Solar system constraints on Rn gravity

A.F. Zakharov∗ National Astronomical Observatories of Chinese Academy of Sciences, Beijing 100012, China†

arXiv:astro-ph/0611051 v1 2 Nov 2006

A.A. Nucita,‡ F. De Paolis,§ and G. Ingrosso¶ Department of Physics and INFN, University of Lecce, CP 193, I-73100 Lecce, Italy
(Dated: November 2, 2006)

Recently, gravitational microlensing has been investigated in the framework of the weak field limit of fourth order gravity theory. However, solar system data (i.e. planetary periods and light bending) can be used to put strong constraints on the parameters of this class of gravity theories. We find that these parameters must be very close to those corresponding to the Newtonian limit of the theory.
PACS numbers: 04.80.Cc, 04.20.-q, 04.25.Nx, 04.50.+h, 95.30.Sf, 96.12.Fe Keywords: Relativity and gravitation, Alternative Theories of Gravity, Gravitational fields, Planets

∗ †

Electronic address: Also at Institute of Theoretical and Experimental Physics, Moscow, 117259, Russia;

Joint Institute for Nuclear Research, Dubna, Russia
‡ §

Electronic address: Electronic address: Electronic address:


for n → ∞ it follows β → 1. INTRODUCTION Since many years different alternative approaches to gravity have been proposed in the literature such as MOND [1. if f (R) = R + 2Λ the theory coincides with General Relativity (GR) with the Λ term. As a result. Capozziello et al. it has been found that the parameter used them to study deviations from the standard Paczynski light curve for gravitational 36n4 + 12n3 − 83n2 + 50n + 1 . Of course. 6. [10. Very recently. 11]: Φ(r) = − where β= 12n2 − 7n − 1 − Gm 1+ 2r r rc β . conformal [4]. scalar-tensor [3]. Consider for example the Sun as the source of the gravitational field and the Earth as the rc varies in the range ≃ 1 − 104 AU. In particular.I. Once rc and β has been fixed. On the other hand. the gravitational potential is found to be [10. (2) √ The dependence of the β parameter on the n power is shown in Fig. 11]. (1) where f (R) is a generic function of the Ricci scalar curvature and Lm is the standard matter Lagrangian. Capozziello et al. 6n2 − 4n + 2 (3) . It is clear that for gravitational microlensing one could detect observational differences between GR and an alternative theory (the fourth order gravity in particular). depending on the considered system and its typical scale. while β is a universal parameter. Yukawa-like corrected gravity theories [5. For the Galactic microlensing case RE is about 1 AU. 10. [10] microlensing [14] and claimed that the implied deviation can be measured [15]. while for n = 1 the parameter β reduces to zero and the Newtonian gravitational field is recovered. 11] considered power law function f (R) theories of the form f (R) = f0 Rn . it has been proposed [9. 2]. Since Earth velocity is ≃ 30 km s−1 . in the framework of higher order theories of gravity – also referred to as f (R) theories – a modification of the gravity action with the form A= √ d4 x −g[f (R) + Lm ]. This is a reason why the authors 2 test particle. 7]. rc in principle is an arbitrary parameter. so that one should have different potentials at the scale RE (the Einstein radius) of the gravitational microlensing . 1. in the weak field limit [13]. For example. and so on (see papers [8] for reviews).

SOLAR SYSTEM CONSTRAINTS Light Bending Constraints As stated above. 16] θ= (1 + γ)Gm (1 + cos φ). depending on the choice of the parameters β and rc . we now discuss some observational consequences of the fourth order gravity.FIG. II. [10] have selected rc at a level of astronomical units to obtain observable signatures for nonvanishing β. A. The aim of the present paper is to show that solar system data (light bending and planetary periods) put extremely strong constraints on both rc and β parameters making this alternative theory of gravity not so attracting. φ is the solar elongation angle (between the Sun and the 3 . It is well-known that in the parameterized post-Newtonian formalism the bending angle through which a electromagnetic light ray from a distant source is deflected by a body with mass m is [8. A constraint on the proposed theory can be derived by considering the light bending effect in the Sun limb. 1: The parameter β as a function of n for fourth order gravity. c2 b (4) where b is the impact parameter.

