Answers are grounded in this paper and cite their sources.
The high-resolution x-ray photoemission spectra (XPS) of the total valence bands of atomically clean diamond, graphite, and glassy carbon, obtained with monochromatized Al Kα radiation, are reported and discussed. By comparing valence-band and carbon-$1s$ photoelectron kinetic energies, the XPS valence-band spectra I^'(E) of diamond and graphite were rigorously affixed to the same energy scale as earlier K x-ray emission spectra I(E). The two spectra---I^'(E) and I(E)---have very different energy dependences of intensity because selection rules and cross-section ratios render I(E) sensitive only to $2p$ character and I^'(E) far more sensitive to $2s$ character. Taken together, I^'(E) and I(E) show that the fractional p character in the diamond valence band increases from {~} 16% at the bottom of the band to {~} 92% at the top, with an average hybridization of ~s1.2p2.8. The spectra agree well with the density of states of Painter et al., but indicate a valence-band width of 24.2(10) eV rather than their 20.8 eV. The C(1s) binding energy of 284.68(20) eV in graphite agrees well with a recent theoretical estimate of 284.4(3) eV by Davis and Shirley. Analysis of I^'(E) and I(E) for graphite resolves the valence bands cleanly into σ and π bands, with the spectrum I^'(E) of the former resembling that of diamond, but with a stronger $2s$ admixture (sp² vs sp³). The XPS cross section of the (pz)π bands was very low, as expected by symmetry. The bandwidth of 24(1) eV somewhat exceeded Painter and Ellis's calculated value of 19.3 eV. Glassy carbon showed an I^'(E) between that of diamond and graphite, consistent with an amorphous lattice containing both trigonal and tetrahedral bonds.
No takes yet. Share an insight, caveat, or question.
McFeely et al. (1974) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: