Kepler’s Laws of Planetary Motion and

Newton’s Law of Universal Gravitation

Abstract
These notes were written with those students in mind having taken (or are taking) AP Calculus and AP
Physics. Newton’s law of universal gravitation is introduced in pretty much the same way as it is in AP
Physics, except that the ensuing discussion relies heavily on differential and integral calculus. Through this
treatment the equation of motion is obtained for an orbiting body and the three possible trajectories (elliptical,
parabolic, and hyperbolic) are classified in terms of the relevant constants. At the end is a collection of guided
exercises designed to deepen the reader’s understanding of the interplay between between the mathematics
and the physics, as well as to provide a useful review of analytical geometry.

Equations of Planetary Motion

In this short discussion I would like to

y
show how Newton’s law of univer-
J
s
J J

.
sal gravitation can be applied to de- R =r r
J =(r cos θ, r sin θ)
riving Keplar’s laws of planetary v
motion. Newton’s law of gravitation planet (mass m)
J
says that the force on a planet of mass R
m exerted by another planet (or star) θ x

of mass M is given by the familiar Sun (mass M)

inverse-square law, and has magnitude
GM m
F = ,
r2 Equation 1: (x−√(7))^2/16+y^2/9=1
Equation 2: x^2+y^2=.2

where r is the distance separating the Figure 1: Heliocentric diagram

two masses (measured from their cen-

ters). In the above, G is the universal gravitational constant, whose measured
value is approximately
G = 6.67 × 10−11 N · m2 /kg2 .
As depicted in the above figure, the sun is at the origin (the “heliocentric”
−→
point of view), and R is the position vector of the orbiting planet. The vector
−→ −→ −→ −→
r is the unit vector in the direction  of R . Since r · r = 1, the product
−→ d −→ −→
rule for differentiation shows that r · r = 0; therefore if s is the unit
dt
d −→ −→ −→
vector in the dirction of r , it follows that r · s = 0, as well. All of this is
dt −→ −→
depicted in Figure 1 above. In fact, from this picture, we see that r and s are
given explicitly as
−→ −→
r = (cos θ, sin θ), s = (− sin θ, cos θ),
from which it follows that
d −→ dθ −→ d −→ dθ −→
r= s and that s=− r .
dt dt dt dt
The vector equation dictating the motion of of the orbiting planet is
d2 −→
 
GM m −→
− r = m R (1)
r2 dt2
since the force on the planet is directed back towards the sun.
d2 −→
In order to compute 2 R explicitly, note first that the velocity vector is
dt
given by
 
−→ d −→ d  −→ dr −→ d −→ dr −→ dθ −→
v= R= r· r = r +r r = r +r s (2)
dt dt dt dt dt dt
The acceleration is the time derivative of Equation (2):
 
−→ d −→ d dr −→ dθ −→
a = v= r +r s
dt dt dt dt
 2
d2 r −→ dr dθ −→ d2 θ −→ dθ −→
= 2
r +2 s +r 2 s −r r
dt dt dt dt dt
 2 !
d2 r d2 θ −→
 
dθ −→ dr dθ
= −r r + 2 +r 2 s
dt2 dt dt dt dt

However, from equation (1) we see that the acceleration vector is given by
−→ GM −→
a=− r
r2
from which we conclude that
2
d2 r

dθ GM
−r =− (3)
dt2 dt r2
and
dr dθ d2 θ
2 + r 2 = 0. (4)
dt dt dt

Equations (3) and (4) will be put to good use momentarily.

Keplar’s Second Law

Keplar’s second law says simply this:

the area swept out by the planet’s or-
t2 −t1 =t4 −t3 implies equal areas
bit between times t1 and t2 , versus the
.
t4 .t 3
.t2.t 1
area of the planet’s orbit swept out be-
tween times t3 and t4 are equal pro-
vided that t2 − t1 = t4 − t3 . In other x
words, equal areas are swept out in
equal times. Referring to Figure 2, and
recalling how to compute areas in polar
coordinates, we get Equation 1: (x−√(7))^2/16+y^2/9=1
Equation 2: x^2+y^2=.2

Figure 2: Keplar’s Second Law

Z θ2 Z t2
1 2 1 dθ
Area swept between times t1 and t2 = r dθ = r2 dt.
2 θ1 2 t1 dt
Note, however, that
2
 
d 2 dθ dr dθ 2d θ
r = 2r +r = r · 0 = 0,
dt dt dt dt dt
dθ
where we have used Equation (4), above. The upshot is that the integrand r2
dt
is a constant—call it L—from which we conclude that
1 t2 2 dθ
Z
1
Area swept between times t1 and t2 = r dt = L(t2 − t1 ).
2 t1 dt 2
Therefore, if t4 − t3 = t2 − t1 the areas will be the same! This proves Kepler’s
Second Law.

