Who can provide me some elegant solution for
[ˆaM,ˆa†N]with[ˆa,ˆa†] = 1
other than brute force calculation?
================================================
O lala!!! Thanks for @Prathyush and @Qmechanic !!! I got the same result with Qmechanic... I think Prathyush's suggestion should be equivalent to the my suggestion of the correspondence up to a canonical transformation. Here is my calculation (I was not confident to post it...)
representation of (ˆa,ˆa†) on polynomial space span{xn√n!}n≥0ˆa[f(x)]=ddxf(x);ˆa†[f(x)]=xf(x);[ˆa,ˆa†][f(x)]=id[f(x)]|0⟩∼1;|n⟩∼xn/√n!
calculate the normal ordering [ˆaM,ˆa†N]:∼[dMdxM,xN]=dMdxM(xN⋆)−xNdMdxM(⋆)∼{min{M,N}∑k=0N!(N−k)!CkM(ˆa†)N−k(ˆa)M−k}−(ˆa†)N(ˆa)M
====================== One comment on 02-12-2012: The representation I was using is actually related to Bergmann representation with the inner product for Hilbert space (polynomials) being:
⟨f(x),g(x)⟩:=∫dxe−x2¯f(x)g(x),x∈R,f,g∈C[x]
Answer
The standard way is to use generating functions (in this case a la coherent states). Usually one would like the resulting formula to be normal-ordered.
Recall the following version eAeB = e[A,B]eBeA
of the Baker-Campbell-Hausdorff formula. The formula (1) holds if the commutator [A,B] commutes with both the operators A and B.Put A=αa and B=βa†, where α,β∈C.
Let [a,a†]=ℏ1, so that the commutator [A,B]=αβℏ1 is a c-number.
Now Taylor-expand the exponential factors in eq. (1).
For fixed orders n,m∈N0, consider terms in eq. (1) proportional to αnβm.
Deduce that the the antinormal-ordered operator an(a†)m can be normal-ordered as an(a†)m = min(n,m)∑k=0n!m!ℏk(n−k)!(m−k)!k!(a†)m−kan−k.
Finally, deduce that the normal-ordered commutator is [an,(a†)m] = min(n,m)∑k=1n!m!ℏk(n−k)!(m−k)!k!(a†)m−kan−k.
No comments:
Post a Comment