Go to previous page Go up Go to next page

8.4 Third post-Newtonian equations of motion

By adding i m1a 1|δAln to Equation (171View Equation), we obtain our 3 PN equations of motion for two spherical compact stars whose representative points are defined by Equation (176View Equation),
dvi1 m1m2 i m1 ----= − --2---n12 dτ r12 [ ] 2m1m2-- i 2 2 3- 2 5m1- 4m2- +ε r212 n 12 − v1 − 2v2 + 4(⃗v1 ⋅⃗v2) + 2(⃗n12 ⋅⃗v2) + r12 + r12 m1m2 +ε2---2--V i[4(⃗n12 ⋅⃗v1) − 3(⃗n12 ⋅⃗v2)] r12 [ 4m1m2-- i 4 2 2 3-2 2 9- 2 2 +ε r2 n 12 − 2v2 + 4v2(⃗v1 ⋅⃗v2) − 2(⃗v1 ⋅⃗v2) + 2v1(⃗n12 ⋅⃗v2) + 2 v2(⃗n12 ⋅⃗v2) 12 2 2 − 6(⃗v ⋅⃗v )(⃗n ⋅⃗v )2 − 15(⃗n ⋅⃗v )4 − 57m-1 − 9 m-2− 69m1m2-- 1 2 12 2 8 12 2 4 r212 r212 2 r212 m ( 15 5 5 39 + --1 − ---v21 + -v22 − --(⃗v1 ⋅⃗v2) +---(⃗n12 ⋅⃗v1)2 r12 4 4 2 2 ) 17 2 − − 39(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + --(⃗n12 ⋅⃗v2) 2 ] m2- ( 2 2 2) + r12 4v2− 8(⃗v1 ⋅ ⃗v2)+2 (⃗n12 ⋅⃗v1) − 4(⃗n12 ⋅ ⃗v1)(⃗n12 ⋅⃗v2)− 6(⃗n12 ⋅⃗v2) [ ( ) +ε4m1m2--V i m1- − 63(⃗n12 ⋅⃗v1) + 55(⃗n12 ⋅⃗v2) + m2- (− 2 (⃗n12 ⋅⃗v1) − 2(⃗n12 ⋅⃗v2)) r212 r12 4 4 r12 2 2 2 ⃗ + v1(⃗n12 ⋅⃗v2) + 4v2(⃗n12 ⋅⃗v1) − 5v 2(⃗n1]2 ⋅⃗v2) − 4(⃗v1 ⋅⃗v2)(⃗n12 ⋅V ) 2 9- 3 − 6(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + 2(⃗n12 ⋅⃗v2) 2 [ ( ) ( ) ] +ε54m--1m2- ni (⃗n ⋅ ⃗V ) − 6 m1-+ 52m2- + 3V 2 + Vi 2m1- − 8 m2-− V 2 5r312 12 12 r12 3 r12 r12 r12 m m [35 15 15 +ε6--12-2ni12 ---(⃗n12 ⋅⃗v2)6 −---(⃗n12 ⋅⃗v2)4v21 + --(⃗n12 ⋅⃗v2)4(⃗v1 ⋅⃗v2) r12 16 8 2 2 2 15- 4 2 3- 2 2 2 + 3(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2) − 2 (⃗n12 ⋅⃗v2) v2 + 2(⃗n12 ⋅⃗v2)v1v2 15 − 12(⃗n12 ⋅⃗v2)2(⃗v1 ⋅⃗v2)v22 − 2(⃗v1 ⋅⃗v2)2v22 +---(⃗n12 ⋅⃗v2)2v42 4 6 2 + 4(⃗v1 ⋅⃗v2)v2 − 2v 2 m1 ( 171 171 + --- − ----(⃗n12 ⋅⃗v1)4 +----(⃗n12 ⋅⃗v1)3(⃗n12 ⋅⃗v2) r12 8 2 723- 2 2 383- 3 − 4 (⃗n12 ⋅⃗v1) (⃗n12 ⋅⃗v2) + 2 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) 455 4 229 2 2 205 2 − ----(⃗n12 ⋅⃗v2) + ----(⃗n12 ⋅⃗v1) v1 −----(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v1 8 4 2 + 191-(⃗n12 ⋅⃗v2)2v21 − 91-v41 − 229(⃗n12 ⋅⃗v1)2(⃗v1 ⋅⃗v2) 4 8 2 225- 2 + 244(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2) − 2 (⃗n12 ⋅⃗v2) (⃗v1 ⋅⃗v2) 91 2 177 2 229 2 2 + ---v1(⃗v1 ⋅⃗v2) −----(⃗v1 ⋅⃗v2) + ----(⃗n12 ⋅⃗v1) v2 2 4 4 − 283-(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v22 + 259(⃗n12 ⋅⃗v2)2v22 − 91v21v22 2 ) 4 4 2 81-4 + 43(⃗v1 ⋅⃗v2)v2 − 8 v2 m ( + --2 − 6(⃗n12 ⋅⃗v1)2(⃗n12 ⋅⃗v2)2 + 12(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)3 r12 + 6(⃗n12 ⋅⃗v2)4 + 4(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2) 2 2 + 12(⃗n12 ⋅⃗v2) (⃗v1 ⋅⃗v2) + 4(⃗v1 ⋅⃗v2) ) − 4(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v22 − 12(⃗n12 ⋅⃗v2)2v22 − 8(⃗v1 ⋅ ⃗v2)v22 + 4v42 2( + m-2 − (⃗n12 ⋅⃗v1)2 + 2(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + 43(⃗n12 ⋅⃗v2)2 r212 2 ) + 18 (⃗v1 ⋅⃗v2) − 9v22 ( m1m2-- 415- 2 375- 1113- 2 + r2 8 (⃗n12 ⋅⃗v1) − 4 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + 8 (⃗n12 ⋅⃗v2) 12 2 2 ) − 615π--(⃗n ⋅V⃗)2+18v2 + 123-π-V 2+33 (⃗v ⋅⃗v )− 33v2 64 12 1 64 1 2 2 2 m2 ( 2069 939 + -21 − -----(⃗n12 ⋅⃗v1)2 + 543 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) −--(⃗n12 ⋅ ⃗v2)2 r12 8 ) 4 471 2 357 357 2 + ---v1 − ----(⃗v1 ⋅⃗v2) +----v2 8 ( 4 ) 8 ( )] 16m32- m21m2-- 547- 41-π2 13m31- m1m22- 545- 41π2- + r3 + r3 3 − 16 − 12r3 + r3 3 − 16 [ 12 12 12 12 +ε6m1m2--V i 15(⃗n ⋅⃗v )(⃗n ⋅⃗v )4 − 45(⃗n ⋅⃗v )5 − 3(⃗n ⋅⃗v )3v2 r212 2 12 1 12 2 8 12 2 2 12 2 1 + 6(⃗n ⋅⃗v )(⃗n ⋅⃗v )2(⃗v ⋅⃗v ) − 6(⃗n ⋅⃗v )3(⃗v ⋅ ⃗v ) 12 1 12 2 1 2 12 2 1 2 − 2(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2)2 − 12 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)2v22 + 12(⃗n12 ⋅⃗v2)3v22 + (⃗n ⋅⃗v )v2v2 − 4(⃗n ⋅⃗v )(⃗v ⋅⃗v )v2+ 8(⃗n ⋅⃗v )(⃗v ⋅⃗v )v2 12 2 142 12 1 41 2 2 12 2 1 2 2 + 4(⃗n12 ⋅⃗v1)v2 − 7(⃗n12 ⋅⃗v2)v2 m2-( 2 2 3 + r12 − 2(⃗n12 ⋅⃗v1) (⃗n12 ⋅⃗v2) + 8(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + 2 (⃗n12 ⋅⃗v2) + 2(⃗n ⋅⃗v )(⃗v ⋅⃗v ) + 4(⃗n ⋅⃗v )(⃗v ⋅⃗v ) 12 1 1 2 12 ) 2 1 2 − 2(⃗n12 ⋅⃗v1)v22 − 4(⃗n12 ⋅⃗v2)v22 m ( 243 565 + --1 − ----(⃗n12 ⋅⃗v1)3 + ----(⃗n12 ⋅⃗v1)2(⃗n12 ⋅⃗v2) r12 4 4 269- 2 95- 3 207- 2 − 4 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) − 12(⃗n12 ⋅⃗v2) + 8 (⃗n12 ⋅⃗v1)v1 137 27 − ----(⃗n12 ⋅⃗v2)v21 − 36(⃗n12 ⋅⃗v1)(⃗v1 ⋅⃗v2) +--(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2) 8 ) 4 + 81-(⃗n ⋅⃗v )v2 + 83-(⃗n ⋅⃗v )v2 8 12 1 2 8 12 2 2 2 + m-2(4(⃗n12 ⋅⃗v1) + 5(⃗n12 ⋅⃗v2)) r212 m1m2 ( 307 479 123π2 ) + --2--- − ---(⃗n12 ⋅⃗v1) + ---(⃗n12 ⋅⃗v2) + ------(⃗n12 ⋅ ⃗V ) r12( 8 8 ) ] 32
in the harmonic gauge.

