CHAPTER

ANGULAR
MOMENTUM
Whereupon I computed what would be the Orb described by the
Planets.... I found now that whatsoever was the law of the forces which kept
the Planets in their Orbs, the areas described by a Radius drawn from them to
the Sun would be proportional to the time in which they were described.
Isaac Newton

13.1 ROTARY MOTION

The world is full of things that exhibit rotary motion. They range in size from galaxies
to electrons orbiting around atoms, and they include such familiar objects as orbiting
planets, amusement park rides, flywheels, and bathtub vortices. What underlying principle
explains the persistence of such motions? Can we describe them all in a unified way?

335

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 336 ANGULAR MOMENTUM

In Chapter 7 we studied two partial answers to such questions. One was that the
moon circles the earth (or a planet circles the sun) by falling toward it continuously.
Another was that an object can undergo uniform circular motion if a centripetal accel-
eration of magnitude v2lr is supplied. But the first answer is restricted to gravitational
attraction, the second is restricted to the special case of uniform circular motion, and
each addresses the persistence of such motion rather obliquely.
The concept that unifies the description of all rotary motion is angular momentum
and the persistence of such motions is most directly described in terms of the conservation
of angular momentum. To be sure, angular momentum conservation, when applicable,
can be derived from Newton's three laws. Any rotational problem can be solved directly
by following Newton's three laws through time without introducing angular momentum.
And yet, like energy conservation (to which the same comments applied), conservation
of angular momentum is a most useful concept which both aids understanding and sim-
plifies calculation.

13.2 TORQUE AND ANGULAR MOMENTUM

In Section 6.6 we gave the general definition of torque,
T = r x F. (13.1)
As we see from this relation, torque is an unusual vector insofar as it depends on the
distance r from a chosen reference point O to the point of application of the force F (Fig.
13.1). The torque tends to produce a rotation about point O, with the axis of rotation
parallel to the vector T.

Figure 13.1 Vector relations for torque about a point O.

We now wish to bring out this relation between torque and rotation, with the aid of
Newton's second law

F = | . (6.1)

Substituting dpldt for F in Eq. (13.1), we obtain

dp
T = r x-f .
dt

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.2 TORQUE AND ANGULAR MOMENTUM 337

Because
d dx dp
Jt(r p) = - x p + r x -
by the product rule as applied to vectors (Sec. 5.6), the relation for T can be rewritten
d dr
, ,
T = ( r X p ) X p
^ - ^ -
The last term vanishes:
dx
— x p = v x (mv) = 0,
dt
because the cross product of a vector with itself is zero, and we obtain the result

T = j ( (r x p). (13.2)

Equation (13.2) bears a striking resemblance to Newton's second law,

(6.1)

On the left-hand side of (13.2) we have the torque, a "twisted" version of force. And
on the right-hand side we have the time derivative of r x p, a "twisted" version of
momentum. This important quantity r x p i s denoted by the symbol L and, in view of
the analogy with momentum, is called angular momentum:

L = r x p. (13.3)

According to Eq> (13,2) the time rate of change of angular momentum equals the torque:

(13.4)

Like its linear analog F = dp/dt, Eq. (13.4) is a very important result. It has a number
of significant consequences which we will explore in this chapter and the next.
When a system consists of two or more parts, the individual bodies composing it
will generally be subject to both internal and external torques. For the system as a whole,
however, the internal torques cancel, just as internal forces do, provided the internal
forces between any pair of bodies are equal and opposite and act along the line joining
the two bodies, as do gravitational and Coulomb forces. To see this, consider the pair
of bodies in Fig. 13.2. The torque r, x F 21 on body 1 about point O, due to the force
F2i from body 2, is the same as r l x x F 21 in Fig. 13.2 because the component of rx
parallel to F 21 does not contribute to the torque:

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 338 ANGULAR MOMENTUM

Figure 13.2 Torques exerted by a pair of bodies on each other.

T2i = r, x F 21 = r 1 ± x F 2 1 . (13.5)
Similarly the torque r 2 x F 12 on body 2 about point O, due to the force F 12 from body
1, is the same as r 2 ± x F 1 2 :
T 12 = r 2 x F 1 2 = r2JL x F 1 2 . (13.6)
But F 21 = - F 1 2 by Newton's third law, and one easily sees from Fig. 13.2 that
r
i± = r2± if F12 a n d F2i have the same line of action. So the net internal torque of the
pair vanishes,

*2i + Ti2 = rIJL x (F 21 + F 12 ) = 0,

about any point O. Proceeding as for internal forces, one can easily extend this result
and show that the sum over all internal torques vanishes. Therefore only Text contributes
to dL/dt in Eq. (13.4) for a system of particles:

dh
Text = — • (13.7)
dt
When an external torque is present, integration of Eq. (13.7) over time yields
'2

= L(r2) - L(r,). (13.8)

By analogy with the relation between impulse and momentum, the quantity on the left-
hand side is sometimes called angular impulse, and the equation reads: ' T h e angular
impulse equals the change in angular momentum."

