Gravitomagnetic Accelerators
Abstract
We study a simple class of time-dependent rotating Ricci-flat cylindrically symmetric spacetime manifolds whose geodesics admit gravitomagnetic jets. The helical paths of free test particles in these jets up and down parallel to the rotation axis are analogous to those of charged particles in a magnetic field. The jets are attractors. The jet speed asymptotically approaches the speed of light. In effect, such source-free spacetime regions act as “gravitomagnetic accelerators.”
pacs:
04.20.CvThe purpose of this Letter is to point out that general relativity in principle permits the existence of dynamic source-free spacetime regions in which the speed of free test particles can rapidly increase so as to approach the speed of light.
Mechanisms for the acceleration of particles to ultrarelativistic speeds are important for the explanation of high-energy astrophysical phenomena 1 ; 2 . Most of the known acceleration processes involve charged particles and are electromagnetic in origin 3 . Gravitational mechanisms involving arbitrary test particles have been the subject of recent investigations 4 ; 5 ; 6 ; 7 ; 8 . We adopt a different approach here and present a simple explicit solution of Einstein’s vacuum field equations for which the geodesic equation has exact analytic solutions that represent gravitomagnetic jets. The jets are attractors and the speed of a free test particle in such a jet asymptotically approaches the speed of light.
Consider a warped product spacetime with a metric of the form
(1) |
where is a 3D stationary metric given by
(2) |
and we use units such that throughout. Here and is the solution of the differential equation
(3) |
The spacetime metric (1) has signature and is Ricci-flat, namely, 9 . The cylindrical coordinates are dimensionless. To transform them to physical coordinates , we need arbitrary lengthscales and ; then, and . The conformal factor , the warping function and the gravitomagnetic potential lead to interesting aspects of the gravitational field under consideration here.
The cylindrical coordinates in the spacetime domain of interest must be admissible. It follows from the Lichnerowicz conditions 10 that the principal minors of the metric tensor and its inverse must be negative. Hence, and , where . The first of these conditions is satisfied if is a monotonically increasing function of . It turns out that Eq. (3) is a special case of the generalized Emden-Fowler equation. Of the two known explicit solutions of this equation, is not acceptable as metric (1) would then degenerate into a 3D metric and is such that is monotonically decreasing with . In fact, with taken as radial coordinate, the solutions of Eq. (3) other than fall into two classes: either is monotonically increasing or decreasing. In this work we focus on the former class, since the latter class involves rotating gravitational waves and was investigated in detail in Ref. 9 .
The solutions of Eq. (3) with monotonically increasing all have the general form depicted in Figure 1. If is a solution of Eq. (3), then so is ; henceforth, we consider only the branch. It is possible to show that there is an open set of initial conditions corresponding to admissible solutions ; each exists on a radial interval , where with limiting values and such that at . Furthermore, increases, decreases and approaches infinity as . Indeed, at the axis of cylindrical symmetry and at the exterior boundary cylinder. The physical region of interest is the open hollow expanding cylindrical domain with inner boundary around the axis and outer boundary at the null hypersurface . The cylindrical coordinates are not admissible at these boundaries, since vanishes at and vanishes at .
With admissible cylindrical coordinates in , we have an algebraically general Petrov type I solution of Einstein’s source-free equations with two commuting spacelike Killing vector fields and associated with cylindrical symmetry. The two-parameter isometry group is not orthogonally transitive. While is hypersurface orthogonal, is not; moreover, the other two coordinates can be invariantly defined via the magnitudes of these Killing vector fields. The hypersurfaces are spacelike, while hypersurfaces are timelike in the physical domain . Near the axis (),
(4) |
where and are constants. Near the boundary (),
(5) |
where is a constant. We note that solutions of Einstein’s gravitational field equations with cylindrical symmetry have been the subject of numerous investigations (see 11 ; 12 ; 13 ; 14 ; 15 and the references cited therein).
The proper radial distance from the axis to an event with is given by
(6) |
At a finite instant of time , the proper radial distance from the axis to the boundary is finite. It is possible to show that the condition of elementary flatness is not satisfied near the axis 16 . Within the spacetime region of interest , however, there are no curvature singularities. To illustrate this fact, we note that there are four algebraically independent scalar polynomial curvature invariants in a Ricci-flat spacetime that can be represented as 12
(7) | ||||
(8) |
These turn out to be real in this case and can be expressed as
(9) | ||||
(10) |
where . These invariants are well behaved in the interior of the physical region and have proper limits at the boundaries. Let us note that and both diverge in the infinite past (), a situation that is consistent with the emergence of the universe from a singular state as in the standard cosmological models.
