Is this a canonical transformation?

  • Thread starter Thread starter Liquidxlax
  • Start date Start date
  • Tags Tags
    Proof Transform
Click For Summary
SUMMARY

This discussion focuses on the conditions for a transformation to be canonical in phase space, specifically through the change of variables defined by the smooth and invertible function f. The transformation from (q,p) to (Q,P) is established using the relations Pj = (∇f-1)Tjk(f(q))pk. The participants clarify that for the transformation to be canonical, the conditions (dQi/dqj)(q,p) = (dpj/dPi)(Q,P) and (dQi/dpj)(q,p) = -(dqj/dPi)(Q,P) must hold true. The discussion also addresses the role of the transpose in the Jacobian matrix and confirms that the transformation is indeed canonical when the matrix multiplication MJMT = J is satisfied, where J is the symplectic matrix.

PREREQUISITES
  • Understanding of canonical transformations in Hamiltonian mechanics
  • Familiarity with symplectic geometry and the symplectic matrix J
  • Knowledge of gradient and Jacobian matrices, specifically (∇f-1)Tjk
  • Proficiency in calculating Poisson brackets and their properties
NEXT STEPS
  • Study the properties of symplectic matrices and their applications in Hamiltonian mechanics
  • Learn about generating functions in canonical transformations and their explicit forms
  • Explore the implications of Poisson brackets in phase space transformations
  • Investigate the symplectic method and its role in proving canonical transformations
USEFUL FOR

Physicists, mathematicians, and students specializing in classical mechanics, particularly those focusing on Hamiltonian dynamics and phase space analysis.

Liquidxlax
Messages
312
Reaction score
0
Consider the following change of variables in phase space f: maps the reals and is smooth and invertable change of coordinates Q=f(q), q = f-1(Q). Given f, define a change of variables on phase space (q,p) -> (Q,P) by the pair of relations

Pj = (∇f-1)Tjk(f(q))pk

q runs from 1 to n

show that its canonical.
I know that for this to be canonical

(dQi/dqj)(q,p) = (dpj/dPi)(Q,P)

(dQi/dpj)(q,p) = -(dqj/dPi)(Q,P)i'm having a couple problems, is f the generating function that i have to find explicitly?

Can i use the sympletic method such that MJM^T = J

what is the point of the transpose for the (∇f-1)Tjk part for? i thought f had to be symmetric.
 
Physics news on Phys.org
for Pj it can equal

(∇f-1)TjkQpk

=(∇f)T *-1jkqpk

but for that to be true then (∇f-1)Tjk has to be symmetric therefore the transpose dissapears

looking at my notes, i think this is supposed to be the associated point transformation in phase space
 
Last edited:
Guess i can answer my own question... i knew how to do it but i had an error in my notes

[dQ, dP)T = [{dQ/dq, dQ/dp}, {dP/dq, dP/dp}]*[dq, dp] = Mij*[dq, dp]

to prove its a canonical transformation

MJMT = J where J = [{0,I},{-I,0}] and T represents the transposeIf i do the matrix multiplication and say:now i will let {a,b} where a and b are some arbitrary coordinates be the poisson brackets

MJMT = [(0,{Q,P}), (-{Q,P},0)]

It is known that {P,P} = 0 = {Q,Q} and {Q,P} = δij

Using the following vvvv
n7n5k.png


and there you have it
 

Similar threads

Replies
19
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K