Now we list some features of our 3 PN equations of motion. In the test-particle limit, our 3 PN equations of motion coincide with a geodesic equation for a test-particle in the Schwarzschild metric in harmonic coordinates (up to 3 PN order). Suppose that star 1 is a test particle, star 2 is represented by the Schwarzschild metric, and ⃗v = 0 2. Then the geodesic equation for star 1 becomes

( ) ai = − m2-ni + ε2m2- − v2ni + 4m2-ni + 4(⃗n ⋅⃗v )vi 1 r212 12 r212 1 12 r12 12 12 1 1 m2 ( m ) + ε4--2 2(⃗n12 ⋅⃗v1)2ni12 − 9-2 ni12 − 2 (⃗n12 ⋅⃗v1)vi1 r312 r12 6m32( 3 i 16m2 i i) 7 + ε -4- − (⃗n12 ⋅⃗v1) n12 + -----n 12 + 4(⃗n12 ⋅⃗v1)v1 + 𝒪 (ε ) (182 ) r12 r12
in the harmonic gauge. Thus, in the test particle limit Equation (181View Equation) coincides with the geodesic equation for a test particle in the Schwarzschild metric up to 3 PN order.

With the help of the formulas developed in [28], we have checked the Lorentz invariance of Equation (181View Equation) (in the post-Newtonian perturbative sense). Also, we have checked that our 3 PN acceleration admits a conserved energy of the binary orbital motion (modulo the 2.5 PN radiation reaction effect). In fact, the energy E of the binary associated with Equation (181View Equation) is

