The first-order orbital equation

Maurizio M. D’Eliseoa兲
Osservatorio S. Elmo, Via A. Caccavello 22, 80129 Napoli, Italy
共Received 27 February 2006; accepted 15 December 2006兲
We derive the first-order orbital equation employing a complex variable formalism. We then
examine Newton’s theorem on precessing orbits and apply it to the perihelion shift of an elliptic
orbit in general relativity. It is found that corrections to the inverse-square gravitational force law
formally similar to that required by general relativity were suggested by Clairaut in the 18th
century. © 2007 American Association of Physics Teachers.
关DOI: 10.1119/1.2432126兴

I. INTRODUCTION II. THE FIRST-ORDER ORBITAL EQUATION

Almost all classical mechanics textbooks derive the ellip- If we multiply Eq. 共1兲 by r*, we have
tical orbit of the two-body planetary problem by means of
rr* r2 ␮
well known methods. In this paper we derive the first-order r̈r* = − ␮ 3 = − ␮ 3 =− . 共3兲
orbital equation by using the complex variable formalism. r r r
The latter is a useful tool for studying this old problem from
If we take the imaginary part of both sides of Eq. 共3兲, we find
a new perspective. From the orbital equation we can extract
all the properties of elliptic orbits. Newton’s theorem of re-
volving orbits,1 which establishes the condition for which a
closed orbit revolves around the center of force, has a wide
Im共r̈r*兲 = − Im 冉冊␮
r
= 0. 共4兲

range of applicability, and its application to an inverse- It is easy to verify that

square force allows us to apply the first-order orbital equa-
tion. d
Im共ṙr*兲 = Im共r̈r*兲 + Im共ṙṙ*兲. 共5兲
A revolving 共precessing兲 ellipse reminds us of the general dt
relativistic perihelion shift of the planet Mercury. We explain
why an approximate general relativistic force found by We have Im共ṙṙ*兲 = 0 because the term in brackets is the
Levi-Civita2 in his lectures on general relativity gives the square of the module 兩ṙ兩, a real quantity. Then from Eqs. 共5兲
same perihelion shift of the r−4 general relativistic force de- and 共4兲 we write
rived in textbooks. Our results suggest an interesting link d
with the work of the 18th century scientist Clairaut.3,4 Im共ṙr*兲 = 0. 共6兲
We identify the plane of the motion of the gravitational dt
two-body problem with the complex plane.5 An object of Equation 共6兲 implies that Im共ṙr*兲 is time independent. We
mass M is at the origin. The position x, y of the second denote its real value by ᐉ and write
object of mass m is given by r = x + iy, and the equation of
motion can be expressed as ṙr* − ṙ*r
Im共ṙr*兲 = = ᐉ. 共7兲
2i
r e i␪
r̈ = − ␮ 3 = − ␮ 2 , 共1兲 Equation 共7兲 is the area integral.
r r
This derivation holds for any central force f共r兲ei␪ = f共r兲
where ␮ = G共m + M兲, r = rei␪ = r共cos ␪ + i sin ␪兲, r* = re−i␪ is ⫻共r / r兲. We substitute Eq. 共2兲 into Eq. 共7兲 and find the fun-
the complex conjugate of r, r = 兩r 兩 = 冑rr* = 冑x2 + y 2, and ␪ damental relation ᐉ = r2␪˙ , which can be cast in three equiva-
= ␪共t兲 is the true longitude, that is, the point 共x , y兲 has the lent forms:
polar form rei␪, where r is the modulus and ␪ is its argument.
From r = rei␪ we have by differentiation with respect to time 1 ␪˙
= , 共8a兲
r2 ᐉ
ṙ = 共ṙ + ir␪˙ 兲ei␪ , 共2兲
r2
dt = d␪ , 共8b兲
where r˙* = ṙ*. The solution of Eq. 共1兲 requires knowledge of ᐉ
the functions r共␪兲 and ␪共t兲, but we are interested here only in
the determination of the function r共␪兲, which describes the d ᐉ d
geometry of the orbit. = . 共8c兲
dt r2 d␪
We denote by Re共r兲 and Im共r兲 the real and the imaginary
parts of r, respectively. Thus Re共r兲 = 共r + r*兲 / 2 = x = r cos ␪ We can rewrite Eq. 共1兲 using Eq. 共8a兲 as
and Im共r兲 = 共r − r*兲 / 2i = y = r sin ␪. It is useful to consider ␮ ˙ i␪ i ␮ d i␪
complex variables as vectors starting from the origin so that r̈ = − ␪e = e , 共9兲
ᐉ ᐉ dt
Re共A兲 is the component of A along the x 共real兲 axis, and
Im共A兲 is the component along the y 共imaginary兲 axis. or

