Download Alternative Perspectives in Classical Physics: Lagrangian & Hamiltonian Approaches and more Study notes Physics in PDF only on Docsity!
Least Action Principle in Physics
Muhammad Gaffar^1
1 Bachelor Student at Department of Physics
Universitas Indonesia, Depok 16425
17 Nov 2017
Abstract. We learn physics by starting it from classical problem, where Newton’s laws govern the classical physics. In high school or often in early physics in undergraduate, we still solve classic problem with newton’s equation of motion. In this article, I will show that there’s more alternative way to see classical problem instead seeing from Newton’s law, where it seems so boring. The alternative ways we will see that is Lagrangian formula and Hamiltonian’s equation of motion. And you will see that Newton’s laws less fundamental from alternative methods where they are derived from more fundamental principle, called Least Action Principle. They are still same physics and same result, but different perspective and advantages. Newton’s laws have their own advantages, so as Lagrangian and Hamiltonian. Yet, from Lagrangian and Hamiltonian we can see further the beautiful of physics and mathematics inside. Furthermore, we will not just see Least Action Principle in classical paradigm, since we are now lives in physics are governed by Quantum mehcanics, so we will see this principle’s completeness from Quantum paradigm. Again, we will see the beautifulness of this principle arise in Quantum Mechanics, and of course the strangeness too.
1 Introduction
The principle of least action has been great tool for discovery law of nature. Thanks to Euler, Lagrange, Hamilton, and Jacobi that transform this useful principle to mathematical sense. We have tackle many classical problem, like particle trajectory, or whatever with Euler-Lagrange or Hamilton-Jacobi equation. Unfortunately, often as student, appreciation to this method often just limited by how it can solve analytical problems. Sometimes, or even rarely we are lack of how we interpret the mathematics behind this principle. So here, we are asking how are actually the meaning behind this principle and meaning of the its mathematics. In addition, I will show new paradigm of this principle arise in quantum mechanics.
2 A Boring Method
Before I dive too deep about least action principle, let me give short explanation about old-method in classical mechanics, newtonian method. Suppose you have particle with position ~r(t), and acted upon by a force F~. By Newton’s Laws, we can write down the equation to be
F^ ~ = m~a (1)
or in more gorgeous way, m~¨r = −∇~V (r) (2)
This second-order differential equation∗^ have two integration constant, initial position ~r(t) and initial velocity ~v(t), where V (r) is potential acted upon on system. And voila!, we can draw the trajectory or prediction about how system behave (usually position × time). Yet, there’s problems in this method. First, it contains vector, where we need consider the relative position, or relative velocity. Second, it just about initial information (position and velocity). Okay, in other persepctive the
∗In more general form, where Cartesian coordinates are not always convenient. In a set of curvilinear coordinates, Newton’s second law is
F a^ = m
( (^) d (^2) ξa dt^2
dξc dt
)
second problem is not really a problem, but in most case we want solve a problem where we know the initial information, and final information that we want to be. Third, it is hard and takes long time to solve in system contains more object to taken into account. Fourth, it is boring. About 4 years we solve mechanics problem with just one method and limited case (high school to second-semester college). At last but not least, we cannot derive least action principle.
3 Action in nutshell
If you read carefully this principle, for first time you must be disturbed by the ’action’ notion (at least for me in the first time). So, why least action, not lowest possible energy? Words like ’energy’ are bandied about as if we know exactly what we mean (when we do not), and these have even been embodied symbolically in many famous equations of physics. (Perhaps there exist more refined concepts.) Among these old concepts are the familiar definitions of potential energy and kinetic energy. But if, as Einstein discovered, all motion (speed) is relative, and if all positions are relative, it appears that these concepts of ’potential and kinetic energy’ need to be, in some sense, qualified^6. So we need alternative way to look how nature works, especially trajectory or history problems. The standard and most used definition of action was defined by Hamilton,
S ≡
∫ (^) t
0
L(x, x˙; t) dt (3)
Where L(x, x˙; t) is Lagrangian,
L ≡
m x˙^2 − V (x) (4)
Universally, energy must be conserved. In same sense, there must be properties need to be hold in action. Yes, nature works where action must be stationary,
δS = 0 (5)
Wait, what’s the meaning behind integral of lagrangian? and where minus sign come from? Well, it may be hard too make action mathematical definition intuitive. But it works, we can derive equa- tion of motion from action mathematical definition where action must be stationary, and widened to scope of Electrodynamics, General Relativity, and Quantum Mechanics. Well, if you still not convinced enough, let me give you some illustration and derivation equation of motion from (3) and (5) after that. This illustration was narrated by Feynman^2 ,
Suppose you have a particle (in a gravitational field, for instance) which starts somewhere and moves to some other point by free motionyou throw it, and it goes up and comes down (Fig. 1). It goes from the original place to the final place in a certain amount of time. Now, you try a different motion. Suppose that to get from here to there, it went as shown in Fig. 2 but got there in just the same amount of time. Then he said this: If you calculate the kinetic energy at every moment on the path, take away the potential energy, and integrate it over the time dur- ing the whole path, youll find that the number youll get is bigger than that for the actual motion.
