Page 149 - Introduction to Colloid and Surface Chemistry
P. 149

The solid-gas interface  139

        versus energy spectrum is produced.  This kinetic energy, £ e|, is given
        in terms of the irradiation frequency,  v, and the ionisation potential,
        /, of the  electron  by Einstein's equation

                 = hv-J                                        (5.14)
             E el
        If it is assumed that photoionisation  occurs without any adjustment of
        the remaining electrons  (Koopman's theorem),  the ionisation poten-
        tial and the orbital energy, c, of the ejected  electron  are  numerically
        equal  (/  =  — c).  A  spectrum  of  electron  orbital  energies  is  thus
        obtained.
          The  ejected  electrons  will interact strongly with  matter because of
        their  charge,  therefore,  when  studying  solids,  the  detection  of
        electrons  as  described  above  will  be  limited  to  those  ejected  from
        atoms at  or  close  to  the surface.
          In  ultraviolet photoelectron  spectroscopy  (UPS  or  U-PES),  the
        irradiation  (usually a  Hc(I)  (21.2  eV)  or  He(II)  (40.8  eV)  source)
        causes the displacement of a valence electron. Although an important
        method  of  studying  the  electronic  nature  of  molecules  in  the  gas
        phase,  it  is less  useful  for  studying the  surfaces of  metals,  since  the
        valence electrons are in a continuous (conducting) energy band with a
        spread  of  about  10 eV.  Adsorbed  layers can,  however,  usefully  be
        investigated  in  terms  of  the  difference  between  the  spectrum
        following  adsorption  and  that for  the  clean metal surface.
          In X-ray  photoelectron spectroscopy  (XPS or  X-PES),  the  irradiation
        (usually a Mg K a  (1253.6 eV) or Al K a (1486.6 eV) source) causes a
        core electron  to be ejected. This is a more useful  technique than UPS
        for  surface  studies,  since  the  binding energies  of  core  electrons  are
        characteristic  of  the  elements  in  question  and  surface  elements  can
        thus  be  identified  by  the  traditional  'spectroscopic  fingerprinting'
        procedure.  In  this  respect,  XPS  is  sometimes  referred  to  by  its
        alternative  name,  electron  spectroscopy  for  chemical  analysis  (ESCA).
        In order to  emphasis  the contribution  from  surface atoms,  the X-ray
        beam is usually set at a grazing angle to the surface. Most of the signal
        originates  from  within a nanometre  of the  surface.
          The  ionisation  potential  of  a  core  electron  depends,  to  a small
        extent,  on  the  chemical  environment  of  the  atom  in  question,  and
        chemical  shifts  of up  to  about  10 eV  can  be observed.  For  example,
        the  C(ls)  XPS signal for  molecular!y  adsorbed  carbon  monoxide on
        polycrystalline  iron  at  290  K  shows  a  peak  at  285.5  eV,  which  is
   144   145   146   147   148   149   150   151   152   153   154