352 Am. J. Phys. 75 共4兲, April 2007 http://aapt.org/ajp © 2007 American Association of Physics Teachers 352
 d
dt
冉 i␮
ṙ − ei␪ = 0.
ᐉ
冊 共10兲 u + iu⬘ =
␮
ᐉ2
共1 + ee−i␪兲. 共17兲

The expression in parentheses is complex and constant in Despite its heterogeneous nature, it is convenient to write the
time. For convenience we denote it as 共i␮e兲 / ᐉ. The reason left-hand side of Eq. 共17兲 in terms of u = u共␪兲 ⬅ u共␪兲
for this choice will soon be apparent. We have thus deduced + iu⬘共␪兲, so that we have
the Laplace integral6 ␮
u= 共1 + ee−i␪兲, 共18兲
i␮ ᐉ2
ṙ = 共ei␪ + e兲, 共11兲
ᐉ which we call the first-order orbital equation.
From Eq. 共18兲 we can immediately deduce the orbit and its
where ṙ is the orbital velocity, e = e exp 共i␻兲 is a complex apsidal points. The orbit is given by the real part of u,
constant that we will call the eccentricity vector, and e is the
共scalar兲 eccentricity. The vector e is directed toward the peri- 1 ␮ ␮
Re共u兲 = = 2 关1 + Re共ee−i␪兲兴 = 2 关1 + e cos 共␪ − ␻兲兴.
helion, the point on the orbit of nearest approach to the cen- r ᐉ ᐉ
ter of force, and ␻ is the argument of the perihelion. 共19兲
If we use the area integral to eliminate the explicit pres-
ence of t in Eq. 共11兲, we obtain a relation between r and ␪. The apsidal points are determined from the condition
One way to integrate Eq. 共1兲 twice with respect to time is to Im共u兲 = 0 because r⬘共␪兲 = 0 at these points. If Im共u兲 = 0 for
substitute into Eq. 共7兲 the expression for ṙ given by Eq. 共11兲: every value of ␪, then e = 0, and we have a circular orbit with

