Jim Lambers

MAT 605
Fall Semester 2015-16
Lecture 7 Notes

These notes correspond to Section 2.5 in the text.

The Two-Body Problem

We revisit the two-body problem
Gm1 m2
m1 r001 = 3 (r2 − r1 ), (1)
r12
Gm1 m2
m2 r002 = 3 (r1 − r2 ), (2)
r12

that describes the motion of two bodies, of mass m1 and m2 located at positions r1 and r2 due to
the gravitational force they exert on each other.

Conservation of Linear Momentum

Adding equations (1) and (2) yields

m1 r001 + m2 r002 = 0, (3)

which, upon integration with respect to t, yields

m1 r01 + m2 r02 = P,

where P is the total linear momentum of the system, given by

P = m1 r01 (0) + m2 r02 (0).

Integrating a second time with respect to t yields

m1 r1 + m2 r2 = Pt + B, (4)

where B is a constant vector.

Uniform Motion of the Center of Mass

Let M = m1 + m2 denote the total mass of the system. From (4), we obtain
m1 m2
r1 + r2 = Vt + C.
M M
The left side of this equation is the center of mass of the system. We see that the center of mass
moves with constant velocity V = P/M and has initial position C = B/M .

1
Jacobi Coordinates
The following change of variables provides a useful alternative perspective on the two-body problem.

Definition 1 (Jacobi Coordinates) Let r1 and r2 be the solutions of the two-body

system (1), (2). The variables r and R, defined by
m1 m2
R = r1 + r2 , (5)
M M
r = r2 − r1 , (6)

are called the Jacobi coordinates of the system.

From (1), (2), (5) and (6), we obtain

R00 = 0, (7)
00
r − = r002 r001
Gm1 Gm2
= − 3 r− 3 r
r12 r12
GM
= − 3 r, (8)
r
where we use the shorthand r = kr2 − r1 k = r12 . A significant advantage of the system (7), (8)
in Jacobi coordinates, compared to (1), (2) in Cartesian coordinates, is that the equations are
uncoupled; that is, (7) depends only on R while (8) depends only on r.
We have seen that (7) has the solution R = Vt + C; what remains is to solve (8). Once we do,
we can use the fact that the Jacobi and Cartesian coordinates are related by the system of linear
equations    m1 m2   
R M M r1
= ,
r −1 1 r2
which can be solved to obtain
1 −m m2
      
r1 1 M
2
R R− M r
= m1 m2 m1 = m1 .
r2 + 1 M r R+ M r
M M

The Central Force Problem

The general central force problem is
f (r)
mr00 = − r,
r
where the function f (r) describes the dependence of the force on the distance r between the two
bodies, the first of which is located at the origin.
For the two-body problem we are considering, if we define the reduced mass
m1 m2
m= ,
m1 + m2
then equation (8) can be rewritten as a central force problem
Gm1 m2
mr00 = − r,
r3
where f (r) = Gm1 m2 /r2 .

2
Conservation of Angular Momentum
Consider the vector
r × mr0 ,
which is normal to the plane spanned by the position vector r and the tangent vector r0 . Differen-
tiating yields
 
d 0

0 0 00 00 f (r)
r × mr = r × mr + r × mr = r × mr = r × m − r=0
dt r

That is, we have

r × mr0 = L,
where L is a constant vector. This is conservation of angular momentum, which means that the
motion of the body remains in the same plane.

Central Force Equations in Polar Coordinates

We now convert the central force problem

kf (r)
r00 = − r,
r
where for convenience we have defined k = 1/m, into polar coordinates x = r cos θ, y = r sin θ. We
begin by writing this system in terms of components,

kf (r)
x00 = − x, (9)
r
kf (r)
y 00 = − y, (10)
r
p
where r = x2 + y 2 . Differentiating x = r cos θ and y = r sin θ with respect to t yields

x0 = r0 cos θ − r sin θθ0 ,

y 0 = r0 sin θ + r cos θθ0 .

Differentiating with respect to t a second time yields

x00 = r00 cos θ − 2r0 sin θθ0 − r cos θ(θ0 )2 − r sin θθ00 ,
y 00 = r00 sin θ + 2r0 cos θθ0 − r sin θ(θ0 )2 + r cos θθ00 .

Substituting these second derivatives into (9), (10) yields

(r00 − r(θ0 )2 ) cos θ − (2r0 θ0 + rθ00 ) sin θ = −kf (r) cos θ,

(r00 − r(θ0 )2 ) sin θ + (2r0 θ0 + rθ00 ) cos θ = −kf (r) sin θ.

Now, if we multiply the first equation above by cos θ and the second equation by sin θ, and add
them, we obtain
r00 − r(θ0 )2 = −kf (r). (11)

3
Kepler’s Second Law
Similarly, if we multiply the first equation by sin θ and the second equation by cos θ, and then
subtract them, we obtain
2r0 θ0 + rθ00 = 0,
which, after multiplying by r, can be simplified to
d 2 0

r θ = 0.
dt
Integrating yields Kepler’s Second Law,
r2 θ0 = c,
where c is a constant. To interpret this law, we integrate both sides of this equation with respect
to t, and obtain
1 t2 2 0
Z
1
r θ (t) dt = c(t2 − t1 ),
2 t1 2
where the left side is the area swept out by the polar curve r = r(t) from θ1 = θ(t1 ) to θ2 = θ(t2 ).
That is, equal areas are swept out in equal times.
To solve (11), we define the new variable ρ by r = 1/ρ(θ). Using f (r) = Gm1 m2 /r2 , Kepler’s
Second Law r2 θ0 = c, and the derivatives
 0
0 1 ρ0 (θ)θ0
r = =− = −cρ0 , r00 = −cρ00 θ0 = −c2 ρ00 ρ2 ,
ρ ρ2

we obtain the equation

ρ00 + ρ = a,
where a = k/c2 . It follows from the method of undetermined coefficients that

ρ(θ) = a + A cos θ + B sin θ,

and therefore
1
r(t) = ,
a + A cos θ(t) + B sin θ(t)
where the constants a, B and C can be determined from the initial conditions. To solve for θ(t),
we note that Kepler’s Second Law, when rewritten as

θ0 (t) = cρ(θ(t))2 ,

is a separable ODE, which can be solved by integration.

Exercises
Section 2.5: Exercises 2, 3