13.3 ANGULAR MOMENTUM CONSERVATION

As with its close relative, momentum, there are many circumstances in which the angular
momentum of a system is conserved. First of all, the external torques applied to a body
may vanish or cancel. In this case

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.3 ANGULAR MOMENTUM CONSERVATION 339

dh „
(13.9)

SO

L = const. (13.10)
Because L is a vector, Eq. (13.10) is really three separate conservation laws for the
components Lx, Ly, and Lz.
Second, even if external forces and torques are present, they may have a negligible
effect during the brief duration of a collision or explosion. In this case the overall
momentum and angular momentum of a system may be conserved in the collision or
explosion:

P, = const,
(13.11)
L, = const,

even though strong internal impulses change the p, and L, of individual particles. For
example, friction supplies both an external force and a torque to a rolling billiard ball,
eventually bringing it to rest. But friction can be neglected in the collision of two billiard
balls - both momentum and angular momentum are conserved.
Third, if something exerts an external torque on a system, then the system exerts an
external but opposite torque on that something, as we saw in the previous section. If we
include that external thing as part of a new, larger system, and that external thing is not
subject to other torques, then the external torque on the whole system vanishes, so its
angular momentum will be conserved. For example, in a classic lecture demonstration,
a student stands on a turntable holding a bicycle wheel in the horizontal plane. When
the student applies a torque to the bicycle wheel, it applies a countertorque to him. Neither
the wheel's L nor the student's L is individually conserved, but the combined angular
momentum of wheel, student, and turntable is conserved. So if the student spins the
wheel clockwise (as viewed from above), starting from rest, the student must rotate
counterclockwise to conserve the total angular momentum of the combined system (Fig.
13.3b). If the student now turns the wheel 180° over so it rotates counterclockwise, his
rotation becomes clockwise to conserve total angular momentum (Fig. 13.3c). In this last

Figure 13.3 Angular momentum conservation for a student holding a

bicycle wheel on a turntable.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 340 ANGULAR MOMENTUM

case it is somewhat complicated to follow the torque in detail, but angular momentum
conservation gives the final direction of rotation easily.
By continuing forever this process of enlarging the system we may include all objects
within the universe in our system. In this case there is nothing left outside to exert a
torque on it. So the total angular momentum of the universe is conserved.
Thus angular momentum joins energy and momentum as the third great conserved
quantity of classical mechanics. These conservation laws all had their origin in Newton's
laws. One can always solve a problem directly from Newton's laws, but use of the
conservation principles leads to great simplifications and insight. We shall find that angular
momentum conservation is just as helpful in simplifying rotational problems as momentum
conservation was for problems of linear motion.

Example 1
A particle of mass m moves with constant velocity v along a straight line which is a
distance b from the origin of a coordinate system.
(a) Find the angular momentum of the particle at any instant.
(b) Show explicitly that the angular momentum is conserved.

(a) We know that the angular momentum is given by

L = mv x v. (13.3)

The magnitude of the angular momentum is L = mrLv, where rL is the component of

the position vector r perpendicular to the velocity v. From the diagram we see that this
distance is b. Therefore
L = mr±v = mbv.

By the right-hand rule, the direction of the angular momentum is into the plane of the
page.
(b) From our expression for L, we see that as the particle continues to move along
the straight line, the component of r perpendicular to v remains equal to b. Thus the
angular momentum is conserved. The reason for the conservation of angular momentum
here is that no force acts on the particle (by Newton's first law), and therefore it feels
no torque.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.3 ANGULAR MOMENTUM CONSERVATION 341

In Example 1 we could have chosen a different origin for the coordinate system,
located at distance b' from the line of motion of the particle. Angular momentum would
still be conserved in this case, but its value would be
V = mb'v.
The conservation law always holds, but the choice of reference point is arbitrary. This
situation reminds us of the similar arbitrariness in the choice of reference point about
which torque is taken. In problems in which the general relation T = dh/dt is used one
can choose any reference point, but both T and L must be taken around the same point.
Example 1 also shows that a body need not be rotating to have angular momentum.
Angular momentum is a very general property.
In the foregoing example energy, momentum, and angular momentum were all
conserved. But it happened they were not needed to solve for the motion, which followed
trivially from Newton's first law. In the next example the angular momentum of the
system is again conserved (along with linear momentum) and it aids greatly in finding
an easy solution for the motion.

Example 2
Two masses, 8.0 kg each, are connected by an extensible rod of negligible mass. Originally
the rod is rotating about its center at an angular speed of 2 rad/s with the masses 0.4 m
from the axis of rotation. What will be the rod's angular speed if the masses are pulled
in to a distance of 0.3 m from the center?

This example is a prototype for rotating ice skaters, ballerinas, and students holding
dumbbells on turntables, all of whom gain angular speed by pulling in their initially
outstretched arms. In each case the friction of the ice on skates, the floor on dance
slippers, the ball bearings on the turntable, or the pivot on the turning rod can be neglected
on the time scale it takes to pull the arms in, so angular momentum is effectively conserved.
We shall learn how to treat the distributed mass of a rotating person quantitatively in the
next chapter. In the present example we analyze the system of the (massless) rod con-
necting (point) masses, which exhibits the same qualitative behavior while being simple
enough to analyze completely with our present methods.
Each mass moves in a circle of radius rx about the axis of rotation, with constant
speed v. At any instant, the velocity v of the mass is perpendicular to the radius vector
r,. Therefore the magnitude of its angular momentum is

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 342 ANGULAR MOMENTUM

L = mvrx sin 90° = mvrx.