r 1 (t)
r 2 (t)
r
t Figure 1.
In other words, the laws of Newton could be stated not in the form F = ma but in the form: the average kinetic energy less the average potential energy is as little as possible for the path of an object going from one point to another.
4 Sophisticated Method
4.1 Euler-Lagrange
Okay, we have derived the equation of motion. Then what? of course, the goal is not just to show derivation of equation of motion. But, to give alternative way to solve mechanics problem. Instead using initial information, how about initial and final information, i.e position. This is what called Euler-Lagrange equation offers. There’s no new physics in Euler-Lagrange equation, just new perspective in understanding nature, where nature works in least action principle. Before, the change value of action δS = 0. So we can derive Euler-Lagrange from the functional of action S. After some math∗, we get Euler-Lagrange (EL) equation
∂L ∂qj
d dt
∂L
∂ q˙j
k
λk(t)
∂fk ∂qj
Where L = T − V , T as kinetic energy of system, and V as potential acted upon by system as I write in equation (4). We can say that q is generalized coordinates. It is therefore worth pointing out that Lagrange’s equations expressed above, the undetermined multipliers λk(t) are closely related to the forces of constraint^7. You see that Lagrangian’s method is simpler where we deal only with energy of the system. It’s scalar not vector. In fact it is super useful in quantum mechanics where we normally know the energies but not the forces.
Although there’s no difference in result of mechanics analysis in body of system. But from philosoph- ical viewpoint, we can make distinction. Let me re-state how nature works in least action principle. In the Newtonian formulation, a certain force on a body produces a definite motion-that is, we al- ways associate a definite effect with certain cause. According to Hamilton’s principle, however, the motion of a body results from the attempt of nature achieve a certain purpose, namely, to minimize the time integral of difference between the kinetic and potential energies^7.
EL equation is really good equation. It follows many rule of physics, well if not, it’s not good equation. Let me present you one of the most beautiful and useful theorems in physics. It deals with two fundamental concepts, namely symmeries and conserved quantities. It’s called Noether’s theorem.
Noether’s theorem: For each symmetry of the Lagrangian, there is a conserved quantity.
Through this beautiful theorem, let me prove the some properties of Lagrange Equation.
Consider a small shift of coordinates,
qi → qi + Ki(q) (15)
Ki(q) may be a function of all the qi. The fact that the lagrangian does not change at first order in means that
0 = dL d
i
∂L
∂qi
∂qi ∂
∂L
∂ q˙i
∂ q˙i ∂
i
∂L
∂qi
Ki +
∂L
∂ q˙i
K^ ˙i
Let me rewrite in momentum, where p ≡ (^) ∂∂L q˙i Then we can write
i
pK ˙ i + p K˙i
It’s easy to check
0 =
d dt
i
pKi (16)
Hence, we get (^) ∑
i
pKi = Q (17)
∗You can easily check the derivation in your textbook, i.e Marion Thronton or David Morin
for K 1 = K 2 = 1, we see that Q = p 1 + p 2 (18)
This is conservation of linear momentum. and for K 1 = q 2 and K 2 = −q 1 , we see that
Q = q 2 p 1 − q 1 p 2 = ~q × ~p (19)
And this is conservation of angular momentum. And let me prove you one last thing where every methods must hold, the conservation of energy. In closed system, the system has time translation symmetry. So, dL dt
So, total derivative of Lagrangian becomes
dL dt
j
∂L
∂qj
q ˙j +
∂L
∂ q˙j
q ¨j (21)
But we know before, the lagrange’s equation for zero constraint are
∂L ∂qj
d dt
∂L
∂ q˙j
Subtitute equation (22) to (21),
dL dt
j
q ˙j d dt
∂L
∂ q˙j
∂L
∂ q˙j
q ¨j (23)
We can easily check that dL dt
j
d dt
q ˙j
∂L
∂ q˙j
so that
d dt
L −
j
q ˙j
∂L
∂ q˙j
We denote the constant by −H
L −
j
q ˙j
∂L
∂ q˙j
= −H (26)
Where ∂L ∂ q˙j
∂(T − V )
∂ q˙j
∂T
∂ q˙j
Then, equation (26) can be written as
T − V −
j
q ˙j
∂T
∂ q˙j
= −H (28)
You can check by yourself^7 that
j q˙j
∂T ∂ q˙j = 2T^ , so T − V − 2 T = −H ⇒ T + V = H (29)
The total energy is a constant of the motion for this case. The function H, called Hamiltonian of the system. We can see that it’s beautiful that simple principle of action sum up the fundamental conservation law.
