It is shown that the photoluminescence (PL) dead layer model, which is widely used in the literature to fit the variations of the room temperature PL intensity versus the surface recombination velocity of III–V compound semiconductor materials, may not be suitable to practical experimental condition. It is proposed a simple analytical derivation of the PL yield which includes the contribution of radiative recombination in the near surface field region.

The probability of hot-hole injection has been measured both on metal nitride-oxide silicon (MNOS) and metal-oxide-semiconductor (MOS) structures in the case where the silicon electric field is one dimensional and normal to the interface. The experiment uses the effect of optically induced hot carrier injection as proposed by Ning et al. [J. Appl. Phys. 48, 286 (1977)]. In the case of MNOS structures, the hot-hole injection currents can be readily measured because the Si-Si3N4 interface barrier is lower than the Si-SiO2 interface barrier. Measurements on MOS structures were carried out using heavily doped silicon. The measurements have been interpreted using the lucky carrier model with some modifications: the hot-hole mean-free path has been found equal to 41±5 Å in the case of MOS structures. Taking into account the accuracy of the measurements, this value is compatible with the value derived in the case of MNOS structures and also with the value derived from ionization measurements.

No abstract

The influence of a strong magnetic field (up to 12 T), parallel to the Si-SiO2 interface on the injection of hot electrons from Si into SiO2, has been measured in the case where the heating electric field is uniform and normal to the interface. The experimental results show that the influence of the magnetic field can be either an increase or a decrease of the injection probability depending on the type of test device used. These results are compared to a numerical simulation based on the hypothesis that injected electrons are in majority ballistic (‘‘lucky’’) electrons. It is shown that a possible interpretation of the experimental results can be found with this model by taking into account the influence of interface roughness on the escape conditions of hot electrons. Numerical results based on a simple model are given that show for the first time that interface roughness may be an important factor to consider in the study of hot-electron injection physics.

No abstract