冋 册
u = Re共u兲 = ␮ / ᐉ2. If 0 ⬍ 兩e 兩 ⬍ 1, then Eq. 共18兲 gives the posi-
i␮ tion of the two apsidal points rmin and rmax. At these points
ᐉ = Im共ṙr*兲 = Im 共r + er*兲
ᐉ we have
␮ ␮
= 关r + Im共ier*兲兴 Im共u兲 = − e sin 共␪ − ␻兲 = 0, 共20兲
ᐉ ᐉ2
␮ so that, by considering the derivative 关Im共u兲兴⬘ = u⬙共␪兲, we
= r关1 + Re共ee−i␪兲兴, 共12兲
ᐉ find rmin when ␪ + 2n␲ = ␻ and rmax when ␪ + 共2n + 1兲␲ = ␻,
where n = 0 , 1 , 2 , . . ..
from which we can solve for r If we denote by D the differential operator d / d␪, Eq. 共18兲
may be written in operator form as
ᐉ 2/ ␮
r= . 共13兲 ␮
1 + e cos 共␪ − ␻兲 u = 共1 + iD兲u = 共1 + ee−i␪兲. 共21兲
ᐉ2
Equation 共13兲 is the relation in polar coordinates of the orbit,
which is a conic section of eccentricity e with a focus at the By multiplying both sides on the left by 共1 − iD兲, we obtain
origin. If the orbit is an ellipse 共0 ⬍ e ⬍ 1兲, we have the rela- ␮
tion ᐉ2 / ␮ = a共1 − e2兲, where a is the semi-major axis 共which 共1 − iD兲共1 + iD兲u = 共D2 + 1兲u = u⬙ + u = , 共22兲
ᐉ2
lies on the apse line兲.7 Thus the names given earlier to 兩e 兩
= e and ␻ are justified. which is Binet’s orbit equation.9
Another way to find the orbit is to transform the time
derivative into a ␪ derivative. We start from Eq. 共11兲, which III. THE PRECESSING ELLIPSE
is a function of ␪ instead of time. From Eq. 共8c兲 we obtain
The two-body solution we have found together with the
ᐉ i ␮ i␪ appropriate corrections due to the presence of other bodies
ṙ = 2 r⬘ = 共e + e兲, 共14兲
r ᐉ does not account for the observed residual precession of the
planetary perihelia.10 An explanation in classical terms is that
where a prime denotes differentiation with respect to ␪. Then a small additional force acts on all the planets causing pre-
cession.
i ␮ i␪
r⬘ = 共rei␪兲⬘ = 共r⬘ + ir兲ei␪ = 共e + e兲r2 . 共15兲 All perturbing central forces of the type F ⬃ r−mei␪, with
ᐉ2 m ⱖ 3, produce a secular motion of the apse of an elliptical
orbit. Conversely, from the observed planetary apse motion
If we multiply by e−i␪, we obtain the complex Bernoulli we can deduce by Newton’s theorem the presence of a per-
equation8 turbing inverse-cube force F ⬃ r−3ei␪. This result was ob-
i␮ tained by Newton in more general terms using the following
r⬘ + ir = 共1 + ee−i␪兲r2 . 共16兲 reasoning.1
ᐉ2 Consider a closed orbit determined by the centripetal force
−f共r兲ei␪. If we let r = r共␴␪兲, where ␴ ⫽ 1 is an arbitrary real
If we take the imaginary and the real parts of both sides of
constant, we will obtain the same orbit as for ␴ = 1, but re-
Eq. 共16兲, we obtain the orbit r共␪兲 and its derivative r⬘共␪兲,
volving around the center of force 共the two orbits are coin-
respectively. However, it is better to change the dependent cident when ␪ = 0兲. From the area integral of the first orbit,
variable, so we divide both sides by r2, make the variable ˙
change r共␪兲 → 1 / u共␪兲, and multiply by i. We find r2␪˙ = ᐉ, we obtain r2␴␪˙ = ␴ᐉ, which we write as r2˜␪ = ᐉ̃. This

353 Am. J. Phys., Vol. 75, No. 4, April 2007 Maurizio M. D’Eliseo 353
integral is for the centripetal force −f̃共r兲ei␪. The radial equa- e i␪ 3 ␣ ᐉ 2 i␪
r̈ = − ␮ − 4 e 共32兲
tions of these two orbits with the same r共t兲 are r2 r

r̈ − r␪˙ 2 = − f共r兲, 共23a兲 and we see that general relativity introduces an effective per-
turbative r−4 force.
˜ From Eq. 共31兲 or Eqs. 共11兲 and 共32兲 we can deduce12 the
r̈ − r␪˙ 2 = − f̃共r兲, 共23b兲 standard formula for the perihelion shift given in general
relativity by
from which we obtain

冉 冊
6␲␣␮ 6␲␣
˜ ᐉ̃ 2 ᐉ2 ᐉ2共␴2 − 1兲 ⌬␻ = = . 共33兲
f̃共r兲 − f共r兲 = r共␪˙ 2 − ␪˙ 2兲 = r 4 − 4 = . 共24兲 ᐉ 2
a共1 − e2兲
r r r3
If we equate Eqs. 共30兲 and 共33兲 and solve for ␴, we obtain
The extra radial force is outward or inward depending on ␴2 ⬇ 1 − 6␣␮ / ᐉ2, and by using Eq. 共25兲 we obtain the
whether ␴ is greater or less than unity. inverse-cube perturbation that gives the same perihelion shift
Thus far the force f共r兲 is arbitrary, but if we specialize to as predicted by general relativity:
the inverse-square gravitational force, then the first-order or-
bital equation 共18兲 with the perturbing inverse-cube force 6 ␣ i␪
F=−␮ e . 共34兲
ᐉ 共␴ − 1兲 i␪
2 2 r3
e = ᐉ2共␴2 − 1兲u3ei␪ 共25兲
r3 That is, we can obtain the same precession using either an
r−3 or r−4 perturbative force.
takes the form
Levi-Civita obtained an effective r−3 force in general rela-
␮ tivity using a method based on a new form of Hamilton’s
u= 共1 + ee−i␴␪兲. 共26兲 principle2 devised to go smoothly from the classical equation
ᐉ2
of motion to the Einstein field equation. His approximation is
Hence not as general as the usual r−4 effective force because it does
not produce the bending of light rays, a subject that Levi-
␮ ␮ Civita treated with another ingenious approximation.
u = Re共u兲 = + e cos 共␴␪ − ␻兲 共27兲 Binet’s equation for the r−3 perturbative force, obtained
ᐉ2 ᐉ2
from the equation of the motion by the use of Eq. 共8c兲 and
is an ellipse precessing around the focus with an angular the variable change r共␪兲 → 1 / u共␪兲, is

