A wavefunction is a Slater determinant iff it's density matrix is a projector in the single-particle Hilbert space .
Theorem (Baranger and Veneroni)
Any density matrix that belongs to a Slater determinant can be decomposed in the following way:
Where and are Hermitian matrices which are even under time reversal.
The above decomposition is unique if:
has only ph and hp matrix elements in the basis in which is diagonal:
( projects onto particle states.)
The eigenvalues of obey:
We are not proving these theorems 🤫.
Some rules for calculating with Slater determinant densities
An arbitrary matrix A has the following "parts" in a basis in which is diagonal:
The three following statements are equivalent:
If two matrices , obey , then:
If , then . For Hermitian matrices , , with vanishing pp and hh elements, we can define these useful vectors:
and find:
Densities of HF-BCS and HFB states
zzz
Wick's theorem
Say we want
Particle-Hole picture
For many-body systems, it is useful to use basis states that are labeled as deviations/excitations from some reference many body Slater determinant (vs states with N creation operators acting on the vacuum):
Occupied single particle states below the Fermi level are hole states, and unoccupied single particle states above it are particle states.
Quasiparticle formalism
Very useful for Hartree-Fock and beyond. We start with a simple redefinition of the creation and annihilation operators, like so (see Notation and more):
Normal ordering w.r.t.
We will use curly braces instead of to distinguish from normal ordering from respect to the vacuum:
Everything else (fundamental contractions, etc) remains the same when swapping in the operators.