The spacetime domain rotates about the axis of cylindrical symmetry. To elucidate the gravitomagnetic aspects of , imagine the class of fundamental observers in . They are spatially at rest by definition, with spatial frames that we can choose to be along the natural directions of the cylindrical coordinate system. A unit gyro carried by these observers precesses about the axis with frequency , indicating the presence of a gravitomagnetic field parallel to the axis 17 . There is experimental evidence for gravitomagnetism; indeed, GP-B has recently measured the exterior gravitomagnetic field of the Earth 18 .
We now turn to the motion of free test particles in this gravitational field. Let be the four-velocity vector of the test particle, where is the proper time along its world line. The components of along the Killing vector fields will be constants of geodesic motion. Thus is the constant linear momentum (per unit mass) of the test particle parallel to the axis, while is the constant angular momentum (per unit mass) of the test particle about the axis. That is,
(11) |
It then follows from that , where
(12) |
Here and we have assumed that monotonically increases with along the geodesic world line. Thus the geodesic equation reduces in this way to the radial equation given by
(13) |
In practice, Eq. (3) must also be solved simultaneously with respect to the particle’s proper time.
In analogy with the motion of a charged particle in a magnetic field, we look for geodesics that are confined to a cylinder of fixed radius . Thus we set and . Equations (11) and (12) yield
(14) |
while it follows from Eq. (13) that
(15) |
These are compatible, the latter with the lower sign only, if and only if and there is a radial coordinate such that for . In this case, the geodesic equation can be solved explicitly for a special three-parameter set of geodesics parameterized by the initial values of such that remains constant at . The geodesics in this class exhibit helical motions up () and down () on the cylinder parallel to the axis except for the measure zero subset () whose motions remain bounded on circular orbits about the axis. It can be shown that for a given , either there is a unique , , or there is none; moreover, there is a non-empty open subset of admissible solutions of Eq. (3), corresponding to in Eq. (5), that allow these special geodesics. For a solution of Eq. (3) in this class, we call the set of special geodesics a gravitomagnetic jet.
Numerical experiments reveal that gravitomagnetic jets are attractors, see Figs. 1 and 2. That is, the union of special geodesics with a given and is a non-compact connected invariant manifold that attracts all nearby geodesics. Figure 2 highlights the helical motion of a gravitomagnetic jet. We note that these helical motions up and down within a double-jet configuration have the same orientation; the helical sense is positive in our case due to our choice of the branch. For recent studies of helical motions in astrophysical jets, see 19 .
Null geodesics can be treated similarly in and it turns out that they have special helical solutions confined to just as in the case of timelike geodesics. Moreover, for , special timelike geodesics go over to special null geodesics as a simple consequence of the dynamical equations of motion.
To explore this important aspect of gravitomagnetic jets further, we consider the speed of jets with respect to the fundamental observers. These are endowed with an orthonormal tetrad such that in coordinates
(16) |
and we can choose the spatial frame to be along the standard cylindrical coordinate axes. Then , where is the jet four-velocity measured by the fundamental observers. Hence, and a straightforward calculation reveals that
(17) |
Here is the Lorentz factor corresponding to circular motion with ; that is, , where . For and , diverges exponentially with time; that is, the jet speed rapidly approaches the speed of light. The gravitational influence of the test particle on the spacetime geometry has been neglected in our work; clearly, this approximation eventually breaks down for the gravitomagnetic jets.
Equation (3) is invariant under the scale transformation , where and for . This scale invariance can be used to reduce Eq. (3) to a first order system of autonomous equations. For instance, let and consider scale-invariant variables and , where and ; then, Eq. (3) reduces to the Lotka-Volterra system
(18) |
A solution is admissible when for all ; similarly, an admissible solution allows jets when for some , where at the symmetry axis and at the outer boundary. A detailed investigation of system (18)—for admissible solutions that allow jets—reveals that for lies in the interval , where . This means that , the speed of the free test particle on a circle of radius , is such that , where .
It follows from these results that free test particles in a gravitomagnetic jet can in principle start out with speeds near zero, but they then inevitably undergo rapid “acceleration” to almost the speed of light. Similarly, one can study the motion of other geodesics in with respect to the fundamental observers. In the simple numerical experiments whose results are presented in Figs. 1 and 2, we followed the variation of the Lorentz factor from the initial point to late times. There is indeed a vast difference between the geodesics that leave via and those that are attracted to the jet—see the right panel of Fig. 1. In the former case, the Lorentz factor initially decreases but then increases as the geodesic exists , remaining within about an order of magnitude of unity. In the latter case, the Lorentz factor quickly diverges to infinity—starting from (for the left-hand plot) and (for the right-hand plot) in Fig. 2.