Keplar’s First Law

Keplar’s first law says that planets follow an eliptical orbit with the Sun at one of
the foci. This is a bit trickier to prove than the second law, but we can proceed
as follows. Recall from the above discussion that the quantity
dθ
L = r2
dt
is a constant; define the new constant
L2
P = ,
GM
and define the (dimensionless) variable
P
u= .
r
This says that
GM L2 u2
=
r2 P3
and that
dθ L Lu2
= 2= 2.
dt r P
We have, using the Chain Rule, that
 
dr d P dθ P dθ du L du
= =− 2 =− .
dt dθ u dt u dt dθ P dθ
Differentiate again and obtain
d2 r L d2 u dθ L2 u2 d2 u
 
d dr dθ
= =− =− 3
dt2 dθ dt dt P dθ2 dt P dθ2

Next, we substitute into Equation (3) and get

   2 2
L2 u2 d2 u P Lu L2 u2
− 3 − = − .
P dθ2 u P2 P3
L2 u2
Dividing by the common factor of − 3 results in the inhomogeneous second-
P
order differential equation:
d2 u
+ u = 1.
dθ2
The general solution of this has the form
u = u(θ) = 1 + e cos(θ − θ0 ),
where e and θ0 are constants which can be determined from the initial conditions.
In terms of the polar radius r, this becomes
P
r= , (5)
1 + e cos(θ − θ0 )
which gives a circle when e = 0, an ellipse when |e| < 0, a parabola when |e| = 1,
and a hyperbola when |e| > 1 (see Exercise 1, below). Note that the coordinate
system can be chosen so that θ0 = 0, (which puts the perihelion1 on the positive
x-axis) giving the equation of the form
P
r= , (6)
1 + e cos θ
1 The perihelion is the point of closest approach of the orbiting planet to the Sun.
from which Keplar’s first law follows.

Determining the Eccentricity

In this section we take a closer look

.
y
at Equation (5) and determine the ec-
centricity e in terms of the initial data. asteroid v(0)
J
We imagine a planet of mass M and (mass m) φ
a nearby massive object—call it an J
asteroid—which has mass m. As we R(0) α x
have seen, the trajectory of the as-
teroid is a conic section; we assume
Planet (mass M)
a coordinate system which places the
planet at a focus. Assume that at time
t = 0 we measure the distance of the Equation 1: r=a/(1+kcos(θ−φ))
Equation 2: x^2+y^2/3=.1
Figure 3: Initial configuration
asteroid to the plant to be r0 and that
the corresponding speed of the asteroid is v0 . Assume that relative to our coordi-
nate system, the initial location of the asteroid corresponds to the angle θ(0) = α
and that the angle between the asteroid’s velocity vector and the position vector
is φ (as indicated) in Figure 3.

If we read Equation (5) at time t = 0, we get

P
r0 = . (7)
1 + e cos(α − θ0 )
It will be convenient to introduce the new constant p = P/r0 , and so Equation
(7) implies that
p
1= ,
1 + e cos(α − θ0 )
which gives
1 + e cos(α − θ0 ) = p. (8)
Next, we differentiate Equation (5) with respect to t; noting the Chain Rule
and setting t = 0, this leads to
dr P e sin(α − θ0 ) dθ ev0 sin φ sin(α − θ0 )
v0 cos φ = = · = .
dt t=0 [1 + e cos(α − θ0 )]2 dt t=0 p
Therefore,
e sin(α − θ0 )
cot φ = ,
p
and so
e sin(α − θ0 ) = p cot φ. (9)
Applying Equations (8) and (9) leads immediately to
e2 = e2 cos2 (α − θ0 ) + e2 sin2 (α − θ0 ) = (p − 1)2 + p2 cot2 φ
which simplifies quickly to
e2 = p2 csc2 φ − 2p + 1. (10)

From the above, we get a useful trichotomy for the eccentricity e:

