Let be the normalized ground state for
Using it as a trial state for , the variational principle gives
The pointwise assumption makes the last expectation nonpositive. Hence
Use the normalized scaled trial state
If
then scaling derivatives and changing variables give
If a finite ground state existed, would minimize this expression, and hence
It would follow that
for . But a bound state of an attractive potential tending to zero at infinity must have negative energy. This is a contradiction.
Indeed, the scaling gives the stronger collapse statement
when . Thus the ideal singular Hamiltonian is not bounded below and has no finite ground bound state.
In the basis ,
Writing , its exact eigenvalues are
For the normalized trial state,
The minimum of is with . Therefore
The variational estimate is exact because the trial family ranges over all real normalized states in the two-dimensional Hilbert space, and the real symmetric Hamiltonian has a real ground eigenvector.

Articles by others on the same topic (0)

There are currently no matching articles.