandt. The Maximal Acceleration Group. In XIIIth International Colloquium on Group Theoretical Methods in Physics, W.W.Zachary, ed., World Scientific, Singapore, pp. 519-522, 1984.
13. H.E.Brandt. (1991) Connections and Geodesics in the Spacetime Tangent Bundle, Found. Phys., 21, 1285-1295.
14. H.E.Brandt. Differential Geometry of Spacetime Tangent Bundle. Int. J. Theor. Phys. 31, 575-580, 1992.
15. H.E.Brandt. Kinetic Theory in Maximal-Acceleration Invariant Phase Space. Nucl. Phys. B (Proc. Suppl.), 6, 367-369, 1989.
16. H.E.Brandt. Kähler Spacetime Tangent Bundle. Found. Phys. Lett., 5, 315-336, 1992.
17. H.E.Brandt. Complex Spacetime Tangent Bundle. Found. Phys. Lett., 6, 245-255, 1993.
18. H.E.Brandt. Spacetime Tangent Bundle with Torsion. Found. Phys. Lett., 6, 339-369, 1993.
19. H.E.Brandt. Solution to Torsion Relations in Finsler-Spacetime Tangent Bundle. Found. Phys. Lett., 7, 297-301, 1994.
20. H.E.Brandt. Quantum Fields in the Spacetime Tangent Bundle. Found. Phys. Lett., 11, 265-275, 1998.
21. H.E.Brandt. Finslerian Quantum Fields and Microcausality. Found. Phys. Lett., 13, 307-328, 2000.
22. H.E.Brandt. Causal Domain of Minkowski-Spacetime Tangent Bundle. Found. Phys. Lett., 13, 581-588, 2000.
23. H.E.Brandt. Particle Geodesics and Spectra in the Spacetime Tangent Bundle. Rep. Math. Phys., 45, 389-405, 2000.
24. H.E.Brandt. Lorentz-Invariant Quantum Fields in the Spacetime Tangent Bundle. Internat. J. Math. and Math. Sci., No. 24, 1529-1546, 2003.
25. H.E.Brandt. Intrinsic Gravitational Regularization of Quantum Fields. Found. Phys. Lett., 17, 295-300, 2004.
26. H.E.Brandt. Finite Vacuum Energy Density in Quantum Field Theory. To appear in J. Mod. Optics, 2004.
27. R.P.Feynman. The Theory of Positrons. Phys. Rev., 76, 749-759, 1949.
28. R.P.Kanwal. Generalized Functions: Theory and Technique. Academic Press, New York, p.1, 1983.
29. F.Oberhettinger, L.Badii. Tables of Laplace Transforms. Springer, New York, p.94, 1973.
30. A.Zee. Quantum Field Theory in a Nutshell. Princeton University Press, Princeton, 2003.
Howard
E.Brandt,
Researcher-scientist (Army Research
Laboratory, Adelphi, USA); he is a theoretical
physicist with primary scientific interests in quantum field theory, general
relativity, differential geometry, and quantum information processing. The
present paper is an expanded version of an invited paper, "Finsler Geometry and
Quantum Field theory", presented at the Fourth World Congress of Nonlinear
Analysts in the session: Applications of Finsler Differential Geometry (in
Engineering, Physics, and Biology) held July 5, 2004, in Orlando, Florida, U.S.
© 1995-2008 Kazan State University