Angular Momentum.

Physics 3550, Fall 2012 Angular Momentum. Relevant Sections in Text: 3.4, 3.5 Angular Momentum You have certainly encountered angular momentum in a previous class. The importance of angular momentum lies principally in the fact that it is conserved in appropriate circumstances, as we shall explore. It is worth noting that the angular momentum of electrons in atoms is one of the most important physical features of atoms and largely makes them what they are. The angular momentum of a single particle is dened by* l = r p, where usually the momentum is just p = mv (more on this later). Notice that the angular momentum involves the position vector, which is dened relative to a choice of origin. Thus the angular momentum is dened relative to an origin. It is easy to forget this. Dont. As I mentioned, we focus on angular momentum because of its conservation under appropriate circumstances. What are these circumstances? Well, let the particle move according to Newtons second law, we have dl = v (mv ) + r F . dt The rst term involves the cross product of two parallel vectors and so it vanishes. We then get dl = , dt where =rF is the torque. Like the position vector and the angular momentum, the torque is always dened relative to a choice of origin. We tacitly assume the torque is dened relative to the same origin as the angular momentum. We see then that the angular momentum of a particle is conserved if and only if the torque vanishes. For a single particle, the torque can vanish and angular momentum is conserved in three possible ways, which you can easily see from the denition of torque as a cross * Why is angular momentum dened this way? Is it always dened that way? Why is the total angular momentum conserved? All these questions will be answered when we explore the Lagrangian form of mechanics. 1

Angular Momentum.

product. Either (i) the particle spends all its time at the origin. Or (ii) the net force on the particle is zero. Or (iii) the force is (anti)parallel to the position vector, i.e., the force points toward or away from the origin. The latter possibility is the most important; such forces are called central forces. What happens if you we use dierent origins to compute our angular momenta? Let the two origins dier by a vector d, the two position vectors are related by r1 = r2 + d, while the velocities and momenta are the same relative to the two choices of origin. So the two angular momenta are related by l1 = l2 + d p. Note, in particular, that this means if the angular momentum relative to a particular choice of origin is conserved it may not be conserved relative to another choice of origin: dl1 dl dp dl = 2 +d = 2 + d F. dt dt dt dt A classic and important application of angular momentum conservation appears in trying to understand Keplers second law of planetary motion. Recall that Keplers rst law says that planets move on ellipses, with the sun at one of the foci. (We will prove this later.) The second law of Kepler says that the motion in time of the planets is such that the relative position vector of the planet and the sun sweeps out equal areas in equal times. This result can be understood via conservation of angular momentum. Lets look into this. First, we assume for simplicity that the sun is so much more massive than the planets that the center of mass can be idealized as being at the center of the sun:* R= 1 msun rsun + mplanet rplanet rsun , msun + mplanet mplanet << msun .

It is this point which the planets orbit around and it is there we place the origin. To verify that the angular momentum (about this origin) is conserved we note that in the chosen reference frame the force on the planet is of the form k F = 3 rplanet , rplanet where k has Newtons constant and the masses. This is a central force; the torque vanishes and angular momentum is conserved. * This assumption is not needed, as we shall see. 2

Angular Momentum.

Now we prove Keplers second law using angular momentum conservation. Consider the position vector at two nearby times, t and t + dt. In the limit as the time interval vanishes, the two position vectors and their relative displacement form a (very skinny!) triangle with edges r(t), r(t + dt), and dr = r(t + dt) r(t). Next we use an important math fact: a, b and c = a b dene a triangle whose area is the magnitude of one-half the cross product of any two of the vectors. You can easily check this and you should check it since this is one of the main geometric intepretations of the cross product. Applying this result to the position vectors we have that the area dA swept out in time dt is 1 dA = ||r dr||. 2 Now, divide by dt to get* dA 1 1 = ||r p|| = l. dt 2mplanet 2mplanet So, assuming l (or at least its magnitude) does not change in time, we see that the rate of change of area is constant in time, i.e., A(t2 ) A(t1 ) = (const.)(t2 t1 ), or A = (const.)t, which is Keplers second law. (We got this formula by integrating.) Let us look at another example. We return to the system consisting of a charged particle (mass m, charge q ) moving in a uniform magnetic eld, Bk . The force is F = qv B. The magnetic eld is homogeneous (but not isotropic); we can x the origin anywhere we like. The torque is = r F = qr (v B ). Next I invoke a very famous vector identity (you should make sure you know this one): A (B C ) = (A C )B (A B )C. * Note this is a nice, generally valid, geometric interpretation of the magnitude of angular momentum. 3

Angular Momentum.

