The Three

Body Problem

Joakim Hirvonen Grützelius

Karlstad University December 26, 2004
Department of Engineeringsciences, Physics and Mathematics
Analytical Mechanics 5p
Examinator: Prof Jürgen Füchs
 Abstract
The main topic of this project is to give a mathematical description of the
three body problem. A direct application to this problem is a rotating two-
body system such as Sun-Jupiter, this rotating system is going to be treated
in detail.
At the end of the project there is a short discussion about the Ascending
nodes and the Lagrangian points.
The Three Body Problem 1

Contents
1 Historical Background 2
1.1 Origin of The Three Body Problem . . . . . . . . . . . . 2
1.2 Introduction to The Three Body Problem . . . . . . . . 2

2 Mathematical Description of The Three Body Problem 3

2.1 The differential equations of the problem . . . . . . . . . 3
2.2 Reduction to the 6th Order . . . . . . . . . . . . . . . . 4

3 The Restricted Three Body Problem 6

A Ascending Node 8

B Lagrangian Points 9

J H Grützelius Analytical Mechanics

The Three Body Problem 2

1 Historical Background
1.1 Origin of The Three Body Problem
Since bodies in the solar system are approximately spherical and their
dimensions extremely small when compared with the distances between
them, they can be considered as point masses. Hence the origin of the
problem can be thought of as being synonymous with the foundation of
modern dynamical astronomy. This part of celestial mechanics, which
connects the mechanical and physical causes with the observed phe-
nomena, began with the introduction of Newton’s theory of gravitation.
From the time of the publication of the Principia in 1687, it became
important to verify whether Newton’s law alone was capable of render-
ing a complete understanding of how celestial bodies move in space. In
order to pursue this line of investigation, it was necessary to ascertain
the relative motion of n bodies attracting one another according to the
Newtonian law.
Newton himself had geometrically solved the problem of the two
bodies for two spheres moving under their mutual gravitational attrac-
tion, and in 1710 Johann Bernoulli had proved that the motion of one
particle with respect to the other is described by a conic section. In
1734 Daniel Bernoulli won a French academy prize for his analytical
treatment of the two body problem, and the problem was solved in
detail by Euler 1744. Meanwhile work was already in progress on the
higher dimensional problem. Driven by the needs of navigation for
knowledge about the motion of the moon, researchers scrutinized the
system formed by the sun, the earth and the moon, and the lunar theory
quickly dominated the early research into the problem.

1.2 Introduction to The Three Body Problem

The three body problem, which was described by Whittaker as ”the
most celebrated of all dynamical problems” [1] and which fulfilled for
Hilbert the necessary criteria for a good mathematical problem, can
be simply stated: three particles move in space under their mutual
gravitational attraction; given their initial conditions, determine their
subsequent motion. Like many mathematical problems, the simplicity
of its statement belies the complexity of its solution. For although the
one and two body problems can be solved in closed form by means of
elementary functions, the three body problem is a complicated linear
problem, and no similar type of solution exists.
Apart from its intrinsic appeal as a simple-to-state problem, the
three body problem has a further attribute which has contributed to
its attraction for potential solvers: its intimate link with the funda-
mental question of the stability of the solar system. Over the years
attempts to find a solution spawned a wealth of research, and between
1750 and the beginning of the twentieth century more than 800 papers

J H Grützelius Analytical Mechanics

The Three Body Problem 3

relating to the problem were published, invoking a roll call of many

distinguished mathematicians and astronomers. And hence, as is often
the case with such problems, its importance is now perceived as much in
the mathematical advances generated by attempts at its solution as in
the actual problem itself. These advances have come in many different
fields, including, in recent times, the theory of dynamical problems.
To clarify the mathematical difficulties associated with the problem
we will begin with a mathematical description.

