QM Slides Eng 03

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

Mathematical Foundations

of Quantum Mechanics

04. Quantization

Tim Shilkin

Saarland University, Spring 2023

www.pdmi.ras.ru/∼shilkin

 web-site containing some materials for our course


Ÿ1. Correspondence principle
and quantization rules

1. Bohr's correspondence principle

Axiom 4. Non-relativistic theory of interaction of a quantum particle


with a classical eld must contain a small parameter
~ (which is related
to the size of the support of the wave function ψ ∈ L2 (R3 , C) in the
case of coordinate respesentation), and when ~ → +0 all relations
for the dynamical characteristics of a quantum particle (coordinates,
momentum, energy, angular momentum), when interpreted correctly,
should transform into the relations of classical mechanics for the
same characteristics of a classical particle.

Axiom 0  probabilistic nature of experiment results


Axiom 1  states
Axiom 2  observables
Axiom 3  probability measure on the spectrum
2. Equations of motion of a classical particle in a potential eld

In classical mechanics the motion of a particle in a force eld with


potential v : R3 → R is dened by Newton's second law:

d2 x
m 2 (t) = −∇U (x(t)),
dt
where U is the potential energy of a particle in a eld with potential v .

For a detrministic description of the dynamics we x the initial data:

dx
x(0) = x0, m (0) = p0
dt
The set of parameters (x0, p0) is the 6-dimensional Euclidean space

R3 × R3
called phase space of a classical particle.
3. Observables in classical mechanics

The motion of a classical particle determines its dynamic characteristics


(coordinate, momentum, energy, angular momentum) in the unique
way. Thus, in classical mechanics the dynamic characteristics of a
particle (i.e. classical observables ) are functions f : R3 × R3 → R of
the coordinate and momentum:

• f (x, p) = x  coordinate
• f (x, p) = p  momentum
|p|2
• f (x, p) = 2m  kinetic energy
• f (x, p) = U (x)  potential energy
|p|2
• f (x, p) = 2m + U (x)  total energy
• f (x, p) = (x − x∗) × p  angular momentum with respect to x∗

Def. Observables for a classical particle are smooth functions on


the phase space. (Compare: observables for a quantum particle
are self-adjoint linear operators on the state space).
4. Hamiltonian of a classical particle

Def. The observable


|p|2
H : R3 × R3 → R, H(x, p) := + U (x),
2m
corresponding to the total energy of a classical particle is called the
Hamiltonian of a classical particle.

5. Equations of motion in Hamilton form

| p| 2
Theorem. Assume H(x, p) = 2m + U (x) is the Hamiltonian of a
classical particle. Then

dx
(t) = ∇pH(x(t), p(t))


Newton's

dt

⇐⇒
second law  dp


 (t) = −∇xH(x(t), p(t))
dt
6. How do the physical characteristics of a particle evolve in time
in classical mechanics?

Theorem. Let F (t) be the values of some physical characteristic of a


classical particle, i.e. F (t) = f (x(t), p(t)), where f : R3 × R3 → R is
a classical observable. Then
d
f (x(t), p(t)) = {H, f }
dt x=x(t),p=p(t)
where we have denoted by {H, f } the expression

{H, f } := ∇pH · ∇xf − ∇xH · ∇pf

Proof. d f (x(t), p(t)) = ∇ f · ẋ +∇ H · ṗ


dt x |{z} p |{z}
∇p H −∇x H

7. Classical Poisson bracket

Def. Classic Poisson bracket is the map

{·, ·} : C ∞(R3 × R3) × C ∞(R3 × R3) → C ∞(R3 × R3),


{f, g} := ∇pf · ∇xg − ∇xf · ∇pg
8. Properties of the classical Poisson bracket

Theorem. Let {·, ·} be the classical Poisson bracket on C ∞(R3 × R3).


