The electronic states of the Si-SiO2 interface.

The presence of carrier traps, at the Si-SiO2 interface, is at the origin of various instability phenomena encountered both in bipolor and FET devices, and described at length in Chaps. 14 and 15 respectively. The nature of these traps (also called interface states), their properties, their impact on C(V) curves and the models used to explain them, have all been developed in Cap. 11. In the present chapter we describe the electrical techniques devised to measure the characteristic parameters of interface traps, and how these techniques can be used for diagnosis. We first describe the quasistatic method (QSM), based on the analysis of the high frequency and low frequency C(V) curves, through two of its variants: the combined lf-hf method and the method of Berglund. The QSM enables us to measure the density of states around mid-gap. Although it is not very sensitive, it is still widely used, but as a quick diagnostic tool. The QSM has found many other applications in the characterization of the silicon substrate underneath the oxide. Each application is described in a separate section. We thus review: the measure of uniform doping concentrations and of doping profiles, the detection of lateral non-uniformities (in surface potential), the study of bulk traps, the measure of the minority carrier generation lifetime, and the study of what may cause the hysteresis of C(V) loops. We next present the conductance method, based on the measurement of the complex admittance of the MOS capacitor. This method let us measure, not only the density of states but also their capture cross-section. The deep level transient spectroscopy (DLTS) is based on the analysis of the transient (capacitance or voltage) signal generated by the re-emission of the carriers trapped in the interface states. We describe that variant called the constant capacitance (CC-DLTS). It allows us to measure the density of those states located between mid-gap and the majority carrier band edge, as well as their