Calculation a reduced matrix element using E-Wigner Th.

2 replies · 2K views · last reply
squareroot
Messages
76
Reaction score
0
Hello.

I fail to follow one step in the process of calculating ⟨la∥Y(L)∥lb⟩ .

The spherical harmonics Yma(L)(r) represent the 2L+1 components of the spherical tensor of rank L. Writing the Eckart-Wigner th. for M = 0 yields:
Equation: <la||ma>... Y(L)... Wigner 3j symbol with rows la L lb and -ma 0 mb, times reduced matrix element <la||Y(L)||lb>, prefactor (-1) to the power la - ma.
(1)

Also one can write
Equation: <la||Y_0^(L)||lb> with ma and mb inside the bra and ket, equals the integral over the solid angle of Y_ma^(la)* Y_0^(L) Y_mb^(lb) evaluated at the direction vector.
(2)

Coupling L and lb to l:
Equation: Y_0^(L)(r) Y_{m_b}^{(l_b)}(r) equals the sum over l and m of sqrt of (2L+1)(2l_b+1)(2l+1) over 4 pi, times the 3j symbol (L l l_b; 0 m m_b), times the 3j symbol (L l l_b; 0 0 0), times Y_m^{(l)*}(r).
(3)

Thus having
Equation: <la ma | Y^(L)_0 | lb mb> equals a sum over l and m of sqrt((2L+1)(2lb+1)(2l+1)/(4π)) times a 3j symbol (L l lb; 0 m mb) times a 3j symbol (L l lb; 0 0 0) times the integral of Y*(la, ma) Y(l, m) dΩ.
(4)

Now solving the integral:
Two integral identities: the integral over solid angle of Y(la,ma)* Y(l,m)* Y(l,m) equals (-1)^m times the integral of Y(la,ma)* Y(l,-m)* Y(l,m), which equals (-1)^m times delta(la,l) delta(ma,-m).
(5)

So:

Equation: <la ma|Y_0^(L)|lb mb> = (-1)^(-ma) times the square root of (2L+1)(2la+1)(2lb+1)/(4 pi) times the 3j symbol (L la lb; 0 -ma mb) times the 3j symbol (L la lb; 0 0 0).
(6)

Here is my problem! After solving the integral (5) and replacing it into (4) I don't understand how that changes the Wigner 3j symbols from
Two 2x3 matrices side by side. Left: top row (L, l, l_b) over bottom row (0, m, m_b). Right: top row (L, l, l_b) over bottom row (0, 0, 0).
(3) into
Matrix product of a row vector (l_a, L, l_b) with a 2x3 matrix of entries (-m_a, 0, m_b) and a row (0, 0, 0)
(6)

Could anyone please help me with this step? I m guessing it has something to do with does kronecker deltas from solving the integral and they act on the wigner symbols after substitution... but i have no idea how!
 
In equation (4), the right hand side is summed over ##l## and ##m##. However upon inserting the result from equation (5), only the term with ##l=l_a## and ##m=-m_a## survives. Yes it has to do with the property of Dirac delta symbol which is for any pair of integer ##i## and ##j##, the Dirac delta ##\delta_{ij}## will give zero if ##i\neq j##. If ##i=j##, ##\delta_{ij} = 1##.