Measurement of the Anomaly at theLII,IIIEdge of Sodium

Abstract

Recently interest in the absorption and emission structure near the onset of inner-shell transitions in simple metals has been revived in a series of theoretical papers. 1 " 9 Previously absorption at the edge was regarded as being due to transitions of core electrons into empty states just above the undisturbed Fermi level. This would explain a step in the absorption coefficient. The influence of the potential due to the deepstate hole was considered to be negligible because of immediate shielding by the conduction electrons. A more refined theoretical analysis yields the result that electron states in the vicinity of the Fermi surface are affected by the core potential in such a way that a singular behavior of the absorption and emission occurs at the edge 1 " 9 :where AE is the distance from the edge to higher (lower) energies in absorption (emission), \i is the absorption coefficient, respectively emission probability, a is expected to be positive for L nm transitions whereas it can also achieve negative values for K transitions. In the latter case a smoothing of the edge instead of a peak should be observed. We have considered Na as being an ideal metal to test these predictions because a peak near the onset could not be caused by a band singularity. Emission spectra of Na show a peak near the L um edge. 10 The only older absorption measurements which came to our attention are those of O'Bryan. 11 Though some peaking near the edge was observed we felt that a remeasurement with better techniques was desirable. s Synchrotron light from the Deutsches Elektronen-Synchrotron accelerator served as the source t.of radiation. The Na films were evaporated from nickel boats onto substrate films of Mg and Al which were surrounded by surfaces cooled to 77 °K. The base pressure in the system outside the cool trap was 10~6 Torr. The cooling served to i-lower water vapor pressure and has been proved * to be a good means 12 to get good metallic Na films without ultrahigh vacuum. The samples themselves were maintained at 77°K. Absorption was measured immediately after evaporation and it v was verified that none of the observed structures are caused by oxide. r Figure 1 shows our results. The absorption coefficient /i is given in arbitrary units since the n film thicknesses were not determined. The reproducibility of curves from different samples j-which were fitted at two points was within 3 % of 3 the total edge discontinuity. The accuracy of absolute energy calibration is ±0.05 eV.A peaking of the absorption towards the edge is 1 clearly recognized. The structure is somewhat 3 complicated by the spin-orbit splitting of the edge showing up as a separation of 0.17 ±0.02 eV. The observed intensity ratio of these two contribu-The absorption structure of sodium metals has been measured in the vicinity of the onset of 2p electron absorption. The results agree with theoretical investigations which claim a sharp absorption peak at the edge. 528

Optical Anisotropy of Anthracene Single Crystals for Photon Energies from 4.5 to 11.5 eV

“…The measurements were taken using the polarized continuum of the synchrotron radiation from the 7.5 GeV Deutsches Elektronen-Synchrotron DESY [15]. The radiation was monochromatized by a near normal incidence monochromator in a modified Wadsworth mount, as described in [16].…”

Optical Plasma Resonances in Solids

“…as in circular accelerators [64]. The spectrum is a broad, pure continuum and the radiation is almost perfectly polarized.…”

The Excitation of Polarization Waves in Bound Electronic States

“…Optical measurements are difficult in the energy region above 20 eV owing to the difficulty in obtaining intense light sources in this region. Synchrotron radiation seems t o be emerging as the most suitable source in this region [3].…”