As I said before, Today, we use the Lagrangian method to describe all of physics, not just me- chanics. All fundamental laws of physics can be expressed in terms of a least action principle. This is true for electromagnetism, special and general relativity, particle physics, and even more specu- lative pursuits that go beyond known laws of physics such as string theory^1. For example, (nearly) every experiment ever performed can be explained by the Lagrangian of the Standard Model
L ∼ R −
Fμν F μν^ + iψγμDμψ + |Dμh|^2 − V (|H|) + hψψ (30)
Dont worry if you dont understand many of the symbols. You are not supposed to. View this equation like art.
Figure 3. Evolution of the ensemble of 1000 systems described by the Hamiltonian^5
In some aspect, we can say that Hamiltonian is better than Lagrangian, also we can say it reversely. In Hamiltonian we have Hamiltonian phase space and Liouville’s theorem, where it represents the time evolution dynamics directly. It seems Lagrangian sound less useful for time evolution. On other hand, Lagrangian actually gives more insight into the symmetries of the system, the Noether’s Theorem.
5 Quantum Is Strange
Previously, we have discussed dynamical of system in classical, and it seems so intuitive and common sense. I hate to say this, that what i told you before, that nature choose path where the action is stationary, is a lie. Don’t be pessimistic quickly. It’s not completely lie, it just our statement is incomplete. Got You!
There’s something less intuitive in least action principle. I should discuss this in section ”Action In Nutshell”, but i suspend it for this section. In Newton’s Law, system moves under the law itself, or we can say it ”cause and effect”. But something wrong when I gave you statement that system choose path where action is stationary. How can system know it the path is stationary? what a decision making! How system evaluate other path? See? something not going right here, as i said, incomplete.
In Quantum Mechanics, particle can behave like wave. So there’s properties like amplitude, phase, and superposition. Particle actually choose all paths! It do not know where the least is. But if we see particle as wave, It goes to all direction! Of course that’s not what we observe. When wave went on a path that took different amount of times, it would arrive at different phase. The total amplitude at some point is superposition of all different ways the light can arrive. All the paths that give wildly different phases dont add up to anything. But if you can find a whole sequence of paths which have phases almost all the same, then the little contributions will add up and you get a reasonable total amplitude to arrive. The important path becomes the one for which there are many nearby paths which give the same phase^2. In Quantum Mechanics, the probability finding particle is the square of amplitude at some point. Waves that superposition each other with same phase, will give highest amplitude among other path, so that’s where particle goes.
It feels incomplete before we put in language of mathematics. Let me give you one more beau- tiful of physics (and mathematics of course, no doubt!), The Feynman Path Integral. Feyman Path
Integral, or simply, path integral is a description of quantum theory that generalizes the action principle of classical mechanics. I will not give you the formulation of Path Integral here, seeing that i still not capable to do that. Briefly, Feynman said that the contribution of amplitude of a path is proportional to A ∝ e−iS/¯h^ (41)
The path integral assigns to all these amplitdues equal weight but varying phase, or argument of the complex number. In macroscopic system, S is very large compared to ¯h. So even a small (but still macroscopic) change in the action for such a particle will cause great changes in the phase. For such particles, many path amplitudes cancel each-other, making them unimportant for the particles way of motion. What we actually discover in reality is that only one certain path is followedthe classical trajectory^8.
Figure 4. Emergence of the classical limit
For deeper understanding we see that the phase is not real function, but complex function. So we cannot see the inteference of this path happens in real space, shortly Euclidean Space. Because Quantum Mechanics lives in Hilbert Space!∗^ Therefore, the overall probability amplitude of the system starting at xa at time ta and ending up at xb at time tb is given by the sum over all possible paths, P =
x
e−iS(x)/h¯^ (42)
If we draw graph of this function with real and imaginary parth separately, it gives
Figure. ∗One of my favorite quotation in physics, what a such astounding statement.