The direction of the angular momentum is perpendicular to the plane of the circle, as
shown in the figure. Since the rod has been assumed massless, its angular momentum
can be neglected.
The total angular momentum is the sum of the individual angular momenta. The
angular momentum of both masses is in the same direction, so the magnitude of the total
angular momentum is
Lx = mvrx + mvrx = 2mvr]
where rx is the distance from the axis and v is the speed of each mass. We can write this
in terms of angular speed by recalling that v = rxu>x:
Lx = 2mrx2o)x.
When the masses are pulled in to a new distance r 2 , the angular momentum will be
L2 = 2mr2(o2.
The force applied to pull in the masses acts radially inward along the rod, so F is parallel
to r, where r is the position vector of the mass relative to the center of the rod. Thus
the torque T = r X F acting on the mass about the center of the rod vanishes and angular
momentum is conserved.
Conservation of angular momentum implies that Lx = L 2 . Equating the two angular
momenta and solving for oo2, we get
co2 = oo^rjVrf).
Substituting o)x = 2 rad/s, rx = 0.4 m, and r2 = 0.3 m, we find <o2 = 3.6 rad/s.
To pull the masses in, a centripetal force had to be supplied. It did work which went
into rotational energy. We shall discuss this aspect of the problem in Chapter 14.

13.4 FORCE AND TORQUE

In equilibrium, as discussed in Sections 6.5 and 6.6, both the net force and the net torque
on the system must vanish:
2 F, = 0, (13.12a)
T
2 / = 0- (13.12b)

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.4 FORCE AND TORQUE 343

yielding a total of six scalar equations. For simplicity, in this book we usually consider
only forces in a plane and torques about an axis normal to this plane. When both F ; and
r, are confined to the xy plane, the torque lies in the z direction. So we make frequent
use only of the three conditions

2 Fix = 0, (13.13a)
X Ffy = 0, (13.13b)

2 ^ = 0. (13.13c)
A system in equilibrium need not be at rest. Its center of mass may move with
uniform velocity in the absence of any net force. And, for example, a rigid body may
undergo uniform rotation about its center of mass in the absence of any net torque.
Momentum and angular momentum are both conserved for a system in equilibrium.
It is important to realize that the equilibrium conditions for force and torque are quite
distinct. A body can accelerate under the force of gravity, while experiencing no torque
about its center of mass.* Conversely, the forces on a body may cancel, but be applied
at different points so as to produce a nonzero torque.
As an example of this last point, Fig. 13.4 shows a couple. The forces balance, but
the torque is nonzero. The torque about point 1 has the value r12 x F 2 .

F2 = - F ,

Figure 13.4 A couple.

A curious property of the couple is that its torque about an arbitrary reference point
O, relative to which point 1 is situated at rx and point 2 airl + r 12 , is independent of
the reference point:
r, x F, + r 2 x F 2 = r, x ( - F 2 ) + (r, + r 12 ) x F 2

= r12 X F 2 . (13.14)
The torque due to an unbalanced force, by contrast, depends on the reference point.
A couple is an exception to the rule, mentioned in Section 6.6, that for any set of
forces a single resultant force can be found which gives the same total torque as well as

*Note, however, that the body does experience a torque about any other point. In general, if arigidbody
is in rotational equilibrium but not translational equilibrium, the sum of the external torques is zero only about
the center of mass.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 344 ANGULAR MOMENTUM

force, and that the resultant can be balanced by a single equilibrant force. A couple has
no resultant and no equilibrant. To balance a couple it is necessary to introduce two more
forces, in other words, another couple.
An important special case in which a force does not produce a torque occurs for
bodies orbiting around the origin under a central force. By central force we mean one
of the form
F =/(r)r (13.15)
which has a radial direction and a magnitude depending only on r. The gravitational and
Coulomb forces exerted by a point mass or charge are examples of central forces, but
f(r) need not vary as r" 2 . A central force exerts a torque about almost all reference points.
But if the reference point is placed at the force center itself, then r is always parallel to
the force, so
T = r X F = r X f(r)r = 0. (13.16)
The torque vanishes and continues to vanish as r changes direction in sweeping around
the orbit. Therefore angular momentum about the origin of a central force field is
conserved:

L = r x m\ = const, (13.17)

if no other forces act.

Note the generality of this result. It holds for uniform circular motion, with L =
rmv. But it also holds for orbits that swing inward and outward, with v alternatively
speeding and slowing. And it holds for any radial dependence/(/•)•
In the next section we'll use this powerful result to understand Kepler's empirical
law of equal areas for planetary motion.

13.5 KEPLER'S LAW OF EQUAL AREAS

As a young man, Johannes Kepler began an immensely fertile search for mathematical
relations describing the orbits of planets.
One question he seized upon was the relation between a planet's distance from the
sun and the time needed to complete one revolution (the orbital period). The orbital
periods of the five planets visible to the naked eye had been known since antiquity. The
ratio of their distances from the sun had been determined by Copernicus on the basis of
his heliocentric model. [For a clear description of how this was done see A. P. French,
Newtonian Mechanics (Norton, New York, 1971), pp. 246-9.] The greater the planet's
distance from the sun, the longer its period. However, a precise mathematical ratio was
lacking. Saturn, for example, is twice as far as Jupiter from the sun, but its 30-year
period is not twice the 12-year period of Jupiter. No simple ratio relates the distances
and periods of the other planets either. The orbital period of a planet increases with the
planet's distance from the sun, but not in the same proportion as the distance. Nobody
before Kepler is on record as having asked why this should be so.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.5 KEPLER'S LAW OF EQUAL AREAS 345

Kepler theorized that there must be a force emanating from the sun which drives the
planets in their orbits around the sun. The outer planets move more slowly because this
driving force diminishes with distance. Kepler's proposal had revolutionary significance.
For the first time an attempt was made not only to describe heavenly motion in geometrical
terms, but to assign a physical cause to it. After a divorce of 2000 years, astronomy and
physics met again. This reunion led to Kepler's three laws, the pillars on which Newton
built his universe.
After years of false starts and drudgery searching for a more accurate description in
the Copernican system of the motion of the planets, Kepler realized that the ancient idea
of circular orbits had to be abandoned to conform to observed data. Through obstinate
perseverance, he discovered on the basis of Tycho Brahe's new observations that the
actual orbits correspond more to elliptical paths rather than the circular paths of the
Copernican system. The sun is at one focus of the ellipse, and the earth, for example,
moves around it as shown in Fig. 13.5. This is Kepler's first law: the planets orbit the
sun along elliptical paths. We'll say more about this law in Chapter 16.