Ellipse (e2 < 1): Equation (10) reduces in this case to
p2 csc2 φ − 2p + 1 < 1 ⇐⇒
p csc2 φ < 2.
However, recall that
L2 dθ
P = , where L = r02 = r0 v0 sin φ,
GM dt t=0
and so
r0 v02
p csc2 φ = .
GM
That is to say, an elliptical orbit will occur precisely when
r0 v02 < 2GM. (11)
Parabola (e2 = 1): Similarly with the above, this will occur when
r0 v02 = 2GM. (12)
Hyperbola (e2 > 1): This is when
r0 v02 > 2GM. (13)

Exercise 1.
(a) Using the equations x = r cos θ, y = r sin θ, show that Equation (6) can
be expressed in cartesian coordinates as
2
y2 P2

Pe
x+ + = , if e 6= 1,
1 − e2 1 − e2 (1 − e2 )2
and as
y 2 + 2P x = P 2 if e = 1.
From this, conclude that one obtains an ellipse if e2 < 1, a parabola if e2 = 1,
and a hyperbola if e2 > 1.
 (b) Given that e2 < 1, show that the semi-major axis√of the ellipse has length
P/(1 − e2 ) and the semi-minor axis has length P/ 1 − e2 .

Exercise 2: Calculations of Eccentricities of Planetary Orbits about

the Sun

The Earth’s orbit

We itemize some estimates of the important constants:
• Distance at perihelion: r0 ≈ 1.471 × 108 km.
• Linear velocity at perihelion2 : v0 ≈ 30.29 km/sec.
• Universal gravitational constant: G ≈ 6.67 × 10−11 N · m2 /kg2
• Mass of Sun: ≈ 1.99 × 1030 kg.

Using Equation (10), compute the eccentricity of the Earth’s orbit about the
Sun.3

Mars’s orbit
The important constants are given below:
• Distance at perihelion: r0 ≈ 2.066 × 108 km.
• Linear velocity at perihelion: v0 ≈ 26.50 km/sec.

Compute the eccentricity of Mars’s orbit about the Sun.4

Mercury’s orbit about Sun

The important constants are given below:
• Distance at perihelion: r0 ≈ 4.600 × 107 km.
• Linear velocity at perihelion: v0 ≈ 58.98 km/sec.

Compute the eccentricity of Mars’s orbit about the Sun.5

2 This is the point at which the Earth is closest to the Sun.

3 You should get e ≈ 0.0167, showing that the Earth’s orbit is very close to a perfect circle.
4 Here, e ≈ 0.0935, showing that the Mars’s orbit is more elliptical that that of the Earth.
5 Here, e ≈ 0.2056, making Mercury’s orbit the most elliptical among the planets. (Actually Pluto’s orbit is more elliptical,

with e ≈ 0.248, but in 2006 Pluto was downgraded from a “planet” to a “dwarf planet.”)
Exercise 3: Minimum and Maximum Orbital Speeds

In this guided exercise you’ll show that for the orbital model given by Equation
(5), the body’s maximum speed occurs precisely at its perigee (point of closest
approach to the central mass), and when the orbit is elliptical, its minimum speed
occurs at its apogee (point of maximum distance from the central mass).
Step 1. Using Equation (2), show that the speed is given by
 2  2
dr dθ
v2 = + r2 .
dt dt
Step 2. Use Equation (5) to write
P
r= .
1 + e cos θ
dθ
Also, you know that r2 is a constant, which we earlier denoted by L:
dt
dθ L
= 2.
dt r
Combine this to get
L2
 
v2 = [e2 sin2 θ + (1 + e cos θ)2 ].
P2
Step 3. Show that
dv sin θ
= constant × 2 .
2v
dt r
Step 4. Finish the proof that the maximum speed occurs at the perigee and the
minimum speed occurs at the apogee (if any).6

