Can a Hamiltonian expression, say $H$ with $(q,p)$ as conjugate variable pair, include the total derivative of $p$ explicitly? That is, can we have $H=H(q,p,\dot{p})$?

And, if so, what does it imply in terms of the usual canonical equations? I can see that it will lead to second order derivatives now. Is this permissible in the Hamiltonian formulation, and what does it mean?

Finally, if a transformation $(Q,P,t) \rightarrow (q,p,t)$ is what had led a regular Hamiltonian $K(Q,P,t)$ to become a new one $H(q,p,\dot{p},t)$ with such dependence on $\dot{p}$, as described above, does that mean that such transformation is necessarily not canonical? Or can a regular canonical generating function $F$, for example, do so?


No, the Hamiltonian $H(q,p,t)$ depends by definition not on $\dot{q}$ or $\dot{p}$. Similarly for the Kamiltonian $K(Q,P,t)$.

For Legendre transformations involving $\dot{q}$ or $\dot{p}$, see e.g. this & this Phys.SE posts and links therein.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.