Planet
Figure 13.5 Illustration of Kepler's first law: the orbits of planets are
ellipses with the sun at one focus.

Since antiquity, astronomers had observed that the planets do not move along their
orbits at constant speed. Each planet moves faster when it is close to the sun than when
it is far away. This means that if we draw the angles swept out by the earth moving
around the sun in equal intervals of time at two different parts of the orbit, as illustrated
in Fig. 13.6, one angle is smaller than the other. In searching for some regularity in this

B1

Figure 13.6 An illustration of Kepler's second law: the radius vector

from the sun to a planet sweeps out equal areas in equal times.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 346 ANGULAR MOMENTUM

orbit, Kepler discovered a law of amazing simplicity. As a planet moves in its orbit, the
radius vector from the sun to the planet sweeps out equal areas in equal times; this is
Kepler's second law. Kepler also succeeded in finding a precise relation between a planet's
orbital period and its mean distance from the sun. This came to be known as his third
law and we'll discuss it in Chapter 17.
Nearly 100 years after Kepler, Isaac Newton formulated his laws of mechanics and
demonstrated that Kepler's laws can be deduced from his mechanics. Thus Newton
realized Kepler's dream of assigning a physical cause to the geometrical motions of the
planets. We'll now demonstrate precisely how Kepler's second law of equal areas arises
from Newton's dynamics. But first we need a mathematical description of the law of
equal areas.
Imagine a radius vector at some time r, r(t), drawn from the sun to a planet as in
Fig. 13.7. If the planet moves through a displacement Ar in a short time At, then its
new position is r(t 4- At) = r + Ar. As illustrated in Fig. 13.7, the three vectors r(t),

Figure 13.7 Position vector of a planet sweeping out an area AA in

time Ar.

Ar, and r(t + At) form a triangle. The area AA of this triangle is one-half the base r
times the height, Ar sin 0; thus

AA = Ir(Ar sin 6)

where 9 is the angle between r and Ar.

Writing the area in this way suggests using the vector cross product to represent the
area. Recall that in Chapter 5 we found that the magnitude of the cross product between
two vectors is equal to the area of the parallelogram formed by the vectors, as Fig. 13.8
shows. And the area of the triangle is half the area of the parallelogram. In vector notation,
we can write the area of the triangle as

AA = i|r x Ar|.

The direction of the cross product, given by the right-hand rule, is perpendicular to the
plane of the vectors. So we can introduce the vector

AA = I r x Ar

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.5 KEPLER'S LAW OF EQUAL AREAS 347

Figure 13.8 Area vector A A as a vector cross product

and call it the area vector A A. Its length is the area of the triangle formed by r and Ar.
The rate of change of this area vector, dkldt, is the limit of A A/A* as the time
interval shrinks to zero:

d\ ,. AA ,. lr x Ar
, ,. Ar
— = hm —- = hm : = 2^r x hm — .
dt Ar->0 Ar Ar-»O Af A/-0 A/

But the last limit is simply the velocity v of the planet. Therefore we find that the rate
of change of the area vector is
dk
— = Il r X v. (13.18)
dt
Now the connection to Newton's dynamics becomes clear. Comparing Eqs. (13.18)
and (13.3), we recognize that dkldt is proportional to angular momentum:

dt 2m
From Newton's second law we deduced that the angular momentum vector L is conserved
in a central force field*; that is,

L is constant in magnitude and direction.

Thus, Kepler's law of equal areas follows from Newton's second law.
The result that L has constant direction also supplies part of Kepler's first law, which
states that each orbit is an ellipse and hence lies in a plane. Since L/m is the cross product
of r and v, each of r and v must be perpendicular to L. But L has constant direction,
so both r and v must lie in a plane perpendicular to this direction. In other words,
conservation of angular momentum implies a planar orbit.

*Strictly speaking, it is the center of mass of the sun-planet system, not the center of the sun, at which
the origin of the central force field is located and about which L is conserved. The reason is that the center of
the sun is always accelerated toward the planet, and the sun as well as the planet revolves about the center of
mass. Although at first sight the center of the sun is also the origin of the central force, in fact the reference
frame with origin fixed at the center of the sun is noninertial, entailing noninertial forces which prevent L from
being conserved about this point. The center of mass, on the other hand, is unaccelerated and thus can be
placed at the origin of an inertial frame. In practice, however, the sun is so much more massive than the planets
that its center lies very near the center of mass, and one can ignore the difference between the two points for
most purposes.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 348 ANGULAR MOMENTUM

Example 3
At its closest approach (perihelion), the earth is 1.47 X 108 km from the sun and its
speed is 30.2 km/s. What is the speed at its farthest (aphelion) point from the sun, a
distance of 1.52 x 108 km?