9998 ± 0. in turn. the planet acceleration a = −∂Φ(r)/∂r is given by Gm a=− 2 1+ 2r r rc β −β r rc β .75′′ . In order to compare the orbital period predicted by the fourth order theory with the Solar System observations. B. (6) where PK = [4π 2 r3 /(Gm)]1/2 is the usual Keplerian period. In particular.001 ± 0. the observed and expected bending angles are related by θobs = (1. β = 0.e.0004). We therefore emphasize that β-values considered in [10] and [11] (i. Let us consider as a toy model a planet moving on circular orbit (of radius r) around the Sun. this means that the deflection of the light path is well described by the GR theory. 0. 0. In other words. γ = 1 and for light rays at Sun’s limb. as radio observations of distant sources have shown [18]. 0.58. at least. within 2σ (grey region) or within 5σ (light grey region) the observed value.source as viewed from Earth) and γ is the post-Newtonian parameter.25. rc ) − 1| PK PK (7) 4 .001)θGR . the deflection angle of light rays at Sun’s limb depends on both the parameters β and rc . (2). the planetary circular velocity v can be evaluated and. [17] measured the bending angles for distant compact sources and concluded that light bending angles follow GR with a very high precision (γ = 0.75) are ruled out by light deflection data. the orbital period P is given by P = PK 2 1 + √ r rc β −β r rc β −1/2 .43. θGR = 1. let us define the quantity |P − PK | ∆P = = |f (β. In the framework of the fourth order gravity theory. it is clear that only β-values nearby zero (corresponding to a completely arbitrariness of rc ) are consistent with the observed deflection angles. Recently. From Eq. Shapiro et al. We explore this dependence in Fig. (5) Accordingly. Inspecting the same figure. For GR. 2. by requiring that the expected value for the bending angle is. Planetary Constraints A stronger constraint on the fourth order gravity theory can be obtained from the motion of the solar system planets.

This is obviously impossible since the fourth order theory defines β as a parameter. There is a question about a possibility to satisfy the planetary period condition – vanishing the Eq. For these cases. As one can see from Eq. (8) (and Fig. respectively. while the specific system under consideration (the Solar system in our case) allows us to specify the rc parameter. (7) – with β parameter which is significantly different from zero. 2: Constraints on the fourth order theory parameters (β and rc ) arising from the deflection angle of light rays close to the solar limb. Hence the right hand side of Eq. rc ) the factor appearing on the right hand side of equation (6) and multiplying the usual Keplerian period. It is noticing that for scale reason we have not plotted values of rc up to 104 AU. β (8) so that Eq. 3). rc ) as a dependence on β parameter for fixed rc and planetary radii r (see Fig. Just for illustration we present the function f (β. being f (β. (8) should be satisfied for all the planetary radii. The grey and light grey regions embed the part of the parameter space allowed by solar system observations at the 2σ and 5σ confidence level. 3 as 5 . Vanishing the right hand side of Eq. (8) even with two (or more) different planetary radii.FIG. (8) is fixed for the Solar system. (7) we obtain the relation ln r = ln rc − ln (1 − β) . the observations can be restored only for β → 0. implying that it is impossible to satisfy Eq.

respectively [19]. corresponding 6 . As an example. differences in the heliocentric distances do not exceed 10 km for Jupiter and amount to 180. there is no solution of Eq. rc ) → 2 for β → 1). Uranus. as a function of β. (8).FIG. 1200 and 14000 km for Saturn. considered in [10]. Table 2 in [20]) so that the relative error in the orbital period determination is extremely low. the orbital period of Earth is T = 365.0 × 10−10 days. if we have at least one radius r ≤ rc . in the case of Mercury (dashed line). 410. so that it is impossible to satisfy Eq. rc )) is given. Moreover. Current observations allow also to evaluate the distances between the Sun and the planets of the Solar System with a great accuracy. only for β approaching zero it is expected to recover the value of the Keplerian period. 1 AU (dashed line) and ≃ 104 AU (solid line). (8). well) for each planetary radius r > rc there is β ∈ (0. (8) if the number of planets is more than one. rc ) is given as a function of β for the two limiting values of rc . 1) satisfying Eq. Earth (solid line) for √ different rc (it is clear that f (β. Both cases imply that the β parameter should be around zero. Moreover. In the above mentioned figure. there is no solution of Eq.e. but the β values depend on fixed rc and r and they are different for a fixed rc and different r. if we have at least one radius r ≤ rc . As one can note.256363051 days with an error of ∆T = 5. (8). In Fig. 4 (left panel). (8). so that they are different for a fixed rc and different r. the calculation has been performed for the Earth orbit (i. Errors in the semi-major axes of the inner planets are even smaller (see e. 3: The orbital period in units of the Keplerian one (the function f (β.g. the factor f (β. Neptune and Pluto. Venus (dotted line). r = 1 AU). In particular. If r > rc there is β ∈ (0. but the β value depends on fixed rc and r. 1) satisfying Eq.

