Phil Lucht Math & Physics Archive
Home / Math and Physics Files / Physics / E&M / junk bin

scratch1

DOCX · 19.7 KB
Open DOCX file

A short scratch document, apparently Phil's own working note from an E&M junk bin folder, with the author calling the argument a hokey proof. It writes a general vector function as three scalar angular functions, expands each in Y_lm, and uses the gradient and r-cross-gradient relations for each partial wave. It then concludes the resulting coefficients still depend on angles, a big so what, and the text cuts off. Greek symbols and arguments are lost in extraction.

AI-written summary; may contain errors. This description is approximate.

Extracted text (machine-read; may contain errors)
Scratch stuff I will try to give a hokey proof right here. Certainly for a particular , we know that rYm(,) = am(,) + bm(,) I could compute a and b by doing the gradient, but let's not bother. Then we know that rxYm(,) = - bm(,) + am(,) We solve to get [ where cm(,) am(,)2 + bm(,)2] = - (am(,)/cm(,)) rYm(,) + (bm(,)/cm(,)) rxYm(,) = (bm(,)/cm(,))rYm(,) + (am(,)/cm(,)) rxYm(,) Rewrite as = - a'm(,) rYm(,) + b'm(,) rxYm(,) = b'm(,) rYm(,) + a'm(,) rxYm(,) Notice that the above is true for each m except when =0 which we don't include in things normally. Now, the most general vector function can be written: A(,) = e(,) + f(,) + g(,) where all functions are dimensionless and do not include any r factors. At this point, we could expand each of the three scalar functions onto the Ym like this e(,) = em Ym(,) em = d Ym*(,) e(,) Thus, A(,) = em Ym(,) + fm Ym(,) +gm Ym(,) Now comes the trick. We can use the expressions shown above for and with each partial wave, sort of custom to that partial wave. Then we have: A(,) = em Ym(,) + fm Ym(,) { - a'm(,)rYm(,) + b'm(,)rxYm(,) } +gm Ym(,) { b'm(,)rYm(,) + a'm(,)rxYm(,) } Now just group the two vector quantities together to get A(,) = em Ym(,) + Ym(,) { - fm a'm(,) + gm b'm(,) } rYm(,) Ym(,){ fm b'm(,) + gm a'm(,) } rxYm(,) Rewrite as, A(,) = em Ym(,) + Ym(,) { cm(,) } rYm(,) Ym(,){ dm(,) } rxYm(,) Now expand each of these c and d coefficients onto the Ym(,), A(,) = em Ym(,) + Ym(,) { C'm' Y'm' (,)} rYm(,) Ym(,){ D'm' Y'm' (,)} rxYm(,) And after all this work, we arrive at a big "so what!" The coefficients of the basis functions are now functions of and , but that is not what we want to be claiming. So we now have shown that A(,) = em