Page 412 - Academic Press Encyclopedia of Physical Science and Technology 3rd Analytical Chemistry
P. 412

P1: GLQ Final pages
 Encyclopedia of Physical Science and Technology  EN012C-568  July 26, 2001  15:32






               62                                                                               Photoelectron Spectroscopy


               with an average value of only a few angstroms. Even for a
               kinetic energy of 10 or 1000 eV, the escape depth is only
                              ˚
               on the order of 20 A. PES probes only the few outermost
               atomic layers of a solid, which can be a disadvantage if
               one wants to study the bulk material. First, the composi-
               tion of the surface is often different from the composition
               of the bulk because of segregation effects or surface con-
               tamination (Section III.A). Even if there is no difference
               in composition, there is usually a strong contribution from
               the outermost layer, especially for kinetic energies around
               100 eV. The outermost layer is chemically always different
               from the interior, since the atoms in this layer have fewer
               neighbors. The surface sensitivity of PES is advantageous,
                                                                 FIGURE 5 Reference schemes for solid samples: (a) conducting
               however, if we want to study the surface itself. Therefore,  sample; (b) insulating sample with reference material on top. S,
               PES has become one of the most powerful tools in surface  Sample; Sp, spectrometer; R, reference.
               science. It allows us to study not only a surface, but also
               atoms or molecules sitting at the surface. Nowadays it is
               possible to detect coverages down to a fraction of a mono-  In the case of solids it is necessary to distinguish be-
               layer. Thus, PES is extremely useful for the investigation  tween conductors and insulators. If the sample is a con-
               of adsorbates.                                    ductor and in electrical contact with the spectrometer, the
                 Another special aspect of solids is the “reference prob-  Fermi levels E F  equilibrate (Fig. 5a). The same is true
               lem.” For an atom or molecule in the gas phase, ionization  for any metal that is used to calibrate the binding energy
               leads to the creation of an electron and a positive ion. The  scale. The binding energy E S B  of an arbitrary conduct-
               electron is either detected or lost at the walls of the instru-  ing sample can therefore be referred to the Fermi level
               ment. The ion also leaves the ionization region rapidly. By  of the spectrometer, which is the reference level used in
                                                                                           S
               calibration with accurately known binding energies (see  most investigations. To refer E to the vacuum level of
                                                                                           B
               Table I), the binding energies of the sample can be re-  the sample, the work functions of the reference material
               ferred to the vacuum level that corresponds to an infinite  and sample must be known.
               separation of electron and ion.                     The situation is more difficult for insulating samples.
                                                                 Photoionization creates positive charges within the sam-
                                                                 ple that are not equilibrated immediately, and the sample
               TABLE I Useful Calibration Lines
                                                                 becomes charged. At the same time there is usually a rel-
               Atom      Level    Compound/phase    Energy (eV) a  atively high density of low-energy electrons close to the
                                                                 sample surface, which can neutralize the positive charges.
                 Ne      1s            Gas            870.37
                                                                 The equilibrium between outgoing and incoming elec-
                 F       1s          CF 4 /gas        695.52
                                                                 trons depends on the measuring conditions, specifically
                 O       1s          CO 2 /gas        541.28
                                                                 on the intensity of the ionizing radiation and the clean-
                 N       1s           N 2 /gas        409.93
                                                                 ness of the surrounding metal parts. Therefore, the actual
                 C       1s          CO 2 /gas        297.69
                                                                 charging potential φ ch  (Fig. 5b) depends on the measuring
                 Ar      2p 3/2        Gas            248.62
                                                                 conditions. The charging may not even be homogeneous
                 Kr      3p 3/2        Gas            214.55
                                                                 over the surface area investigated (differential charging),
                 Kr      3d 5/2        Gas             93.80
                                                                 resulting in a broadening of the observed lines. Sam-
                 Ne      2s            Gas             48.47
                                                                 ple charging can be reduced by use of very thin sam-
                 Ne      2p            Gas             21.59
                                                                 ples or a separate source of low-energy electrons (flood
                 Ar      3p            Gas             15.81
                                                                 gun). Alternatively, sample charging can be taken into ac-
                 Cu      2p 3/2       Metal           932.8
                                                                 count by depositing small amounts of a reference ma-
                 Ag      3p 3/2       Metal           573.0
                                                                 terial (usually gold) onto the sample surface or by us-
                 Ag      3d 5/2       Metal           368.2
                                                                 ing the carbon that is found on nearly every surface as a
                 Cu      3s           Metal           122.9
                                                                 reference material. Assuming that the reference material
                 Au      4f 7/2       Metal            83.8
                                                                 and the sample are at the same potential in the irradi-
                 Pt      4f 7/2       Metal            71.0
                                                                 ated area (Fig. 5b), the binding energies for the sample
                 a                                               are then referred to the binding energies of the reference
                  For the metals the energies refer to the Fermi level instead of the
               vacuum level.                                     material.
   407   408   409   410   411   412   413   414   415   416   417