Cathodoluminescence of cerium dioxide: Combined effects of the electron beam energy and sample temperature

Abstract

In-situ cathodoluminescence (CL) spectroscopy is used to study point-defect formation in cerium dioxide (CeO2) by high-energy electrons (400 keV-1,250 keV) at ∼100 K, 200 K, and 300 K in a high-voltage electron microscope (HVEM). Complementary CL spectra are also obtained for 20-keV electron excitation at ∼300 K in a scanning electron microscope (SEM). Experiments were carried out on a single crystal and polycrystalline sintered sample. The more prominent and broad emission band centered at a photon energy of ∼4.2 eV is ascribed to F+ centers (oxygen vacancies) produced by the high-energy electron irradiation. Two other weaker CL bands centered at ∼2.3–2.4 eV and 2.8–2.9 eV at 300 K are related to trivalent cerium ions, corresponding to 5d → 4f radiative transitions, regardless of electron energy. Similar spectra are recorded at 100 K and 200 K, yet with shifts and broadening of the latter emission bands. A maximum of all CL bands is found at ∼600 keV electron energy for the polycrystalline sample regardless of temperature, and for the single crystal at 300 K. In contrast, a continuous increase versus electron energy is observed for the 4.2-eV band of the single crystal at 100 K and 200 K above a threshold corresponding to the oxygen displacement energy. The dependence of CL intensities on the primary electron energy is analyzed on the basis of the interplay of the luminescence cross section and oxygen displacement cross section. The electron energy dependence of ionization cross sections is addressed by relying on secondary electron spectra computed with the PHITS code which is benchmarked against other computer codes such as PENELOPE, as implemented in Geant4. The F+ center band intensity is strongly reduced in the polycrystalline sample with respect to the single crystal in the same irradiation conditions regardless of temperature due to grain boundaries.