From this identity the torque takes the form: = q (r B )v (r v )B. From this result we can recover the familiar result that for circular motion in a plane perpendicular to B , the angular momentum is conserved. The angular momentum in this case is parallel to B and has magnitude mvr, where v is the particles speed and r is the radius of the circle. Make sure you can prove this to yourself. Angular momentum of a system of particles Let us now consider a system consisting of two particles; the generalization to more than two particles is pretty straightforward. We dene the total angular momentum as L = r1 p1 + r2 p2 . The time rate of change of this observable is dL = r1 F1 + r2 F2 = 1 + 2 = . dt We have denoted by the total torque on the system. The force on, say, particle 1 is a superposition of the force F21 from particle 2 and any external forces F1ext : F1 = F21 + F1ext . Similarly, for the force on particle 2 we have F2 = F12 + F2ext . Let us assume that the interparticle force obeys Newtons second law. We have F21 = F12 F . We can then write dL = (r1 r2 ) F + ext . dt We now generalize our notion of central force. We say two particles interact via a central force if the mutual force is always directed along a line joining the two particles. In this case, we have dL = ext . dt If there are no external torques the angular momentum is conserved. 4

Angular Momentum.

A good example of a central force is the gravitational force, where F21 = G m1 m2 (r2 r1 ). |r1 r2 |3

So, for a binary star system, ignoring all external inuences, the total angular momentum is conserved. As a nice exercise you can show that if, say, m2 >> m1 , (as in the system of a planet and the sun) the usual 1-particle central force law discussed earlier in the context of Keplers laws is recovered. You can easily generalize this discussion to three or more particles. The key result is that, for a system of particles interacting via central forces, the time rate of change of total angular momentum is given by the net external torque.

Motion in a plane Let us consider a nice application of some of our ideas developed thus far. Consider the motion of two particles under a mutual central force with no other forces (e.g., gravitational forces only). Again, think of a binary star system, or the Earth-Sun system, Earth-Moon system, etc. We can now show that there always exists a reference frame in which the motion must lie in a plane. The proof is pretty simple. First of all, since the forces are internal and obey Newtont third law the center of mass of the system must have no acceleration. Let us work in the reference frame which is at rest relative to the center of mass. Put the origin at the center of mass. In this reference frame we have m1 r1 + m2 r2 = 0, m1 v1 + m2 v2 = 0. The rst equation just says the center of mass is at the origin. The second equation is just the time derivative of this equation telling us the center of mass is at rest in this reference frame. Note that this reference frame has the particles always having diametrically opposed position vectors. Now, consider the conserved total angular momentum; we have L = r1 p1 + r2 p2 = r v, where r = r1 r2 , are the relative position and velocity and = m1 m2 m1 + m2 5 v = v1 v2 ,

Angular Momentum.

is the reduced mass. To prove this, you need to use r1 = m2 r, m1 + m2 r2 = m1 r. m1 + m2

The point of this result is the following. The total angular momentum L is conserved. This means that, as time runs, positions and velocities change, but r v is a xed vector in space. In particular, its direction is xed. Since r and v are perpendicuar to this xed vector, as time evolves the relative position (and velocity) moves in a plane orthogonal to L. But this means that each of the particles moves in that plane. This is why planetary motion is always in a plane the plane being determined by the angular momentum of the system. Now for an amusing little paradox. Consider two bodies moving under an attractive central force, such as just described. The system could be the Earth-Sun system, or it could be the hydrogen atom. Under what circumstances is the angular momentum zero? Well, evidently, we can get L = 0 by either having (i) r = 0, or (ii) v = 0, or (iii) r || v . Physically these mathematical relations say: (i) the particles are on top of each other, (ii) the particles are moving with the same speed in the same direction, (iii) the particles are moving toward each other or away from each other. The rst situation is a collision. The second situation only happens when there is no force. The third situation corresponds to a collision in the future, or one in the past. Now for the paradox: the ground state of the hydrogen atom has zero angular momentum. How can this be?