respectively. 1 AU (dashed line) and ≃ 104 AU (solid line). < |f (β. These values can be used in order to constrain the possible values of both the parameters β and rc introduced by the fourth order gravity theory. the two limiting cases considered by Capozziello et al. in the case of Earth. since the planetary orbits are nearly circular. A more precise analysis which takes into account the planetary semi-major axes and eccentricities leads to variations of at most a few percent with respect to the results in Fig.0 × 10−12 be inferred from the middle and right panels of Fig. only for β approaching 0 it is expected to recover the value of the Keplerian period. [10]) we find the allowed upper limits on the β parameter to be 4. rc ) is given as a function of β (left panel) for the two limiting values of rc . Therefore. For rc = 1 AU and rc = 104 AU (i.FIG. r = 1 AU). respectively (since ∆P/PK = ∆β[−1 + ln (r/rc )]/4).9 × 10−13 . This can be done by requiring that ∆P/PK ∼ ∆T /T so that. rc ) − 1| ∼ 10−12 which can be solved with respect to β once the rc parameter has been < fixed to some value. we found that these data are important to constrain parameters of the fourth order gravity theory. in spite of the fact that and 3. 7 . As one can note. 4. These results can also orbital periods of planets are not generally used to test alternative theories of gravity (since it is taken for granted that the weak field approximation of these theories gives the Newtonian limit). 4.e. 4: The factor f (β. Note that only for values of β close to 0 the Solar System observation can be restored (see text for more details). the quantity f (β. Here. In the middle and right panel. to a relative error of ∆T /T less than 10−12 . the calculation have been performed at Earth (i.e. rc ) − 1 is given as a function of β for rc = 104 AU and rc = 1 AU.

124. Phys. Rev. 599. Bekenstein. for rotation curves [11] and cosmological SN type Ia [21] are ruled out by solar system data.prot. 069901(E) (2005).H. Rev. the parameter β of fourth order gravity should be very close to zero (it means that the gravitational theory should be very close to GR). Rev. laser ranging for distant objects in the solar system. parabolas and hyperbolas and so on. Acknowledgments AFZ is grateful to Dipartimento di Fisica Universita di Lecce and INFN. No doubt that one could also derive further constraints on the fourth order gravity theory by analyzing other physical phenomena such as Shapiro time delay. As a result. 2004020323 004. AAN. Bekenstein. 365 (1983). [3] C. 270. The authors are grateful to the referee whose comments improved the paper.. J. Brans and H. D. DISCUSSION GR and Newtonian theory (as its weak field limit) were verified by a very precise way at different scales. Dicke. 925 (1961). 083509 (2004). Magueijo. ApJ.D. 70. [1] M. J. [2] M. frequency shift of radio photons [22]. J. Sanders. FDP and GI have been partially supported by MIUR through PRIN 2004 . deviations of trajectories of celestial bodies from ellipses.D. Milgrom and R. D. the β parameter values considered for microlensing [10]. There are observational data which constrain parameters of alternative theories as well.III. Phys. Milgrom. 8 . 71. But our aim was only to show that only β ≃ 0 values are not in contradiction with solar system data in spite of the fact that there are a lot of speculations to fit observational data with β values significantly different from zero. astro-ph/0602266. In particular. Phys. L25 (2003). ApJ. Bekenstein. Sezione di Lecce where part of this work was carried out and to the National Natural Science Foundation of China (Grant # 10233050) and to the National Key Basic Research Foundation (Grant # TG 2000078404) for partial financial support. J.

