XPS Study of Alkali Graphite Intercalation Compounds

  • PDF / 474,301 Bytes
  • 9 Pages / 420.48 x 639 pts Page_size
  • 115 Downloads / 284 Views

DOWNLOAD

REPORT


G. K. WERTHEIM AND S. B. DICENZO Bell Laboratories, 600 Mountain Avenue, Murray Hill, New Jersey 07974, USA

ABSTRACT

Carbon is x-ray photoemission spectra of alkali graphite intercalation compounds are shown to provide detailed information about the distribution of charge among the carbon layers and atoms. The interior layers of higher stage compounds contain little charge and remain similar to graphite. The charge in the bounding layers is strongly localized to screen the alkali intercalate ions. The c-axis charge distribution in KC.1 2n is in reasonable accord with the electrostatic screening model of Safran and Hamann.

INTRODUCTION We have investigated graphite intercalation compounds (GIC's) using x-ray photoelectron spectroscopy (XPS), and have obtained information about the distribution of charge between the intercalate and host, and even about the details of the distribution of charge among inequivalent carbon layers and atoms. XPS is a suitable technique because the carbon Is core electron spectra are sensitive to changes in the density of electronic charge near the carbon atoms and to changes in the band structure. A drawback of XPS is its surface sensitivity. The sharp photoemission peaks are due to photoelectrons that emerge from within the bulk of the sample without an inelastic collision. Electrons with 1200 eV kinetic energy, such as make up the C is peak in our experiments, have a mean-free-path of between 10 and 203 in a wide variety of solids. Thus we may assume that the region probed by XPS includes only 5 layers of graphite. Because it is therefore crucially important to minimize the effects of surface degradation, all data were taken on surfaces freshly cleaved in the spectrometer vacuum. We derive our information primarily from shifts in the C Is binding energy measured relative to the Fermi level, EF. These shifts are due to both initial and final state effects. The final state shift is due to changes in the screening of the core hole that is produced by photoemission. In general, when the amount of mobile charge increases, the screening becomes more effecIn the initial state, tive, and hence the measured binding energy is lowered. charge transfer shifts binding energies by the direct electrostatic interaction between the additional valence charge and the core electrons. Also, charge donated to the graphite lattice fills empty states, causing the Fermi level to move relative to the core levels. These two initial state phenomena have effects of opposite sign and comparable magnitudes.

Mat. Res.

Soc. Symp. Proc. Vol. 20 (1983) C Elsevier Science Publishing Co.,

Inc.

72 A spectrum for highly oriented pyrolytic graphite (HOPG), for which the C Is core electron binding energy is 284.4 eV, is shown in Fig. 1. All data discussed here were obtained with monochromatized 1486 eV Al KW radiation. The observed width of the peak is largely due to the 0.6 eV instrumental resolution. The asymmetry and long tail toward greater binding energy are due to electronic excitations in the final state;J1,2] w

Data Loading...