Page 335 - PRINCIPLES OF QUANTUM MECHANICS as Applied to Chemistry and Chemical Physics
P. 335

326                             Appendix G

                                                   ø lm (è, j) ˆ c lm P jmj (cos è)e imj       (G:40)
                                                                   l
                             where c lm are the normalization constants. A comparison of equation (G.40) with
                             (5.59) shows that the functions ø lm (è, j) are the spherical harmonics Y lm (è, j).



                             Radial equation for the hydrogen-like atom
                             The radial differential equation for the hydrogen-like atom is given by equation (6.24)
                             as
                                                 2
                                                d S  2 dS     1   ë   l(l ‡ 1)
                                                   ‡     ‡ÿ ‡       ÿ        S ˆ 0             (G:41)
                                                dr 2  r dr    4   r     r 2
                             where l is a positive integer. If a power series solution is applied directly to equation
                             (G.41), the resulting recursion relation involves a k‡2 , a k‡1 , and a k . Since such a three-
                             term recursion relation is dif®cult to handle, we ®rst examine the asymptotic behavior
                             of S(r). For large values of r, the terms in r ÿ1  and r ÿ2  become negligible and
                             equation (G.41) reduces to
                                                              2
                                                             d S   S
                                                                 ˆ
                                                             dr 2  4
                             or
                                                            S ˆ ce  r=2
                             where c is the integration constant. Since r, as de®ned in equation (6.22), is always
                             real and positive for E < 0, the function e r=2  is not well-behaved, but e ÿr=2  is.
                             Therefore, we let S(r) take the form
                                                         S(r) ˆ F(r)e ÿr=2                     (G:42)
                               Substitution of equation (G.42) into (G.41) yields
                                             2
                                            r F 0 ‡ r(2 ÿ r)F9 ‡ [(ë ÿ 1)r ÿ l(l ‡ 1)]F ˆ 0    (G:43)
                                                             2 r=2
                             where we have multiplied through by r e  . To solve this differential equation by the
                             series solution method, we substitute equations (G.2), (G.3), and (G.4) for F, F9, and
                             F 0 to obtain
                                 1                                    1
                                X                                k‡s  X                  k‡s‡1
                                    a k [(k ‡ s)(k ‡ s ‡ 1) ÿ l(l ‡ 1)]r  ‡  a k (ë ÿ 1 ÿ k ÿ s)r  ˆ 0
                                 kˆ0                                  kˆ0
                                                                                               (G:44)
                                                                        s
                             The indicial equation is given by the coef®cient of r as
                                                      a 0 [s(s ‡ 1) ÿ l(l ‡ 1)] ˆ 0            (G:45)
                             with solutions s ˆ l and s ˆÿ(l ‡ 1). For the case s ˆÿ(l ‡ 1), we have
                                               F(r) ˆ a 0 r ÿ(l‡1)  ‡ a 1 r ÿl  ‡ a 2 r ÿl‡1  ‡      (G:46)
                                                    4
                             which diverges at the origin. Thus, the case s ˆ l is the only acceptable solution.
                               Omitting the vanishing ®rst term in the ®rst summation on the left-hand side of
                             (G.44) and replacing k by k ‡ 1 in that summation, we have

                             4  The reason for rejecting the solution s ˆÿ(l ‡ 1) is actually more complicated for states with l ˆ 0.
                              I. N. Levine (1991) Quantum Chemistry, 4th edition (Prentice-Hall, Englewood Cliffs, NJ), p. 124,
                              summarizes the arguments with references to more detailed discussions. The complications here strengthen
                              the reasons for preferring the ladder operator technique used in the main text.
   330   331   332   333   334   335   336   337   338   339   340