冉 冊
velocity proportional to the radius vector. This description
becomes more accurate as ␴ approaches unity. The apsidal ␮ ␮
points are given by Im共u兲 = 0. From Eq. 共26兲 and e u⬙ + 1 − 6␣ 2 u = 2. 共35兲
ᐉ ᐉ
= e exp 共i␻兲, we have at the apsidal points
The null-geodesic equation of light rays requires that we for-
sin 共␴␪ − ␻兲 = 0. 共28兲 mally put ␮ / ᐉ2 = 0 in Eq. 共35兲,13 so that it becomes
In particular we have rmin when u⬙ + u = 0. 共36兲
␴␪ − ␻ = ␪ − 关␻ + 共1 − ␴兲␪兴 = 0, 共29兲
The solution of Eq. 共36兲 is k sin ␪ where k = const and 0
so that after one complete revolution the angular perihelion ⱕ ␪ ⱕ ␲. In terms of the radius r = 1 / u, the solution becomes
shift is r sin ␪ = 1 / k. Because r sin ␪ is the Cartesian coordinate y,
the solution represents a straight line parallel to the x axis, so
⌬␻ = 2␲共1 − ␴兲. 共30兲 that the light ray is not deflected at all by the sun’s gravita-
tional field in this approximation.
If ␴ ⬍ 1, then ⌬␻ ⬎ 0 and the shift is positive, while for ␴ As we have seen, the weak-field approximation of general
⬎ 1 we have ⌬␻ ⬍ 0 and the shift is negative. relativity adds an effective r−3 force or a r−4 force 共depend-
ing on the approximation method used兲 to Newton’s inverse
square force to explain the perihelion motion. It is interesting
IV. GENERAL RELATIVITY that these results were proposed in the mid-18th century by
the mathematician and astronomer Clairaut who proposed
The foregoing considerations have a direct application in the addition of a small r−n force to the r−2 gravitational force
general relativity. The general relativistic Binet’s orbit equa- to explain the swift motion of the lunar perigee. In particular,
tion, which is obtained from the geodesic equation in the he examined the influence of both r−4 and r−3 terms.14 It was
Schwarzschild space-time, is11 later recognized by Clairaut that this addition was unneces-
sary, because a purely r−2 force law could completely explain
␮ the motion of the Moon. The apparently anomalous secular
u⬙ + u = + 3␣u2 , 共31兲 motion of the perigee was due to discarded noncentral force
ᐉ2
terms in the process of successive approximations.15 No
where ␣ = GM / c2 ⬅ ␮ / c2 is the gravitational radius of the doubt Clairaut would have again made his suggestion if he
central body, and c is the speed of light. The corresponding had known about the anomalous motion of Mercury’s peri-
equation of motion is helion. Without a new first principles gravitational theory16