Planetary orbit

As illustrated with the ellipse above, when a planet is either at its closest or farthest
point from the sun, its velocity is perpendicular to the radius vector. For these points Eq.
(13.18) and the fact that dA/dt is constant imply that

\rPvP = \rAvA.
Solving for vA, the speed of the planet at its farthest distance, we have
vA = vP(rP/rA) = (30.2 km/s)(l.47/1.52) = 29.2 km/s.

13.6 VORTICES AND FIRESTORMS

We've just found that the law of conservation of angular momentum explains why planets
orbiting the sun sweep out equal areas in equal times - an empirical law which Kepler
discovered by examining voluminous data. The conservation of angular momentum ap-
plies not only to planets moving silently around the sun, but to any moving body for
which r x m\ is constant. One interesting application of this law is to vortices.
Suppose you have a huge bowl (a bathtub will suffice), which you fill with water
through a hose, as shown in Fig. 13.9. As water pours into the bowl, it moves in some
type of circulatory motion. If you turn off the water, that rotary motion will die out very
slowly; it may not completely disappear for hours or even days. What happens if you
pull the plug on the bottom of the bowl while the water is still rotating? Of course, the
water will run out.
When you first pull the plug, the water flows straight out through the hole, but after
a while, the water goes into a new kind of flow state. Instead of flowing straight out, it
forms a whirlpool. Why? Aside from a small amount of viscosity from the walls of the
bowl, there are no torques acting on the water. Consequently, the water conserves angular
momentum as it runs out.
The angular momentum of a small portion of water at any instant as it spirals toward
the center of the bowl is L = r x m\ about an axis through the center of the bowl.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.6 VORTICES AND FIRESTORMS 349

Figure 13.9 Vortex formation in a bowl of water as a result of

angular momentum conservation.

Because r and v are perpendicular, the magnitude of the angular momentum of this chunk
is
L = mvr — const.
The precise value of the constant depends on the initial conditions of rotation.
As each small bit of water moves down toward the hole, it conserves angular mo-
mentum. So as the distance from the central axis becomes smaller, the velocity becomes
increasingly larger, according to
v = const/wr.
When the water is moving in very small circles, it is moving very rapidly. But an inward
force must keep this bit of water moving in a circle. That force which provides the
centripetal acceleration is the tensile strength of the water - the ability of the water to
hold itself together.
As the distance r becomes very small, the velocity correspondingly becomes ex-
tremely large. A large force is required to hold the water together. But when this force
exceeds the tensile strength of the water, the water can no longer keep itself moving in
a circle, so the surface ruptures and forms a hole. The hole in the center of the whirlpool
is called a vortex.
The same dynamics occurs in other instances on a much larger scale. Whenever there
is a large fire, such as a forest fire, great destruction can occur. This happened in the
Chicago fire of 1871, and in the bombing of Hamburg, Germany, during the Second
World War. The cause of the destruction was not a simple fire, but something more
devastating, a firestorm. Firestorms are vortices similar to those in a bathtub, and are a
result of conservation of angular momentum.
The heat of the fire causes the air to swell upward (just as we had water flowing
downward). The resulting low-pressure region at the bottom draws in oxygen from the
sides, so the fire burns faster. If there is any circulatory motion in the air, the air moves
faster and faster near the center of the fire when the upwelling occurs. Since the air cannot
provide the centripetal force to keep itself moving so rapidly, a vortex forms. Thus the
same phenomenon occurs as in a bowl of water, except upside down. This vortex is a
long-lived and violent state - the firestorm.
Meteorologists investigating historical records for the Great Fire of London in 1666
discovered that the air in the vicinity at the time had no distinct circulatory motion.
Consequently, they believe that a firestorm never developed. Their conjecture is borne
out by a comparison in the tolls: in the London fire, only four people died directly as a

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 350 ANGULAR MOMENTUM

result of the fire and only 436 acres were burnt in 87 hours; on the other hand, in the
Chicago fire, 2124 acres were burnt and 250 people killed in two days.
Another example of the same phenomenon occurring without a fire is upwellings of
air in the atmosphere. If there is an upwelling of air, such as that above warm water,
and if there is any circulatory motion, a vortex can form. In this case it is called a
hurricane. Hurricanes can be long-lived (several days), persistent, and very destructive.
The longest-lived hurricane we know of is on the planet Jupiter; it is the Great Red Spot
of Jupiter, which has persisted for at least 300 years (we've only been able to observe
it since the invention of the telescope in 1609). Figure 13.10 is a photograph of the Red
Spot taken by Voyager 1 on its flight past the planet. Years before the Voyager fly-by,
it was suggested that the Red Spot was a huge hurricane, three times the size of the earth.
Data from Voyager support this idea.

Figure 13.10 Photograph from Voyager 1 of the Great Red Spot on

Jupiter. (Courtesy JPL/NASA.)

13.7 CONSERVATION OF ANGULAR MOMENTUM AND ENERGY

When a particle moves in a gravitational or Coulomb force field, its direction changes
continuously so momentum is not conserved. However, its angular momentum about the
force center is conserved, as we have seen, and its total mechanical energy is also
conserved.
Because of angular momentum conservation, the motion is confined to a plane
perpendicular to the angular momentum vector. The situation is most readily analyzed
by resolving the velocity vector into two perpendicular components, a radial component
vT and a transverse component vB, as indicated in Fig. 13.11.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.7 CONSERVATION OF ANGULAR MOMENTUM AND ENERGY 351

0-
Figure 13.11 Components of velocity in perpendicular directions.

