Page 154 - Introduction to Colloid and Surface Chemistry
P. 154

The solid-gas  interface  143

          AES  can  be  used for  studying specific crystal  faces.  For  example,
        nitrogen  chemisorption on iron  has been  shown to be much  stronger
        on  the  more  open Fe(lll)  crystal  faces  than  on  the  more  compact
        Fe(lOO)  and  Fe(llO)  faces  (see  Figure  5.16),

        Low energy electron diffraction (LEED)

        We now turn from  the study of surface composition to that of surface
        structure.  Low energy electron  diffraction  is the  basic technique  for
        studying  the  arrangement  of  atoms  at  solid  surfaces,  just  as X-ray
        diffraction  is the  basic technique for  studying the  three-dimensional
        arrangements  of  atoms  in  crystalline solids.  The  electron  energy
        range  exploited  in  LEED  is roughly 20-300 eV.  In this range, two
        basic,  but  conflicting,  requirements are  satisfied.

        1.  The  electron  energy is high enough  for the  de Broglie wavelength
           (A  =  hip  — h/(2m  e)^)  to  be  of  the  same  order  of  magnitude  as
                                                               10
           interatomic  spacings  (e.g.  c =  150 eV corresponds  to A =  10~  m).
        2.  The  energy  is  low  enough  to  ensure  that  most  of  the  electrons
           (incident  at  right-angles  to  the  surface)  do  not  penetrate  much
           beyond the outer atomic layer of the solid, thus giving a technique
           appropriate  for surface  studies.

          Crystalline solids consist  of periodically repeating  arrays of  atoms,
        ions  or  molecules.  Many  catalytic  metals  adopt  cubic close-packed
        (also called face-centred  cubic) (Co,  Ni, Cu, Pd, Ag, Pt) or hexagonal
        close-packed  (Ti,  Co,  Zn)  structures.  Others  (e.g.  Fe, W) adopt  the
        slightly  less  efficiently  packed  body-centred  cubic  structure.  The
        different  crystal faces which are  possible  are  conveniently  described
        in  terms  of  their  Miller  indices.  It  is  customary  to  describe  the
        geometry  of a crystal in terms of its unit cell. This is a  parallelepiped
        of characteristic  shape  which generates  the  crystal  lattice when many
        of  them  are  packed together.
          In  two dimensions,  the  equivalents of unit cell and  lattice  are  unit
        mesh  and  net,  respectively.  Crystallography  in  two  dimensions  is,
        obviously,  simpler  than that in three dimensions,  and there  are only
        five types  of net  (illustrated  in Figure  5.15). The  choice  of unit  mesh
        is  arbitrary.  The  primitive  unit  mesh  (illustrated  at  the  bottom  left
        hand corner  of each  net)  is the  smallest  possible repeating  quadrilateral
        with  lattice points only at  the  corners.  However,  it may be  appropriate
   149   150   151   152   153   154   155   156   157   158   159