20

Suppose we have a Hamiltonian that depends on various real parameters. When tuning the values of these parameters, the energy eigenvalues will often avoid crossing each other. Why?

Is there a physically intuitive justification for level repulsion and avoided crossings? It would be nice to see a general argument.

ChickenGod
  • 2,155
  • The question (v1) is essentially a duplicate of http://physics.stackexchange.com/q/32041/2451 – Qmechanic Apr 08 '13 at 08:45
  • @Qmechanic Actually, the question you linked is what inspired my question. Perhaps I should have added more detail in my question, but I was thinking Adiabatic Theorem, not perturbation theory. I will edit my question correspondingly. – ChickenGod Apr 08 '13 at 10:59
  • Concerning the Adiabatic Theorem, see also e.g. Wikipedia. – Qmechanic Apr 08 '13 at 11:17

1 Answers1

9

Consider what happens if there is a crossing. A crossing would imply a degeneracy in the system. A degeneracy would imply a symmetry. It would be unnatural for a perturbation to introduce a symmetry into a system, and so the eigenvalues cannot cross generically, but can under special circumstances.

alemi
  • 13,521
  • 1
    Only upto accidental degeneracies. For e.g. it happens in H-atom due to circular symmetric Coulomb potential V(r) – L.K. Apr 16 '19 at 14:29
  • Can you show it implies a symmetry? How does the symmetry act? – Marten Jun 29 '22 at 16:58