Since v2 = v2 + VQ by the theorem of Pythagoras, the conserved energy can be

written

E = U(r) + K = U(r) + \mvl + [mv}. (13.20)

For the gravitational or electrical applications we have in mind

D
U(r) = - - , (13.21)

where D is GMm or - Kcq {q2, at large r, but U{r) may have a less singular r dependence
at small r if the source is spread out rather than being pointlike.
The angular momentum of the particle about the force center has magnitude

so we can rewrite the nonradial part of the kinetic energy as

(1322)
¥=£? •
This is interpreted as the rotational kinetic energy of the particle. The total energy now
has the form
j2 2

E = U(r) + + — . (13.23)

If we call the first two terms on the right-hand side

t/effto = Uir) + - ^ 4 (13.24)

and plot them as in Fig. 13.12, we find that Uerf(r) rises rapidly at small r and exceeds
the total energy E at some radius r min . Because the radial kinetic energy jmv2 must be
positive, rmin is the closest the particle can approach the origin.
Another way to see that the particle cannot reach arbitrarily small r is by considering
forces. Although the attractive gravitational or electrical force urges the particle to come
closer, angular momentum conservation requires the particle to keep circulating as well.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 352 ANGULAR MOMENTUM

Figure 13.12 Potential, rotational kinetic, and total energy of a

particle moving under a gravitational or electrical force.

The centripetal force

(13.25)
r mr3 mr3
required to keep the particle circulating grows faster at small r than the attractive force,
which grows as \F\ = D/r2 for a point gravitational or electrical source or more slowly
for a spread out source. The available attractive force is insufficient to supply the required
centripetal force when

(13.26)
mrJ
as happens for example at
L2
r < rn = (13.27)
mD
if |F| = D/r2. Beyond this point an inward-moving particle is slowed down, and eventually
turns around.
If E < 0, that is, if a particle's initial radial velocity is less than its escape velocity,
there is also a maximum distance rmax as pictured in Fig. 13.12. So a particle with E < 0
is constrained to shuttle back and forth between rmin and r max , with the radial kinetic
energy ^mv2 making up the difference between £/eff(r) and the constant E at each r.
Although /7eff(r) is a hybrid of U(r) and rotational kinetic energy L2/2mr2, it depends
only on r for constant L, and is often called the effective potential because of the potential-
like role it plays in Fig. 13.12.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.7 CONSERVATION OF ANGULAR MOMENTUM AND ENERGY 353

If there is other matter present to supply resistive or frictional forces, the particle
will tend to lose radial kinetic energy and settle down into a circular orbit at the minimum
or equilibrium point of t/eff (r), located at rm in Fig. 13.12. In determining rm it is necessary
to be careful about one point, though: the interaction with other matter will also tend to
change the angular momentum of the particle we are considering. Nevertheless, its L is
not likely to fall to zero, because the overall angular momentum of the system is conserved
and must be shared among the particles. Thus the shape of £/eff(r) for a typical particle
will still resemble Fig. 13.12, although the particle's angular momentum will no longer
have its initial value and the particle will still settle down into a circular orbit.
The process we have just gone through - setting up an idealized problem in which
the energy and angular momentum of a particle in a spherically symmetric system are
conserved, and then admitting that in reality dissipative forces change the energy and
angular momentum of the particle (and even the spherical symmetry of the system), is
typical of physics. The idealizations allow us to express the essence of the problem in
terms of simple mathematics. Once this is done, the outlines of how realistic complications
will change the answer can often be seen without having to calculate their effects in
complete detail.
The conservation of overall angular momentum has interesting implications for a
system that changes its size or shape. An especially fascinating application is something
which not even the imaginative Kepler would have envisioned - the contraction of colossal
clouds of gas and dust to form galaxies and solar systems.
Suppose initially the cloud is spherical, but has a rotation about a single axis. If a
particle is initially on the "equator" near the outside of this mass distribution and sharing
the general circulation, it will experience a £/eff(r) of the qualitative nature we have
discussed in Fig. 13.12, although U(r) will have a somewhat different r dependence
because the attracting matter is not all concentrated at the center.
In a condensing gas cloud, a particle spirals in, losing energy by collisions with
other particles until it eventually reaches the lowest energy possible for its angular mo-
mentum. The stable orbit reached at this energy is a circle, as described earlier. As Fig.
13.13 indicates, particles with different values of angular momentum will have different
minimum values of energy corresponding to circular orbits of different radii.
For a given U(r) and L we can mathematically determine the radius of the circular
orbit that corresponds to minimum energy; it is the value of r for which the slope of the
curve is zero. In other words, the minimum energy occurs where dUtii{r)fdr = 0. Taking
the derivative of £/eff(r) = U{r) 4- L2/2mr2 in the approximation where the potential has
the form U — - GMm/r> we obtain
_ L Mm
3
dr mr r2
Setting this equal to zero, and solving for r m , the corresponding distance, we find
L2 (13.28)
GMm2
Looking back at Eq. (13.27) we recognize rm as r 0 , the distance where the gravitational
attraction supplies exactly enough centripetal force to maintain circular motion. The radius
of the circular orbit can be found either by minimizing Ueff or from this force requirement.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 354 ANGULAR MOMENTUM

Energy

Figure 13.13 Stable orbit of a particle subject to dissipative collisions

depends on the angular momentum.