2 Mathematical Description of The Three Body

Problem
2.1 The differential equations of the problem
Let us suppose that the three bodies under consideration to be at the
points Pi , with masses mi and coordinates qij in an inertial reference
frame (i, j = 1, 2, 3). The distance between them are large, so we can
think of them as point particles. We denote the distance between them
as Pij = rij , where rij = |qi − qj |. Due to Newton’s law of gravitation,
2
the force of attraction between the ith and jth becomes Gmi mj /rij , and
the corresponding term in the potential energy becomes −Gmi mj /rij .
Then the potential energy of the hole system is
 
m1 m2 m1 m3 m2 m3
V = −G + + (1)
r12 r13 r23
where G is the gravitational constant. With the help of Newtons equa-
tion of motion and that F = − ∂V ∂r
, then choosing units so that G is
equal to one, the equations of motion become
d2 q1i ∂V
m1 = − (2)
dt2 ∂q1i
d2 q2i ∂V
m2 2
=− (3)
dt ∂q2i
d2 q3i ∂V
m3 2
=− (4)
dt ∂q3i
or
d2 q1i (q2i − q1i ) (q3i − q1i )
2
= m2 3
+ m3 3
(5)
dt r12 r13
d2 q2i (q1i − q2i ) (q3i − q2i )
2
= m1 3
+ m3 3
(6)
dt r21 r23
d2 q3i (q1i − q3i ) (q2i − q3i )
2
= m1 3
+ m2 3
(7)
dt r31 r32

J H Grützelius Analytical Mechanics

The Three Body Problem 4

where i = 1, 2, 3. The problem is therefore described by nine second-

order differential equations or by 18 equations of the first order
dq1i dq2i dq3i
= q̇1i , = q̇2i , = q̇3i
dt dt dt
dq̇1i ∂V dq̇2i ∂V dq̇3i ∂V
m1 =− , m2 =− , m3 =− (8)
dt ∂q1i dt ∂q2i dt ∂q3i
So for a closed solution to the problem, the system needs 18 independent
integrals. However, it is only possible to find 12 such integrals, and the
system can therefore only be reduced to one of order six. As will be
shown below, this is achived through the use of the so-called ten classic
integrals, the six integrals of the motion of the centre of mass, the three
integrals of angular momentum, and the energy integral, together with
the elimination of the time and the elimination of what is called the
ascending node. Illustration of the ascending node and other orbital
parameters can be found in appendix A.

2.2 Reduction to the 6th Order

When multiplying equation (5), (6) and (7) by mi a summation can be
performed to give three equations
3
X d2 qij
mi 2 = 0 , (j = 1, 2, 3), (9)
i=1
dt

if we integrate these equation twice we get the equations

3
X
mi qij = Aj t + Bj , (j = 1, 2, 3), (10)
i=1

in which the Aj and Bj are constants of integration. These equations

show that the centre of mass of the three particles either remains at
rest or moves uniformly in space in a straight line. This is expected
since there are no forces acting except the mutual attractions of the
particles. The six constants serve to describe the motion of the centre
of mass in the original arbitrary inertial coordinate system and play no
part in the motion of the bodies about the centre of mass.
If the first equation of (5) multiplied by −q12 , the first equation of
(6) by −q22 and the first equation of (7) by −q32 , and in equation (5)
the second equation is multiplied by q11 , the second equation of (6) by
q21 , and the second equation of (7) by q31 , and these two sets are added
together, this will give us
3 3
X d2 qi2 X d2 qi1
mi qi1 − m q
i i2 =0, (11)
i=1
dt2 i=1
dt 2

J H Grützelius Analytical Mechanics

The Three Body Problem 5

and two similar equations can be obtained by a cyclic change of the

variables (x, y, z). The three equations can then be integrated to give
3  
X dqi3 dqi2
mi qi2 − qi3 = C1 (12)
i=1
dt dt