Then for any f , g , h ∈ C ∞(R3 × R3) and any λ ∈ R

• {f, g} = −{g, f } (skew-symmetry)


• {f + g, h} = {f, h} + {g, h} (additivity)
• {λf, g} = λ{f, g} (homogeneity)
• {f, {g, h}} + {g, {h, f }} + {h, {f, g}} = 0 (Jacobi identity)

9. Properties of the commutator

Theorem. The commutator [A, B] of linear operators in a Hilbert


space H has the following properties:

• [A, B] = −[B, A] (skew-symmetry)


• [A + B, C] = [A, C] + [B, C] (additivity)
• [λA, B] = λ[A, B] (homogeneity)
• [A, [B, C]] + [B, [C, A]] + [C, [A, B]] = 0 (Jacobi identity)
10. Dirac's canonical quantization

Def. Dirac's canonical quantization is a map

πh : C ∞(R3 × R3) → { A : D(A) ⊂ H → H, A = A∗ }


algebra of classical observables into a set of self-adjoint linear operators
on a Hilbert space such that

• πh(f + g) = πh(f ) + πh(g)


• πh(λf ) = λπh(f ), ∀ λ∈R
• πh(0) = O, πh(1) = I
• πh({f, g}) = ~i [πh(f ), πh(g)]  Bohr's correspondence principle

Axiom 5. Quantum observables are obtained from the corres-


ponding classical observables by applying to the latter the quan-
tization operator πh.
11. Diculties related to the denition of the quantization

The product of self-adjoint operators is not necessarily a self-adjoint


operator.

So, the Dirac canonical quantization should be supplemented with a


rule according to which the product of two classical observables f and
g is associated with a self-adjoint operator.

For example, the product of functions f g is assigned Jordan product


of operators, i.e. instead of matching

fg 7→ πh(f )πh(g)  non self-adjoint

we use the matching

1 (formally)
 
fg 7→ πh(f )πh(g) + πh(g)πh(f ) 
2 self-adjoint
However, such matching is convenient only in the case of bounded
operators. For unbounded operators this denition is formal, since
unbounded operators must be carefully dealt with domains of the
product and sum of operators.
Therefore, in order to get a working quantum theory, it is necessary
to drop the requirement that the quantization πh is an isomorphism of
algebras of classical and quantum observables. Instead, we require that
the mapping f 7→ πh(f ) becomes an isomorphism of the corresponding
algebras only in the limit as h → 0.

12. General denition of quantization

Def. A quantization of a classical system is an invertible map

πh : C ∞(R3 × R3) → { A : D(A) ⊂ H → H, A∗ = A }


of the algebra of classical observables into a set of self-adjoint linear
operators on a Hilbert space such that

 
−1
1) lim 1 π πh(f )πh(g) + πh(g)πh(f ) = f g
h→0 2 h

 
2) lim πh−1 ~i [πh(f ), πh(g)] = {f, g}  Bohr's correspondence principle
h→0
Ÿ2. Ñommutation relations
of quantum mechanics

1. Commutation relations of classical mechanics

Theorem. Classical observables of coordinates Xj ∈ C ∞(R3; R3) and


momentums Pj ∈ C ∞(R3; R3)

X1(x, p) = x1, X2(x, p) = x2, X3(x, p) = x3,


P1(x, p) = p1, P2(x, p) = p2, P3(x, p) = p3,
satisfy the commutation relations

{Xk , Xj } = 0, {Pk , Pj } = 0, {Xk , Pj } = −δkj , k, j = 1, 2, 3

Proof: {f, g} := ∇pf · ∇xg − ∇xf · ∇pg


       
0 0 1 1
{X1, P1} =  0  ·  0  −  0  ·  0  = −1 
       
0 0 0 0
2. Commutation relations of quantum mechanics

Theorem. Observables of coordinates Qj : H → H and momenta


Pj : H → H in quantum mechanics, obtained from the corresponding
classical observables by the canonical quantization, must satisfy the
commutation relations

[Qk , Qj ] = 0, [Pk , Pj ] = 0, [Qk , Pj ] = i~δkj I, k, j = 1, 2, 3 (∗)

Def. Operator relations (∗) are called the canonical commutation


relations of quantum mechanics.

3. The impossibility of constructing quantum mechanics on the


basis of the algebra of bounded operators

Theorem. If self-adjoint operators Q and P satisfy the relation

QP − P Q = i I
then Q and P can not be bounded simultaneously.
Proof. By contradiction. Let A, B be bounded and satisfy

AB − BA = i I
Then

∀ n∈N AnB − BAn = i n An−1


Base of induction: n = 1  by assumption.

Inductional step: Assume AnB − BAn = i n An−1. Then

An+1B − BAn+1 = An+1B−ABAn + ABAn − BAn+1 =


= A |(AnB −
{z
BA n ) + (AB − BA) An = i(n + 1)An
} | {z }
= inAn−1 = iI
As A = A∗ , we have kAn k = kAkn and hence

A=0
n kAkn−1 ≤ 2 kAknkBk =⇒ (?!!)
∀ n∈N n ≤ 2 kAk kBk


You might also like