We now have a picture of the evolution of a rotating gas cloud. As Fig. 13.14 depicts,
the cloud begins to contract toward a point under the mutual attraction of gravity. As it
contracts in the equatorial direction, the cloud can't move all the way in because of
overall conservation of angular momentum. Particles will settle into circular orbits whose
radii are prescribed by Eq. (13.28). However, in the vertical direction, contraction can
take place without any increase in rotational kinetic energy and therefore continues after
the limit to contraction in the equatorial direction has been reached. Consequently, the
cloud tends to flatten out, forming galaxies like those in Fig. 13.15.
In addition to the pancakelike shape, some galaxies exhibit spiral arms. The simple
explanation we found for the flattening is not sufficient to explain the structure of the
spiral arms. Different phenomena are occurring. Likewise, when gas clouds condense to
form solar systems, planets are formed, rather than a uniformly flattened structure as we
have in galaxies. The same thing happens in planetary systems; moons are formed out
of the gas and dust condensing into planets. Sometimes rings are formed. Since at least
three planets in our solar system have rings, we know that ring formation is not unusual.
In all these phenomena, there presumably is no underlying physics involved other
than Newton's laws. If we could solve the differential equations for all sorts of conditions,
then we should be able to predict all the spectacular phenomena, such as Saturn's braided
rings, which were unexpected before the Voyager exploration. There are solutions to
Newton's simple equations which nobody ever guessed existed. Commenting on this,
Nobel laureate Richard P. Feynman wrote in 1962:
There are those who are going to be disappointed when no life is found on other planets.
Not I. I want to be reminded and delighted and surprised once again through interplanetary
exploration of the infinite variety and novelty of phenomena that can be generated from such
simple principles. The test of science is its ability to predict. Had you never visited earth,

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 13.8 A FINAL WORD 355

Figure 13.14 Successive stages in the contraction of a rotating sphere

of gas.
could you predict the thunder storms, the volcanoes, the ocean waves, the auroras and the
colorful sunsets? A salutory lesson it will be when we learn of all that goes on, on each of
those dead planets - those eight or ten balls - each agglomerated from the same dust cloud
and each obeying exactly the same laws of physics.

13.8 A FINAL WORD

We now have derived the third of the three great conservation laws in physics: conservation
of energy, conservation of linear momentum, and conservation of angular momentum.
All three laws came from Newton's law, F = ma. On the other hand, the conserved
quantities involved are simple things which must be true all the time; they are useful
bookkeeping devices for applying Newton's laws in complicated situations. For example,
to have written Newton's laws for the water in the vortex bowl as it spirals down and
out through the hole would have been exceedingly difficult. But the principle of why a
hole develops in the middle is not difficult to understand, once we understand that angular
momentum is conserved. This suggests that the conservation laws may be more than
simply special ways of applying Newton's laws.
In the twentieth century the world witnessed a scientific revolution which is com-
parable in magnitude to the original scientific revolution of the period from Copernicus
to Newton. We no longer believe that Newton's laws are a sufficient, adequate description

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 356 ANGULAR MOMENTUM

Figure 13.15 Pictures of galaxies exhibiting pancakelike structure.

(Palomar Observatory photograph.)

of the way the universe actually works. His laws are a special case, an approximation to
deeper and more exact laws, known as the laws of quantum mechanics. Surprisingly,
though, the three conservation laws of energy, linear momentum, and angular momentum
exist in quantum mechanics also. Remember that they were derived from Newton's laws,
which we no longer believe are universally correct. But in the most profound and correct
theory that we have, the conservation laws themselves persist.
We've revealed some of the most profound aspects of nature - the three conservation
laws, laws you'll never have to abandon. Hold on to them; they'll serve you well.

Problems

Torque and Angular Momentum

1. A worker finds it difficult to twist a stubborn bolt with a wrench, so he attaches a

rope to the wrench as shown in (b), and pulls just as hard on the rope. How does
the torque applied with the rope compare to that without the rope? How does the
applied torque change if the worker runs the rope over a fixed pulley as shown in
(c)?

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 PROBLEMS 357

(a)

(b)

(c)

2. A 1.5-kg particle has position and velocity as shown with r = 0.4 m and v =
3.0.m/s. The magnitude of the force acting on it is 4.0 N. Relative to the origin
find

160°

(a) the angular momentum of the particle,

(b) the torque acting on it.

3. A particle at position r = (3i + 6j - 2k) m is acted upon by a force F equal to

( — 4i - 3j + 5k) N. Find the torque about the origin acting on the particle.

4. In a certain coordinate system, a 0.5-kg particle is at a position given by the vec-

tor r = (3i - 2j + k) m moving with a velocity v = ( - 5 i + 3j - 2k) m/s.
(a) Find the angular momentum of this particle with respect to the origin of the
coordinate system.
(b) Calculate the angular momentum relative to the point (1,1,1).
5. A particle of mass m is at the end of a rod of negligible mass and length R.

(a) If at any instant the rod makes an angle 8 with the horizontal as it falls, calculate
the torque about 0 acting on the mass as it falls.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 358 ANGULAR MOMENTUM

(b) If the mass is started from the vertical postion (9 = IT/2) with initial speed v,
find its angular momentum about 0 as a function of 8.
(c) Use your results to find the angular speed dQ/dt of the mass.
6. Two identical masses, each of mass m, move along straight lines with velocities v
and - v , respectively, as shown in the diagram.