354 Am. J. Phys., Vol. 75, No. 4, April 2007 Maurizio M. D’Eliseo 354
 12
he probably would have employed a phenomenological ap- See, for example, B. Davies, “Elementary theory of perihelion preces-
proach and introduced one of the two forces by empirically sion,” Am. J. Phys. 51, 909–911 共1983兲; N. Gauthier, “Periastron preces-
sion in general relativity,” ibid. 55, 85–86 共1987兲; T. Garavaglia, “The
adjusting the numerical factors.
Runge-Lenz vector and Einstein perihelion precession,” ibid. 55, 164–
165 共1987兲; C. Farina and M. Machado, “The Rutherford cross section
a兲
Electronic mail: s.elmo@mail.com and the perihelion shift of Mercury with the Runge-Lenz vector,” ibid.
1
S. Chandrasekhar, Newton’s Principia for the Common Reader 共Claren- 55, 921–923 共1987兲; D. Stump, “Precession of the perihelion of Mer-
don, Oxford, 1995兲, pp. 184–187. In particular, Proposition XLIV- cury,” ibid. 56, 1097–1098 共1988兲; K. T. McDonald, “Right and wrong
Theorem XIV: The difference of the forces, by which two bodies may be use of the Lenz vector for non-Newtonian potentials,” ibid. 58, 540–542
made to move equally, one in a fixed, the other in the same orbit revolv- 共1990兲; S. Cornbleet, “Elementary derivation of the advance of the peri-
ing, varies inversely as the cube of their common altitudes. helion of a planetary orbit,” ibid. 61, 650–651 共1993兲; B. Dean, “Phase-
2
T. Levi-Civita, Fondamenti di Meccanica Relativistica 共Zanichelli, Bolo- plane analysis of perihelion precession and Schwarzschild orbital dynam-
gna, 1929兲, p. 123. ics,” ibid. 67, 78–86 共1999兲.
13
3
A. C. Clairaut, “Du Systeme du Monde, dan les principes de la gravita- R. Adler, M. Bazin, and M. Shiffer, Introduction to General Relativity,
tion universelle,” Histoires de l’Academie Royale des Sciences, mem. 2nd ed. 共McGraw-Hill, New York, 1975兲, p. 216, Eq. 共6.149兲.
14
1745 and Ref. 4. A. C. Clairaut, “Du Systeme du Monde, dan les principes de la gravita-
4
We have reproduced many papers of historical interest at 具gallica.bnf.fr/典. tion universelle,” Histoires de l’Academie Royale des Sciences, mem.
5
T. Needham, Visual Complex Analysis 共Oxford U. P., New York, 1999兲. 1745, p. 337: Clairaut wrote that “The moon without doubt expresses
6
The Laplace integral can be found in P. S. Laplace, Ouvres 共Gauthier- some other law of attraction than the 关inverse兴 square of the distance, but
Villars, Paris, 1878兲, Tome 1, p. 181, formula P. To obtain Eq. 共11兲, we the principal planets do not require any other law. It is therefore easy to
need to use z = 0, c = xẏ − ẋy = ᐉ, f = ␮ Re共e兲, and f ⬘ = ␮ Im共e兲, and add the respond to this difficulty, and noting that there are an infinite number of
first relation to the second one multiplied by −i 共see Ref. 4兲. To keep the laws which give an attraction which differs very sensibly from the law of
customary notation we use the same letter e for the eccentricity and for the squares for small distances, and which deviates so little for the large,
the complex exponential. that one cannot perceive it by observations. One might regard, for ex-
7 ample, the analytic quantity of the distance composed of two terms, one
A parallel treatment of the two-body problem with vectorial methods is
given by V. R. Bond and M. C. Allman, Modern Astrodynamics 共Prince- having the square of the distance as its divisor, and the other having the
ton U. P., Princeton, NJ, 1998兲. square square.” On p. 362, Clairaut examined the effect of a perturbing
8
R. E. Williamson, Introduction to Differential Equations 共McGraw-Hill, inverse-cube force. This memoir is dated 15 November 1747 and can be
New York, 1997兲, p. 84. found in Ref. 4. Clairaut was also the first to introduce a revolving ellipse
9
R. d’Inverno, Introducing Einstein’s Relativity 共Oxford U. P., New York, as a first approximation to the motion of the moon. This idea is some-
2001兲, p. 194. times called Clairaut’s device or Clairaut’s trick.
15
10
For a complete calculation of all the perturbing effects see M. G. Stewart, See F. Tisserand, Traité de Mecanique Celeste III 共Gauthier-Villars, Paris,
“Precession of the perihelion of Mercury’s orbit,” Am. J. Phys. 73, 730– 1894兲, p. 57; reproduced at Ref. 4.
16
734 共2005兲. Clairaut was also a first-class geometer, specializing in curvature. See his
11
Reference 9, p. 196. Recherches sur le courbes a double courbure at Ref. 4.

355 Am. J. Phys., Vol. 75, No. 4, April 2007 Maurizio M. D’Eliseo 355