No abstract

The excitation spectrum of the 2.8-eV luminescence band of crystalline Si02 has been measured in a photon energy range between 6 and 14 eV at 77 K. We find that the onset of the 2.8-eV luminescence occurs at 8.3 eV, which is nearly equal to the fundamental optical absorption edge of SiOz. This result supports strongly the model that the band is due to the radiative recombination of the self-trapped excitons.Irradiation of Si02 with ionizing radiation at low temperatures gives rise to strong blue luminescence, ' the origin of which has been a topic of interest. The blue luminescence, which is known to be a composite, has been suggested to include the radiative recombination of the self-trapped excitons. So far only the excitation spectra for the composite have been obtained, and the spectrum for the luminescence due to the self-trapped excitonsis not yet known.The composite nature of the blue luminescence was first demonstrated by the change in the luminescence spectrum with temperature.Two distinct bands at 2.5 and 2.8 eV have been identified by measurements of the polarization characteristics of the x-ray-induced luminescence in crystalline Si02 by Tanimura and Halliburton.They showed that the 2.5-eV band is polarized perpendicular to the z axis (threefold axis of the crystal) and that the 2.8-eV band is parallel to the z axis. We showed that the 2.5-eV band has a decay time of 1.2 ms at 80 K, while the 2.8-eV band has a decay time of 940 ps at the same temperature.Other workers have obtained several peaks for the blue luminescence by comparing the luminescence spectra at several temperatures.For instance, Alonso etal. have found peaks at 2.8, 2.9, and 3.2 eV. Since their spectrometer was not calibrated, the peak energies are not accurate and their 2.8 and 2.9-eV bands have been ascribed to the 2.5 and 2.8-eV band, respectively.Trukhin and Plaudis and Trukhin have observed a peak at 2.6 eV but their peak appears to be a composite of 2.5 and 2.8 eV bands, in view of their studies of the temperature dependence. Kristianpoller has observed the peaks at 2.8, 3.26, and 3.54 eV. Grinfelds, Aboltyn, and Plekhanov have observed a broad peak at 2.5 eV, and showed also that introduction of Cu impurities produces a band at 3.4 eV. Summarizing these observations, all of the authors have observed a luminescence band around 2.8 eV, but the presence of other peaks depends on the specimen. Furthermore, the results that the intensity of the 2.5-eV band saturates with increasing the intensity of the excitation appears to indicate that the band arises from defects or impurities. Thus it is most likely that only the 2.8-eV band is intrinsic.Further detailed information has been accumulated on the 2.8-eV luminescence band. It has been shown' that its intensity is proportional to the density of excitation up to about 10' cm . Time-resolved optical absorption measurements have indicated the presence of two allowed optical transitions at 5.2 and 4.2 eV from the initial state of the luminescence to higher states. A transient volume chan...

The transient volume change of o-quartz and fused silica induced by irradiation with an electron pulse has been measured above 80 K. It is shown that transient changes of volume and optical absorption due to the E{ centers (oxygen vacancies) decay in parallel and that the volume change per E{ center is of the order of a unit molecular volume. The results show unambiguously that recombination-induced defect formation occurs in Si0 2 but the defects created are not stable even at low temperatures.PACS numbers: 61.70.Bv, 78.20.Hp, 78.50. It is known that intrinsic photolytic reactions, the creation of intrinsic lattice defects by electronic excitation, occurs in some ionic solids such as alkali and silver halides. 1 In alkali halides, of which the photolytic reaction has been understood most satisfactorily, the primary process is known to be the ejection of a halogen atom to a neighboring interstitial position from a highly excited nonbonding state of the selftrapped exciton. 1 ' 2 Whether such intrinsic photolytic processes are induced in other solids, such as oxides and semiconductors, is a problem of general interest. In these materials it is generally believed that the knockon process creates point defects 3 ' 4 but not the electronic excitation.In Si0 2 , Griscom 5 and Hobbs et al. 6 have suggested that a close pair of an oxygen vacancy and an interstitial oxygen is generated by electronic excitation. Griscom has found that a transient optical-absorption band in the ultraviolet region is induced by bombardment of a-quartz and fused silica with an electron pulse, and he ascribed the optical-absorption change to creation of a singly ionized oxygen vacancy (the E x ' center), of which the yield is much higher than that expected from the knockon process. It is not yet clear, however, whether the transient opticalabsorption band arises from momentary trapping of holes at isolated neutral oxygen vacancies existing before irradiation or from newly created ionized vacancies by a photolytic reaction. Hobbs and co-workers have observed, by an electron microscope, formation of heterogeneously nucleated disordered strain centers and a subsequent homogeneous crystal-amorphous transformation by irradiation with low-energy electrons that does not cause any knockon processes. They suggested that the strain centers are vacancy-interstitial pairs originating from an intrinsic photolytic reaction. The phenomena they observed are nonlinear effects that are observed Ec only under dense electronic excitation and apparently are not a simple photolytic process. Thus direct experimental proof of the intrinsic photolysis for Si0 2 is needed. Since the generation of vacancies and interstitials by photolytic reactions is proportional to the absorbed energy of ionizing radiation and gives always a considerable volume change, 7 the volume change per absorbed energy is the most direct measure whether such intrinsic photolysis is effective. The present Letter reports the result of measurements of the volume change of a -quartz and fused silic...

No abstract

No abstract