(a) Calculate the angular momentum of the system with respect to the origin 0.
(b) In this particular problem, L is the same with respect to any reference point, as
you can readily check by considering some simple cases such as L about a point
on the upper dotted line. Construct a proof that the angular momentum of two
mass points is independent of the reference point whenever (as in the present
problem) the total linear momentum 2 ntfUi of the system vanishes.
7. A body of mass m falls from rest in the earth's gravitational field according to
Galileo's law, z = z0 — Vi gt2. Its horizontal coordinates are x = JC0, y - 0.

(a) Determine the position vector r and velocity vector v for this case.
(b) Find the angular momentum vector L about the origin as a function of time.
(c) Find the torque vector T about the origin.
(d) Show explicitly that T = dLldt and that the angular impulse equals the change
in angular momentum.
8. A particle of charge e moving with velocity v in a magnetic field B feels a force
F = e\ x B. The field is uniform along the z axis, B = Bok.

(a) For what direction of v is momentum p conserved?

(b) if v = 3j + 5k and the particle is located at r = 2i + 2j, what is the torque
on the particle around the origin?
c
( ) At an instant when the particle is at r = 5i, for what directions of v is
dLldt = 0?
(d) If the particle has initial position r = 3j and initial velocity v = 5i, in what
plane does it move? Show that the motion is a circle of radius mv/(eB0) in that
plane.

Angular Momentum Conservation

9. Suppose an object were acted upon by the force F = -Ax, where r is the radius
vector from some origin and k is constant. Would the angular momentum about
the origin of the particle be conserved? Explain why.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 PROBLEMS 359

10. A mass m is twirled in a circle of radius rl with a constant speed V\. If the string
is pulled so that the mass moves in a circle of radius r2 as shown, find

(a) the speed of the particle,

(b) the angular speed of the particle.

11. Two objects of mass 2.0 kg and 3.0 kg are connected by a light rod and move in
a horizontal circle as shown. The speed of each is 1.5 m/s.

(a) What is the total angular momentum of the objects about the center?
(b) If the rod contracts uniformly to one-third of its original length, will the speed
of the objects change? If so, by how much?
12. A mass m revolves at a distance of rx about an axis with an angular velocity coj.
Suddenly the distance from the axis is decreased to r2 by application of a radially
inward force.

(a) What is the new angular velocity?

(b) By how much is the kinetic energy increased?
(c) Show that your answer to part (b) is precisely the amount of work done by the
centripetal force during the contraction from rY to r2.
13. A toy train slides with coefficient of friction fx on a track fastened to a horizontal
bicycle wheel. The wheel turns about its vertical axis without friction. Initially the
train and wheel are at rest.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 360 ANGULAR MOMENTUM

(a) The train is struck horizontally, setting it into clockwise motion. Describe the
subsequent motion of the train and the wheel.
(b) If, after losing half its initial speed, the train is again struck horizontally, bringing
it suddenly to rest while the wheel still rotates, describe the ensuing motion.
Give the relation between the final wheel speed in this case and the final wheel
speed achieved in part (a) when the wheel was not struck a second time.

Force and Torque

14. (a) Find the resultant for the unbalanced pair of forces — Fj applied at the origin
and (F + AF)j applied at JCI.
(b) Show that the point of application of this resultant moves to larger x as AF is
decreased, and in fact moves to x = °° as AF approaches zero.
(c) In the limit as AF -» 0 we are left with a couple - F and F. Show that - F
and F could point in any direction in the xy plane and still provide the same
torque.

Kepler's Law of Equal Areas

15. Referring to the elliptical orbit shown in Fig. 13.6, where is the velocity of the
planet a maximum? Where is the velocity a minimum? What does Kepler's second
law imply about the speed of a planet moving in a circular orbit?
16. If the speed of the earth is 29.8 km/s when it is 1.49 x 108 km from the sun,
what angle does the velocity vector make with its position vector? (Use data in
Example 3 on p. 348.)

Vortices
17. In what way would you expect the radius of a hole in the center of a whirlpool of
given angular momentum to depend on the tensile strength of the liquid?
18. In an attempt to provide a physical mechanism that causes the planets to orbit the
sun, Descartes (working before Newton) formulated a vortex theory. In this
model, the universe is filled with matter. The sun is located at the center of a
whirlpool of air. The earth and other planets are embedded in the matter and
hence are swept around the sun by the vortex motion of the surrounding medium.
Based upon your understanding of vortices, can you find fault with this model?

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

 PROBLEMS 361

Conservation of Angular Momentum and Energy

19. The orbits of all the planets lie roughly in the same plane. If the planets were
formed from a condensing gas cloud, why would you expect this feature?
20. Suppose two gas clouds initially have the same angular momentum but one is
much more massive than the other. If these clouds eventually form galaxies,
which will form a disk with larger radius?
21. In what type of orbit, elliptical or circular of the same angular momentum, does a
planet have greater total energy?
22. A rocket fired off the earth with speed ve > 2GMJRe will escape the earth no
matter which direction it is fired in, as pointed out in Section 10.4. However, the
maximum height attained by a rocket fired with speed v < ve is greater the more
vertical the initial direction. Prove this by drawing the curve

for a vertical launch at energy

E = l- mv2r + Uc(f < 0,

drawing £/cff for a nonvertical launch at the same E, and comparing the
maximum distance rmax for the two cases. On the same plot, show that E >
0 leads to escape for any initial direction of firing.

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press

https://doi.org/10.1017/CBO9780511818493.014 Published online by Cambridge University Press