84
4 Measurement of Work Function
electron beam, or ion beam. Examples of the secondary electron cutoff region of
XPS spectra for polycrystalline Cu and Pt are shown in Fig. 4.15b [10]. The position
of the secondary electron cutoff for Cu is located at a lower kinetic energy (a higher
binding energy) than that for Pt, meaning that the work function of Cu is smaller than
that of Pt. Because the energy difference of the secondary electron cutoff position
indicates the absolute difference in the work function, the relative work function
can be obtained from any secondary electron cutoff spectrum. Note that a low work
function does not mean a higher secondary electron emission current, as demonstrated
in Fig. 4.16. In the figure, inner-shell photoelectron intensity images, a secondary
electron image, and an emission image at the low electron kinetic energy (work
function image, taken at the energy marked at the bottom left spectrum in Fig. 4.16)
measured on mesh-shaped patterns of Cu and Pt are shown. The patterns of the
secondary electron image and the work function image are opposite, i.e., higher
secondary electron emission from Pt, but a higher intensity for electron with low
kinetic energy from Cu [10].
This technique is also applicable to the work function measurement of semiconductors that do not charge up (i.e., they have relatively high electric conductivity).
The principle of the measurement is illustrated in Fig. 4.17. Because there are no
electrons in the Fermi level in this case, the energy difference (E MAX − E MIN ) is
equal not to hν − φ but to hν − φ − (E F − E V ), where (E F − E V )(≡ E V B ) can be
obtained from the binding energy of electrons with the highest kinetic energy. From
experimentally obtained values for (E MAX − E MIN ) and E V B , the work function of
semiconductors is obtained as follows.
φ = hν − (E MAX − E MIN ) − E V B
(4.12)
Figure 4.18 demonstrates an example of the work function measurement of ZnO.
In this example, hν = 21.22 eV, (E MAX − E MIN ) = 14.39, and E V B = 2.75, which
give φ = 4.08 eV.
The above method cannot be applied to bulk insulators, which charge up. However,
in the special case that a very thin insulator film is on a metallic substrate and does
not charge up, the work function of the insulator film can be measured by the same
principle as that discussed for a semiconductor. The example of an ultrathin epitaxial
alumina film on an NiAl(110) single crystal is shown in Fig. 4.19. The bottom
spectrum is for clean NiAl(110), and the change in the spectrum upon the oxidation
of NiAl(110) with the growth of an epitaxial alumina film can be seen, where unit
of oxygen dosage is L (Langmuir, 1 [L] = 1.3 × 10
−4 [Pa] × 1 [s]). The secondary
electron cutoff position shifts towards a higher binding energy, indicating a decrease
in the work function. This type of measurement is possible when an insulator film
is sufficiently thin for electrons generated inside the insulator to be able to tunnel
either to the vacuum or to the substrate.
Here are some tips for the measurement of the secondary electron cutoff position.
Because secondary electrons have very low kinetic energy, they are easily affected by
weak electric and magnetic fields. For most commercially available equipment for
electron spectroscopy, the magnetic field is shielded, which is not usually a problem.
4 Measurement of Work Function
electron beam, or ion beam. Examples of the secondary electron cutoff region of
XPS spectra for polycrystalline Cu and Pt are shown in Fig. 4.15b [10]. The position
of the secondary electron cutoff for Cu is located at a lower kinetic energy (a higher
binding energy) than that for Pt, meaning that the work function of Cu is smaller than
that of Pt. Because the energy difference of the secondary electron cutoff position
indicates the absolute difference in the work function, the relative work function
can be obtained from any secondary electron cutoff spectrum. Note that a low work
function does not mean a higher secondary electron emission current, as demonstrated
in Fig. 4.16. In the figure, inner-shell photoelectron intensity images, a secondary
electron image, and an emission image at the low electron kinetic energy (work
function image, taken at the energy marked at the bottom left spectrum in Fig. 4.16)
measured on mesh-shaped patterns of Cu and Pt are shown. The patterns of the
secondary electron image and the work function image are opposite, i.e., higher
secondary electron emission from Pt, but a higher intensity for electron with low
kinetic energy from Cu [10].
This technique is also applicable to the work function measurement of semiconductors that do not charge up (i.e., they have relatively high electric conductivity).
The principle of the measurement is illustrated in Fig. 4.17. Because there are no
electrons in the Fermi level in this case, the energy difference (E MAX − E MIN ) is
equal not to hν − φ but to hν − φ − (E F − E V ), where (E F − E V )(≡ E V B ) can be
obtained from the binding energy of electrons with the highest kinetic energy. From
experimentally obtained values for (E MAX − E MIN ) and E V B , the work function of
semiconductors is obtained as follows.
φ = hν − (E MAX − E MIN ) − E V B
(4.12)
Figure 4.18 demonstrates an example of the work function measurement of ZnO.
In this example, hν = 21.22 eV, (E MAX − E MIN ) = 14.39, and E V B = 2.75, which
give φ = 4.08 eV.
The above method cannot be applied to bulk insulators, which charge up. However,
in the special case that a very thin insulator film is on a metallic substrate and does
not charge up, the work function of the insulator film can be measured by the same
principle as that discussed for a semiconductor. The example of an ultrathin epitaxial
alumina film on an NiAl(110) single crystal is shown in Fig. 4.19. The bottom
spectrum is for clean NiAl(110), and the change in the spectrum upon the oxidation
of NiAl(110) with the growth of an epitaxial alumina film can be seen, where unit
of oxygen dosage is L (Langmuir, 1 [L] = 1.3 × 10
−4 [Pa] × 1 [s]). The secondary
electron cutoff position shifts towards a higher binding energy, indicating a decrease
in the work function. This type of measurement is possible when an insulator film
is sufficiently thin for electrons generated inside the insulator to be able to tunnel
either to the vacuum or to the substrate.
Here are some tips for the measurement of the secondary electron cutoff position.
Because secondary electrons have very low kinetic energy, they are easily affected by
weak electric and magnetic fields. For most commercially available equipment for
electron spectroscopy, the magnetic field is shielded, which is not usually a problem.