3  
X dqi1 dqi3
mi qi3 − qi1 = C2 (13)
i=1
dt dt
3  
X dqi2 dqi1
mi qi1 − qi2 = C3 . (14)
i=1
dt dt
These equations represent the conservation of angular momentum for
the system. That is, they show that the angular momentum of the three
particles around each of the coordinate axes is constant throughout the
motion.
Equation (5), (6) and (7) can be written in the form
d2 qij ∂V
mi 2 = − . (15)
dt ∂qij
dq
Multiplying by dtij and summing gives, since V is a function of the
coordinates only,
3
X d2 qij dV
pij 2 = − . (16)
i,j=1
dt dt
This equation can then be integrated to give
3
X p2ij
= −V + C , (17)
i,j=1
2m i

where C is a constant of integration. Furthermore, since the left-hand

side of the equation represents the kinetic energy T of the system,
the integral can be put in the form T + V = C, which expresses the
conservation of energy.
Two final reductions can then be made to the order of the system.
First, the time can be eliminated by using one of the dependent vari-
ables as an independent variable which is used in the section treating
the restricted three body problem, and, second, a reduction can be
made by the so called elimination of the nodes.
Thus through use of the classical integrals and these last two inte-
grals, the original system of order 18 can be reduced to a system of
order six. Furthermore, this result can be generalised to the n body
problem. In this case the differential equations constitute a system of
order 6n. By using the same integrals this system can be reduced to a
system of order (6n − 12).

J H Grützelius Analytical Mechanics

The Three Body Problem 6

3 The Restricted Three Body Problem

If we consider a body of unit mass (e.g., an asteroid) moving in the
field of a heavy body of mass M (e.g., the Sun) and a much lighter
body (e.g., Jupiter) of mass m. Also assume that the heavy bodies are
in circular orbit around each other with angular frequency Ω; the effect
of the light body on them is negligible. All three bodies move in the
same plane. This is the restricted three body problem.
When ignoring the effect of the asteroid, we can solve the two body
problem to get
Mm GM m G(M + m)
RΩ2 = , ⇒ Ω 2
= . (18)
M +m R2 R3
When we choose the center of mass of the heavy bodies as the ori-
gin of a polar coordinate system. Then the position of the Sun is at
(νR, π − Ωt) and Jupiter is at ((1 − ν) R, Ωt), where R is the distance
between them, and ν = Mm +m
. The distance from the asteroid to the
Sun is p
ρ1 (t) = r2 + ν 2 R2 + 2νrR cos (θ − Ωt) (19)
and to Jupiter is
q
ρ2 (t) = r2 + (1 − ν)2 R2 + 2 (1 − ν) rR cos (θ − Ωt). (20)

The Lagrangian for the motion of the asteroid is

 
1 2 1 2 2 1−ν ν
L = ṙ + r θ̇ + G (M + m) + . (21)
2 2 ρ1 (t) ρ2 (t)
In this coordinate system the Lagrangian has an explicit time depen-
dence: the Hamiltonian is not conserved. We change variables to
χ = θ − Ωt to get
 
1 2 1 2 2 1−ν ν
L = ṙ + r (χ̇ + Ω) + G (M + m) + (22)
2 2 r1 r2
where p
r1 = r2 + ν 2 R2 + 2νrR cos χ (23)
and q
r2 = r2 + (1 − ν)2 R2 − 2 (1 − ν) rR cos χ (24)
are now independent of time.
Now the Hamiltonian in the rotating frame,
r2
 
∂L ∂L 1 1 1−ν ν
H = ṙ + χ̇ −L = ṙ2 + r2 χ̇2 −G (M + m) + +
∂ ṙ ∂ χ̇ 2 2 2R3 r1 r2
(25)
is a constant of the motion. This is called the Jacobi integral in classical
literature.