S. S. Capozziello. 38. Lett. Allemandi. S.S. 711 (2006). Sudarsky. Capozziello.N. 506. Bludman. D. A. A. A. V. 1418 (2001). V. [8] C. V.D. S. Phys. Lett. J. Astropart. Proskurin. Cardone. Phys. M. Living Rev. Borowiec. Zinchuk. 103503 (2004). Francaviglia.A. astro-ph/0605198. Francaviglia. 458 (2006). 73. U. Phys.M. in Proc. A. S. Fischbach and C. V. 3 (2006). 4839 (2005). [6] E. 2005. Francaviglia. Phys. Erratum: E. 86. 71. E. E. 103005 (2005). Phys. J. B 634.N. 72. V. S. S. Cardone. Rev. 207 (1992). D. B 530 20 (2002). Dunsby. Hoyle. S. astro-ph/0610026. 1 (2006). 56. and H. Lett. Blaschke. B. Phys. G. Paczynski. M. 3. Troisi.[4] D. astro-ph/0511824. Fischbach. A. Odintsov.J. Intern. J.. Sami. A.. Schmidt.A. Phys. Allemandi. S. D.F.B. Phys. B.A. Capozziello. Lett. J. Rev. M. Copeland. Phys. Quantum Grav. A. Heckel. 043503 (2005).N. Clifton. D. Capozziello. Nojiri.F. G.F. B 633. Rev. Gen. S.D.H. A.F. Lett. V. V. Cardone. [10] S. 043524 (2004). M. 483 (2002).E. D. Cardone. Zinchuk. 356. V. 1 (1986). Tsujikawa. 70. 22.J. A. Troisi. 044022 (2005). Phys. B. V. Pervushin. Phys. D.N.F. Rev.D. [7] C. Gundlach. D. Troisi. Francaviglia. Lett. A. Troisi. Capozziello. Szafer. 362 (2006). Fischbach. Pervushin. Rev. Capozziello. 9 . Cardone. et al. [9] S. [13] Cosmological and weak-field constraints for gravity theories of derived by extending GR by means of a Lagrangian proportional to R1+δ in δ → 0 limit was analyzed by Clifton and Barrow [12]. V. 08. Barrow. AIP Conf. A 326. Mod.M. M.L. Phys. Rev. A. J. D. Troisi. Carloni. [14] B. A. Cardone. 56. of the 28th Spanish Relativity Meeting (ERE05): A Century of Relativity Physics.D. G. Borowiec. Behnke. Relativ. 292 (2004). 1083 (2006). D. Gravit. Zakharov.G. Proc. Lett. Rev. S. Phys. Carloni. B. Will. Phys. A. V. 70. 72. A. 1986. astro-ph/06034552 (2006). S. Szafer. Relativity. Odintsov. 063505 (2005). V. Adelberger. Allemandi. Sudarsky. [5] E. D. 93 (2006). [12] T. 841. Mod. Borowiec. 11. Capozziello. S. P.K. Kapner. V. Swanson. ApJ. Troisi. Barbashov. Intern. 21.F.F. 72. astro-ph/0603057. D. Phys. Mod. Phys. Straumann. Barbashov. Rev. Rev. Nature. [11] S. (accepted). 104019 (2006). Zakharov. 9.R.F. Phys. Cosm. V. S. A.F. A. Pervushin. Capozziello.. Talmadge. D. Rev. Troisi. Phys. Capozziello. Class. N. Lett.M. 1427 (1986). A. et al. Pervushin. Zakharov. Starobinsky. Capozziello and A. Barbashov. Zinchuk. Sahni. D.

Bertotti. Nature 425. Quantum Grav. 10 . Dillinger. 1993). Pitjeva. Moffat. Godlowski. 121101 (2004). [21] A. [19] E.S. Shapiro. Rev. Carter and W.S. 043502 (2006). Lebach and J. Phys. Theory and Experiment in Gravitational Physics (Cambridge University Press. Brownstein and J. 374 (2003).E. Cambridge.[15] We will not discuss here other more natural effects (such as the extended source effect) that could mask the phenomenon in real microlensing searches since the range of rc and β parameters considered by [10] is ruled out by solar system data. 768 (1991). Phys.L. Rev. M. Davis. [17] S. D. W. [22] B. Robertson. Borowiec. Nature 349. [16] C. Iess and P. Gregory. W. Lett. 74. [18] D.V.E.H. Szydlowski. J. Will.W. Solar System Research 39. 23 3427 (2006). 176 (2005) [20] J. 92.R. D. Class. Tortora.S. L.

Sign up to vote on this title
UsefulNot useful