taking the boundary value asτ→0. This fixes the branch of the square root in
23.12 and one finds (see for instance section 7.6.7 of [107]) that the square root
factor needs to be taken to be
√
ω
i 2 πsin(ωT)
=e−iπ 4
e−inπ 2√
ω
2 π|sin(ωT)|forωT∈[nπ,(n+ 1)π].
23.5 The Bargmann transform
The Stone von-Neumann theorem implies the existence of:
Definition.Bargmann transform
There is a unitary map called the Bargmann transform
B:HS→Fintertwining the Schr ̈odinger representation and the Bargmann-Fock represen-
tation, i.e., with operators satisfying the relation
Γ′S(X) =B−^1 Γ′BF(X)BforX∈h 3.
In practice, knowingBexplicitly is often not needed, since the representation
independent relation
a=1
√
2
(Q+iP)can be used to express operators either purely in terms ofaanda†, which have
a simple expression
a=d
dz, a†=zin the Bargmann-Fock representation, or purely in terms ofQandPwhich have
a simple expression
Q=q, P=−id
dqin the Schr ̈odinger representation.
To compute the Bargmann transform one uses equation 23.11, for non-
normalizable continuous basis states|δu〉, to get
〈q|δu〉=(ω
π)^14
e−u 22
e−ω 2 q 2
e√
2 ωuqand
〈δu|q〉=(ω
π)^14
e−u 22
e−ω 2 q 2
e√
2 ωuq (23.13)