1- 2 m1m2-- E = 2 m1v 1 − 2r [ 12 2 ( ) ] + ε2 3-m1v4 + m-1m2-+ m1m2-- 3v2− 7(⃗v1 ⋅⃗v2) − 1(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) 8 1 2r212 2r12 1 2 2 [ 5 m3 m 19m2 m2 + ε4 ---m1v61 − --13-2− ----13--2 16 ( 2r12 8r12 ) m21m2 2 7 2 29 2 13 2 + ---2-- − 3v1 + -v2 + ---(⃗n12 ⋅⃗v1) − ---(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + (⃗n12 ⋅⃗v2) 2r12 ( 2 2 2 m1m2-- 3- 3 3- 2 2 9- 2 + 4r12 2(⃗n12 ⋅⃗v1) (⃗n12 ⋅⃗v2) + 4 (⃗n12 ⋅⃗v1) (⃗n12 ⋅⃗v2) − 2 (⃗n12 ⋅⃗v1)(⃗n12 ⋅ ⃗v2)v 1 13 21 13 − ---(⃗n12 ⋅⃗v2)2v21 +---v41 + --(⃗n12 ⋅⃗v1)2(⃗v1 ⋅⃗v2) 2 2 2 )] 55- 2 17- 2 31- 2 2 + 3(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2) − 2 v1(⃗v1 ⋅⃗v2) + 2 (⃗v1 ⋅⃗v2) + 4 v1v 2 [ 4 3 2 + ε6 -35-m v8+ 3m-1m2- + 469m-1m-2- 128 1 1 8r412 18r412 m2 m2 ( 547 3115 123π2 + --13-2 ----(⃗n12 ⋅⃗v1)2 −-----(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) −------(⃗n12 ⋅⃗v1)(⃗n12 ⋅ ⃗V) 2r12 6 24 ) 32 575 2 41π2 4429 − ----v1 + -----(V⃗ ⋅⃗v2) + ----(⃗v1 ⋅⃗v2) ( 9 32 72 m31m2- 437- 2 317- 2 301- 2 + 2r3 − 4 (⃗n12 ⋅⃗v1) + 4 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + 3(⃗n12 ⋅⃗v2) + 12 v1 12 ) − 337-(⃗v ⋅⃗v ) + 5-v2 12 1 2 2 2 m m ( 5 5 5 + --1--2 − --(⃗n12 ⋅⃗v1)5(⃗n12 ⋅⃗v2) −---(⃗n12 ⋅⃗v1)4(⃗n12 ⋅⃗v2)2 −--(⃗n12 ⋅⃗v1)3(⃗n12 ⋅⃗v2)3 r12 16 16 32 19- 3 2 15- 2 2 2 + 16 (⃗n12 ⋅⃗v1) (⃗n12 ⋅⃗v2)v 1 + 16 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) v1 3 19 21 + --(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)3v21 +---(⃗n12 ⋅⃗v2)4v21 −---(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v41 4 16 16 − 2(⃗n ⋅⃗v )2v4 + 55v6 − 19-(⃗n ⋅⃗v )4(⃗v ⋅⃗v ) − (⃗n ⋅ ⃗v )3(⃗n ⋅⃗v )(⃗v ⋅⃗v ) 12 2 1 16 1 16 12 1 1 2 12 1 12 2 1 2 15- 2 2 45- 2 2 − 32 (⃗n12 ⋅⃗v1) (⃗n12 ⋅ ⃗v2)(⃗v1 ⋅⃗v2) + 16(⃗n12 ⋅⃗v1) v1(⃗v1 ⋅⃗v2) 5 11 139 + --(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v21(⃗v1 ⋅⃗v2) +--(⃗n12 ⋅⃗v2)2v21(⃗v1 ⋅⃗v2) −----v41(⃗v1 ⋅⃗v2) 4 4 16 − 3-(⃗n ⋅⃗v )2(⃗v ⋅⃗v )2 + -5-(⃗n ⋅⃗v )(⃗n ⋅ ⃗v )(⃗v ⋅⃗v )2 + 41v2(⃗v ⋅⃗v )2 4 12 1 1 2 16 12 1 12 2 1 2 8 1 1 2 -1- 3 45- 2 2 2 23- 2 2 79- 4 2 + 16 (⃗v1 ⋅⃗v2) − 16(⃗n12 ⋅⃗v1) v1v2 − 32(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v1v2 + 16 v1v2 ) − 161-v21v22(⃗v1 ⋅⃗v2) 32 m2 m2 ( 49 75 187 11 + --12--- − --(⃗n12 ⋅⃗v1)4 + --(⃗n12 ⋅⃗v1)3(⃗n12 ⋅⃗v2) − ----(⃗n12 ⋅⃗v1)2(⃗n12 ⋅⃗v2)2 +---v41 r12 8 8 8 2 247- 3 49- 2 2 81- 2 + 24 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2) + 8 (⃗n12 ⋅⃗v1)v1 + 8 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v1 21 15 3 − ---(⃗n12 ⋅⃗v2)2v21 −---(⃗n12 ⋅⃗v1)2(⃗v1 ⋅⃗v2) −--(⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)(⃗v1 ⋅⃗v2) 4 2 2 + 21-(⃗n12 ⋅⃗v2)2(⃗v1 ⋅⃗v2) − 27v2(⃗v1 ⋅⃗v2) + 55(⃗v1 ⋅⃗v2)2 + 49(⃗n12 ⋅⃗v1)2v2 4 1 2 4 2 27- 2 3- 2 2 55- 2 2 2 − 2 (⃗n12 ⋅⃗v1)(⃗n12 ⋅⃗v2)v2 + 4(⃗n12 ⋅⃗v2)v2 + 4 v1v2 − 28(⃗v1 ⋅⃗v2)v2 135 ) ] + ----v42 16 + (1 ↔ 2) + 𝒪 (ε7). (183 )
This orbital energy of the binary is computed based on that one found in Blanchet and Faye [27Jump To The Next Citation Point], the relation between their 3 PN equations of motion and our result described in Section 8.5 below, and Equation (167View Equation). (After constructing E given as in Equation (183View Equation), we have checked that our 3 PN equations of motion make E to be conserved.)

We note that Equation (171View Equation) as well gives a correct geodesic equation in the test-particle limit, is Lorentz invariant, and admits the conserved energy. These facts can be seen by the form of i a1|δAln, Equation (175View Equation); it is zero when m1 → 0, is Lorentz invariant up to 3 PN order, and is the effect of the mere redefinition of the dipole moments which does not break energy conservation.

Finally, we here mention one computational detail. We have retained during our calculation ℛ-dependent terms with positive powers of ℛ or logarithms of ℛ. As stated below Equation (76View Equation), it is a good computational check to show that our equations of motion do not depend on ℛ physically. In fact, we found that the ℛ-dependent terms cancel each other out in the final result. There is no need to employ a gauge transformation to remove such an ℛ dependence. As for terms with negative powers of ℛ, we simply assume that those terms cancel out the ℛ dependent terms from the far zone contribution. Indeed, Pati and Will [129Jump To The Next Citation Point130Jump To The Next Citation Point], whose method we have adopted to compute the far zone contribution, have proved that all the ℛ-dependent terms cancel out between the far zone and the near zone contributions through all post-Newtonian orders.


  Go to previous page Go up Go to next page