A tetrahedral model is presented to explain the bonding properties of nonstoichiometric amorphous silicon oxynitride (a-SiOxNy) alloys, grown under highly nonequilibrium conditions, whose structures obey neither the random bonding model nor the random mixture model. Based on our approach, a numerical procedure is proposed to obtain the relative atomic percentages of each component structural phase from the deconvolution of the high-resolution x-ray photoelectron spectroscopy (XPS) spectra in the Si 2p3∕2 region. The tetrahedral model is then used to study the bonding properties of a-SiOxNy films grown by electron-cyclotron resonance plasma-enhanced chemical-vapor deposition, having relatively low values of the O/Si atomic ratio (⩽0.37) incorporated in their networks. The experimental results show that five tetrahedral phases (tetrahedrons Si–Si4, Si–Si2ON, Si–N4, Si–Si3O, and Si–O4) are present in a-SiOxNy films with low N/Si atomic ratios (⩽0.93), while only three phases (Si–SiON2, Si–N4, and Si–O2N2) are present in samples with higher N/Si atomic ratios (⩾1.12). The Si3N4 phase is the most important bonding unit and it is the only phase present in all our samples. These results are corroborated by survey scans and by comparison with the high-resolution XPS spectra in the N 1s region. They support the validity of the model proposed for a-SiOxNy alloys and the XPS analysis, correlated with growth conditions, presented in this work.

No abstract

A new formalism is reported for the analysis of the current–voltage (I–V) characteristics of a tunnel metal-insulator-semiconductor (MIS) device, which considers a bias dependent distribution of interface states and barrier lowering due to the image force. Our theoretical expression for the I–V characteristics is general in the sense that it is applicable even under conditions when both the thermionic emission and the diffusion mechanisms of current transport compete with each other. The method is ideal for new epitaxial materials and devices where the carrier density is not known precisely beforehand. A self-consistent method of analysis is reported to determine the characteristic parameters of MIS diodes, using simultaneously the I–V and capacitance–voltage data as a function of temperature. This computational analysis has been used to examine the current transport mechanism in an Au/p-InP epitaxial MIS diode. The experimental verification of the theory and computational analysis is done by comparing the values of the interface state density distribution in thermal equilibrium with the semiconductor Nss, obtained from the forward I–V characteristics, with those directly measured by the multifrequency admittance method. Excellent agreement from these comparisons strongly supports the validity of the theory. Over the temperature range of 200–393 K, our results indicate that the interfacial layer-thermionic emission was clearly the dominant mechanism of the forward current transport in an MIS fabricated on a lightly doped InP:Zn epitaxial layer. The transmission coefficient through the insulator layer obtained from the reverse I–V characteristics was θp=1.43×10−3±7% from which we estimate an oxide thickness of 2.2 nm. The analysis of the barrier height φb0 versus temperature, obtained from 1 MHz C–V data provided a value φ0=1.06 V±10% for the zero bias and zero temperature barrier height.

Abstruct-A new degradation mechanism of PM-HEMT's subsequent to hot electron stress tests or high temperature storage tests is presented. A noticeable increase in drain-to-source current, IDS, is observed after the tests. We show that this IDS variation is slowly recoverable and is correlated with the presence of deep levels in the device. Stress tests cause a variation of trapped charge. Trapping of holes created by impact-ionization and/or thermally stimulated electron detrapping induce a variation of the net negative trapped charge, leading to a decrease in the threshold voltage, VT and a consequent increase in IDS. The correlation between %AVT and AIDS clearly demonstrates that the variation of trapped charge induced by hot electron tests is localized under the gate.

No abstract