Page 297 - PRINCIPLES OF QUANTUM MECHANICS as Applied to Chemistry and Chemical Physics
P. 297

288                              Appendix B

                                                            a n ÿ ib n
                                                        c n        ,   n . 0
                                                               2
                                                            a n ‡ ib n
                                                      c ÿn         ,   n . 0                    (B:9)
                                                               2
                                                            a 0
                                                        c 0
                                                             2
                             The coef®cients c n in equation (B.8) may be obtained from (B.9) with a n and b n given
                             by (B.2). The result is
                                                              … ð
                                                            1        ÿinè
                                                       c n ˆ     f (è)e  dè                    (B:10)
                                                           2ð  ÿð
                             which applies to all values of n, positive and negative, including n ˆ 0. We note in

                             passing that c ÿn is the complex conjugate c of c n .
                                                                 n
                               In terms of the distance variable x, equations (B.8) and (B.10) become
                                                                1
                                                               X
                                                        f (x) ˆ    c n e inðx=l                (B:11)
                                                              nˆÿ1
                                                           1  … l   ÿinðx=l
                                                      c n ˆ     f (x)e    dx                   (B:12)
                                                          2l  ÿl
                             while in terms of the time t, they take the form
                                                                1
                                                               X      inùt
                                                         f (t) ˆ    c n e                      (B:13)
                                                               nˆÿ1
                                                          ù  … ð=ù    ÿinùt
                                                      c n ˆ       f (t)e  dt                   (B:14)
                                                          2ð  ÿð=ù




                             Parseval's theorem
                             We now investigate the relation between the average of the square of f (è) and the
                             coef®cients in the Fourier series for f (è). For this purpose we select the Fourier series
                             (B.8), although any of the other expansions would serve as well. In this case the
                                           2
                             average of jf (è)j over the interval ÿð < è < ð is
                                                            … ð
                                                          1     jf (è)j dè
                                                                    2
                                                          2ð  ÿð
                               The square of the absolute value of f (è) in equation (B.8) is
                                                                2
                                                      1             1    1
                                                     X             X    X
                                                2                                i(nÿm)è
                                           jf (è)j ˆ     c n e inè    ˆ     c c n e            (B:15)
                                                                             m

                                                    nˆÿ1          mˆÿ1 nˆÿ1
                             where the two independent summations have been assigned different dummy indices.
                             Integration of both sides of equation (B.15) over è from ÿð to ð gives
                                              ð             X    X        ð
                                             …                           …
                                                                  1
                                                            1
                                                     2                       i(nÿm)è
                                                jf (è)j dè ˆ         c c n  e      dè
                                                                      m
                                              ÿð           mˆÿ1 nˆÿ1      ÿð
                             Since m and n are integers, the integral on the right-hand side vanishes except when
                             m ˆ n, so that we have
   292   293   294   295   296   297   298   299   300   301   302