J H Grützelius Analytical Mechanics

The Three Body Problem 7

The Hamiltonian is of the form H = T + V where T is the kinetic

energy and V is an effective potential energy:
 2 
r 1−ν ν
V (r, χ) = −G (M + m) + + . (26)
2R3 r1 r2
It consist of the gravitational potential energy plus a term due to the
centrifugal barrier, since we are in a rotating coordinate system.
The effective potential V (r, χ) is conveniently expressed in terms of
the distances to the massive bodies,
  2   2 
r1 1 r2 1
V (r1 , r2 ) = −G M + +m + (27)
2R3 r1 2R3 r2
using the identity
1 2 1 2 1
r1 + r2 = r 2 + R2 . (28)
ν 1−ν ν(1 − ν) 2
(We have removed an irrelevant constant from the potential).
Sometimes it is convenient to use cartesian coordinates, in which the
lagrangian and hamiltonian are
1 1
L = ẋ2 + ẏ 2 + Ω (xẏ − y ẋ) − V (x, y). (29)
2 2
1 1
H = ẋ2 + ẏ 2 + V (x, y). (30)
2 2
It is obvious from the above formula for the potential as a function of
r1 and r2 that r1 = r2 = R is an extremum of the potential. There
are two ways this can happen: the asteroid can form an equilateral
triangle with the Sun and Jupiter on either side of the line joining
them. These are the Lagrange points L4 and L5 . These are actually
maxima of the potential. In spite of this fact, they correspond to stable
equilibrium points because of the effect of the velocity dependent forces.
A discussion of the Lagrangian points can be found in appendix B. [2].

J H Grützelius Analytical Mechanics

The Three Body Problem 8

A Ascending Node
The ascending node is one of the orbital nodes, a point in the orbit
of an object where it crosses the plane of the ecliptic from the south
celestial hemisphere to the north celestial hemisphere in the direction of
motion. Because of this, the ascending node of the orbit of the Earth’s
moon is one of only two places where a lunar or solar eclipse can occur.
The line of nodes is the intersection of the object’s orbital plane with
the ecliptic, and runs between the ascending and descending nodes.

Figure 1: Illustration of orbital parameters.

J H Grützelius Analytical Mechanics

The Three Body Problem 9

B Lagrangian Points
A location in space around a rotating two-body system (such as the
Earth-Moon or Sun-Jupiter) where the pulls of the gravitating bodies
combine to form a point at which a third body of negligible mass would
be stationary relative to the two bodies. There are five Lagrangian
points in all, which can be seen in figure 1 below, three of which are

Figure 2: Illustration of the Lagrangian points.

unstable because the slightest disturbance to any object located at one

of them causes the object to drift away permanently. Until recently, this
meant that the unstable Lagrangian points seemed to have no practical
application for spaceflight. Now, however, they are known to have
immense significance and have become the basis for chaotic control. In
addition, growing numbers of spacecraft are being placed in halo orbits
around the L1 and L2 points; station-keeping, in the form of regular
thruster firings, are needed to maintain these orbits (which are around
empty points in space!). The NASA Sun-observing probes SOHO and
ACE currently orbit around L1 , while future spacecraft to be placed
in L2 halo orbits include the Next Generation Space Telescope and the
European Space Agency’s Herschel, GAIA, and Darwin spacecraft. In
many ways these points are ideal for observing both near and far reaches
of space since spacecraft can orbit around them far from disturbing
influences, such as that of Earth’s magnetosphere. L1 is well-suited to
solar observations; L2 offers uninterrupted observations of deep space,
since the spacecraft can be oriented so that the Earth, Moon and Sun
remain ”behind” it at all times, and enables the entire celestial sphere
to be observed over the course of one year. L3 hasn’t been utilized for
spaceflight because it lies on the opposide side of the Sun from Earth.
The remaining two Lagrangian points, L4 and L5 , lie at the vertices of
equilateral triangles formed with the two main gravitating masses and
in their orbital plane. They are also referred to as libration points since
if any objects located at them are disturbed, the objects simply wobble
back and forth, or librate.

J H Grützelius Analytical Mechanics

The Three Body Problem 10

References
[1] A treatise on the analytical dynamics of particles and rigid bodies,
4th edition, E.T Whittaker, Cambridge University Press (1937).
[2] Methods of Celestial Mechanics, D.Brouwer and G.M. Clemence,
Academic Press, NY (1961)
[3] Poincaré and the three body problem, June Barrow-Green, Amer-
ican Mathematical Society, Printed in USA (1997).
[4] http://www.daviddarling.info/encyclopedia/L/Lagpoint.html
[5] Dynamical systems, G.D. Birkhoff, American mathematical society
1927.

J H Grützelius Analytical Mechanics