The existence of a jet is a scale-invariant property; that is, for every admissible solution with a jet, scaling leads to a one-parameter family of solutions of the same kind, where the jet now occurs at . One can use this property to set for every jet solution. Thus Eq. (3) can be integrated with initial conditions that at , and ; in this case, for . The solution is admissible once ; that is, . For example, in Fig. 1 we have and hence for the corresponding jet depicted in Fig. 2. A detailed treatment of gravitomagnetic jets is contained in Ref. 20 .
The empty annular region of physical interest that allows gravitomagnetic jets could in principle be joined to two concentric circular cylindrical regions with matter such that the inner one surrounds the axis () and the outer one has its inner boundary at . Our treatment has been greatly simplified by the assumption of cylindrical symmetry; nevertheless, we hope that in the context of general relativity similar gravitomagnetic accelerators may emerge under physically more realistic circumstances. In any case, this appears to be a promising approach in the search for the origin of high-energy astrophysical jets. Further recent results in this direction are contained in Ref. 21 .
The work of CC was supported in part by the NSF grant DMS 0604331.
References
- (1) A.M. Hillas, Ann. Rev. Astron. Astrophys. 22, 425 (1984); B. Mashhoon and J. C. McClune, Mon. Not. R. Astron. Soc. 262, 881 (1993).
- (2) F. de Felice and O. Zanotti, Gen. Rel. Grav. 32, 1449 (2000); B. Punsly, Black Hole Gravitohydromagnetics, 2nd. ed. (Springer-Verlag, Berlin, 2008).
- (3) V. Petrosian and A.M. Bykov, arXiv:0801.0923.
- (4) C. Chicone and B. Mashhoon, Class. Quantum Grav. 19, 4231 (2002).
- (5) C. Chicone, B. Mashhoon and B. Punsly, Int. J. Mod. Phys. D 13, 945 (2004); Phys. Lett. A 343, 1 (2005).
- (6) C. Chicone and B. Mashhoon, Class. Quantum Grav. 21, L139 (2004); Class. Quantum Grav. 22, 195 (2005); Ann. Phys. (Berlin) 14, 290 (2005); Astron. & Astrophys. 437, L39 (2005); Ann. Phys. (Berlin) 14, 751 (2005); Class. Quantum Grav. 23, 4021 (2006).
- (7) Y. Kojima and K. Takami, Class. Quantum Grav. 23, 609 (2006) [arXiv: gr-qc/0509084].
- (8) C. Chicone and B. Mashhoon, Phys. Rev. D 74, 064019 (2006).
- (9) B. Mashhoon, J. C. McClune and H. Quevedo, Class. Quantum Grav. 17, 533 (2000) [arXiv: gr-qc/9609017].
- (10) A. Lichnerowicz, Théories Relativistes de la Gravitation et de l’Électromagnétisme (Masson, Paris, 1955).
- (11) J.L. Synge, Relativity: The General Theory (North-Holland, Amsterdam, 1960).
- (12) H. Stephani, D. Kramer, M. MacCallum, C. Hoenselaers and E. Herlt, Exact Solutions of Einstein’s Field Equations, 2nd ed. (Cambridge University Press, Cambridge, 2003), ch. 22.
- (13) W.B. Bonnor, J.B. Griffiths and M.A.H. MacCallum, Gen. Rel. Grav. 26, 687 (1994).
- (14) L. Wylleman and N. Van den Bergh, Phys. Rev. D 74, 084001 (2006).
- (15) R. Opher, N.O. Santos and A. Wang, J. Math. Phys. 37, 1982 (1996).
- (16) M.A.H. MacCallum, Gen. Rel. Grav. 30, 131 (1998); J. Carot, J.M.M. Senovilla and R. Vera, Class. Quantum Grav. 16, 3025 (1999).
- (17) B. Mashhoon, Phys. Lett. A 173, 347 (1993); B. Mashhoon, in The Measurement of Gravitomagnetism: A Challenging Enterprise, edited by L. Iorio (NOVA Science, Hauppauge, New York, 2007), pp. 29–39 [arXiv: gr-qc/0311030].
- (18) C.W.F. Everitt et al., Space Sci. Rev. 148, 53 (2009).
- (19) A. E. Broderick and A. Loeb, arXiv:0908.2999; F.M. Rieger, Chin. J. Astron. Astrophys. 5S, 305 (2005); X.Y. Hong et al., New Astron. Rev. 43, 699 (1999).
- (20) C. Chicone and B. Mashhoon, arXiv:1005.1420 [gr-qc].
- (21) C. Chicone, B. Mashhoon and K. Rosquist, arXiv:1011.3477 [gr-qc].