Exercise 4: The Moon’s Orbit about the Earth

You are given that the Moon’s distance from Earth at perigee is about 364,397
km and its distance to the Earth at apogee is about 406,731 km. The mass of
the Earth is ≈ 5.98 × 1024 kg. Given this information, compute the eccentricity
of the moon’s orbit and its minimum and maximum linear speeds.7
6 Note that L2 /P 2 = GM/P ; therefore the velocity at perigee is
r
GM
vmax = (1 + e) ,
P
and the velocity at apogee (if any) is r
GM
vmin = (1 − e) .
P
7 Simple algebra leads to
rmax − rmin
e= ≈ 0.055.
rmax + rmin
 .
Exercise 5. On the earth an as- y
tronomer spots a very slow-moving as- J
v(0)
teroid at time t = 0 moving at a speed asteroid
φ
of roughly v0 ≈ 42 m/sec and whose
J
initial distance from the earth is ap- R(0) θ x
proximately r0 ≈ 7.5 × 108 km. As-
sume at the time of observation the an-
gle φ between the velocity vector and Earth
−→
− R (as shown) is approximately 15◦ ,
and recall that the mass of the earth Equation 1: r=a/(1+kcos(θ−φ))
Equation 2: x^2+y^2/3=.1
Figure 4: Asteroid initial conditions
is M ≈ 5.98 × 1024 kg. The value of
the universal gravitational constant is
roughly
G = 6.67 × 10−11 N · m2 /kg2 ,
and the initial angle relative to the coordinate system indicated in Figure 4 is
θ(0) ≈ 120◦ .
(i) Will this asteroid be captured into an elliptical orbit about the Earth?8
(ii) Compute the eccentricity of the asteroid’s orbit relative to Earth.9
(iii) Relative to Figure 4 and the model
P
r= ,
1 + e cos(θ − θ0 )
determine the angular shift θ0 and plot the trajectory of the asteroid.10
(iv) Compute the speed of the asteroid at its perigee. (Just use the result for
v 2 in Step 2 of Exercise 2, above.)11

Exercise 6. Prove Kepler’s Third Law, namely, that the square of the orbital
period (for an elliptical orbit) is proportional to the cube of the semi-major
To compute the corresponding velocities, just use the equation
P
r= ;
1 + e cos θ
2 v2
the perigee occurs at θ = 0, and P = rmin max /GM . Put this together and get vmax = 1.074 km/sec. Likewise, get vmin = 0.963
km/sec.

8 No, because
r0 v02 ≈ 1.32 × 1015 > 8 × 1014 = 2GM.

9 Using Equation (10), get e ≈ 1.14, which dictates a hyperbolic orbit.

10 From Equation (8), get θ(0) − θ0 ≈ 133◦ , and so θ0 ≈ −13◦ .
11 The constant L is given by L = r v sin φ, which implies that L2 = P GM . Therefore, the velocity v
0 0 max at perigee is
determined through  
2 GM
vmax = (1 + e)2 =⇒ vmax ≈ 105 m/sec
P
axis.
To do this, just carry out the indicated steps.
Step 1. Denoting by T the orbital period, note that the area of the ellipse swept
out by the orbiting body is
1 t=T 2 dθ
Z
A= r dt.
2 t=0 dt
dθ
However, we’ve seen that r2 is a constant, denoted L. Therefore, the
dt
above area is given by A = 12 LT.
Step 2. Denoting by a the semi-major axis and by b the semi-minor axis, the
area of the corresponding ellipse can be expressed by the integral
4b a p 2
Z
A= a − x2 dx = abπ.
a 0
Step 3. Recall from Analytical Geometry that an√ellipse with semi-major axis a
and eccentricity e has semi-minor axis b = a 1 − e2 . (See also Exercise 1,
above.) Conclude from Steps 1 and 2 that
√
2A 2a2 1 − e2 π
T = = .
L L
Step 4. Using the equation
P L2
r= , P = ,
1 + e cos θ GM
together with
P P
2a = rmin + rmax = +
1+e 1−e
leads immediately to
L2 = a(1 − e2 )GM.
T2 4π 2
Step 5. Conclude by showing that = .
a3 GM

Exercise 7: A Paradox?

As we Suppose that we return to the Earth-Moon system of Exercise 4, where the

Moon’s orbit about the Earth was determined to be elliptical, having eccentricity
e ≈ 0.055. Furthermore, a moment’s thought reveals that relative to observers
on the Moon, the Earth can be said to orbit about the Moon in an
elliptical orbit with the same eccentricity and the same velocities at perigee and
apogee. However, suppose that we adopt a coordinate system which places the
Moon at the origin and the positive x-axis in the direction of the Earth’s perigee
relative to the Moon. Then the equation of motion predicting the Earth’s motion
relative to the Moon should be
P
r= ,
1 + e cos θ
2 2
rmin vmax
where P = , and where m is the mass of the Moon and is
Gm
m ≈ 7.36 × 1022 kg.
However, using rmin ≈ 3.64 × 105 km and vmax ≈ 1.05 km/sec, the above
equation returns an eccentricity of
P
e= − 1 ≈ 80.8,
rmin
which isn’t the eccentricity of an ellipse at all! What gives?